By Jamie L. W. Kennedy, MD, FACC

Medical Director, Heart Transplant, Inova Heart and Vascular Institute, Falls Church, VA

SYNOPSIS: Discontinuation of beta-blockers in heart failure with preserved ejection fraction patients significantly improved quality of life scores and expanded exercise capacity.

SOURCE: Palau P, Seller J, Domínguez E, et al. Effect of ß-blocker withdrawal on functional capacity in heart failure and preserved ejection fraction. J Am Coll Cardiol 2021;78:2042-2056.

Heart failure with preserved ejection fraction (HFpEF) is a complex syndrome of exercise intolerance and fluid retention despite normal or near-normal left ventricular systolic function. Cardiac contributions have taken center stage based on findings such as elevated left atrial pressure, impaired left ventricular diastolic function, and chronotropic incompetence. However, Houstis et al revealed the heterogenous and multiorgan system nature of HFpEF by using cardiopulmonary exercise testing (CPET) to identify the cause of exercise limitation in a series of HFpEF patients.1 Every step of the oxygen delivery pathway, from alveolar ventilation to peripheral oxygen use, was abnormal in at least one patient. Most patients studied exhibited defects at multiple pathway points.

Patients with HFpEF often take beta-blockers for several reasons, including comorbid conditions (e.g., atrial fibrillation) and coronary artery disease, as well as less compelling indications (e.g., hypertension). Beta-blocker use has been associated with increased risk of heart failure hospitalization in a retrospective analysis of the TOPCAT study for patients with left ventricular ejection fraction (LVEF) > 50%. There was no increased risk in patients with LVEF 45% to 49%.

Palau et al assessed the effects of beta-blocker use on exercise capacity for HFpEF patients. They enrolled adults with New York Heart Association class II-IV symptoms, LVEF > 50%, left ventricular internal dimension end-diastolic < 6 cm, N-terminal-pro hormone BNP measurement of greater than 125 pg/mL, echo evidence of HFpEF (i.e., left ventricular hypertrophy, left atrial enlargement, or diastolic dysfunction), prior heart failure hospitalization, and blunted heart rate response during CPET (defined as chronotropic index < 0.62). Patients with significant valvular heart disease, uncontrolled arrhythmias or hypertension, recent acute coronary syndrome or ischemia during CPET, significant pulmonary disease, chronic treatment with other rate control medications, resting heart rate > 75 bpm, or other life-limiting comorbid conditions were excluded.

The authors screened 250 patients, narrowing the enrollment to 52 patients (average age was 74.5 years, 60% were women, and average BMI was 31 kg/m2). Comorbid conditions included hypertension (88%), dyslipidemia (71%), diabetes mellitus (40%), atrial fibrillation (38%), and ischemic heart disease (23%). Prescribed medications included loop diuretics (85%), ACE inhibitors or ARBs (75%), statins (67%), and mineralocorticoid antagonists (11%). Bisoprolol was by far the most common beta-blocker prescribed (88% of patients at a median dose of 2.5 mg daily).

Baseline assessment included echo, CPET, Minnesota Living with Heart Failure Questionnaire, and cognitive assessment. Patients were randomized to two groups. Group A was tapered off beta-blockers over three days, while group B continued beta-blocker at baseline dose. Complete assessment was repeated at day 15, after which group A restarted beta-blockers and group B tapered off. Complete assessment was repeated at day 30. One patient in each arm developed atrial fibrillation with rapid ventricular response and was removed from the study. One patient in group A suffered a stroke and was withdrawn. Three patients in group B withdrew consent.

Baseline CPET showed patients were fairly limited, with mean peak VO2 at 12.4 ± 2.9 mL/kg/min, 72 ± 17.8% predicted. All enrolled patients completed a maximal baseline test with respiratory exchange ratio > 1.05. Mean resting heart rate was 64.8 ± 8.8 bpm and peak 97.2 ± 14.7 bpm for a chronotropic index of 0.41 ± 0.14. Patients in Group A improved their peak VO2 to 14.06 ± 3.35 mL/kg/min when off beta-blocker; after re-introduction, the rate lowered to 12.26 ± 3.24 mL/kg/min (P < 0.001). Group B produced similar results, with continued beta-blocker use peak VO2 12.24 ± 3.05 mL/kg/min and discontinued 14.48 ± 3.79 mL/kg/min (P < 0.001).

Overall, the average increase in peak VO2 off beta-blocker was 2.1 ± 1.3 mL/kg/min. There was no evidence of period effect. Similar results were obtained when VO2 percent predicted was analyzed (average increase of 11.7 ± 2.3%). Peak heart rate during exercise increased from 97 bpm to 127 bpm when beta-blockers were withdrawn. A mediation analysis suggested the increase in peak heart rate accounted for 36% of the improvement in peak VO2. Quality of life scores improved with discontinuation of beta-blockers (from 27.4 to 22.3). There was no change in scores on cognitive testing when beta-blockers were discontinued.

Adverse events included atrial fibrillation and rapid ventricular response in two patients; both underwent successful cardioversions. At the 60-day follow-up visit, the beta-blocker dose decreased by at least 50% in 19 patients and completely discontinued for 27 patients. There were three cardiovascular hospitalizations for three patients at one month, and four cardiovascular hospitalizations for three patients at six months; none occurred during the beta-blocker withdrawal period. There were no deaths during six months of follow-up.

The authors demonstrated a significant improvement in exercise capacity with discontinuation of beta-blockers in HFpEF patients. Also, there was a corresponding improvement in quality of life scores, albeit over short-term follow-up. A longer-term study of morbidity and mortality is needed.


Palau et al did not analyze the data based on indication for beta-blocker therapy. While awaiting more data on this question, transitioning HFpEF patients with weaker indications for beta-blocker therapy (e.g., treating hypertension with alternative agents) seems likely to result in symptomatic improvement with minimal risk of adverse events. I would be much more cautious when treating patients with atrial fibrillation or those who present with significant coronary artery disease.


  1. Houstis NE, Eisman AS, Pappagianopoulos PP, et al. Exercise intolerance in heart failure with preserved ejection fraction: Diagnosing and ranking its causes using personalized O2 pathway analysis. Circulation 2018;137:148-161.