Deaths in a neonatal intensive care unit: a 10-year perspective.

Objective: To examine changes in the characteristics and management of infants dying in a regional neonatal intensive care unit in 1987-1988 vs. 1997-1998.

Setting: The level III Neonatal Intensive Care Unit (NICU) at Rikshospitalet, Oslo, Norway, handles both regional and national referrals.

Design/methods: The study was retrospective and observational. Patients who died in the neonatal intensive care unit were identified using our own and the hospital's data records. Charts were reviewed by the principal author.

Results: The mortality rate relative to admissions decreased significantly from 1987-1988 to 1997-1998 (6.9% vs. 3.4%, p <.0001). Infants who died in 1997-1998 were more mature and had higher birth weights than those who died in 1987-1988 (34.0 +/- 5.5 vs. 32 +/- 6.0 wks gestational age [mean +/- sd], p <.05; and 2,186 +/- 1,207 vs. 1,699 +/- 1,038 g, p <.05). There was a significantly higher proportion of infants with complex congenital malformations among those who died in 1997-1998 (54% vs. 28%, p <.005). Forgoing intensive care treatment was more commonly associated with the process of dying in 1997-1998 than 10 yrs earlier (63.5% vs. 22.8%, p <.0001). Parental involvement in the process leading to a decision to forgo life support was more frequently described in the charts from 1997-1998 (72.7% vs. 23.8%, p <.001). During the last time period, parents were also present at the time of death significantly more often.

Conclusions: The mortality rate of sick infants decreased significantly between 1987-1988 and 1997-1998, showing the improvements in neonatal intensive care during that decade. In 1997-1998, congenital malformations had become the leading cause of death. Parental involvement in life-and-death questions seems to have become the rule, and almost two thirds of neonatal intensive care unit deaths followed a decision to forgo life support.