how to choose a ventilator

BG-09 Ventilators:BG-09 Ventilators

09

94

20/8/08

23:13

Page 94

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 95

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

09

Correspondence C. Gregoretti Dipartimento Emergenza Accettazione ASO CTO-CRF-Maria Adelaide Via Zuretti 29 10129 Torino Italy Fax: 39 116933266 E-mail: c.gregoretti@tiscali.it

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR C. Gregoretti, P. Navalesi, I. Tosetti and P. Pelosi The purpose of this guide A range of ventilators is available to the intensive care unit (ICU) physician, something that may in theory enable him or her to treat different forms of respiratory failure more effectively. New ventilatory modes are frequently introduced, but could this novelty sometimes be considered no more than a marketing tool?

Although these new modes might to some extent assist the clinician in daily practice, scientific evidence proving their effectiveness is often lacking [1]. When purchasing a ventilator, clinicians rely on personal experience and on the results of small observational trials showing positive effects on physiological variables, such as oxygenation and work of breathing; and on subjective variables, such as patient comfort [1].

Ideally, the process of purchasing a ventilator should be based on a strong scientific rationale [2, 3], founded on predetermined requisites and scores [4, 5]. Because manufacturers’ specifications alone are of limited importance, they should not be a prominent factor in decision making. A basic knowledge of the principles of

ventilator functioning may be helpful when choosing a mechanical ventilator, enabling the purchaser to consider its technical performance in relation to the clinical characteristics of the patients to be treated, the healthcare environment and the financial resources available [6, 7].

The primary purpose of this guide is not to compare the individual features of each ventilator, but rather to provide information about their technical aspects, which may assist in the purchasing decision [6].

Table 1 Fundamental elements of a positive pressure ventilator

1. Pneumatic system 2. Inspiratory and expiratory valves and output variable control 3. Phase variables: trigger, cycling off and limit 4. Modes of mechanical ventilation 5. Control system 6. User interface 7. Safety and alarm systems 8. Monitoring system

Basic principles of ventilator function Briefly, a mechanical ventilator can be considered as a series of consecutive functions that turn an input (energy) into an output (ventilatory variable), such as pressure, flow or volume. It can transfer energy by applying positive pressure to the airways, acting as a positive pressure ventilator (PPV), or by applying subatmospheric pressure externally to the chest, acting as a negative pressure ventilator. This article will focus solely on PPVs. There are several fundamental elements to a PPV [6, 8], as shown in table 1.

9. Respiratory circuit and accessories

1. Pneumatic system All ICU PPVs, except for portable transport ventilators that may be driven pneumatically, require electricity (AC external power or a DC internal battery).

The gas source can be: a) an external high-pressure gas (centralised gas system or tanks); b) an internal compressor;

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

95

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:13

Page 96

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

c) a turbine or piston; or d) a combination of (a +c) or (a+b).

Based on the gas source, ICU PPVs can practically be divided into two categories: those that work with oxygen and air at high pressure (4 atm (400 kPa)); and those that work with oxygen at high pressure (4 atm (400 kPa)) and atmospheric air.

Category 1. Oxygen and air at high pressure PPVs High-pressure air and oxygen are the external source of energy for this type of ventilator (figs 1–8).

Pressure inside the ventilator is then reduced to atmospheric pressure by a “pressure reducer device” to allow the patient to inhale. While some old ICU PPVs allowed the physician to manipulate the working (driving) pressure, for safety reasons none of the current generation of PPVs allow this [6]. The peak flow output of these PPVs is usually up to 200 L·min-1 (with constant flow about 130–140 L·min-1) with a very fast pressure rise time (pressure in a given time with flow as the dependent variable) when using assisted pressure-limited, timecycled (e.g. assisted pressure-

Table 2 Ways to set pressure rise time

Percentage of pressure slope Percentage or a given value of flow# Steepness of pressure ramp gradient Pictograms depicting different pressure ramps # : In this case, pressure slope changes according to flow setting.

controlled ventilation (AC/PCV)) or flow-cycled ventilatory support (e.g. pressure support ventilation (PSV)) [9]. Pressure rise time may be

Figure 1. eVent Inspiration LS category 1 ventilator

Figure 2. Hamilton Medical Galileo, G5 and Raphael XTC category 1 ventilator

Figure 3. Taema Extend category 1 ventilator

96

Figure 4. Maquet Servo i category 1 ventilator

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

Figure 5. Covidien PB 840 category 1 ventilator

Figure 6. Dräger Evita XL category 1 ventilator

Figure 7. GE Engstrom Carestation category 1 ventilator

Figure 8. Viasys Avea category 1 ventilator (Cardinal Health)

BG-09 Ventilators:BG-09 Ventilators

09

21/8/08

09:11

Page 98

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR PRODUCT LISTINGS

regulated in most of these ventilators and its setting may have a real clinical application [6, 10, 11].

Table 2 shows the methods of define pressure rise time.

Category 2. Oxygen at high pressure (4 atm) plus atmospheric air PPVs In these ventilators, a piston or turbine sucks atmospheric air from the environment. Although pistonor turbine-driven ventilators are usually used for home-care positive pressure ventilation [12], the simultaneous use of high-pressure oxygen and a sophisticated user interface make these PPVs suitable for critical care use. In the past, a major problem with turbine-driven ventilators was that in spite of a very high peak flow output (up to 240 L·min-1) in basal conditions (without any additional resistance), the application of a given resistance significantly decreased the ventilatory output (on occasion to <100 L·min-1). In addition, during volume-targeted ventilation, it was difficult to maintain a given tidal

98

volume when facing increased patient elastic or resistive load.

Fast turbines (“dynamic blower systems”; e.g. Viasys Vela and Viasys Pulmonetic LTV 1200 (both Cardinal Health), ResMed Elisée 350, VersaMed iVent 201 IC, Dräger Carina) or turbines rotating at constant speed (“constant-revolution blower systems”) driven by a proportional valve (e.g. Dräger Savina, Respironics Esprit, Respironics Vision, Taema Neftis ICU, Hamilton Medical C2) make the latest generation of turbine-driven PPVs (figs 9–19) as efficient as those driven by high-pressure gas [9].

From the standpoint of high responsiveness to patient’s flow demand, constant-revolution blower systems with a proportional valve perform very well. However, although in the past these systems had a clear responsiveness advantage over dynamic blower systems (which change speed to reach the preset ventilatory ouput), recent

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

developments show that dynamic blowers with a small blower wheel diameter and a very high revolution rate per minute are also extremely responsive to patient demand [13]. Some turbine-driven ventilators use a bleed oxygen flow to increase responsiveness; however, this has the disadvantage of high oxygen consumption.

In this ventilator category, too, pressure rise time can be adjusted.

However, the clinical significance of this may be different from that in category 1 ventilators, because pressure rise time is critical at the very beginning of inspiration (0.3 s), particularly in dynamic blower systems [9].

Gas blending Ventilators no longer incorporate accumulators, as for instance in the old Servo 900 [6]. Instead, all category 1 and some category 2 ventilators (e.g. Dräger Savina, Respironics Esprit, Respironics Vision) blend gases using an internal

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 99

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

blender driven by a proportional valve. Some PPVs use a proportional solenoid valve (e.g. Covidien PB 840, Respironics Esprit). The preset oxygen tension is constant, whatever the minute volume. Most category 2 PPVs do not have a real blender: oxygen is delivered by a proportional valve and mixed with air coming from the turbine. Again, the preset oxygen tension is constant regardless of the minute volume.

Some category 2 ventilators can also have a low-pressure oxygen inlet (e.g. VersaMed iVent 201 AB/IC,

Viasys Vela and Viasys Pulmonetic LTV 1200, ResMed Elisée 350, Dräger Carina and Savina, Taema Neftis ICU, Hamilton Medical C2).

In spite of the Fi,O2 that an internal oxygen sensor may read, oxygen delivery is not constant and Fi,O2 cannot reach 100%.

Although helium–oxygen breathing is known to reduce airway resistance and may therefore be indicated in conditions such as acute asthma, acute exacerbations of chronic obstructive respiratory disease or acute upper airway

09

obstruction, so far its clinical use is limited, in part because of the lack of adequate ventilators. As a further recent development, some commercially available ICU ventilators now incorporate the technology for helium–oxygen ventilation (e.g. Viasys Avea (Cardinal Health), Maquet Servo i, eVent Inspiration LS) [6, 14].

