Current issues of ACP Journal Club are published in Annals of Internal Medicine


Withdrawal of digoxin after treatment of chronic heart failure led to clinical deterioration and worsening heart failure

ACP J Club. 1994 Jan-Feb;120:1. doi:10.7326/ACPJC-1994-120-1-001

Related Content in the Archives
Digoxin reduced hospitalizations in patients with heart failure and normal sinus rhythm

Source Citation

Packer M, Gheorghiade M, Young JB, et al. Withdrawal of digoxin from patients with chronic heart failure treated with angiotensin-converting-enzyme inhibitors. N Engl J Med. 1993 Jul 1;329:1-7.



To compare clinically stable patients with chronic heart failure (HF) treated with angiotensin-converting-enzyme (ACE) inhibitors who discontinued digoxin with patients who continued taking digoxin.


Randomized, double-blind, placebo-controlled trial.


43 U.S. and Canadian clinics.


178 adults (136 men, mean age 60 y) with HF (dyspnea and fatigue on exertion in association with left ventricular fraction ≤ 35% and end-diastolic dimension of ≥ 60 mm), reduced exercise capacity, exercise duration of 2 to 14 minutes, and normal sinus rhythm. Exclusion criteria were uncorrected primary valvular disease; myocarditis; cardiomyopathy; hypertension; angina-limited exercise; lung disease; claudication; angina requiring continuous treatment; myocardial infarction within 3 months; stroke within 1 year; or severe primary pulmonary, renal, or hepatic disease.


During an 8-week run-in period, digoxin was adjusted to achieve a serum level of 0.9 to 2.0 ng/mL; diuretics, to achieve optimal fluid balance; and captopril, to ≥ 25 mg/d, or enalapril, to ≥ 5 mg/d. Patients were assigned to continuing digoxin therapy (n = 85) or placebo (n = 93) for 12 weeks. Patients withdrew because of worsening HF or adverse effects.

Main outcome measures

Worsening HF, time to withdrawal, and change in exercise tolerance. Secondary objectives were quality of life, symptoms, and cardiac function.

Main results

Patients taking placebo had more incidents of worsening HF (23 vs 4 patients, relative risk 5.9, 95% CI 2.1 to 17.2), decreased exercise tolerance (mean difference 42 s, P = 0.006), more deterioration in functional status (27% vs 10%, P < 0.04), more self-reported deterioration (31% vs 9%, P = 0.007), and less improvement (33% vs 47%) and more deterioration in quality of life (48% vs 41%) (P = 0.04 for both). The groups did not differ for overall rates of adverse effects, but those receiving placebo had more adverse effects requiring study withdrawal (37% vs 14%, P < 0.001).


Patients with chronic heart failure and normal sinus rhythm who were taking angiotensin-converting-enzyme inhibitors had clinical deterioration and worsening heart failure after discontinuing digoxin.

Source of funding: Burroughs Wellcome.

For article reprint: Dr. M. Packer, Division of Circulatory Physiology and Center for Heart Failure Research, Columbia University, College of Physicians and Surgeons, 630 West 168th Street, New York, NY 10032, USA FAX 212-305-7439.


This innovative study by Packer and colleagues adds to clinical knowledge about digoxin. Clinicians must decide whether to start digoxin in patients with HF or to continue digoxin in those receiving ACE inhibitors. Until data from the large ongoing clinical trial are available, this study provides evidence that digoxin should not be discontinued in patients who have stabilized on a regimen of digoxin, diuretics, and ACE inhibitors. The improvements in functional status, quality of life, and left ventricular function provide a convincing picture. Several key nuances, however, must be mentioned.

Only patients with substantial systolic dysfunction and ventricular dilation were included, because previous studies have shown they benefit the most from digoxin (1). Patients with HF caused by the inability of the left ventricle to relax during diastole (diastolic dysfunction) may not benefit as much (2). Therefore, left ventricular function should be measured in patients with symptomatic HF using echocardiography, scintigraphy, or angiography because of possible clinical misdiagnosis, especially diastolic dysfunction. This study also provides supportive but not direct evidence that patients with systolic dysfunction, ventricular dilation, and symptoms of HF could, initially, be given digoxin.

Digoxin doses were higher than those commonly used (0.38 vs. 0.125 to 0.25 mg/d). However, digoxin blood levels were 0.9 to 2.0 ng/dL, a reasonable level to strive for in patients with HF. For higher digoxin levels, the trade-off between clinical usefulness and increased adverse effects and mortality (3) needs to be studied in a large-scale mortality trial. Until then, studies such as this provide evidence to continue digoxin therapy with humility.

Robert M. Califf, MD
Duke UniversityDurham, North Carolina, USA


1. Curfman GD. Inotropic therapy for heart failure—an unfulfilled promise. N Engl J Med. 1991;325:1509-10.

2. Grossman W. Diastolic dysfunction in congestive heart failure. N Engl J Med. 1991;325:1557-64.

3. Yusuf S, Garg R, Held R, Gorlin R. Need for a large randomized trial to evaluate the effects of digitalis on morbidity and mortality in congestive heart failure. Am J Cardiol. 1992;69:64G-70G.