Digoxin is an old drug, perhaps ancient by some standards. It is a purified cardiac glycoside extracted from the purple foxglove (digitalis purpurea) plant.

Trade names for digoxin include Lanoxin, Digitek, and Lanoxicaps. It is also a generic medication. Digoxin has been used in patients with atrial fibrillation and congestive heart failure for decades and even centuries.

Spurious use of digoxin as a weight loss drug has been reported, presumably driven by its side effect of anorexia and nausea. Digitoxin, one of the older cardiac glycosides clinically used, became infamous when employed as a murder weapon in Agatha Christie’s Hercule Poirot 1938 novel Appointment With Death.

Continue Reading

William Withering, in 1785, published his treatise on the use of “foxglove tea” to treat dropsy, which was a generic term for conditions associated with edema and fluid retention. Who knows what the diagnosis of his patients actually was.

Could they have had congestive heart failure with atrial fibrillation, nephrotic syndrome, hepatic cirrhosis, or profound hypothyroidism with myxedema. Certainly some patients responded to the concoction.

Since that time the active ingredients of the tea have been characterized and used liberally in many patients with heart disease largely because of the therapeutic wasteland apparent until the past half century. More recently digoxin has been vanishing from our heart failure treatment protocols.

Comparatively little is written and published about digoxin today, arguably an indication of interest in the drug. For example, PubMed in the 1980 to 1985 window listed 1,911 citations for “digoxin” compared to only 766 in 2005 to 2010.

The drug has become generic and is no longer “marketed.” The proportion of patients on digoxin in large scale clinical trials done between 1980 and 2010 to evaluate angiotensin-converting enzyme inhibitors, beta-blockers, angiotensin receptor blocking agents, aldosterone antagonists, as well as pacemaker resynchronization and defibrillators has dropped rather dramatically.

For example, in the SOLVD study of enalapril done in the late 1980s and published in 1991 about 67% of participants were concomitantly on digoxin compared to 27% in Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF) and 22% in Systolic Heart Failure Treatment (SHIFT), which were recently completed. That is unfortunate because digoxin is effective in certain clinical situations, particularly congestive heart failure with or without atrial fibrillation.

When used appropriately and in low dose, while being mindful of drug-drug interactions, digoxin is quite safe. To effectively use this drug, the clinician, as with any agent prescribed, needs to be familiar with pharmacokinetic properties, adverse effects, interactions with other drugs, anticipated physiologic actions and responses, and most importantly, clinical indications.

Clinical trial evidence as well as opinions of experts dealing with heart failure and cardiac arrhythmia management has been distilled into several therapeutic guidelines, clearly indicating a significant niche for the drug in many treatment protocols. Clearly the agent still has value.

Evidence of this is contained in many well done, controlled, small, short-term follow-up clinical trials, as well as two larger, nonmorbidity endpoint “digoxin withdrawal” studies.

The Prospective Randomized study Of Ventricular failure and the Efficacy of Digoxin (PROVED) trial assessed the effect of digoxin withdrawal in patients with mild to moderate congestive heart failure and sinus rhythm, and the Randomized Assessment of Digoxin on Inhibitors of Angiotensin-Converting Enzyme (RADIANCE) study had the same design and patient characteristics with the exception of patients being on concomitant ACEi therapy.

In aggregate, these studies demonstrated that benefits were primarily related to improvement in symptoms, exercise capabilities, and functional capacity. They were done in the early 1980s, a time where beta-blockers were not routinely employed in this population, but there is no reason to believe their observations and conclusions are not applicable to contemporary patients.

The Digitalis Investigation Group (DIG) trial was a large 6,800 patient effort that did explore the effects of digoxin on mortality and morbidity (hospitalization in particular). The results published in 1997 revealed that after 3 years of follow up, when digoxin was added to an ACEi-based protocol for symptomatic congestive heart failure and sinus rhythm with left ventricular ejection fraction below 45%, there was no statistically significant mortality differences between treatment and control group but hospitalization for worsening heart failure and all-cause hospitalizations were significantly diminished in the digoxin group.

