An evaluation of the autopsy following death in a Level IV neonatal intensive care unit

Objective: To ascertain the determinants of neonatal autopsy, define clinic al errors in the causes of death, and elucidate the possible audit and gene tic value of the autopsy following death in a Level IV neonatal intensive c are unit (NICU). Methods: A review and correlation of clinical and autopsy information in a case series of infants who died during the period 1991-97. Results: Two hundred and twenty-nine of 4057 infants admitted to the NICU d ied and 91 (39.7%) underwent an autopsy. The underlying cause of death was significantly different in infants who had an autopsy compared with infants who did not (P = 0.02). The autopsy rate was higher for deaths from miscel laneous causes (52.9%), lethal malformation (46.8%) and infection (45.4%) t han deaths from prematurity (25.9%) and asphyxia (19%). Clinical errors in the causes of death were found in 22% of the infants, and in 4.4% a change in management may have been curative or prolonged life. The autopsy had aud it value in 26% of infants and genetic value for a single gene (Mendelian) disorder in 4.4%. Conclusions: Although the autopsy following death in a Level IV NICU yields potentially useful information in more than one-third of cases, this does not seem sufficient to ensure a high neonatal autopsy rate.