Relationship between customised birthweight centiles and neonatal anthropometric features of growth restriction

Objective To determine the relationship between customised birthweight centiles (adjusted for maternal and fetal physiological variables) and neonatal anthropometric features of intrauterine growth restriction (IUGR). Design Observational study. Population Two-hundred and seventy women with low risk pregnancies participating in a cohort study of serial ultrasound biometry. Methods Customised birthweight centiles were calculated following adjustment for maternal weight, height and ethnic origin, gestational age at delivery, birth order, and sex of the infant. Three separate neonatal anthropometric measures were used to define IUGR: subscapular or triceps skinfold thickness <10th centile; ponderal index <25th centile; and mid-arm circumference to occipito-frontal circumference ratio (MAC/OFC) <-1 standard deviation (SD). Relationship of the centiles to these outcomes was evaluated using likelihood ratios (LR) and kappa statistic. These approaches allowed us to examine the strength of the association: an LR of 5-10 would be expected to generate moderate changes in the pre-test probability of IUGR, whereas a kappa value of 0.2-0.4 would reflect fair agreement between customised birthweight centiles and neonatal anthropometric measures. Results Customised birthweight centile of 10 or less had the following LR values for the various anthropometric criteria for IUGR: 5.1 (95% CI 3-8.5) for low skinfold thickness; 4.3 (95% CI 2.5-7.1) for low ponderal index; and 3.9 (95% CI 2-6.6) for low MAC/OFC ratio. The kappa values were: 0.4 (95% Cl 0.26-0.51) for low skinfold thickness; 0.33 (95% Cl 0.21-0.46) for low ponderal index; and 0.13 (95% CI 0-0.26) for low MAC/OFC ratio. Conclusion In a low risk population, customised birthweight centiles can only be moderately useful in the identification of neonates with low skinfold thickness and low ponderal index.