O objetivo deste trabalho foi descrever o perfil epidemiológico e a tendência da mortalidade materna no Brasil, por meio de revisão de estudos sobre o tema. Foi realizada busca eletrônica de artigos científicos publicados entre 1980 e 2010, nas bases de dados LILACS e MEDLINE. Identificaram-se inicialmente 486 artigos. Após aplicação dos critérios de inclusão e exclusão restaram 50 artigos. A comparação dos dados mostrou queda da RMM de forma diferenciada nas regiões brasileiras. Os estudos sobre determinação do óbito materno apontaram desigualdades sociais relacionadas à cor da pele e escolaridade. O preenchimento incompleto da declaração de óbito e a subnotificação ainda persistem. Prevaleceram as causas obstétricas diretas, com predomínio das doenças hipertensivas. Quando analisada, a evitabilidade apontou falhas na assistência pré-natal e ao parto. Apesar de sua relevância, são poucos os artigos sobre mortalidade materna no Brasil. A RMM, embora em declínio, permanece em níveis elevados. Melhorias na qualidade da assistência pré-natal e ao parto são necessárias.

Objective: To investigate severe maternal morbidity/near misses in a tertiary public maternity in the state of Rio de Janeiro, using different identification criteria. Methods: This is a cross-sectional study, performed in a regional reference hospital between June and October 2009, on severe maternal morbidity/near miss cases identified from the log books of the maternity hospital and review of medical records. This study focused on women who, during pregnancy, delivery, or the postpartum period, showed no clinical symptoms compatible with the defining criteria for severe maternal morbidity/near miss of Waterstone et al, Mantel et al. and the World Health Organization (WHO). Results: Among the 1,544 admissions during the period studied, 89 women with severe maternal morbidity were identified, considering all criteria. The occurrence of severe maternal morbidity/ near misses ranged from 81.4 to 9.4 per 1,000 live births (LB), depending on the criterion used. The mortality rate was 3.2%, reaching 23% in the WHO criteria. Only 40% of these women had more than six prenatal visits and 10% did not have any visit at all. The most common markers found were severe preeclampsia, followed by severe hemorrhage, ICU admissions, HELLP syndrome, and eclampsia. There were three maternal deaths with a MMR = 280/100.000 LB and one late death. The WHO criterion showed greater specificity, identifying more severe cases, while the Waterstone criterion was more sensitive. Conclusions: The study of severe maternal morbidity/near misses in a regional reference hospital can contribute to the knowledge of this event's magnitude, as well as to identify its most frequent characteristics and clinical conditions, being essential for dealing with maternal morbidity and mortality.