Article

Why do women die during Childbirth?

Robert C. Pattinson, Hilde Vandecruys, A.P. Macdonald, Gerald D. Mantel*
MRC Maternal and Infant Care Strategies Research Unit, O&G Department, University of Pretoria
* Department of Obstetrics and Gynaecology, University of Natal, King Edward VIII Hospital, Durban, South Africa

Despite global advances in healthcare, women are still dying during and shortly after childbirth. Why? The Medical Research Council of South Africa recently undertook a 2- year audit maternal morbidity and near miss cases (where a women with organ failure or dysfunction would have died had not luck or good care been on her side).  This review aims to answer these questions, to show the feasibility of studies such as these and to give some recommendations to health care workers on how to prevent maternal morbidity.  

IN BRIEF: An audit of severe acute maternal morbidity and maternal mortality was conducted in the Pretoria Academic Complex for the year 2000 and the findings compared with a similar study performed over two years between 1997-9. There has been a significant reduction in referrals from outside the Pretoria Region due to the establishment of a regional hospital in Mpumalanga Province at Witbank. Severe acute morbidity and mortality has dramatically declined in women with abortions and this is related to the liberalisation of termination of pregnancy laws in South Africa and the provision of the services by private clinics. Complications due to hypertension in pregnancy and obstetric haemorrhage have increased and the underlying causes need to be identified and rectified. Complications due to pregnancy related sepsis and non-pregnancy related sepsis have increased and is directly related to the increasing prevalence of HIV infected pregnant women. Pregnant women with AIDS who develop severe acute morbidity have a very high chance of subsequently dying (with a current conversion rate of 75%). This factor will be very important when allocating scare adult intensive care resources. 

INTRODUCTION

Severe Acute Maternal Morbidity (SAMM) also known as "near miss" case means a woman with organ dysfunction or failure who would have died had it not been that luck or good care was on her side. These clinical definitions are simple to use and allow for identification of SAMMs. There are nearly five times as many SAMMs as maternal deaths and by combining the two, a meaningful audit of maternal care can be performed at frequent intervals.

The aim of this study was to track changing patterns of severe acute maternal morbidity and maternal mortality in the Academic Complex of Pretoria. By performing this audit, priorities for future interventions in the Pretoria Academic Complex can be identified and acted on.


METHODS

The Pretoria Academic Complex serves comprises 4 hospitals, two of which receive tertiary referrals from outside the Gauteng Province. The data were separately analysed for women living in the Pretoria Region and those referred to our complex. The data of the women living in the Pretoria Region provide population based data.

SAMMs and maternal deaths were identified at daily audit meetings and a simplified SAMM audit form was completed for all cases fulfilling the definition of "near misses" (1). The audit was performed from 1.01.2000 to 31.12.2000. The data were compared with the data obtained from the original two-year audit from the Pretoria Academic Complex (1.02.1997-31.01.1999). 

The definitions for primary obstetric causes are the same as those used for the confidential enquiry into maternal deaths in South Africa (2). Non-pregnancy-related sepsis as a primary cause of SAMM or maternal death are cases where the pregnant women has an infection that is incidental to the pregnancy, for example malaria, tuberculosis, or AIDS. Pregnancy-related sepsis is those cases where the sepsis is related to the pregnancy, for example septic abortions or puerperal sepsis. AIDS is only given as a primary cause if the woman has an AIDS defining condition or a CD4 count of less than 200. HIV infection is not regarded as a primary cause unless AIDS is diagnosed. A woman who has for example a septic abortion and who is also HIV infected, but does not have the criteria for diagnosis of AIDS is classified as septic abortion for the primary cause.

Data is expressed as percentages and rate/100 000 births. Only standard statistical techniques were used. The conversion rate is introduced here for the first time. The conversion rate is the number of maternal deaths / SAMMs + maternal deaths and expressed as a percentage. It reflects the proportion of women who present as a SAMM and subsequently dies. It gives an indication of how successful clinicians are in treating that particular complication.

RESULTS AND DISCUSSION

The total number of births in the Pretoria Academic Complex in the first 2-year period was 26152 and for the year 2000 was 13854 births. The total number of SAMMs and maternal deaths for 1997-1999 was 364 (305 SAMMs and 59 maternal deaths) and for the year 2000 was 147 (121 SAMMs and 26 maternal deaths). The equivalent data for women living in Pretoria Region, and those referred to the Complex is shown in Table 1. The data is expressed in rates per 100 000 births.


Table 1: Comparison of SAMM and Maternal Deaths
(data expressed at rate/100 000 births)

SAMM - Severe acute maternal morbidity, MD - Maternal death, NS - Non significant at p=0.05 level.

