OBJECTIVE: The increasing prevalence of obesity in young children emphasises the need for accurate measures of total and regional fat at this age. Thus the aim of this study was to evaluate the ability of waist circumference, waist-to-height ratio (WHtR) and the conicity index (CI) to discriminate between children with low and high levels of trunk fat mass. METHODS: Trunk fat mass was measured by dual-energy x-ray absorptiometry (DXA) in 301 predominantly Caucasian children (150 girls) aged 3-5 years. High trunk fat was defined as an internal z-score of > or =+1. Receiver-operating characteristic (ROC) curves and the areas under each curve (AUC) were constructed to compare the relative ability of waist circumference, WHtR and CI to identify children with low and high trunk fat mass. RESULTS: Girls had more truncal fat than boys (P<0.001). AUCs indicated that waist circumference correctly discriminated between children with low and high trunk fat mass 87% (for girls) to 90% (for boys) of the time. Waist circumference performed better than WHtR (AUCs: 0.79 in girls and 0.81 in boys; P=0.164 and P=0.011, respectively) and the CI (AUCs: 0.53 in girls and 0.65 in boys, P<0.0001). A z-score of 0.55 correctly identified 79% of girls and 81% of boys with high trunk fat mass, and 82% of girls and 84% of boys with low trunk fat mass. Suggested waist circumference cut-offs for each half-year of age in both sexes are reported. CONCLUSION: Waist circumference performs reasonably well as an indicator of high trunk fat mass in preschool-aged children.