Internal battery Most category 1 PPVs have an internal battery, usually with a

Figure 12. Respironics Esprit category 2 ventilator

Figure 9. Taema Neftis ICU category 2 ventilator

Figure 13. Viasys Pulmonetic LTV 1200 category 2 ventilator (Cardinal Health)

Figure 15. ResMed Elisée 350 category 2 ventilator

Figure 16. VersaMed iVent 201 category 2 ventilator Figure 17. Dräger Carina category 2 ventilator

Figure 10. Dräger Savina category 2 ventilator

Figure 14.

Hamilton Medical C2 category 2 ventilator

Figure 11. Viasys Vela category 2 ventilator (Cardinal Health)

Figure 19.

Newport HT50 category 2 ventilator

Figure 18. Covidien Airox Supportair category 2 ventilator

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

99

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:13

Page 100

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

short life per charge. Interestingly, some category 1 PPVs have a “plug and play” battery system (e.g.

Maquet Servo i, Dräger Evita XL, Hamilton Medical G5) to increase battery life. Some category 1 ventilators (e.g. eVent Inspiration LS, Viasys Avea) have an internal battery that can power an internal compressor.

Figure 20. Medisize Zephyros category 1 ventilator

Some PPVs, especially those of category 2, have a very long-lasting (4–9 h) internal battery (e.g.

Covidien Airox Supportair, ResMed Elisée 350, Viasys Vela). Others may be equipped with a supplemental battery providing energy for up to 4–7 h (e.g. Dräger Savina, VersaMed iVent 201 IC, Hamilton Medical C2).

This can be very useful when transporting critical patients inside the hospital.

Recommendations to the buyer

Figure 21. Respironics Vision category 2 ventilator

1. The buying team must be aware of the differences in terms of gas input and type of control system when choosing a crtical care ventilator. In areas where compressed air is not available, a turbine-driven ventilator with additional oxygen (category 2) may replace compressed air.

2. In all category 2 ventilators that also have a high-pressure oxygen inlet preset oxygen tension remains constant regardless of the minute volume.

3. Some turbine-driven ventilators also have the option of a lowpressure oxygen inlet. When using this option, oxygen tension is never constant.

4. The charge-life of the internal battery as well as the possibility of using a supplemental battery must be taken into account in areas where mains power is frequently interrupted and where electrical back-up is inconsistent.

5. When choosing category 2 PPVs in areas where oxygen is provided only in tanks, ventilator oxygen consumption for a given minute ventilation

100

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

must be taken into account. In some markets, oxygen may be a real cost issue.

2. Inspiratory and expiratory valves and output variable control The inspiratory valve is meant to control respiratory cycle phases, along with the expiratory valve. In category 1 PPVs and some category 2 PPVs, the valve manages the output of the ventilator (set point, auto-set point, servo, adaptive and optimal control with controlled proportional valve). For instance, auto-set point control is called: volume-assured pressure support (VAPS) in the Respironics Esprit; AutoFlow in the Dräger Evita 2 Dura, 4 and XL and Savina; or Automode in the Maquet Servo i.

Servo control is called proportional assist ventilation (PAV+) in the Covidien PB 840; adaptive control is known as pressure-regulated volume control (PRVC) in the Maquet Servo i and Taema Extend, or pressure-controlled volume guaranteed in the GE Engstrom Carestation; optimal control is called Adaptive Support Ventilation (ASV) in the Hamilton Medical G5 [6, 10].

In many category 2 PPVs, the inspiratory valve has only an on-off function: pressure and flow both depend on the mechanical system (the piston or on the rotational speed of the turbine; set point control without a proportional valve) [6, 8].

In category 1 PPVs, the proportional valve (in currentgeneration models always a solenoid valve) can function as either a flow- or a pressurecontroller. With pressure or flow control, the delivered waveform is a function of both ventilator setting and respiratory system impedance. In current-generation category 1 ventilators and in some category 2 ventilators, three waveforms are available:

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 101

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

rectangular; descending ramp in volume-targeted mode; and exponential decay in pressuretargeted mode [6]. Although the clinical usefulness of inspiratory waveform manipulation is unclear, peak pressure is lower and gas distribution is improved with a decelerating flow pattern (descending ramp).

The expiratory valve can be a simple valve that is closed in counter-phase with the inspiratory one (a mushroom or diaphragm valve), or a proportional-aperture valve. Usually, expiratory valves also control the baseline phase, which can be atmospheric pressure or the positive end-expiratory pressure (PEEP). When baseline pressure is increased to create a threshold resistor (which, unlike a flow resistor, ideally maintains constant force regardless of gas flow) the efficiency of the exhalation valve functioning may decrease, prolonging the expiratory time constant. The use of a microprocessor-controlled expiratory valve or an electromagnetic valve operated by an actuating shaft (as in all Hamilton Medical ventilators, for instance) can modify this inefficiency by reducing pressure in the exhalation valve to atmospheric pressure early in the expiratory phase [6, 12].

The use of active expiratory valves makes it possible for the patient to breathe spontaneously during a mandatory pressure-controlled breath in controlled or assisted mandatory breathing.

Manufacturers refer to pressurelimited ventilation that allows patients to breathe spontaneously throughout the respiratory cycle by various names (e.g. assisted positive-release ventilation (APRV) as well as BIPAP in the Dräger Evita 2 dura, 4 and XL and in the BIPAP Savina; BiLevelTM in the Covidien PB 840 and GE Engstrom Carestation; BIVENT in the Maquet Servo i; BiPhasic in the Viasys Vela and Avea, etc.) [8].

09

Recommendations to the buyer 1. Control of output variables and valve functioning are very important when choosing a PPV, regardless of its pneumatic system.

2. The use of microprocessorcontrolled expiratory valves or electromagnetic valves operated by an actuating shaft can reduce exhalation system inefficiencies

3. Phase variables: trigger, cycling and limit Three phase variables define inspiration [6, 8]: • The trigger that begins inspiration (pressure-, volume-, flow- and time-dependent).

• The limit that cannot be exceeded during inspiration (pressure, volume, and flow).

• The cycling-off criteria.

Trigger variable During mechanical ventilation a breath can be initiated by [6, 10, 15–17]: • time (timed mandatory breaths set by the operator); • pressure (assisted breath pressure-triggered); • flow (assisted breath flowtriggered); • volume (assisted breath volumetriggered); or • diaphragmatic activity (Neurally Adjusted Ventilatory Assist (NAVA)).

The term “mandatory breath” defines a breath that is initiated and cycled by the ventilator.

During partial ventilatory assistance (assisted breath), the inspiratory synchronisation system (trigger) detects any patient inspiratory effort and activates a mechanical act. The goal of a good inspiratory trigger is to reduce

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

101

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

Page 102

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

as much as possible the duration and intensity of the muscular effort that comes before mechanical support, while avoiding autotrigger effects.

While pressure triggering allows detection of a pressure drop within the circuit (at the airway opening or inside the ventilator) due to the patient’s inspiratory effort, flow triggering is achieved with the measurement of flow by using a pneumotachograph at the airway opening (e.g. Hamilton Medical ventilators, Viasys Pulmonetic LTV 1200) or inside the ventilator (most category 1 and category 2 ventilators). Some flow triggers work with the “flow-by system”, which provides a continuous bias flow into the circuit with triggering occurring when the difference between the flows entering and exiting the respiratory double-limb circuits equals the trigger sensitivity [6].

This bias flow may be automatically adjusted when increasing flow trigger sensitivity (e.g. Covidien PB 840) or else manually adjusted (e.g.

Viasys Avea).

New trigger algorithms also aim at improving patient–ventilator interaction during sudden changes in flow or respiratory rate or in the presence of air leaks during noninvasive ventilation (NIV). This can be achieved with volume triggers, triggers linked to flow waveform algorithms (e.g.

Respironics Vision), combining pressure and flow signals in the same trigger algorithm (e.g. Dräger Evita 2 dura, 4 and XL) or using both pressure and flow triggers (e.g.

Smart trigger Viasys Avea). Table 3 shows the types of available triggers.

