Hypoglycemia and the full-term newborn: how well does birth weight for gestational age predict risk?

To determine whether anthropometric characteristics could be used to accurately predict risk of hypoglycemia in full-term newborns during the early post-birth period.

Descriptive, utilizing newborn anthropometric measurements singly and in combination to determine risk of neonatal hypoglycemia. The following measurements were obtained twice for each newborn: weight, head circumference, chest circumference, abdominal circumference, mid-arm circumference, thigh circumference, and length. The investigator was blind to all measurements except weight.

Mothers' rooms or the newborn nursery in a community hospital.

All measurements were obtained twice, and a physical examination was completed on each newborn by the principal investigator. These newborns were classified as large-, average-, and small-for-gestational age, using a tool typically used in many newborn nurseries.

One hundred fifty-seven full-term newborns (94 White and 63 African American).

The differences in anthropometric measurements by race and gender were calculated using two-way analysis of variance. The risk of hypoglycemia was calculated using logistic regression modeling.

There were significant differences in measurements by race and by gender. Additionally, there was a subset of newborns classified as average for gestational age who had an increased risk of hypoglycemia (OR = 4.17, 95% CI = 1.33-13.08). Newborns with a mid-arm circumference/head circumference ratio that varied from .26 to .29 have an odds ratio of 6.10 (95% CI = 1.89-19.66) for risk of hypoglycemia. Plotting a newborn's birth weight on a published fetal growth curve clearly did not accurately predict his or her risk of hypoglycemia.

These findings indicate that extremes in newborn birth weight are not always correctly defined, may vary by race and ethnic group, and may not be the best method for determining under- or overnourishment and risk of neonatal hypoglycemia.