Cardiac medications and their association with cardiovascular events in incident dialysis patients: cause or effect?

BACKGROUND: There are no large randomized, controlled trials evaluating the efficacy of common cardioprotective medications on cardiovascular outcomes in the dialysis population. We aimed to determine the association between cardioprotective medications and a composite end point of cardiovascular events or death in an incident dialysis cohort.
METHODS: Medicare claims data were utilized to determine outcomes in participants of the Dialysis Morbidity and Mortality Wave 2 cohort. Information was gathered at baseline regarding demographics, comorbidities, medication use, and lab data. Cox proportional hazard models were developed to determine the association between cardioprotective medication use at baseline and the subsequent development of a composite of cardiovascular events or death.
RESULTS: Two thousand two hundred thirty-four of 3044 individuals experienced a cardiovascular event or death. Use of the following medications at baseline was associated with an increased event rate: nitrates, antiplatelet agents, and warfarin. There was no association between the use of angiotensin-converting enzyme inhibitors, beta-blockers, or 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors and a composite cardiovascular outcome or death. Only use of a calcium channel blocker at baseline was associated with a reduced risk of events. This beneficial association did not persist when death was excluded as an outcome. Many of the associations observed varied substantially depending on the outcome utilized.
CONCLUSION: Use of cohort data to determine an association between medication use and outcome is difficult because the indication for the medication confounds the observed association. Before widespread practice changes, rigorous clinical trials of common cardioprotective medications on cardiovascular morbidity and mortality in dialysis should be performed.