OBJECTIVES: To evaluate the association between birth weight and adult body composition in twins, controlling for maternal and genetic influences.

DESIGN: Twins were randomly selected from the East Flanders Prospective Twin Survey, a population-based historic twin cohort.

SUBJECTS: Male members of 229 twin pairs between 18 and 34 y of age who participated in the Prenatal Programming Twin Study.

MEASUREMENTS: Adult body mass, height, BMI, lean body mass, sum of skinfolds and waist-to-hip-ratio. Lean body mass, sum of skinfolds and waist-to-hip-ratio were also adjusted for body mass. Intra-pair difference in adult anthropometric measures between the heaviest and the lightest twin at birth.

RESULTS: Per kg increase in birth weight, body mass (4.2 kg), height (3.3 cm), lean body mass (3.1 kg) and to a lesser extent BMI (0.49 kg/m(2)) increased, whereas waist-to-hip-ratio (-1.4%) and sum of skinfolds (-0.11 s.d.) decreased, when adjusted for body mass. In a pair-wise analysis, the heavier twin at birth was taller and heavier as an adult, but, when adjusted for body mass, he had a lower waist-to-hip-ratio, less subcutaneous fat, and more lean body mass, compared to his lighter sib. Intra-pair difference in body composition was associated with intra-pair birth weight difference in monozygotic and dizygotic twins.

CONCLUSIONS: An adverse intra-uterine environment, as measured by birth weight, is associated with more subcutaneous and abdominal fat and less lean body mass in adulthood. This association is independent of maternal and genetic influences. However, we cannot exclude the existence of genes that act on both birth weight and adult body composition.