GLP-1 drugs versus other weight loss methods and their link to atrial fibrillation returning after catheter treatment

A systematic review and meta-analysis have been performed according to the PRISMA guidelines [8] to investigate the effect of GLP-1 RA versus other weight loss interventions on the risk of AF recurrence after ablation in overweight or obese patients. “Exercise”, “Diet”, “Risk factor modification”, “Lifestyle management”, “Weight loss”, “Overweight”, “Obesity”, “Obese”, “Surgery”, “Glucagon-like peptide 1 receptor agonists”, “Exenatide”, “Lixisenatide”, “Liraglutide”, “Dulaglutide”, “Semaglutide”, “Albiglutide”, “Efpeglenatide” and “Atrial fibrillation ablation” were used as the keywords in the search process. There were no restrictions on publication year. Studies published in English, French, German, Italian, Spanish, and Chinese languages were included. Eligible studies involved patients using GLP-1 RA or patients who were overweight or obese and received other perioperative weight loss interventions before or after AF ablation, compared with a control group receiving standard treatment or no weight loss intervention. We included randomised controlled trials (RCTs) and cohort studies in the analysis, excluding case reports, reviews, and cross-sectional studies. The primary outcome was AF recurrence post-ablation with different weight loss interventions, with the secondary outcome being the timing of weight loss intervention.

The literature search was performed on PubMed, Scopus, Embase, and Cochrane Library, with the last search conducted on 09 Oct 2024. An additional search was conducted on Heart Rhythm to include relevant conference abstracts. All search was performed twice by two independently investigators (H.Y.L. and T.K.C.), and further verified by a third senior investigator (Y.H.C.). All included studies were reviewed independently and assessed for quality. STROBE checklist [9] was used for the quality assessment of cohort studies, while the CONSORT checklist [10] was used to assess the quality of randomised controlled trials (RCTs). Studies that met less than 50% of the checklist items would be excluded from the analysis to avoid the inclusion of poor-quality studies.

As this is a meta-analysis of published studies, there is no direct patient clinical contact. Therefore, no specific ethical approval was obtained or deemed necessary.

Data regarding AF recurrence 12 months post-ablation, methods of weight-loss intervention and initial and completion time of interventions were extracted from each study. Weight loss interventions were grouped into three categories: GLP-1 receptor agonists (GLP-1 RA), Lifestyle risk factor modification (LRFM), and bariatric surgery. GLP-1 RA involves the perioperative use of medications such as Semaglutide, Liraglutide, and other agents that target glucagon-like peptide-1 receptors to facilitate weight loss [11–14]. LRFM focuses on promoting a healthy lifestyle and managing cardiovascular disease (CVD) risk factors, including hypertension, diabetes, and sleep apnoea. Studies in this category implemented various strategies, such as intermittent fasting, lifestyle counselling, physical training, and professional-led risk factor modification clinics to support participants in achieving weight loss [15–21]. In one study (Jia et al.) [22], the weight loss intervention used was not specified. It is assumed that patients at least partially adopted lifestyle modifications to achieve weight loss. Bariatric surgery was defined as a surgical weight loss method, particularly used in morbidly obese patients, that was conducted pre-ablation to reduce weight and improve comorbidities prior to ablation [23].

For studies with follow-up periods that did not correspond to 12 months, recurrence data closest to this duration were collected to ensure more accurate comparisons between studies. Additionally, baseline characteristics and BMI at the end of the follow-up period were also collected and summarised.

The primary outcome was defined as recurrence of clinical AF, documented AF or atrial arrhythmia with a duration ≥ 30 s, or a requirement for further rhythm control post-ablation including new initiation of class I/III anti-arrhythmic drug therapies, cardioversion, or further ablation.

All analyses used the “meta” package in R Studio version 2024. A random-effects model using the Inverse Variance method was conducted to derive risk ratio (RR) with 95% confidence intervals (CI) estimates of AF recurrence. Between-study variance (τ²) was estimated using the Restricted Maximum Likelihood (REML) method to account for heterogeneity. Heterogeneity between studies was evaluated using the I² statistic and τ², with I² > 50% and τ² > 0.1 suggesting a significant heterogeneity. Forest plots were drawn to visualise the effects of each outcome.

Subgroup analysis was done using a Chi-squared test for group differences to assess the effects on the types and time of weight-loss interventions. Meta-regression analyses were also conducted to assess factors that affect the heterogeneity in outcomes between studies. Sensitivity analysis was conducted using the leave-one-out method and exclusion of conference abstracts to assess the influence of the individual study. Publication bias was assessed using funnel plots and Egger’s test.

