SAJOG Vol 23, No 1 (2017) by HMPG

ISSN 2305-8862

May 2017 Vol. 23 No. 1

• Abnormal umbilical artery Doppler velocimetry and placental histopathological correlation in fetal growth restriction • Preventive role of low-molecular-weight heparin in unexplained recurrent pregnancy loss • A cross-sectional descriptive study of breastfeeding behaviour and galactogogue use among private-sector patients in Cape Town, South Africa • Bilateral synchronous benign ovarian neoplasm: A rare occurrence

SAJOG May 2017 Vol. 23 No. 1

THE SOUTH AFRICAN JOURNAL OF OBSTETRICS AND GYNAECOLOGY

Editor William Edridge Editorial Board: SAJOG Alan Alperstein (Cape Town) Geoffrey Buga (Walter Sisulu) Hennie Cronje (Free State) Franco Guidozzi (Witwatersrand) Justus Hofmeyr (East London) Thinus Kruger (Stellenbosch) Gerhard Lindeque (Pretoria) Eddie Mhlanga (KwaZulu-Natal) Sam Monokoane (Limpopo) Jack Moodley (KwaZulu-Natal) Dan Ncayiyana (Durban) Hein Odendaal (Stellenbosch) Zephne van der Spuy (Cape Town) HEALTH & MEDICAL PUBLISHING GROUP (HMPG):

CONTENTS

CEO and Publisher Hannah Kikaya Email: hannahk@hmpg.co.za

Editorial

Executive Editor Bridget Farham

Classification, confusion and misclassification

W Edridge

Research 3 Puerperas᾽ knowledge regarding postpartum exercises in a tertiary hospital in the Capricorn District of Limpopo Province, South Africa

M O Mbombi, M K Thopola, J C Kgole

7 Knowledge, attitudes and practices of health professionals in public health institutions on emergency contraception in Pietermaritzburg, KwaZulu-Natal Province, South Africa N van der Westhuizen, G Hanekom

12 Abnormal umbilical artery Doppler velocimetry and placental histopathological correlation in fetal growth restriction

R Agarwal, A Tiwar, N Wadhwa, G Radhakrishnan, S Bhatt, P Batra

Managing Editors Ingrid Nye Claudia Naidu Technical Editor Naadia van der Bergh Kirsten Morreira Emma Buchanan Paula van der Bijl Production Manager Emma Jane Couzens DTP and Design Travis Arendse Clinton Griffin Chief Operating Officer Diane Smith | Tel. 012 481 2069 Email: dianes@hmpg.co.za Online Support Gertrude Fani Tel. 021 532 1281 | Cell. 072 463 2159 Email: publishing@hmpg.co.za

17 Preventive role of low-molecular-weight heparin in unexplained recurrent pregnancy loss

ISSN 2305-8862

Journal website: www.sajog.org.za

E S Khan, A Basharat, M Jamil, S Ayub, M A Khan

20 A cross-sectional descriptive study of breastfeeding behaviour and galactogogue use among private-sector patients in Cape Town, South Africa

N Steyn, M Zunza, E H Decloedt

Use of editorial material is subject to the Creative Commons Attribution – Non-commercial Works Licence. https://creativecommons.org/licenses/ by-nc/4.0

24 Beliefs and practices in using misoprostol for induction of labour among obstetricians in Zimbabwe

M G Madziyire, B Mateveke, M F Gidiri

Case Reports 28 Secondary postpartum haemorrhage with uterine artery pseudoaneurysm after caesarean section, managed with uterine artery embolisation

S Barahmeh, S H Edwan

31 Bilateral synchronous benign ovarian neoplasm: A rare occurrence

I U Takai, U A Umar, H A Nggada, M Bukar, B M Audu

CPD Questions

Listed in DOAJ, AIM, AJOL, Excerpta Medica (EMBASE), EBSCO, Scopus, Sabinet, Emerging Sources Citation Index (Web of Science) Published by the Health and Medical Publishing Group, a subsidiary of the South African Medical Association HEAD OFFICE Block F, Castle Walk Corporate Park, Nossob Street, Erasmuskloof Ext. 3, Pretoria, 0181 Tel: 012 481 2069 EDITORIAL OFFICE Suite 11, Lonsdale Building, Lonsdale Way, Pinelands, 7405 Tel: 021 532 1281 Cell: 072 635 9825 E-mail: publishing@hmpg.co.za All letters and articles for publication must be submitted online at www.sajog.org.za ©Copyright: Health and Medical Publishing Group (Pty) Ltd.

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

EDITORIAL

Classification, confusion and misclassification The classification of objects and phenomena in science and nature has fascinated academics since Carl Linnaeus, the Swedish botanist and zoologist, created his binomial description of living things in the 1700s and probably long before in accounts of others in textbooks long since gone. It must have concerned human beings and animals since they first became aware. Classification is designed to assist understanding. The response to a phenomenon, object or substance may be assisted by knowing which class the subject belongs to: a chemist may know what to expect from a reaction by knowing that an aldehyde, not an ester, is being investigated. Classifications change and in that process, we can see that someone or some group has recognise that a previous classification hindered understanding or moulded the response to a finding incorrectly. This change is a beneficial event, but it tells us that at any given time we should question classifications to look for flaws. In 2014, the World Health Organization’s[1] classification of endometrial hyperplasia changed and was simplified to include two categories, compared with four previously: from with atypia and without (each of those complex or simple) to hyperplasia with atypia or without atypia. The previous classification, which was suited to a pathologist’s ability to distinguish, had led to confusion and mismanagement by gynaecologists. What is worrying if not the word ‘complex’? But this is just a histological term. The important distinction was atypia. The latest Royal College of Obstetricians and Gynaecologists (UK) guidelines on placenta previa and accreta, published in 2011,[2] recognise a change in the classification of placenta previa. Placenta previa was once classified into four categories: entering the lower segment, approximating the os, covering the os, or mainly in the lower segment. This has now been replaced by major and minor: approaching the os in the lower segment, or covering the os. Why change? Because the elongation of the lower segment in pregnancy renders many previas insignificant, exposing overtreatment and needless anxiety when mere confirmation of ‘emigration’ was all that was necessary. Emigration is rendered less likely if the placenta covers the os by 1.5 - 2 cm or if it is anterior when there is a previous caesarean section. But all is not yet simple: what remains confusing is the doubt, highlighted in the text, that surrounds the correct safe management in labour of a placenta that is classified as approaching the os. Many of the recommendations in the guideline relating to this problem are level ‘D’. In out-patient managed antenatal cases in highly resourced settings, with good transport, the likelihood of heavy bleeding seems independent of the degree of previa.[2] In other words, there is no clear evidence to support one course of action against another based on the identification of a particular case within this classification. The classification is incomplete, partly due to the difficulty of studying a sporadic problem scientifically or perhaps because of the difficulty, once again, of recognising subsets within the classification. The guideline also details an important change in another classification: placenta accreta, the abnormal attachment of the placenta to the myometrium that may be associated with placenta previa. The risk of accreta is up to 40% in cases of previa with three previous caesareans. Incorrectly managed placenta previa may be cataclysmic, and even more so with accreta. But how to classify placenta accreta?

Previously it was said that accreta involved early myometrial invasion, increta further and percreta beyond. Abnormal vasculature adjacent to the uterus on imaging may represent penetration of the placenta through the uterus or mere recruitment and enlargement of separate vessels beyond to enhance blood supply to the myometrium. The guideline discusses magnetic resonance imaging findings and Doppler findings (grey scale, colour and three-dimensional power Doppler) that cast light on the extent of invasion. The truth is, you cannot easily tell the extent of penetration. If there is little invasion of the placenta you can tell, but if there is more, you cannot definitively say until the operation. So, if in doubt, take a senior surgeon, a cell saver (if available), and be prepared. Hence the simplification – it is all ‘accreta’. The old classification is accurate to the pathologist, not the obstetrician. Some classifications naturally evolve with time. Premature labour was once simply classified as being before 37 weeks. Modern care of the newborn makes the difference between delivery at 38 weeks, 37 weeks, and even 36 weeks most often meaningless. Hence the further classification into very preterm and extremely preterm newborn. Similarly, the arbitrary cut-off of 2.5 kg as the classification of a low birthweight baby, malnourished in the uterus and at risk of longterm injury, has been replaced by the description growth restricted – a baby of 2.8 kg may be at risk if it was intended/capable of growing to 3.6 kg, and such a fetus may not suffer the rigours of birth and delivery or even persisting pregnancy well. This failure to reach potential and be classified as growth restricted might be recognised clinically by poor symphysis fundal height development and palpated low liquor levels, which are unreliable, or asymmetric growth on sonar, which may not be available in all settings. The at-risk fetus, despite attempts to classify and identify, may still often go unrecognised. The classification of vulval disease, too, has changed four times in 40 years. Clearly, something is not understood correctly. The common woodlouse that inhabits the insect world is, in fact, a crustacean, which includes crabs, lobsters and prawns. It is a close relative and survivor of the ammonites, many now extinct, which once inhabited and occasionally still do inhabit the ocean floor. The woodlouse, however, has many segments and many legs, as do the myriapods, millipedes and centipedes that, like the woodlouse, inhabit the insect world. Evidently, classification can help or hinder; it can make sense in one way but not in another and it can be confusing and misleading. Anyone who goes to the seaside to look for woodlice would be cruelly disappointed. There must be other similar confusions and misclassifications that inhabit the world of obstetrics and gynaecology.

Will Edridge Editor

1. Emons G, Beckmann MW, Schmidt D, Mallmann P, for the Uterus Commission of the Gynecological Oncology Working Group (AGO). New WHO Classification of Endometrial Hyperplasias. Geburtshilfe Frauenheilkd 2015;75(2):135-136. https://doi.org/10.1055/s-0034-1396256 2. Royal College of Obstetricians & Gynaecologists (RCOG). Placenta Praevia, Placenta Praevia Accreta and Vasa Praevia: Diagnosis and Management (Green-top Guideline No. 27). London: RCOG, 2011. https://www.rcog.org.uk/en/guidelines-research-services/guidelines/gtg27/ (accessed 29 April 2017).

S Afr J Obstet Gynaecol 2017;23(1):2. DOI:10.7196/SAJOG.2017.v23i1.1191

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

Puerperas’ knowledge regarding postpartum exercises in a tertiary hospital in the Capricorn District of Limpopo Province, South Africa M O Mbombi, MCur; M K Thopola, PhD; J C Kgole, DLit et Phil Department of Nursing Science, School of Healthcare Sciences, University of Limpopo, Polokwane, South Africa Corresponding author: M O Mbombi (masenyani.mbombi@ul.ac.za)

Background. Postpartum exercises (PPEs) are critical for the involution process post delivery. PPE affects the physical and psychological wellbeing of puerperas. Clinical observation shows a lack of exercise by puerperas in postnatal units. Objectives. To determine puerperas’ knowledge regarding PPE, as well as to develop a health education programme on PPE. Methods. A quantitative, non-experimental, descriptive research design was employed. Probability random sampling was used to ensure that all puerperas had an equal opportunity to be selected. The sample comprised 50 puerperas who were admitted to a postnatal unit at a tertiary hospital in the Capricorn District, Limpopo, South Africa. The researchers distributed questionnaires that contained closedended questions to the puerperas on the days they visited the institution. Reviews of the literature and consultations with midwifery experts were conducted to ensure content validity. Data analysis was done through descriptive statistics. Results. Our data revealed that 68% of puerperas participating in the study lacked knowledge regarding PPE, whereas 72% of puerperas were not exercising due to perineal pains, discomfort, exhaustion and a lack of educational programmes at clinics and hospitals. The study showed that there was a high rate of ignorance among puerperas regarding the importance of PPE. Conclusion. We recommend that registered midwives initiate, develop and implement a sustainable educational programme on PPE in postnatal units. S Afr J Obstet Gynaecol 2017;23(1):3-6. DOI:10.7196/SAJOG.2017.v23i1.1059

Exercises are an important part of daily life for many women. According to Davidson et al.[1] postpartum exercises (PPEs) are regarded as basic therapy, which may improve the health of puerperas. Body image is a great concern for puerperas.[2] However, most puerperas never worry about the exercises following parturition and their benefit. This suggests limited knowledge and awareness regarding PPE that might result in backache, uterine subinvolution, urinary incontinence and flabby abdominal muscles.[3] The study findings by Gaffield et al.[4] indicated that puerperas have limited information with regard to the importance and benefits of PPEs. Sellers[5] asserts that physiological changes such as the weakening of the abdominal and pelvic floor muscles, stiffness, and swelling of tissues caused by trauma during delivery inhibit the performance of PPEs, which are necessary to bring the stretched abdominal and pelvic muscles back to normal.[6] A study by Ashrafinia et al.[7] has shown that PPEs help to strengthen pelvic and abdominal muscles, help in controlling haemorrhage and ensure a speedy uterine involution and recovery to the non-pregnant weight and physique. Davies et al.[6] further indicated that puerperas should be engaged in PPE to reduce postpartum complications such as postpartum haemorrhage and/or puerperal sepsis. Fine et al.[8] recommend that, in the absence of either obstetric or medical complications, puerperas should perform moderate exercises to maintain cardiopulmonary functions, muscular fitness and so that the internal structures could return to the pregravid state.

Methods

to the entire population in the tertiary hospital of Capricorn District. Measurements focused on specific variables, namely age, parity, knowledge and PPE, that were quantified through rating scale and frequency count. The study hypothesis was that young and educated puerperas had knowledge regarding PPE.

Population and sampling The population was puerperas in a postnatal unit of a tertiary hospital in Capricorn District, Limpopo Province, South Africa (SA).[3] Probability random sampling was used to ensure that all puerperas had an equal opportunity to be selected for the study. The sample study consisted of 50 puerperas who completed the questionnaire in the postnatal unit. Puerperas were assembled in one cubicle and small papers with numbers or without numbers were placed in a box where each respondent picked a paper. Those who picked papers with numbers were included in the study and those without numbers were excluded from the study. [11]

Data collection Questionnaires with closed-ended questions were used as a method of data collection. Puerperas were assembled in the hospital cubicle and briefed about the self-administered questionnaires with fixed alternative questions.[11,12] Fifty puerperas completed the questionnaires, which were designed to collect demographic data (section A) as well as data on the mothers’ knowledge of PPE. Data collection took 5 days.

Research design

Ethical considerations

A quantitative, non-experimental and descriptive research design was employed in this study.[9] This design enabled the researchers to study a selected sub-group of the population and to extrapolate the findings

Permission to conduct the study was requested from the Medunsa Research Ethics Committee (MREC) and from the Limpopo Department of Health and Social Development. Informed consent

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH

Validity and reliability

14 12

Knowlegde Yes No

8 7

2 1

Grade 10

Grade 11

Grade 12

Fig. 2. Puerperas’ knowledge regarding PPE in relation to their level of education. 20

Knowlegde Yes No

Puerperal women, n

10 8

5 1

2 Parity

Fig. 3. Puerperas’ knowledge regarding PPE in relation to their parity. 25

Knowlegde Yes No

15 11

Puerperal women, n

7 6

Normal

Post-grade 12

Education

Puerperal women, n

Puerperas aged 17 - 19 years had the highest lack of knowledge regarding PPE. This supported the hypothesis from the researchers. Only 2 (4%) had knowledge about PPE. There were no differences in knowledge among puerperas between 20 and 22 years old. Fifty percent of the puerperas had knowledge regarding PPE, while the other half had no knowledge regarding PPE. Nine (18%) puerperas between 23 and 25 years old, had no knowledge, while only 7 (14%) had knowledge with regard to PPE. Only 1 (2%) of the puerperas, who was between 26 and 28 years old, had knowledge with regard to PPE, while 9 (18%) puerperas had no knowledge about puerperal exercises (Fig. 1). Nine (18%) puerperas who were learners in grade 10 had no knowledge about PPE and only 1 (2%) puerpera in grade 11 had no knowledge of PPE. Five (10%) puerperas who were learners in grade 11 had knowledge of PPE. Fifteen (30%) puerperas in grade 12 lacked knowledge about PPE and only 5 (10%) had knowledge about PPE. Nine (18%) post-grade 12 puerperas had no knowledge while only 6 (12%) had knowledge about PPE (Fig. 2). Nineteen (38%) puerperas who were para 1 had no knowledge regarding PPE. Fifteen (30%) puerperas who were para 2 had knowledge, while 8 (16%) had no knowledge regarding PPE. Seven (14%) puerperas who were para 3 had no knowledge and only 1 (1%) had knowledge regarding PPE (Fig. 3). Twenty-one (42%) puerperas who had normal vaginal deliveries had no knowledge and only 4 (8%) puerperas had knowledge

Midwifery experts and literature reviewing ensured content validity. Pretesting of the questionnaire contributed towards reliability of the data collection instrument, measured in terms of asking about things the puerperas were likely to be able to answer and clarity of the questions.

Results

Knowlegde Yes No

Puerperal women, n

was obtained from all puerperas after explaining the goals of the investigation, as well as the possible advantages and disadvantages of participating. Participation to the study was voluntary. Each respondent signed the consent forms after the researchers had explained the purpose of the study based on the research design in their own language. Puerperas were assured of confidentiality. A number was allocated to each respondent to ensure anonymity and confidentiality.

C/S

Vacuum

Forceps

Type of delivery

Fig. 4. Puerperas’ knowledge regarding PPE in relation to the type of delivery

2 1

17 - 19

20 - 22 23 - 25 Age group (years)

26 - 28

Fig. 1. Puerperas’ knowledge regarding PPE in relation to their age group.

regarding PPE. Eleven (22%) puerperas who delivered by caesarean section (C/S) had knowledge, whereas three who delivered by C/S had no knowledge regarding PPE. Five (10%) puerperas who delivered by vacuum extraction had no knowledge, and only one

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH Table 1. Educational programme for PPE Day

Type of exercise

Importance

Activities

Pelvic floor (Kegels)

Strengthen the muscles of the pelvic floor Regain full bladder control Prevent puerperal stress incontinence Ensure sexual satisfaction

Squeeze/tighten the anal sphincter for 3 seconds, relax the muscle for 3 seconds, and squeeze again. Begin with 10 3-second squeezes, thrice a day and increase gradually. Work up to 50 - 100 Kegels per day.

Abdominal breathing

Increase the tonicity of the deep transverse muscles

Lying supine, inhale deeply using abdominal muscles to expand the abdomen. Exhale slowly through pursed lips while tightening the abdominal muscles. Repeat 2 - 3 times.

Abdominal tightening

Strengthen abdominal muscles

Lie on back with knees bent. Tighten abdominal and buttock muscles and allow pelvis to tilt upwards.

Ankle/foot circulation

Enhance circulation, reduce oedema and prevent deep-vein thrombosis

Do ankle circles clockwise and anti-clockwise in different positions, such as sitting or lying down, etc. Repeat circular pattern 3 - 5 times daily.

Leg sliding

Increase circulation in legs

Lie on back with knees bent. Do the pelvic tilt to keep back flat while sliding one heel up and down the bed.

Arm and upper back stretch

Relieve backache and strengthen the ligaments

Raise arms over head with elbows straightened and palms facing one another, and hold position for 5 - 10 seconds. Lower arms out to the side, palms facing downward, while maintaining a straight back.

Abdominal strengthening

Strengthen the abdominal muscles

Lie on back with knees bent and feet flat on the floor. Slowly move chin to chest and raise the head and shoulders until the neck is 15 - 20 cm off the floor, while one arm is stretched out in front of the body.

Straight curl-ups

Strengthen the abdominal muscles

Lie on back with knees bent. Breathe in slowly through nose. Tuck chin in and raise head while hands are pointed toward the knees. Exhale and lift shoulders off the floor. Hold the position for 5 seconds. Inhale and slowly lower body to the count of five.

Sit-ups

Strengthen the abdominal muscles

Elevate head with a pillow and bend knees. Tuck chin in, exhale and reach towards the knees. Hold the position for 5 seconds and inhale as you release. Repeat 3 - 5 times.

Aerobic activity

Promote cardiorespiratory and muscular fitness

Repeat all the above exercises and start to initiate gentle aerobic activities, such as walking, as soon as you are able to tolerate them, within ~1 week, where possible. Vigorous aerobic activity can usually be resumed after your postpartum check-up.

3-7

(2%) had knowledge regarding PPE. All 5 (10%) puerperas who delivered by forceps had no knowledge regarding PPE (Fig. 4).

