|
INTRODUCTION
Digitalis preparations have been a common remedy in the treatment of heart disease for centuries. Oral digoxin first became available in the twentieth century, and a large amount of data from clinical trials has demonstrated it to be both safe and effective in the treatment of symptomatic heart failure (HF) with or without atrial fibrillation. Due to its wide availability and lack of patent protection, it did not have extensive support from the pharmaceutical industry, but modern clinical trial data led to its approval by the U.S. Food and Drug Administration in 1999 for use in chronic HF, and recommendations for its use by the American College of Cardiology and American Heart Association (ACC/AHA) as well as the Heart Failure Society of America (HFSA) followed soon after.1,2 However, the role of digoxin therapy has recently been challenged after its use was associated with increased mortality in women.3 In addition, its worth has not been studied in the presence of modern therapy with b-blockers and aldosterone antagonists.Accordingly, the new ACC/AHA guidelines no longer recommend digoxin as routine therapy for patients with chronic HF and systolic dysfunction who are in sinus rhythm.4
BACKGROUND
The beneficial effects of digitalis preparations have been recognized for centuries, however, they were not formally introduced to the allopathic community until 1785 when first described by Sir William Withering, an English botanist and physician, in his textbook describing the medical uses of foxglove.5 Withering described the ability of digitalis to cause diuresis and slow the heart rate of patients with irregular pulse. Beginning in the twentieth century, many studies in animals and humans demonstrated positive inotropic properties of digitalis in normal as well as failing myocardium.
In the late 1970s, the use of digoxin was challenged when several nonrandomized studies in patients with HF in normal sinus rhythm (many of which did not assess left ventricular [LV] function) failed to show clinical benefit. In addition, there was a high incidence of digoxin intoxication that was associated with a mortality as high as 40%.6 These findings led to a decreased emphasis on its use, and newer therapies that included potent diuretics, vasodilators, and new inotropic agents were developed that provided clinicians with important alternative therapies in a growing population of HF patients.
In the 1990s, interest was renewed when (1) newer inotropic agents were found to worsen survival, (2) randomized studies demonstrated clinical benefits of digoxin in combination with diuretics and ACE inhibitors, and (3) lower incidences of digoxin toxicity were demonstrated due to increased recognition of drug interactions, lower dosing, and the monitoring of serum digoxin concentration (SDC). The safety and clinical benefits of digoxin gained widespread acceptance after the publication of the Digitalis Investigation Group (DIG) trial in 1996, when digoxin was shown to significantly decrease hospitalizations and improve symptoms in patients with congestive heart failure (CHF). These results led to strong recommendations for its use by both the ACC/AHA in 2001 and the HFSA in 1999 (Table 11-1).
In the last decade, several landmark studies demonstrating significant mortality benefits of b-blockers and aldosterone antagonists in patients with chronic heart failure have been published. The interest in digoxin faded when these new treatments became widely available. However, background digoxin therapy ranged from 51% to 90% in these clinical trials, leading to speculation about the usefulness of these drugs in the absence of digoxin therapy.7–10 Digoxin was further challenged by a post-hoc analysis of the DIG trial that reported an increased mortality in women,a finding that has been recently disputed.3
Mechanisms of Action
For decades, physicians have debated the exact mechanism by which digitalis increases cardiac performance. Basic science studies have demonstrated a clear affinity of the digitalis molecule for the potassium (K+) receptor of the sodium-potassium adenosine triphosphatase (ATPase). It is through this action that digitalis inhibits the enzyme, resulting in increased levels of intracellular Na. This results in increased transmembrane sodium-calcium (Na-Ca) exchange, increasing intracellular Ca levels, and improving myocardial contractility.11 This mechanism is also thought to account (at least in part) for the neurohormonal effects of digitalis by increasing baroreceptor sensitivity.12
Neuroendocrine Effects
HF causes neurohormonal activation, many phases of which are countered by digitalis and may explain its long-term beneficial effect in this population. These include the following: (1) Baroreceptor function: In patients with CHF, the failure of the carotid sinus to respond properly may lead to stimulation of the sympathetic nervous system, which will increase the production of plasma renin and vasopressin. In low-output HF models, this decrease in baroreceptor function improves with digoxin administration.12 (2) Vagomimetic effect: At therapeutic doses, digitalis increases vagal tone, resulting in decreased sinoatrial (SA) and atrioventricular (AV) conduction.13 (3) Direct sympathoinhibitory effect: It is mediated by direct inhibition of sympathetic nerve discharge.14 This effect is, if independent of the increase in cardiac performance, produced by digoxin, and is not seen in the administration of other medications that increase cardiac output (e.g., dobutamine).
