The Perinatal Conferences Are Held Under the Auspices of The

The Perinatal Conferences are held under the auspices of the

Department of Obstetrics and Gynaecology, University of Pretoria

and sponsored by Abbott Laboratories SA (Pty) Ltd

Proceedings : 16th Priorities in Perinatal Care

Editor's Note:

An editorial board was appointed to look at the papers in this year's Proceedings. Where there were marked errors in the papers these were referred back to the author for correction. If, however, these were not corrected the paper has been printed in its original form. The editors thus accept no responsibility for any mistakes found. If corrections were suggested by the editorial board and these did not in any way affect the meanings intended in the paper, these changes were made without contacting the author. We apologise if this causes offence but, in view of the amount of administrative work involved in contacting the relevant authors, it was felt this was the easiest course to follow.

Proceedings : 16th Priorities in Perinatal Care

TABLE OF CONTENTS

CONSENSUS STATEMENTS

ANTENATAL SCREENING FOR SYPHILIS. SD Delport...... 1

DISCUSSION DOCUMENT. NEONATAL ETHICS. DE Ballot...... 3

CONTINUING EDUCATION

STAFF TRAINING FOR PERINATAL AUDIT. S Bergström...... 6

THE IMPACT OF THE MATERNAL CARE MANUAL OF THE PERINATAL EDUCATION

PROGRAMME ON THE PRACTICAL SKILLS OF MIDWIVES. GB Theron...... 10

SUCCESS WITH THE PERINATAL EDUCATION PROGRAMME. D Woods...... 12

PREPARING AND MAINTAINING SYSTEMATIC REVIEWS - THE SA COCHRANE CENTRE

(SAFRICC). C Nikodem...... 14

PROMOTING EVIDENCE-BASED PRACTICE IN DEVELOPING COUNTRIES - THE WHO REPRODUCTIVE HEALTH LIBRARY. GJ Hofmeyr 17

HIGHER EDUCATION RESOURCE ORGANISATION (HERO). D Woods...... 19

LABOUR

SUBAPONEUROTIC HAEMORRHAGES - A PREVENTABLE PERINATAL PROBLEM. H Saloojee 20

INTRAPARTUM FETAL DEATHS AT BARAGWANATH - HOW PREVENTABLE? EJ Buchmann,.22

VAGINAL MISOPROSTOL FOR INDUCTION OF LABOUR. S Bergström...... 25

MISOPROSTOL FOR THIRD STAGE OF LABOUR MANAGEMENT: A DOUBLE BLIND,

PLACEBO CONTROLLED CLINICAL TRIAL. L Rose...... 28

THE USE OF DEXATHASONE IN WOMEN WITH PRETERM PREMATURE RUPTURE OF MEMBRANES: A MULTICENTRE, PLACEBO CONTROLLED RANDOMISED CONTROLLED TRIAL.

The DEXIPROM Study Group...... 31

FETAL AND INFANTS GROWTH

FETAL GROWTH AND LOW BIRTH WEIGHT IN MOZAMBICAN PREGNANT WOMEN.

NB Osman...... 34

GROWTH AND DEVELOPMENT OF VERY LOW BIRTHWEIGHT INFANTS TO AGE FIVE YEARS.

PA Cooper...... 36

MINIMAL ENTERAL FEEDING : AN IMPORTANT ROLE FOR MILK IN THE GUT. SJ Newell....39

IS EARLY DISCHARGE OF PREMATURE BABIES SAFE? M Mokhachane...... 42

MATERNAL NUTRITION AND LOW BIRTH WEIGHT. K Kyriazis...... 45

ANTENATAL CARE

CAN DOPPLER FLOW STUDIES EXPLAIN THE PRESENCE OF OLIGOHYDRAMNIOS IN POST TERM

PREGNANCIES? S Morris...... 47

THE VALUE OF ROUTINE HEART AND LUNG AUSCULTATION IN ANTENATAL CARE.

E Divanovic...... 50

AN INVESTIGATION OF ANTENATAL PATIENTS REPORTING REDUCED FETAL MOVEMENTS.

A Theron...... 52

FETAL VIBRO-ACOUSTIC STIMULATION WITH A CAN: A CLINICAL STUDY. M de Jager.....54

INFECTIONS

TRANSMISSION OF HIV FROM MOTHER TO CHILD, STRATEGIES FOR PREVENTION.

GE Gray...... 57

PERINATAL HIV: A NEW PERSPECTIVE. THE PRETERM NEONATE - AN EMERGING PROBLEM.

M Adhikari...... 62

BACTERIAL VAGINOSIS, LOW BIRTH WEIGHT INFANTS AND PERINATAL DEATH

LR Pistorius...... 64

HIDDEN PREGNANCY WASTAGE ATTRIBUTABLE TO SYPHILIS IN RURAL SOUTH AFRICA.