Discussion

The results of the study indicated that 18 (36%) puerperas from the sample were not exercising due to perineal pains and discomfort experienced following labour and delivery. Table 1 recommends the types of exercises and also contains information on various exercises to improve puerperas knowledge regarding PPE. Downs et al.[2] hold that puerperas fail to initiate PPE because of exhaustion, perineal pains and discomfort that result from a sutured perineum. The study revealed that 34 (68%) puerperas lacked knowledge regarding PPE and there were no educational

programmes offered to them either at the clinic level during puerperal care or at the hospital after delivery. However, 4 (8%) of the puerperas engaged actively in PPE. The findings also indicated that puerperas were unable to understand the significance of exercising after delivery. Twenty (40%) puerperas reported that breastfeeding could return their bodies to the pregravid state. Only 11 (22%) puerperas had knowledge that through PPE, their bodies could return to the pregravid state. Nineteen (38%) puerperas reported that eating a normal balanced diet could return their bodies to the non-gravid state. Only 3 (6%) puerperas reported that other methods such as the tying of the abdomen could return their bodies to the nongravid state.

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

RESEARCH

Table 2. Exercises following C/S Day

Types of exercises

Importance

Activities

Foot and leg

To reduce oedema and prevent deep-vein thrombosis

Do ankle circles in different positions, when sitting and/ or lying down. Repeat circular pattern 3 - 5 times.

Deep breathing

To ensure full expansion of the lungs

Exhale slowly against pursed lips, while tightening the abdominal muscles.

Coughing

Helps to loosen secretions

Patient should cough while seated, with sutures supported by both hands and/or a pillow.

Pelvic floor, curl-up and hip hitching

To regain full bladder control and ensure normal sexual satisfaction

Tilt the pelvis by flattening the hollow of the back on the floor and squeeze pillow between knees.

3-7

Breathing and abdominal

To promote abdominal muscle contraction and tonicity, as well as maintain cardio-respiratory and muscular fitness

Lie down on a comfortable surface with knees bent. Relax and allow body weight to sink into the surface on which you are lying. Contract the abdominal muscles during exhalation.

Recommendations The findings of the study suggest that the initiation and implementation of health education programmes on PPE in postnatal units would improve the knowledge of puerperas regarding PPE. Wagner et al.[13] hold that educational programmes can be initiated and implemented in postnatal units for the improvement of PPE and enriching the puerperas’ knowledge. Table 1 outlines the recommended health education programme. Midwives should emphasise the importance of PPE so that puerperas can be assisted to regain their pregravid state.

Study limitations The findings of this study could not be generalised to other hospitals in the Limpopo Province because the study was conducted in one tertiary hospital in the Capricorn District of Limpopo. However, the study could be replicated at other healthcare institutions.

Conclusion

The study revealed that 36 (72%) puerperas did not exercise because of a lack of knowledge regarding PPE. The high rate of ignorance with regards to the importance of PPE indicates that there is an urgent need for registered midwives to initiate, develop

and implement sustainable educational programmes on PPE in maternity units. 1. Davidson MC, London ML, Ladewig PW. Contemporary Maternal-Newborn Nursing Care: Nurse, Family, Community. 8th ed. USA: Pearson, 2014. 2. Downs DS, Hansenblas HA. Women’s exercise beliefs and behaviours during their pregnancy and postpartum. J Midwifery Women’s Health 2004;49(2):138-144. https://doi.org/10.1016/j. jmwh.2003.11.009 3. Olds S, London M. Maternal Newborn Nursing and Woman’s Health Care. 7th ed. New Jersey: Upper Saddle River, 2004. 4. Gaffield ME, Egan S, Temmerman M. It᾽s about time: WHO and partners release programming strategies for postpartum programs. Global Health Sci Pract 2014;2(1):4-9. https://doi.org/10.9745/ ghsp-d-13-00156 5. Sellers PM. Midwifery. South Africa: Juta and Company, 2004. 6. Davies GA, Wolfies LA, Mottola ME, Mackinnson C. Exercise in pregnancy and the postpartum period. J Obstet Gynaecol Canada 2013;25(6):516-529. 7. Ashrafinia F, Mirmohammadali M, Rajabi H, Kazemnejad A, Sadeghniiat HK. Effect of Pilates exercises on postpartum maternal fatigue. Singapore Med J 2015;56(3):169-173. https://doi. org/10.11622/smedj.2015042 8. Fine P, Burgio K, Borello-France D, et al. Teaching and practicing of pelvic floor muscle exercises in primiparous women during pregnancy and the postpartum period. Am J Obstet Gynecol 2007;197(1):107.e1-107.e5. https://doi.org/10.1016/j.ajog.2007.02.052 9. Burns N, Grove SK. The Practice of Nursing Research: Appraisal, Synthesis and Generation of Evidence. 6th ed. Saint Louis: Saunders, 2009. 10. Brink HI. Fundamentals of Research Methodology for Health Professionals. Cape Town: Juta and Company, 2006. 11. Babbie E, Mouton J. The Practice of Social Research. Cape Town: Oxford University Press, 2009. 12. LoBiondo-Wood G, Harber J. Nursing Research Methods and Critical Appraisal for Evidencebased Practice. 7th ed. New York: Mosby Elsevier Inc., 2010. 13. Wagner DL, Bear M, Davidson NS. Measuring patient satisfaction with postpartum teaching methods used by nurses within the interaction model of client health behavior. Res Theory Nurs Prac 2011;25(3):1-7. https://doi.org/10.1891/1541-6577.25.3.176

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

Knowledge, attitudes and practices of health professionals in public health institutions on emergency contraception in Pietermaritzburg, KwaZulu-Natal Province, South Africa E Sibanda, MB ChB, Dip Fam Med, Dip Obst, FCOG (SA); M J Titus, MB ChB, FCOG (SA), LLM, PG Dip Intl Res Ethics Department of Obstetrics and Gynaecology, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa Corresponding author: E Sibanda (emanuelsibanda@yahoo.com)

Background. Although emergency contraception (EC) is widely available, its use is surrounded by many controversies. Overall, it seems to be underutilised worldwide. Objectives. To determine healthcare professionalsáž˝ knowledge, attitudes and perceptions regarding EC, and how frequently they encounter, educate and issue it to patients. Methods. A questionnaire-based survey of doctors and nurses (volunteers) working in obstetrics and gynaecology was conducted in 3 public hospitals and 17 clinics in Pietermaritzburg, KwaZulu-Natal Province, South Africa. Data were analysed using SPSS. Results. Sixty-seven (25%) doctors and 201 (75%) nurses participated in the survey. Awareness of the three ECs available in the public sector overall was 56.4%, and 62.7% of participants could prescribe one EC correctly. Only 39.6% knew that EC pills prevent ovulation. Seventy-six percent thought that the use of EC could lead to high-risk sexual behaviour, high risk of transmission of HIV and non-use of other forms of contraception. Only 7.8% saw patients seeking EC often, 5.6% issued it often and 23.5% educated patients about it often. Conclusion. Participants were familiar with EC, but lacked accurate and detailed knowledge about its mechanism of action and had misperceptions on its social impact. They seldom prescribed it. S Afr J Obstet Gynaecol 2017;23(1):7-11. DOI: 10.7196/SAJOG.2017.v23i1.1091

It is estimated that about 41.0% of the 208 million pregnancies that occurred worldwide in 2008 were unintended.[1] East and Central Africa have the highest rates of unintended pregnancies and about 14 million unintended pregnancies are estimated to occur in subSaharan Africa annually.[1,2] Unintended pregnancies are resolved differently by women. Globally, there were an estimated 42 million induced abortions in 2003, of which 48% were unsafe and 97% of the unsafe abortions were performed in developing countries.[3] The World Health Organization (WHO) estimated that 21.6 million unsafe abortions occurred worldwide in 2008, an increase from 19.7 million in 2003,[4] with Eastern and Central Africa having the highest rates.[5] Unsafe abortions have a negative impact on maternal morbidity and mortality. A reduction in unintended pregnancies may result in a reduction of unsafe abortions, hence lowering maternal mortality. In situations where there is unprotected sexual intercourse or contraceptive method failure, emergency contraception (EC) may be used to prevent unintended pregnancies. Although EC is widely available in many countries, its use has been marred with controversies and misperceptions, especially concerning its mode of action, impact on behaviour and safety. EC pills prevent pregnancy by inhibiting or delaying ovulation, but they cannot disrupt an established pregnancy. However, an analysis of 1 077 articles in 113 newspapers between 1992 and 2002 showed that 44.5% of them included at least one instance of confusion between EC and medical abortion, with 31.0% of the articles inaccurately portraying the mode of action of EC as medical abortion.[6] The use of EC varies in different countries, and overall it is underutilised

worldwide. In a cross-sectional survey of 600 pregnant teenagers requesting termination of pregnancy in China, 47.7% had heard of EC and among them, 44.0% had used it a least once within the 6 months before the pregnancy.[7] In a survey of college students in Pennsylvania, USA, 74.0% had heard of EC.[8] In Turkey, a survey of married women showed that 39.6% knew about EC.[9] A study conducted in South Africa (SA) in 2001 showed that only 22.8% of patients interviewed at public primary healthcare facilities had heard about EC.[10] A survey of tertiary students in Durban, SA, showed that 56.5% of them had heard of EC,[11] and a Kenyan study showed that only 11.0% of family planning patients surveyed had heard of EC.[12] Lack of knowledge and misperceptions about EC also exists among healthcare professionals, therefore hindering the dissemination of knowledge to patients. One such example is a Florida survey, which revealed that 56% of pharmacists thought EC caused birth defects and 46% thought it caused abortion.[13] A national survey of obstetricians and gynaecologists in the USA between October 2008 and January 2009 showed that gender, religion and divergent beliefs about EC influenced their practices.[14] In SA, the public sector is the main provider of contraceptives. EC is available in the form of the combined oral contraceptive (COC) pill, the copper intra-uterine contraceptive device (Cu-IUCD) and progesterone-only pills (POPs). However, promotion of EC seems to be low. Healthcare professionals have a responsibility to counsel and offer their patients knowledge on all forms of contraception, including EC. This study therefore aimed to establish the knowledge and attitudes of healthcare professionals (doctors and nurses) towards issuing EC in the Pietermaritzburg area of KwaZulu-Natal Province, SA.

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

RESEARCH Furthermore, it aimed to establish if the healthcare workers might be a contributory factor in the lack of knowledge on EC in the public sector. Promotion of EC by healthcare professionals would have a positive impact on its use, which would decrease unintended pregnancies.

Methods

A purposive sample of doctors and nurses working in the three public hospitals in the obstetrics and gynaecology units, and 17 selected public clinics in Pietermaritzburg, SA, rendering antenatal, postnatal and contraceptive services, participated in the study in January 2013. All the selected facilities provided primary healthcare and family planning services except the tertiary hospital, which only provided a family planning service. Participation in the study was voluntary and only those who completed the questionnaire were included. The healthcare professionals included in the study were specialist obstetricians and gynaecologists, registrars, medical officers, medical interns, midwives, registered nurses and enrolled nurses. Those who participated were given a 25-item questionnaire containing closed- and open-ended questions. In 22 of the questions on the questionnaire, the participants were given choices from which they could choose the correct answer. The questionnaire was completed anonymously for 20 - 30 minutes under the supervision of the researcher and the forms were collected immediately after completion. To assess their knowledge, participants were asked to mention the ECs they were aware of, how they were prescribed, the time limit within which specific ECs should be taken and their effectiveness. They were also asked questions about ECs᾽ indications, mechanisms of action, teratogenicity, side-effects and questions pertaining to their prescription. Attitudes towards prescription were assessed by finding out if they agreed with the use of EC and if they thought its use could lead to high-risk sexual behaviour and transmission of sexually transmitted infections (STIs). Practices were assessed by asking how frequently they encountered, taught patients and prescribed EC to them. The answers to the questions on the questionnaire were coded. The data were then captured and subsequently analysed using the Statistical Package for Social Sciences version 21 (IBM Corp., USA). Descriptive statistics were used to summarise the results. This study was approved by the Biomedical Research Ethics and Postgraduate Committees of the University of KwaZulu-Natal (ref. no. BE269/12).

Results

Out of 355 healthcare professionals who were approached (80 doctors, 95 clinic nurses and 180 hospital nurses), 268 (75.5%) participated and completed the questionnaires fully. Of the participants, 67 (25%) were doctors and 201 (75%) were nurses. All the doctors who participated in the study worked in one of the three public hospitals in Pietermaritzburg, while two-thirds of the nurses who participated worked at one of the three hospitals and one-third worked at the clinics. The mean age of the doctors was 29.1 years, with an average work experience of 3.8 years, and the mean age of nurses was 38.1 years, with an average work experience of 10.8 years. Table 1 shows the detailed demographics.

Knowledge Most participants could name an EC method. The COC (Yuzpe) regimen was the most commonly mentioned method (82.5%),

Table 1. Demographics of all participants (N=268) n (%)* Profession Enrolled nurse

29 (10.8)

Professional nurse

172 (64.2)

Medical intern

41 (15.3)

Medical officer (obstetrics and gynaecology)

11 (4.1)

Registrar (obstetrics and gynaecology)

14 (5.2)

Specialist (obstetrician and gynaecologist)

1 (0.4)

Total

268 (100)

Work places Clinic nurses

67 (25)

Hospital labour ward nurses

43 (16)

Hospital antenatal ward nurses

41 (15.3)

Hospital postnatal ward nurses

26 (9.7)

Hospital O&G OPD nurses

10 (3.7)

Hospital gynaecology ward nurses

11 (4.1)

Hospital family planning nurses

3 (1.1)

Hospital O&G doctors

67 (25)

Sex Male

40 (14.9)

Female

228 (85.1)

Age distribution (years) Mean

35.73

Range

20 - 61

O&G = obstetrics and gynaecology; OPD = out-patient department. *Unless otherwise specified.

while ulipristal acetate was the least mentioned (0.4%). When asked to prescribe two ECs of their choice, 62.7% of participants could prescribe their first choice and 31.2% could correctly prescribe their second choice. Their correct responses on the time limit within which an EC of their choice could be used were: 55.6% for the COC time limit of 72 hours; 53% for the 72-hour time limit of POP (Norlevo) and 20% for the 120-hour time limit of POP (Norlevo); 16.7% for the Cu-IUCD time limit of 120 hours. The participants’ responses to a question on their knowledge of the indications of EC were: 89.6% mentioned rape or sexual assault, 79.5% mentioned unprotected sex, 43.3% mentioned a burst condom and 20.1% mentioned missed pills. Concerning the issues around prescription of EC, 70.2% stated that a prescription was not needed to access EC in SA, 50.7% stated that a repeat dose of EC should not be given within 1 month of giving another dose and 53% stated that EC should be issued to women who are >12 years of age (Table 2). Concerning the mechanisms of action of EC pills, 39.6% of participants selected that they prevent ovulation, 76.5% that they prevent implantation, 24.3% that they cause abortion, 64.6% that they prevent fertilisation and 8.6% that they cause malformations that are incompatible with life (Table 3). When asked about the effect of EC pills on the fetus if taken unknowingly by a woman who is already pregnant, their responses were as follows: 42.2% thought that the woman would have a spontaneous miscarriage, 35.1% thought that the fetus might have minor congenital abnormalities, 14.2% thought that the fetus would have major congenital abnormalities and 34.0% said that there would be no effect on the baby (9.0% of doctors and

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH Table 2. Knowledge on methods of EC Doctors (N=67), n (%)

Nurses, n (%) Hospital (N=134)

Clinic (N=67)

Total (N=201)

Total (N=268), n (%)

Type of EC COC pill (Yuzpe)

52 (77.6)

110 (82.1)

59 (88.1)

169 (84.1%)

221 (82.5)

POP (Norlevo)

46 (68.7)

63 (68.0)

49 (73.1)

112 (55.7)

158 (59.0)

Cu-IUCD

44 (65.7)

26 (19.4)

4 (6.0)

30 (14.9)

74 (27.6)

Mifepristone

4 (6.0)

4 (1.5)

Ulipristal acetate

1 (1.5)

1 (0.4)

Choice 1

39 (58.2)

77 (57.5)

52 (77.6)

129 (64.2)

168 (62.7)

Choice 2

33 (49.3)

29 (21.6)

20 (29.9)

49 (24.4)

82 (30.5)

Correct prescription of EC

Time frame for administration of EC 23 (34.3)

49 (36.5)

18 (26.9)

67 (33.3)

90 (33.6)

POP (Norlevo) (72 hours)

COC (72 hours)

17 (25.4)

17 (12.7)

19 (28.4)

36 (17.9)

53 (19.8)

POP (Norlevo) (120 hours)

3 (4.5)

8 (6.0)

9 (13.4)

17 (8.5)

20 (7.5)

1 (1.5)

1 (0.4)

Rape/sexual assault

59 (88)

127 (94.8)

54 (80.6)

181 (90)

240 (89.6)

Unprotected sex

55 (82.1)

100 (74.6)

58 (86.6)

158 (78.6)

213 (79.5)

Cu-IUCD (120 hours) Indications for EC

Burst condom

34 (50.1)

43 (32.1)

39 (58.2)

82 (40.8)

116 (43.3)

Missed pills

7 (10.4)

32 (23.9)

15 (22.4)

47 (23.4)

54 (20.1)

35 (52.2)

68 (50.7)

33 (49.3)

101 (50.2)

136 (50.7)

Number of times EC pills can be used in a month Once Twice

7 (10.4)

7 (5.2)

3 (4.5)

10 (5.0)

17 (6.3)

Thrice

1 (1.5)

1 (0.5)

1 (0.4)

Unlimited

9 (13.4)

20 (14.9)

10 (14.9)

30 (14.9)

39 (14.6)

Do not know

16 (24)

39 (29.1)

20 (29.9)

59 (29.4)

75 (28)

>12 years

30 (44.8)

74 (55.2)

38 (56.7)

112 (55.7)

142 (53)

>16 years

3 (4.5)

6 (4.5)

3 (4.5)

9 (4.5)

12 (4.5)

Age restriction to access EC in SA

>18 years

7 (5.2)

1 (1.5)

8 (4)

8 (3)

No age restriction

27 (40.2)

32 (23.9)

17 (25.4)

49 (24.4)

76 (28.3)

Do not know

7 (10.4)

15 (11.2)

8 (11.9)

23 (11.4)

30 (11.2)

36.3% of the nurses). Concerning knowledge on the side-effects of EC pills, the majority (90.3%) mentioned nausea and vomiting (Table 3).

Attitudes Most participants (95.9%) stated that they supported the use of EC. Their responses on the age limit for issuing of EC were diverse; however, 41.0% said >12 years of age. They also gave diverse views on who should issue EC, with 23.1% saying doctors only, while 69.4% said either doctors, nurses or pharmacists. Nearly two-thirds (66.4%) of participants disagreed with the idea that EC should be issued only with a doctor’s prescription. More than three-quarters of participants (78.8%) agreed with the notion that EC should be taught in the schools’ curriculum. Concerning the impact of EC, 79.5% said it could lead to highrisk sexual behaviour, 73.9% said it could lead to a high risk of transmission of HIV and 74.6% thought it could lead to non-use of

other forms of contraception. Regarding the issuing of EC pills to a patient who came for a repeat dose after 1 week, 57.1% said they would not give a repeat dose. Those who said they would not issue a repeat gave different reasons for not doing so, with the majority (32.7%) saying that they would instead advise the client on other forms of contraceptives. Some participants (16.8%) felt it would be too soon to repeat the dose, while others said that they would not give it because the patient would be abusing it (19.5%). On the issue of advance issuing of EC, most participants (75.7%) said that they would not issue it in advance (Table 4).

