Intradialytic hypotension, blood pressure changes and mortality risk in incident hemodialysis patients

Abstract

Background: Intradialytic hypotension (IDH) occurs frequently in maintenance hemodialysis (HD) patients and may be associated with higher mortality. We hypothesize that nadir intradialytic systolic blood pressure (niSBP) is inversely related to death risk while iSBP change (Δ) and IDH frequency are incrementally associated with all-cause mortality.

Methods: In a US-based cohort of 112 013 incident HD patients over a 5-year period (2007-11), using niSBP, ΔiSBP (pre-HD SBP minus niSBP) and IDH frequency (proportion of HD treatments with niSBP <90 mmHg) within the first 91 days of HD, we examined mortality-predictability at 1, 2 and 5 years using Cox models and restricted cubic splines adjusted for case-mix, comorbidities and laboratory covariates.

Results: We observed that niSBP of <90 and ≥140 mmHg had a 5-year mortality hazard ratio (HR) (95% confidence interval) of 1.57 (1.47-1.67) and 1.25 (1.18-1.33), respectively, compared with niSBP 110 to <120 mmHg. ΔiSBP of <15 and ≥50 compared with 21-30 mmHg had mortality HR of 1.31 (1.26-1.37) and 1.32 (1.24-1.39), respectively. Among patients with >40% IDH frequency, we observed a mortality HR of 1.49 (1.42-1.57) compared with 0% IDH frequency in fully adjusted models. These associations were robust at 1 and 2 years of follow-up.

Conclusion: In conclusion, we observed a U-shaped association between niSBP and ΔiSBP and mortality and a direct linear relationship between IDH frequency and mortality. Our findings lend some prognostic insight of HD blood pressure and hemodynamics, and have the potential to guide blood pressure management strategies among the HD population.

Keywords: blood pressure; hemodialysis; intradialytic hypotension; mortality; nutrition.

FIGURE 1

Algorithm of patient selection for cohort construction. TWICHD, thrice-weekly in-center HD.

FIGURE 2

Association of niSBP with all-cause mortality. (A) Restricted cubic splines showing 5-year mortality for unadjusted, case-mix and case-mix + MICS adjustments. (B) Five-year mortality for unadjusted, case-mix and case-mix + MICS. (C) One-, 2- and 5-year mortality for case-mix + MICS. In splines, dashed lines represent HR and solid lines and error bars represent 95% CI.

FIGURE 3

Association of ΔiSBP with all-cause mortality. (A) Restricted cubic splines showing 5-year mortality for unadjusted, case-mix, and case-mix+MICS adjustments. (B) Five-year mortality for unadjusted, case-mix and case-mix+MICS. (C) One-, 2- and 5-year mortality for case-mix+MICS. In splines, dashed lines represent HR and solid lines and error bars represent 95% CI.

FIGURE 4

Association of proportion of intradialytic hypotension with all-cause mortality. (A) Restricted cubic splines showing 5-year mortality for unadjusted, case-mix and case-mix + MICS adjustments. (B) Five-year mortality for unadjusted, case-mix and case-mix + MICS. (C) One-, 2- and 5-year mortality for case-mix + MICS. In splines, dashed lines represent HR and solid lines and error bars represent 95% CI.

FIGURE 5

Association of nadir intradialytic systolic blood pressure and 5-year all-cause mortality indicated by HR with 95% CI for various stratifications and adjusted for case-mix + MICs covariates. Abbreviations: CVC, central venous catheter; HD, hemodialysis; SBP, systolic blood pressure; KRU, residual urea clearance (mL/min).

FIGURE 6

Association of ΔiSBP and 5-year all-cause mortality indicated by HR with 95% CI for various stratifications and adjusted for case-mix+MICs covariates. Abbreviations: CVC,central venous catheter, HD, hemodialysis, SBP, systolic blood pressure, KRU, residual urea clearance (mL/min).

FIGURE 7

Association of proportion of IDH and 5-year all-cause mortality indicated by HR with 95% CI for various stratifications and adjusted for case-mix + MICs covariates. Abbreviations: CVC,central venous catheter; HD, hemodialysis; SBP, systolic blood pressure; KRU, residual urea clearance (mL/min).