Article révisé par les pairs
Résumé : Background: The aim of this study is to establish factors explaining perinatal death rates in the city of Lubumbashi. Methods: We have carried out a case controlled study in the maternity ward of Jason Sendwe hospital. Perinatal death cases have been compared to those of surviving newborn children among parturient women in the course of 2008. Sociodemographic characteristics, maternal morbidity, children's typical features, have been studied as independent variables. Their effect on perinatal mortality has been assessed using an adjusted odds ratio value at a 5% confidence interval and a logistic regression model. Results: In total, we considered 2279 births (mother and child pairs) for our study. Among these were 415 perinatal mortality cases and 1864 control cases. After adjustment for several parameters, household chores (AOR = 1.8; 95% IC = 1.2-2.9), multiple pregnancies (AOR = 1.9; 95% IC = 1.2-2.9), malaria (AOR = 1.4; 95% IC = 1.1-1.8), primiparity (AOR = 1.7; 95% IC = 1.3-2.4), stillbirth (AOR = 5.2; 95% IC = 2.5-11.0) and prematurity (AOR = 2.9; 95% IC = 1.5-5.5) in previous pregnancies, onset of antepartum ferver (AOR = 3.0; 95% IC = 1.2-7.3) and antepartum hemorrhage (AOR = 6.8; 95% IC = 3.1-15.0), lack of fetal motions near delivering time, dystocias (AOR = 2.0; 95% IC = 1.3-3.0), low birthweight (AOR = 15.7; 95% IC = 11.2-22.0), very low birthweight (AOR = 49.0; 95% IC = 28.6-85.1) and foetal macrosomia (AOR = 3.5; 95% IC = 1.8-7.0) were the main factors explaining perinatal mortality. Conclusion: Perinatal mortality in Lubumbashi remains associated with several avoidable factors. Basic and emergency obstetrical-neonatal care (B-EMONC) should be improved. Significant efforts should be made in this direction. Perinatal audits should be established for a good heath care quality follow-up. Obstetrical care should be offered as a continuum in order to facilitate communication between the different caregivers. © 2012 Elsevier Masson SAS.