Practices Slightly more than half of participants (51.9%) rarely saw patients requesting EC, 7.8% often had requests for EC and 40.3% had never had a request for EC. Regarding how frequently they taught patients about EC, 28.4% never taught, 48.1% rarely taught and 23.5% taught

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH Table 3. Knowledge on mechanism of action, effects on pregnancy and side-effects of EC pills (doctors, N=67; nurses, N=201)

Mechanism of action Prevents ovulation Prevents implantation Causes abortion Prevents fertilisation Causes malformations incompatible with life Effect of EC pills on pregnant women Spontaneous abortion Minor congenital abnormalities Major congenital abnormalities No effect on the mother and baby Common side-effects of EC pills Nausea and vomiting Menorrhagia Delayed menses Amenorrhoea Diarrhoea

Doctors

True, n (%) Nurses Total

Doctors

False, n (%) Nurses Total

Do not know, n (%) Doctors Nurses Total

27 (40.3) 60 (89.5) 19 (28.3) 37 (55.2) 4 (6.0)

79 (39.3) 145 (72.1) 46 (22.9) 136 (67.7) 19 (9.5)

106 (39.6) 205 (76.5) 65 (24.3) 173 (64.6) 23 (8.6)

36 (53.7) 5 (7.5) 43 (64.2) 19 (28.4) 50 (74.6)

97 (48.3) 35 (17.4) 124 (61.7) 33 (16.4) 103 (51.2)

133 (49.6) 40 (14.9) 167 (62.3) 52 (19.4) 153 (57.1)

4 (6.0) 2 (3.0) 5 (7.5) 11 (16.4) 13 (19.4)

25 (12.4) 21 (10.5) 31 (15.4) 32 (15.9) 79 (39.3)

29 (10.8) 23 (8.6) 36 (13.4) 43 (16.0) 92 (34.3)

39 (58.2) 26 (38.8) 9 (13.4) 18 (9.0)

75 (37.3) 68 (33.8) 29 (14.4) 73 (36.3)

114 (42.5) 94 (35.1) 38 (14.2) 91 (34)

23 (34.3) 26 (38.8) 41 (61.2) 42 (62.7)

95 (47.3) 78 (38.8) 107 (53.2) 80 (39.8)

118 (44.0) 104 (38.8) 148 (55.2) 122 (45.5)

5 (7.5) 15 (22.4) 17 (25.4) 7(10.4)

31 (15.4) 55 (27.4) 65 (32.0) 48 (23.9)

36 (13.5) 70 (26.1) 82 (30.6) 55 (20.5)

64 (95.5) 29 (43.3) 41 (61.2) 17 (25.4) 17 (25.4)

178 (88.5) 66 (32.84) 76 (37.8) 44 (21.9) 54 (26.9)

242 (90.3) 95 (35.4) 117 (43.7) 61 (22.8) 71 (26.5)

1 (1.5) 28 (41.8) 15 (22.4) 38 (56.7) 30 (44.8)

10 (5.0) 69 (34.3) 78 (38.8) 113 (56.2) 86 (42.8)

11 (4.1) 91 (36.2) 93 (34.7) 151 (56.3) 116 (43.3)

2 (3.0) 10 (14.9) 11 (16.4) 12 (17.9) 20 (29.8)

13 (6.5) 66 (32.8) 47 (23.4) 44 (21.9) 61 (30.3)

15 (5.6) 76 (24.6) 58 (21.6) 56 (20.9) 101 (30.2)

Table 4. Attitudes on EC (doctors, N=67; nurses, N=201)

Risky sexual behaviour Agree Disagree High risk of transmission of HIV Yes No Non-use of other forms of contraception Yes No Repeat dose of EC pills within 1 week Yes No Advance prescription of EC Yes No

Nurses, n (%)

Doctors, n (%)

Hospital (N=134) Clinic (N=67)

Total (N=201)

Total, n (%)

51 (76.1) 16 (23.9)

110 (82.1) 24 (17.9)

52 (77.6) 15 (22.4)

162 (80.6) 39 (19.4)

213 (79.5) 55 (20.5)

45 (67.2) 22 (38.2)

104 (77.6) 30 (22.4)

49 (73.1) 18 (26.9)

153 (76.1) 48 (23.9)

198 (73.9) 70 (26.1)

50 (74.6) 17 (25.4)

101 (75.4) 33 (24.6)

49 (73.1) 18 (26.9)

150 (74.6) 51 (25.4)

200 (74.6) 68 (25.4)

28 (41.8) 39 (58.2)

56 (41.8) 78 (58.2)

31 (46.3) 36 (53.7)

87 (43.3) 114 (56.7)

115 (42.9) 153 (5.1)

18 (26.9) 49 (73.1)

28 (20.9) 106 (79.1)

19 (28.4) 48 (71.6)

47 (23.4) 154 (76.6)

65 (24.3) 203 (75.7)

Table 5. Practices regarding EC (doctors, N=67; nurses, N=201)

Frequency of seeing clients requesting EC Never Rarely Often Frequency of educating patients about EC Never Rarely Often Frequency of issuing EC Never Rarely Often

Nurses, n (%)

Doctors, n (%)

Hospital (N=134)

Clinic (N=67)

Total (N=201)

Total (N=268), n (%)

30 (44.8) 32 (47.8) 5 (7.4)

67 (50.0) 62 (46.3) 5 (3.7)

11 (16.4) 45 (67.2) 11 (16.4)

78 (38.8) 107 (53.2) 16 (8.0)

108 (40.3) 139 (51.9) 21 (7.8)

22 (32.8) 37 (55.2) 8 (12.0)

45 (33.6) 58 (43.3) 31 (23.1)

9 (13.4) 34 (50.8) 24 (35.8)

54 (26.9) 92 (45.8) 55 (27.3)

76 (28.4) 129 (48.1) 63 (23.5)

34 (50.7) 29 (43.3) 4 (6.0)

90 (67.0) 40 (30.0) 4 (3.0)

15 (22.4) 45 (67.2) 7 (10.4)

105 (52.2) 85 (42.3) 11 (5.5)

139 (51.9) 114 (42.5) 15 (5.6)

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

RESEARCH patients often. On the issue of issuing EC to patients, 51.9% never issue, 42.5% rarely issue and 5.6% issue EC often (Table 5).

Discussion

Although most participants were doctors and nurses working in one of the three public hospitals where they do not frequently encounter patients seeking EC, the level of awareness of the available methods of EC in the public sector in SA was high, particularly for the COC pill method (82.5%), the POP (59.%) and the Cu-IUCD (27.6%). Overall awareness was high when compared with a similar survey of healthcare providers in Lagos, Nigeria, where 64.1%, 50.7% and 19.3% mentioned the COC, the POP and the Cu-IUCD, respectively.[15] In this study, awareness of the COC and POP methods was highest among the clinic nurses, at 88.1% and 73.1%, respectively. Hospital nurses and doctors were slightly less aware of these methods but awareness of the use of the Cu-IUCD was highest among the doctors (65.7%). In contrast, only 19.4% of hospital nurses and 6% of the clinic nurses were aware of the Cu-IUCD. When compared with an EC survey conducted in Turkey, where 35.3% of general practitioners (GPs) and 32.6% of nurses knew that the IUCD could be used for EC, the nurses’ level of awareness of IUCD for EC was lower, while that of doctors was higher.[16] However, when compared with a survey of family planning providers in Ghana, where only 8.3% of family planning providers were aware that the IUCD could be used as EC,[17] their level of awareness was much higher. Two-thirds of the healthcare professionals could prescribe at least one EC and about half (54.3%) knew the 72-hour time limit for EC pills as in the Nigerian survey,[15] where knowledge of the indications of EC was high. Knowledge of the mechanisms of action of EC pills was lacking as only 39.6% knew that they prevent ovulation, while the majority thought that they prevent implantation (76.5%) and fertilisation (64.6%). The results are similar to the survey in Turkey, where 87.1% of participants (96.1% GPs v. 82% nurses and midwives) thought that the mechanism of action of EC was prevention of implantation.[16] However, the knowledge was higher when compared with the survey in Ghana, where only 2.5% of the providers knew that EC pills prevent or delay ovulation.[17] Knowledge on the side-effects and issuing of EC was particularly high. However, knowledge of how frequently it should be issued and the age restriction was low. Most participants (95.9%) agreed that it was okay to use EC. However, about three-quarters of the participants had a misperception that EC might lead to high-risk sexual behaviour (79.5%), high risk of transmission of HIV (73.9%) and non-use of other forms of contraception (74.6%). These results were similar to those in the survey in Turkey, where 88.2% of GPs and 75.3% of nurses and midwives thought that if patients knew about EC, there would be a reduction in the use of condoms, and 76.4% thought that the use of EC could result in an increase in the rate of HIV and sexually transmitted infections.[16] The majority (41%) said that women should be offered EC from the age of 12 years. They gave diverse responses on who they thought should issue EC, with the majority preferring nurses (28.7%). Most of the participants (66.4%)

disagreed with the suggestion that EC should only be issued with a doctor’s prescription. The majority of the healthcare professionals never or rarely saw patients seeking EC and never or rarely issued EC, unlike in the Nigerian survey, where 58% had prescribed EC.[15] It is important to note that two-thirds of them work in a hospital setting where they rarely encounter these patients, but the clinic nurses, where most patients should present, also had similar results. Of concern is that they never or rarely educate their patients on EC.

Conclusion

Although most participants work in facilities where they do not encounter patients seeking EC, the level of awareness of ECs is high, but most health professionals lack accurate and detailed knowledge about the mechanisms of action of EC pills. Two-thirds of participants could prescribe at least one EC correctly and ~50% knew the time limit of EC pills. The majority of participants had misperceptions on the impact of EC and the provision of EC service is poor, even among the health professionals who work in clinics. There is a great need to do in-service training of healthcare professionals to dispel their misperceptions and increase their knowledge, thereby changing their attitudes and practices on EC. 1. Singh S, Sedgh G, Hussain R. Unintended pregnancy: Worldwide levels, trends, and outcomes. Stud Fam Plann 2010;41(4):241-250. https://doi.org/10.1111/j.1728-4465.2010.00250.x 2. Hubacher D, Mavranezouli I, McGinn E. Unintended pregnancy in sub-Saharan Africa: Magnitude of the problem and potential role of contraceptive implants to alleviate it. Contraception 2008;78(1):7378. https://doi.org/10.1016/j.contraception.2008.03.002 3. Sedgh G, Henshaw S, Singh S, Åhman E, Shah IH. Induced abortion: Estimated rates and trends worldwide. Lancet 2007;370(9595):1338-1345. https://doi.org/10.1016/s0140-6736(07)61575-x 4. World Health Organization. Unsafe abortion: Global and regional estimates of the incidence of unsafe abortion and associated mortality in 2008. 6th ed. http://www.who.int/ reproductivehealth/ publications/unsafe_abortion/9789241501118/en (accessed 26 June 2013). 5. Shah I, Åhman E. Unsafe abortion in 2008: Global and regional levels and trends. Reprod Health Matters 2010;18(36):90-101. https://doi.org/10.1016/s0968-8080(10)36537-2 6. Pruitt SL, Mullen PD. Contraception or abortion? Inaccurate descriptions of emergency contraception in newspaper articles, 1992-2002. Contraception 2005;71(1):14-21. https://doi.org/10.1016/j. contraception.2004.07.012 7. Xu J, Cheng L. Awareness and usage of emergency contraception among teenagers seeking abortion: A Shanghai survey. Eur J Obstet Gynecol Reprod Biol 2008;141(2):143-146. https://doi.org/10.1016/j. ejogrb.2008.08.002 8. Miller LM. College student knowledge and attitudes toward emergency contraception. Contraception 2011;83(1):68-73. https://doi.org/10.1016/j.contraception.2010.06.005 9. Ertem G, Kalkim A, Topçu S. Knowledge of emergency contraception among married women in Izmir, Turkey. Int J Gynecol Obstet 2010;110(3):270-271. https://doi.org/10.1016/j.ijgo.2010.04.018 10. Smit J, McFadyen L, Beksinska M, et al. Emergency contraception in South Africa: Knowledge, attitudes, and use among public sector primary healthcare clients. Contraception 2001;64(6):333-337. https://doi.org/10.1016/s0010-7824(01)00272-4 11. Roberts C, Moodley J, Esterhuizen T. Emergency contraception: Knowledge and practices of tertiary students in Durban, South Africa. J Obstet Gynaecol 2004;24(4):441-445. https://doi.org/10.1080/0 1443610410001685619 12. Muia E, Ellertson C, Lukhando M, Elul B, Clark S, Olenja J. Emergency contraception in Nairobi, Kenya: Knowledge, attitudes and practices among policymakers, family planning providers and clients, and university students. Contraception 1999;60(4):223-232. https://doi.org/10.1016/s00107824(99)00089-x 13. Richman AR, Daley EM, Baldwin J, Kromrey J, O'Rourke K, Perrin K. The role of pharmacists and emergency contraception: Are pharmacists᾽ perceptions of emergency contraception predictive of their dispensing practices? Contraception 2012;86(4):370-375. https://doi.org/10.1016/j. contraception.2012.01.014 14. Lawrence RE, Rasinski KA, Yoon JD, Curlin FA. Obstetrician-gynecologist physicians᾽ beliefs about emergency contraception: A national survey. Contraception 2010;82(4):324-330. https://doi. org/10.1016/j.contraception.2010.04.151 15. Ebuehi OM, Osaretin AT, Ebuehi VI. Health care providers᾽ knowledge of attitudes toward and provision of emergency contraceptives in Lagos, Nigeria. Int Fam Plann Perspect 2006;32(2)89-93. https://doi.org/10.1363/3208906 16. Sevil U, Yanikkerem E, Hatipoglu S. A survey of knowledge, attitudes and practices relating to emergency contraception among health workers in Manisa, Turkey. Midwifery 2006;22(1):66-77. https://doi.org/10.1016/j.midw.2005.03.004 17. Creanga AA, Schwandt HM, Danso KA, Tsui AO. Knowledge about emergency contraception among family-planning providers in urban Ghana. Int J Gynecol Obstet 2011;114(1):64-68. https://doi. org/10.1016/j.ijgo.2011.01.024

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

Abnormal umbilical artery Doppler velocimetry and placental histopathological correlation in fetal growth restriction R Agarwal,1 MS (O&G); A Tiwari,1 MBBS; N Wadhwa,2 MD (Pathology); G Radhakrishnan,1 MS (O&G); S Bhatt,3 MD (Radiology); P Batra,4 MD (Paediatrics) Department of Obstetrics and Gynaecology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi, India Department of Pathology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi, India 3 Department of Radiology, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi, India 4 Department of Paediatrics, University College of Medical Sciences and Guru Teg Bahadur Hospital, Delhi, India 1 2

Corresponding author: R Agarwal (rachna_anila@yahoo.co.in)

Background. Doppler velocimetry (DV) is widely used to assess the vascular formation of the placenta in fetal growth restriction (FGR) and to estimate the haemodynamic condition of the growth-restricted fetus. Umbilical artery (UA) flow is essentially placental, rather than fetal. Hence, DV provides information about the fetal side of the placenta and, alongside placental histopathology, it could possibly help to decipher aetiopathogenesis in FGR cases. Objective. To correlate UA DV findings occurring in FGR with placental findings. Methods. The study was prospective and conducted in a low-income setting. A total of 130 non-anomalous singleton FGR pregnancies (≥24 weeks) were included in the study. All pregnancies were confirmed to be small for gestational age (SGA) after the birth of the neonate. The placental lesions and neonatal outcomes were correlated with DV findings before delivery: 65 cases with normal DV results constituted group 1, and group 2 had 65 cases with abnormal DV results such as reduced flow, absent UA end diastolic flow or reversal of UA end diastolic flow. Results. Group 2 had significantly lower mean (standard deviation) birth weights of 1.59 (0.4) kg v. 1.87 (0.23) kg for group 1 (p<0.001). Considerably higher NICU mortality was seen in group 2 (30.5%) compared with group 1 (6.7%) (p<0.001). The group 2 placentas weighed less, had a higher number of maternal underperfusion (MUP) lesions, higher levels of calcification. Among lesions of MUP, 4 lesions i.e. villous infarction (p<0.001), villous agglutination (p<0.001), syncytial knots (p=0.003) and intervillous fibrin deposition (p=0.001) were present in significantly higher numbers in the abnormal Doppler group compared with the normal Doppler group. Abnormal Doppler had a sensitivity of 80% and specificity of 92.3% for abnormal placental pathology (placental lesions >3). Conclusions. There was a significantly higher number of MUP lesions and neonatal morbidity in SGA patients with abnormal DV findings. S Afr J Obstet Gynaecol 2017;23(1):12-16. DOI:10.7196/SAJOG.2017.v23i1.1109

Fetal development is very closely related to placental development. Primary villus maldevelopment with evidence of reduced placental villus stem arteries and small, fibrotic, hypovascular terminal villi, have been shown in various pathological studies of the placenta of pregnancies complicated by fetal growth restriction (FGR).[1-5] Clinically, many abnormal umbilical artery (UA) waveforms are associated with these pathological findings such as increased Doppler resistance, reduced flow, absent UA end diastolic flow (AEDF) or reversal of UA end diastolic flow (REDF) flow.[4] However, the true clinical implications of these placental findings and Doppler velocimetry (DV) associations are still not well established.[6] We determined whether the abnormal UA DV occurring in FGR patients when correlated with the placental findings could possibly give insight into the aetiopathogenesis of FGR and guide towards early treatment and better neonatal outcomes using DV as a tool.

Methods

The study was conducted in a low-income setting between 2014 and 2016. Prior ethical committee clearance (IEC-UCMS dated 30 October 2014) and patient consent were obtained for the study.

The inclusion criteria for the study was a singleton pregnancy at ≤24 weeks of gestation suspected of having FGR (sonographically estimated fetal weight <10th percentile for the gestational age). All these cases were confirmed postnatally as small for gestational age (SGA). SGA was defined as actual birth weight <10th percentile for that gestational age.[7,8] Pregnancies with an anomalous fetus or associated with any known maternal diseases were excluded. A standard protocol was followed to determine UA Doppler indices for all patients. Ultrasound was performed using a convex transducer at a frequency 3.5 - 5 MHz. The free-floating loop of the umbilical cord was selected for investigation. The angle of insonation was always kept lower than 30°. The pulsatility index (PI) of the UA was determined and any absence or REDF in the UA was also recorded. The patients were divided into two groups based on UA Doppler indices at delivery: Group 1: Normal DV findings (based on PI) in FGR subjects Group 2: Abnormal DV findings (based on PI >95th percentile and/or absent or reversed end diastolic flow) in FGR subjects. The study methodology is detailed in Fig 1. Stained histopathological slides were evaluated by a trained pathologist who was blind to

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH both the clinical and imaging results. Histological lesions were broadly defined according to the nomenclature and the diagnostic criteria proposed by the placental classification provided by Redline.[9] The discrete histopathological findings were grouped into 15 main findings and four major pathological patterns for the purpose of statistical analysis:[10] (i) maternal vascular underperfusion (MUP) including non-marginal, recent, and organised infarction involving >10% of parenchyma, agglutination, placental syncytial (ST) knots, intervillous fibrin deposition involving more than 20% of the intervillous space, peri-villous fibrin deposition, villous hypoplasia, intervillous haematoma or retroplacental haematoma; (ii) fetal vascular underperfusion (FUP) including fetal vasculopathy and/or avascular villi; (iii) inflammatory with villitis, chorioamnionitis or vasculitis lesions; (iv) others (stromal fibrosis and calcification). The placental lesions were then compared with DV findings (at delivery) and prospectively with neonatal outcome.

Statistical analysis Statistical software SPSS version 20.0 (IBM Corp., USA) was used for statistical analysis. Upon histopathological examination, six placental lesions (villous infarction, agglutination, syncytial knots, intervillous fibrin deposition, stromal fibrosis, calcification) were Singleton, non-anomalous pregnancy ≥24 weeks suspected of having FGR, i.e. symphysial fundal height <4 weeks than period of gestation, enrolled as suspected subject (N=160)

Calculation of gestational age by last menstrual period or first trimester ultrasound

Detailed history, obstetric examination and routine antenatal investigations Fetal biometry by ultrasound. Suspected FGR subject with estimated fetal weight <10% for GA admitted. UA Doppler (PI) performed.

Normal UA Doppler results

Abnormal UA Doppler results

PI >95th percentile Repeat Doppler weekly until delivery

AEDF or REDF Repeat Doppler fortnightly until delivery

Management as per hospital protocol

Delivey of FGR subject (confirmed postnatally as SGA) (N=130)

Group 1: FGR with last Doppler normal (n=65)

Group 2: FGR with last Doppler abnormal (n=65)

Delivery details (term ≥37 weeks or preterm <37 weeks) and neonatal outcome noted

Placenta collected and sent for histopathological analysis

Fig 1. Flow diagram depicting study methodology. (*Thirty subjects were excluded because of unconfirmed SGA based on antenatal ultrasound biometry (n=8), unconfirmed SGA postnatally (n=6), inability to collect placental tissue (n=10), and some cases were lost to follow-up (n=6).

found to present in a significantly higher number in group 2. Taking these six lesions into consideration, a receiver operating characteristic (ROC) curve was drawn. The optimal threshold point for the lesions (≥3 placental lesions; classified as abnormal placental pathology) was obtained where the sum of sensitivity and specificity was at a maximum (Youden Index) and the best cut-off point was found. This was further used to calculate the screening potential of UA indices.

