Auditory pathway functioning in prelingual deafness The clinical consequences for cochlear implantation
Marc Lammers
104 ISBN 978-90-393-6435-2
Auditory pathway functioning in prelingual deafness The clinical consequences for cochlear implantation
Marc Jan-Willem Lammers
The research described in this thesis was performed the University Medical Center Utrecht, Utrecht, the Netherlands. The research was partly funded by an unrestrictive research grant by Cochlear Ltd.
Financial support for the publication of this thesis was kindly supported by: Cochlear Ltd. Afdeling Keel- Neus- Oorheelkunde UMC Utrecht Nederlandse Vereniging voor Keel- Neus- Oorheelkunde en Heelkunde van het HoofdHalsgebied Stichting ORLU Brain Center Rudolf Magnus ChipSoft B.V. Meda Pharma EmiD audiologische apparatuur Meditop B.V. Olympus Nederland B.V. Specsavers Tramedico B.V. Daleco Pharma B.V. Dos Medical BV/ KNO-winkel.nl ZEISS ALK Oticon Medical Cover design by Marc Lammers Cover image represents a CochlearÂŽ CI512 cochlear implant with the ABR waves III and V, and the cortical P1-N1-P2 peaks along its array
Layout and printed by Gildeprint, Enschede, the Netherlands ISBN: 978-90-393-6435-2 Copyright Š by M.J.W. Lammers 2015. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature or transmitted in any form of by any means without prior permission of the author, or when appropriate, the publishers of the papers.
Auditory pathway functioning in prelingual deafness The clinical consequences for cochlear implantation De werking van het auditieve systeem als gevolg van prelinguale doofheid De klinische consequenties voor cochleaire implantatie (met een samenvatting in het Nederlands)
Proefschrift ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op vrijdag 6 november 2015 des ochtends te 10.30 uur
door
Marc Jan Willem Lammers geboren op 21 augustus 1985 te Nijmegen
Promotor:
Prof. dr. W. Grolman
Copromotoren:
Dr. H. Versnel Dr. G.A. van Zanten
Table of contents General introduction
7
Part I Chapter 1.1 Chapter 1.2 Chapter 1.3
The influence of the newborn hearing screening on the age at cochlear implantation in children Bilateral cochlear implantation in children: a systematic review and best-evidence synthesis Bilateral cochlear implantation in children and the impact of the inter-implant interval
Part II Chapter 2.1 Chapter 2.2 Chapter 2.3
Predicting performance and non-use in prelingually deaf 83 and late-implanted cochlear implant users Delayed auditory brainstem responses in prelingually 101 deaf and late-implanted cochlear implant users Altered cortical activity in prelingually deafened cochlear 121 implant users following long periods of auditory deprivation
Part III Chapter 3.1 Chapter 3.2
The clinical feasibility of postoperative evoked potential recordings in cochlear implant users Cortical auditory evoked potentials to frequency changes with varied size, velocity and direction
29 43 65
145 155
General discussion
169
Summary Nederlandse samenvatting
189 197
Dankwoord List of publications About the author
205 211 215
General introduction & aim and outline of this thesis
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 8
General introduction
Introduction In this thesis we discuss the consequences of prelingual deafness on auditory pathway maturation and functioning after cochlear implantation. Prelingual deafness can be defined as severe to profound hearing loss with its onset during early language acquisition, i.e. before the age of 2 years. In the first part we focus on factors that play a role on therapy outcomes after cochlear implantation in children with prelingual deafness. We discuss the role of neonatal screening of hearing, and we evaluate the benefit of a second implant. In the second part we investigate the consequences of prelingual deafness in adult cochlear implant users, who only received their implants during adulthood. By evaluating their performance after cochlear implantation and by studying the activation of their auditory pathway with electrophysiological recordings we gain a better insight in the consequences of long term hearing loss on auditory pathway maturation. In the third part we discuss the role of objective neurophysiologic recordings for clinical and research goals. Finally, in the general discussion we elaborate on our findings and propose a model combining the multifactorial consequences of prelingual deafness on auditory pathway maturation.
Time and cochlear implantation For centuries, hearing and deafness has been a mystery for physicians, philosophers and scientists. In ancient civilizations, hearing loss was thought to be an act of demons and evil, but thorough observations by the Egyptians eventually led to the recognition of associations between diseases and deafness (Mudry, 2006; Psifidis, 2006). In medical papyri, like the Ebers and Edwin Smith papyrus, dating back to 3500 - 1500 B.C., they were the first to describe the correlation between temporal bone fractures and deafness and already proposed the use of honey for the treatment of chronic ear discharge (Mudry, 2006; Psifidis, 2006). With the rise of the ancient Greek civilization more anatomical knowledge of the ear canal and the tympanic membrane was acquired by physicians like Hippocrates (460 – 379 B.C.) and the first theories on sound waves and hearing were proposed by philosophers like Plato and Aristotle (Psifidis, 2006). It was not until the Roman era before anatomical studies by Galen (130 – 210 AD) shed light on the structure of the inner ear and the vestibulocochlear nerve (Psifidis, 2006). With the Renaissance the interest in acoustics revived. Technological developments, like the invention of the microscope made it possible to study the inner ear structures in detail. In 1704 Antonio Maria Valsalva (1666 – 1738) published his ‘De aure humana tractatus’ in which he clearly described the anatomy of the cochlea and the different scalae (Meirelles R.C. et al., 2008). Although his studies can be considered as a scientific breakthrough, it was not well
9
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
known how the inner ear works. Nevertheless, at the end of the eighteenth century Alesandro Volta, the inventor of the voltic pile, was the first to experiment with electrical stimulation of the ear (Blume, 2009). He placed metal rods in his ears and connected these to his battery. The electrical current which passed through his head produced the sensation of “a boom within the headâ€?, followed by sounds similar to the boiling of thick soup (Blume, 2009). In the following centuries this experiment has been altered by various scientists like Duchenne of Boulogne and Brenner, but it took until 1957 before the first detailed description of electrical stimulation of the auditory nerve was published by Djourno and Eyriès (Blume, 2009; Djourno et al., 1957; Eisen, 2003). They were the first who investigated whether electrical stimulation of the inner ear could restore hearing in deaf patients. By placing a single electrode in the auditory nerve they provided the patient the ability to discriminate high and low frequencies (Djourno et al., 1957; Eisen, 2003). Since then, technologies have improved rapidly leading to the first commercially available House 3M single-electrode cochlear implant (FDA approval in 1972) and the development of the multichannel cochlear implants (CI) in the eighties (Fig. 1) (Blume, 2009).
Fig. 1. Schematic depiction of a modern multichannel cochlear implant (CI). The cochlear implant system consists of an external part comprising the behind-the-ear processor and transmitting coil, and an internal part containing the receiver and the multichannel electrode array. Sound is picked up by the microphone in the speech processor, and is encoded to a digital signal which is led to the transmitter coil. This coil transmits the signals through the skin to the subcutaneous receiver. From there, the signals are carried to the electrode array, which is surgically placed within the cochlea. Sounds are encoded in such a way that different frequencies stimulate different electrodes within the array, resulting in depolarization of separate neuronal populations within the cochlea. Since the CI uses the specific tonotopic organization of the cochlea, stimulating the various intracochlear electrodes results in the perception of different pitches. Courtesy of Cochlear Ltd.
10
General introduction
In 1985, the first cochlear implantation in the Netherlands was performed in the University Medical Center Utrecht. The first patients in our center received the House 3M single electrode cochlear implants. Although this implant had only one electrode in its array, patients were able to perceive sounds. However, their performance was limited due to the small number of electrodes. In the following years the various implant companies, including the Dutch company Philips, designed and produced implant arrays with multiple electrodes. With the introduction of these multichannel electrodes, performance significantly improved and speech perception became possible for most CI users. Since then cochlear implantation has become a very successful treatment for severe to profound sensorineural hearing loss in both children and adults. Although speech perception scores in quiet are fairly good for most patients, not everyone benefits equally from their implants (Holden et al., 2013; Teoh et al., 2004; Wilson et al., 2008). Especially early implanted children and adults with late onset deafness can obtain high levels of speech perception, with star performers who can even appreciate new musical compositions. If the duration of deafness increases, the results are found to deteriorate (Boons et al., 2012; Holden et al., 2013). In children it has been observed that they can show a remarkable catch-up in speech and language development as long as they receive their implants early in life (Boons et al., 2012; Niparko et al., 2010).
Fig. 2. Maximum postoperative CVC phoneme scores. Univariate logistic regression analyses reveals a significant group effect of age at cochlear implantation on postoperative speech perception (F = 7.0, P < 0.001). The average speech perception score for the age group under 18 months is 91% (n = 11), for the age group 18-36 months 85% (n = 41), for the age group 36-54 months 77% (n = 17), and for the children implanted older than 54 months 69% (n = 54).
Analyses of the speech perception development of 123 children in our clinic by Krijnen and Lammers revealed a similar pattern. Children who received their implants under the age of 18 months obtained highest scores with an average of 91% (Fig. 2). When they were treated after 4.5 years of age average speech perception scores decreased to 69%
11
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
correct. Extending this period of severe hearing loss during childhood reduces the success of cochlear implantation even more. Postoperative speech perception of adults with early onset deafness who received their implants during adulthood are known to be rather poor and highly variable (Caposecco et al., 2012; Klop et al., 2007; Straatman et al., 2014; Teoh et al., 2004; Yang et al., 2011). The few reports in the literature on these, so called prelingually deaf patients who became deaf before early language acquisition (i.e. before the age of 2 years), show that average speech perception scores range from 1% to 45% (van Dijkhuizen et al., 2011; Yang et al., 2011). As described in more detail in Chapter 2.1, the average maximum speech perception scores of all prelingually deaf and late-implanted (>18 years) adults in our clinic was 41%. Half of the patients (48%) was not able to obtain any speech perception and could only use their implants for sound detection. Ten patients (21%) eventually decided not to use their implants due to the limited benefit. These poor results contrast with the level of functioning of postlingually deaf adult CI users and is most likely caused by differences in auditory pathway development, as discussed in the second part of this thesis.
The auditory pathway and its development The auditory pathway extends from the cochlea to the auditory cortex (Fig. 3). In the cochlea peripheral nerve fibers of the spiral ganglion cells run towards the organ of Corti where they innervate the approximately 3000 inner hair cells (Kandel et al., 2000). Their central axons eventually form the vestibulocochlear nerve and terminate in the cochlear nucleus of the brainstem. The highly specific tonotopic organization of the cochlea is preserved by the arrangement of the auditory nerve fibers which enter the brainstem in the cochlear nucleus and is maintained throughout the entire auditory pathway. Neurons in the cochlear nucleus send their axons via three main pathways further towards the brain: the dorsal acoustic stria, the intermediate acoustic stria and the trapezoid body (Fig. 3). The most important pathway is via the trapezoid body where the nerve fibers cross the brainstem and innervate the contralateral superior olivary nucleus (Kandel et al., 2000). The olivary nuclei are important in sound localization as the cells within these structures are sensitive to both time and sound level differences between the two ears. Axons from the superior olivary nucleus form, together with crossed and uncrossed fibers, the lateral lemniscus which terminate in the inferior colliculus. From there, most axons project to the medial geniculate nucleus of the ipsilateral thalamus, but a few fibers cross to the contralateral side. From the medial geniculate nucleus the primary auditory cortex is innervated, which is located in the superior and transverse temporal gyri (Heschlâ&#x20AC;&#x2122;s gyrus).
12
General introduction
Fig. 3. The auditory pathway from the cochlea to the auditory cortex. From the medulla crossed and uncrossed fibers run via the three main pathways: the dorsal acoustic stria, the intermediate acoustic stria and the trapezoid body towards the pons. From there axons form the lateral lemniscus and terminate in the inferior colliculus. From the midbrain the pathway runs thalamus to eventually terminate in the primary auditory cortex. Adapted from Kandel et al. 2000
The auditory cortex can be roughly divided into the more centrally located primary auditory cortex and the secondary auditory cortex which comprises the peripheral areas. Each cortical area consists of six layers predominately composed of pyramidal and stellate cells and their axons (Fig. 4). Axons of pyramidal cells are known to give off collateral branches, which can terminate in other cortical regions. Whereas axons of the stellate cells
13
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
branch towards cortical layers near the cell body. The most superficial layer, layer I, is made up largely of glial cells, whereas layer II and III, the supragranular layers, are closely packed with pyramidal cells giving rise to cortico-cortical connections to other cortical regions. The granular layer, layer IV, is rich in stellate cells and receives most of the afferent input from the medial geniculate nucleus in the thalamus. Layers V and VI, the infragranular layers, contain large pyramidal cells which project back along descending pathways to the thalamus (layer VI) and the inferior colliculus in the auditory brainstem (layer V).
Fig. 4. Maturation of the human primary auditory cortex (adapted from Conel 1939 â&#x20AC;&#x201C; 1967). These illustrations demonstrate the increase in dendritic complexity occurring after birth in all six cortical layers. Numbers I to VI denote the six cortical layers found in the auditory cortex.
The development of the inner ear and the auditory pathway already starts in the first weeks after gestation. The first signs of the developing inner ear structures become visible when the otic vesicle divides into a vestibular and cochlear segment (Moore et al., 2007).
14
General introduction
Thereupon the cochlear segment evolves into the cochlear duct, which elongates and starts to coil (Moore et al., 2007). By the 8th week after gestation the cochlear duct has already reached the full two and a half turns. Afterwards the first cells of the organ of Corti become visible and the first hair cells are formed. Parallel to the formation of the cochlea, the auditory nerve and brainstem develop. Around the 4th week the first ganglion cells become present and begin to form axonal processes in the direction of the organ of Corti and towards the auditory brainstem, forming the cochlear nerve. In the second trimester further maturation of the cochlea and neurons in the auditory pathway continues. By the 15th fetal week the scala tympani and scala vestibule are created and by the end of the second trimester the cochlea has a matured appearance. With this rapid development of the cochlea in the second trimester, also the cochlear nerve grows and by the 22nd gestational week the myelination process commences in the cochlea (Moore et al., 2007). The myelination process follows a peripheral to central pattern. In the third trimester myelin can be identified from the cochlear nerve through the brainstem up to the auditory thalamus (Matschke et al., 1994; Moore et al., 2007; Moore et al., 1995). At the beginning of the second trimester the cerebral cortex only consists of closely packed immature neurons. During the second trimester the cortex thickens and vertical columns of neurons start to originate in the cortex (Moore et al., 2007). At the 27th week of gestation the temporal lobe has been formed and consists of highly packed superficial layers. It is at this stage that fetuses begin to display the first responses to sounds. Besides behavioral responses which have been detected with ultrasound imaging in response to sounds, auditory brainstem responses (ABRs) and even cortical auditory evoked potentials (CAEPs) can be recorded in premature infants with a conceptional age of only 27 weeks (Ponton et al., 1992; Ponton et al., 1993; Rotteveel, 1992; Starr et al., 1977). Although the cochlea is already present in its mature form by the end of the second trimester, it is not until term birth and sometimes even longer before all synapses are fully developed and mature otoacoustic emission patterns can be recorded (Eggermont et al., 1996; Hof et al., 2013). In the third trimester and the first months after birth the auditory brainstem further increases in size and myelination continues. By the age of one year the myelination of the cochlear nerve is comparable to that in adults (Moore et al., 1995) and is thought to be completed in the auditory brainstem only a few years later (Eggermont et al., 1988; Inagaki et al., 1987; Moore et al., 2007; Moore et al., 1995). This process of myelination probably explains the decreasing ABR peak latencies found in this early stage of life (Eggermont et al., 1988; Inagaki et al., 1987; Moore et al., 1995; Ponton et al., 1992). In the late fetal period the temporal lobe further develops, resulting in the formation of the transverse and superior temporal gyrus containing the auditory cortex (Moore et al., 2007). During childhood these gyri slowly increase in size, to reach full maturation around early adulthood. By the end of the fetal period, the auditory cortex is around 1.2 mm thick
15
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
and only the cortical layers I, II and IV can be distinguished. After birth, neuronal cell size increases and by 4.5 months all six cortical layers can be distinguished (Fig. 4). Around the age of one year the cortex reaches its adult thickness of 1.8-2.0 mm and more axons arising from the thalamus become visible (Conel, 1939; Moore et al., 2007). During early childhood cortical neurons enlarge and the density of dendrites and axons increases. In the deeper cortical layers axonal maturation and myelination progresses slowly up to the age of six years (Moore et al., 2007; Pujol et al., 2006). With the increasing density of dendrites within the cortex also the number of synapses increases, reaching its maximum around the age of 4 years (Conel, 1939; Huttenlocher et al., 1997; Kral et al., 2010; Kral et al., 2005). Thereupon synaptic counts decrease, possibly indicating the process of synaptic pruning in which unused synapses and cortical connections are eliminated (Huttenlocher et al., 1997). The synaptic pruning process can be seen as a maturational process of the brain reflecting its process of specialization. In later childhood (age six to twelve years) axonal maturation progresses in the supragranular layers II and III (Conel, 1939; Moore et al., 2001). It is not until the end of childhood before these layers are thought to be fully matured (Albrecht et al., 2000; Conel, 1939; Moore et al., 2001; Ponton et al., 2001; Ponton et al., 2000). The neurons in these layers are found to have connections with adjacent areas of the auditory cortex and even from the auditory cortex on the contralateral side (Kral et al., 2007; Moore et al., 2007). This maturational process is therefore important in the development of cortico-cortical connections between the auditory cortex and other cortical regions (Kral et al., 2007). Since more complex auditory processing and language skills are acquired in the second half of childhood it is thought that this arises from the development of more and more complex cortico-cortical connections within and between the two hemispheres. This complex and protracted process of language acquisition probably underlies the relative slow maturation of the auditory cortex in comparison to other primary sensory cortices which are thought to be mature within the first years of life (Kral, 2013).
Electrophysiological recordings of auditory pathway activation Voltage changes within the neurons and their synapses of the auditory pathway can be registered with electrodes placed closely to the activated cells or at more distant sites on the scalp. The continuous electrical activity from multiple cortical neurons in the different brain regions can be visualized in an electroencephalogram (EEG). Depolarizations of several, equally lined cortical neurons result in changes in the continuous EEG. During sleep the EEG shows large periodic, low-frequency Beta waveforms, whereas in an active brain more highfrequency oscillations are seen. By using multiple scalp electrodes, location specific brain
16
General introduction
activity can be determined. Occipital placed electrodes are most sensitive to activity changes in the visual cortex, whereas activity of the auditory cortex is most optimally recorded at temporal locations. Electrodes at the vertex are frequently used in the clinic, since they record activity from both hemispheres (Burkard et al., 2007). Presenting a certain sensory stimulus results in depolarization of several neurons and subsequently to a change in the continuous EEG activity. This stimulus induced response is called an evoked potential or evoked response. Sound stimuli evoke specific EEG changes at a fixed time-interval after the stimulus onset and are called auditory evoked potentials (AEPs). The amplitudes of evoked potentials are in generally smaller than the continues EEG activity, therefore repetitive stimuli are required to record these evoked potentials. This process of signal averaging makes use of the time-locked fashion in which evoked potentials occur and the similarity in waveform shape of the successive responses, which is in contrast with the highly variable background EEG activity (Burkard et al., 2007).
1. Classification of Auditory Evoked Potentials The AEPs can be categorized according to the peak latencies of the recorded waveforms in short-latency, middle-latency and long-latency AEPs (Burkard et al., 2007; LamorĂŠ, 2011).
Fig. 5. Auditory evoked potentials (AEPs). The short-latency AEPs occur within the first 10 ms after stimulus onset. Middle-latency AEPs between 10 and 50ms, and the long-latency AEPs afterwards. Adapted from LamorĂŠ 2011.
Short-latency AEPs The short-latency AEPs have latencies of less than 10 ms and constitute of the cochlear microphonics (CM), summating potential (SP), the compound action potential (CAP) and the auditory brainstem response (Fig. 5). The CM and SP are associated with the receptor potentials of both the outer and inner hair cells. The CAP is the result of synchronous neural action potentials in the cochlear nerve and is best evoked with transient stimuli, such as clicks or tone pips. The CAP can be recorded in the ear canal, with transtympanic electrocochleography or with scalp located electrodes (Burkard et al., 2007). It is composed of a dominant negativity (N1) followed by a positive peak (P1). When measured with scalp
17
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
electrodes, the CAP is represented by the first positive wave (wave I) after stimulus onset, occurring at approximately 1.5 ms (click stimulus). The auditory brainstem response (ABR) can be recorded with scalp electrodes during the first 10 ms after stimulus onset. The ABR consists of six waves of which wave I, III and V are most reliably recorded (Fig. 6) (Burkard et al., 2007). As mentioned, wave I represents the CAP, the activity of the auditory nerve. Wave III is generated in the cochlear nucleus, and wave V is thought to be generated by cells in the lateral lemniscus or inferior colliculus (Fig. 6). Wave V is the most prominent peak occurring approximately 4 ms after wave I. The short-latency AEPs are little affected by alertness or attention, but while amplitudes are rather small, they are easily contaminated by stimulus or muscle artefacts at high stimulation levels (Burkard et al., 2007).
Fig. 6. The auditory brainstem response (ABR) and localization of the neuronal generators of the six ABR waves within the auditory brainstem. Illustration adapted from the â&#x20AC;&#x153;Ear Anatomyâ&#x20AC;? series by Robert Jackler and Christine Gralapp.
Middle-latency AEPs The middle-latency AEPs are frequently referred to as the middle-latency responses (MLRs). MLRs consist of three positive (P0, Pa, Pb) and two negative peaks (Na, Nb) occurring between 10 and 50 ms after stimulus onset (Fig. 5) (Burkard et al., 2007). Neural generators of the MLR are believed to be located in the thalamus, in subcortical regions and in the primary auditory
18
General introduction
cortex. The most consistent and prominent component is the Pa with a peak latency of about 25 ms. The following positive peak Pb, follows Pa by about 25 ms but is less consistent and may be absent in adults. Furthermore this peak overlaps with the shortest-latency response of the long-latency AEPs (P1). Besides the relative inconsistency of the MLR waves, the Pa peak is particularly affected by contamination of activation of the postauricular muscle, which especially occurs at higher stimulus levels (Burkard et al., 2007). Long-latency AEPs The long-latency AEPs comprise all responses after 50 ms (Fig. 5). These potentials, sometimes also referred to as event related potentials (ERPs), can be classified as two types, obligatory (or exogenous) and cognitive (or endogenous). Amplitude of the obligatory potentials is more dependent on the acoustic parameters of the stimulus, whereas the endogenous potentials are affected by the listeners cognitive status and attention. The first components of the long-latency AEPs consist of the P1-N1-P2-(N2) complex and are commonly referred to as the cortical auditory evoked potentials (CAEPs), since their generators are located in the primary and secondary auditory cortex (Burkard et al., 2007; Martin et al., 2008). Although the P1, N1, and P2 are considered to be obligatory or sensory-evoked potentials, they are not purely sensory and can be affected by alertness and attention. The second negative peak, N2, has been associated to higher level, discriminative processes and is more sensitive to attention and tasks (Burkard et al., 2007; Martin et al., 2008). The following responses like the P300 and the mismatch negativity (MMN), occurring around 300 ms, can only be elicited using oddball-paradigms and are considered to be predominantly cognitive potentials (Burkard et al., 2007). Since the cortical P1-N1-P2 complex is an obligatory onset response it primarily reflects sound detection, whereas the MMN and P300 reflect both the pre-attentive and attentive discrimination of two or more sounds. CAEP morphology The first cortical P1 peak appears about 50 ms after stimulus onset and coincides with Pb peak of the MLR (Fig. 5). In adults the P1 peak is rather small, whereas the N1 is the most robust and best discernible component of the CAEP (Albrecht et al., 2000; Burkard et al., 2007; Martin et al., 2008). The N1 follows around 100 ms and is composed of at least three components with their generators in the primary and secondary cortex (Burkard et al., 2007; Martin et al., 2008). Latency of the second positive wave, P2, is more variable and can be found between 120 and 250 ms in adults. This variability might be explained by its multiple generators located in various auditory areas, including the primary and secondary cortex (Burkard et al., 2007; Martin et al., 2008).
19
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
2. The maturation of the cortical auditory evoked potential During childhood the CAEP is dominated by a large positive peak around 100ms, followed by a large negativity at approximately 200ms (Albrecht et al., 2000; Ponton et al., 2000; Ponton et al., 2001). Around the age of nine years the first signs of the N1 peak become visible as a small negativity within the positive peak (Bishop et al., 2011; Ponton et al., 2000; Ponton et al., 2001). Over the following years this small N1 gradually becomes larger and more dominant resulting in the more mature P1-N1-P2 complex (Fig. 7). With this development also the latency of the P1 peak decreases over the years to around 50 ms in adults (Albrecht et al., 2000; Ponton et al., 2000; Ponton et al., 2001; Sharma et al., 2002; Sharma et al., 2007; Wunderlich et al., 2006a; Wunderlich et al., 2006b). Evidence on latency changes of the N1 peak is less consistent, but the response seems to be less robust in childhood (Albrecht et al., 2000). During adulthood peak latencies are typically found around 90-100ms (Albrecht et al., 2000; Ponton et al., 2000; Wunderlich et al., 2006a).
Fig. 7. Development of the cortical auditory evoked potential during childhood as measured with the recording electrode at Cz. Adapted from Ponton et al. 2000
Recordings of cortical responses in hearing impaired children, treated with cochlear implants revealed that latencies of the P1 peak decrease following the normal pattern, but only if children are implanted early in life (preferably under the age of 3.5 years) (Dorman et al., 2007; Sharma et al., 2002; Sharma et al., 2009). If prolonged deafness during childhood results in absent or aberrant CAEP waveforms is still unknown while little attention has been paid to congenitally deaf patients with long durations of hearing impairment. In Chapter 2.3 we therefore investigate the cortical responses of prelingually deaf and late-implanted CI users.
20
General introduction
Aim and outline of this thesis In this thesis we elaborate on the consequences of early-onset deafness by evaluating clinical functioning after cochlear implantation and by studying auditory pathway maturation using evoked responses. The main goal of this thesis is to reveal the detrimental effects of deafness on auditory pathway maturation and propose evidence based strategies to prevent maturational deficits to occur. This thesis consists of three parts, with the following specific aims: â&#x2C6;&#x2019; In the first part we focus on cochlear implantation in children with prelingual deafness. We aim to evaluate whether early recognition of hearing impairment actually results in earlier cochlear implantation in children. Besides, we review the current literature to determine the benefit of bilateral implantation in children and investigate if a delay between the first and second implant affects postoperative outcomes. â&#x2C6;&#x2019; In the second part we investigate whether long term auditory deprivation in prelingually deaf adults leads to an impaired maturation of the auditory pathway possibly explaining postoperative performance. â&#x2C6;&#x2019; In the third part we assess the clinical feasibility of objective electrophysiological measures which can be used in the daily clinic and give a first insight whether a more complex cortical potential could be of clinical interest for the future. In Chapter 1.1 we evaluate if the introduction of the newborn hearing screening programs in the Netherlands and Germany actually has resulted in earlier treatment by means of cochlear implantation. Following early recognition of bilateral hearing impairment, the optimal treatment should be chosen for the individual patient. In the last decade there has been a long discussion if bilateral cochlear implantation is superior over unilateral implantation. In Chapter 1.2 we address this topic by reviewing the current literature on the benefit of bilateral implantation in prelingually deaf children on various listening skills, such as speech perception in quiet and noise and sound localization. In the following Chapter 1.3 we elaborate further on the effect of bilateral implantation by addressing the aspect of time between the first and second implantation. In this chapter both the prognostic factor of the inter-implant interval and the therapeutic question if simultaneous bilateral implantation is preferable over sequential implantation is reviewed. In the second part we investigate the effects of early-onset and long term hearing impairment in prelingually deaf and late-implanted adult CI users. Chapter 2.1 describes their postoperative performance and several prognostic factors which are independently related to speech perception after implantation. Furthermore, we aim to characterize those patients who eventually decide not to use their implants and become non-users. In the
21
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
following chapters we use electrophysiological measures to assess the activation patterns of the auditory brainstem (Chapter 2.2) and auditory cortex (Chapter 2.3). By comparing the response of these prelingual deaf subjects with postlingual deaf CI users, we can discover differences in activation possibly caused by an impaired maturation during the period of profound deafness. In the final part we discuss the clinical feasibility of recording postoperative CAPs, ABRs and CAEPs evoked by electrical stimuli (Chapter 3.1). For a valid objective clinical tool it is necessary that these measures can be easily and reliably recorded in almost all subjects. By recording the various measures and changing stimulus parameters we formulate an optimal clinical paradigm for using objective measures. In Chapter 3.2 we present the results of a pilot study on a more complex obligatory cortical response, the acoustic change complex (ACC). By altering the frequency modulated sweeps which elicit the ACC we gain more insight in the factors which contribute to the magnitude of this response. This knowledge can be used in future studies evaluating the potential role of the ACC as on objective measure for clinical evaluations.
22
General introduction
References Albrecht R, Suchodoletz W, Uwer R (2000). The development of auditory evoked dipole source activity from childhood to adulthood. Clinical Neurophysiology 111:2268-2276. Bishop DV, Anderson M, Reid C, Fox AM (2011). Auditory development between 7 and 11 years: an eventrelated potential (ERP) study. PloS ONE 6, e18993. Blume S (2009). The Artificial Ear: Cochlear Implants and the Culture of Deafness Rutgers University Press. Boons T, Brokx JP, Dhooge I, Frijns JH, Peeraer L, Vermeulen A, Wouters J, van Wieringen A (2012). Predictors of spoken language development following pediatric cochlear implantation. Ear and Hearing 33:617639. Burkard RF, Eggermont JJ, Don M (2007). Auditory evoked potentials: basic principles and clinical application Lippincott Williams & Wilkins. Caposecco A, Hickson L, Pedley K (2012). Cochlear implant outcomes in adults and adolescents with earlyonset hearing loss. Ear and Hearing 33:209-220. Conel JL (1939). The postnatal development of human cerebral cortex Harvard University Press, Cambridge, MA Djourno A, Eyries C, Vallancien B (1957). [Electric excitation of the cochlear nerve in man by induction at a distance with the aid of micro-coil included in the fixture]. Comptes rendus des seances de la Societe de biologie et de ses filiales 151:423-425. Dorman MF, Sharma A, Gilley P, Martin K, Roland P (2007). Central auditory development: evidence from CAEP measurements in children fit with cochlear implants. Journal of Communication Disorders 40:284-294. Eggermont JJ, Salamy A. (1988). Maturational time course for the ABR in preterm and full term infants. Hearing Research 33:35-47. Eggermont JJ, Brown DK, Ponton CW, Kimberley BP (1996). Comparison of distortion product otoacoustic emission (DPOAE) and auditory brain stem response (ABR) traveling wave delay measurements suggests frequency-specific synapse maturation. Ear and Hearing 17:386-394. Eisen MD (2003). Djourno, Eyries, and the first implanted electrical neural stimulator to restore hearing. Otology & Neurotology 24:500-506. Hof JR, Stokroos RJ, Wix E, Chenault M, Gelders E, Brokx J (2013). Auditory maturation in premature infants: a potential pitfall for early cochlear implantation. Laryngoscope 123:2013-2018. Holden LK, Finley CC, Firszt JB, Holden TA, Brenner C, Potts LG, Gotter BD, Vanderhoof, SS, Mispagel K, Heydebrand G, Skinner MW (2013). Factors affecting open-set word recognition in adults with cochlear implants. Ear and Hearing 34:342-360. Huttenlocher PR, Dabholkar AS (1997). Regional differences in synaptogenesis in human cerebral cortex. The Journal of Comparative Neurology 387:167-178. Inagaki M, Tomita Y, Takashima S, Ohtani K, Andoh G, Takeshita K (1987). Functional and morphometrical maturation of the brainstem auditory pathway. Brain and Development 9:597-601. Kandel E, Schwartz J, Jessell T (2000). Principles of Neural Science, Fourth Edition McGraw-Hill Companies, Incorporated. Klop WM, Briaire JJ, Stiggelbout AM, Frijns, JH (2007). Cochlear implant outcomes and quality of life in adults with prelingual deafness. Laryngoscope 117:1982-1987. Kral A (2013). Auditory critical periods: a review from system’s perspective. Neuroscience 247:117-133. Kral A, Eggermont JJ (2007). What’s to lose and what’s to learn: development under auditory deprivation, cochlear implants and limits of cortical plasticity. Brain Research Reviews 56:259-269. Kral A, O’Donoghue GM (2010). Profound deafness in childhood. The New England journal of Medicine 363:1438-1450.
23
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General introduction
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Kral A, Tillein J, Heid S, Hartmann R, Klinke R (2005). Postnatal cortical development in congenital auditory deprivation. Cerebral Cortex 15:552-562. LamorĂŠ J (2011). 4.5.1 Hersenstampotentialen, Nederlands leerboek audiologie. Nederlandse vereniging van Audiologie. Martin BA, Tremblay KL, Korczak P (2008). Speech evoked potentials: from the laboratory to the clinic. Ear and Hearing 29:285-313. Matschke RG, Stenzel C, Plath P, Zilles K (1994). Maturational aspects of the human auditory pathway: anatomical and electrophysiological findings. Journal for Oto-Rhino-Laryngology, Head and Neck Surgery 56:68-72. Meirelles RC, Neves-pinto RM (2008). Antonio Maria Valsalva - Biographical Profile of A Pioneer of Otology. Archives of Otorhinolaryngology 12:274-279. Moore JK, Guan YL (2001). Cytoarchitectural and axonal maturation in human auditory cortex. Journal of the Association for Research in Otolaryngology 2:297-311. Moore JK, Linthicum FH (2007). The human auditory system: a timeline of development. International Journal of Audiology 46:460-478. Moore JK, Perazzo LM, Braun A (1995). Time course of axonal myelination in the human brainstem auditory pathway. Hearing Research 87:21-31. Mudry A (2006). Otology in Medical Papyri in Ancient Egypt. International Advanced Otology 2:133-142. Niparko JK, Tobey EA, Thal DJ, Eisenberg LS, Wang NY, Quittner AL, Fink NE (2010). Spoken Language Development in Children Following Cochlear Implantation. Journal of the American Medical Association 303:1498-1506. Ponton CW, Eggermont JJ (2001). Of kittens and kids: altered cortical maturation following profound deafness and cochlear implant use. Audiology and Neurotology 6:363-380. Ponton CW, Eggermont JJ, Coupland SG, Winkelaar R (1992). Frequency-specific maturation of the eighth nerve and brain-stem auditory pathway: evidence from derived auditory brain-stem responses (ABRs). The Journal of the Acoustical Society of America 91:1576-1586. Ponton CW, Eggermont JJ, Coupland SG, Winkelaar R (1993). The relation between head size and auditory brain-stem response interpeak latency maturation. Journal of the Acoustical Society of America 94:2149-2158. Ponton CW, Eggermont JJ, Kwong B, Don M (2000). Maturation of human central auditory system activity: evidence from multi-channel evoked potentials. Clinical Neurophysiology 111:220-236. Psifidis A (2006). The Prehistory of Audiology and Otology. International Advanced Otology 2:41-46. Pujol J, Soriano-Mas C, Ortiz H, Sebastian-Galles N, Losilla JM, Deus J (2006). Myelination of language-related areas in the developing brain. Neurology 66, 339-43. Rotteveel J (1992). Development of brainstem, middle latency and cortical auditory evoked responses in the human. Development of Auditory and Vestibular Systems 2:321-356. Sharma A, Dorman MF, Spahr AJ (2002). A sensitive period for the development of the central auditory system in children with cochlear implants: implications for age of implantation. Ear and Hearing 23:532-539. Sharma A, Nash AA, Dorman M (2009). Cortical development, plasticity and re-organization in children with cochlear implants. Journal of Communication Disorders 42:272-279. Sharma A, Gilley PM, Dorman MF, Baldwin R (2007). Deprivation-induced cortical reorganization in children with cochlear implants. International Journal of Audiology 46:494-499. Starr A, Amlie RN, Martin WH, Sanders S (1977). Development of auditory function in newborn infants revealed by auditory brainstem potentials. Pediatrics 60:831-839. Straatman LV, Huinck WJ, Langereis MC, Snik AF, Mulder JJ (2014). Cochlear implantation in late-implanted prelingually deafened adults: changes in quality of life. Otology & Neurotology 35:253-259.
24
General introduction
Teoh SW, Pisoni DB, Miyamoto RT (2004). Cochlear implantation in adults with prelingual deafness. Part I. Clinical results. Laryngoscope 114:1536-1540. van Dijkhuizen JN, Beers M, Boermans PP, Briaire JJ, Frijns JH (2011). Speech intelligibility as a predictor of cochlear implant outcome in prelingually deafened adults. Ear and Hearing 32:445-458. Wilson BS, Dorman MF (2008). Cochlear implants: a remarkable past and a brilliant future. Hearing Research 242:3-21. Wunderlich JL, Cone-Wesson BK (2006a). Maturation of CAEP in infants and children: A review. Hearing Research 212:212-223. Wunderlich JL, Cone-Wesson BK, Shepherd R (2006b). Maturation of the cortical auditory evoked potential in infants and young children. Hearing Research 212:185-202. Yang WS, Moon IS, Kim HN, Lee WS, Lee SE, Choi JY (2011). Delayed cochlear implantation in adults with prelingual severe-to-profound hearing loss. Otology & Neurotology 32:223-228.
25
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Part I Prelingual deafness in pediatric cochlear implant users
Chapter 1.1 The influence of the newborn hearing screening on the age at cochlear implantation in children
M.J.W. Lammers, T.T.G. Janssen, W. Grolman, T. Lenarz, H. Versnel, G.A. van Zanten, V. Topsakal, A. Lesinski-Schiedat
Laryngoscope 2015; 125(4):985-90
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Objectives: To evaluate the influence of the introduction of newborn hearing screening programs on the age at cochlear implantation in children. Study Design: Retrospective, multicenter cohort study. Methods: All 1.299 pediatric cochlear implant users who received their implants before the age of 5 years between 1995 and 2011 in the Medical University Hannover, Germany and University Medical Center Utrecht, the Netherlands were enrolled in this study. Age at implantation and the number of children implanted within the first year of life was assessed for each center. Results: Age at cochlear implantation gradually declined over the years in both centers. The introduction of the screening resulted in significant decline in the age at implantation in the Netherlands; simultaneously, the number of children implanted within their first year of life increased significantly. Comparing 4-year epochs immediately before and after introduction of the screening, the mean age decreased from 2.4 to 1.2 years, and the percentage of early implanted children increased from 9% to 37%. In the German population, a similar effect of the introduction of the hearing screening program was absent. Conclusion: The introduction of the national newborn hearing screening programs has reduced the age at cochlear implantation in young children in the Netherlands but not in Germany. Correspondingly, it resulted in an increase in the number of children implanted early in life. The difference between the Dutch and German population might be due to differences in the follow-up and referral after the hearing screening.
30
The newborn hearing screening and CI
Introduction Over the last years the evidence indicating the importance of early cochlear implantation on the development of the auditory pathway and postoperative outcomes in children has rapidly increased (Boons et al. 2012; Houston et al. 2010; Lammers et al. 2014; Lesinski-Schiedat et al. 2004; Niparko et al. 2010; Sharma et al. 2002; Tajudeen et al. 2010). This has resulted for most children with early onset and severe sensorineural hearing impairment in receiving their cochlear implants (CI) within the first one or two years of life. Delays in the detection or recognition of severe hearing impairment could, however, introduce undesired retardation of speech development and verbal communication skills. With the introduction of the newborn hearing screening programs, it has become possible to detect hearing impairment in a very early stage (Korver et al. 2010), which in turn could result in early auditory rehabilitation. When hearing impairment is suspected with the hearing screening, patient and hospital delays should be minimized to reach optimal care for young children. Providing adequate information to the parents can minimize patient delays, and critical evaluations of the health care process around cochlear implantation are important to point out unnecessary hospital delays. An important step in this is fine-tuning the communication between audiology clinics and cochlear implant centers. The combination of early detection of hearing impairment with swift and adequate referral and an efficient rehabilitation program is therefore essential for the hearing outcomes with CIs. To date, however, it is unknown to what extent the introduction of the newborn hearing screening programs has influenced the age at which cochlear implantations are performed and little is known about hospital delays that could interfere with early cochlear implantation. With this study we aimed to evaluate the impact of the national introduction of the newborn hearing screening program on the pediatric populations of the cochlear implant clinics of the Medical University of Hannover (MHH), Germany and the University Medical Center Utrecht (UMC Utrecht, Utrecht, the Netherlands. Second, we critically evaluated hospital delays in the healthcare process of the cochlear implant team of the UMC Utrecht.
Materials and Methods Design and study population We performed a retrospective cohort study, including all children diagnosed with severe to profound hearing loss who were fitted with a CI before the age of 5 years between 1995 and 2011, at the MHH and the UMC Utrecht. The implementation of the newborn hearing screening program was completed in the Netherlands in July 2006, in Germany in January 2009. The Dutch hearing screening consists of an otoacoustic emission (OAE) test within the first week after birth. After a negative result, a second OAE measure is performed and in case 31
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
of a uni- or bilateral alarm an automated auditory brainstem response (AABR) is performed during a third visit. If hearing impairment is suspected after the AABR assessment, the child is referred to an audiology clinic. Careful registration takes place to ensure that all children who are referred are actually seen and followed by a specialized audiology clinic. In Germany, the hearing screening is performed within the first days after birth by using OAEs or AABRs or both. If a child fails these tests, the parents are advised to contact a qualified otolaryngologist for further investigation. Data extraction and analysis For each patient, we gathered information on the year of implantation, age at implantation, and the cause of deafness. Although we only included children receiving their CIs before the age of 5 years, a wide variation in age at onset of deafness might still be present within this cohort. Due to the retrospective nature of this study, exact details on the age at onset of deafness were not available for all subjects. Therefore, we have restricted our inclusion to children with bilateral deafness who received their implants before the age of 5 years to create a relatively homogenous study cohort. To further refine this cohort, we performed subgroup analyses by lowering the cutoff age of 5 years with 1-year intervals to assess whether refining the cohort in such a way would alter the results. Furthermore, we performed additional analyses by excluding children diagnosed with meningitis or cytomegalovirus (CMV) infection, because the onset of deafness in this subgroup could have been highly variable and later in life than in children with congenital deafness. For our analyses on the impact of the start of the national newborn hearing screening programs on the age at implantation, we used the year 2007 as cutoff value for the start of the program in the Netherlands and 2009 for Germany. We evaluated both the mean age at implantation in each year and the number of children implanted each year within the critical periods for cochlear implantation of <3.5 years of age, as proposed by Sharma et al. (Sharma et al. 2002) and <1 year as suggested by Lesinski-Schiedat et al. (Lesinski-Schiedat et al. 2004). Linear regression analyses were used to evaluate the decline in the age at implantation. Deviations from the regression model were analyzed using one-sample binomial tests. Mean and median differences in the age at implantation were analyzed using Student t tests or Mann-Whitney U tests. Differences in the number of children implanted in specific years were calculated with Ď&#x2021;² tests. Evaluation the CI health-care process In a consecutive series of 17 children who received a CI in the UMC Utrecht in 2011, we evaluated the time between referral to the cochlear implant team and surgery. After the first visit, radiological imaging is performed and the patients are seen by the anesthetic team. At the second visit, the results of the assessment are discussed with the patient and parents, and the surgery is scheduled. 32
The newborn hearing screening and CI
Results Between 1995 and 2011, a total of 1.299 children received their implants before the age of 5 years and could be included in our analyses (Table 1) (UMC Utrecht n = 139, MHH n = 1.160). Table I. Study Population Characteristics Total no. No. of bilateral Mean age at implantation, yr Etiology Congenital Meningitis/CMV Unknown/other
MHH 1.160 482 (42%) 2.3
UMCU 139 2.3
255 (22%) 111 (10%) 794 (68%)
89 (64%) 32 (23%) 18 (13%)
Age at implantation The age at implantation decreased significantly over the period 1995 to 2011 in both centers (MHH r = 0.30, P < 0.0001; UMC Utrecht r = 0.58, P < 0.0001). In the UMC Utrecht, the median age at implantation decreased in these periods from 3.4 to 0.9 years, and in the MHH the mean age decreased from 3.1 to 1.9 years (Fig. 1).
Fig. 1. Average age at cochlear implantation presented for all children treated in the Medical University Hannover (MHH) (dotted line) and University Medical Center Utrecht (UMC Utrecht) (solid line) in the period 1995 to 2011. The vertical dotted lines indicate the introduction of the national newborn hearing screening programs in the Netherlands (NL) in July 2006 and in Germany (GER) in January 2009. Error bars indicate the standard error of the mean. ** P < 0.0001
33
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
The number of children receiving their CIs before the age of 3.5 years increased in the MHH from 67% in 1995 to 1998 to 86% in the period 2008 to 2011 (Ď&#x2021;² = 25.7 (df 1), P < 0.001) and in the UMC Utrecht from 60% in 1995 to 1998 to 96% in 2008to 2011 (Ď&#x2021;2 = 8.6 (df 1), P = 0.003) (Fig. 2). Before 2004, only up to 15% of the children received their implants within their first year of life, whereas in the period 2008 to 2011 this number increased to 33% to 60% (Fig. 2). The number of children receiving bilateral CIs increased in both centers over the years. In 2010, around 80% of the children in both centers received bilateral Cis, of whom the majority were simultaneous.
Fig. 2. The percentage of children implanted under the age of 3.5 (light blue bars) and 1 year (dark blue bars) presented for both centers separately. MHH = Medical University of Hannover, UMCU = University Medical Center Utrecht.
Impact of newborn hearing screening In the period prior to the introduction of the newborn hearing screening there was a gradual decrease in the age at implantation in both centers (UMC Utrecht 1995-2006 r = 0.282, P = 0.004; MHH 1995-2008 r = 0.306, P < 0.0001). After the introduction of the national screening program in the Netherlands in July 2006, a significant decline in the age at implantation can be seen in the UMC Utrecht population. The mean age decreased from 2.4 years in 2003 to 2006 (n = 43) to 1.2 years in 2007 to 2010 (n = 30) (t71 = 4.9, P < 0.0001) (Fig. 1). Regression analysis revealed that after 2006, 90% of the children received their implants earlier than expected with the regression model based on the data before 2006 (Fig. 3) (one-sample binomial test, P < 0.001). The average age at implantation in the individual years 2007, 2008, 2010 and 2011 was significantly lower than expected with the regression model (t7 = -3.2, P = 0.014; t6 = -4.3, P = 0.005; t7 = -6.4, P < 0.001, t4 = -6.1, P = 0.004 respectively; 2009: t5 = -0.47, P = 0.66). The average age at implantation over the total period 2007 to 2011 was 0.6 years lower than expected with the regression model (t34 =-5.2, P < 0.001). In 2009, one clear outlier can be noted, who received the CI at the age of 4.3 years (Fig. 3). Detailed analysis
34
The newborn hearing screening and CI
revealed that this patient became deaf after the age of 2 and had progressive hearing loss, which resulted in deafness at the age of 4. If this outlier was excluded from the analyses, the average age at implantation in 2009 was significantly earlier than expected with the regression model (t4 = -16.6, P < 0.0001).