Unfortunately, reducing autotriggering often implies a lower trigger sensitivity. This is particularly true when ventilating uncuffed airways. However, it has been generally suggested that a trigger (independently of its algorithm) must have a response time <100 ms.

102

23:13

Table 3 Types of inspiratory trigger

Pressure trigger where a given negative pressure is set by the operator (e.g. -1 cmH2O) Genuine flow trigger where a given flow value is set by the operator (e.g. 1 L·min-1) Flow trigger with predetermined flow-by system, where a given flow value is set by the operator# Flow trigger with adjustable flow-by system, where a given flow value for the trigger and a given flow-by value for the system is set by the operator¶ Combined genuine flow trigger and predetermined pressure trigger where the operator sets only a flow value+ Combined flow-by and pressure trigger where the operator sets pressure and a flow value+ Trigger working with sequential algorithm linked to flow§ or turbine speed variation. The operator has nothing to set Volume triggerƒ Neural trigger (NAVA)## #: In this case, flow-by increases with decreasing flow sensitivity (e.g. flow-by increases 1.5 L·min-1 for each increment in flow sensitivity (e.g. Covidien PB 840); ¶: In this case patient effort is detected first by one of the two triggers. There is no ranking logic (e.g. Viasys Avea); +: In this case the pressure trigger acts only as a safety mechanism to avoid auto-triggering (e.g. Dräger Evita 2 dura, 4 and XL and Savina); §: AutoTrak system by Respironics; ƒ: Usually adopted with a default volume value (e.g. 20 mL) in some ventilators to avoid auto-triggering; ##: Only in Maquet Servo i.

All inspiratory trigger drawbacks may be overcome by using a neural trigger obtained by means of a dedicated nasogastric tube with a multiple array of electrodes placed in the distal oesophageal portion (NAVA; e.g. Maquet Servo i) [17, 18].

Cycle variable A breath can be pressure-, time-, volume- or flow-cycled [8]. A breath in a flow-cycled ventilation mode (PSV, inspiratory positive airway pressure (IPAP) or assisted spontaneous breathing (ASB) for Dräger ventilators) is cycled when inspiratory flow reaches a given threshold value (default at 25% of peak flow in most PPVs, but adjustable in others). The inspiratory flow threshold value, also called “expiratory trigger”, thus controls the inspiration-toexpiration switch in these

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

modalities [19]. The aim is to detect the very end of patient inspiration through inspiratory flow measurement. Its goal is to optimise synchronisation between spontaneous patient inspiratory time and ventilator inspiratory time. In a flow-cycled ventilatory modality there is often a ranking logic of expiratory cycling criteria that allow cycling to expiration.

Table 4 shows criteria for PSV/IPAP cycling to baseline pressure (atmospheric pressure or PEEP/expiratory positive airway pressure (EPAP)).

Dyssynchrony at the onset and termination of a PSV breath can be corrected by varying the inspiratory rate of pressurisation and/or the off-cycling criteria (e.g. modulating the expiratory trigger threshold ) [6, 20–24]. All category 1 ventilators and most category 2 ventilators have this feature.

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 103

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

Table 4 Criteria for PSV/IPAP/ASB to cycle to baseline pressure

Flow reaches a predetermined percentage of inspiratory peak flow (usually 25%) Flow reaches a preset percentage of inspiratory peak flow (e.g. from 1% to 80%) According to particular algorithms linked to flow value or wave form (e.g. Respironics AutoTraK system) According to a given preset or adjustable inspiratory time (safety cycling) According to a given default pressure (2–3 cmH2O) above the peak inspiratory pressure (safety cycling) According to a given default flow value after a given time (usually 1–3 s; safety cycling) PSV: pressure support ventilation; IPAP: inspiratory positive airway pressure; ASB: assisted spontaneous breathing.

Limit variable Mechanical ventilators apply positive pressure at the airways with different modalities. Although these are increasingly sophisticated, they are usually divided into volume- or pressure-limited ventilatory supports. With the former, the independent variable (ventilatory output) is flow and its integral over time, namely volume. The dependent variable is pressure, which varies according to the respiratory system impedance. With the latter, the independent variable is pressure while flow and volume (dependent variables) are consequent to ventilatory impedance [6, 8].

In some category 2 ventilators (the so-called bilevel ventilators), the set-up governing the peak pressure applied to the respiratory system during pressure-limited flow or time-cycling ventilatory modes (usually called spontaneous/timed (ST) ventilation with inspiratory pressure, labelled as IPAP, and PEEP labelled as EPAP) may differ from category 1 ventilators, where the same ventilatory modes are called PSV/PEEP (spontaneous) or AC/PCV/PEEP (timed).

To give an example, when using PSV 10 cmH2O plus PEEP 5 cmH2O

in a category 1 ventilator, the actual pressure applied to the patient’s airway above PEEP is 10 cmH2O, while applied peak pressure is 15 cmH2O (10 + 5: the so-called “above PEEP” setting). When using IPAP 10 cmH2O plus EPAP 5 cmH2O using a bilevel ventilator (e.g. Respironics Vision, Covidien Supportair) the actual pressure delivered to the patient above EPAP is 5 cmH2O, so the applied peak pressure remains at 10 cmH2O (the so-called “below PEEP” setting). However, most bilevel ventilators have a reminder on the main screen of the peak pressure applied to the airways.

This fact has generated a lot of confusion in the past. To add more confusion, a modified algorithm of assisted positive pressure ventilation was named Bilevel Inspiratory Positive Airway Pressure (BIPAP). This mode has often been mistaken for BiPAPTM by Respironics, particularly because BIPAP, like all APRV algorithms, uses the below PEEP setting.

Recommendations to the buyer 1. All commercial PPVs, except ones dedicated mainly to NIV (e.g.

Respironics Vision) offer both pressure- and volume-targeted ventilation.

09

2. In spite of clear evidence that a flow trigger performs better than a well-set pressure trigger [6] the sensitivity of pressure or flow trigger setting is mandatory to avoid patient–ventilator dyssynchrony.

3. When combining different kinds of inspiratory trigger, manufacturers should specify their ranking logic clearly. One ventilator (Maquet Servo i) offers the option of using a neural trigger through NAVA. 4. Almost all category 1 and category 2 PPVs enable the user to set the sensitivity of the expiratory trigger threshold (expiratory trigger). The wider the range (e.g. 1–90% of peak expiratory flow), the higher the chance of matching the patient’s neural inspiratory time.

5. When using turbine-driven or piston ventilators, the buyer must be aware of the set-up governing peak inspiratory pressure above PEEP/EPAP during pressurelimited ventilatory modes.

4. Modes of mechanical ventilation It is beyond the aim of the present guide to explain all modes of ventilation. The authors will try to inform the reader of those ventilatory modalities that go beyond the difference between pressure- and volume-limited ventilatory support (automatic tube compensation (ATC), PAV, proportional pressure support (PPS) and NAVA, where neither pressure nor volume must be set by the operator). We will also deal with those modes of ventilation that switch from one limit variable to another one (e.g. from pressure- to volume-limited ventilatory support, so-called dual modes) and ASV [6, 8].

PAV, PAV+ and PPS Proportional ventilation was first described in 1992 [25]. PAV can be viewed as a ventilation mode in

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

103

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

Page 104

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

which the pressure within each breath is titrated by the ventilator in proportion to patient inspiratory flow, which is used as an estimate of patient respiratory muscle effort .

The proportionality between flow and airway pressure is determined by a “gain” setting, which is adjusted by the clinician to determine the proportion of the total work of breathing to be performed by both ventilator and patient. In contrast with previous modes of ventilation, in PAV, the clinician sets the gain (proportion of work of breathing, expressed as a percentage) based on patient respiratory mechanics. In other words, the resistance (Rrs) and compliance (Crs) of the respiratory system are evaluated and the percentage of assistance adjusted accordingly. However, although the difficulties associated with assessing Rrs and Crs in actively breathing patients are being overcome, (e.g. by the Covidien PB 840 in PAV+ mode), PAV may become unstable in certain situations, or if the gain is set improperly. This drawback of PAV has been termed “runaway”, and results when algorithm-induced changes affect an input variable, thereby creating an undesirable change to the output, in a cyclical fashion (increasing pressure results in increased flow, which may increase measured Rrs, resulting in further increases in pressure) [26–29].

