Beta-Blockers Associated With Reduced Mortality in Diastolic Heart Failure

Pam Harrison

November 16, 2014

STOCKHOLM, SWEDEN  — The use of beta-blockers in patients with heart failure and preserved ejection fraction (HFPEF) is associated with a lower risk of all-cause mortality but not the combined end point of all-cause mortality or first hospitalization for heart failure, data from the Swedish Heart Failure Registry indicate[1].

Dr Lars Lund (Karolinska University Hospital, Stockholm, Sweden) and colleagues found that after a median of 755 days, 36% of patients who received a beta-blocker had died compared with 46% of those who did not receive one (P<0.001).

Of those who received a beta-blocker, 84% were still alive at 1 year compared with 78% of those who did not. At 5 years, 51% of those who received a beta-blocker were still alive compared with 41% of those who did not (P<0.001). The unadjusted hazard ratio (HR) for all-cause mortality throughout follow-up was 0.73 (P<0.001).

"These findings should not change management of HFPEF at this time since this was not a randomized trial and the magnitude of risk reduction was modest," Lund told heartwire . "But the main important finding is that beta-blockers were associated with reduced mortality in heart-failure patients with preserved ejection fraction, so we feel that they should be examined in a large randomized clinical trial in patients with HFPEF." The study was published online November 16, 2014 and in the November 19, 2014 issue of the Journal of the American Medical Association.

The Swedish Heart Failure Registry collects data from 67 hospitals (with in- and outpatient units) and 95 outpatient primary-care clinics in Sweden. Patient data were entered into the registry between July 1, 2005 and December 30, 2012.

Investigators identified 19 083 patients with HFPEF; 15 786 had received beta-blockers and 3297 did not. Of these patients, 8244 patients with HFPEF were matched using a 2:1 ratio based on age and propensity scores for beta-blocker use. In this propensity-matched cohort, 5496 patients had received beta-blockers and 2748 had not.

In the propensity-score–matched HFPEF cohort, all-cause mortality at a median follow-up of 709 days was also lower, at 41% in the beta-blocker group compared with 45% in the non–beta-blocker group (P=0.04).

At 1 year, 80% of the beta-blocker group were still alive compared with 79% of those who did not receive beta-blockers. At 5 years, 45% of the beta-blocker group vs 42% of the non-beta-blocker group were alive.

The HR throughout follow-up for all-cause mortality in the propensity-score–matched HFPEF was 0.93 (P=0.04). Of the overall cohort who received beta-blockers, 62% were also free from heart-failure hospitalization at 1 year compared with 58% of the non-beta-blocker group. In the propensity-score matched cohort, 58% of the beta-blocker group were free from heart failure hospitalization at 1 year compared with 59% of those who did not receive beta-blockers.

Positive-Control Analysis

Investigators also performed a positive-control analysis in heart-failure patients with reduced ejection fraction (HFREF). "We already know that beta-blockers reduce mortality in systolic heart failure," Lund explained, "so we performed this positive-control analysis to see if results from our registry study would match findings that we already know are true."

As expected, all-cause mortality was also lower in the HFREF positive-control analysis; 84% of patients who received beta-blockers were alive at 1 year compared with 73% of those who did not receive beta-blockers. The HR for all-cause mortality throughout follow-up in the HFREF cohort was 0.63 (P<0.001).

Investigators did a similar analysis of the HFREF cohort where they propensity-score–matched 4054 patients who received beta-blockers and 2027 who did not.

In contrast to the HFPEF cohort, beta-blockers were associated with an 11% reduced combined end point of mortality or first hospitalization for heart failure in the HFREF cohort (HR 0.89; P=0.001). Lund suggested hospitalization for heart failure was not reduced by beta-blocker use in the HFPEF cohort because it was not an adjudicated end point, so it is very unreliable.

"We also know that rehospitalization for heart failure is not as much affected by therapy as is mortality or hospitalization over the long term, so in this HFPEF cohort that is again what we saw: that treatment effect on hospitalization is less than its effect on mortality."

Reasonable to Postulate

In an accompanying editorial, Drs Susan Cheng and Marc Pfeffer (Brigham and Women's Hospital, Boston, MA) suggest that because heart failure with either reduced or preserved ejection fraction involves myocardial demand in excess of reserve, it is reasonable to postulate that the benefits of beta-blocker therapy in HFREF might also apply to HFPEF[2].

"As expected, the large proportion of HFPEF patients treated with beta-blockers were significantly different from the smaller proportion of those not treated with beta-blockers across multiple characteristics in the cohort," they observed.

Investigators therefore used propensity-score matching to create a subsample of beta-blocker treated and untreated patients who were more balanced in terms of demographics and clinical characteristics. When the matched groups were compared, "presence of beta-blocker therapy was associated with an absolute 3% and a relative 8% lower risk of all-cause mortality at 5 years [P=0.02]," they write.

On the other hand, the editorialists point out that a large number of individuals treated with beta-blockers who could not be matched with patients not treated with beta-blockers using propensity-score matching indicates "strong selection bias" in beta-blocker-prescribing patterns.

They therefore conclude, as did the authors, that more definitive information about whether beta-blocker therapy is effective for preventing important outcomes in HFPEF requires well-designed and well-conducted randomized clinical trials.

 Lund has severed on speaker's bureaus or received consulting fees from AstraZeneca, Novartis, Vifor Pharma, and Boston Scientific. Cheng reported no support from industry. Pfeffer has served as a consultant to pharmaceutical companies including Merck, Novartis, and Sanofi.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.