Fig. 3. Regression analysis based on all children implanted in the University Medical Center Utrecht (n =139) in the period 1995 to 2011. The regression line is based on the period prior to the introduction of the hearing screening in 2007 (y = -0.12*(Year-1995) + 3.53). After the introduction in 2007, 90% of the children received their implants earlier than expected with the regression model. Blue squares indicate the individual data for all children implanted prior to 2007; red dots denote the individual data for the children implanted after the introduction.
In the MHH, a marked decline after the national introduction of the newborn hearing screening program in 2009 in Germany could not be seen (2006-2008 vs 2009-2011: t418 = 0.372, P = 0.710). The regression model also demonstrated that after 2009, almost 50% of the children received their implants later than expected with the model (Fig. 4). Average age at implantation in 2010 and 2011 did not differ significantly from the expected age with the regression model (2010 P = 0.140, 2011 P = 0.068) and was even higher for the year 2009 (t70 = 2.6, P = 0.011) and when the average age over the period 2009 to 2011 was analyzed (t194 = 3.6, P < 0.001). The percentage of children who received their implants before the age of 1 year increased from 9% in 2003 to 2006 to 37% in the period 2007 to 2010 in the UMC Utrecht population (χ2 =8.1 (df 1), P = 0.004). In the MHH population, this percentage increased from 18% (2006-2008) to 24% (2009-2011) (χ2 =2.7 (df 1), P = 0.10). The percentage of children implanted before the age of 3.5 years did not change significantly after the introduction of the national newborn hearing screening programs in both centers (UMC Utrecht χ2 =3.8 (df 1), P = 0.051; MHH χ2 =0.4 (df 1), P = 0.52).
35
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig. 4. Regression analyses based on all children implanted in the Medical University of Hannover (n =1.160) in the period 1995 to 2011. The regression line is based on the period prior to the introduction of the hearing screening in 2009 (y =-0.09*(Year-1995) +3.05). After the introduction in 2009, 56% of the children received their implants earlier as expected with the regression model. Blue squares indicate the individual data for all children implanted prior to 2009; red dots denote the individual data for the children implanted after the introduction.
Fig. 5. Subgroup analyses including children with different cutoff ages at implantation in the Medical University of Hannover. The solid line denotes all included children implanted under the age of 5 (n = 1.160; r = 0.306, P <0.0001), the short dashed line children implanted under the age of 4 years (n = 1.023; r = 0.331, P <0.0001), the dotted line under the age of 3 years (n = 810; r = 0.310, P <0.0001), the long dashed line indicates children under the age of 2 years (n = 492; r = 0.233, P <0.0001) and the lowest solid line denotes the children who received their implant within their first year of age (n = 157; r = 0.08, P = 0.33).
36
The newborn hearing screening and CI
Subgroup analyses revealed that lowering the cutoff age with 1-year intervals did not reveal a significant decline in the age at implantation after the introduction of the hearing screening in the MHH population (Fig. 5). Excluding children with meningitis or CMV infections in both the main and subgroup analyses did not significantly affect the results either. Health care process The evaluation of the duration of the pre-operative assessment in 17 children who received their first CIs in 2011 in the UMC Utrecht revealed an average duration of 142 days between the first visit and their surgery (Fig. 6). The longest delays were found prior to the preoperative imaging and surgery, 53 and 62 days, respectively (Fig. 6).
Fig. 6. The consecutive steps in the healthcare process around cochlear implantation in the University Medical Center Utrecht and the average duration between each step for 17 patients who received a CI in 2011. Visit indicates the visit to the multidisciplinary cochlear implant team.
Discussion The results of this retrospective analysis of the pediatric cochlear implant programs of the MHH and the UMC Utrecht revealed that the age at (first) implantation has decreased gradually over the last years (Fig. 1). The introduction of the newborn hearing screening program in the Netherlands resulted in a significant further decline in age at implantation (Fig. 1) and an increase in the number of children implanted before the age of one year (Fig. 2).
37
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Although the literature suggests large benefits if CIs are provided within the first year of life (Houston and Miyamoto 2010; Lesinski-Schiedat et al. 2004; Niparko et al. 2010; Tajudeen et al. 2010), the vast majority of the early diagnosed children still did not receive their CIs within this period (Fig. 2). First of all, this can be caused by the difficulties encountered during the evaluation of the degree of hearing impairment in very young children. When hearing impairment is suspected at the hearing screening, it can be a long-winded process of evaluations and hearing aid trials before a definite diagnosis of deafness can be made. Furthermore, a patient or hospital delay or a combination of both could also introduce a delay between diagnosis and implantation. The evaluation of the health care process in the UMC Utrecht showed a considerable delay between the first visit and the actual surgery (Fig. 6). In particular, the waiting periods before the preoperative imaging and surgery contributed for more than half of the delay. In the MHH, hearing tests and preoperative imaging are scheduled on the same day, thereby reducing the time between the first visit and surgery. Nowadays, in the UMC Utrecht the waiting periods between the first visit and the final visit prior to surgery has been reduced from 78 to 14 days. By optimizing the health care process in such a way hospital delays can be reduced, the likelihood increases that children will receive their implants within their first year of life. In addition to a hospital delay, the holdup between diagnosis and surgery might also be influenced by a patient delay. Explaining the differences between the two populations In the Dutch population, a significant effect of the introduction of the national newborn hearing screening program was seen on the age at cochlear implantation, whereas a similar effect was absent in the German population (Fig. 1). This difference might be explained by differences in the populations of the two centers, and differences between the countries and centers with respect to referral and follow-up of children, after a negative result at the hearing screening test. Before we discuss aspects in countries and/or centers that could explain the different results, we point out that concerning cochlear implantation in young children with an early diagnosis of severe sensorineural hearing loss, the views of the two centers are identical. In both centers, the aim is to treat these children well within their first year of life, but to wait until they are at least six months old. Only in cases that require urgent cochlear implantation, such as meningitis, are children under the age of six months implanted. Comparison of the national hearing screening programs, however, shows that there are certain differences between the two countries that could explain the differences in the age at implantation. In the Netherlands, children are referred by the screening program to an audiology clinic after hearing impairment is suspected, and they are subject to a national tracking system. In Germany, the parents are advised to contact an otolaryngologist for a diagnostic hearing assessment. No national tracking system is organized by the
38
The newborn hearing screening and CI
insurance companies or the government. In only some parts of Germany, volunteers who are not financially compensated are organizing a tracking system. This self-referral system in Germany might lead to delays between the hearing screening and referral, which has to be initiated by the parents themselves. The consequence might even be that children are not seen by an audiologist or otolaryngologist after a negative hearing screening test at all, and that they present themselves at a later stage in their life when the consequences of hearing impairment become more obvious. Another argument for the unexpected course of age at implantation in Hannover is due to the fact that since 1996, children are intended to be implanted under the age of 2 years. Details from the national tracking system in the Netherlands showed that in 2011, 50% of the children with a negative hearing screening test at the third visit were seen in an audiology clinic within the desired standard of 24 days (Van der Ploeg et al. 2013). As a result, a definite diagnosis of unilateral or bilateral hearing impairment was made in 82% of the children within the desired 3 months (Van der Ploeg et al. 2013). Undesired delays in referral might lead to a delayed initiation of hearing aid fitting and even cochlear implantation. An inadequate referral and follow-up system might therefore cause unnecessary delays and reduce the likelihood that children receive their implants early in life. After hearing loss is confirmed, adequate hearing aid trials are initiated in both centers for at least several months. Around the age of six months, these children are again evaluated and additional hearing tests are performed. In the UMC Utrecht, the hearing aid trials are performed by the audiologists of the cochlear implant team and CIs are considered if aided hearing thresholds are worse than 60 dB and parents and educators report little or no benefit of the hearing aids. In the MHH at the end of the hearing aid trial, auditory brainstem responses are recorded. If thresholds are worse than 80 dB, the child is considered for cochlear implantation. In this respect, there are slight differences between the two centers, but these small differences do not lead to a delay between diagnosis and cochlear implantation and do not explain the different results. If cochlear implantation is recommended by the cochlear implant teams, parents do have the possibility to not have their child receive implants. The number of deaf children who were not implanted because of the parentsâ&#x20AC;&#x2122; choice is in both centers very low and estimated to be less than 1%. This factor is therefore not likely to severely influence the results and to contribute to the differences in age at implantation found between the countries. On the other hand, geographical differences and differences in the size of the study populations between the two centers, might explain the differences between the two centers. Because the two centers are not similar with respect to referral and geographical differences, the MHH population might be more heterogeneous than the UMC Utrecht population.
39
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Methodological considerations Due to the retrospective nature of this study, there are several potential limitations which should be considered. First, in our evaluation of the impact of the introduction of the newborn hearing screening program, no other external factors could be identified by the authors which could have influenced the results, such as changes in health care policy. After the introduction of the hearing screening, one would expect a certain delay before an obvious effect can be expected, but in the UMC Utrecht population a fast decline in the age at implantation was observed immediately after 2007. In addition to the increasing evidence showing the beneficial effect of early implantation (Boons et al. 2012; Houston and Miyamoto 2010; Lammers et al. 2014; Lesinski-Schiedat et al. 2004; Niparko et al. 2010; Sharma et al. 2002; Tajudeen et al. 2010) no changes in indication criteria or other health care policy changes were implemented in the UMC Utrecht prior to July 2006, except for the implementation of the hearing screening. It is therefore most likely that the decline in the age at implantation is indeed caused by the earlier detection due to the hearing screening. Second, incompleteness of data, differences in annotation of data in the databases and medical charts, and differences in preoperative analyses between the two centers have resulted in the fact that data on the cause of deafness are limited or absent, and that insufficient data were available in the medical charts to evaluate the age at onset of deafness, residual hearing, comorbidities, and the duration and effect of previous hearing aid trials. For our analyses, we therefore only used factors which were available for all children (i.e., year of implantation, age at implantation, and number of implantations). To create a relatively homogenous study population of children who became deaf early in life, we used a cutoff age at implantation of 5 years. The onset of deafness could, however, vary in this group of children therefore influencing the results. We therefore performed several subgroup analyses to create a more homogeneous subgroup of children. These subgroup analyses, however, did not significantly influence the results. Moreover, the evaluation of the effect of the introduction of the newborn hearing screening is a within-center comparison, and variations in the age at onset of deafness are likely to be present in the population both before and after the introduction of screening, thereby not significantly influencing this within-center comparison.
Conclusion The results of this retrospective study revealed that the national introduction of the newborn hearing screening programs led to significant decline in the age at cochlear implantation in young children with bilateral severe to profound hearing loss in the Dutch population. Consequently, it resulted in an increase in the number of children implanted early in life. The difference between the Dutch and German population might be attributed to differences
40
The newborn hearing screening and CI
in the referral and follow-up after the hearing screening. To further increase the number of children implanted early and to reduce undesired delays, special attention should be paid to optimize the healthcare process around cochlear implantation and assure that all children are adequately followed and referred after a negative result at the newborn hearing screening.
Acknowledgement The authors thank all members of the cochlear implant teams of the Medical University Hannover and the University Medical Center Utrecht for their support.
41
1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
References Boons T, Brokx JP, Dhooge I, Frijns JH, Peeraer L, Vermeulen A, Wouters J, van Wieringen A (2012). Predictors of spoken language development following pediatric cochlear implantation. Ear and Hearing 33:617639. Houston DM, Miyamoto RT (2010). Effects of early auditory experience on word learning and speech perception in deaf children with cochlear implants: implications for sensitive periods of language development. Otology & Neurotology 31:1248-1253. Korver AM, Konings S, Dekker FW, Beers M, Wever CC, Frijns JH, Oudesluys-Murphy AM; DECIBEL Collaborative Study Group (2010). Newborn hearing screening vs later hearing screening and developmental outcomes in children with permanent childhood hearing impairment. Journal of the American Medical Association 304:1701-1708. Lammers MJ, Venekamp RP, Grolman W, van der Heijden GJ (2014). Bilateral cochlear implantation in children and the impact of the inter-implant interval. Laryngoscope 124:993-999. Lesinski-Schiedat A, Illg A, Heermann R, Bertram B, Lenarz T (2004). Paediatric cochlear implantation in the first and in the second year of life: a comparative study. Cochlear Implants International 5:146-159. Niparko JK, Tobey EA, Thal DJ, Eisenberg LS, Wang NY, Quittner AL, Fink NE (2010). Spoken Language Development in Children Following Cochlear Implantation. Journal of the American Medical Association 303:1498-1506. Sharma A, Dorman MF, Spahr AJ (2002). A sensitive period for the development of the central auditory system in children with cochlear implants: implications for age of implantation. Ear and Hearing 23:532-539. Tajudeen BA, Waltzman SB, Jethanamest D, Svirsky MA (2010). Speech perception in congenitally deaf children receiving cochlear implants in the first year of life. Otology & Neurotology 31:1254-1260. Van der Ploeg CPB, Rijpstra A, Verker PH (2013). Monitoring van de neonatale gehoorscreening door de jeugdgezondheidszorg in 2011. TNO/CH/2013 R11837
42
Chapter 1.2 Bilateral cochlear implantation in children: a systematic review and best-evidence synthesis
M.J.W. Lammers, G.J.M.G. van der Heijden, V.E.C. Pourier, W. Grolman
Laryngoscope 2014; 124(7):1694-9
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Objective: To evaluate the effectiveness of bilateral cochlear implantation over unilateral implantation in children with sensorineural hearing loss. Data sources: Pubmed, Embase and Web of Science. Review methods: All studies comparing a bilateral cochlear implant group with a unilateral implant group were included. Results: Twenty-one studies compared a bilateral cochlear implant group with a unilateral group. No randomized trials were identified. Due to the clinical heterogeneity, statistical pooling was not feasible and a best-evidence synthesis was performed. The results of this best-evidence synthesis indicate the positive effect of the second implant for especially sound localization and possibly for preverbal communication and language development. There was insufficient evidence to make a valid comparison between bilateral implantation and a bimodal fitting. Conclusion: Although randomized trials are lacking, the results of our best-evidence synthesis indicate that the second cochlear implant might be especially useful in sound localization and possibly also in language development.
44
Bilateral cochlear implantation review
Introduction Over the last few years, the number of studies suggesting the effect of bilateral cochlear implantation in children has rapidly increased. This has resulted in a growing belief in the additional benefit of the second cochlear implant in the treatment of children with severe to profound sensorineural hearing loss, as recently expressed by Hodges and Balkany (Hodges and Balkany 2012). Therefore, in numerous countries, bilateral cochlear implantation has become the standard of practice for the treatment of children with bilateral severe to profound sensorineural hearing loss. But the second implant comes at a high cost and evidence for the cost-effectiveness and cost-utility of the second implant is still lacking (Lammers et al. 2011). As in other countries, health care commissioners in the Netherlands have for long maintained their opinion that there is insufficient evidence for the benefit of bilateral implantation to warrant its reimbursement. Two systematic reviews have evaluated the effectiveness of bilateral implantation (Bond et al. 2009; Sparreboom et al. 2010b). Both studies found that to date no randomized controlled trials have been performed in children. Moreover, they concluded that the level of evidence of the included studies was too low to draw robust conclusions (Bond et al. 2009; Sparreboom et al. 2010b). Both systematic reviews show that the majority of the included studies used within-subject comparisons rather than between-subject comparisons. Such studies lack control for changes due to the development of the child, regardless of the cochlear implantation. Since the publication of the most recent systematic review several studies have been published comparing bilateral cochlear implant groups with unilateral control groups. With this article, we aim to provide a thorough update and systematic review of the current literature, examining the additional benefit of the second cochlear implant in children with sensorineural hearing loss.
Materials and methods Search strategy and study selection We conducted a systematic search in the PubMed, Embase and Web of Science databases from inception up to July 25, 2013, by using the search terms ‘cochlear implant’, ‘children’ and their synonyms in title and abstract fields (Appendix 1). Two review authors (M.J.W.L. and V.E.C.P.) independently screened titles and abstracts of all retrieved publications using predefined selection criteria (Fig. 1). Subsequently, the full text of eligible studies was screened for a more detailed selection (Fig. 1). All studies in which a comparison was made between bilateral and unilateral cochlear implantation, with or without a contralateral
45
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
hearing aid, were included. Cross reference search for all selected studies and reviews was used to increase the yield of relevant studies (Fig. 1). Disagreement between the authors was resolved by discussion.
Fig. 1. Flow-diagram of search strategy (Search date: July 25, 2013)
Risk of bias assessment The methodological quality of all included studies was independently assessed by two authors (M.J.W.L. and V.E.C.P.) according to the risk of bias procedure as used by the Cochrane Collaboration (Higgins et al. 2011. Available from www.cochrane-handbook.org.), notably randomized and concealed allocation, blinding of study outcomes and completeness of data. In addition, we assessed the reported information on external factors, notably standardization of surgical factors, such as inter-implant interval and standardization of outcome measures. Each aspect was judged as good, moderate, poor or unknown. One point was awarded if the criterion was fulfilled. We used a predefined cut-off value of â&#x2030;Ľ 4 points (â&#x2030;Ľ two-thirds of the total score) in order to separate high-quality studies (low risk of bias) from studies with a moderate to high risk of bias. Disagreements were resolved through discussion.
46
Bilateral cochlear implantation review
Data extraction and analysis Information was gathered for each study on design, study population, number of included patients, number of unilateral and bilateral implanted patients, number of simultaneous bilateral implanted patients, and the mean ages of implantation. All reported outcome data for either or both bilateral and unilateral implant groups were extracted. The data analysis was performed with the meta-analysis software Comprehensive Meta-Analysis (Borenstein M, Hedges L, Higgins J, Rothstein H. Comprehensive Meta-analysis Version 2, Biostat, Englewood, NJ, 2005). Outcome measures are reported as mean differences with their 95% confidence intervals as presented in the original articles. If results are only presented in figures or graphs, data was extracted from these whenever possible. To maximize accuracy of this form of data extraction, figures were enlarged by approximately 200% and rounded off to the nearest number. These extracted data are denoted with ‡. To compare the results of the various studies it was necessary to transform the data to standardized mean differences with their 95% confidence intervals. When studies are clinically homogenous, that is, they are similar with respect to the degree of hearing loss of the patients included, the duration of deafness and time at implantations, the treatment comparisons, and the type of outcome measures, we consider statistical pooling. We excluded studies with a moderate to high risk of bias (risk of bias score < 4) from such pooling. If statistical pooling is not feasible and the studies are clinically heterogeneous, we consider a best-evidence synthesis. A best-evidence synthesis is a systematic qualitative summarization of available evidence, which helps to reduce the chance of conflicting results and conclusions (Slavin 1995). We stratify the best-evidence synthesis for studies that are similar with respect to the degree of hearing loss of the patients included and the treatment comparisons. For the best-evidence synthesis, we combine the outcome of the quality assessment with the study results. In order to reduce the risk of obtaining biased results, we distinguish different levels of evidence, according to the following decision rules: • Strong evidence, provided by generally consistent results in at least three treatment comparisons in multiple studies with a low risk of bias (≥ 4 points) • Weak evidence, provided by generally consistent results in at least three treatment comparisons in studies with a moderate to high risk of bias (< 4 points). • Insufficient/inconsistent evidence, when less than three treatment comparisons are available, or results were inconsistent in more than three comparisons. • Consistent evidence is defined as ≥ 75% of comparisons with similar direction of effect.
47
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Results Search results Our search identified 4.821 unique articles (Fig. 1). After screening titles and abstracts, we excluded 4.719 non-relevant articles. The 102 remaining articles were retrieved in full text for formal review. Eventually, 21 articles (Litovsky et al. 2006b; Litovsky et al. 2006a; Schafer and Thibodeau 2006; Beijen et al. 2007; Mok et al. 2007; Grieco-Calub et al. 2008; GriecoCalub et al. 2009; Nittrouer and Chapman 2009; Eustaquio et al. 2010; Lovett et al. 2010; Mok et al. 2010; Sparreboom et al. 2010a; Tait et al. 2010; Baudonck et al. 2011; Murphy et al. 2011; Boons et al. 2012; Grieco-Calub and Litovsky 2012; Nittrouer et al. 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a; Nittrouer et al. 2013) were eligible for further analysis because they compared a bilateral cochlear implant group with a separate unilateral implant group. Risk of bias assessment The risk of bias assessment is presented in table I. There were no randomized controlled trials identified. All studies were qualified as having a moderate to high risk of bias (< 4 points). Six studies (Litovsky et al. 2006b; Litovsky et al. 2006a; Schafer and Thibodeau 2006; Beijen et al. 2007; Mok et al. 2007; Grieco-Calub et al. 2008) (29%) were previously included in the systematic review by Sparreboom et al (Sparreboom et al. 2010b). Only one study provided a clear description of their study participant selection procedure and carefully matched their control group on various important confounding factors (Boons et al. 2012). Thirteen of the 21 studies (Litovsky et al. 2006b; Litovsky et al. 2006a; Mok et al. 2007; Grieco-Calub et al. 2008; Grieco-Calub et al. 2009; Nittrouer and Chapman 2009; Mok et al. 2010; Sparreboom et al. 2010a; Grieco-Calub and Litovsky 2012; Nittrouer et al. 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a; Nittrouer et al. 2013) (62%) reported the results on different outcome measures of the same or partially the same group of children, resulting in 14 unique study populations. Study characteristics The median age at which the children received their first and second implant was 1.8 and 4.9 years respectively (Table II). Only 71 out of 375 children received their implants simultaneously, whereas in the other children the age at which their second implant was provided, varied considerably from 0.6 to 15.8 years (Table II).
48
Bilateral cochlear implantation review
Table I. Risk of bias assessment of the included studies Author, year RA Nittrouer et al, 2013
CA
BL -
ST Det ST Out ATT
Boons et al, 2012 Grieco-Calub et al, 2012 Nittrouer et al, 2012
-
Score Comment 2 Part Identical study population Nittrouer 2009 3 3
-
2.5
Sparreboom et al, 2012a Sparreboom et al, 2012b Baudonck et al, 2011 Eustaquio et al, 2011 Murphy et al, 2011 Lovett et al, 2010 Mok et al, 2010
-
1.5
-
1.5
-
?
2.5 2 2.5 2 2.5
Sparreboom et al, 2010a Tait et al, 2010 Grieco-Calub et al, 2009 Nittrouer et al, 2009 Grieco-Calub et al, 2008 Beijen et al, 2007 Mok et al, 2007 Litovsky et al, 2006a Litovsky et al, 2006b
-
2.5
-
3 3
-
2.5 3
-
3 2.5 2.5 2.5
Schafer et al, 2006
-
2
Identical study population Nittrouer 2009 Identical study population Sparreboom 2010a Identical study population Sparreboom 2010a
1.2
Identical study population Mok 2007
Identical study population Grieco-Calub 2012 Identical study population Grieco-Calub 2012
Part Identical study population Litovsky 2006a
Random sequence generation (RA); :Yes, randomization used and method adequately described; : No randomization used; ?: unclear/not adequately described. Concealment of allocation (CA); : Yes; : No; ?: unclear/not adequately described. Blinding (BL); : blinding of participant and/or outcome assessor; : no blinding of participant and/or outcome assessor; ?: unclear/not adequately described; -: not applicable Standardization of determinant (ST Det) (standardization of surgical factors, notably inter-implant interval); : small or no inter-implant interval range (<2 years) within BICI group; , varying interimplant interval range (>2 years) within bici group; , different CI brands used within patients or unknown; Standardization of outcome (ST Out) (i.e. are standardized tests used?); : Yes; : No; ?: unclear/not adequately described. Attrition / Completeness of data (ATT); : ≤ 10% loss to follow-up; : > 10% loss to follow-up; ?: unclear/not adequately described. Score; For each of the 6 criteria one point is awarded if the criterion was fulfilled (). A half point was awarded if the criterion was partially fulfilled (), and zero points if the criterion was not fulfilled (). A predefined cut-off value of ≥ 4 points (≥ 2/3 of the total score) was used in order to separate high quality studies (low risk of bias) from studies with a moderate to high risk of bias. 49
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table II. Study characteristics
Author, year Nittrouer et al. 2013 Boons et al, 2012 Grieco-Calub et al, 2012 Nittrouer et al, 2012 Sparreboom et al, 2012b Sparreboom et al, 2012a Baudonck et al, 2011 Eustaquio et al, 2011 Murphy et al, 2011 Lovett et al, 2010 Mok et al, 2010 Sparreboom et al, 2010a Tait et al, 2010 Grieco-Calub et al, 2009 Nittrouer et al, 2009 Grieco-Calub et al, 2008 Beijen et al, 2007 Mok et al, 2007 Litovsky et al, 2006a Litovsky et al, 2006b Schafer et al, 2006
N BICI 28
N UCI 13
 N CIHA 6
N sim 1st CI BICI group UCI 2 ?
Mean age at implantation, years 1st CI 1st CI 1st CI 2nd CI 2nd CI Range BICI Range BICI Range ? ? ? 3.8 2-5.7
25 27
25 12
-
8 6
1.7 1.1
0.8-2.2 0.6-1.3
1.5 1.2
0.6-2.3 0.7-2.4
2.7 1.7
0.8-4.9 0.8-2.4
18
8
-
4
2.0
0.7-5.5
1.7
1.0-4.5
3.0
1.2-5.5
30
30
-
0
5.2
2.5-8.1
1.8
0.9-2.7
5.3
2.4-8.5
30
9
-
0
1.6
0.8-2.0
1.8
0.9-2.7
5.3
2.4-8.5
13
14
-
0
1.7
?
1.2
?
4.5
?
12
26
-
?
7.4
?
2.9
?
?
?
6 30 4 29
12 20 9
9 -
3 12 0 0
3.3 3.3 6.8 1.6
1.8-5 1.1-11.6 3.9-11 0.8-2.0
4.7 3.1 1.9 1.8
1.7-9.25 1.2-12 1.5-2.6 1.1-2.7
6.8 5.6 8.9 5.3
3.4-10.8 1.3-16.3 5.8-11.3 2.8-8.5
26 17
27 9
-
18 6
1 1.2
0.4-1.8 0.8-1.3
1.1 1.3
0.6-1.8 0.8-2.4
1.2 1.7
0.6-2.1 0.9-2.4
26
15
17
7
1.4
0.8-2.1
1.4
0.8-1.9
2.4
0.9-3.3
10
8
-
1
1.3
0.9-2.2
1.2
0.8-2.4
1.8
1-2.4
5 4 10
5 -
9 10
4 0 0
1.4 6.8 4.9
0.7-1.9 3.9-11 1.4-8.5
1.6 1.9 3.4
1.4-1.9 1.5-2.6 1.8-6
1.7 8.9 6.5
1.4-2 5.8-11.3 2.8-12.3
13
-
6
0
5.5
3.5-8.5
4.6
1.5-13
7.9
3-15.8
12
-
10
0
3.1
1.6-5.8
2.5
1-5.8
6
2-10.1
Study characteristics displaying the number of subjects in each group and the mean ages and age ranges of implantations. N sim indicates the number of children undergoing simultaneous implantation included in the studies. Mean age and age ranges for the study by Eustaquio et al. 2010 could not be extracted and are therefore excluded from the analysis. Age ranges for the study by Baudonck et al. 2011 could not be extracted and are therefore excluded from the analysis. Mean age at second implantation in the study by Nittrouer et al. 2009 is a weighted mean, based on the mean ages of second implantation of the simultaneous and sequential groups. Age ranges in the study by Nittrouer et al. 2009 are estimated by adding and subtracting one standard deviation of the mean. BICI = bilateral cochlear implantation, CI = cochlear implant, CIHA = bimodal (CI+HA), sim = simultaneous, UCI = unilateral cochlear implantation. ? = not reported.
50
Bilateral cochlear implantation review
Data analysis Due to limited number, the clinical heterogeneity between the studies, that is differing outcomes and comparisons with unilateral and bimodal groups, and the lack of studies classified as having a low risk of bias (â&#x2030;Ľ 4 points), statistical pooling was not feasible. Therefore, we performed a best-evidence synthesis (Table III). The overall results of the best-evidence synthesis show that there is weak or insufficient evidence in favour of bilateral implantation, and there is insufficient evidence to make a valid comparison between bilateral implantation and bimodal fitting (CI+HA). In the following section, results for the various outcome measures are presented qualitatively. Details and results for all outcome measures for each study are summarized in Table IV. Preverbal communication The multicenter study by Tait et al. reported significant benefit of the second implant on several domains of the preverbal communication, as assessed with the Tait video analysis (Tait et al. 2010). The largest improvements were noticed on the non-looking vocal turn and vocal turn domains, indicating that bilaterally implanted children were more responsive to vocal cues in the absence of visual reinforcement. Language development The study by Boons et al. compared a group of 25 bilaterally implanted children with a carefully matched and clearly described control group of 25 unilaterally implanted children (Boons et al. 2012). Their results demonstrated a large benefit of bilateral implantation on spoken language comprehension and expression. Nittrouer et al. compared the basic language skills between bilaterally implanted children, children with both a CI and hearing aid (CI+HA) and unilaterally implanted children (Nittrouer and Chapman 2009; Nittrouer et al. 2012). The results indicated no differences between the three groups when oral language scores were assessed with the Preschool Language Scales-4 (PLS-4) and the Expressive One-Word Picture Vocabulary Test (EOWPVT) (Nittrouer and Chapman 2009; Nittrouer et al. 2012). When other aspects were tested such as literacy, phonological awareness, mean length of utterances, or the number of pronouns, no differences were seen between both groups (Nittrouer and Chapman 2009; Nittrouer et al. 2012).
51
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
-
Speech perception in quiet BICI vs UCI BICI vs CI+HA
-
Sound localization BICI vs UCI BICI vs CI+HA -
-
Preverbal communication BICI vs UCI
Speech perception in noise BICI vs UCI S0,N0 S0,NCI1 S0,NCI2 BICI vs CI+HA
-
Overall BICI vs UCI BICI vs CI+HA
Number of comparisons
Number of comparisons 3 1
6 3 2 2 13
12 2
5
49 20
Validity score < 4
4
2
2
12 1
4
23 5
Favouring BICI
Validity score ≥ 4
3 1
2 7
4 3
1
1
26 13
No difference
52
Outcome measure
2
2
Favouring UCI/CI+HA Weak evidence: no difference Insufficient evidence
Inconsistent evidence Weak evidence: no difference Insufficient evidence Insufficient evidence Inconsistent evidence
Weak evidence in favour of BICI Insufficient evidence
Weak evidence in favour of BICI
Conclusion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table III. Validity score best evidence synthesis
Chapter 1.2
-
-
Language expression and comprehension BICI vs UCI
Quality of life Generic questionnaires BICI vs UCI
-
Speech production¥ BICI vs UCI 5
4
6
3
7 4
2
3 3
1
6
3
7 4
Insufficient evidence
Weak evidence in favour of BICI
Weak evidence: no difference
Weak evidence in favour of BICI
Weak evidence: no difference Weak evidence: no difference
Best-evidence synthesis presenting all comparisons made in the 19 studies presenting functional outcomes. Studies describing other outcome measures were excluded from the analysis (Eustaquio et al. 2010, Sparreboom et al. 2012b). The rows indicated with BICI vs UCI present comparisons between bilateral and unilateral cochlear implantation. The rows indicated with BICI vs CI+HA present comparisons between bilateral cochlear implantation and bimodal fitting (CI+HA). The validity scores were based on the risk of bias assessment. Numbers denote the number of comparisons made in the various studies. Results are presented separately for studies with either a high (≥4 points) or a low (<4 points) validity score. Subdomains of specific questionnaires or analysis in the studies marked with ¥, have been combined into one comparison. The study by Grieco-Calub et al. 2008 has been excluded in this analysis, since same data is reported in Grieco-Calub et al. 2012. Since there were no studies with a validity score of ≥4 points, we have omitted the columns indicating the direction of effect. Minus signs indicate that no comparisons have been performed. BICI = bilateral cochlear implantation, UCI = unilateral cochlear implantation.
-
Disease specific questionnaires BICI vs UCI ¥
-
Language development Basic language measures BICI vs UCI BICI vs CI+HA
Bilateral cochlear implantation review
1.2
53
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table IV. Summary of outcome measures Author, year
Outcome measure
Nittrouer et al, Speech in quiet 2013 SRM noise 2nd CI Boons et al, Language development: 2012 Reynell Schlichting Word Schlichting Sentence Grieco-Calub Sound localization: et al, 2012 MAA fixed intensity MAA roved intensity Left-right ±50° Nittrouer et al, Emergent literacy, Word reading, 2012 Comprehension, Phonological awareness, Syllable and initial counting, Final consonant
Sparreboom et al, 2012b Sparreboom et al, 2012a
Baudonck et al, 2011
Eustaquio et al, 2011
54
Results No sign difference BICI vs. UCI No sign difference BICI vs. UCI Sign benefit BICI over UCI: MD: 9.4 (95% CI: 0.3-18.6) MD: 15.7 (95%CI: 5.9-25.4) MD: 9.7 (95% CI:1.5-17.9) Sign benefit BICI over UCI: MD: 39° (95% CI: 27°-52°) MD: 34° (95% CI: 20°-48°) t=3.9, p<0.001 No sign difference BICI vs. UCI
Oral language: Auditory comprehension (PLS-4), Expressive vocabulary (EOWPVT), Narrative score
No sign difference BICI vs. UCI
Device use
No sign difference BICI vs. UCI
Generic QoL (VAS, HUI3, PedsQL)
No sign difference BICI vs. UCI
GCBI, NCIQ Disease specific QoL (SSQ)
No sign difference BICI vs. UCI Sign benefit BICI over UCI for spatial domain of SSQ: U=37.0 (p<0.01) Perceptual voice: GRBAS: Sign difference in favor of BICI for G (p=0.005), S (p=0.004) domains Instability: sign difference in favor of BICI (p=0.01)
Perceptual voice (GRBAS and instability)
Speech intelligibility (SIR) Nasality and resonance Articulation (phonetic and phonological process analysis)
No sign difference BICI vs. UCI No sign difference BICI vs. UCI Articulation: sign more distortions in UCI group as compared to BICI group (p=0.003) Phonological process analysis: no sign differences
Balance portion of the BOT2
No sign difference BICI vs. UCI
Bilateral cochlear implantation review
Table IV. Summary of outcome measures Author, year
Outcome measure
Murphy et al, Speech perception in noise 2011 Sound localization: Left-right ± 120° Left-right ± 60° Left-right ± 30° Lovett et al, Sound localization: 2010 Left-right ± 60° Left-right ± 30° Localization test Movement tracking SRM noise CI/1st CI SRM noise 2nd CI Generic QoL (VAS, HUI3) Disease specific QoL (SSQ)
Mok et al, 2010
Binaural advantage Head Shadow Effect (HSE)
Sparreboom et al, 2010a
Speech perception in quiet (ATT) Speech perception in noise (ATT)
Tait et al, 2010 Tait Video Analysis: Vocal turns Non-looking vocal turns Gestural turns Gestural autonomy Vocal autonomy Grieco-Calub Word recognition in quiet et al, 2009 Word recognition with competing talkers (+10 dB SNR)
Results No sign difference BICI vs. UCI Sign benefit of BICI over UCI: MD 27%‡ (p<0.001) MD 24%‡ (p<0.001) MD 25%‡ (p<0.001) Sign benefit of BICI over UCI:
1.2
Median difference: 35, p<0.01 Median difference: 30, p<0.01 Median difference: 29, p<0.01 Median difference: 30, p<0.01 Median difference: +5.23 dB, p<0.01 Median difference: -1.99 dB, p=0.16 No sign difference BICI vs. UCI Sign benefit of BICI over UCI: spatial domain: med. diff: 2.62, p<0.01, speech domain: med. diff: 1.65, p=0.04 Sign benefit CIHA over BICI: MD: 3.7% (95% CI 1.0-6.3%) BICI subjects showed a sign HSE. Bimodal subjects did not show a HSE 12 & 24 months: No sign difference BICI vs. UCI 12 months: No sign difference BICI vs. UCI 24 months: sign lower SNRs in BICI group as compared to UCI group: MD: 6.3 dB‡, p<0.01. Sign benefit of BICI over UCI: Median difference: 30 (p<0.0001) Median difference: 50 (p<0.0001) Median difference: -18 (p<0.0001) Median difference: -11.5 (p=0.0001) No sign difference BICI vs. UCI No sign difference BICI vs. UCI No sign difference BICI vs. UCI
55
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table IV. Summary of outcome measures Author, year Outcome measure Nittrouer et al, Auditory comprehension (PLS-4), 2009 Expressive vocabulary (EOWPVT) Mean length of utterance (MLU) Number of pronouns Grieco-Calub MAA fixed intensity et al, 2008 Beijen et al, Sound localization: 2007 Left-right ±90° Left-right ±30° Generic QoL (PedsQL) Disease specific QoL (SSQ) Mok et al, 2007
Spatial unmasking
Binaural advantage
Litovsky et al, SRTs in: quiet, S0N0, S0NCI1, S0NCI2 2006a SRM Bilateral benefit MAA Litovsky et al, MAA 2006b Schafer et al, Speech perception in noise 2006
Results No sign difference BICI vs. UCI / CIHA No sign difference BICI vs. UCI / CIHA No sign difference BICI vs. UCI / CIHA Sign benefit BICI over UCI: MD: 27° (95% CI: 5.5°-49°) Sign benefit BICI over UCI: MD: 59% (p<0.01) MD: 52% (p<0.01) No sign difference BICI vs. UCI Sign benefit BICI over UCI for spatial domain of SSQ, MD: 2.3 (p=0.03) Noise to side HA/2nd CI: no sign differences Noise to side CI/1st CI: Sign benefit of BICI over CIHA: MD: 0.8 dB (p=0.004) Noise at front: sign benefit of CIHA over BICI, MD: 1dB (p=0.004). Noise to side CI/1st CI: sign difference BICI over CIHA: MD: 1.6dB (p=0.002). Noise side HA/2nd CI: no sign differences No sign difference BICI vs. CIHA No sign difference BICI vs. CIHA Bilateral benefit: sign larger for BICI group than for CIHA group No sign difference BICI vs. CIHA Sign benefit BICI over CIHA: MD: 28°‡ No sign difference BICI vs. CIHA
Summary of outcome measures presented for all individual studies. ATT = Automated Toy Test, BICI = bilateral cochlear implantation, BOT2 = Bruininks-Oseretsky Test 2nd edition, CIHA = bimodal (CI+HA), EOWPVT = Expressive One-Word Picture Vocabulary Test, GCBI = Glasgow Children’s Benefit Inventory, GRBAS = Grade, Hoarseness, Roughness, Breathiness, Asthenicity, Strain, HA = hearing aid, HSE = head shadow effect, HUI3 = Health Utilities Index Mark 3, MAA = Minimum audible angle, MD = mean difference, MLU = mean length of utterance, NCIQ = Nijmegen Cochlear Implant Questionnaire, PedsQl = Pediatric Quality of Life Inventory, PLS-4 = Preschool Language Scales-4, QoL = quality of life, SNR = signal-to-noise ratio, SIR = speech intelligibility rating, SRM = spatial release from masking, SRT = speech reception in noise threshold, SSQ = Speech, Spatial, and Qualities of Hearing Scale, TVA = Tait video analysis, UCI = unilateral cochlear implantation, VAS = visual analogue scale.
56
Bilateral cochlear implantation review
Sound localization Five studies have evaluated the benefit of bilateral implantation over unilateral implantation with several age-appropriate localization measures (Fig. 2) (Beijen et al. 2007; Grieco-Calub et al. 2008; Lovett et al. 2010; Murphy et al. 2011; Grieco-Calub and Litovsky 2012). The best-evidence synthesis demonstrates that there is consistent evidence indicating the benefit of bilateral implantation for sound localization (Table III). The results on the left-right discrimination tasks performed by the various study groups indicated superior performance when children are bilaterally implanted (Beijen et al. 2007; Lovett et al. 2010). These findings are supported by other sound localization tests used by Lovett et al. (Lovett et al. 2010) and Grieco-Calub et al. (Grieco-Calub et al. 2008; Grieco-Calub and Litovsky 2012). Results for the more challenging minimal audible angle tests revealed that children in the unilateral group did not perform better than chance, whereas 63% of the children in the bilateral group were able to obtain localization scores better than chance (Grieco-Calub and Litovsky 2012). In two studies, Litovsky et al. compared a bilateral group with a group with bimodal stimulation (CI+HA) (Litovsky et al. 2006b; Litovsky et al. 2006a). Their studies demonstrated conflicting results when these groups were compared in terms of sound localization assessed with the minimal audible angle. In one study with equal sample sizes in both groups, no differences were found (Litovsky et al. 2006a), whereas in the other study (Litovsky et al. 2006b) in which the bilateral group was more than twice as large as the bimodal control group (13 vs. 6 subjects) the bilateral implant users performed better than the bimodal users.
Fig. 2. Standardized mean differences for the various localization tests presented in the five studies comparing a bilateral group with unilateral controls. The study by Grieco-Calub et al. 2008 (GriecoCalub et al. 2008) has been excluded in this analysis, since identical data is reported in Grieco-Calub et al. 2012 (Grieco-Calub and Litovsky 2012). BICI = bilateral cochlear implantation, CI = confidence interval, UCI = unilateral cochlear implantation.
57
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Speech perception in quiet Four studies investigated the benefit of the second implant in terms of speech perception in quiet (Litovsky et al. 2006a; Grieco-Calub et al. 2009; Sparreboom et al. 2010a; Nittrouer et al. 2013). One year after activation of the second implant, no differences could be demonstrated between both groups in all four studies. Two studies followed their children for two years or more and showed that, even after a longer period of bilateral use, speech perception in quiet was not better than with only one cochlear implant (Sparreboom et al. 2010a; Nittrouer et al. 2013). Speech perception in noise Five studies compared speech perception in noise in a bilateral group with unilateral controls (Grieco-Calub et al. 2009; Lovett et al. 2010; Sparreboom et al. 2010a; Murphy et al. 2011; Nittrouer et al. 2013). Three studies did not find any significant differences between the bilateral and unilateral groups when noise was presented from the front (Grieco-Calub et al. 2009; Murphy et al. 2011; Nittrouer et al. 2013). The two other studies, however, did find a major benefit of the second implant when the noise was presented on the side of the first or only CI (Table IV) (Lovett et al. 2010; Sparreboom et al. 2010a). When the noise was presented on the side of the second CI or contralateral ear no significant difference was seen between both groups (Lovett et al. 2010; Nittrouer et al. 2013). Four studies evaluated differences between bilateral implantation and bimodal fitting (Litovsky et al. 2006b; Litovsky et al. 2006a; Schafer and Thibodeau 2006; Mok et al. 2010). Three studies did not find any differences between the bilateral and the bimodal group when noise presented from either the front or the side (Litovsky et al. 2006b; Litovsky et al. 2006a; Schafer and Thibodeau 2006). The results presented by Mok et al. showed an advantage of the second CI when the noise was presented from the side of the first or only CI (Mok et al. 2010). When noise was presented from the front or the other side, no difference or even an advantage of the hearing aid over the second CI was seen (Mok et al. 2010). Speech production One study assessed the quality of the children’s’ speech production (Baudonck et al. 2011). Speech intelligibility as measured with the Speech Intelligibility Rating (SIR) did not differ between the two groups (Baudonck et al. 2011). With respect to the voice quality, as measured with the Grade, Roughness, Breathiness, Asthenia, Strain scale, the perceptual evaluation of the unilateral group was scored as being more hoarse, more fluctuating (instability) and more strained compared to the bilateral group. In terms of subjects’ articulation, it was found that the bilaterally implanted children showed less distortions than the children with only one implant (Baudonck et al. 2011).
58
Bilateral cochlear implantation review
Generic and disease specific quality of life Generic quality of life was assessed in three studies by using different proxy questionnaires (Beijen et al. 2007; Lovett et al. 2010; Sparreboom et al. 2012a). In none of these studies did the use of a second implant result in an increase in generic quality of life (Fig. 3). The results of the hearing-specific Speech, Spatial and Qualities (SSQ) questionnaire demonstrated in all three studies a higher score on the spatial domain for the bilateral groups (Beijen et al. 2007; Lovett et al. 2010; Sparreboom et al. 2012a). The study by Lovett et al. also found a higher score on the speech domain of the SSQ (Lovett et al. 2010). No differences were found for the qualities domain in all three studies (Beijen et al. 2007; Lovett et al. 2010; Sparreboom et al. 2012a).
Fig. 3. Forest plot for generic health related quality of life. BICI = bilateral cochlear implantation, CI = confidence interval, GCBI = Glasgow Childrenâ&#x20AC;&#x2122;s Benefit Inventory, HUI3 = Health Utilities Index Mark 3, NCIQ = Nijmegen Cochlear Implant Questionnaire, PedsQl = Pediatric Quality of Life Inventory, SSQ = Speech, Spatial, and Qualities of Hearing Scale, UCI = unilateral cochlear implantation, VAS = visual analogue scale.
Other outcome measures Sparreboom et al. examined the device use of the implants after bilateral and unilateral implantation (Sparreboom et al. 2012b). Although the bilaterally implanted children used their second device less often than their first, comparisons between the unilateral and bilateral groups did not reveal a difference after 12 and 24 months of CI use (Sparreboom et al. 2012b). Eustaquio et al. assessed differences between unilaterally and bilaterally implanted children in terms of balance disorders. In their study, no differences between the two groups were found (Eustaquio et al. 2010).
