Objective: To examine the association between weight loss during the first 10 days of life and the incidence of death or bronchopulmonary dysplasia (BPD) in small for gestational age (SGA) and appropriate for gestational age (AGA) extremely low-birth-weight infants.
Design/methods: This is a retrospective analysis of a cohort of ELBW (birth weight <1000 g) infants from the NICHD Neonatal Research Network's database. The cohort consisted of 9461 ELBW infants with gestational age of 24-29 weeks, admitted to Network's participating centers during calendar years 1994-2002 and surviving at least 72 h after birth. The cohort was divided into two groups, 1248 SGA (with birth weight below 10th percentile for gestational age) and 8213 AGA (with birth weight between 10th and 90th percentile) infants. We identified infants with or without weight loss during the first 10 days of life, which we termed as 'early postnatal weight loss' (EPWL). Univariate analyses were used to predict whether EPWL was related to the primary outcome, death or BPD, within each birth weight/gestation category (SGA or AGA). BPD and death were also analyzed separately in relation to EPWL. Logistic regression analysis was done to evaluate the risk of death or BPD in SGA and AGA groups, controlling for maternal and neonatal demographic and clinical factors found to be significant by univariate analysis.
Results: SGA ELBW infants had a lower prevalence of EPWL as compared with AGA ELBW infants (81.2 vs 93.7%, respectively, P<0.001). In AGA infants, univariate analysis showed that death or BPD rate was lower in the group of infants with EPWL compared with infants without EPWL (53.4 vs 74.3%, respectively, P<0.001). The BPD (47.2 vs 64%, P<0.001) and death (13.8 vs 32.9%, P<0.001) rate were similarly lower in the EPWL group. The risk-adjusted odds ratios (ORs) showed that EPWL was associated with lower rate of death or BPD (OR 0.47, 95% CI: 0.37-0.60). In SGA infants, on univariate analysis, a similar association between EPWL and outcomes was seen as shown in AGA infants: death or BPD (55.9 vs 75.2%, P<0.001), BPD rate (48.3 vs 62.1%, P=0.002) and rate death (19 vs 40.8%, P<0.001) for those with or without EPWL, respectively. Multiple logistic regression showed that as in AGA ELBW infants, EPWL was associated with lower risk for death or BPD (OR 0.60, 95% CI: 0.41-0.89) among SGA infants.
Conclusions: SGA infants experienced less EPWL when compared with their AGA counterparts. EPWL was associated with a lower risk of death or BPD in both ELBW AGA and SGA infants. These data suggest that clinicians who consider the association between EPWL and risk of death or BPD should do so independent of gestation/birth weight status.