Low-dose digoxin in patients with heart failure with reduced or mildly reduced ejection fraction: a randomized controlled trial

Abstract Digoxin is the oldest drug in cardiovascular medicine, but its value in the current management of heart failure is unclear. Earlier studies have suggested that low-dose digoxin might be beneficial, but evidence from rigorous randomized clinical trials is lacking. In this double-blind, placebo-controlled trial (the DECISION trial), 1,001 patients with symptomatic chronic heart failure and a left ventricular ejection fraction of 50% or less were randomized to low-dose digoxin or placebo, with a target serum digoxin concentration of 0.5–0.9 ng ml−1. The mean age of the participants was 72 ± 9 years, 28% were women and 29% had atrial fibrillation. The primary outcome was a composite of total worsening heart failure events, defined as total hospitalizations or total urgent hospital visits for worsening heart failure and cardiovascular mortality. Over a median follow-up of 36.5 months, 238 primary-outcome events occurred in 131 of 500 patients in the digoxin group, and 291 primary-outcome events in 152 of 501 patients occurred in the placebo group (rate ratio 0.81; 95% confidence interval (CI) 0.61–1.07, P = 0.133). The total number of worsening heart failure events was 155 and 203 in the digoxin and placebo groups, respectively (rate ratio 0.76, 95% CI 0.54–1.05) and cardiovascular mortality occurred in 83 patients (17%) and 88 (18%) in the digoxin and placebo groups, respectively (hazard ratio 0.93, 95% CI 0.69–1.26). Low-dose digoxin was generally well tolerated and safe, and results were similar between men and women. The results of this trial indicate that in patients with heart failure and reduced or mildly reduced ejection fraction, low-dose digoxin did not significantly reduce the composite endpoint of total worsening heart failure events or cardiovascular mortality. ClinicalTrials.gov registration: NCT03783429. Similar content being viewed by others Main Digoxin is commonly used in patients with heart failure and atrial fibrillation and is the oldest drug in cardiovascular medicine1. In 1997, the Digitalis Investigation Group (DIG) trial, a large placebo-controlled study in patients with heart failure with sinus rhythm demonstrated, that digoxin—when added to diuretics and angiotensin-converting enzyme inhibitors—did not reduce all-cause mortality, its primary endpoint2. The trial also showed, however, that digoxin was associated with a significant 28% reduction in hospitalization for worsening heart failure. Despite these findings, digoxin use subsequently declined for several reasons, in particular the introduction of other effective drugs and devices for heart failure, as well as numerous post hoc analyses suggesting that digoxin use was associated with unfavorable outcomes, often due to prescription bias3. Additional post hoc studies from DIG and other databases showed that higher serum digoxin concentrations (SDC) (>1.2 ng ml−1), which were present in 24% of patients, were associated with an increased mortality, while lower SDC were associated with favorable outcomes4,5. Although these findings suggest that low-dose digoxin might be beneficial and guidelines advocated to avoid SDC >1.2 ng ml−1 (ref. 6), no larger study with low-dose digoxin has been conducted in heart failure. In atrial fibrillation, one study in 160 patients showed that low-dose digoxin was as effective as bisoprolol regarding quality of life (primary endpoint) but it performed better on several heart failure-related endpoints7,8. Recently, the Digitoxin to Improve Outcomes in Patients with Advanced Heart Failure (DIGIT-HF) trial showed, that low-dose digitoxin, when added to contemporary heart failure treatment, led to a significant 15% reduction in the combined endpoint of all-cause death and first hospital admission for heart failure9, further supporting the value of low-dose digitalis glycosides in heart failure. Indeed, in a recent seminar on heart failure with reduced (or mildly reduced) ejection fraction, it was suggested that the results of DIGIT-HF “could lead to a renaissance in the use of digitalis glycosides”10. In the present study, we examined the effect of low-dose digoxin against a background of contemporary guideline-recommended treatment in the Digoxin Evaluation in Chronic heart failure: Investigational Study In Outpatients in the Netherlands (DECISION) trial11. Results From 3 July 2020 to 8 December 2023, a total of 1,002 patients were randomly assigned to receive digoxin or matching placebo (Fig. 1). One patient who was randomized never received study medication and was excluded from the analysis according to