Author(s): Lilford RJ, van Coeverden de Groot HA, Moore PJ, Bingham P
OBJECTIVE: To compare maternal mortalities attributable to vaginal delivery, elective caesarean section (CS) and intrapartum CS. DESIGN: The number of deaths associated with each method of delivery was ascertained among unselected and among low-risk women by detailed retrospective review of the case-notes of women who died after delivery. The frequency of each method of delivery throughout the study period was ascertained from the computer database and enhanced by analysis of the case-notes of unselected groups of women. SETTING: The Peninsula Maternity Services (Cape Town) during the years 1975-1986 inclusive. SUBJECTS: A total of 108 maternal deaths arising from 263,075 maternities provided accurate information. The relative frequency of vaginal and abdominal delivery was determined from the computer database. The ratio of elective CS to emergency prepartum CS to intrapartum CS was obtained by review of the first 200 operations in the years 1975, 1977, 1979, 1982 and 1984. MAIN OUTCOME MEASURES: (i) Mortality rates associated with the different methods of delivery in unselected women and in women who were healthy before surgery; (ii) mortality rates apparently attributable to the method of delivery. RESULTS: The overall relative risk of mortality associated with caesarean section compared with vaginal delivery was 7 decreasing to 5 after the exclusion of women with medical or life-threatening antenatal complications (eg, haemorrhage, hypertension). The relative risk associated with intrapartum compared with elective sections was 2.3 decreasing to 1.4 after the exclusion of women with medical disorders or life-threatening complications. The relative risk of maternal mortality which was apparently attributable to intrapartum compared with elective sections was 1.7. However, the 95% confidence intervals of these values, even from this large data-set, are wide. Nevertheless, these rates are in broad agreement with an approximation derived from the British confidential enquiries into maternal deaths. CONCLUSION: The attributable relative mortalities of caesarean section compared with vaginal delivery and intrapartum compared with elective caesarean section are lower than the overall relative mortalities of these modes of delivery and are approximately 5:1 and 1.5:1 respectively. These data are crucially important in the decision to recommend elective caesarean section compared with trial of labour.