The association of low blood pressure (BP) with high mortality is a characteristic for hemodialysis patients. This analysis clarifies the association of BP with mortality and stroke in Japanese hemodialysis (HD) patients and examines the association separately for patients with and without antihypertensive medication (BP meds). We analyzed 9134 patients from Japan in phases 1-4 (1999-2011) of the Dialysis Outcomes and Practice Patterns Study (DOPPS), a prospective cohort study of in-center HD patients. The association of patient systolic (SBP) and diastolic (DBP) blood pressure with all-cause and cause-specific mortality was assessed using adjusted Cox regression. A U-shaped association between BP and all-cause mortality was observed, with lowest mortality for baseline SBP 140-159 mmHg and DBP 65-74 mmHg. Both SBP and DBP were positively and monotonically associated with stroke-related death: hazard ratio (95% confidence interval) was 1.24 (1.01-1.53) per 20 mmHg higher SBP and 1.23 (1.05-1.44) per 10 mmHg higher DBP. No evidence of interaction was found between SBP and use of BP meds regarding all-cause mortality (P for interaction = 0.97); the association between SBP and stroke-related death was slightly stronger among patients not on BP meds than patients on BP meds (P for interaction = 0.09). In Japanese HD patients, both low and high BP are associated with all-cause mortality. This analysis also documents a positive and monotonic association of BP with stroke-related deaths. Although our analysis indicates that the prescription of BP meds to hypertensive patients might protect against stroke-related death, additional study is warranted.
Keywords: Hypertension; change in blood pressure over time; outcomes research; reverse epidemiology; systolic blood pressure.
© 2014 International Society for Hemodialysis.