NAVA NAVA is a new form of assisted ventilation that takes into account most principles of proportional ventilation. Pressure is applied by the ventilator in proportion to the electrical activity of the diaphragm, recorded with a dedicated nasogastric tube with a multiple array of electrodes placed in the distal oesophageal portion [17].

NAVA is now commercially available on the Maquet Servo i.

ATC In order to reduce patient work of breathing arising from endotracheal tube resistance, some ventilators

104

23:13

(e.g. Dräger Evita 2 dura, 4 and XL, Covidien PB 840, Viasys Avea, GE Engstrom Carestation and all Hamilton Medical ventilators) keep tracheal pressure constant during inspiration and expiration. The output delivered by the ventilator adapts automatically and immediately to the inspiratory effort of the patient. This control is named ATC. Recent studies have shown that, in contrast to PSV without ATC, the latter is a suitable mode to compensate for the additional work of breathing due to the endotracheal tube in patients with increased ventilatory requirements. They found that when the additional work of breathing due to the endotracheal tube was compensated, no additional level of PSV was required [30, 31].

Dual modes To generate further confusion among buyers, the so-called “dual modes of ventilation” (e.g. PRVC or VAPS) have recently been introduced in the critical care area.

The buyer must be aware that even if the target of the dual mode is a given tidal volume, this volume can be reached by increasing flow or by increasing pressure to the airway to reach the preset volume.

Another difference is that gas delivery may be adjusted either within each breath (e.g. VAPS in the Respironics Esprit) or breath by breath (e.g. PRVC in the Maquet Servo i; Auto Flow in the Dräger Evita 2 dura, 4 and XL and Savina and the Taema Extendl; VV+ in the Covidien PB 840; PCV-VG and BiLevel-VG in the GE Engstrom Carestation, etc.) [6].

ASV ASV may be thought of as an “electronic ventilator protocol” (defined as optimal control) [8] that incorporates the most recent and sophisticated measurement tools and algorithms in an attempt to

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

make ventilation safer, easier, and more consistent [32, 33]. This mode is designed to accommodate not only ventilated patients who are passive, but also those who are breathing actively. ASV recognises spontaneous respiratory activity and automatically switches the patient between mandatory pressure-controlled breaths and spontaneous pressure-supported breaths. With ASV, the clinician determines the desired minute ventilation, and the algorithm determines the optimal respiratoryrate/tidal-volume combination according to the patient’s respiratory mechanics. Any change in respiratory mechanics or patient effort results in an updated optimal breathing pattern, and ASV continuously and gently moves the patient to the new, updated, target.

Intelligent breath-to-breath safety rules maintain ventilation parameters within safety ranges, and if for any reason the patient fails to breathe actively, ASV automatically increases the number of mandatory pressure-controlled breaths that are needed in order to maintain the minute volume target.

The intrinsic requirement for determination of the optimal breathing pattern is the breath-tobreath measurement of respiratory mechanics, including the expiratory time constant, based on the volume–flow loop method [34].

Recommendations to the buyer 1. The buyer should determine which ventilatory modes will be used regularly in order not to purchase sophisticated ventilatory mode options that will seldom be needed. This should result in cost savings.

2. The buyer must be aware of the limitations of dual modes of ventilation such as PRVC/AutoFlow, etc.: when a patient’s ventilatory demand shows an increasing tidal volume, once the target tidal volume is achieved the patient is no longer supported.

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:13

Page 106

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

3. Buyers must be aware that only PAV+ (Covidien PB 840) offers the possibility of carrying out proportional ventilation using a noninvasive measurement of elastance and resistance. Only one ventilator has NAVA. (Maquet Servo i).

4. ASV is provided only by Hamilton Medical. In spite of some limitations, such as the presence of severe leaks or fast changes in respiratory drive causing ventilator algorithm malfunctioning, it is the current authors’ opinion that ASV may have an important role where the number of operators is limited due to cost restraints (e.g. in emerging countries).

5. Evidence is lacking that any ventilation mode provides better outcomes, though there are different potential advantages suggested by surrogate physiological variables [1].

monitoring and safety. In the adaptive modes (e.g. ASV), the servo valve is driven by the proximal flow sensor while the internal flow sensor is used to check values given by the proximal flow sensor (plausibility).

In ventilation modes dedicated to pre-term babies, some category 1 PPVs switch the control of output variables and their feedback system to the proximal airway sensor (e.g.

Dräger Evita 2 dura, 4 and XL, Maquet Servo i, GE Engstrom Carestation). The Viasys Avea allows the user to choose between two types of proximal pneumotachograph (sensor): “hot wire” and “Osborne’s variable orifice”.

Recommendation to the buyer 1. The buyer must be aware of the different places where the feedback signal is recorded. The advantages and disadvantages of different locations of output sensor are discussed in [6].

5. Control system The control system controls or servo-controls output variables and activates safety systems. It can work with either a pressure or a flow feedback. All category 1 and 2 ventilators use a closed-loop control that maintains a constant ventilatory output by using output as a feedback signal that is compared with the operator-set input [8].

Most manufacturers use flow inside the ventilator as a feedback signal for safety reasons. In those ventilators where the output variables are monitored at the airway opening (proximal airway sensor), the feedback control is switched to the proximal transducer (e.g. Hamilton Medical ventilators) depending on ventilatory mode. On the Hamilton Galileo and G5, during controlled modes, the servo valve is driven by an internal flow sensor while the external flow sensor is used for

106

6. User interface The control panel allows the user to interact with the ventilator in order to set ventilator parameters and verify through monitoring that they are correctly applied. Some parameters are set directly whereas other are derived from measurements.

ventilators with their built-in user interface (ResMed Elisée 350, Viasys Pulmonetic LTV 1200).

Recommendations to the buyer 1. User interfaces vary significantly among PPVs. Unfortunately there is no consistency among manufacturers in terms of labelling ventilator functions, particularly ventilation modes [6].

2. The large number of ventilator modes, settings and monitoring can result in systems that are poorly designed, nonintuitive and almost impossible to remember, with the subsequent need for clinicians to refer to manual ventilator instructions in order to interpret poorly labelled controls or to navigate multiple levels of software [3]. As a consequence, the buyer must be aware that there will often be a learning curve to get a “feel” for full navigation of the interface.

3. The buyer must choose the ventilator according to current need (number of beds, emergency room area, different operators rotating in the shifts, etc.).

4. Last but not least, another practical consideration is whether the monitor displaying the settings, curves, alarms, etc., is easily readable.

7. Safety and alarm systems

With the current generation of PPVs, the user interface is commonly a touch pad and/or rotary encoder with or without a touch-screen control. In most PPVs, a two-step process of changing and then accepting a given parameter is used [6]. Some ventilators use an intelligent interface to decrease the demand on the user's cognitive resources (e.g. "dynamic lungs" Hamilton Medical G5).

The ventilator safety system aims at avoiding any damage to the patient due to ventilator malfunction.

Only two category 2 PPVs can be removed from their conventional user interface (as docking stations of a PC) to be used as transport

Another safety system is the presence of an overpressure valve, positioned between the inspiratory and expiratory valves, which can

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

In case of electrical failure, there is a room-air inlet that will let the patient breathe through a simultaneous opening of the inspiratory and expiratory valves.

This may be of little or no help, however, in patients who are sedated and paralysed.

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 107

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

unload any excess pressure in the circuit. In most ventilators, it is usually set to open above a 100 cmH2O threshold and thus it does not limit the risk of barotrauma.

On the contrary, the use of microprocessor-controlled functions, with automated alarm responses for events such as retrograde ventilation in case of apnoea (back-up ventilation) during assisted ventilation or pressuretargeted modes, or expiratory valve opening when a maximum tolerable pressure is reached, is more effective in reducing mechanical ventilation-related risks. Most ventilators can be set to start backup ventilation in all ventilatory modes in case of ventilator malfunction.

Some ventilators (e.g. Dräger 2 dura, 4 and XL, GE Engstrom Carestation, Covidien PB 840, Viasys Avea) can differentiate between a leak and a real disconnection, allowing the ventilator to stop the high continuous flow produced in case of an actual circuit disconnection (e.g. during patient suctioning). This may avoid environmental pollution.

