PURPOSE: to evaluate all maternal deaths that occurred between 1927 and 2001, among 164,161 patients admitted to the Maternidade Therezinha de Jesus, the obstetrical service of the "Universidade Federal de Juiz de Fora", Brazil. METHODS: a retrospective study of 144 maternal deaths that occurred in the maternity hospital in 75 years, with 131,048 live births in the same period of time, analyzing all patients's records regarding their clinical history and data from death certificates. Autopsies were not performed. Data obtained were age, parity, gestation length, complications, moment, and causes of death. The index of maternal mortality (IMM) period 100 thousand live births was utilized. For statistical analysis the chi2 test and the exponential smoothing technique were used (alpha=0.05). RESULTS: IMM decreased from 1544 in the period 1927-1941 to 314 (p<0.001) between 1942 and 1956 and from 1957 to 1971 it was reduced to 76.4 per 100 thousand live births (p<0.001). Nevertheless, since 1972 there was no further significant improvement (IMM=46 in the last 15 years, p=0.139). Maternal mortality was more frequent in the 15 to 39 years age group, in nulliparous patients with term pregnancies and mostly in the immediate postpartum period (53%). Direct obstetric causes occurred in 79.3% and indirect causes in 20.7% of the cases. Analyzing the evolution of the causes of death, it was found that in the first period of time the most frequent direct obstetric causes in descending order were puerperal infection, eclampsia and uterine rupture, while in the second period they were prepartum hemorrhage and eclampsia, and from 1977 to 2001 hemorrhage, abortion and preeclampsia. Analysis of the past 15 years showed the absence of maternal deaths by either preeclampsia or puerperal infection and the main causes were peripartum hemorrhage, abortion and indirect obstetrical causes. Relating maternal mortality to the type of delivery by the relative risk between cesarean section and vaginal delivery, it was found that when the indication of cesarean section is inevitable its risk is lower (relative risk = 0.6) than through vaginal delivery. CONCLUSIONS: despite the reduction along the 75 years of study, maternal mortality of 46 per 100,000 live births is still very high, and there was no significant decrease since 1972. Many deaths are avoidable. Hemorrhage is at present the most frequent cause of maternal death, the decision to intervene should be fast, and a proper indication for a cesarean section is a safe option. Maternal mortality caused by abortion is increasing alarmingly and family planning is essential.
Maternal mortality; Abortion; Hemorrhage; Preeclampsia; Eclampsia; Puerperal infection; Retrospective studies
