OBJECTIVE:To compare body composition of large-for-gestational-age (LGA) with appropriate-for-gestational-age (AGA) newborns and to identify antenatal predictors of LGA. STUDY DESIGN: This cross-sectional study included 536 term, singleton infants. Anthropometric measurements were performed within 48 h of birth and included determination of body fat percentage (%BF) by air displacement plethysmography. Associations were investigated using logistic regression. RESULT:LGA infants had greater %BF (P o 0.001) compared with AGA infants. Significant predictors of LGA infants included parity (odds ratio (OR) = 1.98, (95% confidence interval (CI) 1.00, 4.02)), paternal height (OR = 1.08, (95% CI 1.03, 1.14)), maternal pregravid weight (65 to 74.9 kg: OR = 2.77, (95% CI 1.14, 7.06)) and gestational weight gain (OR = 1.09, 95% CI (1.03, 1.16)). Gestational diabetes mellitus was not associated with LGA infants (P = 0.598). CONCLUSION: Paternal height, parity, maternal pregravid weight and gestational weight gain were strongly associated with LGA infants. These results may allow early prediction and potential modification, thereby optimising clinical outcomes.