The association of interdialytic blood pressure variability with cardiovascular events and all-cause mortality in haemodialysis patients.

Background: Long-term pre-dialysis blood pressure variability (BPV) in haemodialysis patients is associated with increased cardiovascular risk. The association of the main haemodynamic culprit in dialysis, that is, short-term BPV, with outcomes has not been investigated. We examine the prognostic role of short-term BPV for mortality and cardiovascular events in this population.

Methods: A total of 227 haemodialysis patients underwent 44-h ambulatory monitoring during a standard interval and were followed-up for 30.17 ± 17.70 months. We calculated SD, weighted SD (wSD), coefficient of variation (CV) and average real variability (ARV) of BP with validated formulas. The primary endpoint was first occurrence of all-cause death, non-fatal myocardial infarction or non-fatal stroke. Secondary endpoints were: (i) all-cause mortality, (ii) cardiovascular mortality and (iii) a combination of cardiovascular events.

Results: Cumulative freedom from the primary endpoint was similar for quartiles of pre-dialysis and 44-h systolic BP (SBP), but was progressively longer for increasing quartiles of 44-h SBP-SD (P = 0.014), wSD (P = 0.007), CV (P = 0.031) and ARV (83.9, 71.9, 70.2 and 43.9% for quartiles 1-4; P < 0.001). Higher quartiles of 44-h SBP-ARV were associated with higher risk of all studied outcomes. Among diastolic BPV indices, 44-h diastolic BP (DBP)-CV and 44-h DBP-ARV were associated with increased risk for the composite cardiovascular outcome. In Cox regression analysis, SBP-BPV was related to the primary endpoint, independently of SBP levels and interdialytic weight gain [ARV: hazard ratio (HR) 1.115, 95% confidence interval (95% CI) 1.048-1.185]. This association become insignificant after adjustment for pulse wave velocity (PWV; HR 1.061, 95% CI 0.989-1.137), and further attenuated after additional adjustment for age, dialysis vintage, gender, comorbidities and prevalent cardiovascular disease (HR 1.031, 95% CI 0.946-1.122).

Conclusions: Increased BPV during the interdialytic interval is associated with higher risk of death and cardiovascular events, whereas ambulatory BP levels are not. This association was not independent after adjustment for PWV, other risk factors and prevalent cardiovascular disease. Short-term BPV could be a mediator promoting the adverse cardiovascular profile of haemodialysis patients.