AF is the most common arrhythmia in clinical practice and its prevalence is increasing. The estimated number of patients with AF globally was 33.5 million in 2010 and this rose to 50 million in 2020.1,2 The estimated lifetime risk of AF among people in Europe was one in four in 2006, rising to one in three in 2020.3 A similar projected prevalence has been estimated for the US population from 5.2 million in 2010 to 12.1 million in 2030.4 AF is associated with higher morbidity and mortality. Quality of life (QoL) is impaired due to a wide array of symptoms, including lethargy, palpitations, dyspnoea, sleeping difficulties and psychosocial distress. Patients with AF are four times as likely to require cardiovascular hospitalisation compared to propensity-matched controls and suffer from higher rates of heart failure and stroke.5 AF is an independent predictor of all-cause mortality, with a two-fold adjusted increase in death.6
Mirroring this, the prevalence of obesity is rising rapidly and brings with it an exponential risk of cardiovascular morbidity and mortality. Globally, the estimated prevalence of obesity (BMI ≥30) increased from 3.2% in 1975 to 10.8% in 2014 in men, and from 6.4% to 14.9% in women.7 In 2019, more than half (53%) of the European population was overweight (BMI ≥25 kg/m2).8 In the US, the prevalence of adult obesity was 42% in 2017–2018.9 While it is well-known that obesity is an upstream risk factor for hypertension, diabetes and obstructive sleep apnoea (OSA), the independent impact of obesity on incidence, recurrence and management of AF is less appreciated. Both European (anticoagulation, better symptom control, comorbidities) and American (lifestyle and risk factor modification) guidelines on the management of AF recommend weight loss before ablation; however, it is less clear how that may be best achieved and over what time course.3,10,11
The aim of this article is to review the available evidence, including emerging therapies that may optimise patient outcomes.
Obesity and Onset of AF
Multiple epidemiological studies have shown a strong association between obesity and the risk of developing AF, although there are many potential confounding factors, including hypertension, diabetes and dyslipidaemia. However, a smaller number of studies have investigated the effect of obesity independent of associated risk factors by recruiting metabolically healthy obese (MHO) patients, defined as having a BMI above 30 but without associated metabolic conditions (Supplementary Table 1).12–14 MHO is characterised by normal insulin sensitivity blood pressure and lipid profiles. Crucially, obesity should be understood as a complex continuum instead of a binary condition. MHO may progress to metabolically unhealthy obesity (MUO) over time unless preventive interventions are not initiated early.
In a Swedish population-based longitudinal study of 4,021 participants with a mean follow-up of 13.6 years, obesity was associated with an increased risk of AF independent of other risk factors (HR 1.75; 95% CI [1.11–2.74]).12 A study that used a national health insurance database of a large population of 389,321 participants in Korea confirmed that MHO individuals had an increased risk of AF (HR 1.3; 95% CI [1.14–1.48]).13 Of importance, a quarter of this cohort became metabolically unhealthy during a mean follow-up of 7.5 ± 1.5 years. The question whether MHO individuals and MUO individuals carry the same risk of AF was investigated in another population-based prospective study of 47,670 participants from Norway.14 Compared to people with a normal weight (BMI <25 kg/m2), both MHO and MUO individuals had a HR of 1.6 for the development of AF and the risk increased according to the severity of obesity. This growing evidence suggests that obesity is an independent risk factor for AF.
Obesity and Persistence of AF
It has long been established that AF starts with the rapid firing of ectopic foci (triggers) from myocyte sleeves within the pulmonary veins.15 Typically, these present as short attacks and can spontaneously resolve with or without treatment. For these triggers to set off persistent AF, conditions (referred to as substrates) must be favourable. Substrates can be structural, such as inflammation, scarring and changes in pressure or volume, or electrical, such as changes in conduction velocity and voltage in the atria. A trigger initiates AF and the substrate enables it to persist. This interplay between triggers and substrate governs the clinical course of AF and obesity is a common interlocutor (Figure 1).