protocol, that is, before unblinding. Therefore, 1,001 patients were included in the primary analysis. By the end of the study, one patient had withdrawn consent for further follow-up in the study after having developed esophageal cancer and was censored. There were no patients lost to follow-up and no unblinding events in any treatment arm. The median duration of follow-up was 36.5 months (interquartile range (IQR) 26.3-47.8 months), and the median time on treatment was 31.3 months in both groups (IQR 20.9-43.9 months). At baseline, the characteristics of the patients and therapies for heart failure were well balanced between groups (Table 1). The mean age was 72 years and 28% were women. The large majority of patients were New York Heart Association (NYHA) functional class II, the mean left ventricular ejection fraction (LVEF) was 33% and 29% of patients had atrial fibrillation. The median N-terminal pro-B-type natriuretic peptide (NTproBNP) concentration was 1,404 pg ml−1, and mean estimated glomerular filtration rate (eGFR) was 56 ml min−1 1.73 m−2. At enrollment, >85% of patients were treated with a beta-blocker and a renin-angiotensin blocker and >70% were treated with a mineralocorticoid receptor antagonist and a diuretic, while 41% were on a sodium–glucose cotransporter-2 inhibitor. In total 38% of patients had an implantable cardioverter defibrillator and 17% had cardiac resynchronization therapy. Both medication and device treatment were significantly higher in patients with an ejection fraction ≤40% (ref. 12). A total of 238 primary outcome events occurred in 131 of 500 patients in the digoxin group, and 291 primary-outcome events occurred in 152 of 501 patients in the placebo group (rate ratio 0.81, 95% confidence interval (CI) 0.61–1.07, P = 0.133) (Table 2 and Fig. 2a). The results for the primary outcome in 11 prespecified subgroups are shown in Fig. 3. As the result for the primary outcome was not significant, subsequent analyses were considered exploratory. The total number of worsening heart failure events was 155 in the digoxin group and 203 in the placebo group (rate ratio 0.76, 95% CI 0.54–1.05) (Fig. 2b). A total of 83 patients (17%) died from cardiovascular causes in the digoxin group versus 88 patients (18%) in the placebo group (hazard ratio 0.93, 95% CI 0.69–1.26) (Fig. 2c). The total number of cardiovascular hospitalizations and urgent cardiovascular visits was 191 in the digoxin group and 169 in the placebo group (rate ratio 1.12, 95% CI 0.85–1.49) (Extended Data Table 1). In a time-to-first event analysis, 131 patients (26%) versus 152 patients (30%) in the digoxin and placebo groups, respectively, experienced either a worsening heart failure event or died from cardiovascular causes (hazard ratio 0.83, 95% CI 0.65–1.04) (Fig. 2d). A total of 118 patients (24%) in the digoxin group and 126 patients (25%) in the placebo group died from any cause (hazard ratio 0.93, 95% CI 0.72–1.19). Secondary outcomes are presented in Table 2 and Extended Data Figs. 1–3. Last, there were no differences in total hospitalizations between the digoxin and placebo group (rate ratio 1.00, 95% CI 0.82–1.22) Of the whole study population, 28% were women (n = 284) and 72% were men (n = 717); there were no differences with regard to the effect of digoxin on the primary endpoint between women and men (rate ratio digoxin versus placebo: 0.71 and 0.85, respectively, P value for interaction 0.61) (Fig. 3). Regarding rhythm, 29% of the patients had atrial fibrillation and the effect of digoxin was not different in these patients as compared to those in sinus rhythm (rate ratio 0.76 and 0.82, respectively, for atrial fibrillation and sinus rhythm, P value for interaction 0.80). In total, 21% of the patients had heart failure with mildly reduced ejection fraction (HFmrEF), the effect of low-dose digoxin was not different between patients with HFmrEF and those with heart failure with reduced ejection fraction (HFrEF; rate ratio 0.72 and 0.82, respectively, P value for interaction 0.75) As defined by the algorithm, 75% of patients started the study with 0.1 mg digoxin once daily and 25% of patients with 0.2 mg once daily. At the 4-week visit, 63.6% of patients with samples available had a SDC in the target therapeutic range of 0.5–0.9 ng ml−1. In the remaining patients, SDC was too high in 6.9% and it was too low in 29.5%. In patients with SDC not in the target range at the 4-week visit, SDC was repeated and after the second measurement 83.6% of patients with samples available were in the target range (Supplementary Fig. 1). Median SDC was 0.6 ng ml−1 (IQR 0.5–0.8 ng ml−1) and 17 patients had a SDC >0.9 ng ml−1 (the highest SDC was 1.6 ng ml−1). At