Interestingly, the DIG trial had a large subset of patients with left ventricular ejection fraction (LVEF) >45%, now labeled “heart failure with preserved systolic function.” Results in the “preserved” LV systolic function group paralleled those noted in the “reduced” LVEF cohort, particularly with respect to reduction in hospitalizations.

This is important because this observation and the relatively more recent Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM)-Preserved trial remain the only large scale clinical trial efforts suggesting significant benefit of a drug in this perplexing group of patients. Though digoxin had been marketed for many decades, it received ^Food and Drug Administration (FDA) approval only recently (1998) for use in mild to moderate heart failure and for the control of ventricular rate in patients with atrial fibrillation.

Differences between drugs within the class

Within the classification of “Cardiovascular Agents” used by the 65th Physicians’ Desk Reference (2011 PDR), digoxin is listed as an “inotropic agent” with no other agents included in that category. Digoxin is distinct from the alpha and beta adrenergic receptor activating drugs and phosphodiesterase inhibitors.

Digoxin was also listed as a “Miscellaneous Antiarrhythmic Agent.” It is a class V antiarrhythmic agent (along with adenosine) using the Vaughn Williams classification scheme (I: agents interfering with Na+ channels; II: antiadrenergic agents such as beta blockers; III: K+ channel interference; IV: Ca++ channel interference; V: other or unknown mechanisms).

Comparing other cardiac glycosides with digoxin is interesting more from an historic perspective, as older compounds are not readily available or clinically used today. At one time great care and attention was given to “digitalizing” patients with intravenous ouabain and digitoxin.

This is evidenced by the large amount of text given over to the subject in classic cardiology textbooks of the 1930 to 1960 era. Digitoxin was first extracted from digitalis purpurea leaves in 1875, with the first structural analysis done in 1925, and determination of glycoside characteristics in 1962.

Digitoxin has a similar structure and function to digoxin but is longer acting and eliminated via hepatic metabolism whereas digoxin is eliminated by the kidneys. Some believed that digitoxin’s contemporary therapeutic niche was in patients with renal dysfunction and an indication for digoxin, or more specifically when digoxin toxicity appeared in a setting of renal insufficiency. But the use of digitoxin today has largely been abandoned primarily driven by the shorter action of digoxin and overall ease of use of that drug.

Ouabain (g-strophanthin) is structurally similar to digoxin but not from the foxglove purpurea plant. Rather, it is found in ripe seeds of the Strophanthus gratus and the bark of Acokanthera ouabaio plant and tree.

It is a poisonous cardiac glycoside ,and the name is from the Somali word for arrow tip poison. Its mechanism of action is similar to digoxin, with digoxin replacing ouabain because its bioavailability was far greater and it was much more lipophilic with a wider therapeutic index.

Ouabain has been largely relegated to the experimental laboratory where it is used to specifically block the sodium-potassium-ATPase cell membrane “pump.” Other plants with significant cardiac glycoside concentrations include Asclepias or “milkweed” and the oleander evergreen shrub, which both can be fatal if ingested in large quantities.

Oleander-based potions, like those made from ouabain, have been used as arrow tip poison.


Digoxin can be administered as a tablet, gelatin capsule, elixir (primarily for children), or intravenous injection. As a tablet, digoxin given by mouth reaches a peak concentration in 1 to 3 hours at a level 80% of the same dose given intravenously.

Steady state equilibrium is reached in 7 to 12 days when renal function is normal. When taken after meals absorption is slowed. High bran fiber meals can decrease the total amount absorbed significantly.

In some patients colonic bacteria will reduce serum digoxin concentration by as much as 40% because of conversion in the bowel to an inactive reduction product. Sometimes, concomitant antibiotic administration with reduction in gastrointestinal flora will reduce this reaction and precipitate a significant rise in serum digoxin concentration.

This rise is related to the elimination half-life of digoxin and so the rate can be increased by renal insufficiency. One should be aware of this during concomitant digoxin and antibiotic administration in patients with decreased creatinine clearance.