The referrals from outside Pretoria significantly declined from 105 cases per year in 1997-9 to 48 cases for 2000 (p<0.05). The reduction in referrals was most marked from the Highveld Region that in 1997-9 was responsible for the referral of 45 cases per year and in 2000 was responsible for the referral of 9 cases). This drop has coincided with the upgrading of Witbank Hospital as a referral hospital for the Highveld Region. The original report of the initial audit of SAMMs and maternal deaths (3) strongly recommended that Witbank Hospital be upgraded to serve as the first referral hospital for cases coming from Mpumalanga Province. This recommendation was made because of the long delays for patients to reach the Pretoria Academic Complex's hospitals, leading to women arriving in extremis. This recommendation has clearly been implemented. The number of SAMMs and maternal deaths at Witbank Hospital will need to be audited to see that the care there is effective.

There has been a drop in abortion complications in both SAMMs and maternal mortality. However, there is a trend to an increase in complications due to hypertension, haemorrhage and sepsis (both pregnancy related and non-pregnancy related). 

The number of Termination of Pregnancies (TOPs) in the public service in the Pretoria Health Region in 1997-1999 was 1182 per year. In the year 2000, there were 1129 TOP's in the public service, and a further 1429 performed in the private sector by the Marie Stopes clinic, from which 22% were second trimester TOP's. This private clinic was not functioning in 1997-9. The increase in the number of TOPs performed correlates with the drop in severe morbidity and mortality related to abortions, and is probably causally related.


For the conversion rates for the primary obstetric causes within Pretoria,  a  high conversion rate is shown for pulmonary embolism and non-pregnancy related sepsis. There were only 2 cases of pulmonary embolism in 1997-9 and 1 case in 2000. The high rate of maternal deaths due to non-pregnancy related sepsis reflects AIDS. The conversion rate for non-pregnancy related sepsis was 50% in 1997-9 and for 2000 was 75%. Of note is the low conversion rates for obstetric haemorrhage (antepartum plus postpartum haemorrhage) of 2.2% in 1997-9 and 3.7% in 2000. This indicates that although obstetric haemorrhage is still a common cause of SAMMs it is not a common cause of maternal death in the Pretoria Region. This is probably due to the availability of resources, expertise and good transport systems in Pretoria.

CONCLUSIONS

This study firstly illustrates that audit of severe acute maternal morbidity and maternal mortality is feasible, and the inclusion of SAMMS with maternal mortality allows for more frequent meaningful audits of maternal care to be performed. This allows for early detection of trends and this in turn allows for timeous changes in health strategies.

The initial audit report (3) highlighted two major problems. The first was for emergency obstetric care to be decentralised and the second was that the provision of TOP services in Pretoria needed to be expanded, especially for second trimester terminations of pregnancy. 

The first problem has been effectively addressed by Mpumalanga Province upgrading Witbank Hospital to a Regional Hospital and diverting the obstetric emergencies from the Province first to Witbank Hospital. 

The second problem had an unexpected solution, namely that the need for easy access to TOP was provided for by a private clinic, rather than Gauteng Province improving their provision of TOP services. This is especially evident in the case of second trimester TOPs. Unfortunately, the provision of the private service excludes the poorest of the poor from access to the service and Gauteng Province still has the obligation to provide accessible TOP service, including second trimester TOPs. However, there was a significant decrease in maternal morbidity and mortality due to a drop in abortion complications and this is correlated with the legalisation of TOP. There is a significant need for improved contraceptive services in the Pretoria Region.

The 2000 audit has highlighted that severe complications of hypertension in pregnancy, obstetric haemorrhage, pregnancy related sepsis and non-pregnancy related sepsis are increasing. The increase in infectious complications can be explained by the increase in prevalence of HIV infected pregnant women. However, the increase in severe complications due to hypertension and obstetric haemorrhage is a cause for concern. The solutions to these factors could be the implementation of new national guidelines (4) for managing hypertension and obstetric haemorrhage at all clinics and hospitals within the Pretoria Academic Complex.

The conversion rate is introduced for the first time and can be used as an indicator of the standard of care within the Pretoria Academic Complex. A rising index for example in obstetric haemorrhage could indicate a poorer quality of care, and would indicate the need for urgent further investigation. It may also be of use comparing the Pretoria Academic Complex with other institutions.

REFERENCES

1. Mantel GD et al. Severe acute maternal morbidity: a pilot study of a definition for a near- miss. British Journal of Obstetrics and Gynaecology 1998; 105: 985-990

2. Saving Mothers: Report on Confidential Enquiries into Maternal Deaths in South Africa - 1998. Published 999 by Government Printer, Pretoria. ISBN 1-875017-35-6

3. Mantel GD, Pattinson RC, Macdonald AP. Maternal mortality and severe acute morbidity (near misses) in the Pretoria Region: 1/2/1997 - 31/1/1999. A report to the Gauteng Health Department. Published MRC Maternal and Infant Health Care Strategies Research Unit, Kalafong, Pretoria

4. Policy Guidelines. Maternal, Child and Women's Health Unit, National Department of Health. In press.

Science in Africa - Africa's First On-Line Science Magazine

Copyright  2001, Janice Limson. All Rights Reserved