ABSTRACT
I developed a single equation that simulated previously published growth charts for very low-birth-weight infants, using birth weight as the only variable. The intent of this study was to validate the equation using infants admitted to our neonatal intensive care unit (NICU) and to determine if this equation could be used as a research tool to compare weight gain among populations. All 171 surviving infants with birth weight ≤1500 g who received care in our NICU were studied retrospectively. The individual daily weights were compared with the predicted daily weights for each infant using the relative error of the prediction. The relative error of the prediction on all days for all infants was 0.45 ± 0.1 3% (mean ± SEM), and the maximum mean daily relative error of 2.73 ± 0.90% occurred on Day 9. Black infants had a more positive relative error than white infants (p = 0.0001), due to faster weight gain. Twins gained faster than singletons (p = 0.023). Infants with patent ductus arteriosus (PDA) gained more slowly than normal infants (p = 0.001) after an initial rapid weight gain. Infants with lung disease had initial rapid weight gain similar to infants with PDA, but subsequently were similar to infants without lung disease. This equation accurately predicts weight gain in our population of very low-birth-weight infants. The relative error of the prediction is a useful research tool for determining the effect of clinical conditions or interventions on weight gain.
Keywords
Weight gain - growth - premature infants - growth equation