59
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Discussion Our review shows that there currently remains no strong evidence to support the additional benefit of the second cochlear implant in children, whereas randomized trials with a low risk of bias have not been performed. The best-evidence synthesis demonstrates that the available cohort studies provide weak but consistent evidence for the effectiveness of bilateral cochlear implantation in terms of sound localization (Fig. 2). Two studies indicated a benefit of the second implant in terms of preverbal communication and language development (Tait et al. 2010; Boons et al. 2012). Because these results have not been reproduced, firm conclusions could not be drawn. In terms of speech perception in quiet and noise, there is weak but consistent evidence that there is no difference between unilateral or bilateral implantation if speech and noise were presented from the front. When the noise was presented from the first implanted side, scores improved in two studies (table III) (Lovett et al. 2010; Sparreboom et al. 2010a). Although the current evidence implies the beneficial effect of the second cochlear implant on especially sound localization, one could question whether the impact of improved sound localization abilities on daily life could justify bilateral implantation. Longer follow-up data and/or studies with adult bilateral cochlear implant users might reveal whether improved sound localization abilities can aid in everyday listening situations, such as localizing traffic or warning signs. When bilateral implantation was compared to children with a bimodal fitting, there was insufficient and inconsistent evidence to make a valid comparison between the two treatment options. The current available literature presented in this study solely consisted of cohort studies that compared a bilateral group with a unilateral control group. Only one study by Boons et al. provided a thorough set of matching criteria and a clear description of the selection procedure, reducing the risk for selection bias (Boons et al. 2012). Rigorously performed randomized trials, which are less susceptible of selection bias have not been performed. Although most studies deemed impossible the blinding of study participants or observers, this criterion was maintained in our risk of bias to indicate the risk of information bias, which could be introduced by the absence of blinding. The present review does not only provides an update of the literature but it is also unique since it only compares bilateral cochlear implant groups with unilaterally implanted control groups. By contrast, the previous reviews also included studies presenting within-subject comparisons, thereby introducing potential sources of bias (Bond et al. 2009; Sparreboom et al. 2010b). However, some potential limitations of the present review should also be addressed. First, we decided not to statistically pool the results, especially because of the clinical heterogeneity between studies, that is, such as the use of different scales and tests and the high risk of bias scores. Instead, we used a best-evidence synthesis, which is a qualitative summarization of the available evidence (Table III) (Slavin 1995). 60
Bilateral cochlear implantation review
The overall score in our best-evidence synthesis shows a large number of comparisons indicating no difference. This can be partially contributed to the numerous comparisons made in which no or little benefit of the second implant is expected, such as generic quality of life. Second, in several studies data could only be extracted from figures or graphs (Litovsky et al. 2006b; Sparreboom et al. 2010a; Murphy et al. 2011). Although we have made an effort to maximize the accuracy, these point estimates do not reflect the actual data and should be interpreted carefully. Third, some study groups presenting various outcome measures in different reports based their results on (partially) identical study populations (Litovsky et al. 2006b; Litovsky et al. 2006a; Mok et al. 2007; Grieco-Calub et al. 2008; Grieco-Calub et al. 2009; Nittrouer and Chapman 2009; Mok et al. 2010; Sparreboom et al. 2010a; Grieco-Calub and Litovsky 2012; Nittrouer et al. 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a; Nittrouer et al. 2013). One should be careful not to interpret these study populations as separate studies. Finally, the included studies differed widely, within and between studies, in the age range at which the children received either their first or second implant, varying from 0.6 to 13 years and 0.6 to 15.8 years respectively (Table II). Because the inter-implant interval is an acknowledged determinant of postoperative performance (Lammers et al. 2014), its variability between studies might have influenced the results.
Conclusion In conclusion, this systematic review demonstrates that the current best available evidence only consists of cohort studies, with moderate to high risk of bias. Therefore, firm recommendations based on randomized trials cannot be made. The results of our bestevidence synthesis indicate that the second cochlear implant might be useful in sound localization and possibly in language development; however, it is arguable whether this is sufficient to support the use of the second implant.
Acknowledgement The authors thank Dr. Grieco-Calub and Prof. Litovsky (University of Wisconsin-Madison) as well as Dr. Lovett (University of York) for providing additional data on their studies.
61
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Appendix 1 Search query: Pubmed: 1. (child[Title/Abstract]) OR (children[Title/Abstract]) OR (pediatric[Title/Abstract]) OR (paediatric[Title/Abstract]) OR (Toddler[Title/Abstract]) OR (Infant[Title/Abstract]) OR (Baby[Title/Abstract]) OR (Newborn[Title/Abstract]) OR (Youngster[Title/Abstract]) OR (Adolescent[Title/Abstract]) OR (babies[Title/Abstract]) 2. (((Cochlear[Title/Abstract]) OR (cochlea[Title/Abstract])) AND ((implant[Title/Abstract]) OR (implants[Title/Abstract]) OR (implanted[Title/Abstract]) OR (implantation[Title/Abstract]) OR (device[Title/Abstract]) OR (devices[Title/Abstract]) OR (prosthesis[Title/Abstract]) OR (prostheses[Title/Abstract]) OR (prosthetic[Title/Abstract]))) 3. (Electro acoustic stimulation[Title/Abstract]) OR (Electric stimulation[Title/Abstract]) 4. (bilaterally implanted[Title/Abstract]) OR bici[Title/Abstract]) 5. #2 OR #3 OR #4 #1 AND #5
62
Bilateral cochlear implantation review
References Baudonck N, Van Lierde K, D’Haeseleer E, Dhooge I (2011) A comparison of the perceptual evaluation of speech production between bilaterally implanted children, unilaterally implanted children, children using hearing aids, and normal-hearing children. International Journal of Audiology 50:912-919. Beijen JW, Snik AFM, Mylanus EAM (2007) Sound localization ability of young children with bilateral cochlear implants. Otology & Neurotology 28:479-485. Bond M, Mealing S, Anderson R, Elston J, Weiner G, Taylor R, Hoyle M, Liu Z, Price A, Stein K (2009) The effectiveness and cost-effectiveness of cochlear implants for severe to profound deafness in children and adults: a systematic review and economic model. Health Technology Assessment 13:1. Boons T, Brokx JPL, Frijns JHM, Peeraer L, Philips B, Vermeulen A, Wouters J, van Wieringen A (2012) Effect of Pediatric Bilateral Cochlear Implantation on Language Development. Archives of Pediatrics & Adolescent Medicine 166:28-34. Eustaquio ME, Berryhill W, Wolfe JA, Saunders JE (2010) Balance in Children With Bilateral Cochlear Implants. Otology & Neurotology 32:424-427. Grieco-Calub TM, Litovsky RY (2012) Spatial Acuity in 2-to-3-Year-Old Children With Normal Acoustic Hearing, Unilateral Cochlear Implants, and Bilateral Cochlear Implants. Ear and Hearing 33:561-572. Grieco-Calub TM, Litovsky RY, Werner LA (2008) Using the observer-based psychophysical procedure to assess localization acuity in toddlers who use bilateral cochlear implants. Otology & Neurotology 29:235-239. Grieco-Calub TM, Saffran JR, Litovsky RY (2009) Spoken word recognition in toddlers who use cochlear implants. Journal of Speech, Language, and Hearing Research 52:1390-1400. Higgins JPT, Altman DG, Sterne JAC, (editors) (2011. Available from www.cochrane-handbook.org.) Chapter 8: Assessing risk of bias in included studies. In: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) Edition (Higgins J.P.T., Green S., eds). www.cochranehandbook.org: The Cochrane Collaboration. Hodges AV, Balkany TJ (2012) Cochlear implants in children and adolescents. Archives of Pediatrics & Adolescent Medicine 166:93-94. Lammers MJ, Grolman W, Smulders YE, Rovers MM (2011) The Cost-utility of bilateral cochlear implantation: A Systematic Review. Laryngoscope 121:2604-2609. Lammers MJ, Venekamp RP, Grolman W, van der Heijden GJ (2014) Bilateral cochlear implantation in children and the impact of the inter-implant interval. Laryngoscope 124:993-999. Litovsky RY, Johnstone PM, Godar SP (2006a) Benefits of bilateral cochlear implants and/or hearing aids in children. International Journal of Audiology 45:S78-S91. Litovsky RY, Johnstone PM, Godar S, Agrawal S, Parkinson A, Peters R, Lake J (2006b) Bilateral cochlear implants in children: Localization acuity measured with minimum audible angle. Ear and Hearing 27:43-59. Lovett RES, Kitterick PT, Hewitt CE, Summerfield AQ (2010) Bilateral or unilateral cochlear implantation for deaf children: an observational study. Archives of Disease in Childhood 95:107-112. Mok M, Galvin KL, Dowell RC, McKay CM (2007) Spatial unmasking and binaural advantage for children with normal hearing, a cochlear implant and a hearing aid, and bilateral implants. Audiology and Neurotology 12:295-306. Mok M, Galvin KL, Dowell RC, McKay CM (2010) Speech perception benefit for children with a cochlear implant and a hearing aid in opposite ears and children with bilateral cochlear mplants. Audiology and Neurotology 15:44-56. Murphy J, Summerfield AQ, O’Donoghuea’c GM, Moore DR (2011) Spatial hearing of normally hearing and cochlear implanted children. International Journal of Pediatric Otorhinolaryngology 75:489-494. Nittrouer S, Chapman C (2009) The effects of Bilateral electric and bimodal electric-acoustic stimulation on language development. Trends in Amplification 13:190-205.
63
1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Nittrouer S, Caldwell A, Lowenstein JH, Tarr E, Holloman C (2012) Emergent Literacy in Kindergartners With Cochlear Implants. Ear and Hearing 33:683-697. Nittrouer S, Caldwell-Tarr A, Tarr E, Lowenstein JH, Rice C, Moberly AC (2013) Improving speech-in-noise recognition for children with hearing loss: Potential effects of language abilities, binaural summation, and head shadow. International Journal of Audiology 52:513-525. Schafer EC, Thibodeau LM (2006) Speech recognition in noise in children with cochlear implants while listening in bilateral, bimodal, and FM-system arrangements. American Journal of Audiology 15:114-126. Slavin RE (1995) Best evidence synthesis: An intelligent alternative to meta-analysis. Journal of Clinical Epidemiology 48:9-18. Sparreboom M, Snik AFM, Mylanus EAM (2010a) Sequential Bilateral Cochlear Implantation in Children: Development of the Primary Auditory Abilities of Bilateral Stimulation. Audiology and Neurotology 16:203-213. Sparreboom M, Snik AFM, Mylanus EAM (2012a) Sequential Bilateral Cochlear Implantation in Children Quality of Life. Archives of Otolaryngology-Head & Neck Surgery 138:134-141. Sparreboom M, Leeuw AR, Snik AFM, Mylanus EAM (2012b) Sequential bilateral cochlear implantation in children: Parentsâ&#x20AC;&#x2122; Perspective and device use. International Journal of Pediatric Otorhinolaryngology 76:339-344. Sparreboom M, van Schoonhoven J, van Zanten BG, Scholten RJ, Mylanus EA, Grolman W, Maat B (2010b) The effectiveness of bilateral cochlear implants for severe-to-profound deafness in children: a systematic review. Otology & Neurotology 31:1062-1071. Tait M, Nikolopoulos TP, De Raeve L, Johnson S, Datta G, Karltorp E, Ostlund E, Johansson U, van Knegsel E, Mylanus EAM, Gulpen PMH, Beers M, Frijns JHM (2010) Bilateral versus unilateral cochlear implantation in young children. International Journal of Pediatric Otorhinolaryngology 74:206-211.
64
Chapter 1.3 Bilateral cochlear implantation in children and the impact of the inter-implant interval
M.J.W. Lammers, R.P. Venekamp, W. Grolman, G.J.M.G. van der Heijden
Laryngoscope 2014; 124(4):993-9
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Objective: To determine the effectiveness of simultaneous versus sequential bilateral cochlear implantation on postoperative outcomes in children with bilateral deafness and to evaluate the impact of the inter-implant interval and age at second implantation on postoperative outcomes in children who already received their first cochlear implant. Data sources: PubMed, Embase and Web of Science. Review Methods: All studies comparing the effects of simultaneous with sequential bilateral cochlear implantation on postoperative outcomes and those evaluating the impact of the inter-implant interval and age at second implantation were retrieved. Results: Four studies compared the effects of simultaneous with sequential bilateral cochlear implantation. All studies lacked randomization. Of these, three reported better speech perception and expressive language development at one year of bilateral experience for simultaneous cochlear implantation. Of the nineteen publications on the impact of the inter-implant interval on postoperative outcomes, the risk of bias was lowmoderate for seven studies which were derived from five different study populations. In two of these populations no impact of the inter-implant interval was found, while in three a longer inter-implant interval was associated with poorer speech and language development. Conclusion: Observational studies suggest that simultaneous implantation in children may be associated with improved speech and language development, and that a prolonged inter-implant interval between both implantations may have a negative impact on these postoperative outcomes. Randomized trials are, however, needed to demonstrate whether simultaneous implantation indeed is superior to sequential bilateral implantation in children with bilateral deafness.
66
Inter-implant interval review
Introduction During the last decade the literature suggesting bilateral cochlear implantation to be superior to unilateral cochlear implantation has rapidly grown. This has resulted in a swift increase in the number of bilateral implantations in children with bilateral severe to profound sensorineural hearing loss (Ramsden et al. 2010). With this change in clinical practice and the reimbursement of the second cochlear implant in a growing number of countries, the question has been raised about the optimal age for providing the second cochlear implant. Neurophysiologic studies suggest that a prolonged interval between both cochlear implantations may have impact on the maturation of the brainstem, and may lead to timing differences of auditory brainstem activity (Gordon et al. 2008a). Experts therefore advocate that simultaneous bilateral cochlear implantation should be preferred over sequential implantation (Ramsden et al. 2010). Yet, studies have demonstrated that children may exhibit a remarkable catch-up with a second implant activated later in life (Smulders et al. 2011). Moreover, concerns have been raised about the long term safety of a prolonged anaesthetic procedure during early infancy which is needed for simultaneous cochlear implantations (Sun 2010). Consequently, bilateral cochlear implantation strategies vary within and between countries. To date, a systematic assessment of the literature on the effectiveness of simultaneous versus sequential bilateral cochlear implantation on postoperative outcomes in children with bilateral deafness has not been reported. In a previous review, a prolonged interval between sequential cochlear implantations has been associated with a negative impact on postoperative outcomes, notably speech and language development (Smulders et al. 2011). This review however, did not consider the age at second implantation as a co-variate of effect, although this is known to strongly correlate with the inter-implant interval in especially early implanted children (Sparreboom et al. 2012b). Furthermore, since its publication new studies on the impact of the inter-implant interval and age at second implantation on postoperative outcomes have been published. The objective of this systematic review therefore is to compare the effectiveness of simultaneous with sequential bilateral cochlear implantation on postoperative outcomes, notably speech and language development, in children with bilateral deafness, and to include age at second implantation in the evaluation of the impact of the inter-implant interval for sequential cochlear implantation on postoperative outcomes.
67
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Materials and methods Search strategy and study selection We retrieved publications from PubMed, Embase and Web of Science from their start up to January 27, 2013, by using the search terms ‘cochlear implant’, ‘children’ and their synonyms in title and abstract fields (Appendix 1). One review author (M.L.) screened titles and abstracts of all retrieved unique publications using pre-defined selection criteria for their relevance for answering our study questions (Fig. 1). Next, the full-texts of potentially relevant publications were assessed in more detail using the same criteria. Studies on children with abnormal cochlear anatomy and those not reporting patient-relevant outcomes (such as electrophysiology, surgical parameters) were excluded. Cross reference search for all selected studies and reviews was used to increase the yield of relevant studies (Fig. 1).
Fig. 1. Flow-diagram of search strategy. Search date: January 27, 2013
68
Inter-implant interval review
Risk of bias assessment Two review authors (M.L., R.V.) independently assessed the risk of bias of all potentially relevant studies. Reviewers resolved initial disagreements by discussion. The risk of bias assessment for studies comparing the effectiveness of simultaneous versus sequential implantation concerned random sequence generation, concealment of allocation, blinding (of both participants and outcome assessment), standardization of outcome measurements, completeness of data (attrition) and standardization of cochlear implantation procedures (Higgins et al. 2011). The risk of bias assessment for studies evaluating the impact of inter-implant interval and age at the second cochlear implantation was assessed according to the procedure described by Hayden et al. (Hayden et al. 2006). For each study we evaluated whether the study population represents the population of interest and if recruitment of the study population was clearly described in the article. Furthermore, we assessed the duration of follow up, the attrition rate, whether the prognostic factor, the given treatment and outcome measures were clearly described and standardized, and finally if appropriate statistical analyses were performed. In addition, we assessed whether studies included a homogeneous inception cohort of only sequentially implanted children with up to 3 years of bilateral deafness. We excluded studies with a cohort of children that suffered from bilateral hearing deprivation for longer than 3 years since this period has been acknowledged as important predictor for childâ&#x20AC;&#x2122;s postoperative performance (Sharma et al. 2005; Lee et al. 2007; Gordon et al. 2008b; Gordon and Papsin 2009). Furthermore, for each of the 9 risk of bias criteria one point is awarded if the criterion was not fulfilled. We used a predefined cutoff value of â&#x2030;¤ 3 points (one-third of the total score) in order to separate high quality studies (low to moderate risk of bias) from studies with a high risk of bias. Studies with a high risk of bias, that is, more than three risk of bias assessment items scored as likely flawed (score > 3 points) were excluded from our analyses. Data extraction and analysis For each study we extracted data on design, number of included patients and the study population characteristics at baseline. Our primary outcome of interest was speech and language development. Secondary outcomes included speech perception in quiet and noise, sound localization ability, including spatial acuity, and other reported clinically relevant outcomes. We extracted all reported outcome data, presented in each study, for our primary and secondary outcomes. Both univariate and multivariate outcomes were eligible and obtained from each study.
69
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Results Search results We retrieved a total of 8.896 records including 5.013 unique publications (Fig. 1). After screening their title and abstract, 96 of these remained for assessment of their full text. Of these, four studies fulfilled our criteria for the review on the effectiveness of simultaneous versus sequential cochlear implantation (Gordon and Papsin 2009; Chadha et al. 2011; Boons et al. 2012; Vincent et al. 2012), while nineteen articles were found to be suitable to answer our study question regarding inter-implant interval and/or the age at second cochlear implantation (Kuhn-Inacker et al. 2004; Bohnert et al. 2006; Manrique et al. 2007; Scherf et al. 2007; Wolfe et al. 2007; Steffens et al. 2008; Zeitler et al. 2008; Gordon and Papsin 2009; Scherf et al. 2009; Grieco-Calub and Litovsky 2010; Sparreboom et al. 2010; Van Deun et al. 2010; Boons et al. 2012; Grieco-Calub and Litovsky 2012; Kim et al. 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a; Strom-Roum et al. 2012a; Strom-Roum et al. 2012b). Two studies reported data that could be used for both our questions (Gordon and Papsin 2009; Boons et al. 2012). Hence, 21 studies qualified for risk of bias assessment and further analyses. Simultaneous versus sequential bilateral implantation Study assessments The four studies comparing the effectiveness of simultaneous with sequential implantation (Table 1a) all carry a high risk of bias, because treatment allocation was neither randomized nor concealed, nor were outcome assessments blinded. Only one study sufficiently standardized outcome assessments (Gordon and Papsin 2009; Chadha et al. 2011; Boons et al. 2012), resulting in a widely variable duration of follow-up in the 3 other studies (Chadha et al. 2011; Boons et al. 2012; Vincent et al. 2012). One study used a matched case-control design (Boons et al. 2012), while the other three studies used a cohort design (Gordon and Papsin 2009; Chadha et al. 2011; Vincent et al. 2012). The age at first implantation in one cohort study was highly variable and ranged from 0.7 to 8.9 years (Vincent et al. 2012), whereas children in the other three studies received their first cochlear implant before the age of 3 years (Gordon and Papsin 2009; Chadha et al. 2011; Boons et al. 2012). Reported outcomes Of the four included studies, three reported statistically significant superior postoperative outcomes of simultaneous over sequential bilateral cochlear implantation, but none of these studies reported effect sizes (Gordon and Papsin 2009; Chadha et al. 2011; Boons et al. 2012). Statistical pooling of outcomes was not feasible due to the heterogeneity in study design and outcomes that were used. The ability of children with simultaneous bilateral
70
Inter-implant interval review
cochlear implants to use both devices was higher than of those with sequential implantation, and this may have resulted in the reported superior understanding of speech in quiet and noise after at least one year of bilateral experience in two studies (Gordon and Papsin 2009; Chadha et al. 2011). In another study speech perception scores in quiet did not differ and were around 83% correct for both groups, when tested in noise scores were 65% and 77% correct in respectively the simultaneous and sequential group (Vincent et al. 2012). Finally, three years after the first cochlear implantation those with simultaneous implantations obtained statistically significant higher scores in terms of speech and language development, as determined with the Word development subscale of the Schlichting Expressive Language Test (Boons et al. 2012). Table Ia. Risk of bias assessment of the included studies comparing simultaneous with sequential bilateral implantation Author, year
N Sim
N Seq
Boons, 2012
8
14
Vincent, 2012 Chadha, 2011 Gordon, 2009
12 10 6
11 12 30*
Design Matched case-control Cohort Cohort Cohort
1.3
RA
CA
BL
ATT
ST Out
ST Treat
N Sim: number of simultaneous implanted children; N Seq: number of sequentially implanted children. Random sequence generation (RA); :Yes, randomization used and method adequately described; : No randomization used; ?: unclear/not adequately described. Concealment of allocation (CA); : Yes; : No; ?: unclear/not adequately described. Attrition / Completeness of data (ATT); : ≤ 10% loss to follow-up; : > 10% loss to follow-up; ?: unclear/not adequately described. Blinding (BL); : blinding of participant and/or outcome assessor; : no blinding of participant and/or outcome assessor; ?: unclear/not adequately described. Standardization of outcome (ST Out); : validated and age specific outcome measures used at standardized evaluation moments after the second implantation; : validated and age specific outcome measures used, but not at standardized evaluation moments after the second implantation; : no standardization of outcome measurement ?: unclear/not adequately described. Standardization of treatment (ST Treat); : standardization of treatment (type and manufacturer of cochlear implantations); : no standardization of treatment (type and manufacturer of cochlear implantations); ?: unclear/not adequately described. *: The study by Gordon et al. 2009 only presented comparisons between the simultaneous group and a long delay group. Comparisons between the simultaneous and short delay group were not performed and therefore not included in this analysis.
71
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
The impact of inter-implant interval and age Study assessments Of the nineteen studies evaluating the impact on postoperative outcomes of inter-implant interval and age at the second cochlear implantation (Table 1b) the risk of bias was low to moderate for 7 studies, and high for 12 studies. A homogeneous inception cohort (duration of bilateral deafness ≤ 3 years) was reported in nine studies (Scherf et al. 2007; Wolfe et al. 2007; Steffens et al. 2008; Gordon and Papsin 2009; Sparreboom et al. 2010; Boons et al. 2012; Grieco-Calub and Litovsky 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a), of which two were excluded for further analysis because of their high risk of bias (Scherf et al. 2007; Steffens et al. 2008). Two studies did not use a standardized evaluation protocol resulting in widely variable duration of follow-up (Boons et al. 2012; Grieco-Calub and Litovsky 2012). Three studies reported on data of children with a widely variable baseline status, who received either sequential or simultaneous bilateral implantation (Gordon and Papsin 2009; Boons et al. 2012; Grieco-Calub and Litovsky 2012). Table Ib. Risk of bias assessment of the included studies evaluating inter-implant interval and/or age at 2nd implantation Author, year
Prognostic Design Pop F/U factor
Interval / Age at 2nd CI Age at 2nd Sparreboom,2012a* CI Age at 2nd Sparreboom,2012b* CI Age at 2nd Sparreboom,2010* CI DurationWolfe, 2007 deafness 2nd CI Age at 2nd Grieco-Calub, 2012 CI Boons, 2012 Interval
Gordon, 2009
Steffens, 2008 Scherf, 2007 Van Deun, 2010 Kim, 2012
72
Interval / Age at 2nd CI Interval Age at 2nd CI Interval
Att
Out
Bl
Prog Ana Treat Score
?
2
?
2
?
2
?
2
?
2
?
3
?
3
?
3.5
?
4.5
?
?
Inter-implant interval review
Age at 2nd CI Strom-Roum, 2012a Interval
Scherf, 2009
?
?
Strom-Roum, 2012b
Interval
?
Kuhn-Inacker, 2004
Interval Age at 2nd CI Interval
?
?
?
?
Grieco-Calub, 2010 Manrique, 2007 Zeitler, 2008
Interval / Age at 2nd CI
?
Bohnert, 2006
Interval / Age at 2nd CI
?
?
?
?
?
*: identical study populationsDesign; : Children included at common point in course of disease (inception cohort): duration of bilateral deafness ≤ 3 years; : Children included at different points in course of disease: wide range of duration of bilateral deafness; ?: unclear/not adequately described. Study population (Pop); : In- and exclusion criteria defined, clear description of sample selection; : In- and exclusion criteria defined, unclear description sample selection; : In- and exclusion criteria and sample selection not specified, unclear description sample selection; ?: unclear/not adequately described. Duration follow up (F/U); : ≥6 months of binaural experience in all children; : ≥6 months of binaural experience in 75-99% of subjects; : ≤ 6 months of binaural experience in 0-75% of subjects or < 6 months in all subjects; ?: unclear/not adequately described. Attrition (Att); : ≤10% loss and reasons explained; : ≤10% loss and reasons not explained or 10-20% loss and reasons explained; : 10-20% loss and reasons not explained or >20% loss; ?: unclear/not adequately described. Standardization of outcome measure (Out); : validated and age specific outcome measures used at standardized evaluation moments after the second implantation; : validated and age specific outcome measures used, but not at standardized evaluation moments after the second implantation; : no standardization of outcome measurement. Blinding for prognostic factor, while determining outcome (Bl); : outcome assessor was blinded for inter-implant interval and/or age at 2nd implantation; : outcome assessor was aware of the interimplant interval and/or age at 2nd implantation; ?: unclear/not adequately described. Prog (standardization of prognostic factor); : Fully defined and available for all children; : Not defined; ?: unclear/not adequately described. Analysis (Ana); : Continuous predictor variable analyzed appropriately with multivariable analysis in cohort of sequentially implanted children; : Continuous predictor variable analyzed appropriately without multivariable analysis performed in cohort of sequentially implanted children, or multivariable analysis in mixed cohort of simultaneous and sequentially implanted children; : inappropriate analysis or no multivariable analysis performed in mixed cohort of simultaneous and sequentially implanted children; ?: unclear/not adequately described. Standardization of treatment (Treat); : standardized treatment (i.e. cochlear implants by identical manufactures used in same patient); : not standardized; ?: unclear /not adequately described. Score; For each of the 9 criteria one point is awarded if the criterion was not fulfilled (). A half point was awarded if the criterion was partially fulfilled (), and zero points if the criterion was fulfilled (). A predefined cut-off value of ≤ 3 points (1/3 of the total score) was used in order to separate high quality studies (low to moderate risk of bias) from low quality studies (high risk of bias). Scores were only calculated for studies including children with a duration of bilateral deafness ≤ 3 years.
73
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Reported outcomes The seven included studies reported clinically relevant outcomes that were derived from five different study populations (Wolfe et al. 2007; Gordon and Papsin 2009; Sparreboom et al. 2010; Boons et al. 2012; Grieco-Calub and Litovsky 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a). Statistical pooling of outcomes was not feasible due to the heterogeneity in the design and conduct of these observational studies and due to the diversity in the reported outcomes as shown in Table 2. Figure 2 presents the age at the second cochlear implantation (mean and range) as a function of study size. Three studies report statistically significant poorer language development and speech perception in quiet for longer inter-implant intervals or an older age at the second implantation (Wolfe et al. 2007; Gordon and Papsin 2009; Boons et al. 2012). Sparreboom et al. and Grieco-Calub et al. report no such impact of older age at the second implantation on postoperative outcomes (Sparreboom et al. 2010; Grieco-Calub and Litovsky 2012; Sparreboom et al. 2012b; Sparreboom et al. 2012a). Table II. Results From all Five Included Study Populations Evaluating the Inter-implant Interval and/or Age at 2nd Implantation Speech perception: bilateral benefit
Speech and language Sound development localization SELT MAA Le/Ri RDLS SELT word sentence
QoL
Device use
-
-
-
-
-
NS
NS
-
-
-
-
-
-
-
NS
NS
-
-
r =0.50, P =0.01
r =0.53, P =0.006
-
-
-
-
Author, year
Silence
Noise
Gordon, 2009
r2 =-0.23, P <0.0001
NS
-
-
-
-
Sparreboom, 2010, 2012a,b
NS
NS
-
-
-
r =0.86, P =0.002
NS
-
-
Grieco-Calub, 2012
-
-
-
Boons, 2012
-
-
r =0.40, P =0.04
Wolfe, 2010
Other
Le/Ri = Left-right discrimination, MAA = Mean audible angle, NS = not significant, QoL = quality of life, RDLS = Reynell Developmental Language Scales, SELT = Schlichring Expressive Language Test
74
Inter-implant interval review
Fig. 2. Mean age at the second implantation of the study population as a function of cohort size. Error bars indicate the age range at the second implantation. Dotted error bars denote studies which did not report an effect of inter-implant interval or age at second implantation. For the study population of Gordon et al. 2009 age ranges could not be extracted, therefore standard deviations are presented instead.
Discussion The available observational studies suggest that simultaneous bilateral cochlear implantation in children with bilateral deafness may provide better postoperative outcomes than sequential cochlear implantation. While in three studies, including young children with bilateral deafness of â&#x2030;¤ 3 years, simultaneous bilateral cochlear implantation provided a statistically significant better speech perception and development of expressive language, the fourth study, including children receiving a relatively late first implant (up to 8.9 years), found no such differences between simultaneous and sequential implantation (Gordon and Papsin 2009; Chadha et al. 2011; Boons et al. 2012; Vincent et al. 2012). It should however be noted that this evidence does not come from randomized trials but from observational comparative studies with relative small samples, and rather low methodological quality. Therefore, these studies do not allow firm conclusions on the benefit of simultaneous over sequential bilateral cochlear implantation. The available studies on the impact of the inter-implant interval with bilateral cochlear implantation are conflicting. However, the studies published to date suggest that a longer inter-implant interval has a negative effect, or at best no impact on postoperative outcomes. These findings are in agreement with the results of a previous review (Smulders et al. 2011), and neurophysiologic studies which indicate that long intervals are associated with prolonged auditory brainstem peak latencies (Gordon et al. 2008a). To our knowledge, this is the first systematic review assessing both the effectiveness of simultaneous versus sequential bilateral cochlear implantation and the impact of the inter-
75
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
implant interval and age at the second implantation on postoperative outcomes in children with bilateral deafness. Before translating our findings to daily practice, some aspects of our study deserve further attention. First, to allow for a comparison between two therapeutic strategies (simultaneous versus sequential bilateral cochlear implantation) and an evaluation of the impact of inter-implant interval (including only sequentially implanted children), we had to perform two separate risk of bias assessments. Although the results of both analyses could therefore not be combined, the findings complement each other as the studies revealed that a longer inter-implant interval has either no or a negative effect on postoperative outcomes. Second, it should be noted that several publications reported data derived from identical study populations (Sparreboom et al. 2010; Sparreboom et al. 2012b; Sparreboom et al. 2012a). The study outcomes that were extracted from such multiple reports were combined. Moreover, it turned out that two studies could be used to answer both our questions (Gordon and Papsin 2009; Boons et al. 2012). Third, sample sizes of all studies comparing simultaneous with sequential bilateral implantation were relatively small. Although one study reported a matching procedure for balancing baseline risk of the compared groups (Boons et al. 2012), all carried a high risk of selection bias due to the lack of random and concealed allocation of treatments. Fourth, not all included studies reported outcomes at standardized evaluation moments after the second implantation (Boons et al. 2012; Grieco-Calub and Litovsky 2012). In these studies the varying duration of bilateral experience could have influenced the results. Furthermore, three of the five study populations in our review on the impact of interimplant interval included children undergoing both simultaneous and sequential cochlear implantation (Gordon and Papsin 2009; Boons et al. 2012; Grieco-Calub and Litovsky 2012). In two of these, a negative effect of inter-implant interval was reported (Gordon and Papsin 2009; Boons et al. 2012). This result, however, might be influenced by the fact that these studies included both children who underwent simultaneous and sequential bilateral implantation. Fifth, the clinical heterogeneity between the various study populations might also have led to the conflicting results found in our review. Preferably, one would like to perform a comparison of multiple homogeneous study populations, which are similar in factors such as duration of deafness and etiology, since this would provide more insight in the direction and magnitude of the effect of the inter-implant interval. In this review we attempt to rule out the potential influence of variable duration of bilateral deafness at baseline on outcomes by selecting studies only including children with bilateral deafness of â&#x2030;¤ 3 years (Gordon and Papsin 2009). Therefore, our findings on the impact of the inter-implant delay on postoperative performance in those who already received their first cochlear implant are only applicable to children with bilateral deafness shorter than 3 years.
76
Inter-implant interval review
Finally, our review did not systematically evaluate whether simultaneous implantation is associated with a higher risk of complications in children as compared to sequential implantation. Although the studies included in our review did not report complications, this observation should be interpreted with caution. That is, the studies included only a relatively small number of children and the follow up in the included studies was relative short. Still, these findings are in agreement with recent reports on the safety of simultaneous bilateral cochlear implantation in young children (Ramsden et al. 2009; Grainger et al. 2012).
Conclusion The results of this review indicate that although evidence from randomized trials is lacking, observational studies suggest that simultaneous bilateral cochlear implantation is associated with better postoperative speech and language development as compared to sequential cochlear implantation in children with bilateral deafness. These results are in line with the findings that a prolonged inter-implant interval may have a negative impact on these postoperative outcomes. Randomized controlled trials are however needed to convincingly show that simultaneous implantation is indeed superior to sequential bilateral implantation in children with bilateral deafness.
77
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Appendix 1 Search query: Pubmed: 1. (child[Title/Abstract]) OR (children[Title/Abstract]) OR (pediatric[Title/Abstract]) OR (paediatric[Title/Abstract]) OR (Toddler[Title/Abstract]) OR (Infant[Title/Abstract]) OR (Baby[Title/Abstract]) OR (Newborn[Title/Abstract]) OR (Youngster[Title/Abstract]) OR (Adolescent[Title/Abstract]) OR (babies[Title/Abstract]) 2. (((Cochlear[Title/Abstract]) OR (cochlea[Title/Abstract])) AND ((implant[Title/Abstract]) OR (implants[Title/Abstract]) OR (implanted[Title/Abstract]) OR (implantation[Title/Abstract]) OR (device[Title/Abstract]) OR (devices[Title/Abstract]) OR (prosthesis[Title/Abstract]) OR (prostheses[Title/Abstract]) OR (prosthetic[Title/Abstract]))) 3. (Electro acoustic stimulation[Title/Abstract]) OR (Electric stimulation[Title/Abstract]) 4. (bilaterally implanted[Title/Abstract]) OR bici[Title/Abstract]) 5. #2 OR #3 OR #4 #1 AND #5
78
Inter-implant interval review
References Bohnert A, Spitzlei V, Lippert KL, Keilmann A (2006) Bilateral cochlear implantation in children: Experiences and considerations. Volta Review 106:343-364. Boons T, Brokx JPL, Frijns JHM, Peeraer L, Philips B, Vermeulen A, Wouters J, van Wieringen A (2012) Effect of Pediatric Bilateral Cochlear Implantation on Language Development. Archives of Pediatrics & Adolescent Medicine 166:28-34. Chadha NK, Papsin BC, Jiwani S, Gordon KA (2011) Speech Detection in Noise and Spatial Unmasking in Children With Simultaneous Versus Sequential Bilateral Cochlear Implants. Otology & Neurotology 32:1057-1064. Gordon KA, Papsin BC (2009) Benefits of Short Interimplant Delays in Children Receiving Bilateral Cochlear Implants. Otology & Neurotology 30:319-331. Gordon KA, Valero J, van Hoesel R, Papsin BC (2008a) Abnormal timing delays in auditory brainstem responses evoked by bilateral cochlear implant use in children. Otology & Neurotology 29:193-198. Gordon KA, Tanaka S, Wong DD, Papsin BC (2008b) Characterizing responses from auditory cortex in young people with several years of cochlear implant experience. Clinical Neurophysiology 119:2347-2362. Grainger J, Jonas NE, Cochrane LA (2012) Simultaneous cochlear implantation in children: the Great Ormond Street experience. Cochlear Implants International 13:137-141. Grieco-Calub TM, Litovsky RY (2010) Sound Localization Skills in Children Who Use Bilateral Cochlear Implants and in Children With Normal Acoustic Hearing. Ear and Hearing 31:645-656. Grieco-Calub TM, Litovsky RY (2012) Spatial Acuity in 2-to-3-Year-Old Children With Normal Acoustic Hearing, Unilateral Cochlear Implants, and Bilateral Cochlear Implants. Ear and Hearing 33:561-572 Hayden JA, Cote P, Bombardier C (2006) Evaluation of the quality of prognosis studies in systematic reviews. Annals of Internal Medicine 144:427-437. Higgins JPT, Altman DG, Sterne JAC, (editors) Chapter 8: Assessing risk of bias in included studies. In: Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) Edition (Higgins J.P.T., Green S., eds). www.cochrane-handbook.org: The Cochrane Collaboration. Kim JS, Kim LS, Jeong SW (2012) Functional benefits of sequential bilateral cochlear implantation in children with long inter-stage interval between two implants. International Journal of Pediatric Otorhinolaryngology 77:162-169. Kuhn-Inacker H, Shehata-Dieter W, Muller J, Helms J (2004) Bilateral cochlear implants: a way to optimize auditory perception abilities in deaf children? International Journal of Pediatric Otorhinolaryngology 68:1257-1266. Lee HJ, Giraud AL, Kang E, Oh SH, Kang H, Kim CS, Lee DS (2007) Cortical activity at rest predicts cochlear implantation outcome. Cerebral Cortex 17:909-917. Manrique M, Huarte A, Valdivieso A, Perez B (2007) Bilateral sequential implantation in children. Audiological Medicine 5:224. Ramsden JD, Papsin BC, Leung R, James A, Gordon KA (2009) Bilateral simultaneous cochlear implantation in children: our first 50 cases. Laryngoscope 119:2444-2448. Ramsden JD, Gordon K, Aschendorff A, Borucki L, Bunne M, Burdo S, Garabedian N, Grolman W, Irving R, Lesinski-Schiedat A, Loundon N, Manrique M, Martin J, Raine C, Wouters J, Papsin BC (2010) European Bilateral Pediatric Cochlear Implant Forum consensus statement. Otology & Neurotology 33:561-565. Scherf F, van Deun L, van Wieringen A, Wouters J, Desloovere C, Dhooge I, Offeciers E, Deggouj N, De Raeve L, De Bodt M, Van de Heyning PH (2007) Hearing benefits of second-side cochlear implantation in two groups of children. International Journal of Pediatric Otorhinolaryngology 71:1855-1863. Scherf F, Van Deun L, van Wieringen A, Wouters J, Desloovere C, Dhooge I, Offeciers E, Deggouj N, De Raeve L, Wuyts FL, Van de Heyning P (2009) Subjective Benefits of Sequential Bilateral Cochlear Implantation in Young Children after 18 Months of Implant Use. Journal for Oto-Rhino-Laryngology, Head and Neck Surgery 71:112-121. 79
1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 1.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Sharma A, Dorman MF, Kral A (2005) The influence of a sensitive period on central auditory development in children with unilateral and bilateral cochlear implants. Hearing Research 203:134-143. Smulders YE, Rinia AB, Rovers MM, van Zanten GA, Grolman W (2011) What is the effect of time between sequential cochlear implantations on hearing in adults and children? A systematic review of the literature. Laryngoscope 121:1942-1949. Sparreboom M, Snik AFM, Mylanus EAM (2010) Sequential Bilateral Cochlear Implantation in Children: Development of the Primary Auditory Abilities of Bilateral Stimulation. Audiology and Neurotology 16:203-213. Sparreboom M, Snik AFM, Mylanus EAM (2012a) Sequential Bilateral Cochlear Implantation in Children Quality of Life. Archives of Otolaryngology-Head & Neck Surgery 138:134-141. Sparreboom M, Leeuw AR, Snik AFM, Mylanus EAM (2012b) Sequential bilateral cochlear implantation in children: Parentsâ&#x20AC;&#x2122; Perspective and device use. International Journal of Pediatric Otorhinolaryngology 76:339-344. Steffens T, Lesinski-Schiedat A, Strutz J, Aschendorff A, Klenzner T, Ruhl S, Voss B, Wesarg T, Laszig R, Lenarz T (2008) The benefits of sequential bilateral cochlear implantation for hearing-impaired children. Acta Otolaryngologica 128:164-176. Strom-Roum H, Laurent C, Wie OB (2012a) Comparison of bilateral and unilateral cochlear implants in children with sequential surgery. International Journal of Pediatric Otorhinolaryngology 76:95-99. Strom-Roum H, Rodvik AK, Osnes TA, Fagerland MW, Wie OB (2012b) Sound localising ability in children with bilateral sequential cochlear implants. International Journal of Pediatric Otorhinolaryngology 76:12451248. Sun L (2010) Early childhood general anaesthesia exposure and neurocognitive development. British Journal of Anaesthesia 105 Suppl 1:i61-68. Van Deun L, van Wieringen A, Scherf F, Deggouj N, Desloovere C, Offeciers FE, Van de Heyning PH, Dhooge IJ, Wouters J (2010) Earlier Intervention Leads to Better Sound Localization in Children with Bilateral Cochlear Implants. Audiology and Neurotology 15:7-17. Vincent C, Bebear JP, Radafy E, Vaneecloo FM, Ruzza I, Lautissier S, Bordure P (2012) Bilateral cochlear implantation in children: Localization and hearing in noise benefits. International Journal of Pediatric Otorhinolaryngology 76:858-864. Wolfe J, Baker S, Caraway T, Kasulis H, Mears A, Smith J, Swim L, Wood M (2007) 1-year postactivation results for sequentially implanted bilateral cochlear implant users. Otology & Neurotology 28:589-596. Zeitler DM, Kessler MA, Terushkin V, Roland JT, Svirsky MA, Lalwani AK, Waltzman SB (2008) Speech perception benefits of sequential bilateral cochlear implantation in children and adults: A retrospective analysis. Otology & Neurotology 29:314-325.
80
Part II Prelingual deafness in adult cochlear implant users
Chapter 2.1 Predicting performance and non-use in prelingually deaf and late-implanted cochlear implant users
M.J.W. Lammers, H. Versnel, V. Topsakal, G.A. van Zanten, W. Grolman
Submitted
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Objectives: Cochlear implantation in adults with prelingual deafness has been found to be far less favorable than in postlingually deaf adults. The large variability in outcomes and rather large number of non-users in the prelingual group remains a concern. The aim of this study was to identify independent predictors of postoperative performance and to characterize the specific subpopulation that has no or little benefit from their implants and eventually become non-users. Design: We performed a retrospective cohort study at the University Medical Center of Utrecht. A total of 48 prelingually deaf patients diagnosed with severe to profound sensorineural hearing loss before the age of 2 years and received their implants during adulthood were included in this study. Univariate and multivariate regression analyses were performed to identify variables which may predict postoperative speech recognition. Results: The average maximum postoperative monosyllabic word score for all subjects was 25% correct. According to univariate analyses, postoperative speech perception was significantly correlated to five variables. Multivariate analyses indicated that preoperative speech perception and the extent of residual hearing in the past were both positive independent predictors of postoperative speech recognition, explaining 47% of the total variance. Almost half of the population experienced only little (27%) or no benefit (21%) of their implants in daily communication. These patients were more likely to have negligible residual hearing and to use sign language. Eventually, twenty-one percent of subjects became non-users. Conclusions: Postoperative performance of adult, prelingually deaf CI candidates, with long-term deafness is dependent on their preoperative speech perception and residual hearing. Candidates who primarily rely on sign language and who have negligible residual hearing are expected to have no or only limited postoperative benefit from their implants and are at risk of becoming non-users.
84
Performance and non-use in prelingually deaf CI users
Introduction Cochlear implantation in adults with postlingual deafness has proven to be a successful treatment leading to a significant improvement in speech recognition and communication. In contrast to these positive results in the postlingually deaf, are the far less favorable results in prelingually deaf and late-implanted adults (Lammers et al. 2015; Teoh et al. 2004). These differences can be explained by the duration and extent of auditory deprivation in childhood. A recent study by our group revealed that prelingually deaf and late-implanted CI users demonstrate an impaired activation of the auditory cortex, probably caused by little or no auditory stimulation during childhood which is necessary for the maturation of the auditory pathway (Lammers et al. 2015). After years of deprivation, this more innate auditory network lacks the ability to develop into a normal mature network using the new auditory input. Stimulation of this immature pathway in conjunction with a certain degree of cross-modal reorganization (Fine et al. 2005; Finney et al. 2001; Lee et al. 2007), lead to poorer postoperative performance compared to those recipients who have experienced a (near to) normal development of their auditory pathway during childhood (Lammers et al. 2015). The extent of these cortical alterations which take place in the prelingually deaf child, may underlie the generally poor postoperative open-set speech perception skills when these patients receive their first implants only during adulthood. While some of them can obtain fairly good speech perception scores, a high percentage of recipients has little or no benefit from their implants if no visual cues are available (Straatman et al. 2014; Teoh et al. 2004). The majority of these patients, however do report that their CI improves their speech understanding if also visual cues are available (Chee et al. 2004) and that their CI improves their communication skills (Chee et al. 2004; Kaplan et al. 2003; Klop et al. 2007; Skinner et al. 1992; Teoh et al. 2004; Waltzman et al. 1992; Zwolan et al. 1996). Furthermore, in most of them cochlear implantation leads to an increase in both generic and disease specific quality of life (Klop et al. 2007; Straatman et al. 2014), suggesting the importance of their CI in daily life even for patients who only use it for sound detection. Considering the large postoperative variability in outcomes, previous studies have tried to identify prognostic criteria. Differences between inclusion criteria of the various studies and the relative small study populations made that previously found prognostic factors such as age at implantation, i.e. adolescent or adult (Schramm et al. 2002), preoperative speech perception (van Dijkhuizen et al. 2011), progressive hearing loss (Caposecco et al. 2012) or time without hearing aid in the implanted ear (Caposecco et al. 2012) have not been consistently confirmed in other studies. In terms of speech production capabilities and language skills more consistent results have been found. Two studies from the same center identified speech intelligibility as a positive predictor for postoperative functioning (Klop et al. 2007; van Dijkhuizen et al. 2011).
