Abstract
Objective
Early rhythm control has demonstrated clinical benefits even among patients with persistent atrial fibrillation. However, some patients continue to experience recurrent atrial tachyarrhythmias despite undergoing catheter ablation and treatment with antiarrhythmic drugs. This study aimed to assess the efficacy of digoxin in restoring sinus rhythm among patients with recurrent atrial tachyarrhythmias after catheter ablation and ongoing medical therapy.
Methods
Between 1 January 2013 and 1 September 2023, a total of 194 patients underwent catheter ablation for persistent atrial fibrillation. Among these, 96 (49.5%) patients experienced recurrence of atrial tachyarrhythmia during follow-up after catheter ablation. Of these, 12 patients with sustained arrhythmias despite at least 1 month of treatment with antiarrhythmic drugs and rate-control medications were treated with digoxin. Patients who achieved sinus rhythm after digoxin administration were categorized as the digoxin-effective group, and those who did not were classified as the non–digoxin-effective group. Clinical characteristics, medications, and echocardiographic parameters were compared between the two groups.
Results
Among the 12 patients analyzed (mean age: 66 ± 7 years; 3 males and 9 females), 7 (58.3%) achieved sinus rhythm after digoxin administration. Patients in the digoxin-effective group exhibited significantly smaller left atrial sizes than those in the non–digoxin-effective group (39.2 ± 3.3 mm vs. 46.8 ± 2.6 mm, p = 0.017). Most recurrences in the digoxin-effective group were atrial tachycardia episodes (5/7, 71.4%). The median duration from digoxin initiation to sinus rhythm restoration was 49 days (range: 21–210 days). No serious adverse effects occurred.
Conclusions
Digoxin effectively restored sinus rhythm in selected patients experiencing recurrent atrial tachyarrhythmias after catheter ablation and ongoing medical therapy, particularly among those with smaller left atrial sizes. Further research is warranted to validate these findings and explore the underlying mechanisms.
Introduction
Atrial fibrillation (AF) remains the most common sustained cardiac arrhythmia, with significant implications for morbidity and mortality. The 2024 European Society of Cardiology guidelines emphasize the importance of early rhythm control in AF, with catheter ablation (CA) established as a cornerstone in the rhythm-control strategy for AF, particularly for patients who remain symptomatic despite antiarrhythmic drug (AAD) therapy.1,2 However, postablation recurrence of atrial tachyarrhythmias is common, particularly among patients with persistent AF (PerAF) or structural heart disease, 3 necessitating further rhythm- or rate-control interventions. 4 In such cases, additional pharmacological strategies, including AADs and rate-control agents, are often required. 4 Among these agents, digoxin, a historically significant but controversial therapy, has garnered renewed interest in the postablation setting. 5
Digoxin, a cardiac glycoside, primarily exerts its effects by increasing vagal tone and reducing atrioventricular (AV) nodal conduction. These properties make it a valuable rate-control agent, particularly in patients with heart failure or left ventricular dysfunction. 6 However, its role in rhythm control postablation remains less well-defined.
Recent studies suggest that digoxin, when used in combination with amiodarone or β-blockers, may enhance rhythm stabilization in patients with recurrent atrial arrhythmias following ablation. 4 A study demonstrated that a triple-drug regimen (comprising digoxin, amiodarone/propafenone, and β-blockers) resulted in a higher rate of sinus rhythm (SR) reversion within 1 month compared with standard therapy. Digoxin remains a viable option for selected patients with recurrent atrial arrhythmias post-CA, particularly when used in combination with other agents. This investigation aimed to assess the efficacy of digoxin in restoring SR in patients with recurrent atrial tachyarrhythmia following CA who were already receiving AADs and rate-control medication.
This study explores the role of digoxin in the post-CA management of recurrent atrial arrhythmias. Specifically, it evaluates its effectiveness in restoring and maintaining SR among patients who remain symptomatic despite initial ablation and AAD therapy.
