Comparison of multifrequency bioelectrical impedance analysis and dual-energy X-ray absorptiometry assessments in outpatient hemodialysis patients

Abstract

Background: Malnutrition is common in hemodialysis patients and closely related to increased morbidity and mortality. As such, simple, reliable, and easily available methods of determining nutritional status and recognition of short-term changes in body composition are desirable for routine clinical practice.

Study design: Diagnostic test study.

Setting & participants: 53 stable adult hemodialysis patients attending for thrice-weekly outpatient hemodialysis in a university tertiary hospital dialysis center.

Index test: Comparison of dual-energy x-ray absorptiometry (DEXA) and multifrequency bioelectrical impedance analysis (BIA) using a tetrapolar 8-point tactile electrode system as 2 index tests of body composition.

Reference test: None.

Results: Assessment of whole-body composition showed that lean body mass measured using the 2 techniques correlated highly, with good method agreement shown using a Bland-Altman plot (r = 0.92; P < 0.001; bias, +1 g [95% CI, -1,173 to 1,175]), as did fat mass (r = 0.93; P < 0.001; bias, -157 g [95% CI, -1,251 to 937]). Similarly, segmental analysis of lean body mass showed strong correlations between lean body mass of the trunk and right and left legs with small bias (r = 0.85, 0.89, and 0.86, respectively; P < 0.001; Bland-Altman bias, -859, +364, and +552 g, respectively), but weaker correlations for lean body mass for the right and left arm (r = 0.69 and 0.75, respectively; P < 0.001; Bland-Altman bias, -240 and +12 g, respectively). Bone mineral content derived using multifrequency BIA overestimated that measured using DEXA (r = 0.77; P < 0.001; bias, +530 g [95% CI, 422-638]).

Limitations: Retrospective study in a healthy ambulant outpatient cohort.

Conclusions: Compared with DEXA, multifrequency BIA appears to be a robust tool for measuring and monitoring total-body fat and lean body mass in hemodialysis patients; however, there is less agreement in bone mineral content assessment between the 2 methods.