Heart Rate Variability and Prognosis in Hemodialysis Patients: A Meta-Analysis

Heart rate variability (HRV) means the variation in time of beat-to-beat interval. Lower HRV has been shown to be related with death and cardiovascular events in previous studies. In the last few years, the number of patients with ESRD has increased steadily. Maintenance hemodialysis is the most prevalent renal replacement therapy in patients with ESRD. This study aims to investigate if decreased HRV is an independent predictor of mortality in maintenance hemodialysis patients.

Pubmed/Medline, EMBASE, Ovid, the Web of Science, and the Cochrane Central Register of Controlled Trials databases were searched up to October 1, 2019, for full-text articles in English. Cohort studies reporting the association between HRV and prognosis in hemodialysis patients were selected. Data extraction was performed by 2 reviewers independently, with adjudication by a third reviewer. Extracted data included the study characteristics, HRV measurement and research outcomes. Hazard ratios (HRs) and 95% confidence interval (CI) were pooled in a random-effects model for outcomes of all-cause and cardiovascular mortality. Heterogeneity assessment, subgroup analyses, and sensitivity analysis were conducted.

A total of 7 studies were eligible. HRV metrics consist of SDNN, SDANN, RMSSD, pNN50, HRVTI, ULF, VLF, LF, HF, LF/HF ratio, HRT, DC, and scaling exponents α1 and α2. Decreased HRV was associated with higher all-cause mortality (HR: 1.63, 95% CI: 1.11-2.39, p = 0.014) and cardiovascular mortality (HR: 1.07, 95% CI: 1.00-1.15, p = 0.045). Among the different HRV metrics, decreased SDANN (p < 0.001) and decreased LF/HF ratio (p = 0.001) were identified as predictors of all-cause death. Decreased SDNN, SDANN, and LF/HF ratio were identified as predictors of cardiovascular death (p = 0.004, p = 0.001, and p = 0.002).

Decreased HRV is associated with higher risk of all-cause and cardiovascular death in the hemodialysis population. Decreased SDANN and LF/HF were identified as predictors of both all-cause and cardiovascular mortality, while the utility of other HRV metrics requires further investigation. The protocol for this study was registered with PROSPERO (CRD42019141886).