Results

Both groups were comparable in terms of age, parity and body mass index characteristics (Table 1). In fetal biometry, the mean (SD) biparietal diameter (group 1: 35.69 (1.93) weeks; group 2: 35.19 (3.32) weeks), mean (SD) femur length (group 1: 34.09 (2.10) weeks; group 2: 33.19 (2.75) weeks and mean (SD) amniotic fluid index (group 1: 6.9 (3.41) cm; group 2: 7.6 (4.33) cm) were nearly same for both groups. There was a significant difference in fetal abdominal circumference and estimated fetal weight between the two groups (Table 1). The findings of the last UA Doppler at delivery showed that all the patients in group 2 had UA PI >95th percentile for their gestation period, 36 had only raised PI, 16 had AEDF, and 13 had REDF. The mean (SD) PI of group 2 was 1.9 (0.56), which was significantly higher than group 1 (0.80 (0.31)). Similarly, the mean systolic:diastolic (S/D) ratio was also significantly higher in group 2, i.e. 5.02 v. 1.99 in group 1 (p<0.001). There were 36 (55.4%) preterm deliveries (<37 weeks) in group 2 v. 24 (36.9%) in group 1 (p=0.054). Term deliveries (≥37 weeks) were more in group 1 (63.0%) than in group 2 (47.0%). There were six intrauterine deaths in group 2. Vaginal delivery was more Table 1. Study group characteristics Characteristics Age (years), mean (SD) Parity, mean (SD) Body mass index (kg/m2), mean (SD) Mean arterial blood pressure (mmHg), mean (SD) Maternal characteristics Pregnancy-induced hypertension Gestational hypertension Pre-eclampsia Eclampsia Ultrasound findings, mean (SD) Abdominal circumference (weeks) Estimated fetal weight (kg) UA Doppler findings Systolic:diastolic ratio PI Abnormal doppler findings Reduced end diastolic flow (PI >95th percentile) AEDF REDF *Unless otherwise specified.

SAJOG • May 2017, Vol. 23, No. 1

Group 1 (n=65), n (%)* 23.89 (2.38) 1.68 (0.86)

Group 2 (n=65), n (%)* 23.42 (2.22) 1.65 (0.84)

22.4 (2.29)

22.32 (1.88)

0.856

84.2 (3.30)

81.3 (2.60)

0.765

16 (24.60)

11 (16.92) 4 (6.15) 1 (1.53)

4 (6.15) 9 (13.84) 3 (4.61)

0.097 0.241 0.619

31.48 (1.90)

30.01 (2.33)

<0.001

1.96 (0.27)

1.67 (0.32)

<0.001

1.99 (0.38) 0.80 (0.31)

5.02 (1.87) 1.90 (0.56)

<0.001 <0.001

36 (55.40)

16 (24.60) 13 (20.00)

p-value 0.240 0.837

RESEARCH commonly seen in group 1 (n=56; 86.20%) v. group 2 (n=29; 44.60%). Among the subjects in group 2, 55.40% (n=36) underwent caesarean deliveries, compared with 13.80% (n=9) in group 1 (p<0.001). Group 2 had a significantly lower mean (SD) birth weight of 1.59 (0.40) kg v. 1.87 (0.23) kg for group 1 (p<0.001) (Table 2).

Considerably higher neonatal intensive care unit (NICU) mortality was seen in 30.50% of babies in group 2, compared with the 6.70% observed in group 1 (p<0.001). The placental histopathological findings are detailed in Table 3. A significantly lower mean (SD) placental weight in group 2

Table 2. Neonatal outcomes of study population Group 2 (n=59)†

Outcomes Birth weight (kg), mean (SD) Apgar score <7 (at 5 min) NICU admission NICU stay (days) Neonatal mortality Complications Respiratory distress syndrome Sepsis Meconium aspiration syndrome Healthy at discharge

Group 1 (n=65), n (%)* 1.87 (0.23) 12 14 (21.5) 3.1 4 (6.7) 14 (21.5) 8 (12.3) 4 (6.1) 2 (3.0) 61 (95.3)

Group 2 (n=59)†, n(%)* 1.59 (0.40) 31 39 (60.0) 5.6 18 (30.5) 49 (83.1) 23 (38.9) 14 (23.7) 2 (3.3) 41 (69.5)

p-value <0.001 0.003 <0.001 <0.001 <0.001

<0.001

Low EDF (n=36), n (%)* 1.77 15 18 (50.0) 3.8 8 (22.2)

AEDF (n=14), n (%)* 1.85 11 13 (92.8) 4.1 3 (21.4)

REDF (n=9), n (%)* 1.14 5 8 (88.8) 4.9 7 (77.0)

8 8 1 28 (77.7)

8 3 1 11 (78.6)

7 3 0 2 (22.2)

*Unless otherwise specified. † Six intrauterine deaths.

Table 3. Comparison of placental histopathological findings in normal (group 1) and abnormal (group 2) Doppler group (N=130)

Outcomes Placental weight (g), mean (SD) Gross examination Haemorrhage Necrosis Calcification Microscopic examination Villous infarction Agglutination ST knots <20% >20% MUP Intervillous fibrin deposition Perivillous fibrin deposition Villous hypoplasia Intervillous haematoma Retroplacental haematoma

Group 1 (n=65), n (%)* 311.6 (67.5)

Group 2 (n=65), n (%)* 228.9 (77.5)

p-value <0.001

Low EDF (n=36) -

Group 2 (n=65) AEDF (n=16) -

8 (12.30) 9 (13.80) 13 (20.00)

19 (29.20) 21 (32.30) 44 (67.70)

0.100 0.007 <0.001

0 4 (6.15)

13 (20.00) 36 (55.38)

<0.001 <0.001 0.003

8 21

2 8

3 7

51 (78.46) 5 (7.69)

36 (55.38) 20 (30.77)

18 13

8 6

10 1

4 (6.15) 2 (3.08) 0 0 1 (1.53)

38 (58.46) 2 (3.08) 2 (3.08) 1 (1.53) 6 (9.23)

<0.001 1 0.496 1 0.115

21 2 0 1 5

9 0 1 0 1

8 0 1 0 0

1 (1.53) 3 (4.60)

2 (3.08) 5 (7.70)

1 0.718

0 2

1 2

1 1

1 (1.53) 5 (7.69) 1 (1.53)

3 (4.61) 9 (13.84) 6 (9.23)

0.619 0.258 0.115

3 8 0

0 1 0

0 0 0

2 (3.08) 26 (40.00) 19 (29.23) 7 (10.77)

24 (36.92) 46 (70.00) 28 (43.08) 18 (27.69)

<0.001 0.001 -

12 0 15 15

7 0 6 1

5 0 7 2

REDF (n=13) -

Fetal underperfusion (FUP) Fetal vasculopathy Avascular villi Inflammatory lesions Villitis Chorioamnionitis Vasculitis Others Stromal fibrosis Calcification <10% >10% *Unless otherwise specified.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH (228.9 (77.5) g) was found compared with group 1 (311.6 (67.5) g). On gross examination, areas of necrosis and calcification were present in significantly higher numbers in the placentas of group 2 compared with those of group 1 (p=0.007 and p<0.001, respectively). Among the lesions signifying MUP, four were found to occur significantly more in the placentas of group 2. These were villous infarction (20% v. 0%), villous agglutination (55.38% v. 6.20%), intervillous fibrin deposition (55.4% vs 6.1%) and ST knots (30.77% v. 7.69%). Among the others histopathological category, stromal fibrosis and calcification were significantly higher in group 2. No significant difference was seen between lesions of fetal vascular underperfusion (FUP) and lesions of inflammatory origin. An ROC curve based on the above 6 placental lesions (4 MUP and 2 others) was plotted and the best cut-off point was found to be >3 placental lesions (Youden Index), which was used as criteria for our study (Table 4 and Fig. 2). The area under the curve (AUC) was 0.898 (95% CI 0.842 - 0.955). There was a total of 57 patients in our study with abnormal placental pathology. Fifty-two patients in group 2 had placental pathologies listed above compared with just five patients in group 1 (p<0.001). Further, based on the ROC curve, abnormal Doppler (reduced flow, AEDF or REDF) had a sensitivity of 80% and specificity of 92.3% as a test for detection of abnormal placental pathology (placental lesions ≥3) in SGA pregnancies. Table 4. Distribution of significant placental histopathological lesions between groups 1 and 2 0 7 1 8

Group 1 Group 2 Total

Number of significant histopathological lesions 1 2 3 4 5 6 30 23 4 1 0 0 3 9 27 18 4 3 33 32 31 19 4 3

1.0

0.8

Sensitivity

0.6

Diagnostic test Chance level

0.4

0.2

0.0 0.0

0.2

0.4

0.6

0.8

1.0

1 – Specificity

Fig. 2. ROC curve for six placental lesions. The area under the curve was 0.898 at 95% CI 0.84 - 0.96. The optimal threshold point for the lesions was obtained where the sum of sensitivity and specificity was at a maximum (Youden Index) and the best cut-off point was found to be ≥3 placental lesions.

Discussion

FGR is still a significant problem despite extensive research into its aetiology, pathogenesis and management.[8] Abnormal UA DV in pregnancies complicated by FGR has been associated with a heterogeneous, nonspecific group of histological lesions of the placenta irrespective of gestational age or FGR aetiology.[2,6,10-14] The exact pathophysiological mechanisms correlating these are not well characterised.[15] Various authors have used different placental histopathological classifications, categorisations of placental lesions, Doppler indices and statistical methods making interpretation and inter-study comparison difficult.[2,6,10-14] Dicke et al.[11] retrospectively investigated the screening efficacy of UA S/D, PI, and AEDF for adverse pregnancy outcomes and placental abnormalities in SGA fetuses. Ninety-four percent of cases with either an S/D or a PI above the 95th percentile, or AEDF, had evidence of placental pathology. They found a higher number of MUP lesions where there was an abnormal Doppler (50%) compared with a normal Doppler (27.6%). Fetal vascular obstructive lesions were lower than MUP lesions, i.e. 27% in group 2 v. 14.9% in group 1. They concluded that abnormal Doppler indices predicted placental lesions. The screening UA Doppler parameters for placental abnormalities in their study had a sensitivity of 42.1% and specificity of 89.3%. We based our analysis on the robust placental classification provided by Redline.[9] In our study, out of the 15 placental lesions investigated, significant differences were noted for 6 placental lesions in the two DV groups (Table 3). These findings supported abnormal utero-placental circulation especially of maternal aetiology. Many other studies have similarly pointed towards a predominant MUP origin for FGR.[6,10,11] Our study derived a much higher sensitivity (80%) and specificity (92.3%) for abnormal Doppler (reduced flow, AEDF or REDF) as a screening tool for detection of abnormal placental pathology. Vedmedovska et al.[6] found major placental microscopic lesions in 50 women with prenatally suspected, and post-delivery confirmed, FGR. They found 56.6% subjects with abnormal DV having >4 placental lesions compared with 25% in the normal Doppler group (p=0.003). The percentage of villous infarction was 48.3% v. 25%, intervillous haematoma 65.5% v. 35%, thickening of the basement membrane 58.6% v. 45%, perivillous fibrin deposition 86.2% v. 80%, stromal fibrosis 37.9% v. 10% in the abnormal v. normal Doppler group. Compared with the control group with normal Doppler, FGR women with abnormal DV had more villous infarctions (p=0.0003), and intervillous thrombi (p<0.0001). Thickening of the basal membrane and villitis were also linked to FGR. In our study abnormal Doppler FGR had significantly higher numbers of villous infarction, a marker of MUP. Moreover, the optimal threshold for classification of abnormal placental pathology was detected at ≥3 placental lesions, similar to the aforementioned study by Vedmedovska et al.[6] Spinillo et al. [10] followed 126 FGR pregnancies for defined placental lesions and correlated findings with UA Doppler velocimetry. They found the PI to be normal in 45 (35.7%) and increased in 44 (34.9%) of the women. End-diastolic UA Doppler flow was absent in 27 (21.4%) and reversed in 10 (7.9%) cases. Increased placental intervillous fibrin deposits, villous hypoplasia, ST knots, giant cells, and immature intermediate trophoblasts were directly related to worsening of UA Doppler results. They also found villous hypoplasia (p=0.031), trophoblast giant cell (p=0.015),

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH and immature intermediate trophoblasts (p=0.014) directly related to worsening of UA Doppler measurements, a finding that was not replicated in our study. Another significant difference was a higher frequency of villous agglutination (p=0.001) in abnormal Doppler subjects in our study compared with theirs (p=0.096). In a more recent study by Parra-Saavedra et al,[12] 126 fetuses with normal UA Doppler indices that were delivered after 34 weeks’ gestation were studied. Among 97 histological findings consistent with placental underperfusion in 84 SGA pregnancies, maternal vascular supply placental injuries were far more in number (79.4%) when compared with fetal vascular supply placental lesions (20.6%). The authors concluded that the presence of histological signs of placental underperfusion implies a poorer neonatal outcome. In our study too, the rate of preterm deliveries, neonatal complications and NICU stay was more pronounced in group 2 with abnormal DV (Table 3). There are many limitations of our study, which we acknowledge. We evaluated only UA indices. Uterine artery, middle cerebral artery and ductus venosus Doppler, which has been evaluated by many others, were not utilised in our investigations.[12,14] Only UA Doppler PI values were used for comparison with placental lesions. The S/D ratio and Resistance Index, although studied, were not used for final analysis; however, the study had a prospective study design and matched SGA subjects for both normal and abnormal Doppler findings. The histopathological blinding and controlled Doppler group ensured precise and unbiased results. The clustering of specific placental lesions into the abnormal Doppler group raised support for MUP as the main aetiology for FGR. The UA Doppler has good screening ability, with a sensitivity of 80% and a specificity of 92.3%, for detection of MUP placental lesions as inferred from our study, making it a potential tool to identify FGR and in turn improve perinatal outcomes through intensive sonographic fetal surveillance and more favourable timing of delivery. The poorer neonatal outcome reported with abnormal Doppler indices/MUP lesions calls for even more stringent clinical supervision of these lesions.

Conclusion

There was a higher number of MUP placental histopathological lesions in FGR patients with abnormal Doppler findings supporting

a maternal aetiology for FGR. With abnormal placental pathology defined as n≥3 placental lesions, these abnormal Doppler findings had a sensitivity of 80.0% and a specificity of 92.3% for predicting abnormal placental pathology in SGA. Furthermore, the abnormal Doppler group had a significantly higher neonatal morbidity. Conflict of interest. The authors declare that they have no conflict of interest.

1. Macara L, Kingdom JCP, Kohnen G, Bowman AW, Greer IA, Kaufman P. Elaboration of stem villus vessels in growth restricted pregnancies with abnormal umbilical artery Doppler waveforms. Br J Obstet Gynaecol 1995;102(10):807-812. https://doi.org/10.1111/j.1471-0528.1995.tb10847.x 2. Macara L, Kingdom JCP, Kaufmann P, et al. Structural analysis of placental terminal villi from growth-restricted pregnancies with abnormal umbilical artery Doppler waveforms. Placenta 1996;17(1):37-48. https://doi.org/10.1016/s0143-4004(05)80642-3 3. Sebire NJ, Goldin RD, Regan L. Histomorphological evidence for chronic vasoconstriction of placental stem vessels in pregnancies with intrauterine growth restriction and abnormal umbilical artery velocimetry indices. J Pathol 2001;195:19A. 4. Apel-Sarid L, Levy A, Holcberg G, Sheiner E. Term and preterm (<34 and <37 weeks, gestation) placental pathologies associated with fetal growth restriction. Arch Gynecol Obstet 2010;282(5):487-492. https://doi.org/10.1007/s00404-009-1255-1 5. Tomas SZ, Roje D, Prusac IK, Tadin I, Capkun V. Morphological characteristics of placentas associated with idiopathic intrauterine growth retardation: A clinicopathologic study. Eur J Obstet Gynecol Reprod Biol 2010;152(1):39-43. https://doi.org/10.1016/j.ejogrb.2010.05.006 6. Vedmedovska N, Rezeberga D, Teibe U, Melderis I, Donders GG. Microscopic lesions of placenta and Doppler velocimetry related to fetal growth restriction. Arch Gynecol Obstet 2011;284(5):1087-1093. https://doi.org/10.1007/s00404-010-1781-x 7. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996;87(2):163-168. https://doi.org/10.1016/0029-7844(95)00386-x 8. Lubchenco LO, Hansman C, Dressler M, Boyd E. Intrauterine growth as estimated from live born birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963:32:793-800. 9. Redline RW. Placental pathology: A systematic approach with clinical correlations. Placenta 2008;29(Suppl A):S86-91. https://doi.org/10.1016/j.placenta.2007.09.003 10. Spinillo A, Gardella B, Bariselli S, Alfei A, Silini E, Dal Bello B. Placental histopathological correlates of umbilical artery Doppler velocimetry in pregnancies complicated by fetal growth restriction. Prenat Diagn 2012;32 (13):1263-1272. https://doi.org/10.1002/pd.3988 11. Dicke JM, Huettner P, Yan S, Odibo A, Kraus FT. Umbilical artery Doppler indices in small for gestational age fetuses: Correlation with adverse outcomes and placental abnormalities. J Ultrasound Med 2009;28(12):1603-1610. https://doi.org/10.7863/jum.2009.28.12.1603 12. Parra-Saavedra M, Crovetto F, Triunfo S, Savchev S, Peguero A, Nadal A. Placental findings in late-onset SGA births without Doppler signs of placental insufficiency. Placenta 2013;34(12):11361141. https://doi.org/10.1016/j.placenta.2013.09.018 13. Vedmedovska N, Rezeberga D, Teibe U, Melderis I, Donders GG. Placental pathology in fetal growth restriction. Eur J Obstet Gynecol Reprod Biol 2011;155(1):36-40. https://doi.org/10.1016/j. ejogrb.2010.11.017 14. Madazli R, Somunkiran A, Calay Z, Ilvan S, Aksu MF. Histomorphology of the placenta and the placental bed of growth restricted fetuses and correlation with the Doppler velocimetries of the uterine and umbilical arteries. Placenta 2003;24(5):510-516. 15. Sebire NJ. Umbilical artery Doppler revisited: Pathophysiology of changes in intrauterine growth restriction revealed. Ultrasound Obstet Gynecol 2003;21(5):419-422. https://doi.org/10.1002/ uog.133

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

Preventive role of low-molecular-weight heparin in unexplained recurrent pregnancy loss E S Khan,1 MBBS, FCPS (O&G); A Basharat,2 MBBS, FCPS (O&G); M Jamil,3 MBBS, FCPS (O&G); S Ayub,4 MBBS, FCPS (O&G); M A Khan,5 MBBS, MSc, FCPS (Nuclear Med) Department of Obstetrics and Gynecology, Agha Khan University, Karachi, Pakistan Department of Obstetrics and Gynecology, Alnafees Medical College, ISRA University, Islamabad, Pakistan 3 Department of Obstetrics and Gynecology, Shifa College of Medicine, Islamabad, Pakistan 4 Department of Pathology, Rawalpindi Medical College, Rawalpindi, Pakistan 5 Department of Nuclear Medicine, Nuclear Medicine, Oncology and Radiotherapy Institute (NORI), Islamabad, Pakistan 1 2

Corresponding author: M A Khan (draleemkhan@yahoo.com)

Background. Recurrent pregnancy loss (RPL) is a source of great distress for couples, and the search continues for an intervention to improve live birth rates in affected women. A daily injection of low-molecular-weight heparin (LMWH) is often prescribed to women with unexplained RPL, although evidence suggesting a benefit is limited. Objective. To compare the efficacy of LMWH with a placebo in terms of live birth rates in women with unexplained RPL. Methods. All pregnant females between 18 and 44 years of age who reported at the unit of obstetrics and gynaecology, Shifa International Hospital, Islamabad, during April 2013 to January 2014, who had a history of ≥2 consecutive first trimester pregnancy losses were enrolled. All participants were randomly allocated to one of two groups. Group A received a daily dose of 40 mg enoxaparin (LMWH) subcutaneously and group B women received a placebo in the form of multivitamin tablets. Efficacy was defined in terms of live births after the age of viability (i.e. 24 weeks’ gestation) and was compared in both treatment and control groups. Risk estimation was also performed and relative risk (RR) along with 95% confidence interval (CI) was calculated. Results. The groups were similar in terms of mean age, gestational age and body mass index. Our results showed no statistically significant difference in live birth rates between the two groups, with 78.8% and 73.8% for group A and B, respectively (p=0.0574). A RR of 1.07 (95% CI 0.9 - 1.3) was calculated for group A. Conclusion. Subcutaneous enoxaparin in a once daily dose of 40 mg did not improve the chance of live births in nonthrombophilic women with unexplained RPL when compared with the placebo. S Afr J Obstet Gynaecol 2017;23(1):17-19. DOI: 10.7196/SAJOG.2017.v23i1.1112

Recurrent pregnancy loss (RPL) is one of the commonly encountered complications in reproductive medicine, as the aetiology is often unknown and there are few evidence-based diagnostic and treatment strategies available. There are varying definitions of RPL reported that include ≥2 failed clinical pregnancies as documented by ultrasonography or histopathological examination;[1] or 3 consecutive pregnancy losses, which are not required to be intrauterine.[2] The definition of RPL also includes non-visualised pregnancy losses (biochemical pregnancy losses and/or pregnancies of unknown location) as they have the same negative impact on future live birth rate as intrauterine pregnancy losses.[3] The European Society of Human Reproduction and Embryology released a 2014 consensus statement proposing that RPL describes repeated pregnancy loss regardless of anatomic location. They did not recommend the number of losses required for the problem to be defined as recurrent. However, the statement advised researchers and clinicians to clearly describe the type of pregnancy loss, gestational age, number of prior pregnancy losses and relevant details of ultrasound measurements.[4] There is a consensus that healthy women should not undergo extensive evaluation after a single first trimester or early second trimester spontaneous miscarriage (up to 20 weeks),

as these are relatively common sporadic events that occur in ~10 - 15% of clinically recognised pregnancies under 20 weeks’ gestation. The overall risk of miscarriage in the next pregnancy remains about 15% after one miscarriage, but rises to 17 - 31% after two consecutive miscarriages and to 25 - 46% after three miscarriages.[5] Based on these and similar data, usual clinical practice is to initiate evaluation and treatment of RPL after two consecutive miscarriages.[5] RPL remains unexplained in ~50% of couples after evaluation; however, the chance of a live birth is good (>50%) with no intervention and this must be considered in evaluating therapies for unexplained RPL.[6] Although women with a history of RPL who become pregnant may be at a higher risk for developing fetal growth restriction and premature delivery, the detection of fetal cardiac activity in early pregnancy is reassuring of subsequent viable delivery.[7] A variety of treatments have been offered to couples with unexplained RPL, including lifestyle modification, progesterone therapy, human menopausal gonadotropin, in vitro fertilisation and preimplantation genetic diagnosis.[8] Anticoagulants have been studied in this context for a number of years. Different thrombophilia polymorphisms have been identified with RPL and intervention with thromboprophylaxis as preventive therapy has been proposed.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH However, the evidence is still insufficient for routine clinical use of anticoagulants in such women.[9] RPL is a source of great distress for couples and a search for an intervention that could result in improved live birth rates in these women is ongoing. Therapeutic recommendations are largely based on clinical experience and data from observational studies. The objective of this study was to compare the efficacy of lowmolecular-weight heparin (LMWH) with a placebo in women with unexplained RPL, to generate evidence-based therapeutic recommendations for these patients.