Obesity and Cardiac Structure and Function
Left atrial (LA) dilatation, whether measured by diameter or volume, and impaired LA function, are known to be important substrates that promote persistent AF. These changes are common in obesity and precede the onset of AF. The prevalence of LA enlargement in obesity ranges from 9.8% to 37% by LA diameter on 2D echocardiography, and obesity is a more important driver for this change than hypertension.16–18 Similar changes are seen on echocardiography in MHO participants, who have a significantly larger LA dimension than people with a normal weight.19 Obese patients have impaired LA global strain which is a measure of how good the left atrium can stretch and contract during a cardiac cycle measured by speckle-tracking echocardiography which tracks tiny dots or speckles all over the left atrium (−10.5 ± 3.7%), compared to non-obese patients (−13.1 ± 5.1%; p=0.004).20 Further, obesity is associated with impaired reservoir (the ability of the LA to store pulmonary venous return during left ventricular (LV) contraction and isovolumetric relaxation) and conduit LA function (the ability of the LA to transport blood passively into left ventricle) (Figure 2).21
Changes in the LA are likely to be secondary to impaired LV diastolic function and LV hypertrophy. Multiple studies have confirmed that diastolic dysfunction is common in obesity, reflected in a higher pressure half-time, lower deceleration slope, high peak A velocity, lower E/A ratio, prolonged isovolumetric relaxation time, and a lower peak systolic velocity of the lateral mitral annulus.22–26 Post-mortem studies of obese individuals reported higher heart weight, LV wall thickness and more frequent dilated cardiomyopathy.27 These changes were corroborated by a meta-analysis of 15 echocardiographic studies that reported obese patients were 4.2-fold more likely to have LV hypertrophy than non-obese patients.28
Obesity, Inflammation and AF
Epicardial adipose tissue (EpAT) and the myocardium are developed from splanchnic mesoderm and receive their blood supply through the coronary arteries.29 By contrast, paracardial fat develops from the primitive thoracic mesenchyme and blood is supplied from branches of the internal mammary artery.29 Unlike paracardial fat, there is no fascia between EpAT and the myocardium, meaning that EpAT is in direct contact and can share microcirculation and local interactions.30 EpAT relation to AF is particularly important. EpAT is positively correlated with BMI.31,32 Moreover, multiple studies have reported a close relationship between the extent of EpAT and incidence, duration and recurrence of AF.33–37 EpAT can be measured using echocardiography, CT or cardiovascular MRI with low inter- and intra-observer variability (Figure 3).35,38–40
EpAT contributes to the development and progression of AF through effects including inflammation and fibrosis. In relation to inflammation, Mazurek et al. compared the inflammatory activity of EpAT and skin adipose tissue (SAT) using samples harvested from elective coronary artery bypass graft (CABG) surgery. Higher levels of inflammatory cytokines such as interleukin (IL)-1b, IL-6 and tumour necrosis factor (TNF)-α were found in EpAT samples (p<0.001).41 Fain et al. replicated these findings to show that EpAT secreted a higher level of IL-1β and TNF-α compared to sternal, abdominal and leg SATs.42 Irrespective of underlying coronary artery disease or hypertension, EpAT from overweight/obese patients showed higher levels of IL-1, IL-6, TNF-α and interferon-g compared to lean controls (p<0.01).43
In relation to fibrosis, the atrial myocardium in rats was exposed to the samples of EpAT and SAT from patients undergoing CABG by Venteclef et al.44 Compared to SAT, EpAT secreted activin A (p=−0.0002), a profibrotic maker, which induced both interstitial and peripheral fibrosis of the rat atria. In a human study comparing EpAT, SAT and pericardial adipose tissue from patients with sinus rhythm and AF undergoing CABG, the expression of connective tissue growth factor was higher in the EpAT of AF patients than in the SAT and PAT of AF patients and the EpAT of patients with sinus rhythm.45 In a study of right atrial appendage tissue excised from patients undergoing CABG with AF, Nalliah et al. reported higher local EpAT volume to be associated with more fibrosis, slower conduction and lateralisation of connexin40 (an important human atrial gap junction protein that regulates successful propagation of action potentials through the myocardium) than in patients in sinus rhythm.46 The lack of fascia between EpAT and the myocardium allows these mediators or ‘adipokines’ to have a local (paracrine) effect, acting as drivers for arrhythmias both in animal models and in human studies.43,47,48 (Supplementary Table 2 for multimodality assessment of persistent AF substrate in obese AF patients).