The alarm system must alert medical and paramedical personnel of an event that requires their intervention [37]. These events can be technical ones (e.g. related to ventilator performance) or clinical ones, due to a change in patient conditions.

All ventilators in a critical care area should have alarms for: • electrical failure; • compressed gas supply problems; • patient disconnection from the ventilator; • changes in airway pressure (in volume-targeted mode); • changes in tidal volume (in pressure-targeted mode); • changes in minute ventilation (in volume- or pressure-targeted modes); and

• changes in Fi,O2.

There are default values for alarm thresholds, beyond which an audible alarm and/or a visual indicator will go off – but these values should be set case by case according to clinical situation, ventilation mode and kind of interface, and possibly modified whenever these conditions change.

Recommendations to the buyer 1. All PPVs for critical care except for those designed only for NIV (Medisize Zephyros) and the Respironics Vision (this ventilator does not feature volume-targeted ventilation) are life-support devices. It is imperative that they respond in a safe manner in case of ventilator malfunction (pneumatic or electrical failure).

2. In spite of the safety system now provided in all PPVs, the buyer must be aware of the presence in the hospital of electrical power generators in case of main power blackout. Users must be informed of the units attached to the emergency generator.

8. Monitoring system The monitoring system is not a part of the ventilator itself and its absence will not jeopardise proper ventilator functioning. However, it is of the utmost importance in optimising ventilatory assistance.

Some turbine-driven ventilators (e.g. ResMed Elisée 350, Viasys Pulmonetic LTV 1200) can be connected to a full-size screen to provide additional monitoring.

Thus, even though clinical and instrumental monitoring based on patient vital signs is very relevant, frequent control of both preset parameters on the ventilator and their actual application through machine-integrated systems is crucial in the ICU setting.

In any ventilator, directly measurable variables are the

09

pressure applied to the airway and the flow, while other parameters can be derived from the analysis of these signals. Measured variables and derived parameters are shown both on a graphical display and as numerical data in all ICU PPVs. Continuous display of airway pressure, flow, and/or volume curves is available with all current ICU ventilators.

Volume is usually calculated through the integration of flow. The flow sensors used in current PPVs are all reliable, but only those in Hamilton Medical ventilators measure flow at the proximal airways in adult patients. Most ventilators measure flow inside the ventilator, so the total calculated volume will include not only the delivered volume, but also the volume that is compressed inside the ventilator circuit. To solve this problem, some ventilators ask for an initial operation to calculate tube compliance (see section 9; estimated at about 2–3 mL·cmH2O-1 in adult circuits). However, the accuracy of the volume displayed varies considerably from that actually measured at the ventilatory circuit “Y-piece”. The discrepancy increases with increased tidal volume [6]. Direct visualisation of flow–time and pressure–time waveforms provides valuable information about the quality of ventilator–patient interaction (e.g. ineffective respiratory efforts, ventilator auto-triggering, dyssynchrony between a spontaneously breathing patient and the ventilator, dyssynchrony due to air leaks in PSV mode or to avoid inadequate inspiration or expiration times during pressurecontrolled ventilation) [36].

In totally mandatory breaths, most PPVs are able to noninvasively measure respiratory system mechanics thanks to functions that regulate inspiratory and expiratory valve closing [37].

Furthermore, most machines allow further display of pressure–volume

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

107

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:13

Page 108

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

or flow–volume loops with the possibility of performing a lowinflation pressure–volume curve (e.g. Taema Extend, GE Engstrom Carestation, Viasys Avea).

Interestingly, few recent ventilator types incorporate the technology to measure an additional pressure, such as tracheal pressure or oesophageal pressure. The Viasys Avea and Hamilton Medical Galileo and G5 do offer this facility [38].

Although there is still a lack of adequate easy-to-use technologies, this type of monitoring may provide useful information on the respiratory mechanics of spontaneously breathing patients at the bedside [39, 40].

Recommendations to the buyer 1. The buyer should determine which monitoring capabilities will be used regularly in order not to purchase sophisticated monitoring systems that will seldom be needed. This should result in cost savings [3].

2. In adult patients, there may be no significant differences between the two sites of measurement: distal (inside the ventilator) or proximal (at the airway opening), except for a better detection of start of inspiration and end of

inspiratory effort. On the other hand, during ventilation in preterms babies where patient flow is very low, a proximal transducer is strongly advised.

3. Regarding the monitoring of respiratory mechanics, pressure–volume curves (lowinflation technique) are still a point of discussion with regard to strategies for setting mechanical ventilation in patients with acute respiratory distress syndrome. However, it should be kept in mind that interpreting the information provided by these curves and drawing therapeutic conclusions may require some experience or at least clear strategies to deal with the results.

4. Finally, it must be borne in mind that the sensor accuracy of many ventilators lacks proper validation, and thus measured data can differ (even substantially) from real values [41].

9. Respiratory circuit and accessories The circuit is not a true part of the PPV and it is often sold separately.

In spite of being a device positioned externally to the PPV, its features may change among different ventilators.

Double-limb respiratory circuit The double-limb circuit is composed of an inspiratory and an expiratory limb whose proximal ends are connected to the inspiratory and expiratory ports, while the distal parts are connected to the so-called Y-piece. The Y-piece can be the connected to the catheter mount with or without a proximal sensor (fig. 22).

In some PPVs (e.g. Hamilton Medical ventilators) the pneumotachograph is positioned at the Y-piece (sensor at the proximal airways; fig. 22). Flow and pressure measured at this site can be used either as simple monitoring tools or to control some ventilator functions (e.g. Hamilton Medical ventilators, Dräger Evita 2 dura, 4 and XL, Maquet Servo i, GE Engstrom Carestation and Viasys Avea when using pre-term ventilation).

The effective compliance of the patient circuit is a combination of tubing compliance and gas compressibility [10]. Most highpressure ICU PPVs provide automatic compensation of circuit compliance according to their circuit place (adult versus paediatric). However some ventilators still lack this automatic compensation (e.g. Dräger Savina, Covidien Supportair)

Figure 23. Dräger Evita 4/XL with its sensor positioned at the airway opening during NeoFlow ventilation.

Figure 22. (left to right). Hamilton Medical G5, Raphael and C2. All use a sensor positioned at the airway opening at the Ypiece before the cathether mount (red arrow).

108

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:13

Page 111

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

09

Table 3 Modalities of CO2 venting in intentional leak circuits

Venting system build-up in the interface (vent slots or holes) or in the circuit proximally to the airway Venting connector (e.g. Whisper swivel by Respironics) Plateau Valve# # : This valve is a reusable exhalation device that provides a continuous leak path in the patient circuit when used with Respironics continuous positive airway pressure (CPAP) and bilevel systems. Although it is not an active valve it does not allow CO2 rebreathing [45].

When the flow sensor is positioned proximal to the airways (e.g.

Hamilton Medical ventilators, Dräger Evita 2 dura 4 and XL, Maquet Servo i and Viasys Avea when using pre-term ventilation) there is no need to calculate compliance (fig. 23).

Single-limb circuit

ventilator pressure such as in the Covidien Supportair). This valve is positioned proximally to the patient close to the catheter mount (fig. 24).

The function of this valve is often an on–off function, which may also work as a PEEP valve. Although they always avoid rebreathing, this kind of valve may prolong the expiratory time constant [42]. The dead space between the proximal valve and patient interface (e.g. the mask) may affect total dead space.

The single-limb circuit is found only in category 2 PPVs, except for some circuits that may have their inspiratory and expiratory limbs embedded coaxially. Single-limb circuits have no Y-piece and the circuit is directly attached to the catheter mount. This circuit is usually limited to anaesthesia use.

Apart from coaxial circuits, three types of single-limb circuit can be defined (figs 24 and 25).

ii. Single circuit without an active nonrebreathing valve. This type of circuit is found only on five PPVs (Respironics Vision, Dräger Carina, Covidien Supportair and Raphael XTC and C2). It is also called an intentional leak circuit (fig. 25). CO2 is vented out through three modalities, as shown in table 3.

i. Single circuit with an active nonrebreathing expiratory valve (e.g. a mushroom valve driven by

In the first two types of intentional leak circuit in table 3, the amount of rebreathing always increases when

Figure 25. A single-limb intentional-leak circuit. The intentional leak is provided through a connector with “vent slots” (Whisper) or a “Plateau Valve” positioned in series in the circuit close to the interface (a and d), through “vent holes” in the mask (c) or through a vent hole positioned in the circuit (b). The red arrows indicate where air flows through the leak.

