Background
Maternal deaths are a heavy burden in developing countries. In Senegal, 4 to 5 women die every day from maternity-related complications. The Tambacounda Region has the highest number of deaths, representing 20% of the national death burden. The objective was to study the determinants associated with these maternal deaths.

Methods
We conducted a case-control study using two controls per case. Registered maternal deaths were included in the study. A case was defined as a maternal death and a control as a pregnant woman hospitalized at the regional hospital of Tambacounda with complications and alive. Following a literature review, data were collected via questionnaire on socio-demographic information, obstetric characteristics, recorded complications, causes of death and patient care management. Data entry and analysis were performed using Epi Info 7. We calculated measures of central tendency for quantitative variables and proportions for qualitative variables. The odds ratio was used to estimate associations and the chi-square test to compare proportions. The significance level was set at p<0.05.

Results
We included 65 cases and 130 controls. The maternal mortality rate was 255 per 100 000 live births. The mean age was 24 ± 6.91 years and 25 ± 7.17 years for cases and controls respectively. Severe anemia, rural residence, eclampsia and <24 hours hospital stay was found in greater proportions among cases with respectively 56.5% versus 36.17% for controls, 92.3% versus 74.69% for controls, 26.1% versus 13.85% for controls and 58.3% versus 20% for controls. A statically significant association was found between rural residence (OR=4.05; 95% CI [1.51 - 11.03]), eclampsia (OR=2.2; 95% CI [1.04 - 4.63]), stay of less than 24 hours (OR=5.6; 95% CI [2.86 - 10.93]) and maternal death.

Conclusion
Maternal deaths are associated with several factors that can be limited by quality prenatal consultations and communication with the communities.