Intradialytic BP Variability is Associated with Cardiovascular Mortality and Hospitalization in HD Patients

Xuelei Zhang,Ling Xu,Peiyi Zhou,Dongqi Song,Jian Wu,Lifang Jia
DOI: https://doi.org/10.1111/1744-9987.13745
2021-01-01
Therapeutic Apheresis and Dialysis
Abstract:Introduction Studies showed that pre-dialysis BP variability (BPV) was an independent risk factor of cardiovascular disease (CVD) among HD patients, but which is limited on how intradialytic BPV affects prognosis. Methods In this study, we designed a retrospective cohort study to examine the association between intradialytic BPV and CVD outcomes in HD patients. A total of 202 patients who underwent HD in our center were included, and all intradialytic BP measurements of November 2017 were obtained from the database. Patients were divided into four groups according to variability independent of the mean (VIM) interquartile. Results The mean age was 62.1 +/- 14.3 years, 60.9% were male, and median VIM was 14.75 (12.60-18.59). Multiple-regression analyses showed patients age, dialysis vintage, serum albumin, and the percentage of intradialytic weight gain as significant predictors of VIM (all p values were <0.05). Kaplan-Meier survival curves showed that CVD mortality was greater in patients with higher VIM (p = 0.05), whereas all-cause mortality had no significant difference between the four groups overall (p = 0.149). Furthermore, multivariate regression analyses demonstrated that VIM (HR = 1.091, p < 0.004) and age (HR = 1.059, p = 0.003) were significant independent predictors for CVD death. Logistic-regression models revealed that higher VIM groups were more likely to have CVD-related hospitalization (OR = 1.085, p = 0.030), whereas the association between VIM and all-cause hospitalization was not statistically significant (OR = 1.015, p = 0.669). Conclusions This retrospective study suggested that higher intradialytic BPV was associated with increasing age, longer dialysis vintage, lower albumin, and greater ultrafiltration; intradialytic BPV could be an effective predictor for CVD mortality and hospitalization in the HD population.
What problem does this paper attempt to address?