After oral administration of digoxin, there is a 6- to 8-hour tissue distribution phase and then a gradual excretion, with the excretion rate determining the actual decline in serum concentration. Ultimately, tissue concentration will reflect serum concentration and relate to pharmacologic actions.

This occurs with chronic administration of the drug. Digoxin crosses the blood brain barrier, which accounts for some of the nausea and depression side effects seen. It also crosses the placenta with clinical effects (bradycardia) noted sometimes in newborns. Approximately 25% of digoxin is protein bound.

Digoxin does have some trivial hepatic metabolism (<20%) and this is not dependent on the cytochrome 450 system, nor does digoxin up or down regulate this enzyme pathway. Renal excretion is proportional to the glomerular filtration rate with about 60% of an intravenous dose excreted unchanged in the urine.

The half-life of digoxin in an individual with normal renal function is about 1.5 days, and this is increased to about 5 days in patients who are anuric. Because of the significant tissue binding of digoxin it is not removed by dialysis, plasma exchange or exchange transfusions, or during cardiopulmonary bypass. Severe drug related toxicity can be managed by the administration of an antidigoxin antibody (Digoxin Immune Fab: Digibind).

Dosing digoxin has been a contentious issue and in the past and associated with much folderol surrounding algorithms and serum drug level monitoring. The dose has been suggested to be 125 to 500 mcg daily in most individuals and is effected by patient weight, renal function, age, and concomitant drugs being taken.

Some have targeted a trough serum digoxin concentration of 0.8 to 2.0n g/ml as “therapeutic range.” In reality, today, most recommend a level closer to 1.0 ng/ml, or even less.

“Toxicity” has been said to be present when the level is >2.0 ng/ml, but this is an arbitrary definition. Many patients exhibit digoxin toxicity at levels well below 2.0 ng/ml while others have desirable effects of this drug without evidence of toxicity when levels are >2.0 ng/ml.

Nonetheless, recent evaluations of clinical trial data suggests that patients do best when the digoxin level is closer to 0.8 ng/ml than to 2.0 ng/ml. A simple and practical approach is to start oral digoxin at 125 mcg daily in just about everyone with reasonable renal function with a careful clinical watch of the patient and understanding of what concomitant drugs can increase the chances for unwanted effects.

Determination of serum digoxin levels should be done when toxicity is suspected or to check on patient compliance and questions of absorption. Routine surveillance monitoring of digoxin levels is unnecessary and titrating the dose of digoxin to the serum level is rarely done anymore except when the risk of toxicity is high. When knowing a level is desired, it should be a trough level drawn about 24 hours after the last dose.

If more rapid “digitalization” of a patient is desired, particularly for the management of atrial fibrillation with rapid ventricular response rate, the intravenous preparation can be given. Slow infusion is preferred over a bolus. Intramuscular injection should be avoided because of pain at the injection site.

Dosage calculation parallels that for digoxin tablets and the same caveats apply to intravenous use of digoxin as for the oral preparation. Intravenous administration provides about 20% more of the drug then when the same dose is given orally. The pediatric elixir provides another oral option for use of this drug when patients are unable to take tablets or capsules but have enteral access of one sort or another.

The gel tablet formulation of digoxin, Lanoxicaps, provides another alternative for giving digoxin and the seeming benefit is better absorption. Argument abounds regarding the justification of increased cost for the capsules.

Indeed, after oral dosing with the capsule 90% to 100% of the product is absorbed when compared to equivalent intravenous dosing, while the bioavailability of the tablet is in the 60% to 80% range. Perhaps this provides some advantages when using digoxin, as the enhanced product absorption might reduce variability of steady-state serum concentrations and provide more reliable pharmacokinetic information to the clinician and steady state pharmacokinetics to the patient.

Pharmacologic action

After centuries, the pharmacologic action of cardiac glycosides generally, and digoxin specifically, are now better known. Specifically, digoxin inhibits sodium-potassium ATPase, the enzyme controlling egress and ingress of cellular sodium and potassium.