Obesity and Electrical Remodelling
Obesity-dependent changes in LA conduction velocity and voltage have been reported in animal studies.49,50 In ovine studies, sheep were fed with hay and high-energy pellets to induce progressive weight gain. This resulted in progressive LA electrical remodelling. Obese sheep developed lower LA conduction velocity, decreased LA voltage, higher LA pressure and more extensive fibrosis. These electrical changes were replicated in obese patients with AF and may be reversible with weight loss.51 In a study of 30 sheep fed over 72 weeks until they were obese, Mahajan et al. showed that with 30% weight reduction, these changes could be reversed resulting in higher conduction velocity, a fall in LA pressure and a reduction in LA fibrosis.52 It is not known, however, whether weight loss in humans can achieve similar reverse remodelling of atrial substrates.
Obesity and Outcome from Ablation
It is well established that obese patients have lower procedural success (usually defined as evidence of AF in a 12-lead ECG or Holter monitoring in previous research studies) and higher complication rates following catheter ablation for AF (Supplementary Table 3). In brief, recurrence of AF following radiofrequency ablation increased by 15% for each 5 kg/m2 increase in BMI (risk ratio 1.15; 95% CI [1.08–1.22]; heterogeneity, 53%) in a meta-analysis of 26 cohort studies.53 Similarly, Ding et al. reported a 1–5% increased risk of recurrence with every increase of one unit of BMI.54 It is important to note that this effect did not seem to be replicated in studies with cryoablation probably because the majority of the study populations were paroxysmal AF (PAF) patients.55,56 Further studies are required to answer which ablation strategy delivers the most benefit to this group of patients.
Effect of Weight Loss on AF Ablation
Weight Loss Before Ablation
There is growing evidence of the benefit of weight loss before ablation for obese patients. A recent meta-analysis of seven cohort studies by Zhao et al. reported that every 10% weight loss from baseline significantly reduced the chance of recurrence (RR 0.54; 95% CI [0.33–0.88]).57 Of the seven studies, four used lifestyle modification and one used bariatric surgery pre-ablation and the duration of intervention varied from 6 to 22 months (Table 1). Weight loss is not only linked to the success of the procedure but a further meta-analysis of six cohort studies showed an extended effect beyond 12 months in reducing recurrence (RR 0.49; 95% CI [0.31–0.80]).58 Moreover, a significant improvement in outcome accrues from less than 10% weight loss compared to no weight loss at all in longer-term follow-up (RR 0.39; 95% CI [0.31–0.49]). Maintaining a lower weight and not gaining weight is also crucial after ablation. Fluctuation in BMI <1 unit after ablation should be the goal for patients (change in BMI <1 unit versus a change in BMI ≥1 unit OR 0.4; 95% CI [0.18–0.9]).59 For daily clinical practice, patients should be advised to achieve some weight loss at least 3–6 months before offering AF ablation and maintaining this weight within 1 unit of BMI after ablation.
Weight Loss After Ablation
Losing weight after AF ablation does not have a similar beneficial effect on recurrence. Gessler et al. randomised 133 patients either to a structured weight reduction programme in a specialised obesity department for the duration of 6 months or a control group.60 Hypertension, diabetes and sleep apnoea were optimised according to current guidelines before ablation in both groups. With implantable monitoring, neither AF burden (defined as the percentage of AF during the observed period) at 3 and 12 months or AF recurrence at 12 months were not significantly different between intervention (BMI changing from 34.9 ± 2.6 to 33.4 ± 3.6 at 12 months) and the control group (no change to BMI). Until further studies investigate a head-to-head comparison between pre-ablation versus post-ablation weight loss, the evidence supports weight loss before ablation and maintenance post-ablation.
