The ABCs of cardioprotection in dialysis patients: a systematic review.

BACKGROUND: Several classes of medications have been shown to decrease all-cause and cardiovascular mortality in the general population. However, dialysis patients have been systematically excluded from these large trials, and the benefits of angiotensin-converting enzyme (ACE) inhibitors, adrenergic beta antagonists (beta-blockers), and calcium channel blockers (CCBs) are uncertain in this population. STUDY
DESIGN: We performed a systematic review using the MEDLINE database (inception to October 14, 2007) to identify studies. SETTING & POPULATION: Incident and prevalent dialysis patients. SELECTION CRITERIA FOR STUDIES: English-language randomized controlled trials (RCTs) and observational studies investigating the use of ACE inhibitors, beta-blockers, and CCBs in humans. INTERVENTION: ACE-inhibitor, beta-blocker, and CCB administration. OUTCOMES: Decreases in all-cause and cardiovascular mortality and cardiovascular morbidity.
RESULTS: 674 reports yielded 13 suitable reports for ACE inhibitors, 12 for beta-blockers, and 6 for CCBs. Because most studies investigated more than 1 class of drug, there were 17 unique reports; 2 were RCTs, 1 was a "pseudo-RCT," and 14 were observational studies. Meta-analysis was not possible because of the heterogeneity of studies. There is considerable discrepancy in the literature about the utility of these agents. ACE inhibitors have not consistently shown survival benefits in either the single RCT or observational studies. beta-Blockers showed mortality benefit in only 1 large cohort study plus an RCT of patients with congestive heart failure, but results were not duplicated in other studies; the magnitude of beta-blocker benefit after myocardial infarction was similar in dialysis and nondialysis individuals in another study. CCBs show the most consistent benefits, albeit only from observational studies, of the classes examined. LIMITATIONS: Several major limitations were present, including a paucity of RCTs and nonrandom treatment assignment and lack of data for longitudinal medication exposure in observational studies.
CONCLUSIONS: Despite considerable uncertainty about the benefits and risks in this population, for individuals with well-established traditional indications for these medications, refraining from prescribing them may be imprudent at this time. However, RCTs, as well as well-designed observational studies that adjust for nonrandom treatment assignment and longitudinal drug exposure, are needed.