Increasing the intracellular concentration of sodium triggers stimulation of sodium-calcium exchange across the cell wall, which results in an increased intracellular calcium concentration. More calcium becomes available to the sarcoplasmic reticulum and myocyte contractility is enhanced.

Equally important, though some might argue even more important, there are many effects on the autonomic nervous system. These become beneficial when managing both arrhythmias and heart failure.

They also relate to digoxin side effects and toxicity. Digoxin has a vagomimetic effect on the sinoatrial and atrioventricular node. This contributes to a decrease in heart rate and slows atrioventricular conduction velocity, the former being important in congestive heart failure, and the later in atrial fibrillation with rapid ventricular response.

Baroreceptor sensitization also occurs with the result being an increased afferent inhibitory activity resulting in reduction of sympathetic nervous and renin-angiotensin-aldosterone signaling system activity. The critical importance of neurohormonal “deactivation” in the heart failure syndrome has only recently been recognized and clarified.

With high doses of digoxin resulting in high serum concentrations, one can demonstrate an increase in central nervous system sympathetic activity and this probably drives much of what we characterize as digoxin toxicity, particularly toxicity manifest as ventricular arrhythmias. Of course the vagomimetic effects account for arrhythmia toxicity as well, with precipitation of profound bradycardia or atrioventricular nodal heart block.

This fact must be remembered when using concomitant agents, which also have an effect on heart rate and atrioventricular nodal conduction such as beta-blockers, some calcium channel blockers, and amiodarone. When using digoxin a focus on low doses and serum levels is important, as enhancing parasympathetic effects is likely more beneficial long term than increasing sympathetic activity.

Indications and contraindications

Indications, contraindications, alternatives

Regulatory labeling of digoxin notes that the drug is indicated for treatment of patients with mild to moderate heart failure. Specifically noted is that digoxin increases left ventricular ejection fraction and improves the symptoms of heart failure and the latter is largely based on evidence from the previously mentioned PROVED, RADIANCE, and DIG trials where, collectively, exercise capacity was increased and heart failure related hospitalizations diminished.

Interestingly, no distinction between heart failure due to left ventricular systolic dysfunction or heart failure with “preserved” left ventricular function is made. For patients with atrial fibrillation the indication is control of the ventricular response rate.

As alluded to, regulatory approval of digoxin for heart failure was driven primarily by the three clinical trials mentioned. Two 12-week, double-blind, placebo controlled, multicenter, randomized trials used a “digoxin” withdrawal approach.

The RADIANCE trial studied 178 patients on a background of digoxin, diuretic, and ACE inhibitor while the PROVED trial randomized 88 patients on digoxin and diuretic baseline therapy.

These trials were not done in an era where beta-blockers were frequently used in these patients. In aggregate, the two trials noted that when participants were blindly withdrawn from digoxin (drug replaced with placebo), exercise tolerance, a global symptomatic benchmark, decreased; NYHA functional classification deteriorated; and heart failure related hospitalizations and need for urgent care increased.

The DIG trial was a classically designed, large scale (6,801 patients), placebo controlled, randomized, mortality and morbidity endpoint clinical trial of patients with both systolic and “preserved” ejection fraction patients with mild to moderate symptomatic heart failure. There were many ramifications and nuances of this study.

The primary focus was on patients with LVEF <45% and in this group over two thirds were NYHA class I or II, 71% had underlying ischemic heart disease, and 44% had been on digoxin (and, thus, after randomization it turned into a digoxin “withdrawal” trial for a substantial number of participants).

Other concomitant medications included use of ACE inhibitors in 94% and diuretics in 82% of patients. Dose was adjusted for patient age, sex, lean body weight, and serum creatinine with a substudy of 1,800 subjects using a dosing algorithm and ending up with a mean serum digoxin level of 1.01 ng/ml at 1 month and 0.97 ng/ml at 1 year.

The median daily dose prescribed was 250 mcg. Risk reduction (with 95% confidence interval) for all-cause mortality or all-cause hospitalization at a median follow-up of 37 months was 0.94 (CI; 0.88-1.00) and risk reduction for heart failure related mortality or heart failure related hospitalization 0.69 (CI; 0.63-0.76).

