Comparison of bioimpedance analysis and dual-energy X-ray absorptiometry body composition measurements in peritoneal dialysis patients according to edema

Abstract

Changes in the difference between bioimpedance analysis (BIA) and dual-energy X-ray absorptiometry (DXA) body composition measurements according to edema is an important issue for peritoneal dialysis (PD) patients. We reviewed the medical records of patients to identify all adults who had undergone PD and composition measurements by both BIA and DXA techniques. The aim of this study was to evaluate any change in the difference between BIA and DXA body composition measurements according to edema. We identified 409 patients whose body compositions were measured by both techniques one or more times, for a total of 1,109 pairs of measurements. The measurements were divided into 4 quartiles on the basis of the edema index. Significant correlations and intraclass correlations were noted between the two methods for lean mass (LM), fat mass (FM), and bone mineral content. Simple linear regression analyses using DXA measurements for the prediction of body compositions by BIA showed that non-standardized-βs of total LM decreased as the grade of edema index increased (from 1.008 to 0.949), whereas non-standardized-βs of total FM increased as the grade of edema index increased (from 1.034 to 1.162). Bias for total LM changed from positive to negative, and this negative bias amplified as the grade of edema index escalated (from 0.406 kg to -2.276 kg). A positive bias was observed for total FM in the 1st quartile, and this positive bias increased with an increase in the grade of edema index (from 0.594 kg to 2.863 kg). LM measured by DXA is overestimated in PD patients with edema compared to the measurements by BIA. However, FM and bone mineral content measured by BIA are is overestimated in PD patients, compared to the measurements by DXA, especially in patients with worsening edema. The difference between the two techniques grows more prominent as the grade of edema increases. A combination of two methods will allow clinicians to conduct more accurate body composition assessments for PD patients with edema.