(4) Effect on circulating neurohormones: Therapeutic doses of digoxin decrease plasma renin activity and circulating norepinephrine levels.15 In the Dutch Ibopamine Multicenter Trial (DIMT), digoxin therapy was associated with a decreased concentration of plasma norepinephrine over a 6-month period. (5) Antifibrotic effects: Aldosterone stimulation of the sodium pump may lead to perivascular fibrosis, which is inhibited experimentally by digoxin.16
Electrophysiological Effects
Administration of nontoxic doses of digoxin slows sinus rate by its parasympathomimetic action. This effect prolongs the refractory period of the AV node. Toxic doses predispose atrial fibers to automatic impulse initiation that does not depend on the autonomic nervous system, high-grade AV block that is mediated by cholinergic mechanisms, and an increase in the rate of spontaneous diastolic depolarization leading to the occurrence of rapid spontaneous rhythms of Purkinje fibers. Although it is clear that digitalis intoxication may produce lethal ventricular arrhythmias, therapeutic doses of digoxin do not appear to increase arrhythmias in the absence of ischemia.17
Hemodynamic Effects
Digitalis administration does not alter cardiac output in normal subjects, although it does cause significant increase in contractility. This lack of effect on cardiac output is likely due to an increase in systemic vascular resistance produced by digitalis that prevents the increase in contractility from translating into increased forward flow. In patients with reduced systolic function and abnormal central hemodynamics who are in sinus rhythm, digoxin improves LV performance and reduces pulmonary capillary wedge pressure while increasing cardiac output both at rest and during exercise.18 These beneficial hemodynamic effects are potentiated in the presence of ACE inhibitors and other afterload-reducing agents. In HF, when hemodynamics are normalized first with diuretics and vasodilators, no further improvement in wedge pressure or cardiac output is achieved after acute administration of digoxin.19 The improvement in hemodynamics persists during chronic therapy due to lack of downregulation of the Na-K-ATPase sites (putative digoxin receptor).
METABOLISM
Digoxin has excellent oral bioavailability, with approximately 80% of the dose being absorbed within 3 hours after ingestion from the distal small bowel and colon. It can be partially inactivated by colonic bacteria, and therefore antibiotic use that depletes enteric flora may increase the amount of active drug that enters the circulation. More than 80% of the active drug is excreted unchanged in the urine. The combination of limited metabolism and relatively large volume of distribution results in a relatively prolonged half-life of 36–48 hours. Steady state serum concentrations therefore generally occur within 7 days after initiation of oral therapy. Digitalis is not removed by dialysis or exchange transfusion.
DRUG INTERACTIONS
Of commonly prescribed medications used to treat cardiovascular illness, it is likely that only warfarin surpasses digoxin in concern regarding dosing and drug interactions (Table 11-2). In particular, many common medications used to treat cardiovascular disease complicate digoxin dosing. Propafenone and verapamil cause decreased renal reabsorption and therefore increase SDC.20,21 Quinidine therapy decreases nonrenal clearance of digoxin.22 Amiodarone, spironolactone, and flecainide all have been shown to increase SDC by unknown mechanisms.23–25 Non-potassiumsparing diuretics could be a major contributing factor to digoxin toxicity by causing hypokalemia.
ELECTROLYTES
Electrolyte disorders commonly accompany and/or potentiate the toxic effects of digitalis. Serum potassium levels should be monitored periodically, as most of the effects of digitalis are through its interaction with the K+ site on the Na-K-ATPase enzyme. Decreased serum K+ levels result in the potentiation of the effects of digitalis, potentially leading to ventricular tachycardia or ventricular fibrillation. Similar effects have been observed with hypomagnesemia. In the presence of hyperkalemia, the autonomic effects of digitalis predominate, resulting in decreased AV nodal and SA conduction, leading to heart block or significant bradycardia. Digitalis toxicity itself can also worsen hyperkalemia and lead to ventricular fibrillation. Hypercalcemia or administration of intravenous calcium may lead to life-threatening arrhythmias or other manifestations of digoxin toxicity even in the setting of normal SDC.
