The effect of customized growth charts on the identification of large for gestational age newborns

Abstract

Objective: To determine the effect of using customized vs. standard population birthweight curves to define large for gestational age (LGA) infants. Methods: We analyzed data obtained from 2,097 singleton pregnancies using three different methods of classifying newborn birthweight: standard population curves, British or Spanish customized curves. We recorded maternal characteristics, proportion of LGA newborns when using each method, percentage of LGA according to one method but not for the others, and concordance between the different methods. Results: The proportion of LGA newborns according to Spanish customized curves was significantly lower than that calculated using either standard general population birthweight curves or British curves (p < 0.001). A third (33.9%) of the infants classified as LGA according to the general population method were adequate for gestational age (AGA) when the Spanish customized curves were used, and 18.5% of non-LGA were LGA according to customized curves (p < 0.001). Concordance between the different models high, but on excluding AGA the concordance coefficient was low (Cohen’s κ <0.4). Conclusions: The use of customized curves allows differentiation between constitutional LGA and cases of fetal overgrowth, leading to a decrease in the rate of both false-positives and negatives as well as the overall proportion of LGA babies.

Keywords::

• Large for gestational age
• macrosomia
• small for gestational age
• customized curves
• population standards
• birth weight percentiles

Declaration of Interest: The authors report no conflict of interest.