Comparison of Late Mortality in Hospitalized Patients >70 Years of Age With Systolic Heart Failure Receiving Beta Blockers Versus Those Not Receiving Beta Blockers.
Am J Cardiol. 2008 Dec 15;102(12):1711-1717
Authors: Pascual-Figal DA, Redondo B, Caro C, Manzano S, Garrido IP, Ruipérez JA, Valdés M
Beta blockers are underprescribed to elderly patients with systolic heart failure (HF). We studied whether the prescription of a beta blocker is associated with a survival benefit in a nonselected population of patients >70 years of age hospitalized with acute HF and systolic dysfunction. We studied 272 consecutive patients >70 years (median 77.0, interquartile range 73.4 to 81.1) hospitalized with acute HF (left ventricular ejection fraction 34 +/- 8%) during a 2-year period. At discharge, beta-blocker therapy was prescribed in 139 patients (51.1%). A propensity score for the likelihood of receiving beta-blocker therapy was developed and showed a good performance (c-statistic = 0.825 and Hosmer-Lemeshow p = 0.820). After discharge, 120 patients (44.1%) died during the follow-up (median 31 months, interquartile range 12 to 46). Cox regression analysis showed a lower risk of death associated with beta-blocker prescription (p <0.001, hazard ratio [HR] 0.450, 95% confidence interval [CI] 0.310 to 0.655), which persisted after risk adjusting for the propensity score (HR 0.521, 95% CI 0.325 to 0.836, p = 0.007). In a propensity-matched cohort of 130 patients, there was a significantly lower mortality in patients receiving beta blockers (log rank 0.009, HR 0.415, 95% CI 0.234 to 0.734, p = 0.003). Risk reduction associated with beta blockade was observed with both high doses (HR 0.472, 95% CI 0.300 to 0.742, p = 0.001) and low doses (HR 0.425, 95% CI 0.254 to 0.711, p = 0.001). In conclusion, beta-blocker prescription at discharge in a nonselected population >70 years of age hospitalized with systolic HF is associated with a significantly lower risk of death even at low doses. This benefit remains consistent after adjustment for potential confounders.
PMID: 19064029 [PubMed - as supplied by publisher]