D Wilkinson...... 67

THE ANTENATAL PREVALENCE AND SEROCONVERSION RATE OF SYPHILIS IN THE TYGERBERG

HOSPITAL AREA: 1993 VS 1996. GB Theron...... 69

EPIDEMIOLOGY

SCREENING FOR CERVICAL NEOPLASIA DURING PREGNANCY. IS IT WORTHWHILE?

HS Cronje...... 72

THE NORTHERN CAPE SAFE MOTHERHOOD NEEDS ASSESSMENT. JA McIntyre...... 74

PROSPECTIVE RISK OF STILLBIRTH COMPARED TO CONVENTIONAL STILLBIRTH RATES FOR THE ATTERIDGEVILLE POPULATION. GR Howarth 77

PROFILE OF A MOU : WORKLOAD, PROBLEMS AND OUTCOMES. DH Greenfield...... 78

WOMEN WHO BOOK LATE IN PREGANCY. D Roussot...... 80

HYPERTENSION

HYPERHOMOCYSTEINAEMIA IN ABRUPTIO PLACENTAE. AA Carolissen...... 83

DOES BETAMETHASONE TREATMENT, TO ENHANCE FETAL LUNG MATURITY IN PATIENTS WITH

SEVERE PROTEINURIC HYPERTENSION LESS THAN 34 WEEKS' GESTATION, HAVE ANY ADVERSE EFFECTS ON THE BASELINE VARIATION AND OTHER PARAMETERS OF THE FETAL HEART? G van Rensburg 85

ORAL KETANSERIN IN MILD MIDTRIMESTER HYPERTENSION - A RANDOMISED CONTROLLED TRIAL.

DW Steyn...... 87

POSTERS

THE CONFIDENTIAL ENQUIRY INTO MATERNAL DEATHS - THE MATERNAL DEATH NOTIFICATION FORM. E Retief 89

MATERNAL MORBIDITY : A DEFINITION FOR A MATERNAL "NEAR MISS". GD Mantel...... 91

THE EFFECT OF THE PERINATAL PROBLEM IDENTIFICATION PROGRAMME IN THE WITBANK

DISTRICT IN 1996. D Kotzé...... 93

THE IMPORTANCE OF LOCAL AUDIT. M Muller...... 98

SEEKING SALEABLE SOLUTIONS: RESULTS OF A REGIONAL PERINATAL AND MATERNAL

MORTALITY AUDIT. RC Pattinson...... 101

A TEN YEAR REVIEW OF BLOOD CULTURE ISOLATES FROM THE NEONATAL UNIT OF GA-

RANKUWA HOSPITAL 1987-1996. M Driessen...... 104

SCREENING FOR PREGNANCY BACTERIURIA. ET Bvuma...... 106

PREVENTION OF REPEAT ABRUPTIO PLACENTAE. L Schoeman...... 108

Proceedings : 16th Priorities in Perinatal Care

CONSENSUS STATEMENTS

Proceedings : 16th Priorities in Perinatal Care

ANTENATAL SCREENING FOR SYPHILIS

SD Delport, RC Pattinson

Department of Paediatrics & Child Health and Obstetrics & Gynaecology, Kalafong Hospital and University of Pretoria

Near patient screening for syphilis should be performed in all pregnant women. This implies performing a screening test in the clinic (on site) or in a nearby laboratory so that the results are available before women leave the clinic or consulting room.

1.All pregnant women should be screened for syphilis at the time their pregnancy is diagnosed.

Motivation:

1.1Many women seek antenatal care late during pregnancy.

1.2The screening test for syphilis is easy to perform, quick and cheap.

1.3Similarly the treatment of maternal syphilis is cheap and prevents congenital syphilis - the cause of 10% of perinatal deaths.

2.Results of the screening test should be reacted on and given to women before they leave the clinic or the consulting room. Women with a positive test must be treated immediately.

Motivation:

2.1Women may not return for a 2nd visit due to social and financial constraints.

2.2Immediate counselling and treatment can be offered. (Patients should not be referred to STD clinics).

2.3Any delay in treatment increases the risk of fetal death.

3.The results of positive screening tests performed on site must be verified in a laboratory and a titre determined. The titre must be documented on the motherhood card. A treponemal test (TPHA or FTA-ABS) should be performed if resources permit.

Motivation :

3.1A baseline titre of a screening test is necessary to determine success of treatment (a declining follow-up titre) or a reinfection, and increasing follow-up titre).

3.2A positive treponemal test in addition to a positive screening test rules out a false positive screening test.

Screening tests for syphilis

1.Rapid plasma reagin (RPR) test

Proceedings : 16th Priorities in Perinatal Care

Can be performed without specialised equipment and will identify all women with RPR titres > 1:16. Negative results need not be verified in a laboratory.