85
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Elaborating on the influence and the extent of possible cross-modal changes during childhood, the effect of sign language on postoperative functioning has been a subject for debate. Imaging studies in deaf signers indicated reorganization of their auditory cortex (Fine et al. 2005; Sadato et al. 2004). This cross-modal reorganization can have a negative effect on auditory rehabilitation if cochlear implants are provided when certain structural changes in the brain cannot be reversed anymore. Various studies already recognized the possible negative role of sign language on performance after cochlear implantation in late-implanted prelingually deaf adults (Caposecco et al. 2012; Kaplan et al. 2003; Watson et al. 2003; Yang et al. 2011), suggesting irreversible effects of cross-modal reorganization. Since a few early studies on late-implanted prelingually deaf adults revealed that they could not obtain openset speech recognition and could only use their implants for sound detection (Skinner et al. 1992; Waltzman et al. 1992; Zwolan et al. 1996), few of these specific patients were considered as good CI candidates, and actually received a CI. For this reason, the number of reports on this specific study population is still limited. Although the aforementioned reports state that, despite poor open-set speech recognition, most patients have benefit from their implants, even less is known about the poorly performing subjects who eventually decide not to use their devices. The clinical experience of the cochlear implant team of the University Medical Center of Utrecht, over the last 30 years suggests that the rate of non-users in this late-implanted prelingually deaf group is much higher than among postlingually deaf CI users, which is around 1.7%. In a clinical perspective to correctly counsel our patients, the objective of this study was to identify preoperative prognostic factors of postoperative speech recognition of prelingually deaf CI users, with long-term early-onset deafness. Specifically, we aimed to characterize the non-users in this population and explore if certain preoperative factors could be identified which could predispose for future device non-use.
Materials and methods Design and study population We performed a retrospective cohort study, including all prelingually deaf CI users, who received their implants at the age of 18 years or older at the University Medical Center of Utrecht (UMC Utrecht) between January 2000 and July 2013. Prelingually deaf subjects were selected based on the following criteria: severe to profound, binaural hearing loss with its onset before the age of 2 years (based on patient information, medical histories and diagnostic audiometry results) and no or insufficient residual hearing during childhood for normal speech and language development. The diagnosis of prelingual deafness was confirmed by the evaluation of the multidisciplinary CI team prior to cochlear implant surgery. Before
86
Performance and non-use in prelingually deaf CI users
cochlear implantation was indicated, extensive hearing aid trials were initiated in each patient to substantiate the impossibility to habilitate their prelingual deafness with hearing aids. The duration of these trials are dependent on the effect of the hearing aid adjustments, and usually takes several months. The study was performed in accordance with the Declaration of Helsinki. Because this is a retrospective study with anonymized data, exemption for a full review from the Local Research Ethics Committee was obtained (WAG/om/15/021600). Data extraction and analysis For each patient we gathered information on the age and cause of deafness, age at first hearing aid fitting, age until hearing aids were used in the implanted ear, communication mode, pre-operative expectations of the cochlear implant team regarding level of postoperative functioning, the presence of some residual hearing on all available audiograms, age at implantation, type of cochlear implant, eventual non-use of implant and the cause of non-use. This information was mainly provided by reviewing the complete medical charts in combination with self-reported information by the patient. For the purpose of data analyses, preoperative communication mode was classified in the following four categories: 1. primarily sign language, 2. sign language and secondary lip reading, 3. oral, lip reading and sign language, 4. oral and lip reading. Preoperative expectations of the cochlear implant team were based on age at deafness, etiology, duration of deafness, hearing aid use, communication skills, patients motivation, effect of preoperative hearing aid trials, and speech recognition tests with auditory only (A), visual only (V), auditory and visual (AV) cues; they were classified in the following 3 categories: detection only, support in speech understanding, and good. Speech perception and benefit from implant Pre- and postoperative speech perception scores were obtained using the Dutch Society of Audiology standard consonant-vowel-consonant (CVC) word list at 65 dB SPL (Versfeld et al. 2000). In this open-set test, speech perception was scored as the percentage of phonemes correct. At the pre- and postoperative evaluations this test was performed in three conditions: A, V, and AV. At the postoperative evaluations after 1, 3, 12, 24, 36, 48 and 60 months after implantation only the scores obtained in the condition with auditory cues was used for our data analyses. Sentence tests were obtained using standardized Dutch sentence lists at 65 dB SPL and scored as the percentage of syllables correct (Versfeld et al. 2000). This test was administered with auditory cues only, visual cues only and both auditory and visual cues. For this study we only used the scores obtained at the preoperative evaluation. Based on the preoperative word and sentence recognition tests performed in the condition with visual cues only (V condition) and with both auditory and visual cues (AV condition), the support of auditory
87
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
cues in communication can be assessed. By subtracting the word or sentences scores obtained in the AV condition from the V condition, an estimate of the support of auditory cues can be modelled, even if patients could not obtain open-set speech perception with auditory cues only. This benefit from auditory cues in word recognitions is coded as AV-V words and for sentence recognition as AV-V sentences. The preoperative support of auditory cues in communication is also evaluated by the speech- and language therapists, based on the results of the aforementioned tests and on observations during communication with the patient. For the purpose of data analyses this benefit from auditory cues as assessed by the speech- and language therapist is classified in the following three categories: good, little, or no benefit from auditory cues. At each of the postoperative evaluation moments it was documented if the patient reported subjective benefit from their implant. The benefit from the implant in daily life was classified as either good (support in daily communication), little (support in sound signaling only) or no benefit. Classification of device non-use was based on the joined decision of patient and cochlear implant team to stop using the implant.
Statistical analyses Statistical analyses were conducted using SPSS version 22.0 software (IBM, Armonk, NY, USA). Mean or median differences in CVC scores were calculated with dependent two-tailed t-tests or Wilcoxon signed rank tests for dependent samples. Differences in the number of events in different subgroups were compared using Fisherâ&#x20AC;&#x2122;s exact tests. Linear regression analyses were performed to assess the relation between postoperative speech recognition and the various continuous predictors such as age at implantation and preoperative CVC scores. The categorical variables like residual hearing, communication mode and expectations of the cochlear implant team were first analyzed for their relation with postoperative speech perception by means of Kruskall-Wallis tests for independent samples. The significant variables were then converted to dummy variables before they could be included in univariate linear regression analyses. Pearson correlations were used for normally distributed data and Spearman correlations for non-normative data. Correlation coefficients of <0.3 were considered weak, between 0.3 and 0.5 moderately strong, and > 0.5 strong (Cohen 2003). Multiple linear regression models with stepwise backward elimination were constructed to analyze independent factors related to postoperative speech recognition. Only variables which were significant at the p level < 0.05 in the univariate analyses were included in the multiple linear regression models.
88
Performance and non-use in prelingually deaf CI users
Results
Patient characteristics and speech perception In the period January 2000 â&#x20AC;&#x201C; July 2013 a total number of 48 patients met the inclusion criteria and thus were included in this study. Mean age at implantation was 41 years (range 18 to 65 years) and the median age at deafness was 0 years (range 0 to 2 years). Medical charts revealed that all included subjects had hearing thresholds of 70 dB HL or worse at any of their audiograms obtained during life. Almost all subjects showed some measurable preoperative low-frequency residual hearing, with best thresholds at or worse than 70 dB HL in the low frequencies of 125 â&#x20AC;&#x201C; 1000 Hz (corner audiogram). Eight subjects displayed residual hearing with best thresholds of 70 dB HL over more than only the low-frequencies (Table 1). The extent of residual hearing was in all cases insufficient for adequate auditory and language development during childhood. More detailed patient characteristics are presented in Table 1. Average of the individualâ&#x20AC;&#x2122;s maximum postoperative CVC scores for all subjects was 25% correct (SD: 24%) (Figure 1) and was significantly better than their preoperative scores with hearing aids (mean: 8% correct, median: 0% correct; Wilcoxon signed rank test t = 12.500, p < 0001). Postoperative speech recognition improved especially over the first year (mean: 23% correct, SD: 22%), to reach a relative stable plateau phase afterwards. Only ten subjects (21%) displayed a further improvement of at least 10% on their CVC word scores after the first year (Figure 1). Ten subjects decided to stop using their implants after the first year and they were classified as non-users. These ten subjects (21%) reported not to have any benefit from their implants, whereas 13 patients (27%) did have some support from their implant in sound detection and the remaining 52% (25 subjects) experienced substantial benefit from their cochlear implants in daily communication.
Fig. 1. Postoperative CVC phoneme scores over time following cochlear implantation (n=48). The thin light grey lines represent individual trajectories. The thick red line indicates the mean scores.
89
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table 1. Patient characteristics and univariate analyses
Characteristics (number of subjects) Age at implantation (n=48) Age at deafness (n=48) Age at first hearing aid (n=35) Age until hearing aid use (n=48) Etiology (n=48) Congenital/unknown Meningitis Hypoxemia Infectious (Rubella/Varicella) Usher Icterus Rhesus antagonism Prematurity Waardenburg syndrome Expectations of CI team (n=48) Detection Support speech understanding Good Residual hearing (n=46) No Low-frequency only Yes Mode of communication (n=48) Sign language Sign language, lip reading Oral, lip reading, sign language Oral, lip reading Support auditory cues (n=45) No benefit Little Good Benefit of CI (n=48) No benefit Little Good Device use (n=48) Yes No Preoperative scores CVC scores, auditory cues only (n=43) Support of auditory cues (AV-V words) (n=37) Support of auditory cues (AV-V sentences) (n=39) Postoperative maximum CVC scores 3 months after implantation (n=39) 1 year after implantation (n=35) 2 years after implantation (n=33) 3 years after implantation (n=26) 4 years after implantation (n=16) 5 years after implantation (n=11)
Mean/median or number (%) 41 0 2 38 23 (48%) 6 (13%) 4 (8%) 7/1 (17%) 1 (2%) 2 (4%) 2 (4%) 1 (2%) 1 (2%) 26 (54%) 16 (33%) 6 (13%) 6 (13%) 32 (70%) 8 (17%) 4 (8%) 12 (25%) 18 (38%) 14 (29%) 4 (9%) 30 (67%) 11 (24%)
Univariate predictors of maximum CVC score Test statistic
P-value
r = 0.849 r = -0.076 r = 0.098 r = 0.204 H = 4.533
NS NS NS NS NS
H = 24.465
< 0.0001
H = 15.776
< 0.0001
H = 6.849
= 0.077
H = 14.870
= 0.001
r = 0.657 r = 0.438 r = 0.228
< 0.0001 = 0.007 NS
10 (21%) 13 (27%) 25 (52%) 31 (79%) 10 (21%) 0 15 20 25 21 23 25 24 23 34
Univariate preoperative factors related to speech perception Univariate linear regression analyses including all preoperative factors described in the methods section are presented in Table 1. Expectations of the cochlear implant team appeared to be the strongest positive predictor (H = 24.465, p < 0.0001), when entered in the linear regression model this factor accounted for 59% of the total variance (r2 = 0.588,
90
Performance and non-use in prelingually deaf CI users
p < 0.0001). The support of auditory cues in communication as assessed by the speech and language therapist (H = 14.870, p = 0.001; r2 = 0.372, p < 0.0001), the preoperative CVC word scores (r2 = 0.432, p < 0.0001) and the amount of residual hearing (H = 15.776, p < 0.0001; r2 = 0.354, p < 0.0001) had a strong positive correlation with postoperative speech perception. The difference between preoperative word recognition with auditory and visual cues and visual cues only (AV-V words) was moderately associated with postoperative speech perception (r = 0.478, p = 0.001), as was the preoperative support of auditory cues in sentence recognition (AV-V sentences: r = 0.334, p = 0.021). The association between postoperative speech perception and mode of communication was weak and not significant (H = 6.849, r2 = 0.130, p = 0.077). Postoperative speech perception was not related to patient characteristics such as age at implantation and etiology of hearing loss (Table 1). Multivariate model predicting speech perception Backward multivariate analyses revealed that the expectations of the cochlear implant team, the pre-operative CVC scores and the presence or absence of some residual hearing remained significant independent predictors of postoperative speech recognition, explaining 69% of the total variance (Figure 2). Speech recognition increased with increase of any of these three predictors. The three factors are related to each other since preoperative CVC scores will be higher with more residual hearing, and the CI teamâ&#x20AC;&#x2122;s expectations are based, among others, on the other two factors (see Methods). Consequently, there was collinearity between preoperative CVC score and the expectations of the CI team. Furthermore, collinearity was found between the preoperative CVC scores and the support of auditory cues as assessed by the speech- and language therapists (confirmed with Kruskall-Wallis test: H = 11.462, P = 0.003). Since preoperative CVC score was a stronger predictor (r2 = 0.432), and a more objective and standardized measure, the assessment of the speech and language therapists (r2 = 0.372) and expectations of the CI team (r2 = 0.588) were omitted in the final model. The final model, including the factors preoperative CVC word scores, residual hearing and AV-V words, identified that the preoperative CVC word scores (r2 = 0.333) and the extent of residual hearing (r2 = 0.136) were both independent variables explaining for 47% of the total variance (r = 0.685, F = 8.242, p < 0.001) (Figure 2b). In this model there was no collinearity between the factors. Based on this model, one can expect that prelingually deafened patients with no or little residual hearing and poor preoperative CVC word scores are least likely to achieve good postoperative speech perception.
91
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
A
B
Fig. 2. Results of the multivariate regression analyses displaying maximum postoperative CVC phoneme scores as a function of preoperative expectations of the CI team and residual hearing (A). Results of the final multivariate regression analyses demonstrating the relation between postoperative phoneme scores as a function of preoperative CVC phoneme scores and residual hearing (B). Red squares indicate patients with no preoperative residual hearing; blue diamonds denote patients with only low-frequency residual hearing and green filled circles patients with more residual hearing. Random jitter has been added to avoid overlapping data points.
Characterizing patients with no or little benefit from cochlear implantation Ten of the 48 subjects (21%) decided to stop using their implants and were classified as non-user. All of them experienced no benefit in daily communication or sound detection and 60% also reported pain around the implant. Etiology was congenital or unknown (n = 8), Meningitis (n =1) or Waardenburg syndrome (n =1). In all cases surgery was uneventful, with full insertion of the electrode array. Due to the poor functioning with their implants, in four of these non-users postoperative CT-scans were performed, which demonstrated a normal position of the electrode array within the cochlea. Eventually, in four non-users the CI has been explanted, which reduced symptoms as pain. Two non-users only used sign language, three others primarily used sign language and secondarily lip reading. The remaining five subjects did also use oral communication and lip reading with or without sign language preoperatively. In nine of the ten non-users the cochlear implant team already expected that they could only use their CI for sound detection. In one patient support in speech understanding had been expected. Â Almost half of the 48 subjects (48%) reported either no (21%, discussed above) or only little benefit (27%) of their implants. After the preoperative evaluations it was expected that these patients would either perform on detection level or have only limited support in speech and language perception (Figure 3). The preoperative expectations in this group were significantly worse than in the group who reported a significant benefit from their implants (Fisherâ&#x20AC;&#x2122;s exact test = 15.47, p < 0.001).
92
Performance and non-use in prelingually deaf CI users
Fig. 3. Postoperative benefit from cochlear implantation as preoperatively expected by the CI team. Red squares indicate patients with no benefit; blue filled circles denote patients with only limited benefit from their implants and green diamonds patients experiencing substantial benefit. Random jitter has been added to avoid overlapping data points.
The patients who experienced no or little benefit were more likely to have either no or only some low-frequency residual hearing, in contrast to the more successful patients, who over the years had either low-frequency residual hearing or low- and high-frequency residual hearing (Fisher’s exact test = 10.52, p = 0.004) (Figure 4a). Four of the patients who had either no or little benefit primarily used sign language, compared to none of the patients who reported benefit from their implants (Figure 4b). Two of these signers became non-users and the other two signers only used their implants for sound detection. Fisher’s exact test did not display a significant association between mode of communication and benefit from the implant if all four categories of the mode of communication were included in the analysis (Fisher’s exact test = 4.84, p = 0.208). Other preoperative factors were not significantly different either between the successful patients and those who had little or no benefit from their implants.
93
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
A
B
Fig. 4. Postoperative benefit from cochlear implantation as a function of residual hearing (A) and communication mode (B). Red squares indicate patients with no benefit; blue filled circles denote patients with only limited benefit from their implants and green diamonds patients experiencing substantial benefit. Random jitter has been added to avoid overlapping data points.
Discussion This study shows that half (52%) of prelingually deaf and late-implanted adult CI users have significant benefit from their implants in daily communication and that their postoperative functioning in terms of open-set speech recognition can be mainly ascribed to their preoperative residual hearing and speech perception. Another 27% mainly uses their implants for sound detection. The remaining 21% stopped using their implants since they experienced no benefit from their CIs in daily life. The level of postoperative performance can be fairly well predicted by experienced cochlear implant teams, already explaining more than half of the total variance. Multivariate regression analyses identified that preoperative speech recognition scores with auditory cues only and the extent of residual hearing are significant independent predictors of postoperative speech perception in this particular group of patients. If patients had no or only low-frequency residual hearing they were more likely to have less benefit from their implants and even become non-users. Although event rates were too low to perform statistical analyses it was striking that the only four subjects who primarily used sign language, eventually had no or only little benefit from their implant. Open-set speech recognition scores found in our study population are comparable to those found in other reports. Average monosyllabic phoneme score of the 48 subjects at 12 months was 23% correct and is comparable to an average phoneme score of 31% correct after one year of implant use in 28 patients studied by Straatman et al. (2014), and monosyllabic word scores of around 30% in other studies (Caposecco et al. 2012; Teoh et al. 2004; Yang
94
Performance and non-use in prelingually deaf CI users
et al. 2011). Klop et al. (2007) and Van Dijkhuizen et al. (2011) reported phoneme scores of about 45%. These better scores found in these studies might be caused by differences in study population regarding preoperative residual hearing. Mean preoperative phoneme score in our study was 8% correct, whereas Klop et al. and van Dijkhuizen et al. studied patients who had average preoperative phoneme scores of 14% to 17% correct. After 12 months of implant use asymptotic levels of speech recognition were reached (Figure 1). This is in line with the findings of other reports which described that a plateau level is reached after 6 to 12 months (Caposecco et al. 2012; Teoh et al. 2004; Yang et al. 2011). So far, reports on device non-use in prelingually deaf, late-implanted adults was lacking. In our study population 21% decided to stop using their implants. This number is considerably higher than the non-user rate of only 1.7% in postlingually deaf CI users in our center during the same period (2000 â&#x20AC;&#x201C; July 2013). Due to the relative small study population and low event rate, only a descriptive presentation of these results in the non-users could be provided, but this indicated that the preoperative expectations in most of these non-users were poor and most likely had little or no residual hearing during childhood (Figure 4a). Furthermore, the four patients who primarily used sign language had either no or little benefit from their implants. This is comparable to the six subjects in the study by Yang et al. (2011) who primarily used sign language and only obtained an average postoperative word recognition score of 1.3% correct. As in their study, speech perception scores increased if patients already used more oral communication skills prior to cochlear implantation. This negative relationship between sign language and postoperative performance in patients with long periods of auditory deprivation has also been described by other reports (Klop et al. 2007; Osberger et al. 1998; Waltzman et al. 2002). Study populations have so far been too small and outcomes too variable for rigorous multivariate statistical analyses to clearly identify a negative effect of sign language in conjunction with long periods of auditory deprivation on postoperative performance. By combining the four signers in our study with the results of the six signers in the study by Yang et al. 2011 it is highly remarkable that all these subjects could not obtain any speech recognition after implantation, indicating more strongly a possible negative effect of sign language on postoperative performance after cochlear implantation. This effect might be explained by extensive cortical reorganization induced by the use of sign language in conjunction with long durations of auditory deprivation. This insight is supported by imaging studies which demonstrated that motion stimuli resulted in activity in the auditory cortex of deaf signers, whereas no activity was present in normal hearing signers, indicating the causality of auditory deprivation and cross-modal reorganization in the auditory cortex (Fine et al. 2005; Sadato et al. 2004). Results from a recent study by our group evaluating cortical auditory evoked potentials in prelingual subjects, revealed that these CI users displayed larger responses with early N1 peaks suggesting activation of a more innate and less complex auditory pathway (Lammers et al. 2015). The combination of stimulation of
95
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
an innate and less complex auditory pathway and a high level of cross-modal reorganization in prelingually deaf and late-implanted CI users might explain their inferior results and higher likelihood to obtain little or even no benefit from their implants. While cochlear implantation still does aid most prelingually deaf subjects it is in the interest for both patient and cochlear implant team to make a preoperative assessment of the expectations which is as accurate as possible. By combining the prognostic factors found in this study, i.e. residual hearing, preoperative level of speech perception and the predominant use of sign language, patient expectations can be more accurately addressed and it can even provide stronger arguments to advise against cochlear implantation in prelingually deaf signers with no residual hearing. Methodological considerations Due to the retrospective nature of this study and the relative small study population there are several potential limitations which should be considered. First, although this study population is one of the larger and most homogenous (including only patients who became functionally deaf before the age of 2 years) presented in the literature, the population and number of events is still relative small. This has a significant effect on the power of statistical analyses. For multivariate regression analyses with a continuous outcome measure as used in this study (postoperative monosyllabic phoneme scores) a minimum of 10 subjects are required for each variable (Peduzzi et al. 1996). Due to the small number of subjects (n=48) not more than four variables were included in our final multivariate analysis. Preferably, one would include a larger sample, but since our indication criteria for prelingual deafness were rather strict (age at deafness <2 years) and a fair number of prelingual candidates declined cochlear implantation after assessment and counseling by the cochlear implant team, larger sample sizes are difficult to obtain and as a consequence have also not been published previously. Although the statistical power of our multivariate model is therefore limited and these results should be interpreted with caution, this is the first multivariate analysis performed in this specific patient population and sheds light on some likely independent predictors which influence postoperative performance. For our analysis of cochlear implant non-use, the number of non-users and event rates where too small for multiple logistic regression analyses to be feasible and only descriptive results of this group could be presented. Second, incompleteness of data resulted in the fact that for some subjects no clear information could be extracted on their preoperative hearing aid use. Recall bias could have influenced the patients statement regarding their first hearing aids, benefit from their hearing aids during childhood and age at which they stopped using their hearing aids. Since in our center in most cases prior to cochlear implantation a thorough hearing aid trial is initiated to assess whether optimization of acoustic amplification could increase their support of their hearing aids, most subjects used hearing aids until implantation. Prior to this final hearing aid trial, most patients have used hearing aids for some period. From reviewing their medical
96
Performance and non-use in prelingually deaf CI users
charts it could not be identified in all patients how long this period lasted. Therefore, the age at which use of hearing aids was ended (see Table 1) could have been underestimated. Â Third, considering the poor results found in prelingually deaf signers who have had long durations of profound hearing loss with no or hardly any residual hearing, cochlear implantation is frequently discouraged by the implant team. Only if some benefit from an implant is expected and patients are highly motivated, cochlear implantation is considered. For this reason numerous prelingually deaf patients did not receive a CI and consequently could not be included in this study. This may have resulted in a relative underestimation of the negative prognostic effects.
Conclusions Although the majority of prelingually deaf, late-implanted adult CI users have benefit from their implants in daily life, still 21% became non-users due to insufficient benefit. Multivariate regression analyses identified that preoperative speech perception and residual hearing are significant predictors of postoperative open-set speech recognition. The non-users and patients with only little benefit from their implants appeared to have at most low-frequency residual hearing prior to implantation and were more likely to use sign language as their primary mode of communication. By combining the preoperative characteristics like preoperative speech perception, the extent of residual hearing and the primary mode of communication a fairly good prediction of postoperative performance can be given. These results are therefore not only helpful for counseling individual patients on their expectations after cochlear implantation, but can also be used for identifying patients who are expected to experience little or no benefit from an implant.
Acknowledgments The authors gratefully acknowledge all current and previous members of the cochlear implant team of the University Medical Center of Utrecht for their contributions and dedicated work over the years. Special thanks to Veronique Kraaijenga for her work on the patient database and Inge Wegner for her advice on the statistical analyses. This study is supported by an unrestrictive research grant from Cochlear Ltd.
97
2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
References Caposecco A, Hickson L, Pedley K (2012). Cochlear implant outcomes in adults and adolescents with earlyonset hearing loss. Ear and Hearing 33:209-220. Chee GH, Goldring J E, Shipp DB, Ng AH, Chen JM, Nedzelski JM (2004). Benefits of cochlear implantation in early-deafened adults: the Toronto experience. Journal of Otolaryngology - Head & Neck Surgery 33:26-31. Cohen J. (2003). Applied Multiple Regression/correlation Analysis for the Behavioral Sciences. L. Erlbaum Associates. Fine I, Finney EM, Boynton GM, Dobkins KR (2005). Comparing the effects of auditory deprivation and sign language within the auditory and visual cortex. Journal of Cognitive Neuroscience 17:1621-1637. Finney EM, Fine I, Dobkins KR (2001). Visual stimuli activate auditory cortex in the deaf. Nature Neuroscience 4:1171-1173. Kaplan DM, Shipp DB, Chen JM, Ng AH, Nedzelski JM (2003). Early-deafened adult cochlear implant users: assessment of outcomes. Journal of Otolaryngology - Head & Neck Surgery 32:245-249. Klop WM, Briaire JJ, Stiggelbout AM, Frijns JH (2007). Cochlear implant outcomes and quality of life in adults with prelingual deafness. Laryngoscope 117:1982-1987. Lammers MJ, Versnel H, van Zanten GA, Grolman W (2015). Altered cortical activity in prelingually deafened cochlear implant users following long periods of auditory deprivation. Journal of the Association for Research in Otolaryngology 16:159-170. Lee HJ, Giraud AL, Kang E, Oh SH, Kang H, Kim CS, Lee DS (2007). Cortical activity at rest predicts cochlear implantation outcome. Cerebral Cortex 17:909-917. Osberger MJ, Fisher L, Zimmerman-Phillips S, Geier L, Barker MJ (1998). Speech recognition performance of older children with cochlear implants. American Journal of Otolaryngology 19:152-157. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR (1996). A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology 49:1373-1379. Sadato N, Yamada H, Okada T, Yoshida M, Hasegawa T, Matsuki K, Yonekura Y, Itoh H (2004). Age-dependent plasticity in the superior temporal sulcus in deaf humans: a functional MRI study. BMC Neuroscience 5:56. Schramm D, Fitzpatrick E, Seguin C (2002). Cochlear implantation for adolescents and adults with prelinguistic deafness. Otology & Neurotology 23:698-703. Skinner MW, Binzer SM, Fears BT, Holden TA, Jenison VW, Nettles EJ (1992). Study of the performance of four prelinguistically or perilinguistically deaf patients with a multi-electrode, intracochlear implant. Laryngoscope 102:797-806. Straatman LV, Huinck WJ, Langereis MC, Snik AF, Mulder JJ (2014). Cochlear implantation in late-implanted prelingually deafened adults: changes in quality of life. Otology & Neurotology 35:253-259. Teoh SW, Pisoni DB, Miyamoto RT (2004). Cochlear implantation in adults with prelingual deafness. Part I. Clinical results. Laryngoscope 114:1536-1540. van Dijkhuizen JN, Beers M, Boermans PP, Briaire JJ, Frijns JH (2011). Speech intelligibility as a predictor of cochlear implant outcome in prelingually deafened adults. Ear and Hearing 32:445-458. Versfeld NJ, Daalder L, Festen JM, Houtgast T (2000). Method for the selection of sentence materials for efficient measurement of the speech reception threshold. Journal of the Acoustical Society of America 107:1671-1684. Waltzman SB, Cohen NL, Shapiro WH (1992). Use of a multichannel cochlear implant in the congenitally and prelingually deaf population. Laryngoscope 102:395-399. Waltzman SB, Roland JT Jr, Cohen NL (2002). Delayed implantation in congenitally deaf children and adults. Otology & Neurotology 23:333-340.
98
Performance and non-use in prelingually deaf CI users
Watson SD, Balko KA, Comer LK, Bishop RD, Reilley D, Backous DD (2003). Benefits of cochlear implantation in pre-lingual adult users: oral and manual communicators. Cochlear Implants International 4 Suppl 1:75-76. Yang WS, Moon IS, Kim HN, Lee WS, Lee SE, Choi JY (2011). Delayed cochlear implantation in adults with prelingual severe-to-profound hearing loss. Otology & Neurotology 32:223-228. Zwolan TA, Kileny PR, Telian SA (1996). Self-report of cochlear implant use and satisfaction by prelingually deafened adults. Ear and Hearing 17:198-210.
2.1
99
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2 Delayed Auditory Brainstem Responses in Prelingually Deaf and Late-Implanted Cochlear Implant Users
M.J.W. Lammers, R.H.M. van Eijl, G.A. van Zanten, H. Versnel, W. Grolman
Journal of the Association for Research in Otolaryngology 2015; 16(5):669-678
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Neurophysiological studies in animals and humans suggest that severe hearing loss during early development impairs the maturation of the auditory brainstem. To date, studies in humans have mainly focused on the neural activation of the auditory brainstem in children treated with a cochlear implant (CI), but little is known about the pattern of activation in adult CI users with early onset of deafness (prelingual, before the age of 2 years). In this study we compare auditory brainstem activation in prelingually deaf and late-implanted adult CI users to that in postlingually deaf CI users. Electrically evoked auditory brainstem responses (eABRs) were recorded by monopolar stimulation, separately using a middle and an apical electrode of the CI. Comparison of the eABR latencies revealed that wave V was significantly delayed in the prelingually deaf CI users on both electrode locations. Accordingly, when the apical electrode was stimulated, the III-V interwave interval was significantly longer in the prelingually deaf group. These findings suggest a slower neural conduction in the auditory brainstem, probably caused by impairment of maturation during the long duration of severe hearing loss in infancy. Shorter wave V latencies, reflecting a more mature brainstem, appeared to be a predictor for better speech perception.
102
eABRs in prelingually deaf CI users
Introduction The duration of auditory deprivation before cochlear implantation is an important predictor for hearing with a cochlear implant (CI). Whereas patients with late onset of deafness can obtain good speech perception in quiet, the hearing performance of patients with longterm early-onset deafness is generally poor (Teoh et al. 2004; Lammers et al. 2015). The large difference might be caused by an impaired development of their auditory pathway in combination with cross-modal changes during a prolonged period of auditory deprivation (Doucet et al. 2006; Lee et al. 2007; Kral and Oâ&#x20AC;&#x2122;Donoghue 2010; Kral and Sharma 2012; Lammers et al. 2015). Recently, we demonstrated that prelingually deaf and late-implanted CI users display relatively early and large N1 peaks of the cortical auditory evoked potential (Lammers et al. 2015). This altered cortical activity raises the question regarding the extent to which the subcortical pathway, particularly the auditory brainstem, is affected in prelingually deaf CI users. It is well documented that after birth, auditory brainstem response (ABR) wave latencies decrease and reach adult levels around the age of 2-3 years (Inagaki et al. 1987; Eggermont and Salamy 1988). This decrease is slower for wave V than for early waves and is hypothesized to result from increasing myelination and/or synaptic efficacy within the auditory brainstem since these developments lead to faster axonal conduction and synaptic transmission (Eggermont and Salamy 1988; Moore et al. 1995; Thai-Van et al. 2007). Long periods of deafness affect the subcortical pathway, resulting in gradual spiral ganglion cell degeneration (Spoendlin 1975; Versnel et al. 2007), and a volume reduction of the cochlear nucleus and its cells (Moore 1990; Leake et al. 2008; Ryugo et al. 2010). On the other hand, electrically evoked ABRs (eABRs) in congenitally deaf cats, demonstrate latencies decreasing with age, similarly to normal-hearing cats (Tillein et al. 2012). This suggests that auditory brainstem structures and pathways develop even in the absence of auditory stimulation. In humans, development of the auditory brainstem following deafness has been studied by recording eABRs in children with CI (Gordon et al. 2006, 2008; Thai-Van et al. 2007; Sparreboom et al. 2010). These studies demonstrated that in children with early onset deafness eABR wave latencies decrease after implantation, irrespective of age at implantation, like they do in normal-hearing children. On the contrary, in bilaterally implanted children, when a response is evoked using a second CI implanted much later than the first, the wave V latency is longer than the responses evoked by the first CI (Gordon et al. 2008; Sparreboom et al. 2010). This suggests impaired maturation of the auditory brainstem of the later implanted ear (Gordon et al. 2008; Sparreboom et al. 2010), or altered neuronal connections induced by the period of unilateral hearing with the first CI. Whereas above-mentioned studies were performed in children and in animal models, we address the effect of early deafness on the auditory brainstem in adults. We compare
103
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
eABRs in prelingually deaf subjects who had little or no auditory stimulation for more than 20 years to eABRs in postlingually deaf CI users. According to various studies (Eggermont and Salamy, 1988; Moore et al. 1995; Thai-Van et al. 2007; Leake et al. 2008; Ryugo et al. 2010), myelination and synapses in the brainstem of the postlingual group should have developed normally because of sufficient auditory input during childhood. In contrast, we expect the coarse structures of the brainstem to develop in the prelingual group but sensory-driven maturation to be impaired reducing axonal myelination and synaptic efficacy among others. Any effect by auditory stimulation after cochlear implantation could only have occurred in the adult system and is expected to be negligible. Therefore, we hypothesize typical eABR waveforms in both groups, but longer wave V latencies in the prelingually deaf.
Methods Participants All adult users of a Cochlear速 CI who visited the outpatient clinic from December 2011 to December 2012 were consented to participate in a study which included eABR and cortical auditory evoked potential (CAEP) recordings. Twenty-three adults, with at least 6 months experience with their CI, agreed to participate in this study. In 20 subjects, eABRs could be recorded, while in the remaining three postlingual subjects no clear eABRs could be evoked, due to electrical artifacts contaminating the waveforms. Their data were therefore not included in the analyses. Due to time constraints, in one subject eABRs without CAEPs were obtained (Post 1). Prelingually deaf subjects were selected based on the following criteria: onset of severe to profound binaural hearing loss before the age of 2 years (based on medical charts including diagnostic audiometry and self-reported patient information), and insufficient residual hearing during childhood for normal speech and language development. Based on these criteria, the diagnosis of prelingual deafness was confirmed by the multidisciplinary CI team prior to implantation. Eleven adults met these criteria, and were thus labeled as prelingually deaf. Nine adults became deaf during adolescence or adulthood (>16 years of age) and were categorized as postlingually deaf. All participants were users of Nucleus multichannel CIs, and in all subjects a full insertion of the electrode array was achieved. Table 1 summarizes detailed patient characteristics. The data of the CAEPs (recorded in 22 subjects, including all 11 prelingual subjects and 8 postlingual subjects enrolled in this study) have been reported in a separate paper (Lammers et al. 2015). Speech perception Speech perception scores were obtained using the Dutch Society of Audiology standard consonant-vowel-consonant (CVC) word list at 65 dB SPL (Versfeld et al. 2000). In this open-
104
eABRs in prelingually deaf CI users
set test, only auditory cues were available. Speech perception was scored based on the number of phonemes correctly identified. For each subject, the most recent scores prior to the evoked potential recordings were used (time intervals between 0 and 9 months). Procedure and stimuli Participants were seated in a comfortable reclining chair in an electrically shielded, sound attenuated booth and were asked to keep their eyes closed and minimize movements. The electric stimulus consisted of a biphasic pulse, with a phase width of 25 µs and an inter-phase gap of 8 or 58 µs. A monopolar stimulation electrode configuration was applied, and two positions of the active electrode were used: at the apical end of the array (typically electrode no. 20) and a central position (typically electrode no. 11). A basal electrode was also used (as it was for CAEPs, Lammers et al. 2015) but due to stimulation artifacts, the signal-to-noise ratio of the eABRs was too low to obtain reliable and reproducible waveforms in several patients. Stimuli were generated using the Cochlear Custom Sound EP 3.1 software and presented at a rate of 35 Hz at the individual’s maximum comfortable loudness level (C-level). For each subject 1.500 accepted sweeps were averaged. Multiple additional replications at the same level and lower stimulus levels were recorded to confirm the response. Stimulation levels were decreased until wave V could not be distinguished anymore. For data analyses, only the response obtained at C-level was used for each subject. Evoked potential recording Responses were recorded by Ag/AgCl electrodes placed according to the 10-20 system at Cz and Fz using a Medelec Synergy T-10 Evoked Potential system. The ground was placed on the forehead and the contralateral mastoid was used as reference. Recordings were filtered from 100 Hz to 5 kHz and recorded with a sampling rate of 20 kHz. Sweeps containing signals of > 50 µV at any electrode were rejected and not included in the average signal. Electrode impedances were kept below 5 kΩ. Data analyses Averaged eABR data were analyzed using custom scripts in MATLAB (version 7.11.0, Mathworks). For each subject, the analysis of the wave III and wave V latencies and the III-V interval was based on the eABRs obtained at maximum comfort level. Waves III and V were manually identified by two authors (MJWL and RHMvE) independently. Disagreements were resolved by discussion. Statistical analyses were completed using SPSS version 22.0 software. Repeated measures ANOVAs with the two different intracochlear stimulus locations (i.e. middle and apical) as within-subjects factor and group (i.e. prelingual or postlingual) as between-subjects factor were used. Significant main effects and interactions (p<0.05) were followed with Bonferroni
105
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
post hoc tests, and the Greenhouse-Geisser correction was applied to compensate for violations of the sphericity assumption. Group differences in peak latencies were calculated with unpaired, two-tailed t tests or the Mann-Whitney test for independent samples. Stepwise multiple linear regression analyses were performed to evaluate the influence of other variables on wave latencies. Linear regression analyses were performed to evaluate the relationship between speech perception and eABR latencies.
Results Patient demographics and speech perception The monosyllabic open-set speech perception scores varied widely among the patients, ranging from 0% to 82% in the prelingual group and from 60% to 97% in the postlingual group (Fig. 1). Median speech perception scores were significantly higher in the postlingual group (median 85% correct) than in the prelingual group (22% correct; Mann-Whitney U = 6.0, P < 0.0001). Patient demographics are presented in Table 1. The prelingual group and the postlingual group differed significantly in implant experience (mean difference, 6.5 years; unpaired two-tailed t-test, t18 = 4.6, P < 0.001). Age at implantation did not differ significantly (unpaired two-tailed t-test, t18 = 1.1, P = 0.305), neither did age at testing (unpaired, twotailed t-test, t18 = 2.0, P = 0.057). Preoperative CVC phoneme scores were not different between the two groups.
106
Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual
Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual
Pre 1 Pre 2 Pre 3 Pre 4 Pre 5 Pre 6 Pre 7 Pre 8 Pre 9 Pre 10 Pre 11 Mean
Post 1 Post 2 Post 3 Post 4 Post 5 Post 6 Post 8 Post 9 Post 11 Mean
Left Right Right Right Right Left Right Right Left
Right Right Left Left Left Left Left Right Left Right Left
Progressive Otitis media Otitis media Unknown Progressive Trauma Progressive Progressive Meningitis
Meningitis Congenital Rubella Varicella Rubella Meningitis Meningitis Congenital Unknown Rubella Unknown
Side CI Etiology
75 80 58 73 57 22 59 53 33 57
Age at test (years) 23 38 56 55 47 55 47 43 31 42 36 43 68 64 27 50 29 16 43 40 15 39
Age at onset deafness (years) 0.5 0 0.3 2 0 0.8 0.8 0 0 0 0 0
Implant experience (years) 1.7 0.6 2.5 0.6 1.5 5.7 5.7 2.9 3.2 1.4 8.9 3.2 6.5 5.3 14.4 13.8 6.9 6.5 13.3 11.9 8.2 9.6
Age at CI (years) 21 37 54 55 46 49 41 40 27 41 27 40 68 75 43 59 50 16 45 41 25 47
0 37 28 0 25 11 0 14
Pre-op CVC score (%) 0 0 0 0 48 12 0 0 0 0 28 8 78 (0) 89 (2) 60 (3) 72 (0) 95 (7) 90 (0) 85 (3) 84 (8) 97 (0) 83
22 (3) 0 (1) 82 (5) 26 (5) 66 (4) 18 (0) 28 (5) 0 (9) 15 (8) 0 (1) 77 (0) 30 Oral Oral Oral and lipreading Oral Oral and lipreading Oral Oral and lipreading Oral Oral
Lipreading Lipreading and sign language Lipreading Lipreading and sign language Oral and lipreading Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language
Post-op CVC Primary mode of communication score (%) (pre-operative)
-: not performed. Months between EABR recording and CVC measures are displayed between brackets in the Post-op CVC score column
Group
Subject
Table I Subject demographics
eABRs in prelingually deaf CI users
2.2
107
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig. 1. Consonant-vowel-consonant (CVC) phoneme scores of the prelingual and postlingual group taken prior to the evoked potential recordings. Each dot represents an individual subject. The horizontal lines represent the median scores (prelingual group, 22 %; postlingual group, 85 %). Subjects with hearing experience (Pre 1, 3, 4, 6, and 7) are marked with ¼; subject Pre 5, who used both visual speech cues and oral communication preoperatively, is marked with §. This plot is a partial replot of Fig. 1 in Lammers et al. (2015); all 11 prelingual data points and 8 of 11 postlingual data points of that paper are plotted here.
eABR waveforms in the pre- and postlingual groups In all nine postlingual subjects reproducible eABR waveforms could be obtained on both electrodes. In 9 out of 11 prelingual subjects eABR waveforms could be obtained on both electrodes, whereas in two subjects no clear waves III and V could be identified on the middle electrode. Figure 2 shows the individual eABR waveforms evoked at an apical electrode (typically electrode no. 20) for prelingual (left) and postlingual (right) subjects in order of CVC score. The waveforms tended to be relatively small for the prelingual subjects with poor speech perception. Depending on factors such as location of recording and stimulation electrodes, head size and skull thickness, eABR wave amplitudes were highly variable among subjects and thus less reliable for group comparisons. Therefore, only wave latencies were considered for group comparisons.
108
eABRs in prelingually deaf CI users
2.2
Fig. 2. Individual eABR waveforms evoked at an apical electrode. In cases in which large stimulus artifacts partially obscured the measurements recorded on Cz, waveforms measured at electrode Fz are presented here (denoted with ‡). Waveforms were corrected for stimulus artifact by fitting a first-order polynomial and subtracting it from the signal. In almost all subjects in both groups, wave III (indicated with first upward arrowhead) and wave V (indicated with second upward arrowhead) could be identified. The vertical lines drawn near the peak latencies of waves III and V are shown to facilitate comparisons between subjects. At the right side of the individual waveforms, the CVC phoneme scores are presented. In the outer margins of the figure the patient numbers are presented. Subjects with hearing experience (Pre 1, 3, 4, 6, and 7) are marked with ¥; subjects who had a preoperative CVC phoneme score higher than 0% correct (Pre 5, 6 and 11) are marked with §.
Wave latencies Grand averages of the eABRs evoked at apical and middle electrodes (Fig. 3) indicate that waves III of the two patient groups coincide, while wave V starts and peaks considerably later for the prelingually deaf patients when compared to the postlingually deaf. Accordingly, wave V latency was significantly longer in the nine prelingual subjects than in the nine postlingual
109
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
subjects across electrode locations (F(1,16) = 6.86, P = 0.019; Fig. 4.a). Analyses for the two electrode locations separately revealed that on the apical electrode the average wave V latency in the 11 prelingual subjects was 3.9 ms, whereas in the postlingual subjects this was 3.6 ms (unpaired two-tailed t-test, t18 = 2.6, P = 0.020). On the middle electrode, wave V latencies were on average 3.9 ms for the prelingual subjects and 3.7 ms for the postlingual subjects (unpaired two-tailed t-test, t16 = 2.4, P = 0.030). If the three best performing prelingual users, Pre 3, 5 and 11, were excluded from the analyses, wave V latency remained significantly longer than in the postlingual group (F(1,13) = 13.48, P = 0.003).
Fig. 3. Grand average eABR waveforms measured at Cz for all subjects in both groups, presented for the two stimulation electrode locations separately. The blue and red traces represent the waveforms of the prelingual and postlingual groups, respectively. Waveforms were first corrected for stimulus artifact by fitting a first-order polynomial and subtracting it from the signal and then they were normalized by dividing the signal by the difference in amplitude between the top of wave V and its preceding trough.
Analysis of the wave III latency did not reveal differences between the prelingual group and the postlingual group (F(1,16) = 1.14, P = 0.301). Average wave III latencies were around 2.0 ms in the prelingual group and 1.9 ms in the postlingual group on both electrode locations (apical electrode: unpaired, two-tailed t-test, t18 = 1.1, P = 0.307; middle electrode unpaired, two-tailed t-test, t16 = 1.2, P = 0.237). Wave III latency remained similar for both groups, if the three best performing prelingual subjects were excluded from the analyses (F(1,13) = 1.71, P = 0.214). The interwave III-V interval was longer for prelingually than for postlingually deafened, which was nearly significant when analyzed across both electrode locations (F(1,16) = 4.43, P = 0.052). On the apical electrode the III-V interval was significantly longer in the prelingual subjects (average III-V interval prelingual group, 1.9 ms; postlingual group 1.7 ms; unpaired two-tailed t-test, t18 = 2.7, P = 0.015). On the middle electrode the difference in III-V interval
110
eABRs in prelingually deaf CI users
between groups was smaller, and not significant (unpaired two-tailed t-test, t18 = 1.5, P = 0.143) (Fig. 4.b). If the three best performing prelingual users, Pre 3, 5 and 11, were excluded from the analyses, the III-V interval in the prelingually deaf CI users, was significantly longer than in the postlingual subjects across both electrode locations (F(1,13) = 7.05, P = 0.020).
Fig. 4. Wave V latencies (A) and III-V interwave intervals (B) of the prelingual (pre) and postlingual (post) groups presented for the two stimulation electrode locations. The box plots represent the lower and upper quartile with the median. Whiskers indicate the 5â&#x20AC;&#x201C;95 percentiles. *P < 0.05.