Methods

This randomised controlled trial was carried out at the Obstetrics and Gynaecology Department of Shifa International Hospital, Islamabad, Pakistan, between April 2013 and January 2014. The study was approved by the hospital ethics committee and a fully informed consent was obtained from participants. During this period, females between 18 and 44 years of age who had a history of ≥2 consecutive first trimester pregnancy losses were enrolled and assessed with adequate history, thorough clinical examination and necessary investigations. Pregnancy was confirmed by the quantitative beta human chorionic gonadotropin (βHCG) test and by ultrasound confirming fetal heart activity. Patients who presented with systemic lupus erythematosus, positive IgG and IgM anticardiolipin antibodies, positive for thrombophilia screening, any platelet function abnormality or a previous thromboembolic event requiring anticoagulant therapy (including heparin, aspirin or warfarin) and having sensitivity to acetylsalicylic acid, heparin or warfarin obtained by self-report, were excluded from the study. Patients who had any genetic cause (identified by karyotype analysis of both partners), anatomical (identified on hysterosalpingogram or sonohysterogram) or hormonal cause (identified by mid-luteal phase evaluation of progesterone, prolactin and thyroid-stimulating hormone) of RPL were also excluded from the study. All the patients were randomly allocated to one of two groups by the sealed envelope method. Group A women received a daily dose of 40 mg LMWH subcutaneously and group B women received the placebo in the form of multivitamin tablets. Women in both groups were evaluated in the antenatal clinic every 6 weeks until delivery or pregnancy loss. Efficacy was defined in terms of live births after the age of viability, i.e. 24 weeks’ gestation and was compared in both treatment and control groups by employing a χ2 test. A p-value of <0.05 was considered as statistically significant. Risk estimation was also performed and relative risk (RR) with a 95% confidence interval (CI) was calculated.

Results

A total of 160 pregnant women were enrolled and equally dis-tributed into groups A (LMWH) and B (placebo). The groups were similar in terms of mean age, gestational age and body mass index (BMI) (Table 1). In group A, 78.8% (n=63) completed the pregnancy with successful live births, while the percentage was slightly lower at 73.8% (n=59) in group B. However, the difference was not statistically significant (p=0.574) (Table 2). An RR of 1.07 (CI 0.9 - 1.3) was calculated for the live birth rate of 78.8% in group A (Table 3).

Discussion

RPL affects 1 - 2% of women. In >50% of such cases the cause remains unknown. High-quality data on the management of RPL are limited and reported studies on the aetiology, evaluation

Table 1. Demographic characteristics of groups A and B (N=160) Variables, mean (SD)

A (LMWH)

B (placebo)

Age (years)

25.9 (4.6)

26.0 (3.4)

Gestational age at enrolment (weeks)

8.2 (0.5)

8.1 (0.3)

BMI (kg/m2)

29.9 (3.5)

29.1 (2.7)

Table 2. Comparison of efficacy in terms of live births between both groups (N=160) Live birth, n (%) p-value

Group

Yes

A (n=80)

63 (78.8)

17 (21.3)

B (n=80)

59 (73.8))

21 (26.3)

Total

122 (76.3)

38 (23.7)

0.574

Table 3. Risk estimates Estimates

Value

95% CI

OR (for group A/B)

1.319

0.635 - 2.741

RR (for live birth = yes)

1.068

0.898 - 1.270

RR (for live birth = no)

0.810

0.463 - 1.416

and management of RPL are mostly observational. For these reasons, therapeutic recommendations are largely based on clinical experience and data from these observational studies.[10] Nevertheless, the prognosis for a successful future pregnancy is generally good: the overall live birth rates after normal and abnormal diagnostic evaluations for RPL are 77% and 71%, respectively.[1-3] Thrombosis of spiral arteries and the intervillous space on the maternal side of the placenta can impair adequate placental perfusion. The resulting abnormalities of the uteroplacental circulation may cause late fetal loss, intrauterine growth restriction, placental abruption or preeclampsia. A relationship with early pregnancy loss is less clear and may be restricted to specific thrombophilic defects; literature on the association between maternal inherited thrombophilia and RPL occurring in the first trimester is contradictory. A systematic review of the association between fibrinolytic defects and RPL found a significant association for factor XII deficiency (odds ratio (OR) 18.11, CI 5.52 - 59.4).[11] Procoagulant microparticles may also contribute to the hypercoagulable state in these women and thus may interfere with successful implantation and fetal growth.[12] Thrombophilia has been identified in ~50% of women with RPL and thromboprophylaxis has been suggested as a treatment option.[1,2] It has been reported in the literature that anticoagulation of women with certain inherited thrombophilias may improve maternal outcome (e.g. prevention of venous thromboembolism), but does not appear to prevent pregnancy loss.[13] We explored the effects of LMWH (subcutaneous enoxaparin) in our settings. The role of LMWH in preventing pregnancy losses is based on the principle that placental circulation can become compromised in women with underlying prothrombotic tendencies, and administration of LMWH may be able to reduce this tendency, resulting in better fetal outcomes.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH In our study, the live birth rate in the placebo group was slightly lower than in the intervention group (73.8% v. 78.8%, respectively). However, the difference was not statistically significant (p=0.574). Our results are similar to those in studies reported in the literature. In a similar study, Pasquier et al.[14] enrolled 258 pregnant women with a history of unexplained recurrent miscarriage (≥2 consecutive miscarriages before 15 weeks’ gestation) and a negative thrombophilia screen. They were randomly assigned to receive one daily subcutaneous injection of 40 mg enoxaparin or placebo until 35 weeks’ gestation. Their results indicated that 66.6% of the women who received enoxaparin had a live birth v. 72.9% of women who received the placebo (p>0.05). They concluded that enoxaparin (40 mg once daily) did not improve the chance of a live birth in nonthrombophilic women with unexplained RPL.[14] In another similar multicentre, randomised controlled trial, Schleussner et al.[15] determined whether LMWH increases live birth rates in women with unexplained RPL. They enrolled 449 women, with at least 2 early or 1 late miscarriages, at 5 - 8 weeks’ gestation after confirmation of a viable pregnancy by ultrasonography. Women in the control group received multivitamin tablets, and the intervention group received vitamins and 5 000 IU of dalteparin-sodium for up to 24 weeks’ gestation. The live birth rates were 86.0% and 86.7% in the intervention and control groups, respectively. They concluded that daily LMWH injections did not increase ongoing pregnancy or live birthrates in women with unexplained RPL. A 2014 Cochrane review including nine trials reported similar results, after evaluating the effects of either LMWH or aspirin, or a combination of the two, on women with RPL or without inherited thrombophilia. The authors found no evidence of an increased frequency of live birth among both groups.[16] Visser et al.,[17] in their randomised, double-blind, multicentre trial, randomly allocated 207 women with ≥3 consecutive first trimester miscarriages into one of three groups: 40 mg enoxaparin and placebo (n=68), 40 mg enoxaparin and 100 mg aspirin (n=63) or 100 mg aspirin (n=76). They found a live birth rate of 71% (RR 1.17; CI 0.92 - 1.48) for enoxaparin and placebo and 65% (RR 1.08; CI 0.83 - 1.39) for enoxaparin and aspirin when compared with aspirin alone (61%). The difference was not statistically significant. In summary, based on the present study results and other similar studies reported in the literature, the evidence is still not sufficient to recommend routine use of anticoagulants in women with unexplained RPL without inherited thrombophilia. Therapeutic intervention for RPL is guided by the underlying cause, and in all cases emotional

support is important in caring for these often-anxious couples, which may enhance therapeutic success. There are limitations in the present study: firstly, placebo injections were not used and secondly, the sample size was relatively small.

Conclusion

Subcutaneous enoxaparin in a dose of 40 mg once daily did not improve the chance of live births in non-thrombophilic women with unexplained recurrent pregnancy loss when compared with the placebo. 1. Practice Committee of American Society for Reproductive Medicine. Definitions of infertility and recurrent pregnancy loss: A committee opinion. Fertil Steril 2013;99(1):63. https://doi. org/10.1016/j.fertnstert.2012.09.023 2. Jauniaux E, Farquharson RG, Christiansen OB, Exalto N. Evidence-based guidelines for the investigation and medical treatment of recurrent miscarriage. Hum Reprod 2006;21(9):2216-2222. https://doi.org/10.1093/humrep/del150 3. Kolte AM, van Oppenraaij RH, Quenby S, et al. Non-visualized pregnancy losses are prognostically important for unexplained recurrent miscarriage. Hum Reprod 2014;29(5):931-937. https://doi. org/10.1093/humrep/deu042 4. Kolte AM, Bernardi LA, Christiansen OB, et al. Terminology for pregnancy loss prior to viability: A consensus statement from the ESHRE early pregnancy special interest group. Hum Reprod 2015;30:495-498. https://doi.org/10.1093/humrep/deu299 5. Jaslow CR, Carney JL, Kutteh WH. Diagnostic factors identified in 1 020 women with two versus three or more recurrent pregnancy losses. Fertil Steril 2010;93(4):1234-1243. https://doi. org/10.1016/j.fertnstert.2009.01.166 6. Practice Committee of the American Society for Reproductive Medicine. Current clinical irrelevance of luteal phase deficiency: A committee opinion. Fertil Steril 2015;103(4):e27-e32. https://doi.org/10.1016/j.fertnstert.2014.12.128 7. Field K, Murphy DJ. Perinatal outcomes in a subsequent pregnancy among women who have experienced recurrent miscarriage: A retrospective cohort study. Hum Reprod 2015;30(5):12391245. https://doi.org/10.1093/humrep/dev044 8. Coomarasamy A, Williams H, Truchanowicz E, et al. A randomized trial of progesterone in women with recurrent miscarriages. N Engl J Med 2015;373(22):2141-2148. https://doi.org/10.1056/ nejmoa1504927 9. Mumford SL, Silver RM, Sjaarda LA, et al. Expanded findings from a randomized controlled trial of preconception low-dose aspirin and pregnancy loss. Hum Reprod 2016;31(3):657-665. https:// doi.org/10.1093/humrep/dev329 10. Christiansen OB, Nybo Andersen AM, Bosch E, et al. Evidence-based investigations and treatments of recurrent pregnancy loss. Fertil Steril 2005;83(4):821-839. https://doi.org/10.1016/j. fertnstert.2004.12.018 11. Sotiriadis A, Makrigiannakis A, Stefos T. Fibrinolytic defects and recurrent miscarriage: A systematic review and meta-analysis. Obstet Gynecol 2007;109(5):1146-1155. https://doi. org/10.1097/01.aog.0000260873.94196.d6 12. Laude I, Rongières-Bertrand C, Boyer-Neumann C. Circulating procoagulant microparticles in women with unexplained pregnancy loss: A new insight. Thromb Haemost 2001;85(1):18-21. 13. Abheiden C, Van Hoorn ME, Hague WM, et al. Does low-molecular-weight heparin influence fetal growth or uterine and umbilical arterial Doppler in women with a history of early-onset uteroplacental insufficiency and an inheritable thrombophilia? Secondary randomised controlled trial results. BJOG 2016;123(5):797-805. https://doi.org/10.1111/1471-0528.13421 14. Pasquier E, de Saint Martin L, Bohec C, et al. Enoxaparin for prevention of unexplained recurrent miscarriage: A multicenter randomized double-blind placebo-controlled trial. Blood 2015;125(14):2200-2205. https://doi.org/10.1182/blood-2014-11-610857 15. Schleussner E, Kamin G, Seliger G, et al. Low-molecular-weight heparin for women with unexplained recurrent pregnancy loss: A multicenter trial with a minimization randomization scheme. Ann Intern Med 2015;162(9):601-609. https://doi.org/10.7326/m14-2062 16. De Jong PG, Kaandorp S, Di Nisio M, et al. Aspirin and/or heparin for women with unexplained recurrent miscarriage with or without inherited thrombophilia. Cochrane Database Syst Rev 2014;CD004734. https://doi.org/10.1002/14651858.cd004734.pub4 17. Visser J, Ulander VM, Helmerhorst FM, et al. Thromboprophylaxis for recurrent miscarriage in women with or without thrombophilia. HABENOX: A randomized multicentre trial. Thromb Haemost 2011;105(2):295-301. https://doi.org/10.1160/th10-05-0334

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

A cross-sectional descriptive study of breastfeeding behaviour and galactogogue use among private-sector patients in Cape Town, South Africa N Steyn,1 BOcc, MB ChB; M Zunza,1 BComm (Applied Statistics), MSc (Clinical Epidemiology), PhD (Paediatrics); E H Decloedt,2 MB ChB, BSc Pharmacol (Hons), MMed, FCCP (SA) 1 2

Faculty of Medicine and Health Sciences, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa Division of Clinical Pharmacology, Department of Medicine, Stellenbosch University and Tygerberg Academic Hospital, Cape Town, South Africa

Corresponding author: E H Decloedt (ericdecloedt@sun.ac.za)

Background. Failure to exclusively breastfeed is often caused by a perception of insufficient breastmilk supply. Galactogogues are frequently prescribed in these circumstances, but this is supported by sparse scientific data with safety concerns for both mother and infant. The exact extent of galactogogue use in South Africa (SA) is not well known. Objectives. To assess breastfeeding behaviour, galactogogue use and perceived galactogogue side-effects among patients attending International Board Certified Lactation Consultant (IBCLC) private practices. Methods. We administered a self-developed, expert-reviewed questionnaire in five IBCLC private practices within the Cape Town Metropole, SA. All patients attending the practices during an 8-week period were invited to participate. Results. Data from 104 participants were included in the study. An exclusive breastfeeding rate of slightly more than 50%, associated with greater parity (p=0.029), was found. Perceived lack of breastmilk predicted galactogogue use (p=0.013). There was a high prevalence of galactogogue use (54%), with 80% of these participants using non-prescription medication. Sulpiride was the most common prescription medication used. Increased milk production was reported by 41% (n=23) of galactogogue users, while 30% (n=17) reported no effect. Most reported side-effects were minor. Conclusions. Prevalence of galactogogue use exceeded other published data. Sulpiride was most frequently prescribed, despite not being recommended during breastfeeding. A large group of participants reported poor efficacy. The effect of vaginal delivery and immediate skin-to-skin contact after delivery on milk production might be smaller than previously reported in mothers who are personally motivated to breastfeed. Healthcare practitioners should acknowledge breastfeeding mothers’ concerns regarding insufficient milk supply and emphasise correct breastfeeding technique. S Afr J Obstet Gynaecol 2017;23(1):20-23. DOI:10.7196/SAJOG.2017.v23i1.1116

Exclusive breastmilk is the optimal feeding for infants up to 6 months of age.[1] Breastfeeding has been associated with short-term benefits such as providing the child’s first immunity and reducing infant morbidity and mortality associated with infections.[2] Long-term benefits include lower mean blood pressure and total cholesterol in adulthood, and improved performance in intelligence tests.[2] However, mothers struggle to achieve exclusive breastfeeding and according the World Health Organization (WHO)᾽s global database on infant and young child feeding the rate of exclusive breastfeeding in South Africa (SA) is among the lowest in the world, at 11.6% in infants <4 months and 8.3% in infants ≤6 months of age, compared with a global rate of 38%.[3] There are many contributing factors leading to low breastfeeding rates, but the most common reason is insufficient breastmilk supply.[4] Galactogogues are often prescribed to augment breastmilk supply in resource-rich environments.[5] Galactogogues are dopamine antagonists such as antipsychotics, antiemetics and natural supplements which enhance lactation by increasing serum prolactin.[6] Galactogoguemediated rise in serum prolactin may transiently increase milk production, but after 2 weeks postpartum it has no effect on lactational performance.[5,7] A recent Cochrane review concluded that the use of galactogogues was not associated with a significant improvement in longer-term outcomes of breastfeeding in preterm infants.[1] In addition, galactogogues have safety concerns in the mother and the infant.

The antipsychotics sulpiride and chlorpromazine cross into breastmilk, while chlorpromazine may adversely affect the developing central nervous system of the infant.[4,8] In turn, mothers may develop movement disorders such as acute dystonic reactions.[4] Antiemetics exploited for their blocking of dopamine receptors also have safety concerns.[8] Metoclopramide is structurally related to the antipsychotic sulpiride and not recommended during breastfeeding.[8] Domperidone has been associated with an increased risk of QT interval prolongation and sudden cardiac death.[5] Natural supplements such as milk thistle and fenugreek have limited safety data during lactation (DRUG-REAX Database, Truven Health Analytics, Inc). Taken together, the exact extent of galactogogue use in SA is unknown and the awareness of the benefit-risk of galactogogue use by breastfeeding mothers has not been documented. The objectives of this study were to describe breastfeeding behaviour, assess galactogogue use and determine perceived efficacy and side-effects in breastfeeding mothers and their infants.

Methods

The investigators developed a questionnaire based on specific research gaps identified in the literature, followed by review from relevant experts in paediatrics, neonatology and breastfeeding consulting. The questionnaire contained three sections. The

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH first section captured demographic information including factors previously shown to influence breastfeeding, such as type of delivery, parity and having skin-to-skin contact. It also assessed duration of breastfeeding, mixed feeding and reasons for not exclusively breastfeeding. The second section assessed galactogogue use, the type, duration, dose, who recommended it and what the perceived efficacy and duration of the effect were. The last section captured data regarding side-effects in the mother and infant (Table 1). According to the information gathered, participants were retrospectively classified into one of the following groups: exclusive breastfeeding, exclusive formula feeding, mixed feeding (breastfeeding and using formula milk concurrently), intermittent mixed feeding (exclusively breastfeeding, but who have used formula feeds in the past) and breastfeeding to formula feeding (mothers who are using only formula feeds, who in the past breastfed exclusively). The first day of data collection was used to pilot the questionnaire. No concerns were identified and the study commenced on the following day. We approached the International Board Certified Lactation Consultants (IBCLCs) practising in the private sector in the Cape Town Metropole via the IBCLC᾽s communication network. A total of 22 had been approached and 5 IBCLCs practising in Rondebosch, Bellville, Stellenbosch and Somerset West, agreed to participate in the study. All the mothers consulting the IBCLCs during an 8-week period (starting 6 October 2015) irrespective of the reason for the consultation, were asked to participate in the study by completing the questionnaire. The IBCLCs avoided data duplication by keeping a record of each participant who had completed a questionnaire. Only data from participants who had signed informed consent and completed the questionnaire were included in the study. Questionnaires were excluded from the data set when, based on consensus by the researchers, answers were ambiguous or incomplete. Ethical approval was obtained from the Stellenbosch University Health Research Ethics Committee (ref. no. S15/04/093).

