Comparison of computed tomography and bioelectrical impedance analysis for measuring body composition in patients with esophageal cancer: A retrospective cohort study

Abstract

Background Low muscle mass is associated with a higher incidence of pulmonary complications and worse survival following esophagectomy for esophageal cancer (EC). The aim of this study was to compare the accuracy of bioelectrical impedance analysis (BIA) with computed tomography (CT) scans as the gold standard reference for assessing low muscle mass. Methods BIA as part of usual care and diagnostic CT were used to retrospectively assess the muscle mass of 75 EC patients prior to neoadjuvant chemoradiotherapy. BIA and CT were performed at different time points prior to treatment. Two validated regression equations were used for both BIA and two for CT to calculate muscle mass. Pearson correlation coefficients and Bland-Altman plots were utilized to assess the correlation and agreement between BIA and CT. Results The median time between BIA and corresponding CT measurement was 17 d. Both equations showed moderate-to-good correlation with CT (R2 = 0.63 and 0.64, respectively). BIA overestimated fat-free mass values compared with CT, and no strict agreement could be observed with respect to skeletal muscle mass values either. However, BIA identified patients in the lower muscle mass quartile with a sensitivity of 80% and a specificity of 93%. The receiver operating characteristic (ROC) curve showed an area under the curve of 0.94 for both equations. BIA was shown to have a moderate-to-good correlation with CT. Conclusion Although BIA showed higher muscle mass than CT, BIA measurements demonstrated the ability to differentiate patients with low muscle mass in the lower muscle mass quartile.