DIGOXIN INTOXICATION
Increased serum digitalis concentrations can lead to intoxication, a clinical diagnosis, which can cause a multitude of symptoms, the most concerning of which are the electrophysiological effects, which can lead to life-threatening arrhythmias by mechanisms described previously. The overall incidence of digoxin toxicity is <1% per patient-year, down from 35% in 1971.6 The commonly described “digitalis effect” on 12-lead electrocardiogram (ECG) should be distinguished from true intoxication. It is commonly manifested by sagging of the ST segments, can occur at normal SDC, and requires no treatment. Digitalis excess is defined as the presence of significant ECG changes such as heart block and/or mild clinical symptoms such as nausea or fatigue. Proper treatment includes holding digoxin therapy with close monitoring of electrolytes. True “digitalis intoxication” is a relatively rare occurrence manifested by life-threatening arrhythmias and severe gastrointestinal and neurological symptoms (Table 11-3). It is best treated with antidigoxin antibodies.26 The results of prospective trials with adults and children have established the effectiveness and safety of antidigoxin antigen binding fragments (Fab) in treating cases of life-threatening digoxin intoxication, including cases of massive ingestion of the drug with suicidal intent (digoxin overdose). Although hemodialysis does not remove digoxin from the body, it can often be useful in cases where hyperkalemia is also present.
DIGOXIN IN HEART FAILURE WITH REDUCED SYSTOLIC FUNCTION
In several studies, digoxin withdrawal in patients with systolic dysfunction and sinus rhythm was associated with a decrease in left ventricular (LV) ejection fraction (EF) and exercise tolerance and an increase in resting heart rate, diastolic pressure, body weight, and/or cardiac size on chest x-ray. The majority of double-blind trials examining the effect of digoxin in patients with HF and systolic dysfunction have noted an improvement in clinical status. Although there has been little evidence that digoxin therapy improves survival, the majority of published data consistently describe a variety of clinical benefits.
The body of clinical research establishing the role of digoxin therapy in HF took place in three distinct phases. Initially, observational studies of digoxin withdrawal were performed in patients with chronic heart failure in normal sinus rhythm. Subsequently, placebo-controlled trials were conducted examining withdrawal of digoxin therapy in patients receiving stable background therapy of diuretics with and without ACE inhibitors. Some of these studies also compared the effects of digoxin withdrawal in patients receiving other oral inotropic agents. Data from the Prospective Randomized Study of Ventricular Failure and the Efficacy of Digitalis (PROVED) and the Randomized Assessment of Digitalis on Inhibitors of the Angiotensin Converting Enzyme (RADIANCE) study provided solid evidence that digoxin therapy improved symptoms and decreased hospitalizations in patients with chronic heart failure.27,28 None, however, addressed the issue of mortality or survival, nor the effects of de novo digoxin therapy in HF. This was the rationale for conducting the DIG trial, a large randomized clinical trial designed to assess the mortality benefit of digoxin therapy for HF patients in normal sinus rhythm who were already receiving angiotensin-converting enzyme (ACE) inhibitors and diuretics (Table 11-4).29
Early Randomized Studies
In a multicenter, double-blind, placebo-controlled study conducted by the Captopril-Digoxin Multicenter Research Group, digoxin was compared to captopril and placebo in a population of 300 patients with mild to moderate HF.30 Patients were randomized to captopril (target dose 50 mg tid), digoxin (target serum level 0.9–3.2 nmol/L), or placebo for 6 months. In the prerandomization phase of the study, only patients who did not deteriorate when digoxin was discontinued were randomized. Despite this initial bias against digoxin, the results for the digoxin group show fewer hospitalizations and emergency room visits for HF, a better EF, and a decrease in diuretic requirements compared to patients randomized to placebo. Except for the changes in EF, similar benefits were also noted in the patients treated with captopril.
However, digoxin had no effect on the New York Heart Association (NYHA) class or exercise capacity. This study was not powered to assess a mortality difference between groups but demonstrated the effectiveness of digoxin therapy in treating the signs and symptoms of HF in patients receiving optimal doses of diuretic therapy.
As intravenous and oral inotropic agents began to emerge as possible therapies for HF, investigators sought to compare these new therapies to digoxin, the only established oral inotropic therapy available at the time. In a 12-week study randomizing patients to milrinone, digoxin, combination therapy, or placebo, a combination of milrinone and digoxin was found to be equivalent to digoxin alone for improvement in exercise tolerance. Digoxin was superior to placebo and milrinone alone in improvement of exercise tolerance.31 Patients randomized to digoxin required fewer cointerventions, defined as additional therapy to control symptoms of HF, than those taking placebo. A trial of oral ibopamine, another oral inotropic therapy, likewise failed to demonstrate an advantage over oral digoxin therapy.32 Both studies confirmed earlier data that digoxin therapy significantly improved exercise tolerance and decreased the frequency of worsening HF compared to placebo.
Prospective Randomized Study of Ventricular Failure and the Efficacy of Digitalis (PROVED) and Randomized Assessment of Digitalis on Inhibitors of the Angiotensin Converting Enzyme (RADIANCE)
The PROVED investigators conducted a randomized, placebo-controlled trial of digoxin withdrawal after a 12-week stabilization period in which all patients received digoxin. Patients received background therapy of diuretics only. A total of 88 patients were randomized. Patients withdrawn from digoxin therapy after 12 weeks of stabilization showed worsened maximal exercise capacity (median change in exercise time -96 s) compared to those who continued therapy. Patients who received digoxin had a lower body weight and heart rate compared to those who received placebo. There was a two-fold increase in the incidence of worsening HF in patients taking placebo.27
Designed as a companion trial to PROVED, RADIANCE was a similarly designed digoxin withdrawal study on the background of ACE inhibitor (captopril or enalapril) and diuretic therapy. All patients had clinical evidence NYHA Class II–III HF, an EF <35%, and LV end-diastolic dimension >60 mm by echocardiography. Exercise testing was performed at baseline and at follow-up intervals and was measured with simple treadmill walk time, and 178 patients were randomized after the stabilization period. Despite the fact that all patients received background ACE inhibitor and diuretics, the investigators found a six-fold increase in worsening HF in the withdrawal group (after a stabilization period as in PROVED) and significant decrease in cardiac performance as measured by decrease in EF and increased heart rate. In addition, body weight was significantly reduced in the digoxin group. Most striking were the data indicating significantly improved quality of life scores in patients who were maintained on digoxin therapy.
Several post-hoc analyses of the PROVED and RADIANCE data have been published. Triple therapy with ACE inhibitors, diuretics, and digoxin was associated with the best outcomes and patients with mild symptoms (NYHA Class II) also benefited from therapy (Fig. 11-1, Table 11-5). The clinical deterioration rate was 5% in patients receiving triple therapy, compared to 30% in patients receiving diuretics alone.33 In a cost-benefit analysis, one study concluded that digitalis therapy has a 90% likelihood of saving the health-care system between $106 million to $822 million annually.34 These figures were based on combined data of the two trials, and had the patients in the PROVED study been taking ACE inhibitor therapy, the figures may have been even more robust.
The Digitalis Investigation Group Trial
In the largest placebo-controlled trial of digoxin therapy ever conducted, 7788 patients with EF <45% received 0.25 mg of digitalis or placebo. The trial was specifically designed to assess mortality differences between the two groups and the effect of de novo digoxin therapy. Investigators also sought to validate an existing formula for the estimation of SDC. Other important questions were the effect of LVEF, NHYA functional class, and cardiothoracic ratio on clinical outcomes of digoxin therapy. All patients were maintained on background CHF therapy of ACE inhibitors and diuretics as needed. Approximately one-half of patients entering the trial were already taking digitalis prior to randomization.
At 37 months’ mean follow-up, there was no difference in mortality between digitalis and placebo. The overall cardiovascular mortality was approximately 30%. However, there was a significant reduction in the primary endpoint of combined death or hospitalization for worsening HF in patients receiving digoxin therapy. Therapy was associated with a 28% reduction in the risk of 6.5% reduction in total hospitalizations compared to placebo. Analysis of the mortality data indicates a reduction in death for worsening HF was offset by death from other causes, which many have presumed to be sudden death from arrhythmias related to digoxin toxicity. In addition, investigators determined that SDC could be reliably estimated in patients with normal serum creatinine (Table 11-6, Fig. 11-2).