Materials and methods
Trial design and study participants
This retrospective cohort study evaluated the efficacy of digoxin in restoring SR in patients with recurrent atrial tachyarrhythmia, including AF, atrial tachycardia (AT), or atrial flutter (AFL), following CA for PerAF. PerAF was defined as AF persisting for more than 7 days without SR recovery. The distinction between AT and AFL is based on a rate threshold of approximately 240–250 beats per minute, and the existence of isoelectric baselines between atrial deflections in AT, which are absent in AFL. 7 Between 1 January 2013 and 1 September 2023, a total of 194 patients received PerAF CA. Among these, 96 (49.5%) patients experienced recurrence of atrial tachyarrhythmia during follow-up after CA. Following the administration of at least one type of AAD and one rate-control medication, 84 (87.5%) patients achieved SR restoration, while 12 patients did not.
Interventions
All patients received digoxin therapy after failure to achieve SR restoration with at least one AAD and one rate-control medication following CA. AAD therapy was continued for a minimum of 3 months postablation and discontinued if no recurrent atrial tachyarrhythmia occurred during this period. Digoxin was initiated only in patients who experienced recurrent atrial tachyarrhythmias that proved refractory to both AAD therapy and standard rate-control medications despite at least 1 month of use. Digoxin was administered at a dose of 0.125 mg once daily, and serum digoxin levels were measured if digoxin toxicity was clinically suspected. In such cases, blood samples were collected within 6–24 h after digoxin administration, in line with standard pharmacokinetic distribution timing. Digoxin was continued until 1 month after SR recovery or until bradycardia episode or digoxin toxicity was observed. The types and dosages of concomitant AADs or rate-control agents were recorded. During digoxin therapy, previously prescribed AADs (amiodarone, propafenone, or dronedarone) were maintained at the same dosage unless contraindications or adverse effects developed. The interval between the recurrence of atrial tachyarrhythmia and the administration of digoxin as well as the time required for SR recovery following digoxin initiation were documented. No patients underwent repeat ablation or electrical cardioversion during the study period. Echocardiographic parameters, including left atrial (LA) size, were assessed before digoxin therapy (Figure 1).
Schematic illustration of postablation management and response to antiarrhythmic drug (AAD) therapy in recurrent atrial fibrillation (AF). Following catheter ablation for AF, patients initially received either rhythm-control or rate-control medications. Upon recurrence of AF, treatment was escalated with additional AADs. In patients refractory to initial escalation, a triple combination therapy was implemented by adding digoxin. Clinical responses varied, with some patients achieving effective rhythm control (“responders”) and others remaining refractory (“nonresponders”). The analysis demonstrated that nonresponders exhibited a significantly larger left atrial (LA) size than responders (bottom right). A representative electrocardiogram from one responder illustrates ongoing AFL despite intensive AAD therapy (bottom left (a)); the patient restored sinus rhythm after 3 weeks of digoxin therapy (bottom left (b)).
Outcomes
The primary outcome was the restoration of SR following digoxin administration, confirmed by electrocardiographic monitoring during outpatient department follow-up in the first week and monthly thereafter. Patients were classified based on their response into the digoxin-effective (hereafter referred to as effective group) or non–digoxin-effective (hereafter referred to as noneffective group) groups. The secondary outcome was the occurrence of any adverse events. Baseline characteristics, comorbidities, echocardiographic data, and medication use were analyzed to identify potential associations with digoxin response.
Consent
As this was a retrospective study utilizing deidentified medical records, the requirement for individual patient consent was waived in accordance with the Chang Gung Medical Foundation Institutional Review Board guidelines (institutional review board number: 202500212B0). The study protocol complied with the ethical principles outlined in the Declaration of Helsinki of 1975, as revised in 2024, and was approved by the local ethics committee. All patient data were fully deidentified to ensure anonymity and prevent any possibility of identification.
Reporting guideline
The reporting of this observational study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. 8
Statistical analysis
Continuous variables were expressed as mean ± SD or median (interquartile range) and compared using the Student’s t-test or Mann–Whitney U test, depending on normality distribution. Categorical variables were expressed as percentages and compared using the χ2 test or Fisher’s exact test, as appropriate. A p-value <0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version 30, IBM Corp.).