Results

A total of 108 participants signed informed consent and completed the questionnaires. Data from four participants were excluded due to incomplete data. Exclusive breastfeeding in this population was achieved by slightly more than 50% of the participants, with a downward trend with increasing age of the infant. The most common reason for the use of formula milk was insufficient breastmilk production in the opinion of the participant. A statistically significant association was found between galactogogue use and perceived insufficient breastmilk supply (p=0.013). There was no statistically significant association between type of delivery and exclusive breastfeeding (p=0.099). A statistically significant increase in exclusive breastfeeding rate was shown in mothers with two or more children (p=0.029), compared with mothers with only one child. Having skin-to-skin contact within 1 hour of delivery was not associated with an increase in exclusive breastfeeding in this study (p=0.308). Galactogogues were used by more than half of the participants (54%, n=56). Counselling on breastfeeding techniques to enhance breastmilk production prior to using galactogogues was given to 64% (n=36) of galactogogue users. Non-prescription medication was used by 80% of galactogogue users; the berry elixir-containing ‘Jungle Juice’, for which there appeared to be no standard ingredients, was most widely used (52%, n=29) (Table 3). The dose and use of ‘Jungle Juice’ varied greatly between 250 mL and 2 L per day. Fenugreek was mostly used in the capsule form, at ≤3 capsules per day (90%, n=19). Prescription medication was used by 48% of galactogogue users, of which low-dose sulpiride (50 mg 2 - 3 times per day) was the most frequently used (Table 2). The median age of infants when starting sulpiride was 2 weeks, while the duration of sulpiride use was 4 weeks, with the interquartile range 2 - 11 weeks. A single galactogogue was used by almost half of the participants (46%, n=26), while two and three galactogogues (including both prescription medication and supplements) were simultaneously

Table 1. Participant characteristics Characteristic

Total number of participants, n (%)* EBF,† n (%)

IMF,‡ n (%)

MF,§ n (%)

BF-FF ¶ n (%)

EFF,|| n (%)

Participants included (N)

104

58 (56)

9 (8)

22 (21)

12 (12)

3 (3)

Caesarean section

78 (75)

40 (69)

7 (78)

22 (100)

6 (50)

3 (100)

Vaginal delivery

26 (25)

18 (31)

2 (22)

6 (50)

1 hour skin-to-skin contact post delivery 62 (60)

37 (64)

6 (67)

8 (36)

8 (67)

3 (100)

Type of delivery

Live children of participants 1

69 (66)

33 (57)

9 (100)

17 (77)

8 (66)

2 (67)

29 (28)

22 (38)

4 (18)

2 (17)

1 (33)

≥3

6 (6)

3 (5)

1 (5)

2 (17)

Galactogogue use

56 (54)

26 (45)

7 (78)

16 (73)

7 (58)

Breastfeeding advice received before galactogogue use

36 (64)

12 (46)

4 (44)

13 (81)

7 (100)

N/A

EBF = exclusive breastfeeding; IMF = intermittent mixed feeding; MF = mixed feeding; BF-FF = breastfeeding to formula feeding; EFF = exclusive formula feeding; N/A = not applicable. *Unless otherwise specified. Mothers only breastfeeding and using no other milk supplement.

†

Defined as mothers exclusively breastfeeding, but who have used formula feeds in the past.

‡

Mothers breastfeeding and using formula supplementation concurrently.

Mothers who are using only formula feeds, who in the past breastfed exclusively.

Defined as mothers using formula feeds only.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH Table 2. Type of galactogogue used (N=56) N (%)

Type

EBF

IMF

BF-FF

EFF

Total galactogogue use

56 (54)

26 (46)

7 (13)

16 (29)

7 (13)

Prescription medication

27 (48)

7 (26)

3 (11)

10 (37)

7 (26)

25 (48)

7 (28)

3 (12)

9 (36)

6 (24)

2 (4)

0 (0)

1 (50)

Sulpiride Metoclopramide Supplements

45 (80)

Herbal (fenugreek)*

21 (47)

7 (33)

5 (24)

7 (33)

2 (10)

Non-herbal (Jungle Juice† and Stoney‡)

33 (73)

18 (55)

4 (12)

9 (27)

2 (6)

12 (27)

2 (17)

1 (8)

8 (67)

1 (8)

Miscellaneous

*Trigonella foenum-graecum is a herb containing phyto-oestrogens, which are plant chemicals similar to oestrogen. Diosgenin is specifically implicated to increase breastmilk production. † A tonic made with blackthorn berry elixir (50 mL), rehydration mixes (sachets containing carbohydrates, electrolytes and possibly antioxidants), fruit juice or rooibos tea (1 L) and water (2 L). Rescue remedy drops are sometimes added in varying quantities. ‡ A ginger-flavoured carbonated drink. § Other galactogogues include: Brewer’s yeast – Saccharomyces cerevisiae Prolac – h omeopathic combination sublingual tablet to stimulate milk production (Helonias ioica, Urtica urens, Ricinus communis, calcarea carbonica, calcarea phosphorica, Vitex agnus-castus, ferrum phosphoricum, Pulsatilla, graphites, calcarea fluorica) Lactogogue tea – herbal tea (can be made with the leaves of different herbal plants, although fenugreek is most commonly used) Lactation cookies – oats-based cookie containing brewer’s yeast and flaxseed Protein powder – a powder containing one of three main proteins: whey, soy and casein Oats – Avena sativa Milk stout – stout containing lactose, often used in the fermentation of beer

Table 3. Reasons for the use of formula milk (N=46)

Reason

n (%)

Association of reason listed and the use of a galactogogue (p-value)

IMF

BF-FF

EFF

Perception of too little breastmilk

23 (50)

0.01

Mastitis, thrush or painful breastfeeding

6 (13)

0.93

Returned to work and could not express

2 (4)

0.29

Use of alcohol, nicotine and other harmful substances

1 (2)

0.17

Use of chronic medication and recommended not to BF

1 (2)

0.17

BF found to be emotionally challenging

5 (11)

0.21

Social pressures to stop BF

Infant refused to take breastmilk

10 (22)

0.25

Medical problems/medical advice to use formula milk

17 (37)

0.22

Other

9 (20)

0.63

used in 21% (n=12) and 12.5% (n=7) of participants, respectively. A subjectively judged good increase in milk production was reported by 41% (n=23) of the galactogogue users and 30% (n=17) of the group reported no effect. The majority of the users who no longer used a galactogogue reported that the effect lasted only during use (56%, n=18), or a few days after stopping (28%, n=9). No statistically significant associations were found between any of the medications used as galactogogues and the effect on breastmilk production, or the duration of the effect after stopping (Table 3). Only a few participants reported that they were counselled on possible side-effects (21%, n=12) before galactogogue initiation. Side-effects were reported in 9 mothers and 2 infants. The sideeffects were mostly associated with sulpiride, varying from improved emotional wellbeing (n=2) to moodiness (n=2), which was subjectively noted to be severe in one case. One participant also mentioned that sulpiride contributed to difficulty in losing weight. Other side-effects were associated with brewer’s yeast (cramps in an infant and vaginal pruritis in a mother), berry elixir (improved maternal energy levels) and fenugreek (increased heart rate and breast congestion in one mother).

In most cases (55%, n=31), galactogogues were recommended by medical practitioners, with obstetricians accounting for 35% (n=20). Nursing practitioners and certified lactation practitioners recommended galactogogues in 38% (n=21) of cases. Forty-five percent (n=25) of galactogogue users reported a recommendation from friends, family or other sources.

Discussion

Our study contributes to the limited SA data regarding breastfeeding behaviour, galactogogue use and perceived efficacy and sideeffects in breastfeeding mothers and their infants. We found a high exclusive breastfeeding rate in all age categories compared with the national breastfeeding rates in SA,[3] which could be explained by the specific population group who were seeking the advice of breastfeeding consultants, indicating a high level of motivation to breastfeed. Successful breastfeeding has been associated with vaginal delivery, skin-to-skin contact in the first hour after delivery and a higher parity.[9,10] In our study this association was found with parity, but not with the type of delivery or skin-to-skin contact. Reasons for

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH this can be attributed to the study population who, while seeking breastfeeding advice and assistance, also commonly make use of private-sector healthcare, where there is a high incidence of caesarean delivery. It may, therefore, also imply a smaller than expected role for these factors in predicting breastfeeding success in a motivated population. We found that galactogogues were frequently prescribed and used by more than half of the participants, which is higher than in high-income settings internationally, where a 5 - 33% prevalence of galactagogue use has been documented.[11,12] In our study, supplements were the most preferred galactogogues, with the use of fenugreek and berry elixir-containing juices being widespread, often in combination with prescription medications. Safety and efficacy data supporting the use of galactogogue supplements are limited. We found that the preferred prescription galactogogue was sulpiride. Internationally, domperidone is most frequently prescribed in high-income settings.[5] Given the high incidence of galactogogue usage observed, it is noteworthy that a relatively high percentage of participants found no effect on breastmilk production when using galactogogues. This, along with the high motivation of this study group to breastfeed, might explain the high incidence of the use of a combination of galactogogues. Although an objective analysis of efficacy is limited by the retrospective study design, our data do not support this practice to improve breastfeeding performance. Most participants used galactogogues for a number of weeks, and often when their infants were beyond the neonatal period. Galactogogues should be prescribed with care with the best scientific evidence available, taking into account that there is no evidence to support the use of a galactogogue for longer than 2 weeks,[5] and that after 2 weeks postpartum it has no proven effect on lactational performance.[5,7] In addition, infant suckling is the most important factor for breast milk production once lactation is established, and not prolactin-induced increases in breastmilk volume.[13] We also assessed the subjective experience of potential side- effects. There are very limited data regarding the safety of galactogogues. The side-effects described by study participants were mostly minor complaints and some could be seen as positive effects. Of note is the relatively low rate of side-effects reported, especially in infants. The study design is such, however, that reliable conclusions cannot be drawn on the safety and potential risks of these medications. Our study findings were limited by a number of factors, the first of which was the small sample size. Second, the study population was prone to selection bias, with only IBCLCs serving the private sector being included. It is likely that the clients attending these practices are highly motivated to breastfeed. While this allows the collection of data from those most likely to use galactogogues, it might skew the actual prevalence of galactogogue use in the general

population. Thirdly, we cannot exclude reporting bias, given that the study was designed to investigate women who are highly motivated to breastfeed and, lastly, the study design was retrospective and uncontrolled. Future research should prospectively collect galactogogue safety and efficacy data given the large proportion of women found in our study who use galactogogues for extended periods of time.

Conclusion

This study found that galactogogues were frequently used, with a prevalence that exceeded other published data. We found that sulpiride was frequently prescribed even though it was not recommended during breastfeeding. Furthermore, the time of initiation and duration of galactogogue use was not in accordance with current guidelines. Doctors and other healthcare practitioners should acknowledge breastfeeding mothers’ concerns regarding insufficient milk supply, and place more emphasis on correct breastfeeding technique and other behavioural factors.[5] Acknowledgements. The authors wish to thank the Undergraduate Research Incentive Fund at the Faculty of Medicine and Health Sciences, Stellenbosch University, as well as the participating IBCLCs practices in the Cape Town Metropolitan Area.

1. Donovan TJ, Buchanan K. Medications for increasing milk supply in mothers expressing breastmilk for their preterm hospitalised infants. Cochrane Database Syst Rev 2012;3:CD005544. https://doi.org/10.1002/14651858.cd005544 2. Horta B, Bahl R, Martines J, Victora C. Evidence on the Long-term Effects of Breastfeeding: Systematic Reviews and Meta-analysis. Geneva: World Health Organization, 2007. 3. World Health Organization. The WHO Global Data Bank on Infant and Young Child Feeding A - Z list. Geneva: WHO, 2010. http://www.who.int/nutrition/databases/infantfeeding/en (accessed 24 May 2015). 4. Zuppa AA, Sindico P, Orchi C, et al. Safety and efficacy of galactogogues: Substances that induce, maintain and increase breast milk production. J Pharm Pharm Sci 2010;13(2):162-174. https://doi. org/10.18433/j3ds3r 5. Anderson PO. The galactogogue bandwagon. J Hum Lact 2013;29(1):7-10. https://doi. org/10.1177/0890334412469300 6. Anderson PO, Valdés V. A critical review of pharmaceutical galactagogues. Breastfeed Med 2007;2(4):229-242. https://doi.org/10.1089/bfm.2007.0013 7. Barguno J, Del Pozo E, Cruz M, Figueras J. Failure of maintained hyperprolactinemia to improve lactational performance in the late puerperium. J Clin Endocrinol Metab 1988;66(4):876-879. https://doi.org/10.1210/jcem-66-4-876 8. United States Committee on Drugs. American Academy of Pediatrics Committee on Drugs: Transfer of drugs and other chemicals into human milk. Pediatrics 2001;108(3):776-789. https:// doi.org/10.1542/peds.108.3.776 9. Moore ER, Anderson GC, Bergman N, Dowswell T. Early skin-to-skin contact for mothers and their healthy newborn infants. Cochrane Database Syst Rev 2012;5(5):CD003519. https://doi. org/10.1002/14651858.cd003519.pub3 10. Al-Sahab B, Lanes A, Feldman M, Tamim H. Prevalence and predictors of 6-month exclusive breastfeeding among Canadian women: A national survey. BMC Pediatr 2010;10(1):20. https://doi. org/10.1186/1471-2431-10-20 11. Grzeskowiak LE, Lim SW, Thomas AE, Ritchie U, Gordon AL. Audit of domperidone use as a galactogogue at an Australian tertiary teaching hospital. J Hum Lact 2013;29(1):32-37. https://doi. org/10.1177/0890334412459804 12. Mannion C, Mansell D. Breastfeeding self-efficacy and the use of prescription medication: A pilot study. Obstet Gynecol Int 2012;2012:1-8. https://doi.org/10.1155/2012/562704 13. Gabay MP. Galactogogues: Medications that induce lactation. J Hum Lact 2002;18(3):274-279. https://doi.org/10.1177/089033440201800311

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

RESEARCH

Beliefs and practice in using misoprostol for induction of labour among obstetricians in Zimbabwe M G Madziyire,1 MMed (O&G); B Mateveke,2 MMed (O&G); M F Gidiri,1 FRCOG 1 2

Department of Obstetrics and Gynaecology, College of Health Sciences, University of Zimbabwe, Harare, Zimbabwe Department of Obstetrics and Gynaecology, Harare Hospital, Harare, Zimbabwe

Corresponding author: M G Madziyire (gynaemadzi@y7mail.com)

Background. Misoprostol is commonly used for induction of labour in term pregnancy. There are different routes and dosing schedules for administering the drug. Objectives. To describe the prescribing pattern (dose, route, duration), beliefs and factors affecting use of misoprostol for inducing term pregnancy among practising obstetricians in Zimbabwe. Methods. A cross-sectional descriptive survey was undertaken among practising obstetricians in Zimbabwe. A questionnaire was sent as an email, WhatsApp or short message service (SMS, or text) web link to all practising obstetricians in Zimbabwe using the SurveyMonkey online tool. All consenting practitioners were requested to respond online. The responses were analysed using the SurveyMonkey software. Results. There were 52 responses from the 63 questionnaires, a response rate of 82.5%. Seventy-six percent preferred oral misoprostol for induction of labour. The most common indication for induction was prolonged pregnancy accounting for 58% of respondents. The largest group of the practitioners who responded (36%) learnt their misoprostol dosing regimen from WHO/FIGO/NICE guidelines. A composite of highly variable dose regimens referred to as ‘other regimens’ was the dosing regimen preferred by 34% of respondents. Fiftyeight percent of practitioners used two cycles of misoprostol dosing before concluding that induction had failed and 52% would resort to caesarean section immediately if induction failed. Conclusion. The results show marked heterogeneity in the dosing schedules employed by obstetricians for induction of labour with the majority not following standard misoprostol guidelines for labour induction. S Afr J Obstet Gynaecol 2017;23(1):24-27. DOI:10.7196/SAJOG.2017.v23i1.1163

Misoprostol is a prostaglandin E1 analogue widely used for the induction of labour in term pregnancy because of its low cost and efficacy when compared with most methods of induction.[1] Induction of labour was estimated to account for 9.6% of births during the World Health Organization (WHO) global survey on perinatal and maternal health.[2] Different routes of administering misoprostol for the induction of labour have been employed with the oral route preferred because of ease of administration and a short half-life.[1,3] A reduced caesarean section rate was reported with the dose of oral misoprostol of 20 - 25 µg 2-hourly compared with vaginal dinoprostone.[1] Oral misoprostol 50 µg 4-hourly at up to 5 doses was also shown to be as safe and as effective as 25 µg 4-hourly when administered vaginally at 5 doses.[4] Various misoprostol induction regimens have been suggested and consequently clinicians have developed diverse prescribing patterns. Various professional bodies have produced recommended dosing regimens. The WHO, International Federation of Gynecology and Obstetrics (FIGO) and the National Institute for Health and Care Excellence (NICE) guidelines recommend 20 - 25 µg oral misoprostol 2-hourly and 25 µg given vaginally every 6 hours.[2,5] The oral titration regimen has emerged as the preferred regimen as it is less likely to cause uterine hyperstimulation.[1,6,7] The guidelines are silent on issues such as maximum number of doses, breech

presentation (if caesarean section is declined) or induction in multiple pregnancies.[2,8,9] In Zimbabwe, the Ministry of Health and Child Care has adapted the WHO guidelines with the dosing regimen outlined above; however, various hospitals follow their own specific guidelines. Harare Central Hospital and Parirenyatwa Hospital, which are tertiary teaching hospitals, follow a titrated oral dose which differs according to parity and is outlined in Table 1. The dosing in Table 1 is administered until onset of labour, or when 200 μg misiprostol has been depleted. There is no instruction on the appropriate action if there is no response to the first cycle of dosing. A cycle of induction of labour is considered to be the doses required before or after completion of Table 1. Recommended oral misoprostol regimen at Harare and Parirenyatwa hospitals Parity

Loading dose (mL)*

Hourly dose (mL)*

1-2

≥4

*1 µg/mL misoprostol solution.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH a total dose of 200 µg of misoprostol, whichever way it has been administered. This dosing schedule was created empirically by senior obstetricians in the unit based on their experience with misoprostol use.

Methods

A questionnaire was sent as either an email, WhatsApp or SMS/ text with a web link to all practising obstetricians in Zimbabwe using the SurveyMonkey online tool. All qualified and practising obstetricians who consented to take part in the survey were contacted. There were 65 registered obstetricians in Zimbabwe at the time of the study (April 2016) according to the Zimbabwe Society of Obstetricians and Gynaecologists (ZSOG) register. Permission to conduct the study was obtained from the ZSOG. Responses were analysed using the SurveyMonkey online analytical tool. The response on misoprostol dosing was open-ended requiring practitioners to write the actual dose they would prescribe for induction of labour. The investigators then categorised the responses according to similarities, or being part of a standard dosing regimen recommended by local or international practice guidelines. All oral doses were mainly in the form of misoprostol dissolved in water to make a 1 µg/mL solution. Table 2 shows the rationale of the dosing categories which were used for analysis.

Results

There were 52 responses from the 63 questionnaires, a response rate of 82.5%. Two responses were incomplete and were removed from the final analysis. All respondents preferred misoprostol for induction of labour with 39 (78%) preferring the oral route. The largest group of respondents (36%) adapted their dosing schedule based on the WHO/FIGO/NICE guidelines. A composite of variable dosing schedules referred to as ‘other regimens’ was the most common dosing schedule, followed by the departmental dosing guideline and the WHO/FIGO/NICE dosing guideline, across all parities. Ten practitioners (20%) did not use misoprostol for ≥ para 4.