Table 4 Causes of CO2 rebreathing in an intenional leak circuit

Patient’s expiratory flow# Ventilator expiratory flow output Level of PEEP/EPAP Position of vents¶ in the circuit as well as in the mask and mask dead space PEEP: positive end-expiratory pressure; EPAP: expiratory positive airway pressure. #: In turbine-driven ventilators featuring a proportional valve to control ventilatory output (e.g. Respironics Vision) patient expiratory flow does not affect rebreathing; ¶ : Manufacturers must always specify “vent system” features.

patient expiratory flow exceeds clearance of the vent system. Table 4 shows the causes of CO2 rebreathing in an intentional leak circuit.

It was previously demonstrated that an EPAP of 8 cmH2O was able to avoid CO2 rebreathing completely [43–46]. The use of a Respironics Plateau Valve nonrebreathing valve may completely avoid CO2 rebreathing in the intentional leak circuit of a dedicated Respironics ventilator (e.g. Respironics Vision) [43].

Recommendations to the buyer Figure 24. A single-limb circuit with a mushroom expiratory valve (red arrows). The valve is driven by the pressure coming from the pipe (green arrow).

1. Coaxial circuits should not be used in a critical care setting

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

111

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

Page 112

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

because of their increased respiratory resistance due to decreased expiratory limb size.

2. The lack of a Y-piece (e.g. coaxial circuit, intentional leak circuit) does not allow a spacer to be positioned on the distal part of the expiratory limb to deliver drugs using a metered dose inhaler (MDI). For the same reason, conventional double-limb circuits should always be detachable at the Y-piece in order to install an in-line spacer for MDI use or use of an active nebuliser [47].

2. Buyers must be aware of respiratory circuit compliance during mechanical ventilation. In adult patients, all category 1 ventilators ask to measure circuit compliance before beginning mechanical ventilation by measuring flow and pressure.

Hamilton Medical ventilators measure pressure at the airway opening and do not need to measure respiratory circuit compliance.

3. While in adult patients there may be no significant difference between using two types of circuit with different compliances (e.g. 2 mL·cmH2O-1), in patients with low tidal volume (e.g. 50 mL; e.g. paediatric or neonates), circuit compliance plays a major role (patient compliance and circuit compliance are connected in parallel). Ventilators featuring pre-term ventilation (e.g. Dräger Evita 2 Dura, 4 and XL, Maquet Servo i, GE Engstrom Carestation, Viasys Avea, Hamilton Medical Galileo and G5) use a sensor at the airway opening.

4. When automatic compliance compensation is not present, it must be calculated as described elsewhere [8].

5. When using a ventilator with an intentional leak circuit (not provided with an active expiratory valve) an antisuffocation valve is strongly

112

23:14

advised in situations where the patient may be left without continuous nursing care (e.g. a patient switched from critical care to the general ward).

Accessories High-pressure hoses and connecting hardware Hoses may have different quick connections to interface with centralised high-pressure gas or gas cylinders. This may also hold true for electrical power connections.

Respiratory circuit See section 9 above.

Heated humidifiers Heated humidifiers are often sold separately. It is beyond the scope of the present guide to discuss the utility of active or passive humidification in critical care.

Carts, brackets and carriers Carts may be a simple base mounted on wheels or a complex storage facility. Brackets are provided in order to mount the PPVs at the patient’s bedside or on a pneumatic column. Carriers allow gas tank and external battery connection, facilitating patient transport.

increase in resistance over time.

Filters can be disposable or reusable.

Nebulisers Some ventilators are capable of powering their own nebuliser without affecting the delivered Fi,O2. This increases their capability to deliver a large volume of aerosol in the therapeutic range (e.g.

Maquet Servo i, eVent Inspiration LS, Dräger Evita 2 dura, 4 and XL and Savina) [6].

Recommendations to the buyer 1. High-pressure hoses and connecting hardware must be compatible with the local gas and electrical plugs.

2. According to team needs, carts, brackets or carriers to facilitate patient transport should be purchased along with the ventilator.

3. In spite of the fact that many ventilators now feature their own nebuliser, drug aerosols within the expiratory circuit may cause expiratory sensor malfunction.

Expiratory limb filters may be used to prevent sensor damage but may have a greater impact on expiratory resistances and on operating costs [6].

Special needs External batteries and external compressors See section 1 above.

Pre-term ventilation See section 5 above.

Filters Heliox PPVs may be equipped with inspiratory as well as expiratory filters. The former may protect patients from particulates in the piped or compressed gas supply, while the latter may protect downstream sensors and components from particulates coming from aerosolised drugs. The filter may have an impact on flow, by adding resistance. Filters also

See section 1 above (gas blending).

Lung-sigh, recruitment and protective lung software SmartCare is a mode of ventilation specifically designed to expedite the weaning process. It utilises only pressure-support breaths, with varying levels of inspiratory

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

HR adv 105+

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:14

Page 115

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

pressure. The principle [48] uses the measured respiratory rate to adjust the level of PSV (increasing PSV if respiratory rate is high, decreasing PSV if respiratory rate is low).

SmartCare analyses ventilation every 2 min: more often than any therapist could. The objective of SmartCare is to obtain a target level of pressure support that is tolerated well and which eventually indicates the patient’s readiness to be discontinued from the ventilator.

The level of pressure support is also adjusted to maintain tidal volume above, and end-tidal carbon dioxide pressure below, certain values. The switch from controlled ventilation to SmartCare is not automatic, and requires manual intervention. This software option is now available on the Dräger Evita XL. The sigh breath is a deliberately increased tidal volume for one or more breaths at regular intervals [49]. The lack of a real evidence base about sighs means that few current-generation PPVs feature this function (e.g. Dräger Evita 2 dura, 4 and XL, Taema Extend).

Only one ventilator allows real manipulation of the sigh function: the eVent Inspiration LS equipped with the Smart Sigh function. Users may define volume- or pressurebased sighs, their frequency and multiples as well as their amplitude (+0–50% of tidal volume or pressure setting). Three recent clinical studies, however, have demonstrated improved oxygenation with the use of sighs as a recruitment manoeuvre in patients with acute respiratory failure [50–52]. Some ventilators have dedicated software for “open lung” and “lung protective strategy” (e.g. Lung Protection Package (LPP) for Dräger Evita XL and Open Lung Tool (OLT) for Maquet Servo i).

Recommendations to the buyer 1. In spite of the lack of a real evidence base about sighs, there is now an increased interest in the

sigh option as well as dedicated software for lung recruitment and open lung techniques.

2. SmartCare in EvitaXL (as a software option) is a mode of ventilation designed to expedite weaning. Patient readiness and ability to wean also need to be assessed by the clinician.

3. As previously mentioned, evidence is lacking that any ventilation mode or ventilator software package can provide better outcomes [1].

Dedicated NIV ventilators and ventilator software for NIV Modes of ventilation The results of a recent bench-model study confirm that leaks interfere with several key functions of ICU ventilators [53]. On most category 1 ventilators, conventional software for invasive ventilation leaks led to an increase in trigger delay and workload, a decrease in pressurisation and delayed cycling [53]. Air leaks during NIV are an obstacle to correct expiratory synchronisation. In fact, any additional flow due to leakage will delay or even prevent reaching the threshold. When this happens, the expiratory trigger is unable to synchronise its action with the end of patient inspiratory effort. On some ventilators, as mentioned previously, the threshold of the expiratory trigger can be manipulated, thus allowing improved expiratory synchronisation [19–23]. The ability to set a maximum inspiratory time beyond which there will always be a mandatory cycling, or else particular algorithms, can also greatly reduce problems related to expiratory I/E cycling during PSV [24]. Most manufacturers have now developed microprocessorcontrolled ventilators with different algorithms that provide both volume- and pressure-limited

09

modes with special features designed to facilitate NIV delivery.

