Most small premature infants are not small weight-for-length

Purpose To compare the categorization of growth status in premature infants using weight-for-gestational age (WT/age) versus weight-for-length3 (WT/LN3) methods. Methods Secondary analysis of infants 25–29 weeks at birth (1991–2003) from the National Institute of Child Health and Human Development Neonatal Research Network Generic Database for Cincinnati infants. Categorized infant growth status by weight-for-age and weight-for-length3 (Lubchenco et al, 1966). Results Sample included 1222 infants (51% males; 68% white) with growth data at birth and at discharge or 120 days of life (whichever came first). The average ± SD at birth was 27.7 ± 1.1 weeks gestational age (GA), 1054 ± 218 gm in WT and 36.3 ± 2.7 cm in LN. The average ± SD at discharge/120 d was 35.5±3.1wk GA, 2213 ± 590 gm in WT and 43.3 ± 3.5 cm in LN. There was poor agreement between WT/age and WT/LN3 methods in assigning small, appropriate or large categories (kappa = 0.02, 0.10 for birth and discharge/120 d, respectively). Of the 151 (12%) small weight-for-age (SGA) infants at birth, 22% were small weight-for-length (SWL), 77% were appropriate weight-for-length (AWL) and 1% were large weight-for-length (LWL). By discharge/120 d, there were more SGA infants (N = 262, 21%) yet fewer (8%) were SWL, 80% were AWL and 11% were LWL. The number of appropriate weight-for-age (AGA) infants declined from birth to discharge/120 d (1068 to 949), but the proportion of AGA infants who were LWL tripled (6% to 19%). Conclusion Using both WT/age and WT/LN3 better defines growth status in premature infants, and may change nutrition care decisions (e.g., high calorie/protein diets may be used less often for SGA infants who are appropriate or large WT/LN3). Overfeeding may be in part responsible for the increase in LWL infants by discharge/120 d. The long-term implications of a premature infant becoming LWL are unknown, but early identification seems prudent.