Estimation of neonatal body fat percentage predicts neonatal hypothermia better than birthweight centile

PEA POD™ air displacement plethysmography quickly and noninvasively estimates neonatal body fat percentage (BF%). Low PEA POD™ BF% predicts morbidity better than classification as small-for-gestational-age (SGA; <10th centile), but PEA PODs are not widely available. We examined whether skinfold measurements could effectively identify neonates at risk; comparing skinfold BF%, PEA POD™ BF% and birthweight centiles' prediction of hypothermia - a marker of reduced in utero nutrition.

Neonates had customized birthweight centiles calculated, and BF% prospectively estimated by: (i) triceps and subscapular skinfolds using sex-specific equations; and (ii) PEA POD™. Medical record review identified hypothermic (<36.5 °C) episodes.

42/149 (28%) neonates had hypothermia. Skinfold BF%, with an area under the curve (AUC) of 0.66, predicted hypothermia as well as PEA POD™ BF% (AUC = 0.62) and birthweight centile (AUC = 0.61). Birthweight <10^th centile demonstrated 11.9% sensitivity, 38.5% positive predictive value (PPV) and 92.5% specificity for hypothermia. At equal specificity, skinfold and PEA POD™ BF% more than doubled sensitivity (26.2%) and PPV increased to 57.9%.

Neonatal BF% performs better to predict neonatal hypothermia than birthweight centile, and may be a better measure of true fetal growth restriction. Estimation of neonatal BF% by skinfold measurements is an inexpensive alternative to PEA POD™.