PCV24 - Comparative effectiveness of angiotensin receptor blockers in chronic heart failure: A network meta-analysis

Abstract

Chronic Heart failure (CHF) is associated with significant morbidity and mortality. Angiotensin receptor blockers (ARBs) are one of most commonly prescribed drug class for patients with CHF among patients who are intolerant to angiotensin converting enzyme inhibitors (ACEI). A recently published Cochrane review concluded that ARBs as a class confer no additional benefit on mortality or total hospitalization as compared to placebo or ACEI. The aim is to compare efficacy of ARBs (losartan, valsartan, candesartan, and telmisartan) on hospitalization and mortality relative to each other and to placebo among patients with CHF. Studies on ARBs were identified from a recently published Cochrane systematic review. A network meta-analysis (NMA) was conducted for three outcomes; all-cause hospitalizations, hospitalization due to heart failure, and all-cause mortality using WinBUGs. Binomial likelihood models were run for each outcome and analyses were conducted on an odds ratio scale. Fixed or random effects models were run to estimate relative treatment effects Sixteen studies were identified from the review. Five studies reported data on all-cause hospitalization and six on hospitalizations due to heart failure, for two drugs (candesartan and losartan).Data on mortality was reported in 16 studies for losartan, valsartan, candesartan, and telmisartan. ARBs did not have significantly different effects from placebo or each other on all the three outcomes studied. NMA can also be used to rank treatments. Within this we found that compared to all other ARBs, telmisartan and losartan had the highest probabilities of being the best treatments to reduce mortality and hospitalizations, respectively. Results of the NMA indicated that there was no significant difference between any of the individual ARBs, ACEI or placebo on mortality and hospitalization. A further analysis that uses meta-regression to adjust for co-morbid conditions such as diabetes, hypertension and ischemic heart disease is warranted.