Factors related to wave V latency and III-V interval In order to assess if other factors, besides age at onset of deafness, could have contributed to the differences in wave V latency, stepwise multivariate regression analyses were performed. On the apical electrode these analyses revealed that group was the only significant predictor of wave V latency (r = 0.516, F(1,18) = 6.535, P = 0.020). Implant experience was not associated with wave V latency on this electrode (P = 0.846; Fig. 5.b). Contrarily, implant experience was found to be the only significant predictor of wave V latency on the middle electrode (r = -0.665, F(1,16) = 12.693, P = 0.003; Fig. 5.a). Group was not a significant predictor of wave V latency on this electrode (P = 0.887). When examining both groups separately, wave V latency was significantly correlated to implant experience in the prelingual group (r = -0.706, P = 0.034; Fig. 5.a), but not in the postlingual group (r = -0.324, P = 0.395). Previous studies on eABRs in children demonstrated that the wave V latency as function of CI experience could be best described by an exponential decay (Gordon et al. 2006; Thai-Van et al. 2007). If we describe the wave V latencies on the middle electrode in the prelingual subjects as an exponential it would yield a time constant of about 4 years. Wave V latency was not correlated to age at implantation on either tested electrode. Moreover, the other preoperative patient characteristics presented in Table I were not a
111
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
factor determining wave V latency. Within the prelingual or postlingual group the wave V latency was not significantly correlated to age at onset of deafness. The III-V interval was not associated with age at implantation, age at onset of deafness or implant experience on either electrode location.
Fig. 5. Wave V latency on the middle (A) and apical (B) electrode as a function of implant experience. Each red square represents a subject in the postlingual group. Each blue dot represents a subject in the prelingual group. On the middle electrode, there was a significant correlation between wave V latency and implant experience for the total group (dotted line, r =-0.665, P = 0.003, n = 18) and for the prelingually deaf subjects (solid blue line, r = -0.706, P = 0.034).
Relationship between eABRs and speech perception As demonstrated in Figure 2, typical eABRs exhibiting both waves III and V could be recorded in almost all subjects irrespective of their speech perception scores or group (pre or postlingually deaf). Waveform morphology was however less clear in the prelingually deaf and poor performing subjects. Besides, in two prelingual subjects no clear eABRs could be recorded on the middle electrode (Fig. 2). Since postlingually deafened subjects show shorter latencies and better speech perception a correlation is expected between these two measures. If all subjects were included in a linear regression analysis and if group was omitted, wave V latency was significantly associated with speech perception on both electrodes (apical r = -0.640, P = 0.002; middle r = -0.706, P = 0.001; Fig. 6, dotted lines). When the groups were analyzed separately a negative correlation between phoneme score and wave V latency was present on the apical electrode for the postlingual group (r = -0.833, P = 0.005; Fig. 6.b solid line) and on the mid electrode for the prelingual group (r = -0.728, P =0.026; Fig. 6.a solid line).
112
eABRs in prelingually deaf CI users
Fig. 6. CVC phoneme score as a function of wave V latency for the middle (A) and apical (B) electrodes. Each red square represents a subject in the postlingual group. Each blue dot represents a subject in the prelingual group. The univariate analyses for the total group revealed significant correlations on both electrodes (dotted lines, middle r = -0.706, P = 0.001; apical r = -0.640, P = 0.002). The blue solid line indicates a significant correlation within the prelingual group for the middle electrode (r = -0.728, P = 0.026), and the red solid line indicates a significant correlation within the postlingual group for the apical electrode (r = -0.833, P = 0.005).
Besides group, implant experience was significantly associated with speech perception (r = 0.580, P = 0.007). Stepwise multivariate regression analyses revealed that on the apical electrode wave V latency and implant experience were not predictive for speech perception, and only group remained a significant predictor (r = -0.755, F(1,19) = 23.89, P < 0.001). On the middle electrode, group and wave V latency were significant predictors of speech perception explaining for 69% of the variance (r = -0.828, F(1,19) = 16.37, P < 0.001; variance explained by group: 37%, P = 0.009 and by wave V latency: 32%, P = 0.018). The III-V interval was significantly associated with postoperative speech perception when assessed over all subjects (apical r = -0.632, P = 0.003; middle r = -0.494, P = 0.037), but not when both groups were analyzed separately. Stepwise multivariate regression analyses confirmed that III-V interval was not a significant predictor besides group.
Discussion In this study, we evaluated differences in auditory brainstem activation between pre- and postlingually deaf subjects who received a cochlear implant in adulthood. The extensive duration of auditory deprivation in the CI users with prelingual deafness appeared to have
113
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
resulted in a delayed wave V while wave III was virtually unaffected. This delayed wave V activation may reflect a hindered neural and synaptic development especially in the more rostral part of the brainstem. To our knowledge this is the first study to examine whether long durations of early onset auditory deprivation leads to delays in human auditory brainstem activation. These findings reveal that in the absence of auditory input throughout childhood, development of the rostral auditory brainstem seems to be hampered, due to degraded myelination and/or synaptic efficacy. Auditory brainstem development and hearing impairment Neuropathological studies in autopsied fetuses and infants revealed that the auditory brainstem rapidly matures during the perinatal period. It is assumed that during the first two trimesters of pregnancy the anatomical structures involved in the auditory pathway are already developed, independent of sound-evoked activity (Moore and Linthicum 2007). The subsequent final maturation of the auditory brainstem is believed to result from sensorydriven processes of myelination, an increase in size of auditory neurons and rapid growth and branching of dendrites (Thai-Van et al. 2007; Moore and Linthicum 2007). Myelination of the auditory pathway starts around the 26th to 29th week and continues in the brainstem during the fetal and postnatal period (Inagaki et al. 1987; Moore et al. 1995; Moore and Linthicum 2007). By the age of 1 year, the myelin density of the cochlear nerve is comparable to that of an adult (Moore et al. 1995). This myelination process and increase in synaptic efficacy of the auditory brainstem is probably reflected by the decreasing ABR wave latencies within the same period (Inagaki et al.1987; Eggermont and Salamy 1988; Ponton et al. 1992, 1996). In the presence of normal auditory stimulation, ABR wave latencies mature over the first years of life (Inagaki et al. 1987; Eggermont and Salamy 1988; Ponton et al. 1992). In normal hearing children, wave V latencies decrease following an exponential model, reaching adult values of 5-6 ms around the age of two. In the absence of normal auditory stimulation, congenitally deaf children show a significantly longer wave V latency when compared to children who became deaf after 1 to 4 years of age (Thai-Van et al. 2007). Among prelingually deaf children implanted between the ages of 1 to 17 years, the III-V wave interval recorded at activation of the CI does not depend on age at implantation (Gordon et al. 2006). In concordance with these findings, the results of our study show that after early onset of deafness and years of auditory deprivation, the wave V latency (and the wave III-V interval) is longer than in deaf subjects who experienced auditory stimulation in early development. In children with early-onset deafness, it has been shown that wave V latencies decrease over the first years following cochlear implantation, in similar fashion as the exponential decrease with age observed in normal hearing children (Gordon et al. 2006; Thai-Van et al. 2007). As suggested by our middle electrode recordings, duration of cochlear implant use
114
eABRs in prelingually deaf CI users
may have been a factor contributing to the wave V latency in addition to onset of deafness (prelingual vs. postlingual). Fitting a decaying exponential function shows the time constant of this effect to be 4 years, which would be much longer than previously reported constants of 68 weeks (Thai-Van et al. 2007) or 5 months (Gordon et al. 2006) observed in children. Although not consistently found over both electrode locations, this effect of implant experience might suggest a possible sensory driven maturation of the auditory brainstem, which is still present even in adulthood after long durations of auditory deprivation. Acquiring longitudinal data will be necessary to clearly identify the role of chronic electrical stimulation in adults. Animal models of deafness show that the major structures and pathways in the brainstem develop, but auditory input is required for the refinement of neuronal connections. Hence, in the absence of auditory stimulation, the cochlear nucleus becomes smaller and the projections received from the auditory nerve are broadened (Moore 1990; Leake et al. 2008), furthermore, cells shrink and synapses loose vesicles (Ryugo et al. 2010). The cochleotopic organization is however maintained despite deafness, but these broader connections might affect frequency resolution (Leake et al. 2008). In ferrets with bilateral cochlear ablation, the number of neurons projecting from the cochlear nucleus to the inferior colliculus was not different than that of normal hearing animals (Moore 1990), suggesting that the coarse structure of the pathway is not affected by deafness. This is supported by a longitudinal study comparing eABRs of congenitally deaf cats with those with normal hearing (Tillein et al. 2012). This study showed that in the absence of auditory input, the normal eABR waves develop and wave latencies decrease at the same rate as in normal hearing cats. Although wave latencies were comparable, eABR waveform morphology was less clear in the congenitally deaf cats. Moreover, the responses revealed a reduction in the amplitude of wave III and an increase of wave V amplitude (Tillein et al. 2012). The ongoing decrease of wave V latency in congenitally deaf cats after birth does not coincide with human data. This might be due to a discrepancy between the two species or due to a difference in stimulation (e.g., lower level in cats than in humans). Changes in brainstem responses and cortical potentials in prelingually deaf Combining the results of the present study with the cortical potentials recorded in the same subjects (Lammers et al. 2015) gives an insight in the spontaneous development of the auditory pathway. In the prelingually deaf subjects both the brainstem and cortical waveforms showed a normal morphology indicating that a coarse neural network from cochlea to cortex develops regardless of sensory input. Thus, even in the absence of sensory input throughout childhood, the development of this innate auditory pathway is not disrupted. Remarkably, whereas the brainstem responses were delayed, the cortical responses in the prelingually deaf subjects showed significantly shorter latencies than the postlingually deaf subjects. We should stress that the cortical latency differences (~15 ms) are almost 100 times
115
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
larger than the brainstem latency differences (~0.2 ms). Whereas myelination and synaptic development explain the relatively subtle ABR latency differences, other mechanisms play a role in cortical potentials. Sensory input representing biologically relevant sounds is crucial for refinement of the cortical network, in particular during childhood (Buonomano and Merzenich 1998; Innocenti and Price 2005; Ohl and Scheich 2005; Kral 2013). Thus, without such input, the cortex maintains a rather coarse organization containing fewer cortico-cortical connections with less top-down inhibition (Kral and Sharma 2012; Lammers et al. 2015). The different mechanisms of brainstem and cortex responses may also explain that eABR wave V latencies tend to decrease with implant experience, whereas such tendency is not found for CAEP latencies. Methodological considerations In our study population a significant correlation between speech perception and wave V latency was found when eABRs were evoked at the middle electrode (Fig. 6). This agrees with a report by GallĂŠgo et al. (1998) who also found an association between speech perception and wave V latencies. Although wave V latency on the middle electrode was found to be an independent predictor of speech perception, these results should be interpreted with caution. Larger patient series should be investigated to clarify the predictive role of eABR wave latencies, especially since there is no clear trend in the current literature studying relative small sample sizes (Abbas and Brown 1991; Brown et al. 1995; GallĂŠgo et al. 1998; Makhdoum et al. 1998; Firszt et al. 2002; Kim et al. 2008; Gibson et al. 2009). Variability among the CI patients was typically substantial. Notably, the speech perception scores among the patients classified as prelingually deaf varied widely (Fig. 1). Three of those subjects had high CVC scores and short wave V latencies, both characteristics being shared by postlingually deaf subjects. Since judgment of early onset of deafness is partly based on selfreported information (see Methods section), one cannot exclude significant use of residual hearing (also discussed in Lammers et al. 2015), which might have contributed to a better development of the brainstem and eventually to a better speech perception with their CI. On the other hand, sensitivity analyses excluding the three best performing prelingually deaf subjects did not change the found effects. Moreover, the early onset of deafness in these three subjects is supported by the cortical potentials which had latencies in the range of the other prelingually deaf subjects, significantly shorter than postlingual subjects (Lammers et al. 2015).
116
eABRs in prelingually deaf CI users
Conclusion Electrically evoked auditory brainstem responses in prelingually deaf late-implanted CI users demonstrate increased wave V latencies and interwave III-V intervals, suggesting delays in neural conduction within the auditory brainstem. These results indicate that long durations of hearing impairment directly or shortly after birth may lead to impaired neuronal connections within the innate and elementary parts of the auditory brainstem.
Acknowledgments We gratefully acknowledge all participants and the members of the cochlear implant team for their time and support. We thank Aren Bezdjian for his valuable suggestions on the manuscript. This study is supported by an unrestrictive research grant from Cochlear Ltd.
2.2
117
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
References Abbas PJ, Brown CJ (1991) Electrically evoked auditory brainstem response: growth of response with current level. Hearing Research 51:123-137 Brown CJ, Abbas PJ, Bertschy M, Tyler RS, Lowder M, Takahashi G, Purdy S, Gantz BJ (1995) Longitudinal assessment of physiological and psychophysical measures in cochlear implant users. Ear and Hearing 16:439-449 Buonomano DV, Merzenich MM (1998) Cortical plasticity: from synapses to maps. Annual Review of Neuroscience 21:149-186 Doucet M, Bergeron F, Lassonde M, Ferron P, Lepore F (2006) Cross-modal reorganization and speech perception in cochlear implant users. Brain 129:3376-3383 Eggermont JJ, Salamy A (1988) Maturational time course for the ABR in preterm and full term infants. Hearing Research 33:35-47 Firszt JB, Chambers RD, Kraus N (2002) Neurophysiology of cochlear implant users II: comparison among speech perception, dynamic range, and physiological measures. Ear and Hearing 23:516-531 Gallégo S, Frachet B, Micheyl C, Truy E, Collet L (1998) Cochlear implant performance and electrically-evoked auditory brain-stem response characteristics. Electroencephalography and Clinical Neurophysiology 108:521-525 Gibson WP, Sanli H, Psarros C (2009) The use of intra-operative electrical auditory brainstem responses to predict the speech perception outcome after cochlear implantation. Cochlear Implants International 10 Suppl 1:53-57 Gordon KA, Papsin BC, Harrison RV (2006) An evoked potential study of the developmental time course of the auditory nerve and brainstem in children using cochlear implants. Audiology and Neurotology 11:7-23 Gordon KA, Valero J, van Hoesel R, Papsin BC (2008) Abnormal timing delays in auditory brainstem responses evoked by bilateral cochlear implant use in children. Otology & Neurotology 29:193-198 Inagaki M, Tomita Y, Takashima S, Ohtani K, Andoh G, Takeshita K (1987) Functional and morphometrical maturation of the brainstem auditory pathway. Brain and Development 9:597-601 Kim AH, Kileny PR, Arts HA, El-Kashlan HK, Telian SA, Zwolan TA (2008) Role of electrically evoked auditory brainstem response in cochlear implantation of children with inner ear malformations. Otology & Neurotology 29:626-634 Kral A (2013) Auditory critical periods: a review from system’s perspective. Neuroscience 247:117-133 Kral A, O’Donoghue GM (2010) Profound deafness in childhood. The New England Journal of Medicine 363:1438-1450 Kral A, Sharma A (2012) Developmental neuroplasticity after cochlear implantation. Trends in Neurosciences 35:111-122 Lammers MJW, Versnel H, van Zanten GA, Grolman W (2015) Altered cortical activity in prelingually deafened cochlear implant users following long periods of auditory deprivation. Journal of the Association for Research in Otolaryngology 16:159–170 Leake PA, Hradek GT, Bonham BH, Snyder RL (2008) Topography of auditory nerve projections to the cochlear nucleus in cats after neonatal deafness and electrical stimulation by a cochlear implant. Journal of the Association for Research in Otolaryngology 9:349-372 Lee HJ, Giraud AL, Kang H, Kim CS, Lee DS (2007) Cortical activity at rest predicts cochlear implantation outcome. Cerebral Cortex 17:909-917 Makhdoum MJ, Groenen PA, Snik AF, van den Broek P (1998) Intra- and interindividual correlations between auditory evoked potentials and speech perception in cochlear implant users. Scandinavian Audiology 27:13-20
118
eABRs in prelingually deaf CI users
Moore DR (1990) Auditory brainstem of the ferret: bilateral cochlear lesions in infancy do not affect the number of neurons projecting from the cochlear nucleus to the inferior colliculus. Developmental Brain Research 54:125-130 Moore JK, Perazzo LM, Braun A (1995) Time course of axonal myelination in the human brainstem auditory pathway. Hearing Research 87:21-31 Moore JK, Linthicum FHJr (2007) The human auditory system: a timeline of development. International Journal of Audiology 46:460-478 Ponton CW, Eggermont JJ, Coupland SG, Winkelaar R (1992) Frequency-specific maturation of the eighth nerve and brain-stem auditory pathway: evidence from derived auditory brain-stem responses (ABRs). Journal of the Acoustical Society of America 91:1576-1586 Ohl FW, Scheich H (2005) Learning-induced plasticity in animal and human auditory cortex. Current Opinion in Neurobiology 15:470-477 Ponton CW, Moore JK, Eggermont JJ (1996) Auditory brainstem response generation by parallel pathways: differential maturation of axonal conduction time and synaptic transmission. Ear and Hearing 17:402410 Ryugo DK, Baker CA, Montey KL, Chang LY, Coco A, Fallon JB, Shepherd RK (2010) Synaptic plasticity after chemical deafening and electrical stimulation of the auditory nerve in cats. Journal of Comparative Neurology 518:1046-1063 Sparreboom M, Beynon AJ, Snik AFM, Mylanus EAM (2010) Electrically evoked auditory brainstem responses in children with sequential bilateral cochlear implants. Otology & Neurotology 31:1055-1061 Spoendlin H (1975) Retrograde degeneration of the cochlear nerve. Acta otolaryngologica 79:266-275 Teoh SW, Pisoni DB, Miyamoto RT (2004) Cochlear implantation in adults with prelingual deafness. Part I. Clinical results. Laryngoscope 114:1536-1540 Thai-Van H, Cozma S, Boutitie F, Disant F, Truy E, Collet L (2007) The pattern of auditory brainstem response wave V maturation in cochlear-implanted children. Clinical Neurophysiology 118:676-689 Tillein J, Heid S, Lang E, Hartmann R, Kral A (2012) Development of brainstem-evoked responses in congenital auditory deprivation. Neural Plasticity 182767 Versfeld NJ, Daalder L, Festen JM, Houtgast T (2000) Method for the selection of sentence materials for efficient measurement of the speech reception threshold. Journal of the Acoustical Society of America 107:1671-1684 Versnel H, Agterberg MJH, de Groot JCMJ, Smoorenburg GF, Klis SFL (2007) Time course of cochlear electrophysiology and morphology after combined administration of kanamycin and furosemide. Hearing Research 231:1-12
119
2.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3 Altered cortical activity in prelingually deafened cochlear implant users following long periods of auditory deprivation M.J.W. Lammers, H. Versnel, G.A. van Zanten, W. Grolman
Journal of the Association for Research in Otolaryngology 2015; 16(1):159-70
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Auditory stimulation during childhood is critical for the development of the auditory cortex in humans, and with that for hearing in adulthood. Age-related changes in morphology and peak latencies of the cortical auditory evoked potential (CAEP) have led to the use of this cortical response as a biomarker of auditory cortical maturation including in studies of cortical development after deafness and subsequent cochlear implantation. To date, it is unknown whether prelingually deaf adults, with early onset deafness (before the age of 2 years) and who received a cochlear implant (CI) only during adulthood, would display absent or aberrant CAEP waveforms as predicted from CAEP studies in late implanted prelingually deaf children. In the current study, CAEP waveforms were recorded in response to electric stimuli in prelingually deaf adults, who received their CI after the age of 21 years. Waveform morphology and peak latencies were compared to the CAEP responses obtained in postlingually deaf adults, who became deaf after the age of 16. Unexpectedly, typical CAEP waveforms with adult-like P1-N1-P2 morphology could be recorded in the prelingually deaf adult CI users. On visual inspection, waveform morphology was comparable to the CAEP waveforms recorded in the postlingually deaf CI users. Interestingly, however, latencies of the N1 peak were significantly shorter and amplitudes were significantly larger in the prelingual group than in the postlingual group. The presence of the CAEP together with an early and large N1 peak might represent activation of the more innate and less complex components of the auditory cortex of the prelingually deaf CI user, whereas the CAEP in postlingually deaf CI users might reflect activation of the mature neural network still present in these patients. The CAEPs may therefore be helpful in the assessment of developmental state of the auditory cortex.
122
eCAEPs in prelingually deaf CI users
Introduction Cochlear implantation in adults has been found to be especially successful in those with postlingual deafness. Postoperative results in terms of speech perception tend to be much poorer in prelingually deaf adults when they received their cochlear implant after a long duration of deafness (Teoh et al., 2004; Klop et al., 2007). These poor performances might be explained by a hampered development of the auditory pathway and cross-modal changes during the period of auditory deprivation (Doucet et al., 2006; Lee et al., 2007; Kral and Oâ&#x20AC;&#x2122;Donoghue, 2010; Kral and Sharma, 2012). Absent auditory input during early childhood is thought to lead to underdevelopment of the auditory cortex including reduced axonal density in supragranular layers (Ponton and Eggermont, 2001; Eggermont and Ponton, 2003) and corticocortical decoupling between the different layers of the primary auditory cortex and between the primary and higherorder auditory cortex (Kral and Sharma, 2012). Corticocortical decoupling between the supragranular and the infragranular layers could affect the descending projections which are thought to be important in the top-down modulation of incoming auditory stimuli, whereas decoupling between the primary and higher order auditory cortex might result in crossmodal cortical reorganization (Kral and Sharma, 2012). Several electrophysiological studies have used cortical auditory evoked potentials (CAEP) to monitor auditory cortex development since the morphology of these cortical responses changes during childhood, possibly reflecting auditory cortex development (Sharma et al., 1997; Ponton et al., 2000; Wunderlich and Cone-Wesson, 2006). In young children the CAEP waveform is dominated by a large and broad positive peak, labeled P1, of which the latency decreases with age. By 9-12 years of age, a negativity, known as the N1 becomes present, resulting in the emergence of the typical P1-N1-P2 complex (Ponton et al., 2000; Ponton and Eggermont, 2001; Wunderlich and Cone-Wesson, 2006). In adults, the N1 peak, occurring at approximately 100 ms after stimulus onset, has been found to be the most robust component and its source is hypothesized to be located in the supragranular layers (II, III) of the auditory cortex (Ponton and Eggermont, 2001). Studies in prelingually deaf children who received a cochlear implant after several years of auditory deprivation and beyond the supposed sensitive period of 3.5 years revealed that the N1 peak did not develop and only an aberrant P1 peak was present (Ponton and Eggermont, 2001; Sharma et al., 2002; Dorman et al., 2007; Kral and Sharma, 2012). This lack of N1 might reflect corticocortical deficits within the superficial supragranular layers, caused by long-term auditory deprivation. In contrast, Gordon et al. (2008) described a large negative peak with a short latency (around 80 ms) in poorly performing, prelingually deaf children, including those who received their implant late. Although the cortical responses of these children resemble the typical P1-N1-P2 waveform morphology, the negativity that appears 20 ms earlier than the typical N1, is thought to reflect abnormal cortical functioning (Gordon et al., 2008). 123
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Thus, after early deafness, the cortex undergoes underdevelopment, possibly followed by cross-modal plasticity and then by providing new auditory input by means of a cochlear implant the cortex may change again. We addressed this cascade of cortical developmental events by recording CAEPs in adult cochlear implant users who became deaf before the age of 2 years and received their implants during adulthood and had therefore much longer durations of auditory deprivation than the late-implanted, prelingually deaf children in the abovementioned studies. The long period of auditory deprivation will most likely prevent neural maturation processes, which may normally occur throughout adolescence, such as synaptic pruning (Huttenlocher and Dabholkar, 1997), to emerge after cochlear implantation. Extrapolating the results of Gordon et al. (2008) a (near-) normal N1 would be possible in these subjects; based on the other studies their CAEPs are expected to lack a significant N1. The aim of our study was therefore to characterize cortical responses in late-implanted prelingually deaf adults and compare these responses with those of adult CI users who became deaf during adulthood after normal development of the auditory cortex.
Methods Participants A consecutive series of adult CI users, who visited the outpatient clinic for their routine evaluations, was asked to participate in this study. Twenty-three adults, who received their CIs after 16 years of age and who had a minimum of 6 months experience with their implants, agreed to participate and could be enrolled in this study. Prelingually deaf subjects were selected based on the following criteria: severe to profound binaural hearing loss with its onset before the age of 2 years (based on patient information, medical histories and diagnostic audiometry results), and impaired speech production as assessed by the speech and language therapist of the CI team. The diagnosis of prelingual deafness was confirmed by the evaluation of the multidisciplinary CI team prior to cochlear implant surgery. Twelve of those 23 adults met these criteria. In one prelingual deaf subject, no reliable CAEPs could be evoked, due to excessive muscle artifacts and this subject was therefore excluded. Eleven adults became deaf during adolescence or adulthood (>15 years of age) and were classified as postlingually deafened. All participants were users of Nucleus multi-channel cochlear implants. Table 1 provides details of all 22 participants including their pre- and postoperative phoneme score on the consonant-vowel-consonant (CVC) word test.
124
Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual Prelingual
Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual Postlingual
Pre 1 Pre 2 Pre 3 Pre 4 Pre 5 Pre 6 Pre 7 Pre 8 Pre 9 Pre 10 Pre 11 Mean
Post 1 Post 2 Post 3 Post 4 Post 5 Post 6 Post 7 Post 8 Post 9 Post 10 Post 11 Mean
Right Right Right Right Right Left Left Right Right Right Left
Right Right Left Left Left Left Left Right Left Right Left
Side CI
Progressive Otitis media Otitis media Unknown Progressive Trauma Otosclerosis Progressive Progressive Otosclerosis Meningitis
Meningitis Congenital Rubella Varicella Rubella Meningitis Meningitis Congenital Unknown Rubella Unknown
Etiology 0.5 0 0.3 2 0 0.8 0.8 0 0 0 0 0 59 64 27 50 29 16 30 43 40 30 15 37
67 80 58 73 57 22 66 59 53 66 33 58
60 75 43 59 50 16 64 45 41 51 25 48
21 37 54 55 46 49 41 40 27 41 27 40 6.4 5.3 14.4 13.8 6.9 6.5 1.9 13.3 11.9 14.4 8.2 9.4
1.7 0.6 2.5 0.6 1.5 5.7 5.7 2.9 3.2 1.4 8.9 3.2 39 37 28 0 33 25 11 0 22
0 0 0 0 48 12 0 0 0 0 28 8 83 (4) 89 (2) 60 (3) 72 (0) 95 (7) 90 (0) 84 (9) 85 (3) 84 (8) 89 (3) 97 (0) 84
22 (3) 0 (1) 82 (5) 26 (5) 66 (4) 18 (0) 28 (5) 0 (9) 15 (8) 0 (1) 77 (0) 30
Oral Oral Oral and lipreading Oral Oral and lipreading Oral Oral Oral and lipreading Oral Oral Oral
Lipreading Lipreading and sign language Lipreading Lipreading and sign language Oral and lipreading Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language Lipreading and sign language
Age at onset Age Implant Pre-op Post-op Primary mode of communication deafness at CI experience CVC score CVC score (pre-operative) (years) (years) (years) (%) (%)
23 38 56 55 47 55 47 43 31 42 36 43
Age at test (years)
-: not performed. Months between CAEP recording and CVC measures are displayed between brackets in the Post-op CVC score column
Group
Subject
Table I Subject demographics
eCAEPs in prelingually deaf CI users
125
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Speech perception Pre- and postoperative speech perception scores were obtained using the Dutch Society of Audiology standard CVC word list at 65 dB SPL (Versfeld et al., 2000). In this auditory-only presented open-set test, speech perception was scored as the percentage of phonemes correct. The most recent scores for each subject, prior to the evoked potential recordings were used (time intervals between evoked potential recordings and most recent postoperative speech perception scores ranged from 0 to 9 months; see Table 1). Procedure and stimuli Participants were seated in a comfortable reclining chair in an electrically shielded, soundattenuated booth and were allowed to watch a silent, captioned movie. They were carefully instructed prior to each recording to minimize movements and to fixate the center of the video screen to minimize muscle and eye movement artefacts. Cortical auditory evoked potentials (CAEP) were evoked by a short train of electrical biphasic pulses at 250 pulses per second for 36 ms. This stimulus was also used by Gordon et al. (2008) and Jiwani et al. (2013). A monopolar stimulation electrode configuration was applied, and the electrode position was varied: it was at the apical end of the array (typically electrode no. 20), at a central position (typically no. 11) or at the basal end of the array (typically no. 2). Pulse trains were generated using the Cochlear Custom Sound EP 3.1 software and presented at a rate of 0.9 Hz at the individual’s maximum comfortable loudness level (C-level). These settings of stimulation rate and level are very similar to settings in the CI literature (Ponton and Eggermont, 2001; Sharma et al., 2002; Gordon et al., 2008; Jiwani et al., 2013). For all participants phase widths of the biphasic pulses were set at 25 µs and interphase gaps at 8 µs. In each participant 100 responses to a pulse train were averaged, and at least one repetition was done. Pilot experiments in normal-hearing subjects and CI users had indicated that 200 accepted trials was adequate to yield clear responses. More repetitions were obtained in cases with higher noise levels due to, e.g., muscle artifacts. Evoked potential recording Responses were recorded by Ag/AgCl electrodes placed according to the 10-20 system using a Medelec Synergy T-10 Evoked Potential system. The active electrode was placed at the vertex of the skull, Cz, the contralateral mastoid was used as reference electrode (A1/A2) and the ground electrode was placed on the forehead, Fpz. Electrode impedances were kept below 5 kΩ. Eye movements and blinks were monitored using electrodes on the forehead above the eye. The electrode signals were filtered from 0.01 to 100 Hz and recorded with a sampling rate of 50 kHz. Responses were acquired in a 500 ms time-window, consisting of a pre-stimulus period of 100 ms and a post-stimulus period of 400 ms. Responses containing amplitudes of >100 µV at any electrode were rejected and not included in the averaged response. 126
eCAEPs in prelingually deaf CI users
Management of stimulus artifact In order to clearly differentiate the auditory response from the stimulus artifact, we used several techniques. First, we directly stimulated the CI by presenting short pulse trains to the individual electrodes. This direct method has no disadvantage of the indirect method by acoustic signal presentations in which FM signals are transmitted by the transmitting coil to the internal device, which could contaminate the recordings. Second, by using a short pulse train of only 36 ms, we expect that the stimulus artifact including possible post-stimulus ringing of the response acquisition system would end before the typical region of interest of the N1 peak (Gordon et al., 2008). Third, in an additional recording channel we used Fz as active electrode since we have found that stimulus artifacts are usually smaller or absent when recorded more frontally. By visual inspection and comparison with the waveforms obtained at Cz and Fz, artifact and response could readily be differentiated in all cases. Fourth, after the actual recordings, we recorded 100 sweeps at a stimulus level just below threshold. In this recording, the stimulus artifact could be identified while the CAEP was absent. Finally, in cases with a large stimulus artifact which could possibly interfere with the CAEP, an additional recording of 100 accepted sweeps at a slower repetition rate of 0.45 Hz was performed. By reducing the repetition rate the CAEP can be easily differentiated from the artifact, while the amplitude of the CAEP increases, whereas the size of the stimulus artifact including poststimulus ringing remains identical (Friesen and Picton, 2010). These recordings were only used to accurately differentiate the CI artifacts from the actual response. For our analyses, only the Cz recordings obtained at the repetition rate of 0.90 Hz were used. Data analyses Averaged evoked potential data were analyzed using custom scripts in MATLAB (version 7.11.0, Mathworks, Natick, MA, USA). The two repetitions of 100 accepted responses obtained at the stimulus rate of 0.9 Hz were averaged for each subject. CAEP peak amplitude and latency analyses were performed on these individual averages recorded at Cz. N1 was defined as the most negative peak at 70 to 150 ms after stimulus onset. P2 was defined as the first pronounced positive peak occurring after N1 at 120 to 250 ms after stimulus onset. Absolute N1 amplitude was computed relative to the start of the N1 wave. Furthermore, the N1-P2 peak-to-peak amplitude was used for data analyses. CAEPs were included for analyses when the N1-P2 amplitude was at least four times the root-mean square amplitude of the 100 ms pre-stimulus trace segment. The CAEP was further evaluated by the correlations between 50 and 250 ms post-stimulus segments of the two recording runs.
127
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Statistical analyses Statistical analyses were completed using SPSS version 20.0 software (IBM, Armonk, NY, USA). Repeated measures ANOVAs with the three different stimulation electrode locations (i.e. basal, middle, and apical) as within-subject factor and group (i.e. prelingual or postlingual) as between-subject factor were used. Significant main effects and interactions (p<0.05) were followed with Bonferroni post hoc tests, and the Greenhouse-Geisser correction was applied to compensate for violations of the sphericity assumption. Group differences in peak latencies and amplitudes were calculated with unpaired, two-tailed t tests or the Mann-Whitney test for independent samples. Stepwise multiple linear regression analyses were performed to evaluate the influence of other variables. A linear regression analysis was conducted on the N1 and P2 amplitudes and latencies with postoperative CVC phoneme scores as independent variables to assess the relation between CAEP characteristics and speech perception.
Results Patient demographics and speech perception Reproducible CAEP waveforms could be obtained in 22 out of 23 participants (96%). In one prelingually deaf subject, no reliable and reproducible waveforms could be evoked on either of the three stimulation locations due to excessive muscle artifacts, and therefore, this subjectâ&#x20AC;&#x2122;s responses were excluded from the analyses. Patient demographics of the 22 included subjects are presented in Table 1. As can be seen, the prelingual group and the postlingual group differed significantly in age at testing (mean difference, 15 years; unpaired, two-tailed t-test, t20 = 2.4, P = 0.024), and implant experience (mean difference, 6.2 years; unpaired, t20= 4.1, P = 0.001). Age at implantation did not differ significantly (unpaired, t20= 1.4, P = 0.193). CVC phoneme scores demonstrated a large variability, varying from 0% to 82% in the prelingual group and from 60% to 97% in the postlingual group (Fig. 1). The difference in scores was highly significant (median difference, 63%; Mann-Whitney U = 116, P < 10-4). In both patient groups, speech perception reached a stable performance level within 1 year after implantation as illustrated in Figure 1b. The CVC phoneme scores taken prior to the CAEP recordings represent stable speech perception in all patients, including the two prelingual patients who had only 6 months CI experience (Pre 2 and Pre 4) and who continued to have similarly low speech perception scores after 2 years follow-up. C-levels were significantly lower in the prelingual group than in the postlingual group (F(1,20) = 5.436, P = 0.030). The C-levels for the prelingual group were around 165 current units (CU), for the postlingual group around 190 CU (mean difference basal electrode, 25 CU unpaired two-tailed t-test, t20 = 2.6, P = 0.016; mean difference middle electrode, 24 CU, t20 = 2.4, P = 0.029; mean difference apical electrode 16 CU, t20 = 0.7, P = 0.098). Detection
128
eCAEPs in prelingually deaf CI users
threshold levels (T-levels) for the prelingual group were around 129 CU and for the postlingual group around 140 CU for all three electrode locations. T-levels were not significantly different between the two groups (F(1,20) = 1.385, P = 0.253).
Fig. 1. A. Consonant-vowel-consonant word (CVC) phoneme scores of the prelingual and postlingual group at the time of the evoked potential recordings. Each dot represents an individual subject. The horizontal lines represent the median scores (prelingual group, 22%; postlingual group, 85%). Subjects who had some hearing experience (Pre 1, 3, 4, 6 and 7) are marked with ¼; subject Pre 5, who used both visual speech cues (as lip reading) and oral communication pre-operatively is marked with §. B. Mean CVC phoneme scores over time following cochlear implantation presented for both groups. Postoperative CVC scores were obtained after 1, 3, 12, 24 and 36 months of cochlear implant use. Four subjects in the prelingual group (Pre 1, Pre 2, Pre 8 and Pre 10) were not able to obtain speech perception at any level if only auditory cues were available; three of these subjects were non-users at the 3-year interval (Pre 1, Pre 2 and Pre 10); at these intervals, a CVC phoneme scores of 0% was noted for these subjects.
CAEP waveforms in the pre- and postlingual groups Figure 2 shows the individual and unfiltered CAEP waveforms evoked at a middle electrode (typically electrode number 11). All subjects in the postlingual group (right column) demonstrated CAEP waveforms comparable to the typical adult waveform morphology found in normal hearing adults in response to acoustic stimuli, with the N1 peaking around 100 ms and the P2 around 200 ms (Wunderlich and Cone-Wesson, 2006; Martin et al., 2008). Although in some waveforms stimulus artifacts were present, careful inspection of the waveforms revealed N1 and P2 peaks in all subjects. In some of the postlingual subjects the N1-P2 complex was weakly present with a low N1-P2 amplitude relative to the pre-stimulus noise level (less than 10 times the root mean square) and/or a low correlation between the two recording runs (r < 0.7). The waveforms in all prelingual subjects (Fig. 2, left column) demonstrated a clear N1-P2 complex. The N1, peaking around 90 ms in the prelingual group, predominates the response and is followed by the P2, with its maximum around 200 ms. All waveforms presented in Figure 2 could be reproduced on both the basal and apical stimulating electrode locations.
129
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig. 2. Individual CAEP waveforms evoked at a middle CI electrode (typically no. 11). The stimulus consisted of a 36-ms pulse train with a pulse rate of 250 Hz. For almost all subjects, waveforms were measured at recording electrode Cz. In case of a large stimulus artifact at the recording electrode Cz, waveforms measured at electrode Fz are presented (indicated with ‡). In all subjects in both groups, a N1 (indicated with downward arrowhead) and a P2 peak (indicated with upward arrowhead) could be identified. At the right side of the individual waveforms, the CVC phoneme scores are presented. At the left side, the correlations between two successive recordings (in the response window 50-250 ms after stimulus onset) are indicated with their coefficient. The vertical line at 100 ms indicates the typical N1 latency found in normal hearing adults; the vertical line at 200ms indicates the typical P2 latency found in normal hearing adults. Subjects who had some hearing experience (Pre 1, 3, 4, 6 and 7) are marked with ¥; subject Pre 5, who used both visual speech cues and oral communication preoperatively is marked with §.
130
eCAEPs in prelingually deaf CI users
Due to the larger stimulus artifact present in the CAEP waveforms obtained after stimulation of the basal electrodes, these responses were less clear as compared to the middle and apical electrodes. In ten subjects (eight postlingual ones) the N1-P2 complexes were weakly present. Nevertheless, cortical responses could be easily differentiated from stimulus artifacts, and on visual inspection, N1 and P2 peaks could be identified. Correlations between two recording repetitions were generally high. For the response segments, 50 to 250 ms after stimulus onset correlation coefficients were on the average 0.72 for basal to 0.88 for apical electrode stimulation. All subjects had reproducible recordings with apical stimulation (r > 0.64). The noise levels in the recordings, assessed on 100 ms pre-stimulus trace segments, were on the average 0.5 ÂľV (root mean square) and they were similar for the pre- and postlingual group (F(1,20) = 1.74, P = 0.24). Figure 3 presents grand mean waveforms recorded from Cz for all three stimulating electrode locations separately for each group. It shows two features: First, the N1-P2 complex appeared earlier and larger in the prelingual group than in the postlingual group. Second, in both groups, the responses evoked by apical electrode stimulation were larger than the basally evoked responses.
Fig. 3. Grand average CAEP waveforms measured at Cz for all subjects in both groups. Grand mean waveforms are presented for the three electrode stimulation locations separately (i.e. basal, middle and apical). The blue and red traces represent the waveforms of the prelingual and postlingual groups, respectively.
Latencies N1 latencies were significantly shorter in the prelingual group than in the postlingual group for all three electrode locations (F(1,20) = 11.78, P = 0.0026) (Fig. 4). The N1 latencies for the prelingual group were around 90 ms and for the postlingual group around 105 ms (mean difference basal electrode, 14 ms unpaired two-tailed t-test, t20 = 2.3, P = 0.029; mean difference middle electrode, 17 ms, t12 = 3.7 P = 0.0013; mean difference apical electrode, 14 ms, t20= 3.5 P = 0.002). There was no interaction effect of stimulating electrode location (F(2,40) = 0.20, P = 0.818), and the N1 latencies did not vary with stimulating electrode location (F(2,40) = 1.89, P = 0.16). 131
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig. 4. N1 latencies of the prelingual (pre) and postlingual (post) groups presented for the three stimulation electrode locations, i.e. basal, middle and apical. At all three stimulation locations, a significant difference in N1 latency between the two groups was found. The box plots represent the lower and upper quartile with the median. Whiskers indicate the 5-95 percentiles. * P < 0.05, ** P < 0.01.
Analysis of the P2 latencies did not reveal differences between the prelingual (average latency per electrode: basal 186 ms, middle 177 ms, apical 177 ms) and the postlingual group (average latency per electrode: basal 174 ms, middle 187 ms, apical 182 ms) (F(1,20)= 0.003, P = 0.96) or the different electrode stimulation locations (F(2,40) = 0.36, P = 0.70). In order to assess whether group differences in age at implantation, age at testing and implant experience could have contributed to the group differences found in N1 latency, stepwise multiple linear regression analyses were performed. Pre- or postlingual deafness accounted for 41% of the variability in N1 latency on the middle electrode (r2 = 0.41, F(1,20)= 13.99, P = 0.0013). Although implant experience and age at testing were identified as significant factors explaining the N1 latency, these variables did not provide further significant improvement to the regression model (P = 0.080 and P = 0.275 respectively). On the basal and apical electrodes, implant experience and age at testing were not even identified as significant factors. Amplitudes N1 amplitudes were significantly larger in the prelingual group than in the postlingual group (mean difference 1.64uV, F(1,20) = 5.03, P = 0.036) (Fig. 5.a). There was no interaction effect of stimulating electrode location (F(2,40) = 0.24, P = 0.64), but the N1 amplitudes did vary with stimulating electrode location (F(2,40) = 6.60, P = 0.009). Bonferroni post hoc analyses revealed that the amplitudes at the basal electrode were significantly smaller than at the middle (mean difference, 0.97 ÂľV, P = 0.027) and the apical electrode (mean difference, 1.80 ÂľV, P = 0.027).
132
eCAEPs in prelingually deaf CI users
Fig. 5. N1 (A) and N1-P2 (B) amplitudes of the prelingual (pre) and postlingual (post) groups presented for the three stimulation electrode locations, i.e. basal, middle and apical. For both N1 and N1-P2 amplitude, a gradual increase in amplitude was found with a more apical location of stimulation. The box plots represent the lower and upper quartile with the median. Whiskers indicate the 5-95 percentiles. * P < 0.05.
The N1-P2 amplitude (Fig. 5.b) showed similar trends as the N1 amplitude, albeit that the difference in N1-P2 amplitude between the two groups was not significant (F(1,20) = 2.31, P = 0.14). The N1-P2 amplitudes varied with stimulating electrode location (F(2,40) = 6.64, P = 0.0032); specifically, the amplitudes at the apical electrode were significantly larger than at the basal electrode (Bonferroni, mean difference 2.39 ÂľV, P = 0.016). Stepwise multiple linear regression analyses showed that age at implantation, age at testing or implant experience were not predictive for N1 amplitudes on either of the three stimulating electrode locations. Figure 6 shows a joint N1-amplitude-latency scatter plot revealing remarkably clear clustering of the two groups. The prelingual CI users mainly displayed large N1 amplitudes with short latencies, whereas the postlingual group had long latencies and/or small amplitudes. This clustering was most clear on the middle (Fig. 6) and apical (not shown) electrodes. Note that the three prelingual subjects with high speech perception scores (indicated with open circles) tended to be near the border between the two clusters.
133
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig. 6. N1 amplitude as a function of N1 latency presented for the middle electrode reveals clustering of the two groups. Blue dots represent individual subjects in the prelingual group; red squares represent postlingual subjects. The three prelingual subjects with postoperative CVC scores of ≥ 66% are marked as a blue open circle: subject Pre 3 (CVC score 82%, N1 latency 90 ms, N1 amplitude 3 µV), Pre 5 (CVC score 66%, N1 latency 92 ms, N1 amplitude 5.4 µV), Pre 11 (CVC score 77%, N1 latency 85 ms, N1 amplitude 2.2 µV).
Relationship between CAEP and speech perception As demonstrated in Figure 2, a CAEP with a typical N1-P2 waveform could be recorded irrespective of speech perception score and onset of deafness. Since the poorly performing prelingual group demonstrated shorter N1 latencies, N1 latency might be related to speech perception. There was only a fair correlation between speech perception and N1 latency in the prelingual group when stimulated at an apical electrode (r2 = 0.43, P = 0.028), independent of implant experience or age at test. On the basal and middle electrode this correlation was not present (Fig. 7). In the prelingual group, there were no significant correlations between speech perception and P2 latency or N1, N1-P2 and P2 amplitudes on any of the three stimulating electrode locations (r2 < 0.28, P > 0.093). Speech perception scores in the postlingual group were not related to the latencies or amplitudes of the N1 and P2 peaks on either of the three stimulating electrode locations (r2 < 0.30, P > 0.082).
134
eCAEPs in prelingually deaf CI users
Fig. 7. CVC phoneme score as a function of N1 latency presented for the middle electrode. Each red square represents a subject in the postlingual group. Each blue dot represents a subject in the prelingual group.