Discussion

This study shows variable dosing schedules used for inducing labour in term pregnancy using misoprostol by obstetricians in Zimbabwe. The largest group of practitioners use dosing schedules which do not comply with any set guidelines – for example, the dosing schedule referred to as ‘other regimens’ comprised highly variable misoprostol regimens, which did not correspond with the main dosing categories selected for analysis (Table 2). The trend for practitioners with <5 years’ experience to follow departmental guidelines (10 (50%)) more than their colleagues with more years of experience (5 (30%)) (Table 3) seems to imply that more experienced colleagues are more likely to depart from guidelines set for uniform practice in the public sector as experience gives them the latitude to use their own regimens. While 36% of practitioners claimed that they adapted their dosing regimen based on the WHO/FIGO/NICE guidelines, only 10 (20%) practitioners used the regimen in actual practice. The fact that 42 (84%) practitioners were either in part-time private practice or fulltime government/university practice does imply that they knew of these departmental guidelines. The most common indication for inducing labour was prolonged pregnancy (58%) (Table 4). This differs from a randomised controlled trial at Harare Hospital in 2013, which looked at factors associated with failed induction of labour with titrated oral misoprostol and found hypertensive disorders to be the commonest indication (38.1%).[7] The difference could be due to Harare Hospital being a tertiary centre for high-risk pregnancies and this would select out hypertensive disorders as a common cause of referral. There is no agreed definition of failed induction of labour using misoprostol, mainly because practitioners have varied desired endpoints of the induction process. The choice of using failure of initiating labour after two cycles of misoprostol as failed induction is arbitrary, and probably chosen as obstetricians try to balance the increased risk of adverse maternal-fetal outcomes and unwarranted intervention. It is noteworthy that a proportion of practitioners have calculated a dosing schedule which begins with 50 µg, then followed with 20 - 30 µg

Table 2. Dosing categories 20 - 30 µg stat, 15 - 20 µg hourly (oral) (departmental regimen)

This is the dosing schedule pinned up in the teaching maternity units of Harare and Parirenyatwa hospitals where most of the obstetricians in Zimbabwe work or have passed through during their training. The higher limit of the dose is recommended for the nulliparous and it is tailored down for women of higher parity.

20 - 25 µg 2-hourly (oral) (WHO/FIGO/ NICE)

This is the WHO/FIGO/NICE recommended dose for inducing a term pregnancy. The ‘Essential Guide to Management of Common Obstetric and Gynaecologic Conditions in Zimbabwe’, produced by the University of Zimbabwe, Department of Obstetrics and Gynaecology also recommends this regimen.[9] The Ministry of Health and Child Care has also adopted this regimen and put it up on wall charts in various maternity units in Zimbabwe.

50 µg 4 - 6-hourly (oral)

The ‘Essential Guide to Management of Common Obstetric and Gynaecologic Conditions in Zimbabwe’, produced by the University of Zimbabwe, Department of Obstetrics and Gynaecology recommends 50 μg 4-hourly of oral misoprostol or 50 μg 6-hourly of vaginal misoprostol.

50 µg stat, followed by 15 - 30 µg 1 - 2-hourly (oral)

This was a common dosing range among the respondents.

20 - 25 µg hourly (oral)

This was found to be preferred by up to 3 respondents.

Other regimens (oral)

This category comprised highly variable dosing regimens.

Vaginal

Any vaginal dosing.

No use

Those who would not administer misoprostol.

SAJOG • May 2017, Vol. 23, No. 1

RESEARCH Table 3. Analysis against experience Years in practice Sex Female Male Total, n (%) Practice Part-time* Government practice Private practice Total Source Departmental guidelines (Table 1) WHO/FIGO/NICE guidelines Experience Colleague Other Total Cycles 2 cycles 1 cycle Other (<1 cycle or >2 cycles) Total Cervical assessments No Yes Sometimes Total Route of administration Oral Vaginal Total

5 - 10

11 - 20

>20

Total, n (%)

7 13 20 (40)

3 5 8 (16)

1 9 10 (20)

3 9 12 (24)

14 (28) 36 (72) 50 (100)

11 7 2 20

6 1 1 8

8 2 0 10

6 1 5 12

31(62) 11 (22) 8 (16) 50 (100)

10 5 2 1 2 20

1 3 1 2 1 8

2 6 0 1 1 10

2 4 4 1 1 12

15 (30) 18 (36) 7 (14) 5 (10) 5 (10) 50 (100)

12 6 2 20

6 2 0 8

5 3 2 10

6 3 3 12

29 (58) 14 (28) 7 (14) 50 (100)

4 7 9 20

0 6 2 8

1 7 2 10

0 11 1 12

5 (10) 31 (62) 14 (28) 50 (100)

18 2 20

5 3 8

8 2 10

8 4 12

39 (78) 11 (22) 50 (100)

*Part-time refers to all practitioners who work for university/government and also practices privately.

Table 4. Common indication for induction (N=50) Answer choices

Responses, n (%)

Hypertensive disorders

21 (42)

Ruptured membranes

Prolonged pregnancy

29 (58)

Non-reassuring fetal heart rate

Table 5. Obstetricians who would use misoprostol in conditions 1 - 6 (N=50) Answer choices*

Responses, n (%)

Breech presentation

3 (6)

Multiple pregnancy

9 (18)

Cardiac disease

19 (38)

Impression of a big baby

12 (24)

Unconscious patient

10 (20)

Previous uterine incision

3 (6)

No use

20 (40)

*Most obstetricians would give misoprostol in cases of more than one of these conditions.

every 1 - 2 hours, and the tendency was to use it mainly in those who were less than para 3 (Table 6). The Ministry of Health has distributed the WHO misoprostol dosing guidelines but they seem not to have been widely accepted, given the above findings. The departmental dosing schedule at Harare hospital is more dose intense than the WHO/FIGO/ NICE schedule of 20 - 25 Âľg misoprostol 2-hourly. Most practitioners did not decrease the dose as parity increased although 10 practitioners stated they would completely avoid misoprostol in those with parities >3. The added caution does not seem to be justified as WHO/FIGO guidelines do not recommend against use in higher parities. It is possible that practitioners would settle for the safest, most efficacious and userfriendly dose. A user-friendly dosing schedule would be one with less dosing frequency, yet maintaining safety and efficacy. It was surprising that three practitioners would use misoprostol in women with a previous caesarean section despite the inherent risk of uterine rupture, should hyperstimulation occur (Table 5). While there were no reported cases of uterine rupture in 160 women with a previous caesarean section in a Cochrane systematic review of oral misoprostol for induction of labour, most guidelines caution against using misoprostol in women with previous uterine surgery.[1] It is interesting and disturbing to note that only 62% of practitioners practise routine cervical assessment before prescribing misoprostol.

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

RESEARCH Table 6. Dosing schedule according to parity (N=50) Parity, n (%) Misoprostol dosing

Other regimens

13 (26)

15 (30)

19 (18)

19 (38)

17 (34)

20 - 30 µg stat, then 15 - 20 µg hourly (departmental recommendations)

12 (24)

10 (20)

11 (22)

20 - 25 µg, 2-hourly (WHO/FIGO/NICE)

10 (20)

50 µg stat, then 15 - 30 µg, 1-2 hourly

9 (18)

8 (16)

7 (14)

3 (6)

2 (4)

20 - 25 µg, hourly

2 (4)

3 (6)

1 (2)

50 µg stat, then 4 - 6-hourly

2 (4)

1 (2)

Vaginal

2 (4)

1 (2)

No use

3 (6)

7 (14)

The argument might be that misoprostol still works whether the cervix is ripe or not but cervical assessment selects out candidates for immediate amniotomy. Amniotomy with or without oxytocin might be more expensive and more laborious than using misoprostol alone. However, cervical assessment allows for other interventions including membrane sweeping which can avoid unnecessary use of misoprostol or increase the chance of a successful induction.

Study limitations This was an online survey and hence respondents had no chance to seek clarity on certain questions.

Conclusion

Obstetricians in Zimbabwe have adopted a variety of misoprostol dosing schedules other than the ones in standard protocols such as the WHO, FIGO or NICE guidelines. There was considerable difference of opinion concerning the conditions which contraindicate

misoprostol use in term pregnancy, and also on the number of cycles of misoprostol to be administered. 1. Alfirevic Z, Aflaifel N, Weeks A. Oral misoprostol for induction of labour. Cochrane Database Syst Rev 2014;6:CD001338. https://doi.org/10.1002/14651858.CD001338.pub3 2. World Health Organization. WHO recommendations for induction of labour. Geneva: WHO, 2011. 3. Zvandasara P, Saungweme G, Mlambo J, Chidembo W, Madzivanzira N, Mwanjira C. Induction of labour with titrated oral misoprostol suspension. A comparative study with vaginal misoprostol. Cent Afr J Med 2008;54(9-12):43-49. https://doi.org/10.4314/cajm.v54i9-12.62626 4. Rahman H, Pradhan A, Kharka L, Renjhen P, Kar S, Dutta S. Comparative evaluation of 50 microgram oral misoprostol and 25 microgram intravaginal misoprostol for induction of labour at term. J Obstet Gynaecol Can 2013;35(5):408-416. https://doi.org/10.1016/s1701-2163(15)30931-2 5. National Institute for Health and Care Excellence. Clinical Guidelines: Inducing Labour. London: NICE, 2008. https://www.nice.org.uk/guidance/cg70 (accessed 12 February 2016). 6. Kundodyiwa TW, Alfirevic Z, Weeks AD. Low-dose oral misoprostol for induction of labour: A systematic review. Obstet Gynecol 2009;113(2 Pt 1):3743-3783. https://doi.org/10.1097/ aog.0b013e3181945859 7. Mateveke B, Chipato T, Guzha B, Mahachi L. Factors associated with failed induction of labour in patients undergoing induction with titrated oral misoprostol at Harare Maternity Hospital. Am Res J Gynecol 2015;1(1):12-16. 8. The National Medicine and Therapeutics Policy Advisory Committee (NMTPAC). 6th Essential Medicines List and Standard Treatment Guidelines for Zimbabwe. Harare: NMTPAC, 2011;93-95. http://apps.who.int/medicinedocs/documents/s21753en/s21753en.pdf (accessed 12 February 2016). 9. Department of Obstetrics and Gynaecology. Essential Guide to Management of Common Obstetric and Gynecologic Conditions in Zimbabwe. Harare: University of Zimbabwe, 2012(8):30-32.

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

CASE REPORT

Secondary postpartum haemorrhage with uterine artery pseudoaneurysm after caesarean section, managed with uterine artery embolisation S Barahmeh,1 MD (O&G); S H Edwan,2 MD (Radiology) 1 2

Department of Obstetrics and Gynaecology, Consultative Arab Hospital, Ramallah, Palestine Interventional Radiology Department, Consultative Arab Hospital, Ramallah, Palestine

Corresponding author: S Barahmeh (samersalman2@yahoo.com)

Uterine artery pseudoaneurysm (PA) is a rare but serious complication of caesarean section (C/S). If inadequately treated, it can lead to life-threatening postpartum haemorrhage. We report the case of a 28-year-old woman who developed secondary postpartum haemorrhage resulting from uterine artery PA after C/S. Angiographic embolisation is a safe and effective procedure for treating postpartum haemorrhage resulting from PA in haemodynamically stable patients. However, uterine artery ligation may be the surgical procedure of choice for haemodynamically unstable patients when preservation of fertility is desired. S Afr J Obstet Gynaecol 2017;23(1):28-30. DOI:10.7196/SAJOG.2017.v23i1.1105

Pseudoaneurysm (PA) of the uterine artery is an uncommon cause of delayed postpartum haemorrhage following cesarean section (C/S) or vaginal delivery. This condition may result in secondary postpartum haemorrhage, which is defined as haemorrhage that occurs between 24 hours, and 6 and 12 weeks postpartum.[1] Although a diagnosis of retained gestational products or endometritis should be considered initially, a diagnosis of PA of the uterine artery and caesarean scar dehiscence should be considered after ruling out disseminated intravascular coagulation as a possible diagnosis when a patient presents with massive uterine bleeding without any associated symptoms, such as fever and tenderness, or subinvolution of the uterus.[2] Haematoma formation involving the uterine artery is the main suggested mechanism associated with PA of the uterine artery. After haematoma formation, there is central liquefaction that leaves a cavity with turbulent blood flow as a result of persistent communication between the parent artery and the hematoma. A uterine artery PA develops when the uterine artery is lacerated or injured in a way that would breach the integrity of the three-layer arterial wall lining. While maintaining contact with the parent vessel, extravasating blood dissects through the tissues, finally establishing a connection with the uterine cavity, which causes a delayed haemorrhage. The boundaries of a false aneurysm are constituted by thrombus, as opposed to the three arterial layers as in a true aneurysm. Although Doppler ultrasound can aid in the assessment, uterine artery angiography is necessary to make the diagnosis and provides the subsequent means for embolisation. We present a case of a uterine artery PA presenting as delayed postpartum haemorrhage.[3]

Case report

A 28-year-old para 3 was transferred to our institution at 27 days post operation with symptoms of excessive bleeding per vagina. She had undergone an elective low transverse segment C/S delivery for breech presentation at a gestational age of 39 weeks at another hospital. Her intraoperative course was smooth

and uneventful, and the estimated blood loss was ~600 mL. Her postoperative course was also uncomplicated until day 13 postoperatively. On postoperative day 13, she was readmitted to the same hospital for management of a delayed postpartum haemorrhage, which had developed abruptly with no previous history of abdominal pain, foul-smelling vaginal discharge or fever. She was hypotensive (blood pressure 80/45) and tachycardic (pulse of 110 bpm) with a haemoglobin of 6.8 g/dL. Her platelet count was 149 and her coagulation profile and biochemistry showed no derangements. A transabdominal ultrasound showed blood in the uterus, but there were no retained gestational products After stabilisation with four units of cross-matched packed red blood cells, a dilatation and curettage was performed and tissue was sent to pathology for analysis. The bleeding stopped, and she was discharged on her second day in hospital, with a haemoglobin level of 10 g/dL. Pathology results showed no evidence of retained gestational products. Two weeks later, she was admitted to the same hospital where she had had her C/S and surgical evacuation, with increased vaginal haemorrhage and severe lower abdominal pain. Although she was haemodynamically stable, with a BP of 110/65 and a pulse of 88 bpm, her blood analysis revealed a haemoglobin level of 8.0 g/dL. Transvaginal sonography, with both colour and power Doppler modes, confirmed an empty uterus without signs of residual placental tissue in the uterine cavity. A hypoechoic lesion was detected in the isthmic region of her uterus and power Doppler imaging revealed blood flow within it. Interventional radiology at our institution was consulted, as the patient desired future fertility. Pelvic angiography demonstrated a left uterine artery PA, which measured 4 cm in diameter, with extravasation of contrast into a pocket that connected to the uterine cavity. After obtaining the patientâ&#x20AC;&#x2122;s consent, selective left uterine artery embolisation was performed with a mixture of Gelfoam and contrast media, followed by one stainless steel coil insertion. The

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CASE REPORT procedure was inititated by placing a number 4 French sheath into the right common femoral artery and passed retrograde into the right external iliac artery, right common iliac artery and aorta, respectively, at which point crossing over to the left common iliac artery, left internal iliac artery, and the left uterine artery. Selective catheterisation of the left uterine artery was carried out with a Cobra glide, Terumo Guidewire and Progreat microcatheter. Angiography showed a PA of the left uterine artery (Figs. 1 and 2), and embolisation was performed with Gelfoam and a stainless steel coil embolus. A post-embolisation angiography investigation was performed to ensure complete occlusion of the left uterine artery. A follow-up with colour Doppler ultrasound one day later showed that the aneurysmal cavity was filled with echogenic content and there was no evidence of blood flow within the cavity (Fig. 3).

Discussion

Postpartum haemorrhage remains one of the major causes of maternal mortality. Secondary postpartum haemorrhage is defined as excessive bleeding that starts any time from 24 hours after delivery up to 6 weeks postpartum, and most commonly occurs between 8 and 14 days postpartum.[1] Common causes include retained gestational products, subinvolution of the placental bed, and endometritis. Rare causes include PA of the uterine artery, arteriovenous malformations, C/S delivery and choriocarcinoma. When the more common causes have been excluded, pelvic angiography may be performed. PA of the uterine artery should be listed as a possible cause of postpartum haemorrhage after C/S.[4] Trauma to the uterine artery during surgery may cause a defect in the arterial wall, through which arterial blood escapes and diffuses to the adjacent tissues, resulting in the formation of a haematoma. When this haematoma is in continuity with the uterine artery, which supplies continuous blood flow, a PA forms. The absence of a three-layered arterial wall lining in a PA differentiates it from a true aneurysm.[5]

Fig. 2. Angiography showing pseudoaneurysm of the left uterine artery.

Fig. 3. Result after embolisation coil.

Fig. 1. Computed tomography angiography showing pseudoaneurysm of the left uterine artery, which measured 4 cm in diameter.

Pelage et al.[6] evaluated the efficacy and safety of selective arterial embolisation of uterine arteries in women with delayed secondary postpartum haemorrhage. In their study (N=14), pseudoaneurysms of the uterine artery were found in two participants and immediate resolution of external bleeding was observed after embolisation. In our study, there were no complications related to this invasive treatment. Other authors have described complications, including muscle pain and bladder necrosis.[1] Our patient developed a PA ~4 weeks post-C/S delivery. Treatment was by angiographic embolisation of uterine arteries with Gelfoam and embolisation coils. In a retrospective review of eight women, Rosenthal and Colapinto[7] observed that angiographic arterial embolisation was the most useful clinical tool in the management of post-operative vaginal haemorrhage.

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CASE REPORT Angiographic embolisation has many advantages when compared with hysterectomy including decreased morbidity, the ability to localise the bleeding site and provide a more distal occlusion than surgical ligation, and preservation of future fertility. Burchell[8] demonstrated that bilateral internal iliac artery ligation was more effective than unilateral ligation in reducing the pulse pressure. It is possible that the redistribution and redirection of blood or hypoxia-induced neovascularisation allows bleeding to occur from the contralateral side after unilateral embolisation. Inadequate embolisation of a PA due to extrauterine feeding arteries, such as the internal pudendal artery, ovarian artery, inferior epigastric artery, or contralateral uterine artery can lead to embolisation failure. Hence, bilateral uterine embolisation is safe and more advantageous than unilateral embolisation.[3,9] Uterine artery embolisation has become an effective and safe treatment for postpartum haemorrhage, allowing the preservation of reproductive function. Recent reports described the use of a thrombin injection directly into the pseudoaneurysm under ultrasound guidance, as a substitute for arterial embolisation; however, its indications and effectiveness have not yet been determined. The surgical approach may be more suitable in cases of acute and massive bleeding in which there is no time for embolisation, and may depend on the specific resources available in each institution.[2] With newer modalities of treatment becoming available, hysterectomy for postpartum haemorrhage should be a last resort. Injuries to the vessel are more likely during a C/S, which could subsequently give rise to a PA. Doppler ultrasound and angiography

are useful techniques to diagnose this condition. Selective embolisation of an affected vessel can be performed and bleeding can be arrested instantaneously. Being a minimally invasive procedure and posing limited risk to the mother, embolisation should be offered whenever feasible, and hysterectomy could be considered when the preservation of fertility is not important. On the other hand, uterine artery ligation and extirpation of uterine artery PA is another surgical choice for preserving fertility.[10] Conflict of interest. The authors report no conflicts of interest. 1. Nanjundan P, Rohilla M, Raveendran A, Jain V, Khandelwal N. Pseudoaneurysm of uterine artery: A rare cause of secondary postpartum hemorrhage, managed with uterine artery embolisation. J Clin Imaging Sci 2011;1:14. https://doi.org/10.4103/2156-7514.76692 2. Yeniel AO, Ergenoglu AM, Akdemir A, Eminov E, Akercan F, Karadadaş N. Massive secondary postpartum hemorrhage with uterine artery pseudoaneurysm after cesarean section. Case Rep Obstet Gynecol 2013;2013:285846. https://doi.org/10.1155/2013/2858463 3. Sharma N, Ganesh D, Devi L, Srinivasan J, Ranga U. Prompt diagnosis and treatment of uterine arcuate artery pseudoaneurysm: A case report and review of literature. J Clin Diagnos Res 2013;7(10):2303-2306. https://doi.org/10.7860/JCDR/2013/6063.3506 4. Zubor P, Kajo K, Dokus K, et al. Recurrent secondary postpartum hemorrhages due to placental site vessel subinvolution and local uterine tissue coagulopathy. BMC Pregnancy Childbirth 2014;14:80. https://doi.org/10.1186/1471-2393-14-80 5. Takeda A, Kato K, Mori M, Sakai K, Mitsui T, Nakamura H. Late massive uterine hemorrhage caused by ruptured uterine artery pseudoaneurysm after laparoscopic-assisted myomectomy. J Minim Invasive Gynecol 2008,15(2):212-216. https://doi.org/10.1016/j.jmig.2007.09.006 6. Pelage JP, Soyer P, Repiquet D, et al. Secondary postpartum hemorrhage: Treatment with selective arterial embolisation. Radiology 1999;12:385-389. https://doi.org/10.1148/ radiology.212.2.r99jl05385 7. Rosenthal DM, Colapinto R. Angiographic arterial embolization in the management of postoperative vaginal hemorrhage. Am J Obstet Gynecol 1985;151(2):227-231. https://doi.org/10.1016/00029378(85)90018-3 8. Burchell RC. Internal iliac artery ligation: Hemodynamics. Obstet Gynecol 1964;24:737-739. 9. Cooper BC, Hocking-Brown M, Sorosky JI, et al. Pseudoaneurysm of the uterine artery requiring bilateral uterine artery embolization. J Perinatol 2004;24(9):560-562. https://doi.org/10.1038/ sj.jp.7211119 10. Chitra TV, Panicker S. Pseudoaneurysm of uterine artery: A rare cause of secondary postpartum hemorrhage. J Obstet Gynaecol India 2011;61(6):641-644. https://doi.org/10.1007/s13224-0110096-6

SAJOG • May 2017, Vol. 23, No. 1

This open-access article is distributed under Creative Commons licence CC-BY-NC 4.0.

