Annual treatment visits with digoxin (A) and digoxin for heart failure (B) in the United States, 1997 to 2012. Source: IMS National Disease and Therapeutic Index, January 1997 to December 2012 (IMS Health Incorporated).
Goldberger ZD, Alexander GC. Digitalis Use in Contemporary Clinical PracticeRefitting the Foxglove. JAMA Intern Med. 2014;174(1):151-154. doi:10.1001/jamainternmed.2013.10432
Over 200 years after William Withering wrote the classic monograph, An Account of the Foxglove and Some of Its Medicinal Uses,1 the indications for and optimal dosing of digitalis glycosides (primarily prescribed as digoxin) continue to be debated. Convincing evidence regarding the purported benefits of digoxin was unavailable until the Digitalis Investigation Group (DIG) trial, published in 1997.2 This study was approved by the institutional review board at each participating center.
The DIG trial was a large-scale, international, prospective trial that randomized 6800 ambulatory adult patients with systolic heart failure (HF) to digoxin treatment or placebo. Enrolled patients were receiving concomitant HF therapy with diuretics (81% of patents) and angiotensin-converting enzyme (ACE) inhibitors (94% of patients). Notably, patients with atrial fibrillation were excluded from the trial. Treatment with digoxin had no significant reduction in all-cause mortality, although it led to a 28% relative risk reduction for hospital admission for worsening HF within a mean follow-up of approximately 3 years (Box).
Ambulatory patients in sinus rhythm
LVEF of 45% or lowera
History of HFb
Eligible patients allowed to be taking digoxin
Age younger than 21 y
Atrial fibrillation or atrial flutter (with or without pacemaker)
MI, cardiac surgery, or PTCA within 4 wk
Unstable or refractory angina within 1 mo
Second- or third-degree AV block (without a pacemaker)
Current treatment with IV inotropes
Hypokalemia/hyperkalemia (range, 3.2-5.5 mmol/L)
Need for cardiac surgeryd or PTCA in near futurec
Listed for cardiac transplant
SSS without pacemaker
Recognizable noncardiac causes of HF
Renal insufficiency (creatinine level, >3.0 mg/dL)
Any noncardiac disease with life expectancy less than 3 years
Baseline LVEF not available
Unlikely to comply with study protocol
Mortality from CV causes
Mortality from worsening HF
Hospitalization for worsening HF
Hospitalization for other causes (including digoxin toxicity)
Results (digoxin, n=3397; placebo, n=3403)
34.8 vs 35.1%: Mortality with digoxin vs placebo (RR, 0.99; 95% CI 0.91-1.07; P = .80)
29.9 vs 29.5%: CV mortality with digoxin vs placebo (RR, 1.01; 95% CI, 0.93-1.10; P = .78)
11.6 vs 13.2%: Mortality from worsening HF with digoxin vs placebo (RR, 1.14; 95% CI, 0.77-1.01; P = .06)
49.9 vs 54.4%: CV hospitalization with digoxin vs placebo (RR, 0.87; 95% CI, 0.81-0.93; P < .001)
26.8 vs 34.7%: Hospitalization for worsening HF with digoxin vs placebo (RR, 0.72; 95% CI, 0.66-0.79; P < .001)
2.0 vs 0.9%: Hospitalization for suspected digoxin toxicity with digoxin vs placebo (RR, 2.17; 95% CI, 1.42-3.32; P < .001)e
ESC (class IIb, level of evidence B): May be considered to reduce the risk of HF hospitalization in patients in sinus rhythm with an LVEF of 45% or lower who are unable to tolerate a β-blocker. Patients should also receive an ACE inhibitor (or ARB) and an MRA (or ARB). May be considered to reduce the risk of HF hospitalization in patients with an EF of 45% or lower and persisting symptoms (NYHA class II-IV) despite treatment with a β-blocker, ACE inhibitor (or ARB), and an MRA (or ARB).4
ACC/AHA (class IIa, level of evidence B): Digoxin can be beneficial in patients with HF with reduced LVEF, unless contraindicated, to decrease hospitalizations for HF. Clinicians may consider adding digoxin in patients with persistent symptoms of HF with reduced LVEF during GDMT. Digoxin may also be added to the initial regimen in patients with severe symptoms who have not yet responded symptomatically during GDMT. Alternatively, treatment with digoxin may be delayed until the patient’s response to GDMT has been defined and may be used only in patients who remain symptomatic despite therapy with neurohormonal antagonists. If a patient is taking digoxin but not an ACE inhibitor or a β-blocker, treatment with digoxin should not be withdrawn, but appropriate therapy with the neurohormonal antagonists should be instituted.5
Heart Failure Society of America (NYHA class II-III, level of evidence B; NYHA class IV, level of evidence C). Digoxin may be considered to improve symptoms in patients with LVEF ≤40%) who have signs or symptoms of HF while receiving standard therapy, including ACE inhibitors and β-blockers.6
Abbreviations: ACC/AHA, American College of Cardiology/American Heart Association; ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; AV, atrioventricular; CI, confidence interval; CV, cardiovascular; EF, ejection fraction; ESC, European Society of Cardiology; GDMT, guideline-directed medical therapy; HF, heart failure; IV, intravenous; LVEF, left ventricular ejection fraction; MI, myocardial infarction; MRA, mineralocorticoid receptor antagonist; NYHA, New York Heart Association; PTCA, percutaneous transluminal coronary angiography; RR, relative risk; SSS, sick sinus syndrome.SI conversion factors: To convert creatinine to micromoles per liter, multiply by 88.4; to convert digoxin to nanomoles per liter, multiply by 1.281.