Results
Ultimately, 12 patients who experienced sustained AF/AT/AFL despite at least 1 month of medical therapy were treated with digoxin and included in the analysis (Table 1). The study population included three male (25%) and nine female patients, with a mean age of 66 ± 7 years. The median duration of arrhythmia recurrence post-CA was 33 (7.5–61.8) days. Of these, seven (58.3%) patients had recurrent AF, while five (41.7%) experienced AFL or AT. All patients had previously undergone an initial PerAF ablation consisting of pulmonary vein isolation with an added cavotricuspid isthmus line. In addition, four (33%) patients developed further LA lesions: two had an anterior LA line (one of whom also had a roof line), one underwent a posterior wall box isolation, and one had a perimitral isthmus line.
Baseline characteristics of the total population and response to digoxin.
AFL: atrial flutter; AT: atrial tachycardia; CA: catheter ablation; CTI: cavotricuspid isthmus; F: female; LA: left atrial; M: male; PMI: perimetral isthmus; PVI: pulmonary veins isolation; SR: sinus rhythm.
SR recovery was achieved in seven patients, representing a recovery rate of 58.3%. Among patients with recurrent AT/AFL, the SR conversion rate was 80% (4/5), while in those with recurrent AF, it was only 43% (3/7). Based on their response to digoxin, patients were categorized into two groups: effective group (n = 7), comprising those who successfully converted to SR, and noneffective group (n = 5), consisting of those who did not achieve rhythm restoration. The clinical factors influencing digoxin effectiveness were analyzed, revealing no significant differences between the two groups in terms of age, sex distribution, comorbidities, and CHA2DS2-VASc scores, suggesting comparable baseline cardiovascular risk profiles. At baseline, nine (75%) patients were on propafenone (150 mg every 8 h), two (17%) on amiodarone (200 mg twice daily), and one (8%) on dronedarone (400 mg twice daily). All these agents were continued concomitantly with digoxin throughout the observation period. The arrhythmia recurrence patterns after CA differed between the two groups. The noneffective group was more likely to experience recurrent AF after CA (4/5, 80%), whereas the effective group exhibited more frequent episodes of AT or AFL after CA (4/7, 57.1%). Interestingly, all four patients who had recurrent rhythm with AFL responded to digoxin therapy. Furthermore, the interval between CA and arrhythmia recurrence was longer in the effective group (37 (6–150) days), while the noneffective group had a much shorter recurrence period post-CA (27 (8.5–34.5) days), although this difference was not statistically significant. The interval between arrhythmia recurrence and digoxin use did not significantly differ between the two groups. In the effective group, this duration was 1 (1–150) days, compared with 1 (1–9.5) day in the noneffective group. The interval between digoxin use and SR recovery did not significantly differ between the two groups. In the effective group, the duration was 49 (21–210) days, compared with 120 (23–175) days in the noneffective group (Table 2).
The demographics and clinical characteristics of all 12 patients and those with/without response to digoxin treatment.
Data are expressed as means ± SD, median (interquartile range), or % (n), as appropriate.
AF: atrial fibrillation; AT/AFL: atrial tachycardia/atrial flutter; CA: catheter ablation; IVS: interventricular septum; LA: left atrial; LVEF: left ventricular ejection fraction; SR: sinus rhythm.
Echocardiographic findings revealed that the LA size was significantly larger in the noneffective group (47.4 ± 5.0 mm) than in the effective group (39.1 ± 5.7 mm, p = 0.026). No significant differences were observed in left ventricular ejection fraction (effective group: 57.7% ± 11.9%, noneffective group: 64.2% ± 12.4%; p = 0.382) or interventricular septal thickness (effective group: 11.6 ± 2.1 mm, noneffective group: 11 ± 2.7 mm; p = 0.687) (Figure 1).
Baseline renal function was similar between the two groups (mean serum creatinine: 0.9 ± 0.2 vs. 1.0 ± 0.4 mg/dL, p = 0.437). Routine serum digoxin concentrations were not assessed in all patients; however, no patient exhibited any clinical signs of digoxin toxicity, and no abnormally high digoxin levels were detected when tested (Table 2).
There was no significant side effect of digoxin use in our patients, including symptomatic tachy- or bradyarrhythmia, gastrointestinal adverse effect, consciousness disturbance, and vision disturbance, during a mean follow-up time of 105 days. Only one patient who had achieved SR recovery developed an asymptomatic first-degree AV block after 7 months of digoxin therapy. In this case, we discontinued the use of digoxin.
