By Van Selby, MD

Assistant Professor of Medicine, University of California, San Francisco Cardiology Division, Advanced Heart Failure Section

Dr. Selby reports he is a consultant for Alnylam Pharmaceuticals and Akcea Therapeutics.

SYNOPSIS: In patients with dilated cardiomyopathy and recovered ejection fraction, discontinuation of heart failure medical therapies was associated with a 44% risk of relapse within six months.

SOURCE: Halliday BP, Wassall R, Lota AS, et al. Withdrawal of pharmacological treatment for heart failure in patients with recovered dilated cardiomyopathy (TRED-HF): An open-label, pilot, randomized trial. Lancet 2019;393:61-73.

With advances in medical therapy for heart failure with reduced ejection fraction (HFrEF), many patients with dilated cardiomyopathy will obtain normalization of left ventricular ejection fraction (LVEF) and volumes, with resolution of heart failure symptoms. Patients and providers are left to decide whether to continue medical therapy once a patient appears to be “cured” of heart failure. No prospective, randomized trial has been conducted to evaluate medication withdrawal in patients who are thought to have recovered dilated cardiomyopathy.

Halliday et al conducted an open-label, randomized trial that included 101 patients with prior dilated cardiomyopathy who demonstrated evidence of recovery. This recovery was defined as an LVEF that had improved from < 40% to ≥ 50% with normalization of left ventricular volume and N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentration < 250 ng/L with no signs or symptoms of heart failure. Patients were randomized to continuation of heart failure medical therapy or a phased withdrawal of medical therapy (starting with diuretics, then mineralocorticoid receptor antagonists, beta-blockers, and finally ACE inhibitors or ARBs) over 16 weeks. The primary endpoint was relapse of dilated cardiomyopathy, defined as any of the following: a reduction in LVEF of at least 10% to a value < 50%, an increase in LV volume by > 10%, a two-fold rise in NT-proBNP to a concentration > 400 ng/L, or clinical evidence of heart failure.

During six months of follow-up, 44% of patients randomized to withdrawal met criteria for relapse, compared to none of the patients who continued medical therapy (P = 0.0001). At the end of the six-month study period, patients in the control arm crossed over and discontinued medical therapy, with a similar (36%) rate of relapse. There were no deaths in either group. All patients with relapse were asymptomatic at the time relapse was identified.

The authors concluded that many patients with recovered dilated cardiomyopathy are at a high risk for relapse following discontinuation of medical therapy. Therefore, drug withdrawal usually should not be attempted in these patients.


As any provider who treats dilated cardiomyopathy can attest, patients often are eager to discontinue medical therapy as soon as their heart failure has “recovered,” often determined by improvement in LVEF. The authors have addressed an important clinical question and demonstrated a high risk of cardiomyopathy recurrence shortly after withdrawal of medical therapy. This provides clinicians with valuable data to support a recommendation to continue medical therapy after cardiac function appears to have normalized.

One of the primary messages from this study is that dilated cardiomyopathy often has not truly “recovered.” As the study authors suggested, it may be more appropriate to describe these patients as “in remission” to emphasize the fact that risk of recurrence persists after the LVEF and LV volumes have normalized and requires ongoing maintenance therapy to prevent relapse. There may be some patients who truly recover. However, our current tools for assessment (echocardiography, BNP levels, clinical exam) were unable to identify these patients in the Halliday et al study. Perhaps newer techniques such as cardiac MRI, strain imaging, and novel biomarkers will facilitate identification of patients with ongoing, subtler cardiac dysfunction or hormonal activation that was not identified using more standard techniques.

When relapse occurred, it generally happened quickly (within eight weeks of drug discontinuation in more than half of those who did relapse). All patients with relapse were immediately restarted on medical therapy. This helps explain why there were no deaths, unplanned cardiovascular hospitalizations, or other major cardiac events during the study period despite the high rate of relapse. This finding emphasizes the importance of early re-evaluation any time a patient with recovered cardiomyopathy discontinues medical therapy for any reason, with prompt re-initiation of heart failure therapies if cardiac function has deteriorated.

This was a small, open-label study with several limitations. Considering the small size, the authors could not explore predictors of successful drug withdrawal. For example, certain etiologies of dilated cardiomyopathy may tolerate discontinuation of medical therapy better than others when the underlying cause of the cardiomyopathy has been removed. It is also possible that withdrawal of some, although not all, medical therapy may be safe in some patients. Hopefully, larger studies will help identify a subset of patients in whom dilated cardiomyopathy has truly recovered and medical therapy can be discontinued safely.

Despite these limitations, patients with dilated cardiomyopathy and recovered ejection fraction should stay on medical therapy indefinitely at this time. Patients often are disappointed to hear this news, but sharing these data with them will help demonstrate the high risk associated with discontinuation.