When using noninvasive continuous positive airway pressure (CPAP), most ventilators impose only a small effort to trigger, but most also provide low-level pressure support and impose an expiratory workload [54, 55].

Pressure triggering during CPAP does require a slightly greater pressure than flow triggering [56] although flow triggering gave better results in NIV-–PSV mode in chronic obstructive pulmonary disease patients. However, even the “best” triggering system may be markedly affected by the interface used (e.g. helmet or other largevolume interface).

Category 2 ventilators with an intentional leak circuit and bilevel ventilation (e.g. BiPAP by Respironics Vision) have been tested in many clinical and bench studies. BiPAP (IPAP/EPAP) is a pressure-limited flow or timecycled mode of ventilation. When the patient is active, BiPAP is, in simple terms, a PSV working in the below PEEP setting. It has a very efficient inspiratory/expiratory trigger algorithm called AutoTrakTM. Many so called other bilevel ventilators (e.g. Covidien Supportair) use the below PEEP setting.

The Respironics Vision also features NIV PAV [57–58].

Circuits Both single- or double-limb circuits are provided in NIV ventilators.

Single-limb circuits without an active nonrebreathing valve are found on only three category 2 PPVs (Respironics Vision, Dräger Carina, Covidien Supportair). The Dräger Carina and Covidien Supportair also have the option to use a single-limb circuit with an active mushroom valve (figs 17 and 24). One category 1 ventilator manufactured only for NIV (Medisize Zephyros) has a

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

115

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:14

Page 116

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

double-limb circuit, like all other category 1 ventilators Alarms During NIV it is also advisable to have alarms for significant air leaks.

These are characterised by a reduction in airway pressure and expiratory tidal volume when in volume-limited mode, and a rising inspiratory tidal volume and decreasing expiratory tidal volume in pressure-limited mode.

Monitoring When using NIV, inspiratory tidal volume measurements can be confounding, because the total volume provided by the ventilator equals the sum of volume actually sent to the patient plus leak volume. Thus expiratory tidal volume is more representative of alveolar ventilation than is inspiratory tidal volume. With ventilators that measure inspiratory and expiratory tidal volumes separately, leak magnitude can be estimated by calculating their difference. This is

true for ventilators that measure both inspiratory and expiratory flow inside the ventilators (e.g.

Medisize Zephyros, Respironics Esprit and all category 1 ventilators) and ventilators that measure flow proximally to the airway (e.g. Viasys Pulmonetic LTC 1200, VersaMed iVent 201 IC and all Hamilton Medical ventilators).

Ventilators designed solely for NIV Only one ventilator (the Medisize Zephyros) is designed solely for NIV. The Respironics Vision and Dräger Carina, although designed mainly for NIV, can be also used in the critical care area for invasive ventilation. The Medisize Zephyros can deliver NIV with the above PEEP setting (PSV) The ventilator has software to set up NIV with different interfaces (e.g. mask or helmet).

Recommendations to the buyer 1. Overall, NIV modes can partially or completely correct drawbacks due to air leaks, but there are

wide variations between machines in terms of efficiency [53]. Clinicians should be aware of these differences when applying NIV with an ICU ventilator.

2. NIV CPAP delivered by ventilators is reliable. However, when using only genuine CPAP, continuous-flow systems have very good performance and are cheaper [52]. CPAP should not be used in ventilators with a double circuit when using a helmet interface [59–61].

Conclusions Purchasing a PPV should be based on a basic knowledge of machinespecific functions and an understanding of the physiological rationale. It is therefore crucial to understand how these functions should interact with the operative environment and patients’ needs.

Once this goal is achieved, a ventilator may be chosen based on specific needs, constraints and costs. ■

REFERENCES 1. Branson RD, Johannigman JA. What is the evidence base for the newer ventilation modes? Respir Care 2004; 49: 742–760.

2. Greenberg D, Pliskin JS, Peterburg Y. Decision making in acquiring medical technologies in Israeli medical centers: a preliminary study.

Int J Technol Assess Health Care 2003; 19: 194–201.

3. Fink JB. Purchasing a ventilator. In: Tobin MJ, ed. Principles and Practice of Mechanical Ventilation.

2nd Edn. New York, McGraw-Hill, 2006; pp. 1365–1370.

4. Chatburn RL, Primiano FP Jr. Decision analysis for large capital purchases: how to buy a ventilator.

Respir Care 2001; 46: 1038–1053.

5. Grace K. The ventilator: selection of mechanical ventilators. Crit Care Clin 1998; 14: 563–580.

6. Kacmarek RM, Chipman D. Basic principles of ventilator machinery.

116

In: Tobin MJ, ed. Principles and Practice of Mechanical Ventilation.

2nd Edn. New York, McGraw-Hill, 2006; pp. 53–95.

7. Fink JB. Device and equipment evaluations. Respir Care 2004; 49: 1157–1164.

8. Chatburn RL. Classification of mechanical ventilators. In: Tobin MJ, ed. Principles and Practice of Mechanical Ventilation. 2nd Edn. New York, McGraw-Hill, 2006; pp.

37–52.

9. Richard JC, Carlucci A, Breton L, et al. Bench testing of pressure support ventilation with three different generations of ventilators.

Intensive Care Med 2002; 28: 1049–1057.

10. Prinianakis G, Plataki M, Kondili E, Klimathianaki M, Vaporidi K, Georgopoulos D. Effects of relaxation of inspiratory muscles on ventilator pressure during

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

pressure support. Intensive Care Med 2008; 34: 70–74.

11. Chiumello D, Pelosi P, Taccone P, Slutsky A, Gattinoni L. Effect of different inspiratory rise time and cycling off criteria during pressure support ventilation in patients recovering from acute lung injury.

Crit Care Med 2003; 31: 2604–2107.

12. Kacmarek RM, Malhotra A. Equipment required for home mechanical ventilation. In: Tobin MJ, ed. Principles and Practice of Mechanical Ventilation. 2nd Edn. New York, McGraw-Hill, 2006; pp.

97–127.

13. Battisti A, Tassaux D, Janssens JP, Michotte JB, Jaber S, Jolliet P. Performance characteristics of 10 home mechanical ventilators in pressure-support mode: a comparative bench study. Chest 2005; 127: 1784–1792.

BG-09 Ventilators:BG-09 Ventilators

20/8/08

23:14

Page 117

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

09

REFERENCES continued 14. Calzia E, Radermacher P: Helium in the treatment of respiratory failure: why not a standard?

Intensive Care Med 2003; 29: 1406–1408.

15. Goulet R, Hess D, Kacmarek RM. Pressure vs flow triggering during pressure support ventilation. Chest 1997; 111: 1649–1653.

16. Takeuchi M, Williams P, Hess D, Kacmarek RM. Continuous positive airway pressure in new-generation mechanical ventilators: a lung model study. Anesthesiology 2002; 96: 162–172.

17. Sinderby C, Navalesi P, Beck J, et al. Neural control of mechanical ventilation in respiratory failure.

Nat Med 1999; 5: 1433–1436.

18. Navalesi P, Costa R. New modes of mechanical ventilation: proportional assist ventilation, neurally adjusted ventilatory assist, and fractal ventilation. Curr Opin Crit Care 2003; 9: 51–58.

19. MacIntyre NR. Respiratory function during pressure support ventilation.

Chest 1986; 89: 403–408.

20. Tassaux D, Gainnier M, Battisti A, Jolliet P. Impact of expiratory trigger setting on delayed cycling and inspiratory muscle workload.

Am J Respir Crit Care Med 2005; 172: 1283–1289.

21. MacIntyre NR, Ho LI. Effects of initial flow rate and breath termination criteria on pressure support ventilation. Chest 1991; 99: 134–138.

22. Hess DR. Ventilator waveforms and the physiology of pressure support ventilation. Respir Care 2005; 50: 166–186.

23. Squadrone V, Gregoretti C, Ranieri VM. Patient ventilator interaction In: Fink MP, Abraham E, Vincent JL, Kochanek PM, eds. Textbook of Critical Care, 5th Edn. Philadephia, Elsevier Saunders, 2005; pp.

505–510.