Discussion
The study yielded several key findings. First, digoxin use demonstrated a satisfactory effect on SR recovery in patients who had undergone CA for AF and were already receiving at least one rhythm- and rate-control drug. There was no significant adverse effect observed in this study during an average period of 3 months of usage. Second, digoxin responded well in the relatively smaller LA size regarding SR recovery. Third, digoxin effectiveness appeared to be associated with the recurrent rhythm of AT or AFL after CA (57.1% in the effective group compared with 20% in the noneffective group), although without achieving a statistically significant difference between both the groups. These findings suggest that digoxin therapy responded well in the AT or AFL rhythm rather than the AF rhythm.
Digoxin in the management of AF and postablation AF
Traditionally, digoxin has been used to control ventricular rate in AF patients, especially those with concomitant heart failure.9–11 Digoxin is frequently utilized as a rate-control agent in AF, primarily in patients with heart failure with reduced ejection fraction who cannot tolerate β-blockers due to hypotension.12,13 By inhibiting the Na+/K+ ATPase pump, digoxin increases intracellular calcium levels, thereby enhancing myocardial contractility while simultaneously slowing AV nodal conduction through increased vagal activity. This dual mechanism renders digoxin particularly useful in patients requiring rate control while maintaining adequate cardiac output.
However, digoxin may contribute to rhythm control in certain clinical scenarios, especially in patients who have undergone CA. By slowing AV conduction and increasing vagal tone, the release of norepinephrine from the sympathetic terminals is expected to be inhibited by the interaction between the sympathetic and parasympathetic nervous systems. Digoxin may help stabilize the ventricular response and potentially reduce AF triggers and rotors. 14 When combined with other antiarrhythmic measures, digoxin may facilitate the maintenance of SR. For instance, a prospective study reported that adding low-dose digoxin to an AAD regimen improved short-term reversion to SR in patients with recurrent arrhythmias after ablation (44.8% reversion at 1 month vs. 6.3% in those without digoxin). 4 This suggests that although digoxin alone is not a frontline cardioversion drug, it may exert a supportive rhythm-stabilizing effect postablation when used in combination with other therapies. In this study, the SR reversion rate was 58.3% with add-on digoxin therapy in patients already receiving at least one kind of AAD and rate-control drug. These results corroborate the previous study findings that a triple-therapy regimen was effective in converting SR in patients with recurrence of atrial tachyarrhythmia after CA of PerAF. In our cohort, digoxin was administered as an add-on therapy while ongoing AADs were maintained. The combined use of digoxin and AADs such as amiodarone or propafenone may have contributed to rhythm stabilization through complementary electrophysiologic mechanisms. However, this overlap also limits the ability to attribute SR restoration exclusively to digoxin. Future studies should include stratified analysis to distinguish the independent effect of digoxin from that of concomitant AADs.
LA size and digoxin efficacy
LA size is a well-established determinant influencing AF maintenance and treatment success. An enlarged LA offers a larger surface area and substrate for multiple reentrant wavelets, promoting AF persistence. Conversely, a smaller LA has less atrial mass and potentially fewer regions of fibrosis or conduction heterogeneity capable of sustaining atrial arrhythmias. Clinical studies have shown that larger LA dimensions are strongly associated with higher AF recurrence rates in patients on rhythm-control therapy. 15 For example, the AFFIRM trial and subsequent analyses demonstrated that incremental increases in LA size significantly predicted AF recurrence despite AAD therapy.5,16 LA dilation often reflects structural remodeling and fibrosis, resulting in an abnormal atrial substrate that resists stable SR. 17 Thus, patients with smaller atria generally possess a more favorable arrhythmic substrate—characterized by less fibrosis and fewer reentry circuits—and are more likely to maintain SR.
