Open Access Highly Accessed Research article

Increasing illness severity in very low birth weight infants over a 9-year period

David A Paul12*, Kathleen H Leef1, Robert G Locke12, Louis Bartoshesky2, Judy Walrath3 and John L Stefano12

Author Affiliations

1 Department of Pediatrics, Section of Neonatology, Christiana Care Health Services, Newark, DE, USA.

2 Department of Pediatrics, duPont Hospital for Children, Wilmington, DE, USA. Department of Pediatrics, Thomas Jefferson University, Philadelphia, PA, USA

3 Eugene duPont, Preventive Medicine and Rehabilitation Institute, Christiana Care Health Services, Newark, DE, USA

For all author emails, please log on.

BMC Pediatrics 2006, 6:2  doi:10.1186/1471-2431-6-2

Published: 6 February 2006

Abstract

Background

Recent reports have documented a leveling-off of survival rates in preterm infants through the 1990's. The objective of this study was to determine temporal changes in illness severity in very low birth weight (VLBW) infants in relationship to the outcomes of death and/or severe IVH.

Methods

Cohort study of 1414 VLBW infants cared for in a single level III neonatal intensive care unit in Delaware from 1993–2002. Infants were divided into consecutive 3-year cohorts. Illness severity was measured by two objective methods: the Score for Neonatal Acute Physiology (SNAP), based on data from the 1st day of life, and total thyroxine (T4), measured on the 5th day of life. Death before hospital discharge and severe intraventricular hemorrhage (IVH) were investigated in the study sample in relation to illness severity. The fetal death rate was also investigated. Statistical analyses included both univariate and multivariate analysis.

Results

Illness severity, as measured by SNAP and T4, increased steadily over the 9-year study period with an associated increase in severe IVH and the combined outcome of death and/or severe IVH. During the final 3 years of the study, the observed increase in illness severity accounted for 86% (95% CI 57–116%) of the variability in the increase in death and/or severe IVH. The fetal death rate dropped from 7.8/1000 (1993–1996) to 5.3/1000 (1999–2002, p = .01) over the course of the study.

Conclusion

These data demonstrate a progressive increase in illness in VLBW infants over time, associated with an increase in death and/or severe IVH. We speculate that the observed decrease in fetal death, and the increase in neonatal illness, mortality and/or severe IVH over time represent a shift of severely compromised patients that now survive the fetal time period and are presented for care in the neonatal unit.