24. Calderini E, Confalonieri M, Puccio PG, Francavilla N, Stella L, Gregoretti C.Patient-ventilator asynchrony during noninvasive ventilation: the role of expiratory trigger. Intensive Care Med 1999; 25: 662–667.

25. Younes M, Puddy A, Roberts D, et al. Proportional assist ventilation. Results of an initial clinical trial. Am Rev Respir Dis 1992; 145: 121–129.

26. Wysocki M, Richard JC, Meshaka P. Noninvasive proportional assist ventilation compared with

noninvasive pressure support ventilation in hypercapnic acute respiratory failure. Crit Care Med 2001; 30: 323–329.

27. Alexopoulou C, Kondili E, Vakouti E, Klimathianaki M, Prinianakis G, Georgopoulos D. Sleep during proportional-assist ventilation with load-adjustable gain factors in critically ill patients. Intensive Care Med 2007; 33: 1139–1474.

28. Rusterholtz T, Bollaert PE, Feissel M, et al. Continuous positive airway pressure vs. proportional assist ventilation for noninvasive ventilation in acute cardiogenic pulmonary edema. Intensive Care Med 2008; 34: 840–846.

29. Varelmann D, Wrigge H, Zinserling J, Muders T, Hering R, Putensen C. Proportional assist versus pressure support ventilation in patients with acute respiratory failure: cardiorespiratory responses to artificially increased ventilatory demand. Crit Care Med 2005; 33: 2125–2126.

30. Banner MJ, Paul B. Blanch PB, Gabrielli A. Tracheal pressure control provides automatic and variable inspiratory pressure assist to decrease the imposed resistive work of breathing. Crit Care Med 2002; 30: 1106–1111.

31. Fabry B, Haberthür C, Zappe D, Guttmann J, Kuhlen R, Stocker R. Breathing pattern and additional work of breathing in spontaneously breathing patients with different ventilatory demand during inspiratory pressure support and automatic tube compensation.

Intensive Care Med 1997; 23: 545–552.

32. Laubscher TP, Frutiger A, Fanconi S, Jutzi H, Brunner JX. Automatic selection of tidal volume, respiratory frequency and minute ventilation in intubated ICU patients as start up procedure for closed-loop controlled ventilation.

Int J Clin Monit Comput 1994; 11: 19–30.

33. Laubscher TP, Heinrichs W, Weiler N, Hartmann G, Brunner JX. An adaptive lung ventilation controller.

IEEE Trans Biomed Eng 1994; 41: 51–59.

34. Brunner JX, Laubscher TP, Banner MJ, Iotti G, Braschi A. Simple method to measure total expiratory time constant based on the passive expiratory flow-volume curve. Crit Care Med 1995; 23: 1117–1122.

35. MacIntyre NR, Day S. Essentials

for ventilatory-alarms system.

Respir Care 1992; 37: 1108–1112.

36. Fabry B, Guttmann J, Eberhard L, Bauer T, Haberthür C, Wolff G. An analysis of desynchronization betweent e spontaneously breathing patient and ventilator during inspiratory pressure support. Chest 1995; 107: 1387–1394.

37. Rossi A, Gottfried SB, Higgs BD, Zocchi L, Grassino A, Milic-Emili J. Respiratory mechanics in mechanically ventilated patients with respiratory failure. J Appl Physiol 1985; 58: 1849–1858.

38. Ivanyi Z, Radermacher P, Kuhlen R, Calzia E. How to choose a mechanical ventilator. Curr Opin Crit Care 2005; 11: 50–55.

39. French CJ. Work of breathing measurement in the critically ill patient. Anaesth Intensive Care 1999; 27: 561–573.

40. Marini JJ. What derived variables should be monitored during mechanical ventilation? Respir Care 1992; 37:1097–1107.

41. Tobin MJ. Monitoring of pressure, flow, and volume during mechanical ventilation. Respir Care 1992; 37: 1081–1096.

42. Lofaso F, Aslanian P, Richard JC, et al. Expiratory valves used for home devices: experimental and clinical comparison. Eur Respir J 1998; 11: 1382–1388.

43. Lofaso F, Brochard L, Touchard D, Hang T, Harf A, Isabey D. Evaluation of carbon dioxide rebreathing during pressure support ventilation with airway management system (BiPAP) devices. Chest 1995; 108: 772–778.

44. Ferguson GT, Gilmartin M. CO2 rebreathing during BiPAP ventilatory assistance. Am J Respir Crit Care Med 1995; 151: 1126–1135.

45. Saatci E, Miller DM, Stell IM, Lee KC, Moxham J. Dynamic dead space in face masks used with noninvasive ventilators: a lung model study. Eur Respir J 2004; 23: 129–135.

46. Schettino GP, Chatmongkolchart S, Hess DR, Kacmarek RM. Position of exhalation port and mask design affect CO2 rebreathing during noninvasive positive pressure ventilation. Crit Care Med 2003; 31: 2178–2182.

47. Jones A, Rowe B, Peters J, Camargo C, Hammarquist C,

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

117

BG-09 Ventilators:BG-09 Ventilators

09

20/8/08

23:14

Page 118

HOW TO CHOOSE AN INTENSIVE CARE UNIT VENTILATOR

REFERENCES continued Rowe B. Inhaled beta-agonists for asthma in mechanically ventilated patients. Cochrane Database Syst Rev 2001; 4: CD001493.

48. Dojat M, Harf A, Touchard D, Laforest M, Lemaire F, Brochard L. Evaluation of a knowledge-based system providing ventilatory management and decision for extubation. Am J Respir Crit Care Med 1996; 153: 997–1004.

49. Branson RD, Campbell RS. Sighs: wasted breath or breath of fresh air? Respir Care 1992; 37: 462–468.

50. Patroniti N, Foti G, Cortinovis B, et al. Sigh improves gas exchange and lung volume in patients with acute respiratory distress syndrome undergoing pressure support ventilation. Anesthesiology 2002; 96: 788–794.

51. Pelosi P, Bottino N, Chiumello D, et al. Sigh in supine and prone position during acute respiratory distress syndrome. Am J Respir Crit Care Med 2003; 167: 521–527.

52. Pelosi P, Cadringher P, Bottino N, et al. Sigh in acute respiratory distress syndrome. Am J Respir

118

Crit Care Med 1999; 159: 872–880.

53. Vignaux L, Tassaux D, Jolliet P. Performance of noninvasive ventilation modes on ICU ventilators during pressure support: a bench model study.

Intensive Care Med 2007; 33: 1444–1451.

54. Pelosi P, Chiumello D, Calvi E, et al. Effects of different continuous positive airway pressure devices and periodic hyperinflations on respiratory function. Crit Care Med 2001; 29: 1683–1689.

55. Takeuchi M, Williams P, Hess D, Kacmarek RM. Continuous positive airway pressure in new-generation mechanical ventilators: a lung model study. Anesthesiology 2002; 96: 162–172.

56. Nava S, Ambrosino N, Bruschi C, Confalonieri M, Rampulla C. Physiological effects of flow and pressure triggering during noninvasive mechanical ventilation in patients with chronic obstructive pulmonary disease. Thorax 1997; 52: 249–245.

57. Ferrer M, Esquinas A, Leon M,

THE BUYERS’ GUIDE TO RESPIRATORY CARE PRODUCTS

Gonzalez G, Alarcon A, Torres A. Noninvasive ventilation in severe hypoxemic respiratory failure: a randomized clinical trial. Am J Respir Crit Care Med 2003; 168: 1438–1444.

58. Layfield C. Non-invasive BiPAP implementation of a new service.

Intensive Crit Care Nurs 2002; 18: 310–319.

59. Gay PC, Hess DR, Hill NS. Noninvasive proportional assist ventilation for acute respiratory insufficiency. Comparison with pressure support ventilation. Am J Respir Crit Care Med 2001; 164: 1606–1611.

60. Taccone P, Hess D, Caironi P, Bigatello LM. Continuous positive airway pressure delivered with a “helmet”: effects on carbon dioxide rebreathing. Crit Care Med 2004; 32: 2162–2163.

61. Racca F, Appendini L, Gregoretti C, et al. Helmet ventilation and carbon dioxide rebreathing: effects of adding a leak at the helmet ports. Intensive Care Med 2008; Epub ahead of print [PMID: 18458874].