A total of 13 studies were included in the analysis, grouped by intervention type to assess the impact on AF recurrence. The overall pooled risk ratio (RR) across all weight-loss interventions was 0.64 (95% CI: 0.47–0.87), indicating a statistically significant reduction in AF recurrence post-ablation (Graphical Abstract). There was high heterogeneity across studies (I² = 86 & τ² = 0.23). The differences between different types of interventions were non-significant (p = 0.28). Comparisons between weight loss and non-weight loss groups (after excluding two studies that did not mention weight change) showed that successful achievement of weight loss, through whichever kind of intervention, was associated with a significant reduction in AF recurrence (RR: 0.63, 95% CI: 0.44–0.89). Study heterogeneity remained high (I² = 81% & τ² = 0.24) (Fig. 1).

Four studies examined the effect of GLP-1 RA on AF recurrence post-ablation, resulting in a pooled RR of 0.63 (95% CI: 0.35–1.15). Overall, there was a trend toward reduced AF recurrence in the intervention group, though the result did not reach statistical significance. Satti et al. [11] reported no difference between intervention and control groups, while Tabaja et al. [12] and Goldberger et al. [13] observed a non-significant reduction. In contrast, Zou et al. [14] showed a significant reduction in AF recurrence with the perioperative use of GLP-1 RA. This subgroup demonstrated significant heterogeneity (I² = 73% & τ² = 0.24), indicating the outcome between studies varied. It should be acknowledged that although the study by Goldberger et al. used GLP-1 RA (liraglutide) for their patients, the exposure for comparison was pre-ablation weight loss of < 3% versus 3–10%, and therefore the exact attributable fraction owing to GLP-1 RA remained uncertain as of the time of this manuscript submission.

Eight studies used LRFM as the weight-loss intervention, with a pooled RR of 0.68 (95% CI: 0.46–1.03). The analysis indicated a reduction in AF recurrence with LRFM to achieve weight loss, though the result was not statistically significant. This subgroup displayed a high heterogeneity (I² = 86% & τ² = 0.27), suggesting high variability between studies.

Only one study investigated the effect of bariatric surgery on AF recurrence post-ablation, with a significantly reduced RR of 0.31 (95% CI: 0.13–0.75), indicating that bariatric surgery is associated with a significant reduction in AF recurrence compared to the control group.

The studies were stratified by the time of weight loss intervention, with the bariatric surgery study [23] excluded due to its one-time intervention nature and Zou et al. [14] excluded for unspecified intervention timing. This resulted in 10 studies for subgroup analysis on initiation timing of interventions, excluding Tabaja et al. [12], which involved patients starting on GLP-1 RA pre- and post-ablation (Fig. 2). Studies with pre-ablation interventions showed a pooled RR of 0.72 (95% CI: 0.43–1.19) for AF recurrence, while post-ablation interventions resulted in an RR of 0.66 (95% CI: 0.38–1.12), with no significant subgroup difference (p = 0.82), which suggested the initiation time of weight-loss interventions did not affect the ablation outcome.

The studies were further divided based on the sustained continuation of weight loss intervention during the follow-up period. Eight studies were included after excluding three retrospective studies with unspecified completion times (Fig. 3). In studies with ongoing weight loss interventions during the follow-up period, a significant reduction in AF recurrence was observed (RR 0.65, 95% CI: 0.43–0.97). In contrast, studies without sustained interventions in the follow-up period showed no significant association with AF recurrence (RR 1.18, 95% CI: 0.80–1.74), with significant between-groups difference (p = 0.03). These suggest that sustained weight loss interventions in the follow-up period can improve ablation outcomes.

Meta-regression analyses were conducted to explore the influence of magnitude of weight loss and clinical characteristics on the study outcome.

Importantly, the magnitude of weight loss, measured from baseline to the end of the follow-up period, significantly predicted risk of AF recurrence (RR 0.94, 95% CI: 0.90—0.99) (Fig. 4). This result also indicated that for each 1% absolute decrease in body weight achieved, there was a 6% relative risk reduction in AF recurrence. This effect was consistent regardless of the choice of weight loss intervention, and helped to account for the variance in effect size observed across studies.

Based on the meta-regression, further sensitivity analyses were performed to further characterize the effect of magnitude of weight loss (Fig. 5). Unlike the meta-regression, which included six studies with detailed baseline and follow-up weight data, the subgroup analysis included 10 studies and stratified them by weight loss magnitude (< 5%, 5–10%, and > 10%). Results displayed a trend of progressively greater risk reduction in AF recurrence with increasing weight loss magnitude. The subgroup achieving > 10% weight loss demonstrated the most significant risk reduction (RR = 0.56, 95% CI: 0.32–0.97), and the 5–10% group also showed a favourable trend (RR = 0.68, 95% CI: 0.34–1.35), while the < 5% weight loss group had the least impact on reducing AF recurrence (RR = 0.78, 95% CI: 0.36–1.70).

