Abstract
Purpose: The purpose of this study was to determine the intra- and interrater reliability of quadriceps and hamstring soft tissue measures using DXA scans. Methods: A total of 44 subjects (23 males) participated in this study. The first total body DXA scan was performed in the standard anterior/posterior scanning position, followed by two additional total body scans while the subjects were lying on their left and right sides with the leg of interest extended and the contralateral leg bent. Unique regions of interest were created for analyses of mineral-free lean masses (MFL) using custom analysis software with manual tracing (by two investigators) of the quadriceps (QUADS) and hamstrings (HAMS) of the right and left thighs. Between-within repeated measure analysis of variance (ANOVA) was used to determine if there were significant differences among the MFL measures, while intraclass correlation coefficients (ICC) and coefficients of variation (CV) were used to assess the intra- and interrater reliability. Results: Between-group analyses revealed that Investigator 2 had small yet significantly higher mean differences for right QUADS (2346.6 ± 602.4 g vs. 2327.4 ± 587.9 g), left QUADS (2337.3 ± 581.9 g vs. 2312.8 ± 581.2 g), right HAMS (2655.9 ± 626.3 g vs. 2543.0 ± 593.5 g), and left HAMS (2686.1 ± 628.1 g vs. 2562.8 ± 596.5 g) when compared to Investigator 1. Intraclass correlation coefficients between (≥0.984) and within (≥0.992) raters were high for all MFL measures, with low variation across all MFL measures (≤1.62%). Conclusions: Despite having significant group mean differences, our results revealed strong and significant reliability, and we recommend that a single investigator analyze the scans twice and that the mean of the two measures be used for final reporting within a given study.