the end of the study visit, 92 of 382 patients (24%) in the digoxin group and 77 of 375 patients (21%) in the placebo group had discontinued study drug treatment for reasons other than death. The risk for study drug discontinuation during COVID in the first year was higher than after COVID, with a hazard ratio of 1.47 (95% CI 1.13–1.91, P = 0.016). To correct for the relatively high discontinuation rate, we performed an as-treated sensitivity analysis. In this analysis, a total of 135 primary outcome events occurred in the digoxin group (incidence rate per 100 patient-years of 10.60), compared to 212 primary-outcome events in the placebo group (incidence rate per 100 patient-years of 16.12), with a rate ratio of 0.66 (95% CI 0.47–0.92, P = 0.015) (Extended Data Fig. 4). At the end of study visit, more patients in the placebo arm received sodium–glucose cotransporter-2 inhibitors as compared to the digoxin arm (80% versus 72%, P = 0.026). There was no difference in the other three guideline-recommended treatment prescription rates. The number of serious adverse events (excluding primary and secondary endpoints) was similar between groups, that is 19.9 events per 100 patient-years in the digoxin group versus 18.3 events per 100 patient-years in the placebo group (rate ratio 1.09, 95% CI 0.86–1.37). In particular there was no difference in hospitalizations or urgent visits for digestive problems (Extended Data Table 2). Treatment-related serious adverse events (which either led to study drug discontinuation or were flagged as such by the local investigator) were reported in 40 patients in the digoxin group and in 25 patients in the placebo group, which was not significantly different (rate ratio 1.53, 95% CI 0.91–2.55) (Extended Data Table 3). There were no significant differences in patient-reported side effects between the digoxin and placebo groups (Extended Data Table 4). The need for device implantation during the study was low and not significantly different between the two treatment groups (Extended Data Table 5). Discussion In patients with heart failure and a reduced or mildly reduced ejection fraction, low-dose digoxin on top of current guideline-recommended treatment did not result in a significantly lower rate of a composite of total worsening heart failure events and death from cardiovascular causes (primary outcome) than placebo. Low-dose digoxin was generally well tolerated and safe and did not lead to an increase in adverse effects or implantation of more pacemakers. By using a simple algorithm, the vast majority of patients were kept within target range and few patients had SDC >1.2 ng ml−1. Earlier post hoc studies have shown that low-dose digoxin (0.5–0.9 ng ml−1) would be more beneficial than higher doses in patients with heart failure4,5, but the present study, in patients with contemporary background medical therapy could not confirm this. Although the absolute number of primary outcome events was lower with digoxin than with placebo, the relative risk reduction was less than the assumed reduction in the sample size calculation (19% versus 22%, respectively)11. Our a priori assumption of the primary outcome event rate was correct, suggesting that a lower than expected event rate was not responsible for the lack of a statistically significant effect. There may be several explanations for the lack of a statistically significant treatment effect in the present trial. Our effect size assumptions were mainly based on analyses of DIG in almost 600 patients, who were all in the target range and had a SDC between 0.5 and 0.9 ng ml−1 and who, by definition, were on the drug5. In DECISION, the digoxin dosing algorithm generally worked well, although not all patients were in the target range. More importantly, a substantial proportion of patients discontinued the (active) study drug, which leads to a loss of statistical power. Interestingly, this did not seem to be primarily due to drug-induced side effects. One reason for the relatively high discontinuation rate may have been the fact that inclusion in our study started in July 2020, when the COVID pandemic was at a peak, which had major implications for heart failure trials13. In some hospitals, visits to the hospital (for research) were no longer allowed and stopped, and often these hospital visits were replaced by telephone calls. COVID lasted until the second quarter of 2022, which was well into our study. Also, the mean age of patients in DECISION was 72 years old, which is markedly higher than almost all heart failure trials, including DIG and DIGIT-HF2,9. Although very representative of the clinical heart failure population nowadays, elderly patients more often stop (study