In the context of digoxin efficacy, this study implicated that LA size may modulate the effectiveness of this therapy for rhythm control. Digoxin’s indirect antiarrhythmic effect (mediated through enhanced vagal tone) might be sufficient to control or terminate arrhythmias in hearts with minimal remodeling. A smaller LA, with less extensive abnormal substrate, may not sustain AF as readily; thus, by slowing the heart rate and increasing AV nodal refractoriness (digoxin’s primary actions), SR restoration might resume. In other words, when the atrial structure is not severely diseased, even modest rhythm-slowing or rate-controlling interventions may be sufficient to terminate the arrhythmia. This finding aligns with the observation that structural factors are critical, with LA size serving as an indicator of atrial “health.” If the LA is not significantly enlarged, the patient’s arrhythmia is more likely driven by trigger firing (e.g. acute pulmonary vein triggers) rather than a broad fibrotic network.18,19 In such cases, digoxin may help suppress these triggers or prevent rapid conduction, thereby aiding rhythm control. Conversely, in patients with a markedly enlarged LA with diffuse fibrosis, AF may continue despite rate control because multiple rotors persist. 15 This study suggested that patients with smaller LA size responded better to digoxin, likely because their atrial substrate was less abnormal. These findings are consistent with the broader understanding that a less dilated atrium corresponds to a higher chance of maintaining SR with medical therapy. 15 Therefore, assessing LA size may help predict which patients are more likely to benefit from therapies such as digoxin for postablation rhythm management.
Digoxin in AFL and AT recurrences
Interestingly, the type of recurrent arrhythmia postablation may influence the effectiveness of digoxin in its management. AFL and AT are typically more organized arrhythmias compared with AF, which exhibits chaotic activity. Clinically, we often observe that AFL or AT recurrences are more amenable to treatment, as they can frequently be terminated with a single shock or suppressed with medications. 20 This is largely because such arrhythmias typically involve a single dominant reentrant circuit or focus. Digoxin’s electrophysiological effects (namely, slowing AV nodal conduction and increasing vagal tone) can significantly impact AFL/AT by reducing the ventricular response and sometimes influencing the atrial cycle length. In typical AFL, for instance, the atrial rate is approximately 300/min, usually with 2:1 AV conduction, resulting in rapid ventricular rates and symptoms. Digoxin can slow the ventricular response (e.g. converting a 2:1 block to 4:1), improving hemodynamic stability and symptom burden. Although merely slowing the ventricular rate does not always terminate the flutter, the increased vagal tone might occasionally interrupt the reentrant circuit or at least make AFL easier to cardiovert with minimal intervention. Our observations indicate that patients whose arrhythmia recurred as AFL showed better response to digoxin than those with recurrent AF. In practice, digoxin often controlled the heart rate well and, in some cases, was associated with SR conversion in AFL patients. This difference likely reflects the underlying mechanisms: AF is maintained by multiple wavelets and often requires more potent antiarrhythmics or repeat ablation, whereas a single-circuit tachycardia might self-terminate once AV nodal conduction is slowed and the cardiac autonomic balance shifts.
Safety of digoxin use in postablation patients
Safety remains a key consideration when incorporating digoxin into a rhythm-control regimen, given the historical concerns about its adverse effects. Notably, the findings of this study indicate that digoxin was well tolerated among the postablation patient population. There were no occurrences of severe bradycardia, high-grade AV block, or other digoxin-related toxicities observed during follow-up. In a similar study combining low-dose digoxin with other antiarrhythmics, no patients developed symptomatic bradyarrhythmia requiring intervention, 4 underscoring that cautious use of digoxin did not introduce additional short-term risk. This real-world safety profile aligns with evidence derived from randomized trials. A comprehensive meta-analysis of 28 trials found no significant difference in short-term all-cause mortality or serious adverse events between digoxin and control treatments. 21 In summary, under controlled trial conditions, digoxin did not increase adverse outcomes compared with placebo or other interventions over the study periods. Our clinical experience reflects that with proper dosing and monitoring, digoxin does not adversely affect hemodynamics or lead to proarrhythmia in the acute postablation phase.
It is important to contextualize these findings with existing literature. Although randomized trials suggest neutral short-term safety, several observational studies have raised concerns about digoxin, associating it with increased mortality in AF patients. 21 This discrepancy is likely due to confounding factors, as sicker patients (e.g. those with heart failure or PerAF) are more frequently prescribed digoxin, which might explain the higher mortality observed in registry studies. Our focused cohort demonstrated no evidence of harm from digoxin, supporting the notion that digoxin can be safe when judiciously prescribed to appropriate patients (such as those without pre-excitation, with normal renal function, and at low risk for toxicity). Concomitant medications (like beta-blockers or amiodarone, if used) were carefully managed to prevent excessive AV nodal depression. Thus, in the setting of post-AF ablation management, a low-dose digoxin strategy appears to carry minimal risk and was not associated with complications. Patients should, of course, be monitored for signs of toxicity; however, our results contribute to the evidence that digoxin, despite its controversial reputation, can be used safely to aid in controlling atrial arrhythmias during the short-term postablation period.
