Quantifying acute changes in volume and nutritional status during haemodialysis using bioimpedance analysis

Aim: To determine the precision of multi-frequency bioimpedance analysis (MFBIA) in quantifying acute changes in volume and nutritional status during haemodialysis, in patients with end-stage renal disease (ESRD).

Methods: Using whole-body MFBIA, we prospectively studied changes in total body water (TBW), extracellular volume (ECV), intracellular volume (ICV), lean body mass (LBM), body cell mass (BCM) and fat mass (FM), pre- and post-haemodialysis and tested the agreement of volume changes with corresponding acute weight change and ultrafiltration volume (UF) using Bland-Altman analysis.

Results: Forty-four prevalent and 17 incident haemodialysis patients were studied (median age 55 years, 56% males). MFBIA-derived TBW, ECV, ICV, LBM and BCM were significantly reduced after haemodialysis (P < 0.001), but FM remained constant. TBW change estimated weight change with mean bias of -0.52 L, with 56/61 (91.8%) data points within limits of agreement (-2.74 L, 1.69 L). TBW change estimated UF with mean bias of -0.62 L, with 55/61 (90.2%) data points within limits of agreement (-2.68 L, 1.43 L). ECV change underestimated weight change and UF with mean bias of -1.17 L and -1.27 L respectively. Similarly, ICV change underestimated both clinical measures with corresponding mean bias of -1.34 L and -1.44 L. Comparing incidents versus prevalent haemodialysis patients, TBW change estimated weight change with smaller mean bias (-0.10 L vs-0.69 L, respectively) and narrower limits of agreement.

Conclusion: Multi-frequency bioimpedance analysis-derived TBW chan e has the best agreement with acute clinical volume change during haemodialysis compared to ECV or ICV change alone, but overall degree of precision remains poor. Nutritional assessment using LBM and BCM measurements is significantly confounded by hydration status.