drug) medication for many other reasons including multidrug use and progression of disease. Last, the introduction and up titration of other guideline-recommended therapies during the course of the study may also have impacted the overall results as we found that more patients in the placebo arm received sodium–glucose cotransporter-2 inhibitors throughout the trial. All these factors together may have contributed to a lower than expected digoxin effect. In a sensitivity analysis when only patients who were on study drug were analyzed, a more pronounced drug effect was observed. The present study provides important data regarding safety of (low-dose) digoxin in women with heart failure. In a post hoc substudy from DIG from 2002, an increased risk of death with digoxin was observed in women, with a P value for interaction between men and women of 0.034 (ref. 14). Interestingly, SDC in women in DIG was shown to be significantly higher after 1 month than in in men. In a later subanalysis from the same DIG study, this adverse effect in women was observed in those with higher SDC, while a favorable clinical effect was found in women with SDC 0.5–0.9 ng ml−1 (ref. 15), which is the same SDC as the target levels in DECISION. In the present study, low-dose digoxin was safe in the 284 women studied, and a similar signal was also seen in 247 women who received (low-dose) digitoxin in DIGIT-HF9. These data therefore provide reassurance regarding the use of (low-dose) digoxin in women. DECISION is now the largest and only randomized, placebo-controlled outcome study investigating the effect of (low-dose) digoxin in patients with atrial fibrillation (and heart failure), since DIG only studied patients in sinus rhythm. The present data are therefore important since the safety of digoxin has been questioned in the past, based on post hoc, retrospective, nonrandomized data. One trial in 160 patients with atrial fibrillation, the RAte control Therapy Evaluation in permanent Atrial Fibrillation (RATE-AF) trial, compared low-dose digoxin to a beta-blocker, with no placebo control group. RATE-AF showed that digoxin and beta-blockers were similarly effective in terms of quality of life, the primary endpoint of the trial8, while digoxin performed better on heart failure-related endpoints. DECISION shows that digoxin (on top of betablockers) is safe and has similar efficacy compared to patients in sinus rhythm. In DIGIT-HF a comparable effect of low-dose digitoxin in patients with heart failure with atrial fibrillation was observed, that is a hazard ratio of 0.72 for their primary endpoint9. The present data should be put into perspective with the other two, large randomized controlled trials with digitalis glycosides, DIG and DIGIT-HF2,9. In DIG, there was no effect on all-cause mortality, but a significant 28% relative risk reduction of worsening heart failure events. In DECISION, the number of worsening heart failure events was, although nonsignificant, lower with digoxin compared to placebo, with a rate ratio (risk) of 0.76. Later analyses of DIG have shown that digoxin, when added to diuretics and angiotensin-converting enzyme inhibitors, reduced the composite endpoint of cardiovascular mortality and (first) heart failure hospitalization by 15%16. In DIG, digoxin also showed the largest effect on heart failure hospitalizations in patients with a reduced ejection fraction, particularly in those with the lowest ejection fraction17, and a similar effect was observed in DIGIT-HF9. In DECISION, we also included patients with heart failure with LVEF >40%, which generally is associated with a lower clinical event rate. Nevertheless, the number of patients who do have LVEF ≤40 is decreasing, and DECISION provides more data in the population with a mildly reduced LVEF11. In total, 21% of the patients in DECISION had heart failure with a mildly reduced ejection fraction and the effect of low-dose digoxin was similar in both groups. In DIGIT-HF9, low-dose digitoxin was examined against a background of contemporary heart failure treatment and the drug led to a 15% lower combined risk of first hospitalization for heart failure or all-cause death. This effect was similar to that observed in DIG2, and also to the (nonsignificant) effect in DECISION. Compared to DECISION, DIGIT-HF examined a more advanced heart failure population: mean ejection fraction was lower (29% versus 33%) and the NYHA functional class was higher, as well as event rates. Natriuretic peptides were not measured in DIGIT-HF, but both trials examined a heart failure population with contemporary background therapy and treatment effect estimates for digitalis glycosides versus placebo were independent of (the