Study limitations and need for prospective trials
Although these findings and hypotheses are intriguing, our study has several important limitations. First, the sample size was relatively small, limiting the power and generalizability of our conclusions. The absence of a control group of patients not receiving digoxin makes it difficult to confirm that digoxin was the determining factor in restoring SR. Without a comparator group, we cannot be certain whether the observed improvements in SR were due to the drug or to other unrelated factors (including the possibility of spontaneous reversion). Additionally, patients in our cohort who received digoxin tended to have relatively smaller LA dimensions, a factor known to favor the restoration of SR. This raises the possibility that the higher rate of SR conversion observed in the digoxin-treated group may be attributable to these favorable baseline anatomical characteristics rather than to an intrinsic pharmacological effect of digoxin. Taken together, these limitations warrant caution in interpreting the apparent benefits of digoxin. Future studies incorporating appropriate control groups and accounting for LA size are needed to definitively establish the efficacy of digoxin in restoring SR.
Besides, IoT technologies now enable continuous remote monitoring of AF, offering real-time heart rate and rhythm data beyond conventional clinic visits. 22 The RATE-AF trial demonstrated that patients receiving low-dose digoxin or β-blockers achieved comparable rate control when monitored using consumer-grade wearable devices. 23 Integrating IoT-based monitoring with digoxin therapy may enhance safety, enable timely dose adjustments, and facilitate personalized patient care.
Conclusions
This study demonstrated that digoxin exhibits promising potential for restoring SR in patients with PerAF who experience recurrent tachyarrhythmia following CA and treatment with AADs. Notably, the therapeutic response to digoxin was more favorable in patients with a smaller LA size, suggesting that atrial remodeling may play a role in its efficacy. These findings highlight the need for further investigation into the clinical utility of digoxin in postablation rhythm management. Future studies should aim to validate its efficacy in larger, prospective cohorts and elucidate the precise mechanisms through which digoxin facilitates SR restoration, particularly in relation to atrial electrophysiological properties and structural remodeling.
Supplemental Material
sj-pdf-1-imr-10.1177_03000605251395565 - Supplemental material for Impact of digoxin on sinus rhythm restoration after catheter ablation of persistent atrial fibrillation and drug-refractory atrial tachyarrhythmia recurrence: A cohort study
Supplemental material, sj-pdf-1-imr-10.1177_03000605251395565 for Impact of digoxin on sinus rhythm restoration after catheter ablation of persistent atrial fibrillation and drug-refractory atrial tachyarrhythmia recurrence: A cohort study by Yu-Sheng Lin, Hsiang-Hsuan Chen, Bang-Yan Hou, Hui-Ting Wang and Yung-Lung Chen in Journal of International Medical Research
Footnotes
Acknowledgments
Not applicable.
Author contributions
Y.-S.L., H.-H.C., and Y.-L.C. led the conception and design of study and revised the draft of the manuscript. B.-Y.H. and H.-T.W. collected the research data and prepared the draft of the manuscript. B.-Y.H. and H.-T.W. performed clinical works and organized the collected data. Y.-L.C. and H.-T.W. performed the statistical analysis and drafted the manuscript. Y.-L.C. and H.-T.W. supervised and validated the clinical works and results. All authors have read and agreed to the published version of the manuscript.
Data availability
The datasets used during the current study are available from the corresponding author on reasonable request.
Declaration of conflicting interests
The authors declare no conflict of interest.
Funding
No funding.
Institutional review board statement
As this was a retrospective study utilizing de-identified medical records, individual patient consent was waived in accordance with institutional review board guidelines (institutional review board number 202500212B0). The study protocol complied with ethical principles outlined in the Declaration of Helsinki and was approved by the local ethics committee.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
