Endocrine Abstracts

JOINT3284

Background: Pediatric obesity is a multifactorial condition characterized by excessive fat accumulation. However, traditional body mass index (BMI)-based classification cannot assess fat mass (FM) or its distribution. Dual-energy X-ray absorptiometry (DXA) is the gold standard for body composition assessment. However, due to its limited accessibility, portable and non-invasive alternatives, such as bioelectrical impedance analysis (BIA), are often used. This exploratory study aimed to describe body composition phenotypes in different types of non-syndromic pediatric obesity and evaluate the agreement between BIA and DXA in assessing FM.

Methods: From September 2024 to January 2025, we performed BIA on patients aged 6–17 years with primary obesity (PO), obesity secondary to tumors now in off-therapy (SO), or genetic obesity (GO), all with BMI>2SDS and scheduled for DXA. DXA provided percentage of total FM and truncal FM (FMtr), fat mass index (FMI, kg/m²) and fat-to-lean ratio (FLR), while BIA provided FMI and FLR.

Results: We assessed 64 patients (61% female), of whom 75% had PO, 12.5% had SO, and 12.5% had GO. The median age was 11 years. By comparing PO, SO, and GO groups, significant differences were observed in median height-SDS (0.9 vs. -1.2 vs. -0.3, P = 0.009) and BMI-SDS (2.9 vs. 2.3 vs. 3.1, P = 0.024). DXA measures revealed significant differences in total FM% (46 vs 48 vs 54, P = 0.029) and FLR (0.87 vs 0.94 vs 1.2, P = 0.028), whereas truncal FM% (47 vs. 48 vs. 54, P = 0.092) and FMI (11.9 vs. 11.8 vs. 14.2, P = 0.246) did not differ significantly. BIA measurements showed a significant difference in FMI (10.6 vs. 7.4 vs. 13.9, P = 0.041), while FLR (0.59 vs. 0.53 vs. 0.78, P = 0.081) was not significantly different. Although a moderate correlation BIA-DXA for FMI was found (r = 0.682, P < 0.001), Bland-Altman analysis indicated that BIA underestimated FMI compared to DXA, with a mean difference of 1.40kg/m² and limits of agreement ranging from -3.34 to 6.16.

Conclusions: Our preliminary findings highlight that pediatric patients with GO exhibit higher FM compared to those with PO and SO, with a distinct fat distribution pattern, characterized by higher FMtr, that may have metabolic implications and thus require further investigation. BIA underestimates FMI compared to DXA, with substantial variability, limiting its use as a method interchangeable with DXA for body composition assessment.