Associations between body composition and intradialytic hypotension (IDH), and between IDH and prognosis, in hemodialysis patients

Abstract

Abstract Background Previous studies describing relationships among body compositions, intradialytic hypotension (IDH), and mortality yielded inconsistent results. We studied associations between body composition and IDH, and between IDH and prognosis, in patients on hemodialysis (HD). Methods Participants were patients on maintenance HD and predilution online hemodiafiltration (HDF) (n = 303). IDH was defined as nadir systolic blood pressure (SBP) < 90 mmHg for ≥ 2 of 10 dialysis sessions during the exposure period (days 1–22). Clinical data at day 1 and post-dialysis body compositions using bioelectrical impedance analysis conducted once during the exposure period were collected. Differences between the IDH and non-IDH groups were analyzed. Kaplan–Meier survival curves of the IDH and non-IDH groups, logistic regression analyses of IDH, and Cox proportional hazard analyses of all-cause and cardiovascular (CV) mortality in all participants were also performed. Results In all participants, the median (median [interquartile range]) age was 67 [56–74] years, median dialysis duration was 76 [37–145] months, and diabetes prevalence was 42.6% (129/303). Compared with the non-IDH group (n = 274), the IDH group (n = 29) had a lower mean pre-dialysis SBP during the exposure period, longer dialysis duration, lower serum albumin levels, and higher median fat tissue index (10.7 [8.6–14.9] versus 9.5 [6.8–11.9] kg/m2, P < 0.05). The IDH group had lower 3-year survival for all-cause and CV mortality (P < 0.05). When adjusted for mean pre-dialysis SBP, mean ultrafiltration volume during the exposure period, HDF, dialysis duration, and serum albumin, fat tissue index, and lean tissue index were associated with IDH (P < 0.05), but body mass index and overhydration/extracellular water were not. After additional adjustments for age, sex, and diabetes mellitus, only fat tissue index was a significant predictor for IDH [odds ratio: 1.12 (95% confidence interval 1.02–1.25), P < 0.05]. IDH was also a significant predictor of 3-year all-cause and CV mortality (P < 0.05). Conclusions Increased fat tissue index was a significant risk factor for IDH in HD and HDF patients. Furthermore, IDH was a significant predictor of 3-year all-cause and CV mortality in HD and HDF patients.