All-cause mortality was 35% and there was no difference in the digoxin and the placebo group. Digoxin was, however, associated with a 25% reduction in heart failure hospitalization and a 6.5% reduction in hospitalization for any cause. Post hoc analyses suggested that patients with lower serum digoxin levels did better.

A rarely evaluated heart failure cohort remains those with “preserved” LVEF. In an ancillary study of DIG, 987 patients with LVEF >45% were studied in the same fashion as were patients entered into the main trial.

In this cohort, the risk of all-cause mortality or all-cause hospitalization was not reduced by digoxin, but the risk of heart failure related mortality or heart failure hospitalization was; hazard ratio (HR) and 95% CI 0.72 (0.53-0.99). Thus, digoxin has some of the most convincing data supporting its use in patients with heart failure in a setting of more normal LVEF.

Comprehensive heart failure practice guidelines have addressed use of digoxin and there appears to be consensus. The Heart Failure Society of America (HFSA) 2010 guideline noted that “digoxin is a drug that is inexpensive and can be given once daily, and it continues to have a therapeutic role in symptomatic patients with HF from reduced LVEF.”

More specifically these guidelines stated that digoxin can improve symptoms in patients with heart failure and LVEF <40% who have signs or symptoms while receiving standard therapy defined as ACE inhibitors and beta-blockers. The level of evidence supporting this was “B”.

Additionally, it was suggested that the dose of digoxin should be 125 mcg daily in the majority of patients and that the serum digoxin level should be <1.0 ng/ml (again strength of evidence listed at “B”). Also stated was that digoxin should be considered for helping control ventricular response in the setting of atrial fibrillation but that maintenance doses of >250 mcg/day should be avoided.

Despite the DIG trial data alluded to above in patients with preserved LVEF, no mention of using digoxin in this group was made in the HFSA document. The American College of Cardiology/American Heart Association 2009 heart failure guideline update is only slightly different; “digitalis (stated instead of digoxin) can be beneficial in patients with current or prior symptoms of HF and reduced LVEF to reduce hospitalizations.” Level of evidence was also said to be “B.”

Digoxin should not be given to patients with known hypersensitivity to the drug. Ventricular fibrillation and Wolff-Parkinson-White syndrome are contraindications as well. Some would argue that a patient at extremely high risk of developing an undesirable effect of the drug (an elderly, frail, patient with multiple comorbidities, on many drugs, and in renal failure or having renal insufficiency) should not be given digoxin.

Digoxin’s effects are partially mediated by positive inotropic actions, as noted. Some believe this is actually trivial when compared to the other actions of the drug.

Nonetheless, there are no other oral available agents that mimic the inotropic action of digoxin. In patients with heart failure, though not strictly alternatives, beta-blockers, ACE inhibitors, angiotensin receptor blockers, and aldosterone antagonists are essential agents to consider as adjunctive therapies to digoxin.

Indeed, these are core baseline drugs that are necessary in heart failure. Also, there are many alternatives that can achieve the antiarrhythmic effects of digoxin. Some are much more effective.

Undesirable effects

Undesirable effects of digoxin can be numerous and, arguably, the drug has a narrow therapeutic range. Despite that fact, when rationally prescribed and patients are watched closely, digoxin is very well tolerated, simple to take, inexpensive, and effective.

One must use the drug with caution when sinus node disease or AV nodal conduction abnormalities present because of the risk of excessive bradycardia and advanced heart block. In the presence of Wolff-Parkinson-White syndrome, after intravenous administration of digoxin, because of the slowing of AV conduction that occurs, precipitous acceleration of heart rate can develop with supraventricular arrhythmias unless other agents have been used to block the accessory conduction pathway.

There is some evidence to suggest that digoxin toxicity is more likely in patients with restrictive cardiomyopathy, constrictive pericarditis, amyloid heart disease, hypertrophic obstructive cardiomyopathy, and acute cor pulmonale. Undesirable effects are more likely in patients with renal insufficiency, and settings of hypokalemia, hypomagnesemia, and hypercalcemia.