Discussion In this study we assessed cortical activity in pre- and postlingually deaf adult CI users by recording CAEP waveforms in response to electrical stimuli presented at basal, middle, and apical CI electrodes. Given the current literature on the development of the CAEP response, the CAEP waveforms including clear N1-P2 peaks that we obtained in prelingually deaf patients, who received their CI only in adulthood and supposedly had an abnormal development of the auditory cortex, were highly unexpected. While on first visual inspection their waveforms were comparable to those obtained in postlingually deaf CI users, who supposedly had a normal development of the auditory cortex during childhood, there seems to be a consistent difference in N1 latency and amplitude between both groups, possibly implying different underlying mechanisms. Development of CAEPs with age In normal hearing children, it has been demonstrated that the CAEP waveform morphology alters with increasing age (Sharma et al., 1997; Ponton et al., 2000; Wunderlich and Cone-Wesson, 2006; Wunderlich et al., 2006). Before the age of 9-12 the CAEP response is predominated by a large positive peak (P1), which is characterized by a decreasing peak latency with age (Sharma et al., 1997; Ponton et al., 2000). Only after this period a negativity becomes present, which eventually develops into a robust negative peak (N1) in adulthood dominating the CAEP (Sharma et al., 1997; Ponton et al., 2000; Wunderlich and Cone-Wesson, 2006). The development of the N1 peak is thought to reflect maturation of
135
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
thalamo-cortical and corticocortical pathways in the superficial layers of the cortex, which are essential in the transfer of auditory input to the cortex and the communication between the two hemispheres and other cortical areas (Ponton et al., 2000; Ponton and Eggermont, 2001; Eggermont and Ponton, 2003). Since the emergence of the N1 peak coincides with the development of complex auditory skills in children it is suggested that these pathways play an important role in the processing of sounds in complex conditions such as speech in noise (Ponton et al., 2000; Eggermont and Ponton, 2003). In prelingually deaf children, who have been fitted with a cochlear implant early in life, the maturation of the predominant positive P1 peak follows the normal pattern as seen in normal hearing children (Sharma et al., 2002; Dorman et al., 2007). In such children, a normal N1 may develop after long use of the CI (Jiwani et al., 2013). After long durations of deafness (> 6 years) there are signs that the maturation of the more superficial layers of the auditory cortex becomes impaired, resulting in the absent development of the N1 and the predominance of the positive peak in these, prelingually deaf and late-implanted children (age at implantation > 6.5 years) (Sharma et al., 2002; Dorman et al., 2007). Also, shorter periods of deafness (~3 years) during early childhood (3.5 â&#x20AC;&#x201C; 6.5 years of age) may block N1 development (Ponton and Eggermont, 2001). With the assumed association of N1 to normal-like cortical development it is puzzling that of prelingually deaf children with several years of auditory deprivation and long CI experience, the ones with poor to fair speech perception show large N1 peaks (Gordon et al., 2008). If existence of an N1 peak reflects normal-like cortical development, one would expect that subjects with prelingual deafness who have received their CI in adulthood would exhibit abnormal CAEP waveforms, in particular missing the N1 around 100 ms. On the contrary, without exception, all prelingually deaf adults that were included in this study did display on first visual inspection the typical, adult P1-N1-P2 morphology (Figs. 2 and 3). Furthermore, the responses in this prelingual group were not different from adult CI users, who have had a normal auditory development during childhood (up to age 16, postlingual group). Our data are in line with the findings of Gordon et al. (2008) in prelingually deaf children, first of all in the sense that a significant N1 can be generated in spite of abnormal cortical development. Furthermore, our study compares with Gordon et al. (2008) with respect to the N1 latency. The poor performing children in the latter study demonstrated a significantly shorter N1 latency (~80 ms) than normal hearing children (~100 ms). In our study, we found a comparable difference in N1 latency between the prelingual and the postlingual group of approximately 15 ms. To a lesser extent, there is agreement in N1 amplitude. We found significantly larger amplitudes in the prelingual than in the postlingual group while Gordon et al reported larger than normal N1 peaks in the poor-performing CI children. In contrast to the study by Gordon et al. (2008), who found a great difference in N1 amplitude between poor/fair performers versus good performers, the N1 amplitude found in our study was only weakly associated with speech perception. Since the adult-like CAEP responses found in our study were not related
136
eCAEPs in prelingually deaf CI users
to duration of implant use, the maturation of the CAEP response might be more driven by age-related changes of the innate auditory pathway independent of auditory stimulation. Future longitudinal studies should address these issues by studying the development of the CAEP waveform morphology over time in prelingual CI users.
CAEP differences between postlingually and prelingually deaf subjects Although the CAEP waveform morphology in the prelingual group might be similar to those in the postlingual group, the generation sites of the different peaks may differ between prelingual and postlingual CI users. The waveforms and peak latencies in the postlingual group are comparable to those found in normal hearing subjects. The average N1 latencies of around 105 ms are comparable to those reported in normal hearing subjects (90 â&#x20AC;&#x201C; 120 ms: (Naatanen and Picton, 1987; Wunderlich and Cone-Wesson, 2006; Martin et al., 2008). The normal-like CAEPs of the postlingually deafened adults follow from the development of the CAEPs in normal-hearing children and probably represent activation of the previously matured auditory pathway. The presence of the typical CAEP response in the absence of sufficient auditory stimulation, however, needs a different explanation. It may be explained by activation of the more innate auditory network in conjunction with residual plasticity which is probably still present even after long durations of deafness (Kral and Sharma, 2012). The consistent early N1 latency might also be a representation of activation of these innate and less complex components of the auditory system. Due to the impaired sensory input, the normal exuberant development and axonal selection might have become impaired, leading to a disrupted selection of axonal populations within the auditory cortex (Innocenti and Price, 2005). In turn, this might lead to activation of a large number of less specific axons, rather than a few specialized neurons (Innocenti and Price, 2005). Since synaptic pruning is thought to be completed by around 12 years of age in the auditory cortex (Huttenlocher and Dabholkar, 1997), these aberrant cortical connections might last and be stabilized due to the long term auditory deprivation in our prelingually deaf group (Innocenti and Price, 2005; Uhlhaas and Singer, 2011). Kral et al. (2006) demonstrated that the naĂŻve auditory cortex of congenitally deaf cats has a reduced sensitivity to auditory input, leading to a smeared representation of auditory stimuli with a deficient representation of auditory features including loss of tonotopy (Kral et al., 2006). Thus, a larger cortical area responds to a stimulus, leading to a larger evoked potential. Furthermore, due to long-term auditory deprivation, corticocortical de-coupling might occur in these prelingually deaf CI users resulting in less corticocortical interactions with other brain regions and less influence of top-down inhibition by higher order neurons (Kral and Eggermont, 2007; Kral and Sharma, 2012). The combination of these factors might result in a wide spread of evoked activity which promotes more synchronous activation of several axonal population, giving rise to the early and large N1 peak. 137
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Our findings that typical CAEP responses could be evoked regardless of postoperative performance (Fig. 2), suggest a limited role of the CAEP as an objective predictor of postoperative speech perception. In contrast to our study, studies in children with auditory neuropathy spectrum disorder suggested that CAEPs can be applied as objective predictor as they found that poor performing children revealed longer P1 latencies (Alvarenga et al., 2012; Cardon and Sharma, 2013). Although our study population did not display consistent relations between speech perception and CAEP morphologies (Fig. 7), the results do demonstrate that the prelingual CI users mainly display large N1 amplitudes with short latencies (Fig. 6). These responses can therefore be helpful in the assessment of developmental state of the auditory cortex. Neither age at test nor implant experience explained the differences found in N1 latency and N1 amplitude between the two groups. In the prelingual group, we included subjects with 0.5 to almost 9 years of implant experience (Table 1). If implant experience would affect N1 latency and amplitude, then we would at least expect a certain trend to be present in this group with this broad range in implant experience. Given the absent relation in both groups we are confident that the N1 differences can be attributed to the age of onset of deafness, rather than to implant experience. Further, it should be noted that the current levels applied to evoke the CAEPs cannot explain the larger and earlier responses in the prelingual group. In fact, because of the lower comfortable levels, the levels used for the prelingually deaf subjects were lower than for the postlingually deaf. This would contribute to smaller and later CAEPs, which indicates that the found differences would have been even larger when same current levels were applied. One may wonder whether the presence of the N1peak is related to the choice of stimulus. The group of Sharma (Sharma et al. 2002; Dorman et al. 2007) who reports a lack of N1 used a different stimulus, a 90-ms speech segment acoustically delivered to the CI processor, than studies reporting an N1 (Gordon et al. 2008; Jiwani et al. 2013; current study) who used a short train of pulses presented through a single electrode or a pair of electrodes. However, Ponton and Eggermont (2001) reporting a lack of N1 also used a short pulse train. Therefore, and because various stimuli as tones, clicks, speech-like sounds, and acoustic changes yield similar CAEP waveforms including the N1 peak around 100 ms (Wunderlich and Cone-Wesson 2006; Martin et al. 2008), the type of stimulus does not seem to be an important factor for this onset response. The rate we used is somewhat slower (0.9 Hz) than the rate of Ponton and Eggermont (2001) and Sharma et al. (2002) who used 1.3â&#x20AC;&#x201C;1.4 Hz, and although the N1 decreases with increase of stimulation rate, it seems unlikely that this difference in rates can explain the different results since only at high rates (93 Hz) the N1 disappears (Wunderlich and Cone-Wesson 2006).
138
eCAEPs in prelingually deaf CI users
Patient selection Despite the careful selection of our prelingually deaf subjects, one could argue that possible residual hearing might have affected the maturation of the CAEP waveforms in these subjects. Although we cannot exclude that some subjects (e.g., Pre 5; see Table 1) used little residual hearing in conjunction with acoustical amplification for sound detection, as in all other prelingual subjects, their deafness has led to serious impairment of their speech production and language skills. Other subjects explicitly stated that they were deaf during childhood and reported that even as a child they did not use their hearing aids due to the absent value, such as subject Pre 7. Also, these subjects had typical CAEP waveforms (Fig. 2). Another issue of debate is the assessment of onset of deafness. Determining the age at which adult individuals became deaf is often not straightforward since it is based on patientreported data about age of first suspicion of hearing impairment, age of first diagnostic audiometry and its outcome, and age of first hearing aid use. In particular, one could wonder in this respect if the three prelingually deaf subjects with high postoperative CVC scores in the same range as the postlingually deaf individuals (Fig. 1.a), were correctly judged as prelingually deaf. Although their N1-latencies were in the range of the other prelingually deaf individuals (Fig. 7), the joint N1-amplitude-latency scatter plot (Fig. 6) indicates these three cases near the border between the prelingual and postlingual clusters. Based on these N1 data, one might surmise that (some of) these three cases have been misjudged as being prelingually deaf. Nevertheless, we conclude that even if some subjects were misjudged according to strict definitions, the objective N1 measures do indicate that the prelingually labeled subjects differed in auditory development from the postlingually labeled subjects. Methodological considerations Since we did not make use of multichannel recordings several stimulus artifact reduction methods proposed in the literature, such as independent component analysis, could not be applied. In order to confidently differentiate the actual cortical response from a possible stimulus artifact, we used several techniques as described in the methods section, including the technique of altering the inter-stimulus interval as proposed by Friesen and Picton (2010). Furthermore, we were able to replicate and reproduce all CAEP responses in all subjects on different stimulating electrodes. Although in some subjects a residual stimulus artifact can still be seen in the unfiltered waveforms, presented in figure 2, we are confident that we could clearly separate CAEP response from stimulus artefacts in all cases, since we used the aforementioned methods to differentiate CAEP response from stimulus artifact. Due to the limited number of recording channels used in our setup, source analysis of the N1 generators could not be performed. Although it would be of interest to assess whether the generators of the N1 would differ between postlingually deaf CI users and late implanted prelingually deaf CI users, the spatial resolution of EEG source localization might be too low to reveal small differences, within the auditory cortex (Pascual-Marqui, 2002). 139
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Conclusion Results from the present study demonstrate that cortical auditory evoked potentials with a typical N1-P2 waveform can be evoked in prelingually deaf and very late implanted cochlear implant users. N1 peak latencies are however significantly earlier and N1 amplitudes are larger in this group, which might be caused by the activation of more innate and less complex auditory cortical network.
Acknowledgements The authors thank all participants for their time and support and the members of the cochlear implant team of the UMC Utrecht for their contributions. This study is supported by an unrestrictive research grant from Cochlear Ltd. Conflict of Interest Disclosure Statement: Wilko Grolman received an unrestrictive research grant from Cochlear Ltd. for this study. Wilko Grolman received unrestrictive research grants from MED-EL GmbH, and Advanced Bionics. No competing interests declared by the other authors.
140
eCAEPs in prelingually deaf CI users
References Alvarenga KF, Amorim RB, Agostinho-Pesse RS, Costa OA, Nascimento LT, Bevilacqua MC (2012) Speech perception and cortical auditory evoked potentials in cochlear implant users with auditory neuropathy spectrum disorders. International Journal of Pediatric Otorhinolaryngology 76:1332-1338. Cardon G, Sharma A (2013) Central auditory maturation and behavioral outcome in children with auditory neuropathy spectrum disorder who use cochlear implants. International Journal of Audiology 52:577586. Dorman MF, Sharma A, Gilley P, Martin K, Roland P (2007) Central auditory development: evidence from CAEP measurements in children fit with cochlear implants. Journal of Communication Disorders 40:284-294. Doucet ME, Bergeron F, Lassonde M, Ferron P, Lepore F (2006) Cross-modal reorganization and speech perception in cochlear implant users. Brain 129:3376-3383. Eggermont JJ, Ponton CW (2003) Auditory-evoked potential studies of cortical maturation in normal hearing and implanted children: correlations with changes in structure and speech perception. Acta Otolaryngologica 123:249-252. Friesen LM, Picton TW (2010) A method for removing cochlear implant artifact. Hearing Research 259:95-106. Gordon KA, Tanaka S, Wong DD, Papsin BC (2008) Characterizing responses from auditory cortex in young people with several years of cochlear implant experience. Clinical Neurophysiology 119:2347-2362. Huttenlocher PR, Dabholkar AS (1997) Regional differences in synaptogenesis in human cerebral cortex. The Journal of Comparative Neurology 387:167-178. Innocenti GM, Price DJ (2005) Exuberance in the development of cortical networks. Nature Reviews Neuroscience 6:955-965. Jiwani S, Papsin BC, Gordon KA (2013) Central auditory development after long-term cochlear implant use. Clinical Neurophysiology 124:1868-1880. Klop WM, Briaire JJ, Stiggelbout AM, Frijns JH (2007) Cochlear implant outcomes and quality of life in adults with prelingual deafness. Laryngoscope 117:1982-1987. Kral A, Eggermont JJ (2007) What’s to lose and what’s to learn: development under auditory deprivation, cochlear implants and limits of cortical plasticity. Brain Research Reviews 56:259-269. Kral A, O’Donoghue GM (2010) Profound deafness in childhood. The New England Journal of Medicine 363:1438-1450. Kral A, Sharma A (2012) Developmental neuroplasticity after cochlear implantation. Trends in Neuroscience 35:111-122. Kral A, Tillein J, Heid S, Klinke R, Hartmann R (2006) Cochlear implants: cortical plasticity in congenital deprivation. Progress in Brain Research 157:283-313. Lee HJ, Giraud AL, Kang E, Oh SH, Kang H, Kim CS, Lee DS (2007) Cortical activity at rest predicts cochlear implantation outcome. Cerebral Cortex 17:909-917. Martin BA, Tremblay KL, Korczak P (2008) Speech evoked potentials: from the laboratory to the clinic. Ear and Hearing 29:285-313. Naatanen R, Picton T (1987) The N1 wave of the human electric and magnetic response to sound: a review and an analysis of the component structure. Psychophysiology 24:375-425. Pascual-Marqui RD (2002) Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details. Methods and Findings in Experimental and Clinical Pharmacology 24 Suppl D:5-12. Ponton CW, Eggermont JJ (2001) Of kittens and kids: altered cortical maturation following profound deafness and cochlear implant use. Audiology and Neurotology 6:363-380. Ponton CW, Eggermont JJ, Kwong B, Don M (2000) Maturation of human central auditory system activity: evidence from multi-channel evoked potentials. Clinical Neurophysiology 111:220-236.
141
2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 2.3
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Sharma A, Dorman MF, Spahr AJ (2002) A sensitive period for the development of the central auditory system in children with cochlear implants: implications for age of implantation. Ear and Hearing 23:532-539. Sharma A, Kraus N, J. McGee T, Nicol TG (1997) Developmental changes in P1 and N1 central auditory responses elicited by consonant-vowel syllables. Electroencephalography and Clinical Neurophysiology 104:540-545. Teoh SW, Pisoni DB, Miyamoto RT (2004) Cochlear implantation in adults with prelingual deafness. Part I. Clinical results. Laryngoscope 114:1536-1540. Uhlhaas PJ, Singer W (2011) The development of neural synchrony and large-scale cortical networks during adolescence: relevance for the pathophysiology of schizophrenia and neurodevelopmental hypothesis. Schizophrenia Bulletin 37:514-523. Versfeld NJ, Daalder L, Festen JM, Houtgast T (2000) Method for the selection of sentence materials for efficient measurement of the speech reception threshold. Journal of the Acoustical Society of America 107:1671-1684. Wunderlich JL, Cone-Wesson BK (2006) Maturation of CAEP in infants and children: A review. Hearing Research 212:212-223. Wunderlich JL, Cone-Wesson BK, Shepherd R (2006) Maturation of the cortical auditory evoked potential in infants and young children. Hearing Research 212:185-202.
142
Part III The clinical use of objective measures and the Acoustic Change Complex
Chapter 3.1 The clinical feasibility of postoperative evoked potential recordings in cochlear implant users
M.J.W. Lammers, H. Versnel, G.A. van Zanten, W. Grolman
Submitted
Chapter 3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract Objective: To evaluate the success rate of postoperative recordings of electrically evoked compound action potentials (eCAP), auditory brainstem responses (eABR) and cortical auditory evoked potentials (CAEP) in cochlear implant users. Design: Eleven adults with unilateral cochlear implants and two adults with bilateral implants were enrolled in this study. Fifteen ears were evaluated at three electrode locations (basal, middle, apical). Results: The postoperative eCAP recording success rate ranged from 53% to 62% for the three electrode locations. The success rate of the eABR recordings was lowest when evoked by a basal electrode (64-70%) and increased to 86% at the apical electrode. CAEPs could be obtained in 93% of the implanted ears for all three electrode locations. The possibility to evoke these recordings was not associated with postoperative functioning. Conclusions: Given the good to excellent postoperative success rate of recording eABRs and CAEPs, we recommend to perform these measures if eCAPs are not measurable or additional information on the auditory pathway is required.
146
Feasibility of evoked potentials
Introduction Objective peripheral and central auditory responses can be useful for postoperative speech processor programming of specific patient populations, such as children, or for evaluating cochlear implant malfunctioning. It is therefore essential that these measures can be reliably recorded in all patients. The most widely used objective response is the electrically evoked compound action potential (eCAP) and is nowadays frequently used during surgery to evaluate implant integrity and to program the processors after hook up. Although intraoperative eCAP recordings can be measured for almost all electrodes, in the postoperative setting the results are more inconsistent varying from approximately half of the electrodes (Smoorenburg et al., 2002; van Wermeskerken et al., 2006) to almost 90% (Cafarelli Dees et al., 2005). Other electrophysiological measures, like the electrically evoked auditory brainstem responses (eABR) and cortical auditory evoked potentials (CAEP) can be a valuable alternative (Abbas and Brown, 2015). Although they are more time-consuming than measuring eCAPs and require EEG recording systems, they can provide additional information on the neural conduction in the auditory pathway and might be used as an alternative objective measure if eCAP recordings fail. Before they can be implemented in the daily clinic, it is essential to know whether these recordings can be evoked reliably in these patients. In this study, we evaluate the success rate of postoperative recordings of eCAP, eABR and CAEP in cochlear implant users and examine if eABR and/or CAEP can be a reliable alternative if eCAP recordings fail.
Materials and Methods Participants Adult users of a Cochlear速 CI who visited the outpatient clinic from December 2011 to December 2012 were consented to participate in this study. Eleven adults with a single CI and two adults with two CIs agreed to participate and could be enrolled in this study, resulting in a total number of fifteen implanted ears which could be used for our analyses. All participants were users of a Cochlear Nucleus Freedom, CI512 or CI513 implant and had at least 6 months experience with their implants. Detailed demographic information of these subjects is listed in Table 1.
147
3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
148
Right Left Right Right Right Left Left Left Left Left Left Left Right Left Right
1R 1L 2 3 4R 4L 5 6 7 8 9 10 11 12 13
Meningitis Meningitis Congenital Congenital Otitis media Otitis media Otosclerosis Rubella Varicella Rubella Meningitis Meningitis Congenital Unknown Rubella
CI512 Freedom Freedom CI512 Freedom CI512 Freedom CI513 CI513 CI513 Freedom Freedom CI512 Freedom CI512
CI type
Age at test (years) 23 23 49 38 80 80 66 56 55 47 55 47 43 31 42
Age at onset deafness (years) 0.5 0.5 0 0 64 64 30 0.3 2 0 0.8 0.8 0 0 0 21 19 49 37 75 79 64 54 55 46 49 41 40 27 41
Age at CI (years)
Implant experience (years) 1.7 3.3 3.3 0.6 5.3 1.3 1.9 2.5 0.6 1.5 5.7 5.7 2.9 3.2 1.4 22 67 40 0 89 74 84 82 26 66 18 28 0 15 0
+ + + + ± + + + ± + +
Post-op CVC eCAP score (%) + + + + + + + + + + ± + +
eABR + + + + + + + + + + + + + +
eCAEP
Recordings: +: on at least 2 of 3 electrodes successful recordings feasible, ±: on only 1 electrode successful recordings, -: no successful recordings feasible on any electrode. The eCAP and eABR success rates when recorded with an interphase gap of 8 or 58µs are combined.
Side CI
Subject
Etiology
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Table I Subject demographics
Chapter 3.1
Feasibility of evoked potentials
eCAP recordings eCAPs were evoked by a biphasic pulse with a pulse width of 25 µs and an interphase gap (IPG) of 8 or 58 µs. Stimuli were generated using the Cochlear Custom Sound EP 3.1 software and presented at a rate of 80 Hz. Stimulus level was increased with steps of 5 current units (CU) until the individual’s maximum comfortable loudness level was reached. eCAPs were recorded by stimulating a basal (typically No. 2), a middle (typically No. 11) and an apical (typically No. 20) electrode. The eCAP threshold was defined as the lowest stimulation level at which a clear N1-P1 waveform could be evoked. An amplitude growth function was judged successful if a total of three or more eCAP responses on different stimulus levels could be recorded. eABR and eCAEP recordings The eABRs were evoked by a biphasic pulse, with a phase width of 25 µs and an interphase gap of 8 or 58µs at a rate of 35 Hz at the individual’s maximum comfortable loudness level. eCAEPs were generated by presenting a 36 ms pulse train at 250 pulses per second and presented at a rate of 0.9 Hz at the individual’s maximum comfortable loudness level. Both eABRs and eCAEPs were recorded by stimulating a basal (typically No. 2), a middle (typically No. 11) and an apical (typically No. 20) electrode. Responses were recorded differentially between electrodes positioned at Cz and Fpz and the contralateral mastoid. An eABR recording was defined as successful if clear and reproducible waves III and V were identified. The presence of the eCAEP waveform was confirmed if clear and reproducible waveforms, with the typical N1-P2 morphology, could be obtained. More detailed stimulus and recording paradigms for the eABR and CAEP recordings are described in previous studies by our group (Lammers et al., 2015a; Lammers et al., 2015b).
Results Postoperative eCAP responses could be obtained in 53% to 62% (Fig. 1) and was independent of electrode location (X2(2) = 0.230, p = 0.892) or IPG of the biphasic pulse (X2(1) < 0.419, p > 0.518). Postoperative eCAP thresholds evoked by the 58 µs IPG pulse were on average 10 CU lower than those evoked by the 8 µs IPG pulse (t(14) = 7.507, p < 0.0001). Amplitude growth functions could be reliably evoked on 33% (middle) to 53% (apical) of the electrodes. At the middle and apical electrodes eABRs could be recorded in 86% of the implanted ears, whereas this number decreased to 64% on the basal electrode (X2(2) = 2.746, p = 0.253). If only the presence of eABR wave V was used as the criterion for a successful recording, success rates increased with 10-20%. Increasing the IPG to 58 µs resulted in a higher success rate of recording eABRs for all electrode locations (varying from 6% to 26%), but this difference was not significantly different (X2(1) < 2.397, p > 0.1). 149
3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
In all but one subject (93%), CAEPs could be evoked (Fig. 1). The most reliable and clearest eCAEP waveforms were obtained by stimulating the middle or apical electrodes; at the most basal electrode, recordings were more likely to be distorted by electrical artefacts. The success rate of recording eCAPs, eABRs or eCAEPs was independent of speech perception or implant experience (linear regression analyses p > 0.05 for all electrodes and conditions).
Fig. 1. Postoperative eCAP, eABR and eCAEP success rates of recording presented for the three stimulating electrode locations. The eCAP and eABR success rates represent the average of success rates obtained using an interphase gap of 8 and 58µs. Due to time constraints, eCAPs on the apical electrode were only obtained in 13 ears, and eABRs were obtained in 14 ears. eCAEPs were obtained in all 15 ears on all electrodes.
Alternative recordings after failed eCAP response In 6 of the 15 implanted ears (40%) eCAPs could not be evoked on two or more electrodes. Instead, eABRs could be recorded in 67% and eCAEPs in 83% of these ‘unsuccessful’ ears, irrespective of recording electrode. In the remaining 9 ‘successful’ cases, clear eCAPs could be evoked in 78% on the basal and middle electrode and in all cases on the apical electrode. Their eABR success rate was 78% for the basal electrode, 89% for the middle electrode and 100% for the apical electrode. eCAEPs could reliably be evoked on all electrodes in all these ‘successful’ subjects.
150
Feasibility of evoked potentials
Discussion As the success rate of postoperative eCAP measures in this study is relative poor (53%- 62%), our results demonstrate that eABRs or eCAEPs can be reliably recorded instead. The eABR success rate is higher than 64-70% on all electrode locations and eCAEP waveforms could be recorded in almost all cases (Fig. 1). Although in the literature a wide variability in the postoperative success rate of recording eCAPs has been reported, ranging from 44 to 90% (Cafarelli Dees et al. 2005; Smoorenburg et al. 2002; van Wermeskerken et al. 2006), the success rates found in these subjects with the newer generation of cochlear implants was still relative poor. This might be related to the stimulation paradigm used in this study, which is different from the automated eCAP paradigms used in other studies. In line with eCAP measures in animal models (Prado-Guitierrez et al., 2006; Ramekers et al., 2014) and psychophysical outcomes in CI patients (McKay and Henshall, 2003; Carlyon et al., 2005), increasing the IPG was found to result in a more efficient stimulus, leading to a decrease in threshold. The number of subjects in whom eCAPs and eABRs could be evoked increased slightly using a larger IPG. Although this effect was not statistically significant, increasing the IPG may be a useful adjustment when measuring these responses in the clinic. The eABR recording success rate ranging from 73% to 87% is in line with a previous study which demonstrated that postoperative eABRs could be evoked in around 80% of the patients (Firszt et al., 2002). Figure 1 displays a tendency towards a lower success rate of recording the eABR when evoked by the more basal electrodes. This might be explained by location effects found in the current study and in previous studies indicating that eABR peak amplitudes on the more basal electrodes are smaller and waveform morphology is less clear (Firszt et al., 2002; Gordon et al., 2007; Lammers et al., 2015b). Electrical artifacts did also diminish the signal-to-noise ratio of the CAEP recordings at especially the more basal stimulating electrodes (Lammers et al., 2015a). However, due to their larger signal and the possibility to separate the time windows of electrical stimulation (0-36 ms) and P1-N1-P2 complex (50-250 ms), it is possible to obtain responses in a large number of patients, even if the more basal electrodes are stimulated.
Conclusion The poor to moderate success rate of the postoperatively recorded eCAP may limits its clinical value. On the other hand, the success rate of postoperative eABR and eCAEP recordings was good to excellent, postulating their role as reliable alternatives for eCAP recordings. Hence, we recommend to perform eABR or eCAEP recordings in cases in which eCAPs are not measurable or additional information on the auditory pathway is required. 151
3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.1
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Acknowledgements The authors thank all participants for their time and support, and the members of the cochlear implant team of the UMC Utrecht for their contributions.
152
Feasibility of evoked potentials
References Abbas PJ, Brown CJ (2015) Assessment of responses to cochlear implant stimulation at different levels of the auditory pathway. Hearing Research 322:67-76. Cafarelli Dees D et al. (2005) Normative findings of electrically evoked compound action potential measurements using the neural response telemetry of the Nucleus CI24M cochlear implant system. Audiology and Neurotology 10:105-116. Carlyon RP, van Wieringen A, Deeks JM, Long CJ, Lyzenga J, Wouters J (2005) Effect of inter-phase gap on the sensitivity of cochlear implant users to electrical stimulation. Hearing Research 205:210-224. Firszt JB, Chambers RD, Kraus, Reeder RM (2002) Neurophysiology of cochlear implant users I: effects of stimulus current level and electrode site on the electrical ABR, MLR, and N1-P2 response. Ear and Hearing 23:502-515. Gordon KA, Papsin BC, Harrison RV (2007) Auditory brainstem activity and development evoked by apical versus basal cochlear implant electrode stimulation in children. Clinical Neurophysiology 118:16711684. Lammers MJ, Versnel H, van Zanten GA, Grolman W (2015a) Altered cortical activity in prelingually deafened cochlear implant users following long periods of auditory deprivation. Journal of the Association for Research in Otolaryngology 16:159-170. Lammers MJ, van Eijl RH, Versnel H, van Zanten GA, Grolman W (2015b) Delayed Auditory Brainstem Responses in Prelingually Deaf and Late-Implanted Cochlear Implant Users. Journal of the Association for Research in Otolaryngology doi:10.1007/s10162-015-0532-x. McKay CM, Henshall KR (2003) The perceptual effects of interphase gap duration in cochlear implant stimulation. Hearing Research 181:94-99. Prado-Guitierrez P, Fewster LM, Heasman JM, McKay CM, Shepherd RK (2006) Effect of interphase gap and pulse duration on electrically evoked potentials is correlated with auditory nerve survival. Hearing Research 215:47-55. Ramekers D, Versnel H, Strahl SB, Smeets EM, Klis SF, Grolman W (2014) Auditory-nerve responses to varied inter-phase gap and phase duration of the electric pulse stimulus as predictors for neuronal degeneration. Journal of the Association for Research in Otolaryngology 15:187-202. Smoorenburg GF, Willeboer C, van Dijk JE (2002) Speech perception in nucleus CI24M cochlear implant users with processor settings based on electrically evoked compound action potential thresholds. Audiology and Neurotology 7:335-347. van Wermeskerken GK, van Olphen AF, van Zanten GA (2006) A comparison of intra- versus post-operatively acquired electrically evoked compound action potentials. International Journal of Audiology 45:589594.
3.1
153
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2 Cortical auditory evoked potentials to frequency changes with varied size, velocity and direction
M.J.W. Lammers, H. Versnel, G.A. van Zanten, W. Grolman
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Abstract This study investigates the effects of frequency modulations on cortical processing. In eight normal hearing subjects, acoustic change complexes (ACC) were evoked by introducing frequency changes with varying size, velocity, and direction of the frequency change. The results demonstrated that the N1 and N1-P2 amplitudes of the ACC waveform increased with increasing size and velocity for both up- and downward frequency changes. Amplitudes of the P2 peak were only associated with the size of the frequency, but not with velocity. Latencies of the N1 and P2 peaks decreased with increasing frequency step and velocity. Analyses of the direction selectivity identified a downward preference for fast and small frequency changes. If the speed of the frequency change decreased or step size was increased, no significant directional preferences were seen. Overall, the results demonstrate that acoustic change complex waveforms are dependent on the size and velocity of frequency changes, indicating enhanced neural synchrony in response to larger and/or faster auditory changes.
156
CAEPs and frequency changes
Introduction The ability of our auditory system to detect modulations within ongoing sounds is essential for daily life. It enables us to detect changes in our environment, to identify vowels and consonants, but also to appreciate pitch differences in musical compositions. Normal hearing can identify very small frequency changes of less than 1% of the base frequency (Amitay et al., 2006; Papakonstantinou et al., 2011; Sek et al., 1995). In the case of sensorineural hearing loss, frequency resolution capabilities become impaired, resulting in poorer speech understanding in especially noisy environments (Dreschler et al., 1985; Horst, 1987; Noordhoek et al., 2001; Papakonstantinou et al., 2011; Strelcyk et al., 2009). Psychophysical experiments revealed that patients with sensorineural hearing loss had frequency discrimination thresholds which correlated well with their speech perception in noise abilities (Noordhoek et al., 2001; Papakonstantinou et al., 2011). These findings indicate the importance of frequency discrimination in oral communication. The underlying neurophysiologic alterations in response to frequency changes have been investigated using cortical auditory evoked potentials (Arlinger et al., 1981; Dimitrijevic et al., 2008; Harris et al., 2008; McCandless et al., 1970; Pratt et al., 2009). The obligatory cortical auditory evoked potential is thought to be the resultant of neuronal activation in the thalamus and auditory cortex in response to the onset or change in auditory stimuli. The response evoked by a change in a continuous stimulus is often called the acoustic change complex (ACC). This auditory change response can be evoked using natural speech (Friesen et al., 2006; Martin et al., 2000; Ostroff et al., 1998; Tremblay et al., 2003) or continues tones (Arlinger et al., 1981; Dimitrijevic et al., 2008; Harris et al., 2008; McCandless et al., 1970; Pratt et al., 2009). These studies already revealed that the amplitude of the ACC waveform growths with increasing size of the frequency change. Harris et al. 2008 demonstrated that ACCs can even be evoked in response to small changes of less than 1% of the base frequency, closely resembling behavioral just noticeable frequency discrimination results (Harris et al., 2008). Although there is consistent evidence that the ACC amplitude is related to the size of the frequency change, little is known on other factors such as the direction of the frequency change, i.e. frequency increase or decrease, or the velocity of the frequency modulated sweep of the auditory change. In the current study, frequency changes with varying size, velocity and direction were used to elicit the ACC in young, normal hearing subjects. We hypothesized that ACC amplitude increases with increasing size and velocity of the frequency change. Based on previous studies (Maiste et al., 1989; Pratt et al., 2009), we expect to find larger ACC amplitudes in response to a frequency increase.
157
3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Methods Subjects Eight healthy, normal hearing volunteers, aged between 24 and 27 years old agreed to participate in this study. Hearing thresholds of each participant measured by standard pure tone audiometry prior to the experiments were < 20 dB HL at each frequency (0.25 - 8 kHz). All subjects did not report a history of hearing loss or tinnitus and were right handed.
Fig. 1. A. Schematic presentation of the stimulus. Stimuli consisted of a 1000Hz pure tone with a duration of 3000ms, an upward or downward frequency modulated sweep, followed by a 300ms continuous tone with a frequency of 1000Hz +/- Î&#x201D;f. An inter stimulus interval of 200ms was provided before the onset of the next stimulus. The insert above demonstrates the corresponding acoustic change response and cortical auditory evoked response elicited by the onset of the new stimulus. B. Schematic overview of all 24 stimulus conditions. Purple lines indicate the fastest sweeps of 85.7 octave/sec, the red lines of 28.6 octave/s, the green lines of 9.5 octave/sec, and the blue lines of 3.2 octave/s. Frequency changes of 0.029, 0.086, and 0.257 octave in both upward and downward direction were used. Downward frequency changes are indicated with a minus sign.
158
CAEPs and frequency changes
Stimuli and recording procedure Acoustic change complexes were evoked using pure tones of 3300 ms including a variable frequency change. The tones consisted of three components a) a reference tone of 1000Hz with a duration of 3000 ms, b) a logarithmic frequency modulation (FM) sweep with a frequency change Δf, c) a 300 ms tone with a frequency 1000-Δf or 1000+Δf Hz (Fig. 1.a). The silent interval between stimuli was 200 ms. The duration of the reference tone was based on pilot data showing larger ACC amplitudes when a reference tone of 3000 ms was used as compared to a tone of 1000 ms. Frequency changes of 0.029 octave (~20 Hz), 0.086 (~60 Hz) and 0.257 (~180 Hz) were used in both the upward (+Δf) and downward (-Δf) direction, resulting in a total of 6 different frequency changes. Four different FM sweep velocities were used: 3.2, 9.5, 28.6 and 85.7 octave/s, giving a total number of 24 conditions for each participant (Fig. 1.b). Stimuli were presented to the left ear in all participants at a level of 75dB SPL. Sound stimuli were generated using MATLAB (version 7.11.0, Mathworks, Natick, MA, USA) and presented monaurally to the left ear through a TDH-39 headphone. All 24 conditions were presented in a random order. Participants were seated in a comfortable reclining chair in an electrically shielded, sound attenuated booth and were allowed to watch a silent, captioned movie. They were carefully instructed prior to each recording to minimize movements and to fixate the center of the video screen to minimize muscle and eye movement artefacts. ACC responses were recorded by Ag/AgCl electrodes placed according to the 10-20 system using a Medelec Synergy T-10 Evoked Potential system. The active electrode was placed at the vertex of the skull, Cz, the contralateral mastoid (right) was used as reference electrode and the ground electrode was placed on the forehead. Eye movements and blinks were monitored using electrodes above and below the eye, contralateral of the stimulated ear. Electrode impedances were kept below 5 kΩ. The electrode signals were filtered from 0.01 to 100 Hz and recorded with a sampling rate of 50 kHz. Responses were acquired in a 1000 ms time-window, consisting of a pre-stimulus period of 100 ms. Responses containing amplitudes of > 100 µV at any electrode were rejected and not included in the averaged response. For each condition 100 accepted sweeps were averaged. Data analyses Averaged evoked potential data were used for determining peak amplitudes and latencies for each subject. The N1 of the ACC was defined as the most negative peak at 70 to 120 ms after stimulus change. P2 was defined as the first pronounced positive peak occurring after N1 at 120 to 250 ms after the change. Statistical analyses were completed using SPSS version 22.0 software (IBM, Armonk, NY, USA). Repeated measures ANOVAs with the three different frequency changes as within-subjects factor and the four velocities as betweensubjects factor were used. Significant main effects and interactions (p < 0.05) were followed
159
3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
with Bonferroni post hoc tests, and the Greenhouse-Geisser correction was applied to compensate for violations of the sphericity assumption. Direction selectivity was computed with the following equation: (Amplitude downwards â&#x20AC;&#x201C; Amplitude upwards) / (Amplitude downwards + Amplitude upwards), as described in more detail by Nelken and Versnel 2000.
Fig. 2. Individual and unfiltered recordings obtained in one subject displaying the acoustic change responses for all 24 conditions separately. The acoustic change complex can be identified in all traces at around 100 ms after the frequency change. The response occurring around 600 ms after the frequency change is the cortical onset response on the next stimulus. The four uppermost traces demonstrate the responses obtained with a frequency change of 0.257 octave, the middle four traces the response obtained with a frequency change of 0.0857 octave, and the bottommost for a frequency change of 0.0286 octave. Purple traces indicate the fastest sweeps of 85.7 octave/sec, the red traces of 28.6 octave/s, the green traces of 9.5 octave/sec, and the blue traces of 3.2 octave/s.
Results Reproducible and clear ACC responses exhibiting the typical P1-N1-P2 waveform morphology could be evoked in all subjects, in all 24 conditions. Figure 2 shows the individual and unfiltered responses evoked in one subject for the 24 different conditions.
160
CAEPs and frequency changes
Fig. 3. N1-P2 (A) and N1 (B) amplitude as a function of size of the frequency change. Purple lines indicate the fastest sweeps of 85.7 octave/sec, red lines of 28.6 octave/s, green lines of 9.5 octave/sec, and the blue lines of 3.2 octave/s. Solid lines represent downward sweeps, dotted lines upward sweeps. Error bars indicate standard errors of the mean.
Effects of frequency change and velocity on amplitude The N1-P2 amplitude of the ACC was significantly affected by the size and the velocity of the frequency step (Fig. 3.a). Amplitudes increased with increasing size of the frequency change, for both upward (F(2,14) = 48.8, P < 0.0001), and downward going frequency changes (F(2,14) = 21.7, P < 0.0001). Besides, N1-P2 amplitudes were significantly influenced by the velocity of the frequency change, regardless of direction of the change (upward changes F(3,21) = 7.7, P = 0.0012; downward changes: F(3, 21) = 10.0, P < 0.001. There was no interaction effect between frequency step and velocity for both directions. The increase of the N1-P2 amplitude (A) with frequency change (Δf) and velocity (v), can be described as follows: A ~ b • log(d) = b • log(Δf/v), with d duration of the FM sweep, thus A ~ = b • log(Δf). The slope b varies from 5.6 for v = 3.2 octave/s to 9.7 for v = 28.6 octave/s (Fig. 4). Amplitudes were strongly correlated to the size of the frequency step for all four different velocities (r2 > 0.86). The N1 amplitudes significantly increased with increasing size and velocity of the change for both upward and downward directions (Fig. 3.b). Amplitudes of the N1 through increased with size of the frequency change (upward changes F(2,14) = 19.6, P = 0.0013); downward changes: F(2,14) = 21.9, P < 0.0001), and velocity of the change (upward changes F(3,21) = 12.0, P < 0.0001); downward changes: F(3,21) = 13.7, P < 0.0001). For both directions there was no interaction effect between frequency step and velocity.
3.2
161
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Fig.4. N1-P2 amplitude as a function of sweep duration. Purple lines indicate the fastest sweeps of 85.7 octave/sec, red lines of 28.6 octave/s, green lines of 9.5 octave/sec, and the blue lines of 3.2 octave/s. The increase of N1-P2 amplitude (A), with frequency change (Δf), can be described as follows: A ~ = b ● log(Δf). The slope b for a velocity of 3.2 octave/s is 5.6, for 9.5 octave/s: 7.78, for 28.6 octave/s: 9.7, and for 85.7 octave/s: 8.34. Downward sweeps are indicated with solid symbols, upward sweeps with open symbols.
Amplitudes of the P2 peak only revealed a significant relation with the size of the frequency step (upward changes F(2,14) = 71.3, P < 0.0001); downward changes: F(2,14) = 10.2, P = 0.0019), but not with the velocity of the change (upward changes F(3,21) = 2.95, P = 0.056); downward changes: F(2,14) = 2.14, P = 0.125). Effects of frequency change and velocity on latency Latencies of the N1 through decreased with increasing frequency step and velocity (Fig. 5.a), for both the upward (frequency step F(2,14) = 48.2, P < 0.0001); velocity F(3,21) = 25.6, P < 0.0001), and downward direction (frequency step F(2,14) = 40.1, P = 0.0001); velocity F(3,21) = 19.0, P < 0.0001). The P2 latencies were also found to decrease with increasing frequency step and velocity (Fig. 5.b), regardless of direction (upward frequency step: F(2,14) = 25.0, P < 0.0001); velocity F(3,21) = 10.7, P < 0.001); downward frequency step F(2,14) = 11.0, P = 0.0091); velocity F(3,21) = 7.8, P = 0.0011).
162
CAEPs and frequency changes
Fig.5. N1 (A) and P2 (B) latencies as a function of size of the frequency change. Purple lines indicate the fastest sweeps of 85.7 octave/sec, red lines of 28.6 octave/s, green lines of 9.5 octave/sec, and blue lines of 3.2 octave/s. Solid lines represent downward sweeps, dotted lines upward sweeps. Error bars indicate standard errors of the mean.
Effects of direction selectivity on amplitudes Analysis of the direction selectivity for both the N1 and N1-P2 amplitudes of the ACC response, revealed that amplitudes tend to be larger for fast and downward changes. One-sample t-tests revealed a significant downward preference for the fastest and smallest frequency change for both the N1 (t = 2.4, P = 0.048) and N1-P2 amplitude (t = 5.8, P = 0.001). Except for the largest frequency change with a velocity of 28.6 octave/s, which revealed larger amplitudes for downward frequency changes (N1 amplitude t = 3.2, P = 0.015), all other conditions were not significantly different from zero. This indicates that for these nine other conditions there was no significant up- or downward direction preference.
Discussion The results of this study show that the amplitude of the acoustic change complex is determined by the size and velocity of the frequency change, i.e. amplitudes increase if the size of the frequency change and/or velocity of the change increase. Interestingly, N1 ACC amplitudes were associated with both size of the frequency change and velocity, whereas the P2 amplitude was only affected by the frequency change and not by velocity. This might suggest that the neural generators of the ACC N1 through are susceptible for both factors, but the neural generators for the P2 peak only for the size of the change and not for the velocity of the change. The latencies of these peaks revealed a similar pattern in response to alterations in size and velocity of the frequency change. Increasing the size and/or velocity of the frequency change resulted in earlier N1 and P2 latencies of the ACC response, despite the longer durations of the change for larger frequency steps. The decreasing peak latencies and increasing peak amplitudes suggest enhanced neural synchrony within the auditory cortex in response to increments in velocity and size of frequency changes. 163
3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
The positive correlation between ACC amplitude and the size of the frequency change is in line with earlier studies (Arlinger et al., 1981; Dimitrijevic et al., 2008; Harris et al., 2008; Martin et al., 2000; McCandless et al., 1970; Pratt et al., 2009). Harris et al. (2008) reported that in young, normal hearing volunteers a frequency change of around 1% of the base frequency was required to evoke a discernable cortical response. In older adults, aged 65-80 years, this threshold was around 2% of the base frequency (Harris et al., 2008). Since Harris et al. (2008) found this difference in age groups, we only included young subjects under the age of 30 years. In line with their results, all subjects in our study displayed a clear response with the smallest frequency change of approximately 2%. In the study by Harris et al. (2008) a FM sweep duration of 10 ms was applied, regardless of base frequency or frequency change (Harris et al., 2008). Our results demonstrated that besides the size of the frequency change, also the velocity of the preceding FM sweep influenced the amplitude of the ACC response. If the sweep duration is constant, and the size of the frequency change is increased the velocity also increases (Fig. 3). As we found that a faster change gives a larger response, one should be aware to keep velocities constant if the aim is to assess the influence of frequency changes on the ACC. This is especially important for future research assessing the possibilities of the ACC as a clinical objective measure for patients with sensorineural hearing loss or cochlear implants. For such purposes, a fast FM sweep is therefore recommended, since this might yield the clearest response. Direction selectivity Considering the importance of auditory change detection for survival it is hypothesized that the auditory system might be highly tuned to auditory changes revealing an approaching object. Based on the Doppler effect, which describes an increase in frequency if an object is approaching, it is suggested that frequency increments might result in larger cortical responses. This theory is supported by the results of Maiste and Picton 1989 and Pratt et al. 2009 who found larger cortical N1 amplitudes in response to an increase in frequency (Maiste et al., 1989; Pratt et al., 2009). On the contrary, in our study there tends to be a downward preference for especially the fast (â&#x2030;¤ 28.6 octave/s) and small (0.029 octave) changes. For larger and slower changes there was no evident directional preference. The different results found in these studies might be a consequence of stimulation paradigms and data analyses. Pratt et al. (2009) recorded the effects of frequency increments or decrements on N1 amplitudes, using tone bursts with a frequency increase of 10% or 50% followed by a frequency decrease after 1 second, whereas we used separate recordings to evaluate the effect of frequency increase or decrease. Since we found in a previous pilot study that duration of the signal prior to the auditory change interacts with the amplitude of the ACC, we chose to use separate recordings in a randomized fashion to control for this possible interaction. Moreover, for our direction selectivity analyses we corrected for variance in
164
CAEPs and frequency changes
amplitudes, rather than using raw peak amplitudes. Inter-subject variance in ACC amplitudes could severely influence calculations of direction selectivity if individual peak amplitudes are averaged. By using the method proposed by Nelken and Versnel a correction is applied to adjust for inter-subject variability (Nelken et al., 2000). More in line with our results are the findings by Arlinger et al. who found no effect of direction of the FM sweep on N1-P2 amplitude (Arlinger et al., 1979; Arlinger et al., 1976). This equipoise in direction preference is also consistent with the psychophysical literature which indicates an overall up/down balance in human speech and in detection of spectrotemporal modulations (Chi et al., 1999; Dooley et al., 1988; Gordon et al., 2002; Luo et al., 2007; Schouten, 1985; Schouten, 1986). Interestingly, the downward preference for fast frequency changes agrees with a downward bias in psychophysical direction identification found by Luo et al. (2007). They showed that Mandarin Chinese speakers had a downward bias for fast frequencies (50 and 100 octave/s), but not if FM sweeps were slower (< 25 octave/s) (Luo et al., 2007). On the other hand, in the same test setup native English speakers demonstrated an upward bias (Gordon et al., 2002). Based on these opposing results it can be argued if an actual directional preference exists in the human auditory system.
