Key result
Serum digoxin concentrations of 0.5-0.7 ng/mL were associated with the lowest risk of all-cause mortality compared with placebo, while concentrations ≥1.5 ng/mL were associated with increased mortality.
Why the study?
Does digoxin targeted to specific serum digoxin concentrations reduce mortality and hospitalizations in patients with heart failure?
Does digoxin targeted to specific serum digoxin concentrations reduce mortality and hospitalizations in patients with heart failure?
Answers are grounded in this paper and cite their sources.
Although secondary analysis of the DIG trial suggests an optimal serum digoxin concentration of 0.5-0.7 ng/mL for mortality benefit, targeting this narrow range in routine clinical practice is likely impractical and should not yet revise existing guidelines.
This article refers to ‘Dose response characterization of the association of serum digoxin concentration with mortality outcomes in the Digitalis Investigation Group trial’, by K.F. Adams et al., published in this issue on pages 1072–1081. Despite >200 years of clinical experience and a pivotal trial,1 the use of digoxin has declined precipitously,2, 3 and the drug remains the most controversial therapy in contemporary cardiovascular medicine.4, 5 One of the major points of contention has been the utility of dosing the drug based on the serum digoxin concentration (SDC). In the landmark DIG (Digitalis Investigation Group) trial,1 an SDC between 0.5 and 2.0 ng/mL was considered to be therapeutic, and this is still the standard reference range employed by most hospital-based and freestanding laboratories. However, several previously published post-hoc analyses of the DIG trial6-8 have raised the hypothesis that higher digoxin doses achieving SDCs thought to be therapeutic in the past (i.e. 1.0–2.0 ng/mL) may not have additional benefit in terms of reducing hospitalizations and may adversely impact survival. In response, the major US guidelines9, 10 currently recommend a target SDC of <1.0 ng/mL, while the European guidelines11 make no specific recommendation. This issue of the journal features a secondary analysis by Adams and colleagues12 that revisits the dose response relationship between SDC and clinical outcomes in the DIG trial. The study population consisted of 5209 out of the 6800 patients enrolled in the DIG trial, including all 3366 patients randomized to the placebo arm and 1843 out of the 3372 patients randomized to digoxin with an available SDC. SDC measurements were made between 6 and 30 h after the last dose of the study drug at week 4. Patients with higher SDCs at week 4 tended to have worse renal function and more advanced heart failure (HF) compared with the placebo group. Notably, there was an association between digoxin and reduced mortality until the SDC exceeded 0.9 ng/mL, while beginning at an SDC of 1.5 ng/mL digoxin was associated with increased mortality. The lowest SDC range reported in the DIG trial (i.e. 0.5–0.7 ng/mL) was associated with the least risk of all-cause mortality compared with placebo. There was also decreased risk of death due to worsening HF for SDCs between 0.5 and 1.1 ng/mL, while above 1.1 ng/mL there was not a significant association between SDC and death due to worsening HF. In contrast, digoxin had a neutral effect on the risk of cardiovascular mortality not due to HF at SDCs <1.0 ng/mL, whereas SDCs ≥1.1 ng/mL were associated with substantially higher risk of death due to cardiovascular causes not related to HF. Finally, there was direct correlation between increasing SDC and risk of all-cause hospitalization, but no association between SDC and risk of hospitalization due to HF. The authors concluded that there is a clear dose response curve between SDC and clinical outcomes, and recommend targeting an SDC range of 0.5–0.7 ng/mL for maximal benefit. There are several weaknesses of the data that should be acknowledged. First, SDCs were only available for a subset of the patients in the DIG trial at a single time point, while outcomes were collected for a mean follow-up duration of 37 months (range 28–58 months).1 Secondly, digoxin dosing in the DIG trial was based on an aglorithm incorporating age, gender, body weight, and renal function, and the DIG trial did not prospectively test the hypothesis that SDC monitoring would improve the efficacy or safety of digoxin therapy. It is conceivable that higher SDCs may simply be a surrogate for other measured or unmeasured confounders not adjusted for in multivariable modelling. This is supported by the fact that Adams et al. found a clear association between higher SDC and more advanced heart and renal failure. Thirdly, it may be impractical for physicians to target an SDC of 0.5–0.7 ng/mL in routine clincial practice. The protocol for the DIG trial allowed blood samples for SDCs to be drawn between 6 and 30 h after the last dose of digoxin. Ideally, SDCs should be drawn at least 8 h and preferably 12 h after the last dose to allow