Hypocalcemia may make digoxin ineffective. Atrial fibrillation associated with hyperthyroidism is difficult to control and sometimes clinicians are tempted to push the dose of digoxin to gain control over the ventricular response rate. There is an increased risk of digoxin toxicity when this tactic is used.

Side effects that the clinicians using digoxin need to be aware of are those primarily centered on the gastrointestinal tract and central nervous system. The drug can cause nausea, vomiting and diarrhea, even when serum levels are <2.0 ng/ml.

One of the most feared complications of the drug is necrotic bowel syndrome due to intestinal ischemia. In our experience, this is seen more often in older, frail patients with systemic atherosclerosis receiving the drug intravenously during digitalization for management of atrial fibrillation or significant congestive heart failure during a hospitalization.

Emboli in a setting of atrial fibrillation is always a concern but rapid digitalization with higher serum levels does cause vascular resist ance to rise in the mesenteric circulation, setting the stage for bowel hypoperfusion. The savvy clinician will know that and be able to take quick and appropriate action if abdominal pain develops in such a setting.

From the central nervous system perspective, digoxin can cause visual disturbance classically described as blurred or yellowed vision (some have speculated that Vincent van Gogh was cardiac glycoside toxic), headache, dizziness, apathy, and depression. Delirium and hallucination has even been reported as side effects of digoxin.

Other side effects include gynecomastia, thrombocytopenia, and a maculopapular rash. Fortunately these side effects are infrequent but do occur more often when the digoxin levels rise, particularly over 2.0 ng/ml.

Many of these side effects improve simply by stopping or reducing the dose of digoxin. The speed with which they dissipate relates to renal function and serum concentration of the drug.

Life-threatening complications of digoxin can occur with massive overdoses or extremely high levels, particularly in the setting of acute renal failure. Ventricular fibrillation or tachycardia can be dramatic, producing bizarre looking QRS complexes and electrocardiogram.

Hyperkalemia can be seen and is a result of the drug effects on the sodium-potassium cell wall pump described above. Patients with massive digoxin ingestion (suicide attempt as an example) should receive large doses of activated charcoal to prevent absorption.

When hemodynamic compromise is evident, usually in the setting of problematic ventricular or supraventricular arrhythmias, and remembering that digoxin is not dialyzable, administration of Digibindshould be considered. Digoxin immune Fab binds to digoxin, making the molecule incapable of activating receptor sites on the myocyte cell wall.

The drug-immune complex accumulates in the blood stream and is excreted by the kidney. Watching the rapid resolution of the malignant arrhythmias induced by digoxin toxicity during Digibind infusion can be an amazing experience. It also is a way to quickly confirm that a problematic arrhythmia is caused by digoxin.

Drug interactions with digoxin are important to understand. Nonpotassium sparring diuretics can cause hypokalemia and hypomagnesemia, which with digoxin increases arrhythmia risk.

Intravenous calcium administration can also increase arrhythmia risk. Reducing digoxin clearance and decreasing the volume of distribution, which will increase serum digoxin levels, is amiodarone, quinidine, verapamil, propafenone, itraconazole, alprazolam, and spironolactone.

Erythromycin, clarithromycin, and tetracycline increase digoxin absorption by inactivating intestinal bacterial metabolism. Antacids, bran, cholestyramine, kaolin-pectin, and metaclopromide decrease digoxin absorbtion.

Rifampin increases renal excretion of digoxin. Sympathomimetics increase myocyte automaticity and increase the risk of arrhythmias with digoxin.

Beta adrenergic blockers, nondihydropyridine calcium channel blockers, flecainide, disopyramide, and bepridil decrease sinoatrial or AV nodal conduction and with digoxin increase the risk of sinoatrial and AV block. ACE inhibitors, angiotensin receptor blockers, and nonsteroidal antiinflammatory agents can decrease renal function, which could diminish renal clearance and increase digoxin levels.