In conjunction, other clinical characteristics including age, sex, AF type, comorbidities (e.g. hypertension, diabetes and obstructive sleep apnea), cardiac function by LVEF, baseline BMI and BMI at the end of follow-up were each assessed at separate meta-regression models. However, none of these variables showed a statistically significant association with the outcome.

A leave-one-out analysis was done to assess the validity of the outcome. After taking out each study at a time, the pooled RR remained statistically significant, and the heterogeneity remained high (I² = 75.5% to 87.6%), suggesting no significant influence from any single study. A funnel plot was drawn to assess publication bias, revealing no significant asymmetry, aligned with the Egger test with a p-value of 0.0848, suggesting no significant evidence of publication bias. Sensitivity analysis was also done by excluding the three conference abstracts, which resulted in a statistically significant pooled RR of 0.68 (95% CI: 0.48–0.98), suggesting the inclusion of these three conference abstracts was justified and supported the findings regarding the effectiveness of weight-loss interventions in reducing AF recurrence.

To our knowledge, this meta-analysis is the first to comprehensively evaluate the effects of different weight loss interventions on the recurrence of AF after catheter ablation in obese patients. Our findings demonstrate a significant reduction in AF recurrence, with a pooled RR of 0.64 (95% CI: 0.47–0.87) among those who underwent weight-loss interventions. Obesity is known to increase blood volume and cardiac output, leading to left atrial enlargement, left ventricular hypertrophy and elevated pulmonary pressures, thus increasing the risk of AF occurrence [24–26]. Additionally, metabolic changes associated with obesity, such as insulin resistance, hypertension, and activation of the renin–angiotensin–aldosterone system (RAAS), contribute to atrial remodelling, further elevating the risk of AF [24, 25]. Obesity-related conditions, such as obstructive sleep apnoea, also increase AF susceptibility by causing recurrent cycles of hypoxia, inflammation, and abnormal autonomic responses [27]. Another contributing factor is epicardial adipose tissue in obese individuals, which releases inflammatory mediators and adipokines, which can lead to myocardial fibrosis and electrical disturbances, triggering AF through paracrine signalling pathways [28–31].

In obese AF patients, weight loss has been shown to significantly reduce AF incidence and severity [4, 18, 32] through the direct effects on cardiac structure and function and indirect effects from the correction of other cardiovascular risk factors [24, 25]. Weight reduction in obese patients has also been associated with decreased left atrial volume and left ventricular wall thickness, thus reducing AF risk [4, 18, 33]. Previous meta-analyses found that achieving sustained weight loss, particularly > 10%, significantly reduced AF recurrence after ablation [34, 35]. We also found that greater weight loss had better ablation outcomes in the meta-regression and subgroup analysis. Our findings along with these studies, highlights the importance of weight management as an essential perioperative care for obese AF patients.

From our study, GLP-1 RA and LRFM each showed a trend towards reduced AF recurrence, but did not reach statistical significance. In comparison, bariatric surgery showed a significant reduction in AF recurrence post-ablation. Bariatric surgery is well-recognized for sustainable and long-term weight loss [36]. Previous studies supported the potential benefit of bariatric surgery in AF prevention among morbidly obese patients. A significant reduction in AF incidence was found among patients undergoing bariatric surgery compared to those receiving other medical treatment [37, 38]. These findings suggested a long-term benefit of bariatric surgery in lowering AF risk in obese patients. Donnellan et al. [23] further showed the additional advantage of bariatric surgery for patients undergoing AF ablation procedures by improving ablation outcome through effective and sustainable weight management.

GLP-1 RA may prevent AF by reducing atrial remodelling [39] through enhanced myocardial metabolism, stabilised blood glucose, and regulating the RAAS in animal studies [40, 41]. However, human studies have shown inconclusive results regarding GLP-1 RA’s clinical efficacy in preventing AF. Prior meta-analyses of RCTs comparing GLP-1 RA to placebo in obese patients showed statistically non-significant reductions in the occurrence of AF by [42, 43]. In contrast, a meta-analysis of 6,031 individuals found a significant reduction in AF recurrence at 12 months post-ablation among obese patients using GLP-1 RA (Hazard Ratio 0.549, 95% CI: 0.315- 0.956) [44]. Additionally, several meta-analyses of RCTs have explored the effectiveness of GLP-1 RA in reducing AF occurrence in patients with type 2 diabetes, comparing GLP-1 RA against placebo or other hypoglycaemic agents. Among these, two studies [45, 46] reported significant reductions in AF occurrence with the use of GLP1 RA, while three [47–49] observed non-significant reductions. The lack of statistical significance or borderline statistical significance from the studies by us and some others may therefore potentially indicate that study samples included for this specific clinical setting was under-powered. Our study indicated that once weight loss was achieved, be it by GLP-1 RA or other weight loss intervention, it reduced the risk of AF recurrence.

Earlier studies showed conflicting results in terms of the effectiveness of LRFM in improving AF ablation outcomes, with both positive [18, 21–23] and negative results [16, 17, 19, 20]. Our study found that the sustained continuation of weight loss post-ablation is an critical determining factor of the ultimate effectiveness any weight loss program. This aligns with the landmark study LEGACY, in which weight fluctuation in obese patients led to a higher risk of AF recurrence compared to those with sustainable weight loss, as greater stability in weight management was associated with improved cardiac remodelling and reduced AF recurrence [4].

Regarding the optimal time for weight loss interventions, both pre-ablation and post-ablation weight loss interventions showed similar outcomes in reducing AF recurrence. However, obesity significantly impacts respiratory mechanics, increasing the risk of perioperative anaesthesia complications [50]. Therefore, pre-ablation weight loss may offer additional benefits by reducing these risks.

This meta-analysis highlights the importance of perioperative weight loss interventions for obese AF patients who undergo AF ablation, as weight management reduces AF recurrence post-ablation. Weight loss programs with long-term support to prevent weight regain should be provided as part of perioperative care for obese AF patients to improve ablation outcomes.

Lifestyle risk factor modifications are always recommended for AF patients, especially for those with obesity and high-risk factors [1]. However, achieving and maintaining weight loss through lifestyle and risk factor changes alone can be challenging. Long-term follow-up and close monitoring are essential to support adherence to the weight loss plan and achieve sustainable weight loss. For patients requiring additional support, both surgical and pharmacological weight loss methods can supplement LRFM to achieve and maintain weight loss goals.

Bariatric surgery showed significant reductions in AF recurrence post-ablation, likely due to its ability to sustain weight loss more effectively than LRFM alone. However, bariatric surgery is generally indicated only for patients with a BMI over 35 kg/m² or a BMI over 30 kg/m² with metabolic disease. Due to the potential risks, bariatric surgery may not be suitable for frail patients or those with multiple comorbidities [51].

For patients who are not suitable or prefer not to undergo surgery, GLP-1 RA offers a non-invasive alternative by reducing appetite and caloric intake to support weight loss maintenance with a healthy lifestyle. Additionally, GLP-1 may potentially provide direct cardiac therapeutic benefits, although further clinical data is needed. The choice of GLP-1 RA type, dosage, and duration will need careful consideration, as weight regain has been observed following the cessation of GLP-1 RA therapy [52]. Different specific types of GLP-1 RA may also affect weight loss outcome [5]. Managing potential adverse reactions, including gastrointestinal side effects, is essential.

In all cases, whether with pharmacological or surgical interventions, maintaining a healthy lifestyle and adjustment of cardiovascular risk factors remain crucial for overall cardiovascular health. Sustainability is the key in any weight loss program to reduce AF recurrence.

This meta-analysis has several limitations. The heterogeneity across studies was high, which may be due to the variation in design, populations, intervention protocols, and definitions of AF recurrence, which limit the generalizability of our study. The definition of AF recurrence varied widely across studies, while several studies defined AF recurrence as atrial arrhythmia episodes ≥ 30 s, one study relied on clinical indicators like the need for AF-related treatments and some did not specify criteria. Additionally, two studies included AF episodes occurring within the blanking period (the initial 3 months post-ablation), which may overestimate the recurrence rates. Moreover, most included studies were observational, and though the majority were propensity score matched, this still increased the potential for bias and confounding. More than half of the included studies were retrospective, which limited the explanatory power due to missing data.

Data-related limitations also impacted the analysis. Adherence to weight loss interventions was inconsistently reported between studies, with some including all patients in the intervention group, regardless of the adherence, while some studies only included participants who strictly adhered to the weight loss strategies and achieved weight loss. This discrepancy limits the comparability and generalizability of the findings. The inclusion of conference abstracts also limits the completeness and quality of data, as these studies lacked details. Subgroup analyses were constrained by the limited number of studies available. Specifically, only one study examined the effect of bariatric surgery, thus limiting statistical power. This is a study limitation that ought to be acknowledged, and findings should be interpreted with caution based on a single study. Furthermore, in two studies (Yaeger et al. [19] and Ding et al. [20]), weight loss interventions were showed to demonstrate a rather opposing trend of being associated with a higher risk of AF recurrence post-ablation, albeit there were not statistically significant. The findings are is not expected, and such discrepancies compared to our overall results warrant to be further explored in future studies. Lastly, there were only six studies included in the meta-regression analysis of the magnitude of weight loss, which may reduce the ability to detect true effects and increase the risk of overfitting. However, this result was consistent with findings in the subgroup analysis and previous studies [34, 35]. This consistency suggests a relationship between weight loss and AF recurrence and highlights the potential of weight loss magnitude as an important factor in explaining the high heterogeneity between studies.

This is a meta-analysis. Ethical approval was not required.

None declared.

Data presented in this study will be made available to researchers upon reasonable request to the corresponding author.