Does sympathetic hyperactivity adversely impact on the effect of pre-ablation bariatric surgery and atrial fibrillation recurrence in morbidly obese patients undergoing atrial fibrillation ablation?

In their innovative study ‘Association between pre-ablation bariatric surgery and atrial fibrillation recurrence in morbidly obese patients undergoing atrial fibrillation ablation’, Donnellan et al.1 concluded that bariatric surgery (BS) is associated with lower atrial fibrillation (AF) recurrence. Also, lower ratios of weight-loss, poor HbA1C improvement levels, and an increased epicardial fat volume [EFV, epicardial adipose tissue (EAT)] were associated with higher rates of AF recurrence post-ablation. As well discussed by the authors, BS was associated with reductions in EFV and improvement in metabolic impairment. Although higher EFV was associated with higher rates of AF recurrence post-ablation, its significance was lost after adjusting for BS in multivariable models.1 Wong et al.2 elegantly reviewed the pivotal role of the autonomic nervous system in the initiation and of AF and the involvement of the ganglionated plexi located in EAT. Their activation causes parasympathetic and sympathetic stimulation, which provokes changes in the atrial myocardium and has been demonstrated to unvaryingly precede the onset of AF.2 Also, this review reported data that injection of botulinum toxin into epicardial fat pads in patients undergoing coronary artery bypass grafting resulted in a sustained and substantial reduction in AF incidence and burden during 3-year follow-up.2 Moreover, Nakahara et al.3 demonstrated that pulmonary vein isolation plus EAT-based ablation efficiently eliminated high-frequency sources and significantly decreased AF recurrence rates.3 Therefore, the authors’ conclusion is entirely plausible.1 Nonetheless, it is relevant to note that at baseline, conditions in which there is a hypersympathetic tone, such as hypertension, morbid-obesity, higher levels of HbA1C, more participants on β-blockers (P = 0.0327, not displayed in the article), and on oral hypoglycaemic (OHG) agents were significantly more common in the non-BS group. Glucose-lowering drugs can have a stimulating effect (insulin, sulphonylureas, dipeptidyl peptidase-4 inhibitors, and thiazolidinediones), a neutral effect (α-glucosidase inhibitors and glucagon-like peptide-1 receptor agonists), or an inhibitory effect (sodium-glucose co-transporter-2 inhibitors and metformin) on the activation of the renin–angiotensin–aldosterone system and sympathetic nervous system (SNS) resulting in or modifying cardiac remodelling.4 Besides, differences between antihypertensive types used for both groups were not available and might have followed a similar pattern to OHG drugs. Indeed, various antihypertensive drugs differentially impact on the renin–angiotensin–aldosterone system and the SNS. Although the blood urea nitrogen was equal between groups, the glomerular filtration rate was not shown, and perhaps chronic kidney disease rates might have been higher in non-BS participants, which itself has been shown to be associated with sympathetic overactivity.5 Importantly, in their uni- and multivariable Cox proportional hazard models, Donnellan et al.1 put the aforementioned variables aside, and it may have ultimately impacted their outcomes. Therefore, a repeat adjusted Cox regression analysis, including those factors, might reveal a different risk ratio between patients who did or did not undergo BS on AF recurrence rates post-ablation. Conflict of interest: M.P.S. is supported by an NHMRC Research Fellowship and has received consulting fees, and/or travel and research support from Medtronic, Abbott, Novartis, Servier, Pfizer, and Boehringer-Ingelheim. MGK, SC, RC and VBM declare that they have no conflict of interest.