Key Points
• In the randomized SMART-DECISION trial conducted in Korea, discontinuation of long-term beta blocker therapy was non–inferior to continuation in stabilized post MI patients without heart failure or left ventricular systolic dysfunction.
• The primary composite endpoint of death from any cause, recurrent MI, or hospitalization for heart failure occurred in 7.2% of the discontinuation group and 9.0% of the continuation group, with a hazard ratio of 0.80 and P = 0.001 for noninferiority.
• Secondary clinical outcomes were generally similar between groups, although uncontrolled tachycardia occurred more often after beta blocker discontinuation during follow up.
The role of long-term beta-blocker therapy after a myocardial infarction in patients without ventricular systolic dysfunction or heart failure is unclear in the era of contemporary coronary-artery reperfusion and secondary prevention interventions. The SMART-DECISION trial assessed whether stopping long term beta blocker therapy was noninferior to continuing treatment with respect to major clinical outcomes.
The results were presented at ACC 2026 with simultaneous publication in NEJM.
SMART-DECISION was an investigator initiated, open label, randomized noninferiority trial conducted at 25 sites in Korea [NCT04769362]. The trial enrolled 2,740 stabilized post MI patients receiving beta blocker therapy for at least 1 year, with LVEF of 40% or greater and no heart failure. After exclusions, 2,540 patients underwent randomization, with 1,246 assigned to discontinue beta blocker therapy and 1,294 assigned to continue treatment. Median time from index MI to randomization was approximately 4.7 years, highlighting the late stable phase targeted by the trial.
Baseline characteristics were well balanced between groups. Mean age was approximately 63 years, about 13% of patients were women, and slightly more than half had experienced STEMI as the index infarction. Nearly all patients had undergone percutaneous coronary intervention, and baseline LVEF was preserved overall, with a median of 59% in both groups.
The primary endpoint was a composite of death from any cause, recurrent MI, or hospitalization for heart failure. Over approximately 4 years of follow up, the primary endpoint occurred in 7.2% of patients assigned to discontinue beta blockers and 9.0% of those assigned to continue therapy. This corresponded to a hazard ratio of 0.80 with a 95% confidence interval of 0.57 to 1.13, meeting the prespecified criterion for noninferiority (P = 0.001 for noninferiority). A per protocol sensitivity analysis was consistent with the main findings, with event rates of 7.2% versus 8.8% and a hazard ratio of 0.79.
Secondary endpoints were broadly similar between treatment groups. Death from any cause occurred in 2.4% of the discontinuation group and 3.4% of the continuation group, while recurrent MI occurred in 2.3% and 2.6%, respectively. Hospitalization for heart failure was infrequent in both groups, occurring in 2.2% of patients who discontinued beta blockers and 2.1% of those who continued therapy. The composite of cardiovascular death, recurrent MI, or heart failure hospitalization was 6.2% in both groups. Stroke was numerically lower in the discontinuation group, 0.7% versus 1.8%, though this should be interpreted cautiously.
Among biomarker and imaging endpoints, changes in LVEF did not differ significantly between groups over time, while changes in NT-proBNP favored discontinuation at 12 months, with a reported P = 0.03, although later differences were not statistically significant.
One notable tradeoff emerged in physiologic follow up. Rates of uncontrolled hypertension were similar between groups, but uncontrolled tachycardia was more frequent after beta blocker discontinuation, occurring in 7.1%, 6.9%, and 6.5% at 12, 24, and 30 months, respectively, compared with 1.5%, 1.9%, and 1.9% in the continuation group.
The investigators acknowledged several limitations, including the open label design, the highly selected stable post MI population, and limited representation of women and patients with mildly reduced LVEF. As such, the findings may not apply to higher risk patients early after MI or to those with a greater burden of comorbidity.
Overall, SMART-DECISION is the first randomized trial to demonstrate that discontinuation of long-term beta blocker therapy was noninferior to continuation in stable post MI patients without heart failure or left ventricular systolic dysfunction. The findings suggest that in carefully selected patients beyond the first year after MI, routine continuation of beta blockers may not be necessary for protection against major clinical events.