Ancillary study of 988 patients with an LVEF greater than 45% performed, with combined primary outcome of death of hospitalization due to worsening HF, with results consistent with main trial.
Current or past symptoms (limitation of activity, fatigue, dyspnea, orthopnea) or signs (edema, elevated jugular venous pressure, rales, S3 gallop), or radiologic evidence of pulmonary congestion.
Eligible after surgery/revascularization.
Severe valvular disease, planned coronary artery bypass surgery.
Mean serum digoxin concentration of 0.80 ng/mL at the 12-month visit, with 88.3% of digoxin group within the range of 0.5 to 2.0 ng/mL.
Little is known about the patterns of digoxin use for the treatment of HF since the publication of the DIG trial, and the use of digoxin in HF therapy remains controversial. Concern about digitalis toxicity, along with the advent of other agents shown to confer a mortality benefit, such as β-blockers, ACE inhibitors and aldosterone antagonists, may have contributed to waning digoxin use.3 We hypothesized that digoxin use for systolic HF has decreased during the past 15 years, despite clinical guidelines supporting its use.4- 6
We used the IMS Health National Disease and Therapeutic Index, an ongoing audit of office-based US physicians that provides nationally representative information regarding disease patterns and treatment. This project was not human subjects research and therefore did not require institutional review board approval. Our primary unit of analysis was a treatment visit, defined as an office visit where digoxin was used for a specific clinical indication. We quantified digoxin use from 1997 through 2012 among all subjects as well as among patient subpopulations. All statistical analyses were performed using SAS 9.2 (SAS Institute Inc). This project was not human subjects research and, as such, did not require institutional review board approval.
Digoxin treatment visits declined by 86%, from 12.9 million visits in 1997 to 1.87 million visits in 2012 (Figure, A). Declines were greater between 1997 and 2001 (12.9 to 6.8 million visits, averaging a 10% decrease per year over 5 years [P < .001 for trend]) than subsequent years (6.9 to 1.9 million visits, averaging a 6% decrease per year over 11 years [P < .001 for trend]). For patients with HF, digoxin treatment visits declined by 91% overall (Figure, B), averaging an 11.2% decrease per year between 1997 and 2001 and a 7% decrease per year from 2002 through 2012. There were no statistically significant differences in these trends based on patient sex or physician specialty. Of note, 23% of treatment visits for digoxin use over the course of the study period were due to HF, compared with 20% for atrial fibrillation or flutter.
There has been a marked reduction in ambulatory digoxin use in the United States since 1997, with the largest declines in use observed from 1997 through 2001, and especially for patients with HF. Our study is limited by the lack of data prior to 1997, and there are a number of potential causes of the declines that we have illustrated. An increasing number of evidence-based therapies for HF, the perceived toxic effects and challenges of digoxin dosing, and the negative results of the DIG trial with respect to its primary end point of all-cause mortality, may all have contributed to reductions in digoxin use. However, the DIG trial demonstrated a significant decrease in hospital admissions for HF in ambulatory patients receiving digoxin therapy. These changes may be particularly salient to contemporary clinical practice in clinical and policy efforts to reduce inpatient health care utilization for HF. However, whether digoxin use will have any direct effect on hospital readmission for HF remains unclear, given that readmissions were not directly measured. As such, this hypothesis merits caution. Unfortunately, new prospective randomized trials of digoxin are unlikely, leaving its fate as an integral part of HF therapy in contemporary practice uncertain.
Corresponding Author: Zachary D. Goldberger, MD, MS, Division of Cardiology, University of Washington School of Medicine, Harborview Medical Center, 325 Ninth Ave, PO Box 359748, Seattle, WA 98104 (email@example.com).
Published Online: November 11, 2013. doi:10.1001/jamainternmed.2013.10432.
Author Contributions: Both authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Goldberger.
Acquisition of data: Both authors.
Analysis and interpretation of data: Both authors.
Drafting of the manuscript: Goldberger.
Critical revision of the manuscript for important intellectual content: Both authors.
Statistical analysis: Goldberger.
Administrative, technical, or material support: Alexander.
Conflict of Interest Disclosures: Dr Alexander reported being an ad hoc member of the Food and Drug Administration’s Drug Safety and Risk Management Advisory Committee, serving as a paid consultant to IMS Health, and serving on an IMS Health scientific advisory board. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies.
Funding/Support: Dr Alexander is supported by grant RO1 HS0189960 from the Agency for Healthcare Research and Quality.
Role of the Sponsor: The funding sources had no role in the design and conduct of the study; analysis or interpretation of the data; and preparation or final approval of the manuscript prior to publication.
Disclaimer: The statements, findings, conclusions, views, and opinions contained and expressed in this article are based in part on data obtained under license from the following IMS Health Incorporated information service—National Disease and Therapeutic Index (January 1997–December 2012)—and are not necessarily those of IMS Health Incorporated or any of its affiliated or subsidiary entities.
Additional Contributions: Matthew Daubresse, MHS, and Hsien-Yen Chang, PhD, provided research assistance.