extent of) background therapy in both trials. Our trial has several limitations. We included a mostly male population with NYHA functional class II heart failure, and extrapolation to other patient populations should be done with caution. A larger than expected number of patients discontinued the study treatment, which may have contributed to the lack of a statistically significant treatment effect of digoxin. In conclusion, the findings from DECISION should be interpreted in the context of the totality of evidence that now exists for digitalis glycosides in patients with heart failure with (mildly) reduced ejection fraction, with comparable treatment effects on worsening heart failure events in all three trials. Considering the safety profile of low-dose digoxin in DECISION, it may be an easy, simple and cheap treatment option in patients with heart failure with (mildly) reduced ejection fraction. Methods Trial design and oversight DECISION was an investigator-initiated, double-blind, randomized, placebo-controlled trial, conducted at 43 sites in the Netherlands (Supplementary Table 1). The trial design as well as the baseline manuscript have been reported11,12, and the study was approved by the ethics committees affiliated with each of the investigational sites. The protocol is included in the Supplementary Information. The trial was conducted according to the principles of Good Clinical Practice and the Declaration of Helsinki11. All patients provided written informed consent. Investigators affiliated with the UMC Groningen designed the study, collected and managed the data. All authors with access to the data (Steering Committee and Epidemiological/Statistical experts) vouch for the accuracy and completeness of the data and analyses, and all authors vouch for the fidelity of the trial as well as the report of this study. Patients Patients were eligible if they were aged ≥18 years, had symptomatic chronic heart failure NYHA functional class II to ambulatory IV, despite optimal treatment for heart failure and had a LVEF ≤50% (ref. 11). They were required to have increased concentrations of NTproBNP. In the very few sites where this was not available, similarly elevated B-type natriuretic peptide (BNP) levels were allowed as previously described in ref. 11 and NTproBNP levels were estimated using a standard formula18. We enrolled patients in sinus rhythm and atrial fibrillation. The full list of inclusion and exclusion criteria is depicted in Supplementary Tables 2 and 3. All patients were on guideline-directed therapies for heart failure and atrial fibrillation at the start of the study19,20. Trial procedures Patients were randomly assigned in a 1:1 ratio to double-blind treatment of either (low-dose) digoxin or matched placebo. The target range for SDC was 0.5–0.9 ng ml−1, and patients received a starting dose of either 0.2 or 0.1 mg depending on age, kidney function, eGFR or interacting medication11. SDC was measured by a central core laboratory at 4 weeks after randomization and every 6 months thereafter (Extended Data Fig. 5). Dose adjustments were made centrally based on SDC measurements with dummy values for patients in the placebo group. Additional SDC measurements were collected 4 weeks after restart of the study medication, after dose adjustment or after the addition of interacting medication. SDC was also measured in case of a ≥30% decrease in eGFR and 2–4 weeks after a heart failure hospitalization. Follow-up visits were scheduled 4 weeks after randomization and every 6 months until the end of the study. Patients were contacted every 3 months in between by telephone. Endpoints The primary outcome was a composite of total worsening heart failure events, defined as total hospitalizations or total urgent hospital visits for worsening heart failure and death from cardiovascular causes. The ranked secondary outcomes were tested in the following hierarchical order: composite of total worsening heart failure events, composite of total cardiovascular events defined as cardiovascular hospitalization or urgent hospital visit (other than heart failure), time to first worsening heart failure event or death from cardiovascular causes, time to first worsening heart failure event or death from any cause, time to first worsening heart failure event, time to death from cardiovascular causes and time to death from any cause (Supplementary Table 4 and ‘Statistical analysis plan’). All primary and secondary endpoints and all serious adverse events were adjudicated by an independent, blinded Clinical Event Committee11. Statistical analysis Under the assumptions of a placebo event rate of 18 events per 100 person-years and a median follow-up of 36 months, we