Using unconditional and conditional standard deviation scores of fetal abdominal area measurements in the prediction of intrauterine growth restriction

Abstract

Objective: To compare the performance of unconditional and conditional standard deviation scores (Z scores) of fetal abdominal area (FAA) measurements in the antenatal identification of infants born with anthropometric features of intrauterine growth restriction. Methods: A prospective observational study, involving 274 low-risk women participating in a longitudinal study of serial ultrasound in pregnancy. Conditional Z scores were established for the last FAA prior to delivery with reference to measurements made both 28 and 56 days previously. Unconditional Z scores (size) were calculated from the last FAA measurement prior to delivery. Receiver-operator characteristics curves were employed to determine an optimal cut-off point for Z scores to predict intrauterine malnourishment. The main outcome measures were: likelihood ratios (LR) for conditional and unconditional Z scores of FAA in the prediction of infants with skinfold thickness < 10th percentile; ponderal index < 25th percentile or mid-arm circumference to occipitofrontal circumference ratio (MAC/OFC) of < -1 SD. An LR of > 10 generates significant changes in the pretest probability of growth restriction, whereas an LR of 5-10 generates only moderate changes. Results: Conditional Z scores with 28- and 56-day separations predicted growth restriction with LR 7.5 (95% confidence interval [CI], 3.7-14.7) and 4.8 (95% CI, 2.8-7.8) for ponderal index but did not usefully predict skinfold thickness or MAC/OFC. Unconditional Z scores did not usefully predict any of the parameters of growth restriction. Conclusions: Quantifying third trimester fetal growth by means of FAA conditional Z scores is moderately useful in predicting infants with a low ponderal index and is superior to unconditional FAA Z scores in late pregnancy.