By Joshua D. Moss, MD

Associate Professor of Clinical Medicine, Cardiac Electrophysiology, Division of Cardiology, University of California, San Francisco

Dr. Moss reports no financial relationships relevant to this field of study.

SYNOPSIS: Significant and sustained weight loss through bariatric surgery reduced the risk of developing new-onset atrial fibrillation in a Swedish cohort of obese individuals.

SOURCE: Jamaly S, Carlsson L, Peltonen M, et al. Bariatric surgery and the risk of new-onset atrial fibrillation in Swedish obese subjests. J Am Coll Cardiol 2016;68:2497-2504.

Obesity is an emerging global epidemic and a well-described risk factor for atrial fibrillation (AF), with a 4-5% increased risk of incident AF per unit increase in body mass index (BMI). Given its independent association with stroke and heart failure, AF may play a role in higher morbidity and mortality seen in overweight individuals. In patients with AF, weight loss in the short term is associated with a reduction in symptomatic AF burden, and aggressive weight and risk factor management improves maintenance of sinus rhythm after AF ablation. Long-term sustained weight loss also is associated with significant reduction in AF burden, particularly with a loss of 10% body weight. Although intensive lifestyle intervention with modest weight loss has not been shown to produce a significant effect on rates of new-onset AF, the effect of more significant weight loss had not been studied previously.

The Swedish Obese Subjects (SOS) study is an ongoing prospective, controlled intervention trial of 4,047 patients between 37 and 60 years of age. Subjects were enrolled between 1987 and 2001 and had a BMI 34 kg/m2 for men or 38 kg/m2 for women. An intervention group of 2,010 patients underwent bariatric surgery, while a control group of 2,037 patients who chose conventional therapies was selected via an automatic matching program based on 18 variables. Although incident AF was not a predefined endpoint of the study, AF data were available from the Swedish National Patient Register, allowing 4,021 patients without baseline AF to be identified and analyzed for this study (2,000 undergoing bariatric surgery and 2,021 treated conventionally).

The study groups were not perfectly matched, with a significantly higher baseline BMI in the surgical group as well as more hypertension, diabetes, and smoking. Surgical patients also were on average 18 months younger than conventionally treated patients. Mean weight reduction in the surgical group was 25% by one year and 18% at 20 years, while the control group experienced no significant weight change over two decades. The significant weight loss in the surgical group was independently associated with a 31% reduction in new-onset AF over a median follow-up of 19 years after adjustment for selected baseline variables. Other conditions independently associated with reduced risk included younger age, shorter height, lower BMI, lower blood pressure, lower alcohol intake, and lower thyroxin levels. The effect of bariatric surgery on reducing new-onset AF was conserved largely among subgroups, although patients younger than 47.7 years benefited more than older patients, and patients with diastolic blood pressure (DBP) 88 mmHg benefited more than those with lower DBP. The authors concluded that sustained weight loss after bariatric surgery compared to usual care reduced the risk of new-onset AF.


Clearly, not all obese patients will be eligible for or interested in bariatric surgery, but, nevertheless, this study has important implications for patient care and researchers. Under the reasonable assumption that the relevant effect of bariatric surgery is predominantly loss of weight, the ability to achieve and maintain dramatic weight loss is associated with a significant reduction in new-onset AF. Some important questions to consider include: Was the SOS cohort representative of our typical obese patient population? Were the two groups matched well enough in this observational study to correctly attribute the large difference in outcomes to the surgery and attendant weight loss? What nonsurgical strategies can be developed to achieve the weight loss needed to affect arrhythmia burden?

The prevalence of an AF diagnosis at baseline in this cohort patients 37-60 years of age was 26 of 4,047 (0.6%), comparable to that found in prior studies of unselected patients, such as ATRIA. The overall incidence of new AF over a median follow-up of 19 years was 587 of 4,021 (14.6%), a relatively high risk compared to the general population and further supporting the idea that obese patients should be targeted specifically for risk-reduction strategies.

The intervention and control groups had several significant differences in baseline characteristics, which the authors believed were probably explained by disparate weight changes in the two groups during the sometimes-long period between matching and baseline measurements. When controlling for these characteristics statistically, weight loss remained independently associated with lower incidence of AF. However, it might be expected that other characteristics for which the authors did not control could have influenced the results — both behavioral (such as motivation to exercise and follow medical recommendations) and clinical (such as likelihood for caregivers to recommend and encourage surgical management). Nevertheless, the effect of weight loss was profound, particularly among younger patients for whom there was perhaps a better chance of significant reverse atrial remodeling.

The major challenge raised by this study is how to achieve the kind of weight loss necessary to affect AF risk without committing all patients to bariatric surgery. In the meantime, this is another strong motivator for clinicians to highlight for patients who want to lose weight.