What's the Evidence?

Withering, W, Willius, FA, Keys, TE. ” An account of the foxglove and some of its medical uses with practical remarks on dropsy and other diseases”. Classics of Cardiology. vol. 1. 1983. pp. 231-52. (This remarkable "case series" is fun to read. It was written by a master clinician and botanist in the late 18th century. It is a "case series" filled with anecdote and insight. Required reading for anyone interested in cardiovascular therapeutics and digoxin in particular.)

Hothi, SS, Chinnappa, S, Tan, LB. ” 200+ years of a misunderstood drug for treating chronic heart failure: Digoxin, why and how should we continue using it?”. Int J Cardiol. 2012 May 17. (A very nice contribution lamenting the fact that digoxin use has diminished dramatically over the last decade, whereas publications about this drug were common a decade ago, few studies of the agent, clinical trials, or use descriptions are currently available. This is the most recent one, with few appearing in the last 5 to 7 years, and thoughtful.)

Gheorghiade, M, van Veldhuisen, DJ, Colucci, WS. “Contemporary use of digoxin in the management of cardiovascular disorders”. Circulation. vol. 113. 2006. pp. 2556-64. (An older but excellent review of digoxin. Nothing has changed since 2006. Written by master cardiovascular clinicians.)

Young, JB. “Wither withering's legacy. Digoxin's role in our contemporary pharmacopeia for heart failure”. J Am Coll Cardiol. vol. 46. 2005. pp. 405-7. (An editorial written in response to a reanalysis of the DIG trial database, which demonstrated that digoxin was safe and quite effective at lower serum concentrations.)

Packer, M. “End of the oldest controversy in medicine – are we ready to conclude the debate on digitalis?”. N Engl J Med. vol. 336. 997. pp. 575-6. (A very well written and curmudgeonly editorial in response to the DIG trial conclusion and publication.)

“The effects of digoxin on mortality and morbidity in patients with heart failure”. N Engl J Med. vol. 336. 1997. pp. 525-33. (Classic, large scale, multicenter randomized clinical trial demonstrating the important contribution digoxin makes to heart failure patients in normal sinus rhythm. The database has been plumbed repeatedly over the last decade and, generally, no matter the technique of analysis or question asked, digoxin comes out in a positive light.)

Uretsky, BF, Young, JB, Shahidi, FE, Yellen, LG, Harrison, MC, Jolly, MK. “Randomized study assessing the effect of digoxin withdrawal in patients with mild to moderate chronic congestive heart failure: results of the PROVED trial”. J Am Coll Cardiol. vol. 22. 1993. pp. 955-62. (First of the digoxin clinical trial troika that led to FDA approval of digoxin for heart failure [though it had been on the "market" for decades]. This study was not done with a background of ACE inhibitor therapy, an emerging strategy that is now essential in heart failure patients.)

Packer, M, Gheorghiade, M, Young, JB. ” Withdrawal of digoxin from patients with chronic heart failure treated with ACE inhibitors”. N Engl J Med. vol. 329. 993. pp. 1-7. (Companion study to PROVED trial done with patients receiving ACE inhibitors as background therapy. In aggregate the PROVED and RADIANCE trials demonstrated important functional improvement in patients with heart failure who were taking digoxin.)

Lindenfeld, J, Albert, NM, Boehmer, JP. “Executive summary: HFSA 2010 comprehensive heart failure practice guideline”. J Card Fail. vol. 16. 2010. pp. 475-539. (Consensus statement that contains suggestions about digoxin's use. Notably absent is a recommendation for using digoxin in patients with preserved LVEF and heart failure.)

Jessup, M, Abraham, WT, Casey, DE. ” Writing on behalf of the 2005 guideline update for the diagnosis and management of chronic heart failure in the adult writing committee. 2009 Focused Update”. Circulation. vol. 119. 2009. pp. 1977-2016. (The opinions of an expert committee put together by the American College of Cardiology and the American Heart Association. It is not much different than the HFSA guideline with respect to digoxin use.)