Conclusions Acoustic change amplitudes and latencies are highly dependent on the size and velocity of sound frequency changes, suggesting enhanced neural synchrony in response to larger and faster auditory changes. In the present study there was no consistent effect of direction selectivity, but ACC amplitudes tended to be larger for the smaller and faster downward frequency changes. This is in line with the psychophysical literature which reports no consistent evidence for an up- or downward preference. The findings of the present study is important for future researchers and clinicians to be aware that ACCs are not only related to the size of the frequency change, but also to the velocity of the change.
Acknowledgements The authors thank all participants for their time, and Thijs Dijkgraaf for technical support. This study is supported by an unrestrictive research grant that Grolman received from Cochlear Ltd.
165
3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Chapter 3.2
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
References Amitay, S., Irwin, A., Hawkey, D.J., Cowan, J.A., Moore, D.R. 2006. A comparison of adaptive procedures for rapid and reliable threshold assessment and training in naive listeners. Journal of the Acoustical Society of America 119:1616-1625. Arlinger, S., Jerlvall, L. 1981. Early auditory electric responses to fast amplitude and frequency tone glides. Electroencephalography and Clinical Neurophysiology 51: 624-631. Arlinger, S.D., Jerlvall, L.B. 1979. Results of psychoacoustic and cortical evoked potential experiments using frequency and amplitude modulated stimuli. Scandinavian Audiology Supplementum:229-239. Arlinger, S.D., Jerlvall, L.B., Ahren, T., Holmgren, E.C. 1976. Slow evoked cortical responses to linear frequency ramps of a continuous pure tone. Acta Physiologica Scandinavica 98: 412-424. Chi, T., Gao, Y., Guyton, M.C., Ru, P., Shamma, S. 1999. Spectro-temporal modulation transfer functions and speech intelligibility. Journal of the Acoustical Society of America 106: 2719-2732. Dimitrijevic, A., Michalewski, H.J., Zeng, F.G., Pratt, H., Starr, A. 2008. Frequency changes in a continuous tone: auditory cortical potentials. Clinical Neurophysiology 119:2111-2124. Dooley, G.J., Moore, B.C. 1988. Duration discrimination of steady and gliding tones: a new method for estimating sensitivity to rate of change. Journal of the Acoustical Society of America 84: 1332-1337. Dreschler, W.A., Plomp, R. 1985. Relations between psychophysical data and speech perception for hearingimpaired subjects. II. Journal of the Acoustical Society of America 78:1261-1270. Friesen, L.M., Tremblay, K.L. 2006. Acoustic change complexes recorded in adult cochlear implant listeners. Ear and Hearing 27:678-685. Gordon, M., Poeppel, D. 2002. Inequality in identification of direction of frequency change (up vs. down) for rapid frequency modulated sweeps. Acoustics Research Letters Online 3:29-34. Harris, K.C., Mills, J.H., He, N.J., Dubno, J.R. 2008. Age-related differences in sensitivity to small changes in frequency assessed with cortical evoked potentials. Hearing Research 243:47-56. Horst, J.W. 1987. Frequency discrimination of complex signals, frequency selectivity, and speech perception in hearing-impaired subjects. Journal of the Acoustical Society of America 82:874-885. Luo, H., Boemio, A., Gordon, M., Poeppel, D. 2007. The perception of FM sweeps by Chinese and English listeners. Hearing Research 224:75-83. Maiste, A., Picton, T. 1989. Human auditory evoked potentials to frequency-modulated tones. Ear and Hearing 10:153-160. Martin, B.A., Boothroyd, A. 2000. Cortical, auditory, evoked potentials in response to changes of spectrum and amplitude. Journal of the Acoustical Society of America 107:2155-2161. McCandless, G.A., Rose, D.E. 1970. Evoked cortical responses to stimulus change. Journal of Speech and Hearing Research 13:624-634. Nelken, I., Versnel, H. 2000. Responses to linear and logarithmic frequency-modulated sweeps in ferret primary auditory cortex. The European Journal of Neuroscience 12:549-562. Noordhoek, I.M., Houtgast, T., Festen, J.M. 2001. Relations between intelligibility of narrow-band speech and auditory functions, both in the 1-kHz frequency region. Journal of the Acoustical Society of America 109:1197-1212. Ostroff, J.M., Martin, B.A., Boothroyd, A. 1998. Cortical evoked response to acoustic change within a syllable. Ear and Hearing 19:290-297. Papakonstantinou, A., Strelcyk, O., Dau, T. 2011. Relations between perceptual measures of temporal processing, auditory-evoked brainstem responses and speech intelligibility in noise. Hearing Research 280:30-37
166
CAEPs and frequency changes
Pratt, H., Starr, A., Michalewski, H.J., Dimitrijevic, A., Bleich, N., Mittelman, N. 2009. Auditory-evoked potentials to frequency increase and decrease of high- and low-frequency tones. Clinical Neurophysiology 120:360-373. Schouten, M.E. 1985. Identification and discrimination of sweep tones. Perception & psychophysics 37:369376. Schouten, M.E. 1986. Three-way identification of sweep tones. Perception & psychophysics 40:359-361. Sek, A., Moore, B.C. 1995. Frequency discrimination as a function of frequency, measured in several ways. Journal of the Acoustical Society of America 97:2479-2486. Strelcyk, O., Dau, T. 2009. Relations between frequency selectivity, temporal fine-structure processing, and speech reception in impaired hearing. Journal of the Acoustical Society of America 125:3328-3345. Tremblay, K.L., Friesen, L., Martin, B.A., Wright, R. 2003. Test-retest reliability of cortical evoked potentials using naturally produced speech sounds. Ear and Hearing 24:225-232.
3.2
167
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion & future research perspectives
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 170
General discussion
Introduction In 2013, three pioneers in the field of cochlear implantation, Graeme Clark, Ingeborg Hochmaier, and Blake Wilson, were honored with the Lasker-DeBakey Award for Clinical Medical Research. Awarding this prestigious prize for the development of the cochlear implant emphasizes the worldwide success story of this sophisticated device for the treatment of severe hearing loss. Every day, hundreds of thousands of children and adults rely on their implants in daily communication. Despite the fairly good results obtained in most CI users, specific patient groups fare less well with their implants. Increasing durations of severe to profound hearing loss prior to implantation are known to negatively affect postoperative performance (Lesinski-Schiedat et al., 2004; Teoh et al., 2004; Niparko et al., 2010; Boons et al., 2012; Lazard et al., 2012; Blamey et al., 2013). In the absence of auditory stimulation, multiple neurophysiologic changes are being observed along the entire auditory pathway, which could affect the final outcome after cochlear implantation. So far, most studies have been conducted in congenital deaf children and adults with postlingual deafness. Little attention has been paid to the specific population of prelingually deaf patients, who only received their implants in adulthood. Since this specific patient group is becoming scarce, as nowadays most congenitally deaf children are treated with cochlear implants within their first years of life, studying neurophysiologic changes induced by their long term auditory deprivation is highly valuable for our understanding of the far stretching consequences of deafness. The general aim of this thesis was to further elucidate the consequences of long term and early-onset deafness on auditory pathway maturation and functioning after cochlear implantation. The first part of this thesis reviewed the clinical results of early (simultaneous) bilateral cochlear implantation in children, whereas the second part has focused on the neurodevelopmental consequences of long term hearing loss in CI users with early-onset deafness. The purpose of this final chapter is to elaborate on the multimodal neurophysiologic alterations induced by long term auditory deprivation, urging the clinical need for early (bilateral) cochlear implantation and to discuss future research perspectives.
Early bilateral cochlear implantation in children The studies presented in the first part of this thesis were conducted to assess the benefit of bilateral cochlear implantation in children and to evaluate the clinical effects of screening for congenital hearing loss. Evaluations of the neonatal hearing screening programs, already revealed that these programs lead to early recognition of hearing loss in the newborn
171
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
(Korver et al., 2010). If this would eventually lead to earlier treatment was to be seen. The results of Chapter 1.1, confirmed that a proper national neonatal hearing screening indeed results in earlier treatment. Interestingly, this effect was only seen in the Dutch, but not in the German population. As discussed, this difference might be attributed to differences in the study populations, but also to differences in the hearing screening programs. Especially the follow-up and tracking of the child after a negative test result are essential to avoid potentially detrimental delays. An inadequate follow up could introduce delays, impeding the effectiveness of the hearing screening program. With the introduction of the screening program in the Netherlands it has been committed that children are directly referred to specialized audiology centers for additional diagnostics and the initiation of hearing aid trials. Unnecessary steps within the diagnostic process have been eliminated as much as possible in order to optimize this process. Following early detection of severe hearing loss, the question arises whether the child should receive unilateral or bilateral cochlear implants. Unilateral implantation is an established cost-effective treatment in both children and adults (Cheng and Niparko, 1999; Barton et al., 2003; UKCISG, 2004; Barton et al., 2006; Bond et al., 2009). Controversy remains on the cost-effectiveness of bilateral implantation (Lammers et al., 2011). In chapter two we present the results of a systematic review on the effectiveness of bilateral implantation in children. Remarkable was the fact that although the second implant comes at a very high cost, no studies to date have thoroughly evaluated the benefit of the second implant in a randomized study design. Although randomization in studies assessing surgical interventions has been a subject for debate, it is expected to be less an issue for cochlear implantation as surgical technique and learning curve have been found to have little effect on postoperative results (Postelmans et al., 2009). The results of our best evidence synthesis, based on nonrandomized studies, demonstrated that the second implant might be helpful in localizing sound sources and possibly aid in the speech and language development. Nevertheless, indisputable evidence derived from randomized studies was lacking. For this reason, in 2012 the randomized controlled trial â&#x20AC;&#x2DC;Bilateral Cochlear implant Effectiveness Study (BiCEPS)â&#x20AC;&#x2122; was initiated in the University Medical Center Utrecht (NTR3232). The aim of this trial was to determine the benefit of the second implant in children, and to assess whether simultaneous bilateral implantation is superior to sequential implantation with a 2-year interval. Before the inclusion of the third child, the study was discontinued, due to the report of the Dutch Health Care Insurance Board (College voor Zorgverzekeringen). This report stated that based on the preliminary results of the systematic review presented in Chapter 1.2, there was sufficient evidence for the reimbursement of bilateral cochlear implantation in children up to the age of 5 years (College voor Zorgverzekeringen, 2012). Following this decision, the federation of cochlear implant centers in the Netherlands (Cochleaire Implantatie Overleg Nederland, CION) established a report proposing the reimbursement of the second implant for all children
172
General discussion
with unilateral implants (Cochleaire Implantatie Overleg Nederland, 2012). Chapter 1.2 and 1.3 of this thesis formed the scientific foundations for this report. The results presented in Chapter 1.3 demonstrate that with increasing inter-implant interval the expected benefit of the second implant decreases. However, a clear cut-off age at which the second implant is deemed to be likely unsuccessful, could not be established. Therefore, studies exploring the additional value of the second implant in older children with a longer inter-implant interval remain interesting. With the conditional reimbursement of the second implant for all children, CI-ON initiated a national multicenter trial to further investigate the effect of the inter-implant interval on speech perception, and speech- and language development. In January 2015 this prospective cohort study was officially started in all Dutch cochlear implant centers. Chapter 1.3 focused on the effect of the inter-implant interval on clinical outcome measures, such as speech perception, sound localization and speech- and language development. The neurophysiological mechanisms underlying the negative correlation with the duration of inter-implant interval, was beyond the scope of this review. It is however, very likely that alterations in the development of the auditory pathway, due to prolonged periods of auditory deprivation in the second ear and the longstanding situation of a dominant ear, are responsible. As described in more detail in the Introduction, the development of the auditory pathway stretches during childhood. During the fetal and neonatal periods the major structures and neuronal tracks of the auditory pathway are being developed (Matschke et al., 1994; Moore and Guan, 2001; Pujol et al., 2006; Moore and Linthicum, 2007; Doria et al., 2010; van den Heuvel et al., 2015). In the absence of sufficient auditory stimulation during this important period, an impaired maturation of the auditory pathway is inevitable. Neurophysiologic studies investigating the development of the auditory brainstem in children, revealed that after bilateral implantation with no or only short intervals (< 1 year) there are no differences in the eABR wave latencies between the first and second implanted ear (Gordon et al., 2008b; Gordon et al., 2011). On the other hand, if the inter-implant interval increased (> 2 years), the wave latencies of the later implanted ear were prolonged and did not normalize over time (Gordon et al., 2008b). These delayed wave latencies found in the second implanted ear, probably represent delays in neural conduction in especially the rostral part of the auditory brainstem. Longer durations of auditory deprivation in the later implanted ear in combination with the long lasting situation of single sided deafness could have resulted in structural changes in the neuronal development and myelination within the auditory brainstem. The neural connections are stabilized during childhood and unalterable after several years of unilateral deafness causing the persistent eABR wave latency differences. Equivalent alterations can be seen in the cortical development in children with uni- or bilateral CIs. In children with unilateral CIs there are signs that they experience
173
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
significant lateralization effects, which could negatively influence their abilities for especially sound localization and speech perception in noise (Gordon et al., 2013a). Unilateral cochlear implant use prior to bilateral implantation was found to result in abnormal strengthening and dominance of the contralateral auditory cortex, which was already unalterable if they received their second implant after a delay of 1.5 years or more (Gordon et al., 2013a). By contrast, cortical activity in children with no or very short inter-implant intervals demonstrated normal lateralized activity in both hemispheres (Gordon et al., 2013a). These neurophysiologic data demonstrate that long-term single sided deafness (even after unilateral implantation) can result in altered maturation of the entire auditory pathway and thereby potentially negatively influence postoperative performance, urging the need for early bilateral implantation with either no or very short intervals. In conclusion, we have found that nowadays there is increasing clinical and neurophysiologic evidence suggesting the benefit of the second cochlear implant in children. If we follow the definition of one of the founding fathers of Evidence Based Medicine, Professor D.L. Sackett, who described Evidence Based Medicine as the integration of clinical expertise with the best available external evidence, alongside with patient preference (Sackett et al., 1996), we could state that the clinical opinion described in a consensus statement of leaders in the field (Ramsden et al., 2012), suggesting a benefit of the second implant in children, is supported by our best evidence synthesis (Chapter 1.2) and by accumulating neurophysiologic evidence (Gordon et al., 2008b; Gordon et al., 2013b; Gordon et al., 2013a). The findings that children experience most benefit after simultaneous bilateral cochlear implantation, highlights the importance of the duration of deafness. Experimental studies are therefore highly valuable for our understanding of the neurophysiologic consequences of prolonged hearing loss which is the key to success or failure after cochlear implantation.
Impaired maturation of the auditory pathway following early-onset deafness The evaluation of cochlear implant performance after long durations of early-onset deafness, presented in Chapter 2.1, revealed that although a majority of such prelingually deaf patients perceive some benefit of their CIs, average speech perception scores were much lower than observed in postlingually deaf CI users. Eventually even 21% of the prelingually deaf late implanted patients decided not to use their implants. These outcomes are not the result of surgical complications or malfunctioning of their devices, but are the consequence of stimulation of an impaired developed auditory pathway resulting in minimal benefit for the patient. As described in the Introduction, auditory stimulation during childhood is essential for
174
General discussion
maturation of the auditory pathway. The development of the cerebral cortex initially follows a bottom-up pattern which is (partly) autonomous and independent of experience (Chapter 2.2. and 2.3; Kral et al., 2013; Tillein et al., 2012). After the development of the coarse structures of the pathway, sufficient stimulation is vital for further maturation and refinement of the pathway since this is more dependent on auditory driven top-down processing (Kral, 2013). The results presented in Chapter 2.2 and 2.3 underline this hypothesis. Despite their earlyonset bilateral hearing impairment the presence of the normal like ABR and CAEP waveforms in the studied prelingual deaf subjects, indicate the development of the coarse structures of their auditory pathway, whereas the latency differences are probably caused by sensory driven maturational deficits. The delayed ABR wave V latencies and increased III-V intervals presented in Chapter 2.2 might represent an impaired neural conduction within the auditory brainstem induced by the long durations of early-onset deafness. Since wave V is thought to be generated in the inferior colliculus and/or lateral lemniscus, the found delay likely reflects prolonged neural conduction along the more rostral part of the brainstem (Eggermont and Salamy, 1988; Ponton et al., 1996). After birth, in normal-hearing children, ABR wave III and V latencies decrease following an exponential function (Eggermont and Salamy, 1988; Ponton et al., 1996). These decreasing ABR wave latencies nicely coincide with myelination patterns found in autopsied infants (Inagaki et al., 1987; Eggermont and Salamy, 1988). This relation suggests that the decreasing ABR wave latencies reflect faster neural conduction due to increasing myelination within the maturing brainstem (Inagaki et al., 1987; Eggermont and Salamy, 1988). Given this correlation between ABR wave latencies and the maturational process of the brainstem, these responses are proposed as a derivate of brainstem maturation (Inagaki et al., 1987; Eggermont and Salamy, 1988). The same holds for the cortical auditory evoked potential, which is however more subject to top-down influences by higher order processing. Earlier studies on the CAEP revealed that this waveform changes during childhood, suggesting its possible role as a biomarker of auditory cortex maturation (Ponton et al., 2000; Ponton and Eggermont, 2001; Sharma et al., 2002; Wunderlich and Cone-Wesson, 2006; Sharma et al., 2007). One major change in this waveform is the formation of the negative N1 peak, which occurs around the age of 8-9 years (Ponton et al., 2000; Ponton and Eggermont, 2001). This N1 peak progressively becomes larger and is in adulthood the most prominent and stable peak of the CAEP waveform (Ponton et al., 2000; Ponton and Eggermont, 2001). The development of the N1 has been thought to coincide with the maturation of the supragranular layers (layer II and III) of the auditory cortex (Ponton et al., 2000; Ponton and Eggermont, 2001; Kral and Eggermont, 2007). In late-implanted children (>6 years) with early-onset deafness a cortical response with a prominent P1 peak and an absent N1 was found (Sharma et al., 2002; Sharma et al., 2007). These findings suggest that the impaired maturation of the supragranular layers in these children prevents the development of the N1. Surprisingly, our results show that
175
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
even after longer periods of early-onset deafness the typical adult-like P1-N1-P2 complex did develop. Comparisons of the N1 peaks, however, revealed that N1 latency was shorter and the peak was larger in the prelingually deaf. This might suggest that the essential cortical structures, necessary to generate this N1 are being developed, but further maturation of the supragranular layers may be hampered. In line with our results are those found by Gordon et al. 2008b. In their study, CAEP responses of 3 prelingually deaf children (age at test >15 years) with relative long durations of deafness (age at implantation 9-15 years), clearly demonstrated a typical P1-N1-P2 waveform morphology (Gordon et al., 2008a), which is in line with our findings that the maturation of this waveform is more likely to be age dependent. Our results presented in chapter 2.3 demonstrated altered activity in the cortex of the prelingually deaf in response to the onset of a stimulus. Long term early-onset hearing loss is even more likely to influence higher order processing, rather than registration of stimulus onset. By using oddball paradigms or introducing temporal or spatial differences within stimuli, higher order neuronal processes can be activated. Acoustic change complex recordings, as introduced in Chapter 3.2, might even demonstrate larger differences between pre- and postlingual deaf CI users. In a follow-up study we will investigate if frequency changes within a continuous stimulus result in an even more pronounced altered cortical activation in the prelingually deaf.
The multifactorial consequences of auditory deprivation Even before birth the coarse functional auditory brain networks appear to be developed, albeit in an immature state (Inagaki et al., 1987; Moore et al., 1995; Moore and Guan, 2001; Moore and Linthicum, 2007; Doria et al., 2010; Tillein et al., 2012; Kral, 2013; van den Heuvel et al., 2015; Wu et al., 2014). Sensory stimulation induces the maturation of this network (Inagaki et al., 1987; Moore et al., 1995; Moore and Guan, 2001; Innocenti and Price, 2005; Pujol et al., 2006; Catani et al., 2013; Miao et al., 2013; van den Heuvel et al., 2015), but in the absence of stimulation this process can be hampered. The results presented in the second part of this thesis, combined with the results of other neurophysiologic studies in both humans and animals show the complexity of the various processes leading to an impaired maturation of the auditory pathway in the absence of sufficient auditory stimulation. These multifactorial consequences of auditory deprivation on the auditory pathway maturation are schematically summarized in Figure 1 and described in more detail in the following section. This diagram not only provides us an overview of the various neurodevelopmental aspects, but can also give us an explanation for the sometimes disappointing results seen after cochlear implantation.
176
General discussion
Fig. 1. Schematic model of the multifactorial consequences of auditory deprivation on the auditory pathway.
Neural degeneration Although the coarse structures and pathways of the auditory system can develop in the absence of sensory input, long-term hearing loss will eventually result in neural degeneration. Animal models demonstrated that severe cochlear hair cell loss leads to degeneration of spiral ganglion cells (Spoendlin, 1975; Versnel et al., 2007). Following this degenerative process of the more peripheral spiral ganglion cells, the consequences in the auditory brainstem become apparent. Experiments in congenital deaf cats showed that hearing loss results in a reduction of the volume of the cochlear nucleus (Leake et al., 2008). The clinical importance of neural degeneration has already been demonstrated in cochlear implant users. In a recent temporal bone study of autopsied cochlear implant users, a trend towards better speech perception in patients with higher spiral ganglion cell counts was observed (Seyyedi et al., 2014). This emphasizes the clinical importance of neural degeneration of the auditory pathway. Impaired myelination Neuropathological and imaging studies revealed that myelination of the auditory pathway already starts around the 26th to 29th week of gestation (Inagaki et al., 1987; Matschke et
177
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
al., 1994; Moore et al., 1995; Moore and Linthicum, 2007), and continues during childhood (Ponton and Eggermont, 2001; Pujol et al., 2006; Moore and Linthicum, 2007; Wu et al., 2014). The process of myelination is likely to follow a peripheral to central pattern, starting in the auditory nerve and gradually evolving to the cortical areas (Moore et al., 1995; Pujol et al., 2006). Although myelination of the auditory cortex and its associated areas takes years to be completed (Ponton et al., 2000; Ponton and Eggermont, 2001; Pujol et al., 2006; Moore and Linthicum, 2007), myelination of the auditory brainstem is thought to be finished around the age of 2 to 3 years (Inagaki et al., 1987; Moore et al., 1995). This maturational process is reflected by the decrease in ABR wave latencies (Inagaki et al., 1987; Eggermont and Salamy, 1988; Ponton et al., 1992). In congenital deaf pediatric CI users, ABR wave latencies were initially prolonged, but decreased in the following years (Gordon et al., 2007; Thai-Van et al., 2007; Gordon et al., 2010). The results of Chapter 2.2, on the other hand indicate that after long-term deafness, ABR wave latencies remain prolonged indicating structural changes possibly reflecting disrupted myelination. Diffusion tensor imaging (DTI) studies in congenitally deaf children and adults also revealed that early-onset deafness results in a reduced fractional anisotropy in the superior temporal and Heschlâ&#x20AC;&#x2122;s gyrus (Miao et al., 2013; Wu et al., 2014), possibly reflecting an impaired development of myelinated tracts within the cortex. Degraded axonal connectivity and impaired synaptic pruning During normal development the density of dendrites and synapses within the cortex increases during the first four years, to gradually decrease afterwards (Conel, 1939; Huttenlocher and Dabholkar, 1997; Kral et al., 2005). Auditory stimulation during this period of cortical maturation, leads to a normal exuberant development and axonal selection (Innocenti and Price, 2005; Kral et al., 2005). Important synaptic connections are being strengthened, whereas unused synapses are eliminated. This process of cortical specialization can become impaired in the absence of sufficient stimulation. Animal models demonstrated that congenital deafness leads to a delayed increase in synaptic activity, followed by an exaggerated overshoot in synaptic activity (Kral et al., 2005). Following this delayed â&#x20AC;&#x2DC;synaptic overshootâ&#x20AC;&#x2122; increased elimination of synaptic connections occurs, resulting in a lower synaptic activity during adulthood compared to normal hearing controls (Kral et al., 2005). This disrupted selection of axonal populations within the auditory cortex might lead to an activation of a larger number of unspecific neurons, rather than a few specialized neurons (Innocenti and Price, 2005). A possible consequence of this impaired process of synaptic pruning has been demonstrated by Kral et al. 2006 who revealed that activation of the naive auditory cortex of congenitally deaf cats, leads to a smeared representation of auditory stimuli with loss of the tonotopical organization (Kral et al., 2006). As synaptic pruning is thought to be completed around the age of 12 years (Huttenlocher and Dabholkar, 1997), prolonged auditory deprivation, as is
178
General discussion
the case in prelingually deaf adults, might even stabilize these aberrant connections in the auditory cortex. These effects of an impaired process of synaptic pruning might also (at least partially) explain the large cortical N1 peak found in the prelingually deaf CI users, presented in Chapter 2.3. Cross-modal reorganization and cortico-cortical decoupling Besides a negative effect on the representation and processing of auditory stimuli, the impeded process of synaptogenesis, makes these cortical regions more prone to alterations in connectivity with other cortical regions. During normal development in childhood and adolescence the brain becomes more and more capable to integrate information from different regions (Pujol et al., 2006; Dosenbach et al., 2007; Fair et al., 2008). Long term auditory deprivation might lead to a reduction and impairment of axonal connections between higher order cortical areas and the supragranular layers of the primary auditory cortex. This process of cortico-cortical decoupling results in a reduction in the number of connections with other brain regions and reduced top-down inhibition (Kral and Eggermont, 2007; Kral and Sharma, 2012; Kral, 2013). The increased amplitude of the cortical response found in the prelingually deaf CI users described in Chapter 2.3, might reflect this decreased top-down inhibition by higher order neurons. Decoupling of the auditory cortex in response to long periods of auditory deprivation has also been observed in imaging studies. Positron emission tomography studies revealed that in postlingually deaf CI users speech stimuli activate both the primary and secondary auditory cortex, whereas in prelingually deaf CI users only the primary auditory cortex was activated, suggesting the decoupling of the primary auditory cortex from higher order centers (Naito et al., 1997; Nishimura et al., 1999; Hirano et al., 2000). Besides reduced top-down inhibition, cortico-cortical decoupling might also lead to deficits in multisensory processing (Schorr et al., 2005; Gilley et al., 2010; Nava et al., 2014). Two study groups assessed the auditory-visual integration abilities of young CI users, using a McGurk task (Schorr et al., 2005; Gilley et al., 2010). The McGurk effect describes the effect that visual cues can modify audition when an incongruent visual signal is presented, e.g. presenting the auditory syllable /ba/ with synchronous lips pronouncing the syllable /ga/, is often perceived as a third syllable:/da/ (McGurk and MacDonald, 1976). Both studies, found that audio-visual integration, was only possible in the early implanted children, whereas later implanted children revealed more visual dominance and decreased audio-visual integration (Schorr et al., 2005; Gilley et al., 2010). Furthermore, altered cortico-cortical connections can also result in the recruitment of the deprived areas by other sensory systems of the brain. This cross-modal reorganization has been extensively studied in the blind (Renier et al., 2014), but an increasing number of studies reveal that long durations of auditory deprivation can result in a cortical reorganization of the
179
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
auditory cortex. Visual stimuli for example can result in an activation of the auditory cortex (Finney et al., 2001; Sadato et al., 2004; Lomber et al., 2010; Kral and Sharma, 2012; Vachon et al., 2013; Dewey and Hartley, 2015; Strelnikov et al., 2015). Recently, an electrophysiological study on the visual evoked potential demonstrated that even in postlingual deaf CI users there is evidence of cross modal reorganization of the auditory cortex (Sandmann et al., 2012). It is, however, expected that the extent of cross modal plasticity is larger during childhood when the cortical layers are still immature, than during adulthood when only little residual plasticity is present. These various neurophysiologic consequences of long-term auditory deprivation within sensitive periods which are essential for proper maturation, will eventually result in an underdeveloped and immature auditory cortex, which is subject to modulations by more local and distant cortical regions. Activation of this innate and less complex auditory cortex results in altered cortical responses, as found in our study presented in Chapter 2.3. Moreover, it provides an understanding for the relative poor postoperative functioning of prelingually deaf CI users with long-term deafness as presented in Chapter 2.1. In this chapter, several prognostic factors are proposed which eventually could influence the postoperative success of implantation. These factors come down to one major criterion which is in line with our neurophysiologic model: the amount of residual hearing. If a child has no or only very little residual hearing he/she is more prone to be subject to the multimodal consequences of hearing loss, schematically presented in figure 1. The extent at which these effects take place, remain a concern in the preoperative clinical evaluation of prelingually deaf patients with long durations of hearing impairment. Especially the extent of cross-modal plasticity or residual plasticity of the impaired auditory cortex is unknown. To further gain insight into the presence of cross-modal reorganization in pre- and postlingually deaf CI candidates we are currently performing 7 Tesla functional MRI scans to assess the extent of visual activation of the auditory cortex by visual stimuli. We expect to find more activation of the auditory cortex in response to visual stimuli in prelingually deaf candidates compared to postlingually deaf. Correlating the extent of preoperative cross-modal reorganization with postoperative performance can eventually give us an idea of the clinical consequences of cross-modal reorganization.
180
General discussion
The clinical use of objective measures Although various objective electrophysiological measures are proposed in cochlear implant users, only the electrically evoked compound action potential (eCAP) has found its way to the clinic. This measure has the advantage that it can be easily recorded intra- and postoperatively without the need of additional equipment. The intraoperative success rate of recording is very high, but decreases in the postoperative setting as demonstrated in Chapter 3.1 and in earlier studies conducted in our clinic (Smoorenburg et al., 2002; van Wermeskerken et al., 2006). In Chapter 3.1 we further explore the postoperative success rates of recording alternative objective measures as the eABR and eCAEP. Although these findings have been obtained in only a fairly small study population, they suggest that postoperative eABR and eCAEP recordings can be used as an alternative measure if eCAPs cannot be evoked. Besides the feasibility to reliably record these measures in (almost) all subjects, a second prerequisite for a valuable clinical tool, is a good correlation with actual performance. As shown in Chapter 3.1, no evident relation between the presence of the various measures and postoperative speech perception could be established. This is confirmed by the lack of any consistent correlations between the various measures and performance in the current literature (Brown et al., 1995; Groenen et al., 1996b; Firszt et al., 2002; Turner et al., 2002; Battmer et al., 2005; Martin et al., 2008; Miller et al., 2008; Kim et al., 2010) or in our findings presented in Chapter 2.2 and 2.3. Although these studies were not designed to identify a possible correlation between the various measures and speech performance, little trust can be put in future studies to identify a clinical useful association between the presence of these onset responses and performance. By changing stimulation paradigms or sound stimuli, more complex and higher order processes can be activated. By using oddball paradigms the P300 and Mismatch Negativity (MMN) can be recorded. Although they seem to be better correlated to performance than the CAEP onset response (Groenen et al., 1996a; Groenen et al., 2001; Beynon, 2005; Martin et al., 2008), subjectsâ&#x20AC;&#x2122; attention and cooperation is required. By introducing a change in a continuous tone, a cortical onset response on this change can be recorded (Maiste and Picton, 1989; Ostroff et al., 1998; Harris et al., 2008; Martin et al., 2008). The advantage of this so called Acoustic Change Complex (ACC) is that it can be recorded without the subjectsâ&#x20AC;&#x2122; attention, and it reflects the ability to detect changes within stimuli, such as changes in frequency. Since psychophysical frequency discrimination tests are found to be related to speech perception in noise in patients with sensorineural hearing loss (Papakonstantinou et al., 2011), it would be of interest to develop an objective derivative of such a test. Before reliable correlations can be made between an objective ACC frequency discrimination test and psychophysical frequency discrimination results, additional studies are needed to identify which alterations in frequency contribute to the ACC waveform. In Chapter 3.2, we present the results of a pilot study which identified that both the size of
181
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
the frequency step and the velocity of the frequency modulation change contributed to the amplitude of the ACC waveform. For future research, studying the ACC in patients with sensorineural hearing loss and cochlear implants, it is important to take this in consideration by altering only one factor, while keeping the other constant. Currently, we are performing a study in normal hearing and patients with sensorineural hearing loss to identify the relation between psychophysical frequency discrimination and the ACC frequency measure. If this proves to be a good predictor of frequency discrimination and/or speech perception in noise, this measure could find its way to the clinical setting and can be studied in other patient groups, such as cochlear implant users.
Conclusion The results from the studies presented in this thesis suggest that cochlear implantation after long durations of deafness results in activation of the more innate and prerequisite components of a less complex and immature neuronal network, which is highly susceptible for cross-modal reorganization. The extent to which these multimodal factors influence the individual patientâ&#x20AC;&#x2122;s auditory pathway seems to be related to the severity and duration of hearing loss. Combining these neurophysiological findings with the accumulating clinical evidence, highlights the importance of early, and bilateral cochlear implantation with no or only short intervals between the surgeries in children with early-onset deafness. Introducing delays will hamper development of the entire auditory pathway and subsequently result in a reduced performance after cochlear implantation.
182
General discussion
References Barton GR, Bloor KE, Marshall DH, Summerfield AQ (2003). Health-service costs of pediatric cochlear implantation: multi-center analysis. International Journal of Pediatric Otorhinolaryngology 67:141-149. Barton GR, Fortnum HM, Stacey PC, Summerfield AQ (2006). Hearing-impaired children in the United Kingdom, III: Cochlear implantation and the economic costs incurred by families. Ear and Hearing 27:563-574. Battmer RD, Lai WK, Dillier N, Pesch N, Killian MJ, Lenarz T (2005). Correlation of NRTâ&#x201E;˘ recovery function parameters and speech perception results for different stimulation rates. In: Fourth International Symposium and Workshops: Objective Measures in Cochlear Implants. Hannover, Germany. Beynon AJ (2005). Electrically-evoked auditory cortical event-related potentials in cochlear implants. The P300 potential. In, p 178: Radboud University Nijmegen. Blamey P et al. (2013). Factors affecting auditory performance of postlinguistically deaf adults using cochlear implants: an update with 2251 patients. Audiology and Neurotology 18:36-47. Bond M, Mealing S, Anderson R, Elston J, Weiner G, Taylor R, Hoyle M, Liu Z, Price A, Stein K (2009). The effectiveness and cost-effectiveness of cochlear implants for severe to profound deafness in children and adults: a systematic review and economic model. Health Technology Assessment 13:1. Boons T, Brokx JP, Dhooge I, Frijns JH, Peeraer L, Vermeulen A, Wouters J, van Wieringen A (2012). Predictors of spoken language development following pediatric cochlear implantation. Ear and Hearing 33:617639. Brown CJ, Abbas PJ, Bertschy M, Tyler RS, Lowder M, Takahashi G, Purdy S, Gantz BJ (1995). Longitudinal assessment of physiological and psychophysical measures in cochlear implant users. Ear and Hearing 16:439-449. Catani M, Thiebaut de Schotten M, Slater D, Dellâ&#x20AC;&#x2122;Acqua F (2013). Connectomic approaches before the connectome. NeuroImage 80:2-13. Cheng AK, Niparko JK (1999). Cost-utility of the cochlear implant in adults: a meta-analysis. Archives of Otolaryngology - Head and Neck Surgery 125:1214-1218. Cochleaire Implantatie Overleg Nederland (2012). Richtlijn indicatie bilaterale cochleaire implantatie voor kinderen van 5 tot en met 18 jaar. https://www.zorginstituutnederland.nl/binaries/content/ documents/zinl-www/documenten/publicaties/rapporten-en-standpunten/2014/1401-tweedecochleair-implantaat-voor-kinderen-tussen-5-en-18-jaar/Tweede+cochleair+implantaat+ voor+kinderen+tussen+5+en+18+jaar.pdf. Accessed July 14 2015 College voor Zorgverzekeringen (2012). Achtergrondrapportage beoordeling stand van de wetenschap en praktijk Bilaterale cochleaire implantaten bij kinderen. https://www.zorginstituutnederland.nl/binaries/ content/documents/zinl-www/documenten/publicaties/rapporten-en-standpunten/2012/1207herbeoordeling-standpunt-bilaterale-cochleaire-implantaten-bij-kinderen/1207-herbeoordelingstandpunt-bilaterale-cochleaire-implantaten-bij-kinderen/Herbeoordeling+standpunt+bilaterale+cochl eaire+implantaten+bij+kinderen.pdf. Accessed 14 July 2015 Conel JL (1939). The postnatal development of human cerebral cortex. Cambridge, MA Harvard University Press. Dewey RS, Hartley DE (2015). Cortical cross-modal plasticity following deafness measured using functional near-infrared spectroscopy. Hearing Research 325:55-63. Doria V, Beckmann CF, Arichi T, Merchant N, Groppo M, Turkheimer FE, Counsell SJ, Murgasova M, Aljabar P, Nunes RG, Larkman DJ, Rees G, Edwards AD (2010). Emergence of resting state networks in the preterm human brain. Proceedings of the National Academy of Sciences of the United States of America 107:20015-20020. Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, Dosenbach RA, Fox MD, Snyder AZ, Vincent JL, Raichle ME, Schlaggar BL, Petersen SE (2007). Distinct brain networks for adaptive and stable task control in humans. Proceedings of the National Academy of Sciences of the United States of America 104:11073-11078.
183
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Eggermont JJ, Salamy A (1988). Maturational time course for the ABR in preterm and full term infants. Hearing Research 33:35-47. Fair DA, Cohen AL, Dosenbach NU, Church JA, Miezin FM, Barch DM, Raichle ME, Petersen SE, Schlaggar BL (2008). The maturing architecture of the brainâ&#x20AC;&#x2122;s default network. Proceedings of the National Academy of Sciences of the United States of America 105:4028-4032. Finney EM, Fine I, Dobkins KR (2001). Visual stimuli activate auditory cortex in the deaf. Nature Neuroscience 4:1171-1173. Firszt JB, Chambers, Rd, Kraus N (2002). Neurophysiology of cochlear implant users II: comparison among speech perception, dynamic range, and physiological measures. Ear and Hearing 23:516-531. Gilley PM, Sharma A, Mitchell TV, Dorman MF (2010). The influence of a sensitive period for auditory-visual integration in children with cochlear implants. Restorative Neurology and Neuroscience 28:207-218. Gordon KA, Papsin BC, Harrison RV (2007). Auditory brainstem activity and development evoked by apical versus basal cochlear implant electrode stimulation in children. Clinical Neurophysiology 118:16711684. Gordon KA, Jiwani S, Papsin BC (2011). What is the optimal timing for bilateral cochlear implantation in children? Cochlear Implants International 12 Suppl 2:S8-14. Gordon KA, Wong DD, Papsin BC (2013a). Bilateral input protects the cortex from unilaterally-driven reorganization in children who are deaf. Brain 136:1609-1625. Gordon KA, Jiwani S, Papsin BC (2013b). Benefits and detriments of unilateral cochlear implant use on bilateral auditory development in children who are deaf. Frontiers in Psychology 4:719. Gordon KA, Tanaka S, Wong DDE, Papsin BC (2008a). Characterizing responses from auditory cortex in young people with several years of cochlear implant experience. Clinical Neurophysiology 119:2347-2362. Gordon KA, Valero J, van Hoesel R, Papsin BC (2008b). Abnormal timing delays in auditory brainstem responses evoked by bilateral cochlear implant use in children. Otology & Neurotology 29:193-198. Gordon KA, Valero J, Jewell SF, Ahn J, Papsin BC (2010). Auditory development in the absence of hearing in infancy. Neuroreport 21:163-167. Groenen P, Snik A, van den Broek P (1996a). On the clinical relevance of mismatch negativity: results from subjects with normal hearing and cochlear implant users. Audiology and Neurotology 1:112-124. Groenen PA, Beynon AJ, Snik AF, van den Broek P (2001). Speech-evoked cortical potentials and speech recognition in cochlear implant users. Scandinavian audiology 30:31-40. Groenen PA, Makhdoum M, van den Brink JL, Stollman MH, Snik AF, van den Broek P (1996b). The relation between electric auditory brain stem and cognitive responses and speech perception in cochlear implant users. Acta Otolaryngologica 116:785-790. Harris KC, Mills JH, He NJ, Dubno JR (2008). Age-related differences in sensitivity to small changes in frequency assessed with cortical evoked potentials. Hearing Research 243:47-56. Hirano S, Naito Y, Kojima H, Honjo I, Inoue M, Shoji K, Tateya I, Fujiki N, Nishizawa S, Konishi J (2000). Functional differentiation of the auditory association area in prelingually deaf subjects. Auris, Nasus, Larynx 27:303-310. Huttenlocher PR, Dabholkar AS (1997). Regional differences in synaptogenesis in human cerebral cortex. The Journal of Comparative Neurology 387:167-178. Inagaki M, Tomita Y, Takashima S, Ohtani K, Andoh G, Takeshita K (1987). Functional and morphometrical maturation of the brainstem auditory pathway. Brain & Development 9:597-601. Innocenti GM, Price DJ (2005). Exuberance in the development of cortical networks. Nature Reviews Neuroscience 6:955-965. Kim JR, Abbas PJ, Brown CJ, Etler CP, Oâ&#x20AC;&#x2122;Brien S, Kim LS (2010). The relationship between electrically evoked compound action potential and speech perception: a study in cochlear implant users with short electrode array. Otology & Neurotology 31:1041-1048.
184
General discussion
Korver AM, Konings S, Dekker FW, Beers M, Wever CC, Frijns JH, Oudesluys-Murphy AM (2010). Newborn hearing screening vs later hearing screening and developmental outcomes in children with permanent childhood hearing impairment. Journal of the American Medical Association 304:1701-1708. Kral A (2013). Auditory critical periods: a review from system’s perspective. Neuroscience 247:117-133. Kral A, Eggermont JJ (2007). What’s to lose and what’s to learn: development under auditory deprivation, cochlear implants and limits of cortical plasticity. Brain Research Reviews 56:259-269. Kral A, Sharma A (2012). Developmental neuroplasticity after cochlear implantation. Trends in Neuroscience 35:111-122. Kral A, Tillein J, Heid S, Hartmann R, Klinke R (2005). Postnatal cortical development in congenital auditory deprivation. Cerebral Cortex 15:552-562. Kral A, Tillein J, Heid S, Klinke R, Hartmann R (2006). Cochlear implants: cortical plasticity in congenital deprivation. Progress in Brain Research 157:283-313. Lammers MJ, Grolman W, Smulders YE, Rovers MM (2011). The cost-utility of bilateral cochlear implantation: a systematic review. Laryngoscope 121:2604-2609. Lazard DS et al. (2012). Pre-, per- and postoperative factors affecting performance of postlinguistically deaf adults using cochlear implants: a new conceptual model over time. PloS ONE 7:e48739. Leake PA, Hradek GT, Bonham BH, Snyder RL (2008). Topography of auditory nerve projections to the cochlear nucleus in cats after neonatal deafness and electrical stimulation by a cochlear implant. Journal of the Association for Research in Otolaryngology 9:349-372. Lesinski-Schiedat A, Illg A, Heermann R, Bertram B, Lenarz T (2004). Paediatric cochlear implantation in the first and in the second year of life: a comparative study. Cochlear Implants International 5:146-159. Lomber SG, Meredith MA, Kral A (2010). Cross-modal plasticity in specific auditory cortices underlies visual compensations in the deaf. Nature Neuroscience 13:1421-1427. Maiste A, Picton T (1989). Human auditory evoked potentials to frequency-modulated tones. Ear and Hearing 10:153-160. Martin BA, Tremblay KL, Korczak P (2008). Speech evoked potentials: from the laboratory to the clinic. Ear and Hearing 29:285-313. Matschke RG, Stenzel C, Plath P, Zilles K (1994). Maturational aspects of the human auditory pathway: anatomical and electrophysiological findings. Journal for Oto-Rhino-Laryngology, Head and Neck Surgery 56:68-72. McGurk H, MacDonald J (1976). Hearing lips and seeing voices. Nature 264:746-748. Miao W, Li J, Tang M, Xian J, Li W, Liu Z, Liu S, Sabel BA, Wang Z, He H (2013). Altered white matter integrity in adolescents with prelingual deafness: a high-resolution tract-based spatial statistics imaging study. American Journal of Neuroradiology 34:1264-1270. Miller CA, Brown CJ, Abbas PJ, Chi SL (2008). The clinical application of potentials evoked from the peripheral auditory system. Hearing Research 242:184-197. Moore JK, Guan YL (2001). Cytoarchitectural and axonal maturation in human auditory cortex. Journal of the Association for Research in Otolaryngology 2:297-311. Moore JK, Linthicum FH, Jr. (2007). The human auditory system: a timeline of development. International Journal of Audiology 46:460-478. Moore JK, Perazzo LM, Braun A (1995). Time course of axonal myelination in the human brainstem auditory pathway. Hearing Research 87:21-31. Naito Y, Hirano S, Honjo I, Okazawa H, Ishizu K, Takahashi H, Fujiki N, Shiomi Y, Yonekura Y, Konishi J (1997). Sound-induced activation of auditory cortices in cochlear implant users with post- and prelingual deafness demonstrated by positron emission tomography. Acta Otolaryngologica 117:490-496. Nava E, Bottari D, Villwock A, Fengler I, Buchner A, Lenarz T, Roder B (2014). Audio-tactile integration in congenitally and late deaf cochlear implant users. PloS ONE 9:e99606.
