Severe maternal morbidity in Angola : studies on postpartum haemorrhage, jaundice and clinic-based audit

Abstract

Background: Angola has recently ended a long period of warfare, which has resulted in an almost complete collapse of public health. In Luanda, the capital, some 70,000 deliveries take place in public facilities. Maternal morbidity and mortality is among the highest in the world. Postpartum haemorrhage is the dominant cause of death. Jaundice of unknown etiology and obstructed labour with uterine rupture are also common. The obstetric network in Luanda involves a system of referrals of emergencies from 10 peripheral birth units to two central hospitals.

Objectives: The aim of these studies is to analyze part of the mortality, partially to elucidate the prevailing pathology and partially to develop safer routines in obstetric care with the focus on avoiding and treating complications and in referral of obstetric emergencies. We wanted to test a new disposable device, UnijectTM for routine administration of oxytocin as part of active management of the third stage of labour to reduce post partum bleeding. We wanted to understand the etiology of jaundice among pregnant women and by clinical audit assess the quality of care by studying the "avoidability" of uterine ruptures and time delay and partograph quality in the referral situation.

Methods: In a comparative prospective study we compared post partum, blood loss of 782 women with expectative management to 814 women with active management of the third stage of labour (AMTL), while evaluating the effect of UnijectTM. A prospective case-control design was used to study jaundice etiology and its clinical consequences. Twenty pregnant women were compared to 40 controls regarding malaria, HIV, hepatitis A, B, C, E, maternal and neonatal outcome. During two periods, women referred from three peripheral units were studied regarding travelling time and waiting time to see a midwife and to see a doctor, partograph quality, clinical outcome and accordance of diagnosis at the two levels.

Results: AMTL reduced postpartum blood loss by almost 50% and women with a severe blood loss of > 1000 nil were reduced by 80%. UnijectTM was well tolerated. Malaria and hepatic E were the most common causes found among the pregnant women with jaundice in the study and differed significantly from controls. Around 2/3 of uterine ruptures were judged to be avoidable at the hospital level. All parameters studied in the referral system showed a dramatic improvement in the second period compared to the first.

Conclusions: AMTL should be introduced as clinical routine. UnijectTM can be recommended of logistic and epidemiologic reasons. To minimize the risk for jaundice among pregnant women, malaria prophylaxis has to be implemented as routine among pregnant women. Hepatitis E reflects the basic sanitary conditions, which need to be improved with the emphasis on clean water, separated from sewage. The conclusion regarding avoidability yield information for future education and improvements regarding the deficiencies disclosed. The dramatic improvement in the referral study shows how audit as a mean of studying current practices can lead to better quality of care.