Objective: To examine the relationship of multiple maternal and cord blood correlates of newborn size to determine the relative strength of the insulin-like growth factor-I association.
Methods: Thirty-seven venous cord blood specimens were obtained at the time of delivery. Ponderal index and birth weight percentile were calculated at birth. Neonatal length estimates were performed with a measuring board. All mothers were nonsmokers and had normal glucose tolerance. There was a wide range of maternal prepregnancy body mass indexes (BMI) (19.6-43.4). Neonates had a wide range of ponderal indexes (2.12-2.75) and birth weight percentiles (7-99th percentile). Univariate correlation coefficients were calculated to determine simple relationships. Stepwise linear regression analyses were performed to determine the relative contribution of potential explanatory variables to both ponderal index and birth weight percentile. Potentially explanatory independent variables included maternal prepregnancy BMI, weight gain in pregnancy, and maternal insulin sensitivity at 32 weeks' gestation. Maternal insulin sensitivity was estimated using the minimal model technique. Neonatal variables included sex, cord blood albumin, insulin, insulin-like growth factor-I, insulin-like growth factor-binding protein-1, and insulin-like growth factor-binding protein-3.
Results: Significant positive univariate correlations were identified between cord blood insulin-like growth factor-I and insulin-like growth factor-binding protein-3 with neonatal ponderal index and birth weight percentile. Maternal insulin sensitivity demonstrated a negative correlation with birth weight percentile (r = -.35, P < .05). Cord blood insulin correlated positively with birth weight percentile (r = .32, P < .05). There were no significant associations of cord blood insulin-like growth factor-binding protein-1 or albumin with either index of newborn size. Stepwise logistic regression analysis demonstrated an independent association of insulin-like growth factor-I with ponderal index (r2 = .41, P < .001). Both insulin-like growth factor-I and male sex were associated independently with birth weight percentile (r2 = .38, P < .001). No additional independent variables contributed to the prediction of ponderal index or birth weight percentile.
Conclusion: These data support a unique relationship between cord blood insulin-like growth factor-I and newborn size under normal growth conditions. This is manifest by the strength and independence of the association between insulin-like growth factor-I and neonatal birth weight percentile ponderal index.