Blood Pressure Level and Risk of All-Cause Mortality in Patients With Kidney Failure on Maintenance Peritoneal...

To identify, appraise, and synthesize relevant epidemiologic studies to better understand the relations between blood pressure (BP) and outcomes in patients with kidney failure receiving maintenance peritoneal dialysis (PD) and to examine how the available evidence aligns with current clinical guidelines.

Systematic review and meta-analysis.

We conducted a comprehensive search of 5 databases (PubMed, Web of Science, Embase, CINAHL, Cochrane Library) through April 2023 (and abstracts from leading nephrology conferences from 2019-2023). Studies were eligible if they included patients receiving maintenance PD and reported associations between BP levels and all-cause mortality.

Systolic BP, diastolic BP, or presence/absence of hypertension.

All-cause mortality.

Dual independent screening and full-text review were performed. Where appropriate, data were then pooled using a random-effects meta-analysis. Adjusted data were not required for inclusion.

Thirty observational studies were included in the systematic review. Twenty-three studies were retrospective; 7 were prospective. Sixteen studies were single center, 13 were multicenter, and one study was unclear as it was presented only in abstract form. More than two-thirds were from Asia. Of the 28 full-length articles, only 9 had BP as the primary exposure, with the rest including BP only as a covariate in multivariable models. In the meta-analysis, compared with systolic BP 100-140 mm Hg, systolic BP >140 mm Hg was associated with a statistically nonsignificant 15% higher risk of all-cause mortality (RR, 1.15; 95% CI, 0.99-1.34). No association was found between diastolic BP >90 mm Hg (vs diastolic BP ≤90 mm Hg) and mortality (RR, 0.92; 95% CI, 0.53-1.57).

High level of heterogeneity and high risk of bias among the included studies.

The existing epidemiology literature is unable to offer definitive guidance regarding BP treatment targets for PD patients due to heterogeneity, risk of bias, and lack of statistical significance.