Skip to main content
Effective Health Care Program

Comparative Effectiveness of Antihypertensive Therapies on Cardiovascular Disease Outcomes In End-Stage Renal Disease

Abstract Archived

Topic Abstract

Background: The optimal antihypertensive therapeutic regimen to improve cardiovascular disease (CVD)-related morbidity and mortality among patients with hypertension and end-stage renal disease (ESRD) is not known. Patients on dialysis have generally been excluded from studies evaluating the effectiveness of antihypertensive therapy. Also, studies evaluating the independent effect of separate classes of antihypertensive therapies on outcomes among dialysis patients have yielded inconsistent results. Given limitations and conflicting results of past studies, clinicians remain uncertain as how to best manage blood pressure in dialysis patients and whether certain classes of antihypertensive therapy offer advantages independent of blood pressure lowering.

Objective: To assess the comparative effectiveness of angiotensin converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), beta blockers, and calcium channel blockers on risk of cardiovascular outcomes (morbidity and mortality), all cause mortality, and other important patient outcomes (e.g., health-related quality of life).

Study Design: Prospective and retrospective cohort studies

Methods: We will use three complimentary and nationally representative data sources to achieve our objective [United States Renal Data System (USRDS) national registry data, Dialysis Clinics Inc (DCI) data, and Choices for Healthy Outcomes in Caring for ESRD (CHOICE) Study data]. Within each data source, we will perform several longitudinal analyses to assess the comparative effectiveness of prescribed antihypertensive therapeutic regimens with all cause and CVD mortality and morbidity.

We hypothesize treatment with ACE inhibitors and/or ARBs is independently associated with reduced all-cause mortality, CVD mortality, and morbidity compared to beta blockers and calcium channel blockers. We also hypothesize that this relationship is primarily mediated by changes in blood pressure, but differs by underlying comorbidity and type of CVD event.

We will model the association between antihypertensive medication use and outcomes using various longitudinal models, including Cox proportional hazards regression. Analyses will adjust for potential confounders of the relation between antihypertensive medication and dependent variables, and we will also account for the interaction between each class of antihypertensive medication, blood pressure, and volume status. We will also utilize models to allow for causal estimates of exposure in presence of time varying confounders, and we will introduce statistical interactions in the models to quantitatively assess the direction and magnitude of effect modification. We will conduct sensitivity analyses to assess the robustness of our findings using alternative techniques including competing risk models and accelerated failure time models. We will look for consistency and precision of the estimates across the studies as a means of validating our findings.

Expected Outputs: At least two scientific reports will be generated regarding the different outcomes examined (mortality and morbidity).

Expected Date of Project Completion: June, 2013

EHC Priority Conditions: End-Stage Renal Disease (ESRD) and Cardiovascular Disease