2.Venereal Disease Research Laboratory (VDRL) test

Sophisticated and more sensitive than an RPR test. Not a suitable test for on-site testing.

Proceedings : 16th Priorities in Perinatal Care

DISCUSSION DOCUMENT : NEONATAL ETHICS

DE Ballot, PA Cooper

The following is a report of a discussion regarding neonatal ethics, held amongst health care workers involved in neonatal care. The problem of selecting patients for neonatal intensive care was discussed previously at the 10th Conference on Priorities in Perinatal Care. At that time it was generally agreed that a birth weight cutoff of 1000g for neonatal intensive care should be applied. It was felt that neonatal admission policies and ways of rationing tertiary care due to limited resources should be reviewed.

There was general agreement that limited health resources in South Africa necessitate the selection of neonates for intensive care. It was felt that a set of general principles could be recommended, but that specific categories should be determined by the individual departments and regions.

ASelection criteria for ICU

Possible methods include:

i)Selection on the basis of anticipated outcome (mortality/ functional and utilisation of resources

Factors which could be considered include:

- Birthweight

- Gestational age

- Severity of illness

- Severe congenital anomalies

- Chromosomal anomalies

- Birth asphyxia

Problems:

-It is difficult to prognosticate at the time of admission. Severity of illness scores e.g. CRIB, SNAP, NTISS seem attractive means for patient selection but these were not designed for this purpose. It is also impossible to accurately predict individual outcome. Many of the scores are cumbersome and require an initial period of admission and treatment.

-Selection of patients may create a self-fulfilling prophecy - sick babies do badly.

ii)Selection on the basis of "worth"

Factors which could be considered include:

- Precious baby

- Poor obstetric history

- Uncaring mother, unwanted

pregnancy

- Unbooked mother

- Poor socio-economic

circumstances

Problems:

-This is a value judgement, which is influenced by an individual's frame of reference.

-These categories of patient often do not correlate with outcome.

-The baby is disadvantaged because of the mother's actions.

iii)Selection on the basis of response to therapy

In this circumstance all babies would be resuscitated and treated for a limited period of time (e.g. 72 hours). Patients would then be assessed and ICU support would be withdrawn in those who were not responding adequately to treatment. Severity of illness scores could be useful in assessing these patients.

Problems:

-Specific criteria for withdrawal of treatment are required (see below).

-Duration of treatment must be specified.

-Withdrawal of treatment appears more aggressive than not starting tertiary care.

-There would be a substantially increased demand for ICU facilities.

-Would parents cope?

iv)Maternal HIV infection

Neonates who have acquired HIV infection from their mothers have a poor prognosis and could be a category of patients denied tertiary care. However, there is no single reliable test to diagnose infection in the neonate at birth at present.

BWithdrawal of therapy

Another means to reduce pressure on ICU beds is aggressive withdrawal of ICU support in those patients with a poor prognosis. This is currently practised in most units but is less well defined than selection of patients for admission. Patient categories include:

- Severe birth asphyxia

- Severe congenital anomalies

- Chromosomal anomalies

- Severe intraventricular

haemorrhage/ leucomalacia

- End stage renal failure

- Multi-organ failure

- Severe bronchopulmonary

dysplasia

If response to therapy is to be used as a means of selecting patients (A iii), other parameters, e.g. oxygen requirement at 72 hours would have to be incorporated into criteria for terminating therapy.

Discussion

It was generally agreed that the best method in our circumstances is to have selection criteria for initial ICU admission. Birth weight is a reasonable criterion for this purpose and most units are using a birth weight cutoff. The 1000g cutoff is widely applied, although cutoffs varied between 900 and 1250 grams. Birth weight remains preferable to gestational age as accurate assessment of gestational age is often unavailable. Studies on neonatal outcome have shown that birth weight and gestational age are closely related.

It was agreed that other factors in addition to birth weight could be considered including:

-Maternal booking status

-Obstetric history and intervention (e.g. in vitro fertilisation)

-Interference in the pregnancy

-Unwanted pregnancy

-Outborn babies

-General condition of the baby (asphyxia, severity of illness, need for resuscitation)

It was suggested that instead of an absolute 1000g cutoff, a graded approach could be used:

-<900 - no ICU support

-900 to 1200g - selective admission of babies to ICU (see above)

->1200g - ICU admission

It was generally felt that all babies should be initially resuscitated and then assessed for ICU admission. The decision to withhold ICU support should be made at senior level. It was emphasised that those babies who are denied ICU admission would still receive level 2 care, including warmth, oxygen, nutrition and antibiotic therapy.

It was agreed that active withdrawal of ICU support in patients with poor prognosis is appropriate. This decision should be made by the medical and nursing staff caring for the baby in conjunction with the parents. The ultimate decision, however, should remain the responsibility of the medical staff and not the parents. Ethics committees may help to review the options but would not take the ultimate decision to withdraw therapy.

