Cause-specific mortality of very preterm infants and antenatal events.

OBJECTIVE: To assess the relationship between antenatal factors and cause-specific risk of death in a large area-based cohort of very preterm infants.

STUDY DESIGN: The ACTION (Accesso alle Cure e Terapie Intensive Ostetriche e Neonatali) study recruited during an 18-month period all infants 22-31 weeks' gestational age admitted to neonatal care in 6 Italian regions (n=3040). We analyzed the data of 2974 babies without lethal or acutely life-threatening malformations. Cause-specific risks of death adjusted for competing causes were calculated, and region-stratified multiple Cox regression analyses were used to study the association between cause-specific mortality and infants' characteristics, pregnancy complications, antenatal steroids, and place of birth.

RESULTS: Deaths attributable to respiratory problems and intraventricular hemorrhage prevailed in the first 2 weeks of life, and those attributable to infections and gastrointestinal diseases afterwards. Antepartum hemorrhage was associated with respiratory deaths (hazard ratio [HR] 1.6, 95% CI 1.1-2.4), and maternal infection with deaths attributable to asphyxia (HR 32.5, 95% CI 4.1-259.4) and to respiratory problems (HR 2.8, 95% CI 1.6-5.2). Preterm premature rupture of membranes increased the likelihood of deaths due to neonatal infection (HR 1.8, 95% CI 1.0-3.1), and preterm labor/contractions of those due to respiratory (HR 1.5, 95% CI 1.1-2.0) and gastrointestinal diseases (HR 5.8, 95% CI 2.1-16.3). In addition, a birth weight z-score<-1 was associated with increasing hazards of death resulting from asphyxia, late infections, respiratory, and gastrointestinal diseases.

CONCLUSIONS: Different complications of pregnancy lead to different cause-specific mortality patterns in very preterm infants.