Importantly, this study using prespecified subgroup analysis showed a reduction in total mortality and total hospitalizations during the first 2 years after randomization in patients with an EF of <25% or patients with moderate to severe symptoms of CHF (NYHA Class III–IV), and patients with cardiomegaly on chest x-ray.
DIGITALIS IN HEART FAILURE AND PRESERVED SYSTOLIC FUNCTION
In the DIG trial, an additional 988 patients with EF >45% were randomized to digoxin or placebo in the same manner as the main trial.54 For the com at least one hospitalization for recurrent HF and a bined endpoint of death or hospitalization for worsening CHF, there was a decrease in the combined endpoint of heart failure hospitalization and cardiovascular mortality at 2 years, but this difference was not significant over the entire course of follow-up. The authors did note a trend toward reduction in hospitalization for worsening heart failure. The initial 2-year data from the DIG trial resulted in FDA approval of digoxin for patients with symptomatic heart failure regardless of LVEF; however, the endpoint reduction for patients with normal systolic function is less clear (Fig. 11-3). Accordingly, digoxin therapy should be considered in this population only after other therapies have failed. However, atrial fibrillation is common in the setting of diastolic dysfunction, and digoxin therapy can safely be initiated if needed for rate control.
DIGITALIS IN WOMEN
A post-hoc subgroup analysis of the DIG trial sought to assess sex-based differences in the outcome of digoxin therapy. Women represented 22% of the patients enrolled in the DIG study, and an absolute mortality difference of 5.8% was found in favor of placebo.35 This difference was not apparent in men, where there was no mortality difference between groups. Women were also found to have a smaller (but still significant) reduction in hospitalization for HF. Not surprisingly, these findings raised strong concern about the role of digoxin therapy for women with HF. The authors suggested that “women may not consider the potential increased risk of death associated with digoxin therapy worth the small reduction in the risk of hospitalization.” However, further analysis has since revealed that for women with serum digitalis concentrations <1 ng/mL, there was no increase in mortality.36 Some have hypothesized that data indicating increased mortality in women may be related to increased risk of toxicity due to overdosing and lower body weights; however, the DIG data do not support this conclusion. Further research is needed to examine sex-based differences in response to all common HF remedies, as women have always been underrepresented in important clinical trials. However, digoxin therapy at low doses remains safe and likely effective to improve symptoms and reduce the rate of hospitalizations for women with symptomatic, chronic HF and reduced systolic function.
CHRONIC HEART FAILURE AND ATRIAL FIBRILLATION
Because it can be taken once daily, is well tolerated, and is inexpensive, digoxin remains an important drug for rate control in atrial fibrillation. However, in patients with normal AV node conduction in atrial fibrillation, digoxin alone will not control the rate of ventricular response, particularly during exercise unless very high doses are given. These doses have been shown to cause digoxin intoxication in a majority of patients. 37,38 Likewise, b-blockers and calcium channel blockers given as monotherapy are unlikely to control ventricular response unless given at doses likely to cause side effects.39 The most likely combination to effectively control ventricular response in patients with atrial fibrillation and reduced systolic function is a combination of digoxin and b-blockers. In one randomized trial of patients with symptomatic heart failure and atrial fibrillation, the combination of carvedilol and digoxin was found to be superior to either drug alone in the management of atrial fibrillation in patients with HF and reduced systolic function.52 This combination reduced symptoms, improved ventricular function, and achieved better ventricular rate control than either agent alone. Clinical experience with this combination has produced similar results, and should be considered the standard treatment for HF patients with reduced systolic function with persistent atrial fibrillation. In patients with atrial fibrillation and preserved systolic function, verapamil or diltiazem may be used, particularly in patients with reactive airway disease or in patients who are unable to tolerate b-blocker therapy. In addition, digoxin should be used to control ventricular response in atrial fibrillation even in patients without HF, unless there is intrinsic conduction disease.