complete distribution from the blood to the tissues and a steady state to be achieved.13 Blood samples drawn before 8–12 h will result in a falsely elevated SDC. In real-world clinical practice, studies have shown that inappropriately drawn specimens constituted 55% of the tests performed.14 Thus, after taking into account changes in clincial status, renal function, potential drug–drug interactions, and limitations of SDC measurements, it does not seem feasible to target such a narrow therapeutic window over the long term. Furthermore, the clinical utility of dosing digoxin to target an SDC of 0.5–0.7 ng/mL is questionable. Although digoxin acts as a neurohormonal modulator at low SDCs by increasing parasympathetic tone and decreasing activation of the sympathetic nervous system and the renin–angiotensin–aldosterone system (RAAS), background therapy for most of these early studies consisted only of a diuretic and possibly an ACE inhibitor.15-18 It seems unlikely that digoxin would have a significant effect on circulating catecholamines and aldosterone levels in the modern era of potent neurohormonal antagonists (i.e. beta-blockers and RAAS inhibitors). In addition, the beneficial effects of digoxin in the DIG trial and earlier studies were probably a result of its haemodynamic properties, principally driven by a reduction in PCWP, which are minimal or absent at lower SDCs.19, 20 The haemodynamic hypothesis is supported by the fact that digoxin led to an early reduction in all-cause and cause-specific hospitalizations in the DIG trial, while for the most part pivotal trials of neurohormonal antagonists demonstrated a robust reduction in cardiovascular morbidity and mortality that emerged only over a more prolonged duration of follow-up.1 Although the management of ambulatory HF patients with a reduced EF has been revolutionized by guideline-directed medical therapies, there has been no such parallel progress in the care of patients admitted for worsening chronic HF, and hospitalization continues to remain an inflection point in the natural history of the condition.21 Digoxin is the only safe inotrope known to increase cardiac output and decrease PCWP without causing deleterious increases in heart rate or decreases in blood pressure. Mutiple well-designed clinical trials have unequivocally shown that digoxin improves signs and symptoms and exercise capacity, and reduces all-cause and cause-specific hospitalizations.1, 22, 23 In addition, at lower SDCs (i.e. <1 ng/mL) and in certain high-risk groups (i.e. patients with NYHA functional class III or IV symptoms, EF <25%, and cardiothoracic ratios >55%), digoxin may improve survival.7, 8, 24, 25 Despite safety concerns regarding supratherapeutic levels and fatal arrhythmias, the absolute incidence of digoxin toxicity requiring hospitalization is low both in the controlled setting of a trial1 and in the context of routine clinical practice.26 Thus, the favourable effects of digoxin on cardiac performance and central haemodynamics translate into tangible short- and long-term clinical benefits. In conclusion, patients with worsening chronic HF with a reduced EF continue to experience unacceptably high rates of cardiovascular morbidity and mortality despite available therapies. Although the use of digoxin has declined considerably over the past couple of decades, there are compelling public health reasons to reconsider the use of digoxin in patients with worsening chronic HF with reduced EF who remain symptomatic and at high risk for admission despite optimal medical therapy. However, the feasaibility and clinical value of dosing digoxin to target an SCD of 0.5–0.7 ng/mL is uncertain and the data presented by Adams et al. should be considered hypothesis-generating only and not used as justification to revise existing guideline recommendations. Additional prospective studies are required to establish the optimal dosing of digoxin as well as the role of serial SDC monitoring in routine clincial practice. Conflict of interest: M.G. has been a consultant for Abbott Laboratories, Astellas, AstraZeneca, Bayer HealthCare AG, CorThera, Cytokinetics, DebioPharm S.A., Errekappa Terapeutici, GlaxoSmithKline, Ikaria, Johnson & Johnson, Medtronic, Merck, Novartis Pharma AG, Otsuka Pharmaceuticals, Palatin Technologies, Pericor Therapeutics, Protein Design Laboratories, Sanofi-Aventis, Sigma Tau, Solvay Pharmaceuticals, Takeda Pharmaceutical, and Trevena Therapeutics. A.P.A. has no conflicts of interest to declare.
No takes yet. Share an insight, caveat, or question.
Ambrosy et al. (2016) conducted an editorial in Heart failure (n=5,209). Serum digoxin concentration (SDC) vs. Placebo was evaluated on All-cause mortality. Serum digoxin concentrations of 0.5-0.7 ng/mL were associated with the lowest risk of all-cause mortality compared with placebo, while concentrations ≥1.5 ng/mL were associated with increased mortality.
Synapse has enriched 10 related papers. Consider them for comparative context: