OBJECTIVE: To determine the variability of measuring regional adipose tissue area using abdominal computed tomography (CT) in normal-weight, prepubertal girls. DESIGN: Measurements of abdominal CT were performed twice, 6 weeks apart. SUBJECTS: Sixty-one normal-weight, prepubertal girls (age: 4.8-10.3 y, mean (s.d.) BMI: 16.7 +/- 1.5 kg/m2). MEASUREMENTS: Abdominal adipose tissue by CT at baseline and 6 weeks later. RESULTS: There were no significant differences between visits 1 and 2 with regard to the children's average abdominal CT derived intraabdominal adipose tissue (21.64 cm2 vs 23.74 cm2) and subcutaneous adipose tissue (62.49 cm2 vs 65.28 cm2). The Pearson coefficient of correlation (r), P-value, total coefficient of variation (CV) and standard error of the difference (SEdelta) for intra-individual measurements between visits 1 and 2 by abdominal CT were: total abdominal adipose tissue, r = 0.94, P < 0.0001, CV = 12.34%, SEdelta 2.25 cm2; subcutaneous abdominal adipose tissue, r = 0.96, P < 0.0001, CV = 10.67%, SEdelta = 1.57 cm2; and intraabdominal adipose tissue, r = 0.67, P < 0.0001, CV = 21.5%, SEdelta = 1.11 cm2. The mean ratios of intraabdominal to subcutaneous adipose tissue on visits 1 and 2 were 0.42 +/- 0.2 and 0.44 +/- 0.24, respectively. CONCLUSION: Regional adipose tissue area measurements using abdominal CT were reliable in healthy, normal-weight, prepubertal girls.