Digoxin is not effective for rate control in patients with high sympathetic tone, low serum K levels, or in patients receiving sympathomimetic medications such as vasopressors or bronchodilators. In addition, digoxin therapy is not recommended for patients with atrial fibrillation and controlled ventricular response. These patients are more likely to have underlying conduction disease, and digoxin could worsen the condition. Digitalis has not been shown to promote conversion to normal sinus rhythm in patients with or without HF with decreased systolic function, but may reduce the frequency of symptomatic episodes.40,41
DIGITALIS AND CORONARY ARTERY DISEASE
A significant number of patients with HF in the United States and Europe have coronary artery disease. Acute hypoxemia predisposes to the manifestations of digitalis intoxication, and myocardial ischemia itself may cause inhibition Na-K ATPase.42,43 Digitalis in animal models has been shown to induce coronary vasoconstriction leading to ischemia, an effect postulated to be caused by a-adrenergic stimulation.44,45 In several studies, digoxin has been reported to increase the early post-discharge mortality in patients who survive a myocardial infarction. In contrast, a regression analysis failed to show digitalis to be an independent predictor of mortality. In the DIG trial, 70% of patients had coronary artery disease. At the time of enrollment, 30% had angina and 60% had a previous myocardial infarction. Hospitalizations for acute coronary syndromes and ventricular arrhythmias were almost identical in the digoxin and placebo groups. However, when the digoxin effects in prespecified subgroups were examined, it appeared that the reduction in hospitalizations and mortality due to worsening HF was lower in patients with coronary artery disease. Similar results were seen in the RADIANCE and PROVED trials. During episodes of ischemia, the presence of digoxin may have a proarrhythmic effect and its beneficial effects in HF may be offset by its potential to cause malignant arrhythmias. If digoxin must be used in the setting of acute ischemia, there is experimental evidence that coadministration of b-blocker therapy will reduce the incidence of ventricular arrhythmias.46 Digoxin therapy should be used with extreme caution, if at all, in patients with acute coronary syndromes. If used, serum concentration should be <0.9 ng/mL. The safety of digoxin may be increased by administration of b-blockers and aldosterone antagonists and decreased by the use of loop diuretics and other inotropic agents including dobutamine, dopamine, and milrinone (Table 11-7).
SERUM DIGITALIS CONCENTRATION
There is increasing evidence that serum digitalis concentrations previously thought to be therapeutic may actually increase mortality. In the Prospective Randomized Milrinone Survival Evaluation (PROMISE) trial, SDC >1.1 ng/mL was associated with a 38% excess mortality rate.48 Early post-hoc analysis of the DIG trial indicated an increased risk of cardiovascular mortality in patients with SDC >0.8 ng/dL.47 More recently, a comprehensive post-hoc analysis was performed that analyzed 1687 patients from the digoxin group who had SDC measurements taken one month after randomization.55 Compared to 3861 patients in the placebo group, it was found that while high SDC (>1.0 ng/mL) was associated with increased mortality, SDC of 0.5 to 0.9 ng/mL reduced overall mortality and hospitalization in all patients compared to placebo, including those with preserved systolic function (Fig. 11-4).48 Although increasing digoxin dose within the therapeutic range may further improve cardiac performance, increasing the dose is not associated with an improvement in the neurohormonal profile.49 These clinical trial data have led to the currently accepted practice of initiating low-dose digoxin therapy for patients with HF and decreased systolic function in sinus rhythm. In a separate analysis of the data from PROVED and Since the development of the radioimmunoassay for digoxin by Dr Thomas Smith in 1969, there has been spirited debate regarding the levels of serum digoxin that result in clinical toxicity. There is significant overlap in SDC between patients who are intoxicated and those who are not. In the 30 years that SDC has been measured routinely, we have learned several important lessons. First, routine SDC should be obtained only under special circumstances— when toxicity is suspected or to monitor compliance, for monitoring of drug interactions, for example, quinidine, verapamil, and amiodarone, or in the setting of renal dysfunction. Levels should be measured just before daily administration to assure that a “trough” level is measured. Second, patients with systolic HF should be maintained on the lowest dose of digitalis, which is likely to provide a clinical effect, and large doses (>0.125 mg for women, >0.25 mg for men) should be reserved for patients with greater body mass and for patients in whom higher doses are needed for rate control in the setting of atrial fibrillation (Table 11-8).
DIGITALIS IN THE SETTING OF MODERN THERAPY FOR CHRONIC HEART FAILURE
ACE Inhibitors
Analysis of PROVED and RADIANCE indicates that the combination of digoxin and an ACE inhibitor with diuretics is superior to either drug alone to prevent worsening HF and improve symptoms. These results were confirmed by the DIG trial, leading to the acceptance of digoxin as standard therapy for symptomatic HF with reduced systolic function in patients stabilized on ACE inhibitor and diuretic therapy. Large randomized studies demonstrating the mortality benefits of ACE inhibitors have reported background digoxin therapy ranging from 50% to 90%. Retrospective analysis from the Assessment of Treatment with Lisinopril and Survival (ATLAS) study indicated that there was an incremental benefit of adding b-blocker therapy and digoxin to high dose ACE inhibitor therapy. Patients receiving high dose ACE inhibitors plus b-blockers plus digoxin had 12% fewer deaths and hospitalizations than patients receiving low-dose ACE inhibitors alone.56
b-Blockers
In recent large randomized studies demonstrating significant mortality benefit of b-blocker therapy in chronic heart failure, concurrent background digoxin therapy was reported. In a 1996 trial demonstrating the mortality benefit of carvedilol, 90% of patients were taking digoxin.7 This percentage fell to 51% in a similar trial demonstrating the benefits of carvedilol published in 2001.8 Digoxin may be safer in patients receiving b-blocker therapy, and useful particularly during the initiation of b-blocker therapy when there is an initial decrease in EF and cardiac output.
Aldosterone Blocking Agents
In the Randomized Aldactone Evaluation Study (RALES) trial, 75% of patients were receiving digoxin therapy.50 Analysis of the mortality data indicates that patients receiving digoxin were more likely to benefit from aldactone therapy compared to placebo. In addition, the potential beneficial effects of aldosterone blockers on perivascular and myocardial fibrosis may be potentiated by the addition of digoxin therapy. Aldosterone-blocking agents also decrease the risk of digitalis toxicity by increasing serum potassium levels. Given these possible synergistic interactions, it is reasonable to recommend continuation of digoxin therapy in patients with CHF receiving aldosterone-blocking agents.
Cardiac Resynchronization Therapy
The Cardiac Resynchronization in Heart Failure (CARE-HF) study was the first randomized study of CRT therapy in HF to demonstrate a clear mortality benefit in patients with reduced systolic function and severe symptoms.51 In both the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) and CARE-HF, digoxin was no longer considered as part of optimal pharmacological therapy.51,52 Nevertheless, in CARE-HF, approximately 40% of patients received digoxin. Digoxin appears to significantly reduce mortality and hospitalization rate particularly in patients with severe symptoms and very low EF (those more likely to receive cardiac resynchronization therapy [CRT]). Accordingly, CRT therapy should be considered in patients who continue to have symptoms of HF despite standard therapy, including digoxin.
RECOMMENDATIONS
Digoxin is inexpensive and has few side effects when dosed appropriately. Published clinical trial data support the current Class IIa recommendation of the ACC/AHA for its use in patients with symptomatic HF and reduced systolic function. It is the only available inotrope that decreases morbidity without increasing mortality. Current data indicate that its effect is enhanced, not diluted, in the presence of other proven therapies for chronic heart failure including b-blockers, aldosterone antagonists, and ACE inhibitors. This multidrug approach to the patient with HF is accepted as the standard of care. Until large randomized trials are performed that indicate the effectiveness of these medications persist in the absence of digoxin, continued use in this patient population should be considered to improve symptoms and prevent significant costs associated with repeated hospitalization. It is likely that digoxin will continue to play an important role in the
REFERENCES
1. Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: summary article. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to revise the 2001 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol. 2005;46:1116–1143.
2. Adams KF Jr., Lindenfeld J, Arnold JM, et al. Executive summary: HFSA 2006 comprehensive heart failure practice guideline. J Card Fail. 2006; 12:10–38.
3. Rathore SS, Wang Y, Krumholz HM. Sex-based differences in the effect of digitalis for the treatment of heart failure. N Engl J Med. October 31, 2002;347(18):1403–1411.
4. Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult– Summary Article. Circulation. 2005;112: 1825–1852 and J Am Coll Cardiol. 2005;46: 1116–1143.
5. Withering W. An account of the foxglove and some of its medicinal uses with practical remarks on dropsy and other diseases. Swinney, Birmingham, England 1785.
1. Beller GA, Conroy J, Smith TW. Ischemia-induced alterations in myocardial Na/K-ATPase and cardiac glycoside binding. J Clin Invest. 1976;57: 341–350.
2. Packer M, Bristow M, Cohn J, et al. The Effect of Carvedilol on Morbidity and Mortality in Patients with Chronic Heart Failure. NEJM. 1996;334(21): 1249–1355.
3. Packer M, Coats J, Fowler M, et al. Effect of Carvedilol on Survival in Severe Chronic Heart Failure. NEJM. 2001;344(22):1651–1658.
4. The MERIT-HF Investigators. Effects of controlled-release metoprolol on total mortality, hospitalizations and well-being in patients with heart failure. JAMA. 2000;283(10):1295–1302.
5. CIBIS II Investigators and Committees. The Cardiac Insufficiency Bisoprolol Study II (CIBIS- II). Lancet. 1999;353:9–13.
6. Smith TW, Antman EM, Friedman PL, et al. Digitalis glycosides: mechanisms and manifestations of toxicity. Prog Cardiovasc Dis. 1984;26: 413–458, 495–530.
7. Wang W, Chen JS, Zucker IH. Carotid sinus baroreceptor sensitivity in experimental heart failure. Circulation. 1990;91:1959–1966.
8. Krum H, Bigger JT, Jr, Goldsmith RL, et al. Effect of long-term digitalis therapy on autonomic function in patients with chronic heart failure. J Am Coll Cardiol. September 1995;26(3):838.
9. Ferguson DW, Berg WJ, Sanders JS, et al. Sympathoinhibitory responses to digitalis glycosides in heart failure patients: direct evidence
from sympathetic neural recordings. Circulation. 1989;80:65–67.
1. van Veldhuisen DJ, Man in’t Veld AJ, Dunselman PHJM, et al. Double-blind placebo controlled study of ibopamine and digoxin in patients with mild to moderate heart failure: results of the Dutch Ibopamine Multicenter Trial (DIMT). J Am Coll Cardiol. 1993; 22:155–161.
2. Zhang J, Sun Y, Weber K. Coronary fibrosis in hyperaldosteronism: prevention by digoxin. Circulation. 1999; 100 (suppl I):1–560.
3. Lown B, Grayboys TB, Podrid PJ, et al. Effect of digitalis drug on ventricular premature beats. N Engl J Med. 1977;296:301–306.
4. Gheorghiade M, St. Clair J, St. Clair C, et al. Hemodynamic effects of intravenous digoxin in patients with severe heart failure initially treated with diuretics and vasodilators. J Am Coll Cardiol. 1987:9:849–857.
5. Gheorghiade M, Hall VB, Jacobsen G, et al. Effects of increasing maintenance dose of digitalis on left ventricular function and neurohormones in patients with chronic heart failure treated with diuretics and angiotensin-converting enzyme inhibitors. Circulation. 1995;92:1801–1807.
6. Woodland C, Verjee Z, Giesbrecht E, et al. The digitalis-propafenone interaction: characterization of a mechanism using renal tubular cell mono-layers. J Pharmacol Exp Ther. 1997;283:39–45.
7. Klein HO, Lang R, Weiss E, et al. The influence of verapamil on serum digitalis concentration. Circulation. 1982;65:998–1003.
8. Hager WD, Fenster P, Mayersohn M, et al. Digitalis-quinidine interaction pharmacokinetic evaluation. N Engl J Med. 1979;300:1238–1241.
9. Paladino JA, Davidson KH, McCall BB. Influence of spironolactone on serum digitalis concentration [letter]. JAMA. 1984;251:470–471.
10. Weeks CE, Conard GJ, Kvam DC, et al. The effect of flecainide acetate, a new antiarrhythmic, on plasma digitalis levels. J Clin Pharmacol. 1986; 26:27–31.
11. Robinson K, Johnston A, Walker S, et al. The digitalis-amiodarone interaction. Cardiovasc Drugs Ther. 1989;3:25–28.
12. Smith TW, Harber E. Digoxin intoxication: the relationship of clinical presentation toserum digoxin concentration. J Clin Invest. 1970;49: 2377–2386.
|
