By Michael H. Crawford, MD, Editor

SYNOPSIS: The use of beta-blockers in cocaine users is controversial, and there are few data on their use in cocaine-associated heart failure. This prospective, observational, registry study of cocaine-associated heart failure patients showed that carvedilol is safe and effective in such patients.

SOURCES: Banerji D, et al. Carvedilol among patients with heart failure with cocaine-use disorder. JACC Heart Fail 2019;7:771-778.

Page RL 2nd, Allen LA. Cocaine, heart failure, and carvedilol: Triangulating the safety of beta-blocker therapy. JACC Heart Fail 2019;7:779-781.

Experience with selective beta-adrenergic antagonists in cocaine toxicity has not proven effective, possibly because of unopposed alpha-adrenergic activity. Nonselective beta-blockers, which feature alpha-blocking properties, currently carry a class IIb recommendation for cocaine-associated NSTEMI. However, their use in heart failure with reduced left ventricular ejection fraction (EF) associated with cocaine use is unclear. Banerji et al tested the hypothesis that carvedilol treatment in cocaine use-associated heart failure is safe.

Researchers used a prospective, observational, registry study of all patients admitted with heart failure at one tertiary medical center in 2011. Cocaine use was self-reported or based on urine toxicology. At this hospital, carvedilol was the only beta-blocker administered for cocaine use-associated heart failure. The primary outcome was major adverse cardiac events (MACE), which included 30-day heart failure readmission.

Out of 2,578 patients admitted with acute heart failure, 503 were associated with cocaine use. Among the latter, 404 were treated with carvedilol and 99 with no beta-blockers. At baseline, those receiving carvedilol recorded lower EF rates, heart rate, and NT-proBNP concentrations and exhibited more coronary artery disease.

Over a median follow-up of 19 months, 169 patients experienced a MACE. In a multivariate model, independent predictors of MACE among cocaine users were: history of coronary artery disease, lower EF, elevated pulmonary artery systolic pressure, higher NT-proBNP, lower education level, unemployment, and less use of standard heart failure medical therapy. Also, MACE rates among those on carvedilol did not differ from those not on carvedilol. Among those with reduced EF (≤ 40%), MACE was lower in those on carvedilol (34% vs. 58%; P = 0.02). MACE rates on carvedilol were no different in those with EF > 40%. The authors concluded that carvedilol is effective therapy for those with reduced EF and is safe in cocaine users.


Despite the “war on drugs,” cocaine use is on the rise again in the United States, especially on the East Coast. It is now the most frequently abused drug in patients with heart failure. In this report from a tertiary center in the Bronx, New York, cocaine use was associated with 20% of their acute heart failure admissions. In patients with acute coronary syndromes, there has been a concern that selective beta-blocker use would unleash unopposed alpha-adrenergic stimulation resulting in peripheral and coronary vasoconstriction and lead to poor outcomes.

However, observational studies have not clearly born this out. Current American Heart Association/American College of Cardiology guidelines for patients with NSTEMI or unstable angina state that combined beta- and alpha-blocking agents such as labetalol may be reasonable for treating sinus tachycardia or hypertension in cocaine users as long as a vasodilator is added (class IIb-c). However, little is known about using beta-blockers in patients with heart failure associated with cocaine use. Current guidelines do not address this issue. Thus, this report is of interest. Banerji et al showed that carvedilol was safe, as the overall results in those who received the drug vs. those who did not were similar. As expected, those who recorded an LVEF ≤ 40% performed better on carvedilol than those who did not take the drug. The results also are biologically plausible since carvedilol blocks alpha- and beta-adrenergic receptors. Although carvedilol use was based on prescribing information only, the lower resting heart rates in those on carvedilol supports that the drug was taken by the patients. Also, the lower NT-proBNP levels in those on carvedilol suggests that cardiac chamber wall tension was reduced. Interestingly, the type of cocaine or the frequency of use did not seem to influence the results.

This study only evaluated carvedilol because that was the only agent used at this institution. Labetalol has similar properties to carvedilol, but has a much higher dosage range (up to 2,400 mg), which will make selecting the optimal dose difficult. Also, carvedilol is lipophilic, and penetration into the brain may reduce central sympathetic output as well. In addition, carvedilol is currently less expensive than labetalol.

There are limitations to this study, starting with the selection bias and unmeasured confounders that plague all observational studies. Also, those not given beta-blockers were a small group and were sicker with more advanced disease. At this point, the safety of using carvedilol in selected patients probably is valid, but clearly more studies are warranted for this growing problem.