estimated that the enrollment of 982 patients yielding 472 primary outcome events would provide 80% power to detect a 22% reduction in the event rate of the composite of total worsening heart failure events and death from cardiovascular causes (primary outcome) with low-dose digoxin compared to placebo11. The primary analysis was performed using a modified intention-to-treat approach excluding randomized patients who never received study medication. The analysis of recurrent events, including the primary analysis, was conducted using the Lin Wei Yang Ying (LWYY) semi-parametric proportional rates model21. A two-sided P value of less than 0.05 was considered to indicate statistical significance. The primary outcome was assessed in 11 prespecified subgroups. The ranked secondary outcomes were tested in hierarchical order: (1) total hospitalizations and urgent hospital visits for heart failure (‘worsening heart failure events’), (2) total cardiovascular hospitalizations and urgent cardiovascular hospital visits, (3) time to first worsening heart failure event or cardiovascular death, (4) time to first worsening heart failure event or all-cause death, (5) time to first worsening heart failure event, (6) time to cardiovascular death and (7) time to all-cause death. The secondary outcomes were analyzed using the LWYY semi-parametric proportional rates model for recurrent-event outcomes and the Cox proportional hazards model using cause-specific hazard ratios for time-to-first event outcomes. Two additional (post hoc) analyses were performed owing to high discontinuation rate. The first was to examine the effect of the COVID pandemic, which is known to have had a major effect on heart failure trials13. The COVID pandemic had a significant effect on health care and the management of patients in the Netherlands for about 2 years, that is, from early 2020 until June 202222, and so this overlapped with the first 2 years of DECISION. Indeed, recent heart failure trials have shown that the COVID pandemic may affect outcomes23. To assess the effect of the COVID pandemic on the discontinuation rate during the first year in the trial, we used a cause-specific Cox regression adjusted for the competing risk of death. The second sensitivity analysis was a supportive on-treatment analysis, in which only events that occurred while patients were using study medication contributed toward the primary endpoint. Patients were censored after they discontinued study medication. The same LWYY model was applied for the on-treatment analysis as for the analysis of the primary trial outcome21. Additional information regarding the statistical analysis is provided in the ‘Statistical analysis plan’ in the Supplementary Information. Analyses were conducted using R (version 4.4.2) and STATA/SE 19.5. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Data availability The datasets generated and/or analyzed during this study are not publicly available, but anonymized participant data can be made available upon requests directed to the corresponding author. The steering committee of DECISION will review requests and will have the right to review and comment on any draft papers based on these data before publication. Data requests will be reviewed on the basis of scientific merit, overlap with any planned secondary analysis, ethical review, available resources and regulatory requirements, and a response can be expected within 2 weeks. Following approvals, unidentified patient data and a data dictionary will be provided. References Gheorghiade, M., van Veldhuisen, D. J. & Colucci, W. S. Contemporary use of digoxin in the management of cardiovascular disorders. Circulation 113, 2556–2564 (2006). The Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in patients with heart failure. N. Engl. J. Med. 336, 525–533 (1997). van Veldhuisen, D. J. & Bauersachs, J. Digitalis in heart failure: declining use and ongoing outcome trials. Eur. Heart J. 44, 1976–1978 (2023). Adams, K. F. et al. Clinical benefits of low serum digoxin concentrations in heart failure. J. Am. Coll. Cardiol. 39, 946–953 (2002). Rathore, S. S., Curtis, J. P., Wang, Y., Bristow, M. R. & Krumholz, H. M. Association of serum digoxin concentration and outcomes in patients with heart failure. J. Am. Med. Assoc. 289, 871 (2003). Dickstein, K. et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008‡. Eur. J. Heart Fail. 10, 933–989 (2008). Kotecha, D. et al. Effect of digoxin vs bisoprolol for heart rate control in atrial fibrillation on patient-reported quality of life. J. Am. Med. Assoc. 