185
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
General discussion
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Niparko JK, Tobey EA, Thal DJ, Eisenberg LS, Wang N-Y, Quittner AL, Fink NE, for the CIT (2010). Spoken Language Development in Children Following Cochlear Implantation. Journal of the American Medical Association 303:1498-1506. Nishimura H, Hashikawa K, Doi K, Iwaki T, Watanabe Y, Kusuoka H, Nishimura T, Kubo T (1999). Sign language ‘heard’ in the auditory cortex. Nature 397:116. Ostroff JM, Martin BA, Boothroyd A (1998) Cortical evoked response to acoustic change within a syllable. Ear and Hearing 19:290-297. Papakonstantinou A, Strelcyk O, Dau T (2011). Relations between perceptual measures of temporal processing, auditory-evoked brainstem responses and speech intelligibility in noise. Hearing Research 280:30-37. Ponton CW, Eggermont JJ (2001). Of kittens and kids: altered cortical maturation following profound deafness and cochlear implant use. Audiology and Neurotology 6:363-380. Ponton CW, Moore JK, Eggermont JJ (1996). Auditory brain stem response generation by parallel pathways: differential maturation of axonal conduction time and synaptic transmission. Ear and Hearing 17:402410. Ponton CW, Eggermont JJ, Coupland SG, Winkelaar R (1992). Frequency-specific maturation of the eighth nerve and brain-stem auditory pathway: evidence from derived auditory brain-stem responses (ABRs). Journal of the Acoustical Society of America 91:1576-1586. Ponton CW, Eggermont JJ, Kwong B, Don M (2000). Maturation of human central auditory system activity: evidence from multi-channel evoked potentials. Clinical Neurophysiology 111:220-236. Postelmans JT, Grolman W, Tange RA, Stokroos RJ (2009). Comparison of two approaches to the surgical management of cochlear implantation. Laryngoscope 119:1571-1578. Pujol J, Soriano-Mas C, Ortiz H, Sebastian-Galles N, Losilla JM, Deus J (2006). Myelination of language-related areas in the developing brain. Neurology 66:339-343. Ramsden JD, Gordon K, Aschendorff A, Borucki L, Bunne M, Burdo S, Garabedian N, Grolman W, Irving R, Lesinski-Schiedat A, Loundon N, Manrique M, Martin J, Raine C, Wouters J, Papsin BC (2012). European Bilateral Pediatric Cochlear Implant Forum consensus statement. Otology & Neurotology 33:561-565. Renier L, De Volder AG, Rauschecker JP (2014). Cortical plasticity and preserved function in early blindness. Neuroscience and Biobehavioral Reviews 41:53-63. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS (1996). Evidence based medicine: what it is and what it isn’t. British Medical Journal 312:71-72. Sadato N, Yamada H, Okada T, Yoshida M, Hasegawa T, Matsuki K, Yonekura Y, Itoh H (2004). Age-dependent plasticity in the superior temporal sulcus in deaf humans: a functional MRI study. BMC Neuroscience 5:56. Sandmann P, Dillier N, Eichele T, Meyer M, Kegel A, Pascual-Marqui RD, Marcar VL, Jancke L, Debener S (2012). Visual activation of auditory cortex reflects maladaptive plasticity in cochlear implant users. Brain 135:555-568. Schorr EA, Fox NA, van Wassenhove V, Knudsen EI (2005). Auditory-visual fusion in speech perception in children with cochlear implants. Proceedings of the National Academy of Sciences of the United States of America 102:18748-18750. Seyyedi M, Viana LM, Nadol JB, Jr. (2014). Within-subject comparison of word recognition and spiral ganglion cell count in bilateral cochlear implant recipients. Otology & Neurotology 35:1446-1450. Sharma A, Dorman MF, Spahr AJ (2002). A sensitive period for the development of the central auditory system in children with cochlear implants: implications for age of implantation. Ear and Hearing 23:532-539. Sharma A, Gilley PM, Dorman MF, Baldwin R (2007). Deprivation-induced cortical reorganization in children with cochlear implants. International Journal of Audiology 46:494-499. Smoorenburg GF, Willeboer C, van Dijk JE (2002). Speech perception in nucleus CI24M cochlear implant users with processor settings based on electrically evoked compound action potential thresholds. Audiology and Neurotology 7:335-347.
186
General discussion
Spoendlin H (1975). Retrograde degeneration of the cochlear nerve. Acta Otolaryngologica 79:266-275. Strelnikov K, Marx M, Lagleyre S, Fraysse B, Deguine O, Barone P (2015). PET-imaging of brain plasticity after cochlear implantation. Hearing Research 322:180-187. Teoh SW, Pisoni DB, Miyamoto RT (2004). Cochlear implantation in adults with prelingual deafness. Part I. Clinical results. Laryngoscope 114:1536-1540. Thai-Van H, Cozma S, Boutitie F, Disant F, Truy E, Collet L (2007). The pattern of auditory brainstem response wave V maturation in cochlear-implanted children. Clinical Neurophysiology 118:676-689. Tillein J, Heid S, Lang E, Hartmann R, Kral A (2012). Development of brainstem-evoked responses in congenital auditory deprivation. Neural Plasticity 2012:182767. Turner C, Mehr M, Hughes M, Brown C, Abbas P (2002). Within-subject predictors of speech recognition in cochlear implants: A null result. Acoustics Research Letters Online 3:95-100. UKCISG UK Cochlear Implant Study Group (2004). Criteria of candidacy for unilateral cochlear implantation in postlingually deafened adults II: cost-effectiveness analysis. Ear and Hearing 25:336-360. Vachon P, Voss P, Lassonde M, Leroux JM, Mensour B, Beaudoin G, Bourgouin P, Lepore F (2013). Reorganization of the auditory, visual and multimodal areas in early deaf individuals. Neuroscience 245:50-60. van den Heuvel MP, Kersbergen KJ, de Reus MA, Keunen K, Kahn RS, Groenendaal F, de Vries LS, Benders MJ (2015). The Neonatal Connectome During Preterm Brain Development. Cerebral Cortex 25:3000-3013 van Wermeskerken GK, van Olphen AF, van Zanten GA (2006). A comparison of intra- versus post-operatively acquired electrically evoked compound action potentials. International Journal of Audiology 45:589594. Versnel H, Agterberg MJ, de Groot JC, Smoorenburg GF, Klis SF (2007). Time course of cochlear electrophysiology and morphology after combined administration of kanamycin and furosemide. Hearing Research 231:1-12. Wu C, Huang L, Tan H, Wang Y, Zheng H, Kong L, Zheng W (2014). Diffusion tensor imaging and MR spectroscopy of microstructural alterations and metabolite concentration changes in the auditory neural pathway of pediatric congenital sensorineural hearing loss patients. Brain Research doi:10.1016/j. brainres.2014.12.025. Wunderlich JL, Cone-Wesson BK (2006). Maturation of CAEP in infants and children: A review. Hearing Research 212:212-23.
187
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Summary
Summary
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 190
Summary
Introduction Cochlear implantation has proven itself as a very successful treatment for both children and adults with severe to profound hearing loss. Although most patients obtain a high level of speech perception, not everyone benefits equally from their implants. Performance is especially variable in prelingually deaf patients, i.e. patients with an onset of severe hearing loss during early-language acquisition around the age of two years. One of the major reasons for this variance is probably related to differences in the maturation of the auditory pathway. In this thesis we have performed several studies to evaluate clinical functioning of children and adults with prelingual deafness after cochlear implantation. Besides, we studied the neurophysiologic consequences of long term hearing loss in prelingually deaf adult cochlear implant (CI) users. In the Introduction a short summary of the development of the auditory pathway is presented. During the fetal period all coarse structures of the pathway are being formed, but further refinement and maturation of this neuronal network takes years to be completed and is thought to be highly dependent on sensory input. Maturation of the pathway follows a peripheral to central pattern. Within the first years of life, maturation of the cochlear nerve and auditory brainstem is completed. However, maturation of the auditory cortex is highly complex and stretches during childhood. Several electrophysiological recordings have been proposed as biomarkers to investigate auditory pathway maturation. An overview of these measures, including the auditory brainstem and cortical response is given in the second part of the Introduction. Especially the auditory brainstem response (ABR) and the cortical auditory evoked potential (CAEP) are found to be potential biomarkers, since alterations in waveform morphology coincide with pathway maturation. Expanding on previous findings reported in the literature, the general aim of this thesis is to reveal the effects of prelingual deafness on auditory pathway maturation and propose evidence based strategies to prevent maturational deficits to occur. In the first part of this thesis, we focus on early detection of hearing loss in children and the clinical effects of bilateral cochlear implantation in prelingually deaf children. In the second part, we use ABR and CAEP recordings to investigate if long term, prelingual deafness leads to an impaired maturation of the auditory pathway affecting postoperative performance. In the third part we assess the clinical feasibility of several electrophysiological measures and give a first insight whether a more complex cortical potential could be of clinical interest for the future to reveal differences in auditory pathway functioning and maturation.
191
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Summary
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Part I In Chapter 1.1 we performed a retrospective cohort study in the University Medical Center Utrecht and Medical University Hannover to evaluate if the introduction of the newborn hearing screening program has resulted in earlier treatment of severe hearing loss. The results revealed a gradual decline in the age at implantation in both centers over the last decades. Additionally, the introduction of the screening in the Netherlands resulted in a further decline in the age at implantation. Comparing 4-year epochs immediately before and after introduction of the Dutch screening showed a significant mean decrease in age at implantation from 2.4 to 1.2 years, and an increase in the percentage of early implanted children (under the age of one year) from 9% to 37%. In contrast, this effect was not seen in the German population. This difference might be explained by differences in the follow-up and referral after a negative result at the hearing screening. Whereas in the Netherlands a rigorous tracking system is available, ensuring a very low loss to follow-up rate, a national tracking system in Germany is not present at the moment. In the first part of this thesis we also present two systematic reviews studying the effect of bilateral cochlear implantation in children (Chapter 1.2 and 1.3). A careful evaluation of the literature revealed that although randomized controlled trials are lacking, the current best evidence suggests the beneficial effect of a second implant in children. By using a best-evidence synthesis we identified that based on the combined results from several studies substantial benefits in sound localization can be expected after bilateral implantation. Moreover, the second implant might also aid in speech- and language development of the child (Chapter 1.2). The second review presented in Chapter 1.3 showed a negative association between the inter-implant interval and postoperative speech- and language development. Multiple studies indicated that a prolonged interval between the first and second implantation resulted in poorer postoperative performance. Comparative cohort studies claiming superior outcomes after simultaneous implantation, underline the negative association between the duration of the inter-implant interval and postoperative performance.
Part II The second part of this thesis discusses the consequences of early-onset and long-term deafness on auditory pathway maturation and performance after cochlear implantation. In Chapter 2.1 we present a retrospective cohort study evaluating the performance of all 48 prelingually deaf, and late-implanted adult CI users who received their implants in the University Medical Center Utrecht between 2000 and 2013. Their average maximum speech perception scores appeared to be only 25% correct. Eventually, even more than 20% of
192
Summary
this population decided to stop using their implant due to insufficient benefit. Multivariate analyses identified that preoperative speech perception and the amount of residual hearing were both positive independent predictors of postoperative speech perception. Almost half of the patients experienced little to no benefit of their implants. These poor performing subjects were found to have little to no residual hearing and to rely more on sign language than the better performing patients. A hampered maturation of the auditory pathway during childhood seems to be responsible for their poorer performance. The neurophysiologic studies presented in Chapter 2.2 and 2.3 provide additional evidence to support this hypothesis. In the study described in Chapter 2.2 we compared auditory brainstem activation in prelingually deaf and late-implanted adult CI users to that in postlingually deaf CI users. Electrically evoked auditory brainstem responses (eABRs) were recorded by stimulation of a middle and apical electrode of the CI. In both groups typical brainstem responses could be obtained, with clear waves III and V. However, analyses of wave latencies demonstrated that wave V was significantly delayed in the prelingually deaf CI users. Accordingly, the wave III-V interwave interval was also prolonged in the prelingually deaf when the apical electrode was stimulated. These results suggest a slower neural conduction of especially the rostral part of the auditory brainstem. This is likely caused by an impaired maturation, due to the long term hearing loss during childhood. The findings that typical waves III and V could be evoked in the prelingually deaf group indicates the development of the coarse structures of the auditory brainstem. Refinement of this coarse network is likely to be more sensory driven, reflected by the delayed wave latencies found in the prelingually deaf CI users. Comparable findings were found in the auditory cortex of these patients. In Chapter 2.3 we studied the cortical activity in both patient groups, using CAEP recordings. As with the ABRs, typical CAEP waveforms with the adult-like P1-N1-P2 morphology could be recorded in both groups. However, latencies of the N1 through were significantly earlier, and amplitudes were significantly larger in the prelingually deaf CI users. Again, the presence of the typical CAEP waveform indicates that even in the absence of auditory stimulation during childhood, the coarse neural network of the auditory cortex has developed. On the other hand, the earlier and larger N1 might represent activation of the more immature and less complex components of the auditory cortex. The N1 wave latencies of the postlingually deaf CI users, which are in line with those found in normal hearing, are likely to reflect activation of the mature neural network which is still present in these patients. These maturational deficits found in the auditory brainstem (Chapter 2.2) and cortex (Chapter 2.3) are likely to contribute to the poor postoperative performance of prelingually deaf CI users with long-term deafness.
193
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Summary
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Part III In the third part of this thesis we discuss the clinical feasibility of recording objective responses of the auditory pathway. Besides, we present a pilot study on an alternative cortical response which might have a better correlation with postoperative functioning. Although numerous objective responses have been proposed to evaluate CI users, only the electrically evoked compound action potential (eCAP) is routinely used in the daily practice. Automated programs, provided by the various cochlear implant manufactures have made it very easy to record these potentials without the need of additional equipment. These responses, which reflect activation of the neuronal cells within the auditory nerve, are widely used during surgery to measure implant integrity. The intraoperative success rate of recording the eCAP is very high and is reported to be around 90%. However, the postoperative success rate of recording the eCAP, as presented in Chapter 3.1, was found to be only 53% to 62%. This relatively poor success rate may limit the clinical value of the eCAP in the postoperative phase. Therefore, we evaluated if eABR and eCAEP recordings could be alternative methods. The postoperative success rate of recording the eABR ranged from 73% to 87% on the more apical electrodes, whereas the eCAEP could be evoked in more than 90% of the patients regardless of stimulating electrode location. These high success rates postulate their value as reliable alternative objective measures. Given these results, we recommend to perform eABR or eCAEP recordings in cases in which eCAPs are not measurable or additional information on the auditory pathway is required. As presented in the Chapters 2.2, 2.3 and 3.1 of this thesis, and in the current literature, there appear to be no consistent correlations between the various measures and postoperative performance. By introducing frequency or intensity changes within ongoing stimuli, more complex and higher order cortical processes can be activated. A change in a continuous tone results in a cortical response, reflecting the ability to discern this alteration. This evoked response is frequently called the Acoustic Change Complex (ACC) and might be better correlated to performance than the cortical responses to the onset of sound stimuli. Before studies can be initiated to evaluate a possible correlation of the ACC with psychophysical frequency discrimination and speech intelligibility, additional knowledge is required to identify which alterations contribute to the ACC waveform. In Chapter 3.2, we present the results of a pilot study which identified that changing the size, and the velocity of a frequency change results in alterations of the ACC waveform. Larger, and faster frequency changes were found to result in larger amplitudes and earlier peak latencies, indicating enhanced neural synchrony. For future studies and clinical use it is important to take these findings in consideration by altering only one factor.
194
Summary
Discussion In the final discussion of this thesis the results of the various studies and the clinical consequences are reviewed. The discussion on the results presented in the first part elaborates on the importance of early, and bilateral cochlear implantation in children. Given the results of the two systematic reviews indicating the positive effect of bilateral cochlear implantation in children, these studies have served as the scientific basis for two reports for the Dutch Health Care Insurance Board, which eventually resulted in the reimbursement of bilateral cochlear implantation for children in the Netherlands. The discussion on the results presented in the second part of this thesis focusses on the neurophysiologic consequences of long term and early-onset deafness in light of the current literature. By combining our results with those found in previous studies in both humans and animals, the multifactorial consequences of auditory deprivation are reviewed. By integrating these various aspects, including impaired myelination and cross-modal reorganization, we have proposed a model of the far stretching effects of auditory deprivation on auditory pathway maturation (Fig. 1, Discussion). This model not only gives an overview of the literature on this subject, but also provides a neurophysiologic explanation for the variable and relative poor results found after cochlear implantation in patients with long-term hearing loss.
195
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Nederlandse samenvatting
Nederlandse samenvatting
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 198
Nederlandse samenvatting
Introductie Met de komst van de meerkanaals cochleaire implantaten is er een succesvolle behandeling voor ernstige slechthorendheid en doofheid met uiteenlopende oorzaken gekomen. Nadat de elektroden van het cochleaire implantaat (CI) chirurgisch in de cochlea (het slakkenhuis) zijn ingebracht, kunnen elektrische stroompulsen de gehoorzenuw activeren en zijn patiënten in staat om geluiden waar te nemen. Hoewel de meeste CI gebruikers een goed spraakverstaan niveau kunnen behalen, blijken de resultaten van vooral dove volwassenen met vroegtijdige doofheid vaak tegen te vallen. Deze zogenaamd prelinguaal doven, die doof zijn geworden in het begin van de taal-spraakontwikkeling vóór de leeftijd van 2 jaar, lijken minder profijt te hebben van hun CI als deze pas op late leeftijd wordt geïmplanteerd. Eén van de belangrijkste oorzaken hiervoor is waarschijnlijk het gevolg van een afwijkende uitrijping van het auditieve netwerk in de hersenstam en hersenen. In dit proefschrift beschrijven we verscheidene onderzoeken die zijn verricht om een beter begrip te krijgen in het functioneren van prelinguaal dove kinderen en volwassenen met een CI. In het bijzonder bestuderen we de neurofysiologische gevolgen van langdurige, prelinguale doofheid op de uitrijping van het auditieve systeem in volwassen CI gebruikers. Om inzicht te krijgen in de gevolgen van doofheid tijdens de ontwikkelingsfase wordt in de Introductie van dit boek een beknopt overzicht gegeven van het verloop van de normale ontwikkeling van het auditieve systeem. Voor de geboorte lijken alle structuren en paden van het auditieve netwerk reeds te zijn aangelegd, maar voor de verdere uitrijping van het systeem is auditieve stimulatie noodzakelijk. De uitrijping van het auditieve systeem is als eerste voltooid in de cochlea en de gehoorzenuw en pas later in de hersenstam. De ontwikkeling van het specifieke hersengedeelte dat verantwoordelijk is voor de verwerking van auditieve signalen (de auditieve cortex) duurt echter nog veel langer en loopt de hele jeugd door. Verschillende elektrofysiologische metingen zijn in de afgelopen jaren onderzocht om de uitrijping van het auditieve systeem te vervolgen. Het meten van de hersen- en hersenstampotentialen, respectievelijk de ‘cortical auditory evoked potential’ (CAEP) en de ‘auditory brainstem reponse’ (ABR), zijn in de literatuur beschreven als biomarkers voor de ontwikkeling van het auditieve systeem, aangezien veranderingen in deze potentialen samenvallen met veranderingen in de uitrijping van het auditieve systeem. Het doel van dit proefschrift is om met behulp van deze bevindingen uit de literatuur na te gaan wat de effecten zijn van prelinguale doofheid op de uitrijping van het auditieve systeem en om na te gaan wat de gevolgen zijn voor het horen met een CI. Daarnaast willen wij aan de hand van best gepubliceerde onderzoeken een wetenschappelijk onderbouwde behandelrichtlijn voorstellen om daarmee ontwikkelingsstoornissen van het auditieve systeem te verlichten of zelfs te voorkomen. Het eerste deel van dit proefschrift richt zich op het vroegtijdig opsporen
199
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Nederlandse samenvatting
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
van slechthorendheid bij pasgeborenen en het vroegtijdig implanteren van twee CIs (bilaterale implantatie) bij kinderen met prelinguale doofheid. In het tweede deel evalueren we het postoperatieve functioneren van volwassen CI patiënten met langdurige prelinguale doofheid en onderzoeken we met behulp van elektrofysiologische ABR en CAEP metingen de gevolgen hiervan op de uitrijping van het auditieve systeem. In het derde deel van het proefschrift gaan we de klinische relevantie na van ABR, CAEP en de veelgebruikte eCAP (electrically evoked compound action potential) metingen. Ten slotte presenteren we de resultaten van een vooronderzoek naar hersenpotentialen opgewekt door complexe geluidsstimuli, welke potentieel veelbelovend zijn voor klinische toepassingen en wetenschappelijke onderzoeken.
Deel I In Hoofdstuk 1.1 presenteren we de resultaten van een retrospectief cohortonderzoek verricht in de databases van de CI centra van het Universitair Medisch Centrum Utrecht en de Medizinische Hochschule Hannover. Het doel van dit onderzoek was om na te gaan of in de afgelopen twintig jaar de leeftijd van cochleaire implantatie bij kinderen met prelinguale doofheid is veranderd. Daarnaast hebben wij onderzocht of de invoering van de landelijke neonatale gehoorscreening heeft geleid tot eerdere cochleaire implantatie. De resultaten toonden aan dat er de afgelopen jaren in beide centra een geleidelijke afname is in de leeftijd van implantatie. De invoering van de neonatale gehoorscreening leidde alleen in de Utrechtse populatie tot een significante daling in de leeftijd van implantatie. Als we de gemiddelde leeftijd van de vier jaar voor de invoering van de screening (landelijke invoering in Nederland in juli 2006) vergelijken met de vier jaar daarna, daalde de gemiddelde leeftijd van implantatie van 2.4 jaar tot 1.2 jaar en nam het aantal kinderen dat onder de leeftijd van 1 jaar werd geïmplanteerd toe van 9% tot 37%. In de populatie van Hannover was dit effect echter niet aanwezig. Dit zou verklaard kunnen worden door verschillen in het vervolgtraject bij een afwijkende testuitslag tussen beide landen. Een surveillance systeem waarborgt in Nederland dat voor alle kinderen met een afwijkende test het juiste vervolgtraject wordt ingezet om daarmee de uitval na de screening tot een minimum te beperken. Een dergelijk nationaal vervolgsysteem is in Duitsland echter niet aanwezig, waardoor het mogelijk zou kunnen zijn dat ouders zich aanvankelijk aan het vervolgtraject onttrekken en zich pas later melden als de gevolgen van de slechthorendheid duidelijk waarneembaar worden. Hoofdstuk 1.2 en 1.3 bestaan uit twee systematische literatuurreviews waarin we de effectiviteit van bilaterale cochleaire implantatie bij kinderen met prelinguale doofheid onderzoeken (Hoofdstuk 1.2) en het effect van het tijdsinterval tussen de eerste en tweede implantatie op het uiteindelijk functioneren met beide CI’s bestuderen (Hoofdstuk 1.3). Hoewel deze literatuuronderzoeken aantoonden dat er geen gerandomiseerde onderzoeken
200
Nederlandse samenvatting
naar de meerwaarde van bilaterale cochleaire implantatie bij kinderen gepubliceerd zijn, geeft een ´best evidence synthesis´ analyse aan dat er een gunstig effect van bilaterale implantatie lijkt te zijn op het lokaliseren van geluiden en mogelijk ook op de taal-spraakontwikkeling van kinderen (Hoofdstuk 1.2). Het literatuuronderzoek in Hoofdstuk 1.3 toont aan dat het interval tussen de 1e en 2e implantatie een negatief voorspellende factor voor de postoperatieve taal-spraakontwikkeling is. Daarnaast lieten (niet-gerandomiseerde) vergelijkende studies zien dat kinderen beter functioneren na simultane bilaterale implantatie dan na sequentiële implantaties. Deze bevindingen benadrukken de negatieve factor van de duur van het interval tussen de 1e en 2e implantatie op het postoperatief functioneren.
Deel II In het tweede deel van het proefschrift bespreken we de gevolgen van vroegtijdige en langdurige doofheid op de uitrijping van het auditieve systeem en de gevolgen hiervan op het functioneren na cochleaire implantatie. In Hoofdstuk 2.1 hebben we het functioneren van alle 48 prelinguaal dove en laat geïmplanteerde volwassen CI patiënten, die tussen 2000 en 2013 in het UMC Utrecht hun CI kregen, onderzocht. Hun gemiddelde maximale postoperatieve spraakverstaanscore was slechts 25% correct. Uiteindelijk besloot 21% van deze populatie hun CI niet meer te gebruiken, omdat ze hiervan onvoldoende baat hadden. Multivariate analyses toonden aan dat het preoperatieve spraakverstaan en de hoeveelheid restgehoor voorspellende factoren zijn voor het postoperatieve spraakverstaan. Bijna de helft van alle patiënten ervaarden weinig of zelfs geen meerwaarde van hun CI. Zij bleken minder preoperatief restgehoor te hebben dan de patiënten die meer baat van hun CI rapporteerden. Deze tegenvallende resultaten kunnen goed verklaard worden door een verstoorde uitrijping van het auditieve systeem ten gevolge van slechthorendheid tijdens hun jeugd. De resultaten in Hoofdstukken 2.2 en 2.3 laten zien dat er ook neurofysiologische onderbouwing is van deze hypothese. In Hoofdstuk 2.2 vergelijken we de herstenstampotentialen van prelinguaal dove en laat geïmplanteerde, volwassen CI patiënten met die van postlinguaal dove, volwassen CI patiënten. ABRs werden opgewekt door stroompulsen die werden aangeboden via een elektrode in het midden of aan het apicale einde van de CI te stimuleren. In beide groepen CI patiënten konden karakteristieke ABR golfvormen met duidelijke pieken III en V worden opgewekt. Analyses van de pieklatenties liet echter zien dat piek V significant later was in de prelinguaal dove groep dan in de postlinguaal dove groep. Daarnaast was ook het piek III-V interval verlengd in de prelinguale groep bij stimulatie van de apicale CI elektrode. Deze bevindingen suggereren een vertraagde neurale voortgeleiding binnen het meer craniale deel van de hersenstam. Deze vertraagde voortgeleiding is mogelijk het gevolg van een gestoorde
201
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Nederlandse samenvatting
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
myelinisatie van de neuronen in de hersenstam. De aanwezigheid van het vertrouwde ABR golfpatroon duidt er op dat de elementaire structuren in de hersenstam zijn aangelegd, maar dat de uitrijping van dit deel van het netwerk verstoord is door onvoldoende auditieve stimulatie tijdens de jeugd. Auditieve hersenpotentialen werden in dezelfde patiënten opgewekt met korte reeksen van stroompulsen aangeboden via basaal, midden of apicaal gelegen CI elektroden. Deze elektrofysiologische metingen lieten eveneens zien dat normale CAEP golfvormen met een volwassen P1-N1-P2 configuratie in zowel prelinguaal als postlinguaal doven konden worden opgewekt (Hoofdstuk 2.3). De N1 piek was echter significant vroeger en groter in de prelinguaal dan in de postlinguaal dove CI groep. Ook hier geeft de aanwezigheid van de normale CAEP golfvorm in de prelinguale groep aan dat de elementaire structuren in de hersenen die nodig zijn om deze golfvorm op te wekken zich ontwikkeld hebben, maar dat langdurige doofheid tijdens de jeugd, nu leidt tot activatie van een nog onvoldoende uitgerijpt corticaal neuraal netwerk. De N1 latenties van de postlinguaal dove CI gebruikers liggen daarentegen in het bereik van normaalhorende mensen, wat er mogelijk op duidt dat het CI bij deze patiënten het reeds uitgerijpte auditieve systeem opnieuw activeert. De bevinding dat vroegtijdige en langdurige doofheid leidt tot een gestoorde uitrijping van zowel de hersenstam als de auditieve cortex draagt zeer waarschijnlijk bij aan de vaak matige tot teleurstellende resultaten na cochleaire implantatie bij prelinguaal dove volwassenen.
Deel III In het derde deel van dit proefschrift bespreken we allereerst de klinische uitvoerbaarheid en toepasbaarheid van drie verschillende elektrofysiologische metingen bij CI patiënten (Hoofdstuk 3.1). Ondanks het feit dat er in de afgelopen jaren verscheidene objectieve metingen zijn ontwikkeld, wordt tot op heden slechts de ‘electrically evoked compound action potential’ (eCAP) frequent in de dagelijkse kliniek gebruikt. Deze meting kan namelijk eenvoudig verricht worden door middel van geautomatiseerde programma’s zonder tussenkomst van aanvullende apparatuur. De eCAP wordt opgewekt door een stroompuls te sturen naar één van de CI elektrodes. De hierop volgende activatie van de gehoorzenuw kan door een naburige CI elektrode worden gemeten, hetgeen wordt weergegeven als de eCAP respons. De gebruiksvriendelijkheid van deze meting en de informatie die verkregen wordt over het functioneren van het CI en de gehoorzenuw, maakt dat de eCAP veelvuldig tijdens en na de operatie gebruikt wordt. Het peroperatieve succespercentage om deze meting uit te voeren is zeer hoog en ligt rond de 90%. In de postoperatieve fase ligt dit percentage echter veel lager. In onze studie beschreven in Hoofdstuk 3.1 bleek dat deze meting slechts in 53% tot 62% van de metingen mogelijk was, waardoor de klinische waarde van deze meting in de
202
Nederlandse samenvatting
postoperatieve fase te betwijfelen valt. In Hoofdstuk 3.1 hebben wij daarom onderzocht of ABRs of CAEPs een beter alternatief zijn. Het postoperatieve succespercentage van de ABR meting lag tussen 73% en 87% en voor de CAEP metingen was dit 90%. Deze resultaten laten zien dat ABRs en CAEPS een betrouwbaar alternatief kunnen zijn. Wij adviseren derhalve om in het geval dat eCAP metingen niet mogelijk zijn of indien meer informatie over het auditieve systeem gewenst is, ABR en/of CAEP metingen te verrichten. Hoewel we met deze drie metingen de activiteit van de verschillende niveaus van het auditieve systeem kunnen weergeven, zijn er in de huidige literatuur geen consistente correlaties beschreven tussen deze verschillende metingen en het postoperatief functioneren van CI gebruikers. Ook de resultaten uit de Hoofdstukken 2.2 en 2.3 lieten geen sterke correlaties zien. Mogelijkerwijs kunnen elektrofysiologische metingen met complexe stimuli, hogere corticale verwerkingsprocessen registreren die een sterkere relatie laten zien met het postoperatieve spraakverstaan. Door veranderingen in bijvoorbeeld frequentie of geluidsniveau te introduceren, veranderingen die ook voorkomen in spraak, worden specifieke neuronen geactiveerd. Deze corticale activatie kan, net zoals de CAEP, geregistreerd worden, en wordt vaak het â&#x20AC;&#x2DC;Acoustic Change Complexâ&#x20AC;&#x2122; genoemd (ACC). Om inzicht te verkrijgen in de verschillende factoren die van invloed zijn op de ACC, hebben wij een vooronderzoek verricht waarin we hebben onderzocht wat de invloed is van de grootte en de snelheid van frequentieveranderingen op de ACC golfvorm. In hoofdstuk 3.2 laten wij zien dat de amplitude van de ACC golfvorm toeneemt als de grootte en de snelheid van de frequentieveranderingen toeneemt. Daarnaast zien wij dat de pieklatenties afnemen bij een toename van de grootte en snelheid. Deze bevindingen worden nu gebruikt voor vervolgonderzoek naar de ACC bij CI gebruikers en patiĂŤnten met perceptieve slechthorendheid.
Discussie In de discussie van dit proefschrift bespreken we de resultaten van de verschillende studies in het licht van de huidige literatuur en dagelijkse praktijk. De resultaten beschreven in het eerste deel van dit proefschrift benadrukken het belang van vroegtijdige en bilaterale cochleaire implantatie bij kinderen. De twee literatuuronderzoeken hebben de wetenschappelijke bewijskracht gevormd waardoor het College van Zorgverzekeringen in 2012 heeft besloten om bilaterale cochleaire implantatie op te nemen in het basispakket van de zorgverzekering in Nederland. De discussie over het tweede deel van het proefschrift richt zich op de neurofysiologische gevolgen van langdurige en vroegtijdige doofheid op de uitrijping van het auditieve systeem. Door onze resultaten te combineren met andere gevolgen welke reeds beschreven zijn in de literatuur, zoals verminderde myelinisatie en cross-modale reorganisatie van de hersenen,
203
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Nederlandse samenvatting
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
hebben wij een multifactorieel model opgesteld dat een overzicht geeft van de gevolgen van doofheid op het auditieve systeem (Fig. 1, Disucssion). Dit model geeft niet alleen inzicht in de huidige literatuur, maar ook een neurofysiologische verklaring voor de wisselende en soms teleurstellende resultaten die we zien na implantatie van patiĂŤnten met langdurige en vroegtijdige slechthorendheid.
204
Dankwoord
Dankwoord
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 206
Dankwoord
Dankwoord Het schrijven van dit proefschrift is een geweldig proces geweest en had nooit volbracht kunnen worden zonder de betrokkenheid en steun van velen. Van 1-jarigen tot 80 plussers, van KNO-artsen tot fysici, jullie hebben mij allen geholpen om mijn doel te bereiken. Op deze plaats wil ik jullie hartelijk danken, in het bijzonder de volgende personen. Professor Grolman, beste Wilko, ik wil u bedanken dat u mij de mogelijkheid hebt geboden om dit promotietraject te doorlopen en daarnaast de opleiding tot KNO-arts te volgen. Hoewel van de oorspronkelijke ideeën van vijf jaar geleden weinig meer terug te vinden is in dit boek, heb ik het heel prettig gevonden dat u mij de vrijheid hebt gegeven om mijn eigen ideeën uit te werken en dat u er op vertrouwde dat dit boek uiteindelijk wel een keer af zou komen. Copromotor dr. H. Versnel, beste Huib. Een begeleider wiens deur altijd open staat. Altijd open voor vragen en het geven van advies of om een kop koffie te drinken. Soms gooide je de deur hard dicht; meestal tussen 9 en 10, als Joost en ik meenden dat de gehele H.02 gang moest meegenieten met het ‘Foute Uur’. Ik ben je in ieder geval ontzettend dankbaar dat jouw deur (meestal) wagenwijd open stond en ik door jouw betrokkenheid en begeleiding dit onderzoek naar een hoger plan heb kunnen tillen. Daarnaast ben ik je dankbaar voor alle dagelijkse onderzoeksrituelen, zoals het ommetje naar het koffie automaat, waarin we niet alleen de bekende grappen, maar ook alle nieuwe ideeën de revue lieten passeren. Ik hoop nog lang deze goede samenwerking voort te zetten met onze vervolgonderzoeken. Copromotor dr. G.A. van Zanten, beste Bert. Sinds mijn wetenschappelijke stage in mijn vijfde studiejaar ben jij bijna onafgebroken bij mijn projecten betrokken geweest en heb jij deze altijd met jouw kennis ondersteund. Aanvankelijk liep ik wel eens je kamer uit met een groot vraagteken op mijn voorhoofd (zoals alle AIOS die weer eens geconfronteerd worden met fysica en de geluidsleer), maar met jouw hulp is het gelukt om een aantal mooie onderzoeken op te zetten. Ik ben je heel dankbaar voor al je adviezen en ik zal in de toekomst nog geregeld even bij je aankloppen voor jouw raad. Prof. Dr. G.J.M.G. van der Heijden, beste Geert, met jou als begeleider op methodologisch vlak hebben we een ontzettend mooi bilateraal project opgezet. Hoewel we na de inclusie van de tweede patiënt alweer konden stoppen met de studie, heb ik veel geleerd van dit proces. Ik ben blij dat een deel van ons werk een mooie plek in dit proefschrift heeft gekregen. Prof. Dr. A. Lesinski-Schiedat, vielen Dank für die gute Zusammenarbeit. Ihre Begeisterung für neue Forschungsprojekte wirkt ansteckend. Ich hoffe, dass wir in der Zukunft noch weiter zusammen arbeiten können. 207
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Dankwoord
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Beste Joost, mijn dankbetuiging aan jou houd ik kort. Niet omdat jij geen lang verhaal waard bent, maar omdat wij slechts weinig woorden nodig hebben om elkaar goed te begrijpen. Ik dank mijn vriend, mijn collega en mijn paranimf. (Oud) collega’s van de H.02, Sarah, Dyan, Juliette, Yvette, Stephanie, Nina, Alice, Inge, Laura, Bernard, Ruben en Hendrik, jullie aanwezigheid op de gang zorgden altijd voor een goede sfeer. In het bijzonder wil ik Joost en Marlien bedanken voor alle gezelligheid op de H02.104. De gemeenschappelijke voorliefde voor hitjes, kaascroissantjes en grappen, maar ook het delen van allemaal ideeën en adviezen maakten het een hele mooie en goede tijd! Dank aan alle studenten die hebben meegewerkt of nog steeds aan het werk zijn aan één van de vele projecten. Ed, Marjolijn, Thijs, Rachelle, Vanessa, Lieke en Wouter, jullie hebben top werk geleverd! In dit rijtje verdient Ed een hele bijzondere plek. Jij hebt niet alleen enorm veel werk gestoken in de kinder CI database, maar samen hebben wij ook de eerste CAEPs gemeten! Voor mij een onvergetelijk moment, net als de dag waarop we de afgedankte dialysestoel naar boven hebben gesleept. Dank voor al je hulp en de mooie tijd! Het gehele BIcEpS team met in het bijzonder Jeanette, Trudi, Adinda, Geert, Bert, Vera, Ellen en Frank. Ik heb nog nooit zulke creatieve brainstorm sessies meegemaakt: van dictafoons ophangen in de wieg tot het bedenken van de mooiste attributen met het BIcEpS logo erop. Het werken in dit multidisciplinaire team was niet alleen leuk, maar ook zeer leerzaam. Alle (oud) arts-assistenten en arts-onderzoekers van de KNO van het UMC Utrecht. Dagelijks merk ik weer wat een voorrecht het is om in zo’n leuke groep te werken. Ondanks dat de samenstelling van de groep elk jaar weer verandert, blijft de sfeer en collegialiteit fantastisch. Het doet mij altijd goed om met jullie een koffie of een klein biertje te drinken… De staf van de afdeling KNO van het UMC Utrecht en het Meander Medisch Centrum in Amersfoort, hartelijk dank voor jullie interesse en ondersteuning van mijn promotietraject. Daarnaast ben ik jullie ook zeer dankbaar voor de altijd prettige samenwerking. Ik ben blij door jullie opgeleid te worden! Beste Hanneke en Daphne, dank voor al jullie ondersteuning en het sturen van alle reminders de afgelopen jaren. KTZAR dr. J.S. de Ru, Kolonel-arts dr. E.L. van der Veen en Kolonel-arts dr. M.C.J. Aarts, beste Sander, Erwin en Mark, dank voor de mooie wetenschappelijke discussies bij de koffie en de mogelijkheid die jullie mij in het CMH boden om naast het dienen van de Koning ook nog aan mijn proefschrift te werken. 208
Dankwoord
Ik wil iedereen van het CI-team hartelijk danken voor alle steun en medewerking aan de verschillende onderzoeken. Zonder al jullie inspanning zouden we nooit zulke mooie onderzoeken kunnen opzetten en de zorg kunnen leveren aan deze bijzondere patiënten. In het bijzonder wil ik Vera Prijs bedanken voor het enthousiasmeren van alle proefpersonen, zonder jou was de inclusie nooit zo goed verlopen! Many thanks to the Planet Earth film crew and David Attenborough for creating this stunning nature documentary, which kept all study participants in the perfect state of mind for cortical response recordings. Mijn bijzondere dank gaat uit naar alle proefpersonen, die helemaal kaal gescrubd en vol beplakt met elektrodes, urenlang naar Planet Earth hebben moeten kijken. Dank voor jullie betrokkenheid en jullie inzet voor het wetenschappelijk onderzoek. De labgenoten van het eerste uur, Sjaak, Ferry, John, Piet, Frits, Frans en natuurlijk ook Huib als nestor van de H.02, dank voor de altijd vermakelijke donderdag koffie. Dank aan iedereen die zo geduldig al mijn ideeën heeft moeten aanhoren. Vooral mijn kameren ganggenoten hebben het moeten verduren als ik wéér zo’n ‘briljant’ idee had bedacht. Het was altijd leuk om met jullie hierover te brainstormen. Wellicht worden een paar van die ideeën toch nog een keer werkelijkheid.... Dyan, nu ik hier weer over nadenk, we moeten echt nog wat gaan doen met onze humane en cavia data…. I would like to thank everyone at Cochlear who has helped me with this project. In particular I would like to thank Niels and Petra for their support. Alle oud huisgenoten van ’t Sticht tot Amsterdam, het was altijd zeer prettig om na een hele dag praten over onderzoek met jullie over de meest zinloze dingen van het leven te praten. De Heeren van jaarclub Vortex, hoewel we elkaar de laatste jaren niet meer zo geregeld zien als ik zou willen door mijn leven in het verre Ooosten en het afronden van dit werk, blijft het ontzettend fijn om zo’n groep vrienden te hebben. De diverse werklocaties op deze planeet, van de muffe kooi van Faraday waarin het zuurstofgehalte langzaam tot nul gezakt was na een hele dag meten, tot het witte zandstrand van Ilha do Sal. Soms kom je op de meest gekke plaatsen tot de beste ideeën.
209
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Dankwoord
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Vielen Dank an meine liebe Familie in Isernhagen, danke Meike, Ulli, Markus und Hendrik. Danke für eure Unterstützung, auch wenn ich heimlich an den Feiertagen einige Korrekturen durchführen musste. Lieve Laureen, als mijn grote zus heb jij mij altijd gesteund en van goede raad voorzien. Niet alleen heb je me geholpen met de keuze van de lay-out van dit boek, maar ook met zoveel andere keuzes in mijn leven, waarover ik met jou altijd ruggespraak kon houden. Het was voor mij dan ook vanzelfsprekend dat jij één van mijn paranimfen zou zijn en vindt het heel fijn dat jij nu weer achter mij wilt komen staan. Lieve Maarten, toen ik tijdens jouw promotie in de senaatszaal zat, wist ik het zeker: ik wil hier ook staan. Dank voor die inspiratie! Lieve Joris, je beseft het waarschijnlijk nog niet, maar jij hebt een aanzienlijke bijdrage geleverd aan dit proefschrift. Jouw uitzonderlijk coöperatieve bijdrage als eerste 1-jarige proefpersoon, heeft mij er uiteindelijk toe doen besluiten dat ik het hele onderzoek over een andere boeg heb gegooid. Mijn dank is groot, dit boek is het resultaat! Lieve Justijn, jij hebt mazzel dat je een oudere broer hebt! Lieve Jan-Willem en Iris, lieve pap en mam, dank voor al jullie steun en mogelijkheden die jullie mij hebben geboden. Jullie enthousiasme voor de medische wereld en het wetenschappelijk onderzoek heeft mij geïnspireerd om deze weg te kiezen. Ik ben ontzettend blij dat wij als familie dit bijzondere moment kunnen vieren. Lieber Marion, Ich danke dir von ganzem Herzen dass du es mir möglich gemacht hast diese Promotionsarbeit in Utrecht zu vollbringen. Obwohl es für uns nicht immer einfach war einen Weg zu finden in diese Situation, haben wir es durchstanden und an einander geglaubt das hinter die Wolken auch noch die Sonne scheint. Jetzt strahlt die Sonne auf unsere Zukunft. Ich hab dich lieb.
210
List of publications About the author
List of publications
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 212
List of publications
List of publications Lammers MJW, van Eijl RHM, van Zanten GA, Versnel H, Grolman W. Delayed auditory brainstem responses in prelingually deaf and late-implanted cochlear implant users. Journal of the Association for Research in Otolaryngology 2015; 16(5):669-678 Lammers MJW, Jansen TTG, Grolman W, Lenarz T, Versnel H, van Zanten GA, Topsakal V, Lesinksi-Schiedat A. The influence of newborn hearing screening on the age at cochlear implantation in children. Laryngoscope 2015; 125(4):985-90 Lammers MJW, Vernsel H, van Zanten GA, Grolman W. Altered cortical activity in prelingually deafened cochlear implant users following long periods of auditory deprviation. Journal of the Association for Research in Otolaryngology 2015; 16(1):159-70 Lammers MJW, Lenarz T, van Zanten GA, Grolman W, B端chner A. Sound localization abilities of unilateral hybrid cochlear implant users with bilateral low-frequency hearing. Otology & Neurotology 2014; 35(8):1433-9 Lammers MJW, van der Heijden GJMG, Pourier VEC, Grolman W. Bilateral cochlear implantation in children: a systematic review and best-evidence synthesis. Laryngoscope 2014; 124(7):1694-9 Lammers MJW, Venekamp RP, Grolman W, van der Heijden GJ. Bilateral cochlear implantation in children and the impact of the inter-implant interval. Laryngoscope 2014; 124(4):993-9 Havenith S, Lammers MJW, Tange RA, Trabalzini F, Della Volpe A, van der Heijden GJMG, Grolman W. Hearing preservation surgery: cochleostomy or round window approach? A systematic review. Otology & Neurotology 2013; 34(4):667-74 Lammers MJW, Lo Galbo AM, Buwalda J. Pathology quiz case 1. Allergic fungal rhinosinusitis (AFRS). Archives of Otolaryngology Head and Neck Surgery 2012; 138(4):426-8 Lammers MJW, Grolman W, Smulders YE, Rovers MM. The cost-utility of bilateral cochlear implantation: a systematic review. Laryngoscope 2011; 121(12):2604-9
213
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
About the author
About the author
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 216
About the author
About the author Marc Jan-Willem Lammers was born on August 21th 1985 in Nijmegen, the Netherlands. He attended the Nieuwe Lyceum in Bilthoven, where he received his gymnasium diploma in 2003. Afterwards, he studied Medicine at Utrecht University and obtained his medical degree in 2010. During the final year of his study he participated in a research project on the sound localization abilities of Nucleus Hybrid cochlear implant users at the Medical University of Hannover, Germany, supervised by Prof. dr. A. B端chner and Prof. dr. T. Lenarz. In 2010, he started his PhD at the department of Otorhinolaryngology and Head & Neck Surgery of the University Medical Center Utrecht under supervision of Prof. dr. W. Grolman, dr. H. Versnel and dr. G.A. van Zanten. The results of his PhD project are described in this thesis and have been presented at national and international conferences. In December 2012, he became a resident at the department of Otorhinolaryngology and Head & Neck Surgery of the University Medical Center Utrecht (supervisor Prof. dr. W. Grolman). During his residency he also worked at the Meander Medical Center in Amersfoort under supervision of Prof. dr. H.F. Mahieu.
217
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39
Auditory pathway functioning in prelingual deafness The clinical consequences for cochlear implantation
Marc Lammers
104 ISBN 978-90-393-6435-2