CASE REPORT

Bilateral synchronous benign ovarian neoplasm: A rare occurrence I U Takai,1 MBBS, MHPM, FCAI, FMCOG; U A Umar,1 MD, FWACS; H A Nggada,2 MBBS, FMC Path; M Bukar,3 MBBS, MHPM, FMCOG, FWACS; B M Audu,3 MBBS, MSc, FMCOG Department of Obstetrics and Gynaecology, Bayero University/Aminu Kano Teaching Hospital, Kano, Kano State, Nigeria Department of Histopathology, University of Maiduguri Teaching Hospital, Maiduguri, Borno State, Nigeria 3 Department of Obstetrics and Gynaecology, University of Maiduguri Teaching Hospital, Maiduguri, Borno State, Nigeria 1 2

Corresponding author: I U Takai (takaiidris@yahoo.co.uk)

Bilateral synchronous ovarian tumours are defined as the occurrence of two or more histologically distinct tumours in the ovaries. Synchronous tumours of the female genital tract are rare and the association of mature cystic teratoma with contralateral serous cystadenoma is uncommon. We report the rare occurrence of a giant mature cystic teratoma with a coexisting contralateral serous cystadenoma in a 32-year-old para 5. The patient refused an initial offer for surgery 10 years earlier, and had three successive term pregnancies and deliveries, with the tumour intact. She later had bilateral salpingo-oophorectomy of a 55 kg left ovarian mass and a 2 kg right ovarian mass, which revealed a left ovarian benign cystic teratoma and a right ovarian papillary serous cystadenoma at histology. Neglected ovarian tumours are still encountered in sub-Saharan Africa and may involve diagnostic and management challenges. There is a need to educate women and the community on ovarian neoplasms and the need to present early for effective management. Womenâ&#x20AC;&#x2122;s reproductive health rights need to be encouraged and possibly legislated in our setting. S Afr J Obstet Gynaecol 2017;23(1):31-34. DOI:10.7196/SAJOG.2017.v23i1.1130

Synchronous tumours of the female genital tract, which are most often ovarian and endometrial,[1] are rare and account for only 0.7 - 1.8% of all tumours.[2] The association of mature cystic teratoma with contralateral serous cystadenoma is uncommon.The most common benign germ cell tumours (GCTs) of the ovary are the mature cystic teratomas (MCTs). Teratomas comprise about 15 - 20% of all ovarian tumours.[3] They are composed of one or more germ layers derived from a pluripotent precursor cell. MCTs are frequently multicystic and contain sebaceous fluid, hair, teeth, bone or skin, and are bilateral in 10 - 12% of cases. Serous cystadenoma is also a benign tumour of the ovary, which is responsible for about 30% of all epithelial ovarian neoplasms. They are commonly unilocular, containing clear yellow fluid, and about 10 - 15% are bilateral.[4] The association of MCT with contralateral serous cystadenoma is uncommon.

Case report

We present a case of bilateral synchronous benign ovarian neoplasm that comprised a giant teratoma weighing 55 kg, with a coexisting contralateral serous cystadenoma that weighed 2 kg. The patient was a 32-year-old para 5 who had last given birth 3 years prior to presentation at the Gynaecology Clinic of the State Specialist Hospital, Maiduguri, Nigeria. She was referred from a private clinic. Her problem started ~10 years prior to presentation with abdominal swelling, when an ultrasound scan revealed a complex left ovarian cyst. She was counselled for surgery but defaulted. She had tapping of the swelling in some of the hospitals she visited and had previously been on treatment with traditional medicine. Despite recurrent swelling, she had three successive term pregnancies and deliveries. At the time of presentation at our facility, the patient believed that she was carrying a twin pregnancy and mentioned she had been informed by traditional healers that her

condition was not suitable for surgery. As such, she was fearful of dying during the removal of the mass. She had also been threatened with divorce if she accepted any form of surgical intervention. These factors had made the patient refuse any surgical intervention at an earlier stage. In the year prior to her presentation at the referring hospital, the abdominal swelling increased rapidly and was associated with respiratory and abdominal discomfort, which necessitated her presentation at the hospital. Occasional abdominal pain and vomiting was accompanied by a loss of appetite but there was no abnormal vaginal bleeding or urinary symptoms. She had no body itching, yellowness of the eyes, cough or chest pain, but had associated dyspnoea and weight loss. She had attained menarche at the age of 15 years and had a regular menstrual pattern. All her previous pregnancies were unsupervised and all deliveries were at home. There was no family history of breast, ovarian or colonic cancers and the patient was neither diabetic nor hypertensive. On examination, she was chronically ill-looking. She weighed 96 kg and was 1.56 m tall. Her chest was clinically clear and her cardiovascular system was stable. An abdominal examination revealed a grossly distended abdomen with distended subcutaneous veins and a uniformly enlarged mass compatible with a term gestation. The mass was firm to hard, not mobile, smooth and non-tender. The liver, spleen and kidneys were not palpable. There was no demonstrable ascites and bowel sounds were present. A pelvic examination revealed a normal vulva and vagina. However, the cervix was flushed and the uterus could not be delineated. Both adnexae were full. The pouch of Douglas was free and the rectal examination was normal. A working diagnosis of huge ovarian tumour was made (Fig. 1A). Her full blood count result was normal with a packed cell volume of 31%. She had normal electrolyte, urea and creatinine levels and

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CASE REPORT

Fig. 1A. Grossly distended abdomen and veins in a patient with ovarian tumour.

at surgery were minimal ascites and a massive twisted left ovarian cystic mass that adhered to the anterior abdominal wall, measuring 62 cm at its widest diameter, and weighing ~55 kg with a thick smooth wall without excrescences on the surface (Fig. 1B). The right ovary was also found to be cystic, measuring 11 cm at its widest diameter and weighing ~2 kg with a thin smooth wall without excrescences on the surface (Fig. 1C). The uterus, tubes, omentum, liver and spleen were normal. The bowels were free and the pouch of Douglas was empty. The pelvis appeared generally clean and the patient had bilateral salpingo-oophorectomy. Macroscopic opening of the left ovary revealed a multilocular cystic cavity containing gelatinous material and a smooth inner lining. A focal solid area, hair shafts and cartilage were noted. Sections of the left ovary showed triphasic elements composed of skin, intestinal epithelium, cartilage, adipose and neural tissue, as well as lymphoid cells, but there was no evidence of malignancy (Fig. 2). Opening of the right ovary revealed a multilocular cyst containing about 800 mL of blood-tinged serous fluid. Sections of the right ovary showed a fibro-collagenous cyst wall lined by a single layer of cuboidal epithelium with focal areas of papillary infolding and haemorrhage consistent with serous cystadenoma (Fig. 3). The ascitic fluid did not show any malignancy

Fig. 1B. Gross appearance of left ovarian benign cystic teratoma weighing 55 kg (ventral view).

Fig. 2. HE stain of a section of the matured left ovarian benign cystic teratoma showing matured cartilage (long arrow), respiratory epithelium (short arrow) and adipose tissue (arrow head).

Fig. 1C. Gross appearance of right ovarian papillary serous cystadenoma weighing 2 kg. tested negative for HIV and the hepatitis B surface antigen (HBsAg). The patientâ&#x20AC;&#x2122;s urinalysis, plain chest X-ray, fasting blood sugar and liver function tests were all normal. A urinary pregnancy test was negative and an abdomino-pelvic ultrasound scan revealed a huge multiseptated, cystic intra-abdominal mass arising from the pelvis, with mixed echogenicity and a thick wall. Measurements of the mass were beyond the capacity of the probe. The liver and spleen appeared normal but the kidneys had dilated calyxes. The uterus was not visualised. She was counselled for exploratory laparotomy and consented on condition that her uterus be preserved. Findings

Fig. 3. HE stain of a section of the right ovarian papillary serous cystadenoma showing cuboidal epithelium (arrow head), papillary projections (long arrow) and connective tissue stroma (short arrow).

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CASE REPORT and the final diagnosis was a left ovarian benign cystic teratoma and right ovarian papillary serous cystadenoma. She had an uneventful postoperative period and was discharged. At a follow-up visit to a gynaecology clinic 3 weeks later, she was clinically stable and the patient was set to be followed up for life.

Discussion

The term synchronous tumour is used when two or more tumours occur simultaneously in a patient. Synchronous tumours of the female genital tract are rare, accounting for 0.7 - 1.8% of tumours in the female genital tract.[2] The most frequently encountered synchronous tumours of the female genital tract are ovarian (as in this case) and endometrial.[1] Although most are double, triple synchronous tumours have been reported.[2] Pathological screening of 957 benign cystic teratomas yielded nine multiple ipsilateral ovarian teratomas.[5,6] Serous tumours rarely occur in combination with GCTs.[7] Common combinations with ovarian tumours include mucinous cystadenoma and a combination of Brenner tumour, mature cystic teratoma, Sertoli-Leidig cell tumour may be seen.[8] Ovarian tumours are currently classified based on the histogenesis of the ovary. Early development of the ovary is characterised by segregation and migration of the primordial germ cells from their sites of origin to the genital ridges (bilateral thickenings of the coelomic epithelium). This is followed by the proliferation of the coelomic epithelium and underlying mesenchyme. The ovary is then divided into the peripheral cortex and central medulla. The cortex develops further, while the medulla undergoes involution.[9] The World Health Organization histological classification of ovarian tumours separates ovarian neoplasms according to the most probable tissue of origin: surface epithelium (65%), germ cells (15%), sex-cord stromal tissue (10%), miscellaneous (5%) and tumours of metastatic origin (10%).[10] Teratomas are the most common GCTs of the ovary. They are composed of multiple cell types derived from one or more of the three germ layers. Mature teratomas are benign cystic teratomas. They occur mainly in women of reproductive age, as in our patient, but may occur in postmenopausal women and in children. These tumours may contain mature tissue of ectodermal (skin, brain), mesodermal (muscle, fat) and endodermal (mucinous or ciliated epithelium) origin.[11] Serous tumours, of which 50% are usually seen in women <40 years of age,[12] develop by invagination of the surface epithelium of the ovary and secrete serous fluid. Serous tumours are generally benign; 5 - 10% have borderline malignant potential, and 20 - 25% are malignant. Serous cystadenomas usually are multilocular and sometimes have papillary projections.[12] Giant ovarian tumours are becoming rare in current medical practice especially in developed countries where most cases are discovered early during routine check-ups or ultrasonography. However, in developing nations such as Nigeria where access to healthcare is poor, coupled with a low literacy level, neglected giant ovarian tumours are still encountered. In our case, the patient had a low literacy level and had been threatened with divorce, which led to an initial delay in intervention. Giant ovarian teratomas commonly present with acute abdominal pain caused by adnexal torsion and abdominal distension due to tumour growth. Some patients, as in ours, may also have nonspecific abdominal complaints from pressure effects of the mass such as dyspnoea, dyspepsia and anorexia.[13]

A review of the literature revealed that giant ovarian tumours were reported in some studies more than 100 years ago. In 1905, Spohn[14] reported on a 43-year-old woman with a simple abdominal cyst that was drained preoperatively over 7 days, yielding ~114 L of gelatinous fluid. The tumour was estimated to be 146 kg.[15,16] In 1994, Oâ&#x20AC;&#x2122;Hanlan[15] removed the largest known cystic tumour, weighing 137.4 kg. Madhu et al.[16] recently reported a complete resection of a giant ovarian tumour weighing 57 kg, similar to our patient. To our knowledge, our patient had the largest tumour (55 kg) ever reported in Nigeria. Estimation of tumour markers such as CA125 and imaging evaluation such as magnetic resonance imaging and computed tomography were indicated in this patient; however, these were not performed due to non-availability in the hospital and the financial constraints associated with referral to other health centres. Although some giant cystic teratomas are amenable to laparoscopic removal after decompression, this is not possible in solid or giant tumours. Our patient was initially counselled for hysterectomy and bilateral salpingo-oophorectomy for fear of malignancy, but did not consent to the former. However, conservative surgery such as ovarian cystectomy and salpingo-oophorectomy is adequate for benign tumours of the ovary.[17] The incidence of malignant transformation in ovarian teratomas is 1 - 2%, and the malignant component may stimulate a separate ovarian mass.[18] Fortunately for this patient there were no histological signs of malignancy. This case attracted our attention for at least three reasons: firstly it was a synchronous tumour which occurred with a GCT and secondly, it was huge. This tumour was also among the largest ever reported in Nigeria. The massive dimension can be attributed to a delay in seeking care, superstitious beliefs, fear of divorce and dying from surgical intervention.

Conclusion

Neglected ovarian tumours are still encountered in sub-Saharan Africa and may involve diagnostic and management challenges. Owing to poverty and illiteracy, patients in rural settings often seek medical advice at a very late stage in disease progression. Our patient delayed surgical intervention for 10 years before finally consenting to have the tumour removed. Had the tumour been malignant, she would have presented terminally ill with advanced disease or she may have died from its complications. There is a need to educate women and their communities on ovarian neoplasms and the need to present early for effective management. 1. Tong SY, Lee YS, Park JS, Bae SN, Lee JM, Namkoong SE. Clinical analysis of synchronous primary neoplasms of the female reproductive tract. Eur J Obstet Gynecol Reprod Biol 2008;136(1):78-82. https://doi.org/10.1016/j.ejogrb.2006.09.010 2. Dudnikova A, Veriozkey I, Pete I. Synchronous tumours of female genital tract: Triple malignant and one benign tumour. Hung Onkol 2012;56(1)55-59. 3. Shanmughapriya S, Senthil Kumar G, Balakrishnan K, Vasanthi N, Vinodhini K, Natarajaseenivasan K. Bilateral ovarian teratoma complicated with carcinosarcoma in a 68 year old woman: A case report. BMC Cancer 2011;11(1):218. https://doi.org/10.1186/1471-2407-11-218 4. Oliver D, Vicki VB. Premalignant and malignant diseases of the ovaries and oviductcts. In: Decherney AH, ed. Current Obstetrics and Gynecologic Diagnosis and Treatment. 9th ed. New York: Lange Medical Books, 2003:933-958. 5. Samuel CJ, George LJ. Sonographic diagnosis of multiple unilateral teratomas. J Ultrasound Med 2001;20(3):279-281. https://doi.org/10.7863/jum.2001.20.3.279 6. Damewood M, Rosenshein NB, Woodruff JD. Multiple benign cystic teratomas of the ovary. Diag Gynecol Obstet 1982;4(3):243-245. 7. Bachhav AA. Serous cystadenoma with co-existing stromal tumour with sex-cord stromal elements: An extremely rare tumour. Open J Obstet Gynecol 2014;4(3):105-108. https://doi.org/10.4236/ ojog.2014.43019 8. Jayalakshmy PS, Poothiode U, Krishna G, Jayalakshmy PL. Ovarian fibroma with serous cystadenoma â&#x20AC;&#x201C; an unusual combination: A case report. Case Rep Obstet Gynecol 2012;2012:1-3. https://doi. org/10.1155/2012/641085 9. DiSaia PJ, Creasman WT. Epithelial ovarian cancers. In: Disala ED, Creasman WT, eds. Clinical Gynecologic Oncology, 4th ed. Saint Louis: Mosby, 1993:333-334.

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CASE REPORT 10. Serov SF, Scully RE, Sobin, LH. Histological Typing of Ovarian Tumours. Geneva: WHO, 1973:17-18. 11. Outwater EK, Siegelman ES, Hunt JL. Ovarian teratomas: Tumor types and imaging characteristics. RadioGraphics 2001;21(2):475-490. https://doi.org/10.1148/radiographics.21.2.g01mr09475 12. Dey M, Pathak N. Giant serous papillary cystadenoma. Med J Armed Forces India 2011;67(3):272273. https://doi.org/10.1016/s0377-1237(11)60059-2 13. Ye L-Y, Wang, J-J, Liu D-R, Ding G-P, Cao L-P. Management of giant ovarian teratoma: A case series and review of the literature. Oncol Lett 2012;4(4):672-676. https://doi.org/10.3892/ol.2012.793 14. Spohn AE. Multicystic ovarian tumour weighing 328 pounds. Texas Med J 1905:1273-1274.

15. Oâ&#x20AC;&#x2122;Hanlan KA. Resection of a 303.2-pound ovarian tumor. Gynecol Oncol 1994;54(3):365-371. https://doi.org/10.1006/gyno.1994.1225 16. Madhu YC, Harish K, Gotam P. Complete resection of a giant ovarian tumour. Gynecol Oncol Case Rep 2013;6:4-6. https://doi.org/10.1016/j.gynor.2013.05.001 17. Kamel RM. A massive ovarian mucinous cystadenoma: A case report. Reprod Biol Endocrinol 2010;8(1):24. https://doi.org/10.1186/1477-7827-8-24 18. Ahmed SE, Ayat SAM. Multiple bilateral huge synchronous ovarian mature cystic teratomas: A rarely encountered condition in practice. Egypt J Radiol Nucl Med 2014;46(1):195-197. http://doi. org/10.1016/j.ejrnm.2014.11.001

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1

CPD

True (A) or false (B): Puerperasâ&#x20AC;&#x2122; knowledge of postnatal exercises 1. It is possible that postpartum exercises may reduce postpartum complications such as haemorrhage and sepsis. Knowledge, attitudes, and practice in emergency contraception 2. Globally there were an estimated 42 million induced abortions in 2001, of which 48% were unsafe. 3. The World Health Organization estimated that between 2003 and 2008 the number of unsafe abortions in the world increased, with the highest rates being in eastern and central Africa. 4. A review of 1 077 articles from 113 newspapers published in 2005 demonstrated a confusion between emergency contraception and medical abortion in almost 45% of those articles. Abnormal umbilical Doppler and placental histology in fetal growth restriction 5. Maternal underperfusion lesions include villous infarction, villous agglutination, syncytial knots and intervillous fibrin deposition. 6. In the study included in this journal, abnormal Doppler had a sensitivity of 65% and specificity of 80% for abnormal placental pathology. Preventive role of low-molecular-weight heparin in early pregnancy loss 7. There are varying definitions of recurrent pregnancy loss, which include two or three pregnancy losses. 8. The risk of miscarriage increases to up to 30% after two consecutive early pregnancy losses. 9. Recurrent pregnancy loss remains unexplained in up to 50% of couples investigated. 10. In the study included in this journal, subcutaneous enoxaparin in a dose of 40 mg once daily improved the chance of live births in non-thrombophilic women with unexplained recurrent pregnancy loss when compared with a placebo.

Galactogogue use and breastfeeding behaviour 11. The global rate of exclusive breastfeeding in the first six months of life is reported as 38%. 12. A recent Cochrane review concluded that the use of galactogogues was associated with a significant improvement in longer term outcomes of breastfeeding in preterm infants. Galactogogues are dopamine agonists such as antipsychotics, 13. antiemetics and natural supplements which enhance lactation by increasing serum prolactin. 14. In high-income settings, galactogogue use has been reported in over 30% of parturient mothers. Beliefs and practices in using misoprostol for induction of labour in Zimbabwe 15. A global survey published by the WHO in 2014 stated that induction of labour was performed in 9.6% of all births. 16. There is currently a clear international definition of failed induction of labour. Embolisation of pseudoaneurysm of the uterine artery 17. Secondary postpartum haemorrhage is defined as haemorrhage occurring 6 hours to 6 weeks postpartum. Bilateral synchronous benign ovarian neoplasms 18. The most common germ cell tumours of the ovary are benign mature cystic teratomas. 19. Benign mature cystic teratomas of the ovary are composed of endoderm and ectoderm, but not mesoderm. 20. Metastases to the ovary constitute 5% of ovarian neoplasms.

The CPD programme for SAJOG is administered by Medical Practice Consulting: Effective in 2014, the CPD programme for SAJOG will be administered by Medical Practice Consulting: CPD questionnaires must be completed online at www.mpconsulting.co.za CPD questionnaires must be completed online at www.mpconsulting.co.za A maximum ofwill 3 CEUs will be awarded per correctly completed test. A maximum of 5 CEUs be awarded per correctly completed test. Accreditation MDB015/173/02/2017(Clinical) Accreditation number:number: MDB015/178/02/2016(Clinical)

SAJOG â&#x20AC;˘ May 2017, Vol. 23, No. 1