Trabecular Bone Score (TBS) Varies with Correction for Tissue Thickness Versus Body Mass Index: Implications When Using Pediatric Reference Norms

Abstract

ABSTRACT Trabecular bone score (TBS) derived from secondary analysis of lumbar spine dual-energy X-ray absorptiometry (DXA) scans improves fracture prediction independent of bone mineral density (BMD) in adults. The utility of TBS to assess fracture risk in younger patients has not been established because pediatric norms have been lacking. Robust TBS reference data from the Bone Mineral Density in Childhood Study (BMDCS) have been published. TBS values for the BMDCS study were derived using an algorithm that accounts for tissue thickness (TBSTH) rather than the commercially available algorithm that adjusts for body mass index (BMI; TBSBMI). We examined the magnitude of differences in TBSTH and TBSBMI in a cohort of 189 healthy youth. TBS values using both algorithms increased with age and pubertal development in a similar pattern. However, TBSBMI values were systematically and significantly higher than TBSTH (mean = 0.06, p < 0.0001). The difference between calculated TBSBMI and TBSTH was not uniform. Differences were greater at lower TBS values, in males, in older individuals, in those at later Tanner stages, and in those with a greater BMI Z-score. These systematic differences preclude the development of a simple formula to allow conversion of TBSBMI to TBSTH “equivalents.” Because of these systematic differences in these two algorithms, using an individual's TBSBMI to calculate a Z-score using the BMDCS TBSTH reference values results in a falsely higher TBS Z-score (differences mean = 0.73, interquartile range [IQR] = 0.3 to 1.6). Until TBSTH software for Hologic DXA equipment becomes commercially available, BMDCS TBS reference norms should not be used. © 2023 American Society for Bone and Mineral Research (ASBMR).