Timing the Weight Loss
Weight loss before ablation must be achieved in a timely fashion as the beneficial effect appears to be lost in those with persistent AF for more than 12 months. Despite significant weight loss (19.1–56.7 kg) with lifestyle modification for 1 year before ablation in a prospective cohort study, the single-procedure recurrence rate was similar between the weight loss and no weight loss group at 1 year (63.8% versus 59.3%; p=0.68).61 Recent data published have indicated that a better outcome may be achieved by interventions (lifestyle modification and drug therapy) being limited to only 3 months before ablation (Table 1). This will be discussed further below.62
Strategies for Weight Loss Before Ablation
The ideal programme to improve ablation outcomes would include interventions that target multiple risk factors, including timely weight reduction, optimisation of hypertension and diabetes, treatment of OSA, reduction of alcohol intake, smoking cessation and increasing physical activity. Such risk factor management (RFM) programmes have been tested but are resource intensive. The outcomes of these studies are shown in Table 2. For example, Gessler et al. tested a programme led by an endocrinologist in a specialised obesity department that required participants to attend twice a month for nutritional advice, physical training and cognitive behavioural therapy.60 Before undergoing ablation, all participants were investigated for sleep disordered breathing and control of hypertension and diabetes was optimised, and the programme continued for 6 months following successful ablation.
A similar approach using a structured motivational and goal-directed programme was employed in the ARREST-AF and LEGACY studies.63,64 The research team supported a dedicated physician to run the programme that included dietary advice, meal replacement sachets and a lifestyle journal (recording diet, weight, blood pressure and exercise). A review occurred every 3 months with additional visits as needed (for the whole duration of the study up to 48 months). Sleep-disordered breathing, hypertension, diabetes and lipids were closely monitored and optimised according to current guidelines. Behavioural support was offered to smokers and counselling follow-ups were provided for those with an excessive alcohol intake.
Less resource-intensive strategies have also been tested, limited to dietary and exercise advice with other risk factors managed by the treating electrophysiologist or a registered nurse with input from a dietitian. The outcomes of these studies are shown in Table 2. For example, the PENN AF Care study was managed by a registered nurse, supported by a cardiac electrophysiologist and tested intermittent fasting and counselling with a dietitian to achieve more than 5% body weight loss from baseline over 6–12 months.61,65,66 The focus of the study was on obesity and OSA, although records were made of the participants’ diet, alcohol and tobacco consumption and physical activity.65 In a variation of this more focused approach, dietitians prepared personalised meal plans to reduce daily calorie intake and participants were asked to perform moderate intensity exercise (150 min/week).61 However, this approach has not been tested in larger randomised controlled studies.
Drug Therapy for Weight Loss and Ablation: the Future?
Recent studies have led to the approval of glucagon-like peptide 1 analogues/receptor agonists (GLP-1/RAs) for use as an adjunct to weight loss. GLP-1RAs mimic the function of the natural hormones called incretins that slow gastric emptying, delay absorption of food, increase satiety, inhibit the release of glucagon and stimulate insulin production. On average, those who take GLP-1RAs lose weight from 0.64–6.5 kg over a 1-year period.67 The first approved GLP-1RAs, such as exenatide and liraglutide, require once or twice-daily administration and were followed by longer-acting versions, such as dulaglutide, exenatide once weekly, semaglutide and tirzepatide (a glucose-dependent insulinotropic polypeptide receptor and GLP-1 receptor co-agonist), which are suitable for once-weekly administration (Supplementary Table 4).
This group of drugs, however, exert pleiotropic effects that make them attractive for short-term use to improve outcomes of ablation in obese patients. However, a recent animal study using swine suggests GLP-1RAs have a direct chronotropic effect on the heart via a calcium signalling response.68 Indirect benefits include not only weight loss and improved diabetic control but also improved blood pressure control, for example, liraglutide achieved a greater reduction in systolic blood pressure (−4.6 versus −1.5 mmHg; p=0.002) and reduction in HbA1C (−0.22 versus −0.06%; p=0.0001) than orlistat.69
Mechanistic studies suggest benefits that go beyond an impact on weight loss. For example, a randomised, double-blind placebo-controlled phase IV trial of liraglutide found a preferential reduction of the visceral adipose tissue over 40 weeks (mean difference −10.86%; 95% CI [−6.97– −14.75]; p<0.001) compared to placebo.70 Additionally, liraglutide preferentially lowered EpAT thickness with a treatment duration as early as 3 months in diabetic patients, an effect also seen with semaglutide.71–73 As discussed earlier, this is important in AF due to the reduction in paracrine and endocrine effects of this EpAT and visceral adipose tissue that increase inflammation. An anti-inflammatory effect of these drugs is supported by studies that found lower proinflammatory markers such as TNF-α (−1.33 versus 0.00 pg/ml), IL-6 (−0.78 versus +0.04 pg/ml), and high-sensitivity C-reactive protein (1.55 versus 0.46 µg/ml; p<0.05) in patients treated with liraglutide over 12 weeks.74 Furthermore, the slow progression of fibrosis was reported in patients with non-alcoholic fatty liver disease treated with liraglutide compared with placebo, therefore further studies are needed to investigate GLP-1Ras’ effect on human atria, especially on anti-fibrosis.75 Although the major cardiovascular risk reduction effect is observed in patients without diabetes treated with GLP-1RAs reported in a recent meta-analysis of 10 randomised controlled trials, few studies confirm or refute its benefit on patients with AF.76 These widespread positive effects raise the possibility that short-duration therapy may improve outcomes of AF ablation in association with risk factor management programmes.
Importance of Weight Loss Regardless of Ablation
As obesity is also an upstream risk factors for other morbidities such as hypertension, diabetes and obstructive sleep apnoea, reduction in weight can optimise the outcomes from these conditions. Hypertension is a well-recognised risk factor (RR 1.5; 95% CI [1.42–1.58]) in the development of AF in a recent systematic review of 56 cohort studies.77 A similar effect was reported in patients with diabetes (RR 1.39; 95% CI [1.1–1.75]; p=0.000) in another meta-analysis.78 OSA also promotes the risk of AF (RR 1.7; 95% CI [1.53–1.89]) as highlighted in a recent meta-analysis.57 Therefore, weight loss for conditions can in turn lower the risk of AF and its progression.
Weight management can also deliver fewer AF episodes, symptoms and a lower burden at 1 year.79 This seems to have a long-term effect of up to 5 years if maintained weight loss without fluctuation with the potential reversal of AF phenotype (persistent to paroxysmal, paroxysmal to sinus rhythm).63,80,81 A weight reduction strategy, therefore, can be considered as a preventive approach for AF progression even before ablation.
Conclusion
Obesity without other associated risk factors is an independent risk factor in AF. The success of catheter ablation is substantially decreased by obesity. The importance of weight management including RFM is highlighted in recently published international guidelines. This review discussed the available evidence that pre-ablation weight loss for at least 3–6 months and weight maintenance post-ablation is crucial for better outcomes from ablation. Moreover, changes in cardiac investigations can potentially be useful to monitor the progress and regress of weight management when selecting an appropriate ablation candidate. Further studies on RFM using newer adjunct therapies might shift the paradigm for the management of obese AF patients in the future.
Clinical Perspective
- Pre-ablation weight loss of at least 5–10% from baseline over 3–6 months appears crucial for improved ablation outcomes in obese AF patients.
- Conventional imaging techniques have not proven to be reliable in assessing left atrial substrate prior to ablation.
- Short-term use of GLP-1 receptor agonists show promise in improving ablation outcomes, potentially through rapid visceral fat reduction, and anti-inflammatory and pleiotropic effects.
- Maintaining weight loss within one unit of BMI after ablation appears important for long-term arrhythmia-free survival in patients with obesity.
- Structured risk factor management programmes, including dietary interventions and exercise, may enhance patient selection and ablation success rates in patients with obesity.