BACKGROUND. There is a paucity of information on the maintenance of body temperature at birth for low birth weight infants.
OBJECTIVES. We examined the distribution of temperatures in low birth weight infants on admission to the NICUs in the Neonatal Research Network centers and determined whether admission temperature was associated with antepartum and birth variables and selected morbidities and mortality.
METHODS. Infants without major congenital anomalies born during 2002 and 2003 with birth weights of 401 to 1499 g who were admitted directly from the delivery room to the NICU were included. Bivariate associations between antepartum/birth variables and admission temperature and selected morbidities/mortality and admission temperature were examined, followed by multivariable linear or logistic regressions to detect independent associations.
RESULTS. There were 5277 study infants and the mean (±SD) birth weight and gestational age were 1036 ± 286 g and 28 ± 3 weeks, respectively. The distribution of admission temperatures was 14.3% at <35°C, 32.6% between 35 and 35.9°C, 42.3% between 36 and 36.9°C, and 10.8% at ≥37°C. The estimate of birth weight on admission temperature with and without intubation was +0.13°C and +0.04°C per 100-g increase in birth weight, respectively. The mean admission temperature for each center varied from 1.5°C below to 0.3°C above a reference center. On adjusted analyses, admission temperature was inversely related to mortality (28% increase per 1°C decrease) and late-onset sepsis (11% increase per 1°C decrease) but not to intraventricular hemorrhage, necrotizing enterocolitis, or duration of conventional ventilation.
CONCLUSIONS. Preventing decreases in temperature at birth among low birth weight infants remains a challenge. Associations with intubation and center of birth suggest that assessment of temperature control for infants intubated in the delivery room may be beneficial. Whether the admission temperature is part of the casual path or a marker of mortality needs additional study.