324, 2497 (2020). Bunting, K. V. et al. Low-dose digoxin improves cardiac function in patients with heart failure, preserved ejection fraction and atrial fibrillation—the RATE-AF randomized trial. Eur. J. Heart Fail. https://doi.org/10.1002/ejhf.70022 (2025). Bavendiek, U. et al. Digitoxin in patients with heart failure and reduced ejection fraction. N. Engl. J. Med. 393, 1155–1165 (2025). Cannata, A., Crespo-Leiro, M. G., Bromage, D. I., Ruschitzka, F. & McDonagh, T. A. Heart failure with reduced ejection fraction. Lancet 407, 529–542 (2026). van Veldhuisen, D. J. et al. Efficacy and safety of low-dose digoxin in patients with heart failure. Rationale and design of the DECISION trial. Eur. J. Heart Fail. 26, 2223–2230 (2024). van der Meer, P. et al. Heart failure patients with reduced or mildly reduced ejection fraction: baseline characteristics of the low-dose digoxin outcome DECISION trial. Eur. J. Heart Fail. https://doi.org/10.1093/ejhf/xuag073 (2026). Anker, S. D. et al. Conducting clinical trials in heart failure during (and after) the COVID-19 pandemic: an Expert Consensus Position Paper from the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Eur. Heart J. 41, 2109–2117 (2020). Rathore, S. S., Wang, Y. & Krumholz, H. M. Sex-based differences in the effect of digoxin for the treatment of heart failure. N. Engl. J. Med. 347, 1403–1411 (2002). Adams, K. F. et al. Relationship of serum digoxin concentration to mortality and morbidity in women in the Digitalis Investigation Group Trial. J. Am. Coll. Cardiol. 46, 497–504 (2005). Castagno, D., Petrie, M. C., Claggett, B. & McMurray, J. Should we SHIFT our thinking about digoxin? Observations on ivabradine and heart rate reduction in heart failure. Eur. Heart J. 33, 1137–1141 (2012). Abdul-Rahim, A. H. et al. Effect of digoxin in patients with heart failure and mid-range (borderline) left ventricular ejection fraction. Eur. J. Heart Fail. 20, 1139–1145 (2018). Kasahara, S. et al. Conversion formula from B-type natriuretic peptide to N-terminal proBNP values in patients with cardiovascular diseases. Int. J. Cardiol. 280, 184–189 (2019). Ponikowski, P. et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur. J. Heart Fail. 18, 891–975 (2016). Kirchhof, P. et al. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur. Heart J. 37, 2893–2962 (2016). Lin, D. Y., Wei, L. J., Yang, I. & Ying, Z. Semiparametric regression for the mean and rate functions of recurrent events. J. R. Stat. Soc. B 62, 711–730 (2000). Bos, J. C., De Boer, R. & Franz, E. Staat van infectieziekten in Nederland, 2023. Dutch National Institute for Public Health and the Environment https://rivm.openrepository.com/entities/publication/e571415a-ac27-43b8-a6ab-d6886d021013 (2024). Ponikowski, P. et al. Ferric carboxymaltose for iron deficiency at discharge after acute heart failure: a multicentre, double-blind, randomised, controlled trial. Lancet 396, 1895–1904 (2020). Acknowledgements We thank all patients, their caregivers, the nurses, physicians and all members of the DECISION study group involved in the trial. We acknowledge the funding provided by the Dutch Heart Foundation for the execution of the study (NHS:2018B024). Disphar International B.V., Tiofarma B.V. and TEVA Nederland B.V. provided the investigational product and financially supported the trial. UMCG cardioresearch foundation provided financial support for the execution of the trial. Author information Authors and Affiliations Contributions All authors contributed to the acquisition, analysis or interpretation of data for this work. All authors critically reviewed the work for important intellectual content. All authors approved the final version and all authors vouch for the fidelity of the trial as well as the report of this study. D.J.v.V., M.R., A.M., A.D.I.v.A., M.A., M.L.B., J.G.P.T. and P.v.d.M. conceived and/or designed the study. K.D., D.P., G.L., J.G.P.T. and G.H.D.V. performed the statistical analysis. Corresponding author Ethics declarations Competing interests D.J.v.V. is an established investigator of the Netherlands Heart Foundation, but has no conflicts of interest to report (and has received no personal payments in relation to the DECISION Trial). M.R. has received an unrestricted research grant from the Netherlands Heart Foundation, which is conducted in collaboration with and supported by the Dutch CardioVascular Alliance, 01-002-2022-0118 EmbRACE. He has also received unrestricted research grants from the Netherlands Cardiovascular Research Initiative: an initiative with support of the Netherlands Heart Foundation; RACE V (CVON 2014–9) and RED-CVD (CVON2017-11). He has also received an unrestricted research grant from the European Union’s Horizon 2020 research and innovation program under grant agreement EHRA-PATHS (945260). In addition, he has received consultancy