Neonatal mortality associated with perinatal asphyxia: a population-based study in a middle-income country

Abstract Background It is challenging to decrease neonatal mortality in middle-income countries, where perinatal asphyxia is an important cause of death. This study aims to analyze the annual trend of neonatal mortality with perinatal asphyxia according to gestational age in São Paulo State, Brazil, during a 10-year period and to verify demographic, maternal and neonatal characteristics associated with these deaths. Methods Population-based study of neonatal deaths associated with perinatal asphyxia from 0 to 27 days in São Paulo State, Brazil, from 2004 to 2013. Perinatal asphyxia was considered as associated to death if intrauterine hypoxia, birth asphyxia or neonatal aspiration of meconium were noted in any line of the Death Certificate according to ICD-10. Poisson Regression was applied to analyze the annual trend of neonatal mortality rate according to gestational age. Kaplan-Meier curve was used to assess age at death during the 10-year study period. Hazard ratio of death during the neonatal period according to gestational age was analyzed by Cox regression adjusted by year of birth and selected epidemiological factors. Results Among 74,002 infant deaths in São Paulo State, 6648 (9%) neonatal deaths with perinatal asphyxia were studied. Neonatal mortality rate with perinatal asphyxia fell from 1.38‰ in 2004 to 0.95‰ in 2013 (p = 0.002). Reduction started in 2008 for neonates with 32–41 weeks, in 2009 for 28–31 weeks, and in 2011 for 22–27 weeks. Median time until 50% of deaths occurred was 25.3 h (95%CI: 24.0; 27.2). Variables independently associated with higher risk of death were