ACCURACY AND PRECISION OF SEGMENTAL BIOIMPEDANCE TECHNIQUE TO ASSESS PERIDIALYTIC CHANGES IN TOTAL BODY WATER
INTRODUCTION AND AIMS: Fluid volume assessment remains a challenge in clinical nephrology. Bioimpedance has gained widespread popularity and has the potential to guide fluid therapy in various aspects. Bioimpedance allows the assessment of total body water (TBW) thus providing insight into fluid status. We aimed to investigate the accuracy of TBW measurements with a segmental multifrequency bioimpedance device (InBody 770; InBody USA, Cerritos, CA, USA) by comparing the peridialytic changes in weight and TBW in hemodialysis (HD) patients. METHODS: Patients underwent pre- and post-dialysis bioimpedance assessments on multiple occasions. Bioimpedance measurements were conducted with patients standing barefooted on the platform of the InBody 770 device. That platform serves as a scale and has also sensing / injecting electrodes integrated. Additional sensing / injecting electrodes are located in hand grips that are hold by the patient. For this analysis we used the peridialytic weight change (Wt) as the reference. We investigated the accuracy and precision of the InBody 770 to detect peridialytic changes in TBW (TBW; expressed as kg) as compared to Wt. We report the mean of the differences of Wt to TBW and employed Bland-Altman analysis to assess accuracy, precision, and proportional error. A regression model with fixed effects accounted for within-patient correlation using generalized estimating equations. RESULTS: We analyzed 303 treatments from 22 patients (average of 14±9 treatments per patient). Patients characteristics were as follows: age 61.4±14 years, 59% male; body mass index 25.7±4.6 kg/m2, and interdialytic weight gain 2.5±0.48 kg. TBW and Wt differed by an average of 0.19 kg with a standard deviation of 0.82 kg. A significant proportional bias for the relationship between difference been TWB and Wt was found (slope estimate -0.15; P=0.047) (Figure 1). CONCLUSIONS: Our analysis of peridialytic change of TBW as measured by the InBody 770 device compared to peridialytic weight change shows clinically acceptable accuracy, while precision is less satisfactory. Studies in a larger population with greater diversity of body composition, a known confounder of accuracy and precision of bioimpedance, are needed to draw final conclusions.