It was suggested that actual practices regarding NICU admission and withdrawal of therapy should be formally evaluated by means of a questionnaire. The results would be made available and would form part of the general guidelines.

The public sector is usually unable to provide support to those parents in the private sector who run out of funds. It would therefore seem appropriate for private practitioners to be made fully aware of policies in public sector hospitals and to agree to refer to public sector hospitals patients who conform to these policies.

Proceedings : 16th Priorities in Perinatal Care

STAFF TRAINING FOR PERINATAL AUDIT

S Bergström

Division of International Health Care (IHCAR), Karolinska Institutet, Stockholm, Sweden

1.The concept of audit

In its original sense audit implies a critical review of events, particularly economic events. It has been defined as "a numerical review by an outside investigator directed at, among other things, the prevention of fraud"(1). Clinical audit does not aim at the prevention of fraud but aims at detecting mismanagement and degree of avoidability of adverse outcomes of clinical management.

The retrospective scrutiny of clinical events, the re-view, is an action close to "re-search". Clinical audit can rightly be considered an action-orientated research, aiming at operationalising findings in order to correct mismanagement, and to improve norms with a focus on avoidability issues.

2.Perinatal audit

Unfortunately, much of perinatal audit exercises tend to focus more on fetal/neonatal outcome than on maternal outcome. It should be underscored, however, that more than half of all maternal deaths in most studies occur in the perinatal period, from the 28th week till the end of the 7th day postpartum. The word "perinatal" alludes only to the period surrounding birth, and does not exclude the mother or the fetus/newborn. Strictly speaking, perinatal mortality, should be specified to be "perinatal fetal/infant mortality" or "perinatal maternal mortality". Audit activities in the perinatal period should therefore cover both maternal and fetal/ neonatal events.

Initially most of perinatal audit efforts should be geared towards auditing deaths, which in itself is an obvious priority. However, it is equally relevant to audit and equally interesting to define the most severe morbidity occurring in the perinatal period both for the mother and the fetus/ newborn, the category of events that we can call "near misses". Apart from such severe morbidity it may also be useful and instructive to include all forms of significant and measurable morbidity among women and newborns in order to gradually improve management and quality of care.

3."The mailbox syndrome"

A well-known phenomenon in most low-income countries is the reasonable and conventional demand from health authorities, ministries and various health authorities to the lower levels that they deliver health statistics for central accumulation of data. Collecting data from peripheral levels of health care without their proper "digestion" on the spot where they can be registered has been said to constitute "the mailbox syndrome". The delivery of collected data to a central authority, like the Ministry of Health, may "legitimise" exportation of data to next higher level instead of importing data to quality-assuring and quality-improving staff activities within the health unit, where the data were elaborated and collected. The latter activities precisely constitute audit, implying visualisation of clinical events for internal consumption, digestion and contemplation. Even if the need for these data in central authorities is perfectly legitimate, a still more important and operation-orientated utilisation is, obviously, in the health unit generating the data. This is particularly important as concerns perinatal data, since the outcome of two individuals is involved.

4.Prerequisites for perinatal audit

A complete and high quality data collection procedure on a daily basis is a sine qua non for a successful perinatal audit. These daily collection activities should be followed by regular, weekly or fortnightly summaries of data for careful case-by-case scrutiny aiming at avoidability analyses. This may concern death, severe morbidity or other important perinatal events. Whereas these steps in the audit procedure could be prepared by a responsible individual, the proper avoidability discussion should follow from a plenary discussion where all health staff involved are present.

Since all forms of audit aim at analysing avoidability, by which mismanagement is in focus, it may appear threatening to staff involved. It has, however, being said rightly that "clinical audit should be carried out as an internal activity, for the benefit of the group and should be under local control". Encouraging participation in and real commitment to audit activities depend upon recognising and managing health staff responses. There are several useful principles that can guide the management of audit groups to constructive working relations. Such principles may comprise the following:

1)Emphasis on local control.

2)Reasonably ambitious and achievable objectives.

3)Emphasis on the value of present and past practice.

4)Identification of current review activities.

5)Acknowledgement of participants' concerns.

It is very important for the proper function of an audit process that the audit itself is not threatening to anyone present. The audit should not produce anxiety adrenaline release and tachycardia.

It has rightly been said that "good leaders are not authoritarian, nor are they shrinking violets. They believe in forging consensus and then helping everyone share in the implementation" (1). The importance of "fostering" co-operation (2) cannot be underscored enough. The following quotation gives a succinct description of some core elements in a good perinatal audit (2): "Co-operation and understanding cannot be taken for granted within groups. Constructive interaction often needs to be nurtured. This requires skilled contributions especially from those who hold leadership responsibilities (such as chair persons). Key activities include: