Application of Bioimpedance Spectroscopy in Taiwan Dialysis Patients

298 MHD patients in 6 hemodialysis centers in Taiwan were enrolled in this randomized controlled study. All the participants were randomized into study group ( dry weight (DW) determined by BCM-BIS) and control group ( dry weight determined by clinical symptoms) with stratification by diabetes mellitus and centers. Body composition monitor based on bioimpedance spectroscopy (BCM-BIS) was performed monthly and concomitant biochemical data and clinical outcomes were collected. Intra-dialysis complications were recorded in every dialysis session. Primary outcome was all-cause hospitalization. Secondary outcomes included 1) complications during DW adjustment 2) intra-dialysis morbidities 3) hypertension 4) mortality 5) acute fluid overload (AFO) or cardiovascular (CV)-related events. The clinical feasibility of the BCM-BIS algorithm was also inspected. Statistical methods: a software (SAS) 9.3 was used for statistical analysis in this study. R 3.0 was used for graphic presentation of parts of the results. Relative risk, incidence rate ratio and Fisher's exact test were used to compare the discrete outcomes in both groups. Survival analysis was used for time-related events. Repeated-measured and longitudinal data were analyzed with the generalized linear mixed model (GLMM) through the procedure for continuous variables (PROC MIXED) and for discrete variables (PROC GLIMMIX) in SAS.

Inclusion criteria

MHD patients with age ≥ 18 and dialysis vintage ≥ 3 months

Exclusion criteria

• coronary stents or pacemaker implantation
• metallic devices in body, such as artificial joints or pins
• contralateral or bilateral amputations
• pregnancy