fees from Bayer, InCarda Therapeutics and Novartis to the institution and speaker fees from Daiichi Sankyo and Pfizer, which all have been paid to his employer UMC Groningen. A.M. is the Chair of the Scientific Advisory Board and Chair of the Clinical Trial Committee of the Netherlands Heart Foundation, for which he receives no payments. He has no other conflicts of interest. M.A. reports no conflicts of interest. A.A.V. has received consultancy fees and/or research support from Anacardio, Astra Zeneca, Bayer AG, Boehringer Ingelheim, Cardurion, Corteria, EliLilly, Merck, Novartis, Novo Nordisk, QTsense, Rycarma and SalubrisBio, which all have been paid to his employer, UMC Groningen. K.D. reports speaker/consultancy fees from Abbott, Astra zeneca, Boehringer Ingelheim, Novartis, FIRE1 and Echosense, which all have been paid to his employer UMC Groningen. A.D.I.v.A. reports no conflict of interest. M.L.B. reports no conflicts of interest. J.S. has received payments for lectures and advisory boards from General Electric, Boehringer, Novo Nordisk, Novartis, Lilly, Astra Zeneca and Pharma Nord, not related to the submitted work. E.E.v.d.W. reports no conflicts of interest. H.J.G.M.C. has received fees and honoraria for lectures, education and scientific advice activities from Atricure, Medtronic and Armgo. D.J.T. reports no conflicts of interest. P.A.M.H. reports no conflicts of interest. J.E.C.V.d.S. has held paid lectures for Abbott; there are no other conflicts of interest. R.J.S. reports no conflicts of interest. A.v.d.S. reports no conflicts of interest. O.B. reports no conflicts of interest. T.J.R. reports no conflicts of interest. T.O. reports no conflicts of interest. G.L.B. reports no conflicts of interest. S.K. has received consultancy fees or educational grants from Astra Zeneca, Novo Nordisk, Boehringer Ingelheim, Novartis, CSL, Amarin, Bayer and Amgen. P.A.D. reports no conflicts of interest. G.C.M.L. reports no conflicts of interest. I.A. reports no conflicts of interest. H.G.R.D. reports no conflicts of interest. A.S. reports no conflicts of interest. M.E.W.H. reports no conflicts of interest. R.G.T. reports grants from Medtronic and Abbott, and personal fees from Boehringer Ingelheim, Daiichi Sankyo, Bayer and Pfizer/Bristol Meyer Squibb all outside the submitted work. D.J.A.L. reports no conflicts of interest. I.C.D.W. reports no conflicts of interest. M.A.T.V. reports no conflicts of interest. G.H.D.V. reports no conflicts of interest. A.H.d.V. has no conflicts of interest. E.L.M.-S. reports no conflicts of interest. D.P. reports no conflicts of interest. G.L. reports no conflicts of interest. J.G.P.T. reports no conflicts of interest. P.v.d.M. is supported by a grant from the European Research Council (ERC CoG 101045236, DISSECT-HF). He has received consultancy fees and/or grants from Novartis, Pharmacosmos, Vifor Pharma, Astra Zeneca, Pfizer, Pharma Nord, BridgeBio, Novo Nordisk, Bayer, Boehringer Ingelheim and Ionis, which all have been paid to his employer, UMC Groningen. Peer review Peer review information Nature Medicine thanks Muthiah Vaduganathan and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Michael Basson, in collaboration with the Nature Medicine team. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Extended data Extended Data Fig. 1 First worsening heart failure event or death from any cause. The Kaplan Meier plot for time to first worsening heart failure event or all-cause mortality. Extended Data Fig. 2 First worsening heart failure event. The Kaplan Meier plot for time to first worsening heart failure event. Extended Data Fig. 3 Death from any cause. The Kaplan Meier plot for time to all-cause mortality. Extended Data Fig. 4 HF hospitalization or CV mortality censored at discontinuation. The cumulative-incidence plot for the on-treatment repeated worsening heart failure events or CV mortality. Patients that discontinued during the trial were censored at the date they discontinued study medication. Extended Data Fig. 5 Digoxin dosing scheme. The DECISION digoxin dosing algorithm. SDC = Serum digoxin concentration. Supplementary information Supplementary Information (download PDF ) Supplementary Tables 1–4, Fig. 1, Statistical analysis plan and Protocol. Rights and permissions Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. About this article Cite this article van Veldhuisen, D.J., Rienstra, M., Mosterd, A. et al. Low-dose digoxin in patients with heart failure with reduced or mildly reduced ejection fraction: a randomized controlled trial. Nat Med (2026). https://doi.org/10.1038/s41591-026-04406-6 Received: Accepted: Published: Version of record: DOI: https://doi.org/10.1038/s41591-026-04406-6