From the AGA Journals

NAFLD may predict arrhythmia recurrence post-AFib ablation


Increasingly recognized as an independent risk factor for new-onset atrial fibrillation (AFib), new research suggests for the first time that nonalcoholic fatty liver disease (NAFLD) also confers a higher risk for arrhythmia recurrence after AFib ablation.

Over 29 months of postablation follow-up, 56% of patients with NAFLD suffered bouts of arrhythmia, compared with 31% of patients without NAFLD, matched on the basis of age, sex, body mass index (BMI), ejection fraction within 5%, and AFib type (P < .0001).

The presence of NAFLD was an independent predictor of arrhythmia recurrence in multivariable analyses adjusted for several confounders, including hemoglobin A1c, BMI, and AFib type (hazard ratio, 3.0; 95% confidence interval, 1.94-4.68).

The association is concerning given that one in four adults in the United States has NAFLD, and up to 6.1 million Americans are estimated to have Afib. Previous studies, such as ARREST-AF and LEGACY, however, have demonstrated the benefits of aggressive preablation cardiometabolic risk factor modification on long-term AFib ablation success.

Indeed, none of the NAFLD patients in the present study who lost at least 10% of their body weight had recurrent arrhythmia, compared with 31% who lost less than 10%, and 91% who gained weight prior to ablation (P < .0001).

All 22 patients whose A1c increased during the 12 months prior to ablation had recurrent arrhythmia, compared with 36% of patients whose A1c improved (P < .0001).

“I don’t think the findings of the study were particularly surprising, given what we know. It’s just further reinforcement of the essential role of risk-factor modification,” lead author Eoin Donnellan, MD, Cleveland Clinic, said in an interview.

The results were published Augus 12 in JACC Clinical Electrophysiology.

For the study, the researchers examined data from 267 consecutive patients with a mean BMI of 32.7 kg/m2 who underwent radiofrequency ablation (98%) or cryoablation (2%) at the Cleveland Clinic between January 2013 and December 2017.

All patients were followed for at least 12 months after ablation and had scheduled clinic visits at 3, 6, and 12 months after pulmonary vein isolation, and annually thereafter.

NAFLD was diagnosed in 89 patients prior to ablation on the basis of CT imaging and abdominal ultrasound or MRI. On the basis of NAFLD-Fibrosis Score (NAFLD-FS), 13 patients had a low probability of liver fibrosis (F0-F2), 54 had an indeterminate probability, and 22 a high probability of fibrosis (F3-F4).

Compared with patients with no or early fibrosis (F0-F2), patients with advanced liver fibrosis (F3-F4) had almost a threefold increase in AFib recurrence (82% vs. 31%; P = .003).

“Cardiologists should make an effort to risk-stratify NAFLD patients either by NAFLD-FS or [an] alternative option, such as transient elastography or MR elastography, given these observations, rather than viewing it as either present or absence [sic] and involve expert multidisciplinary team care early in the clinical course of NAFLD patients with evidence of advanced fibrosis,” Dr. Donnellan and colleagues wrote.

Coauthor Thomas G. Cotter, MD, department of gastroenterology and hepatology, University of Chicago, said in an interview that cardiologists could use just the NAFLD-FS as part of an algorithm for an AFib.

“Because if it shows low risk, then it’s very, very likely the patient will be fine,” he said. “To use more advanced noninvasive testing, there are subtleties in the interpretation that would require referral to a liver doctor or a gastroenterologist and the cost of referring might bulk up the costs. But the NAFLD-FS is freely available and is a validated tool.”

Although it hasn’t specifically been validated in patients with AFib, the NAFLD-FS has been shown to correlate with the development of coronary artery disease (CAD) and was recommended for clinical use in U.S. multisociety guidelines for NAFLD.

The score is calculated using six readily available clinical variables (age, BMI, hyperglycemia or diabetes, AST/ALT, platelets, and albumin). It does not include family history or alcohol consumption, which should be carefully detailed given the large overlap between NAFLD and alcohol-related liver disease, Dr. Cotter observed.

Of note, the study excluded patients with alcohol consumption of more than 30 g/day in men and more than 20 g/day in women, chronic viral hepatitis, Wilson’s disease, and hereditary hemochromatosis.

Finally, CT imaging revealed that epicardial fat volume (EFV) was greater in patients with NAFLD than in those without NAFLD (248 vs. 223 mL; P = .01).

Although increased amounts of epicardial fat have been associated with CAD, there was no significant difference in EFV between patients who did and did not develop recurrent arrhythmia (238 vs. 229 mL; P = .5). Nor was EFV associated with arrhythmia recurrence on Cox proportional hazards analysis (HR, 1.001; P = .17).

“We hypothesized that the increased risk of arrhythmia recurrence may be mediated in part by an increased epicardial fat volume,” Dr. Donnellan said. “The existing literature exploring the link between epicardial fat volume and A[Fib] burden and recurrence is conflicting. But in both this study and our bariatric surgery study, epicardial fat volume was not a significant predictor of arrhythmia recurrence on multivariable analysis.”

It’s likely that the increased recurrence risk is caused by several mechanisms, including NAFLD’s deleterious impact on cardiac structure and function, the bidirectional relationship between NAFLD and sleep apnea, and transcription of proinflammatory cytokines and low-grade systemic inflammation, he suggested.

“Patients with NAFLD represent a particularly high-risk population for arrhythmia recurrence. NAFLD is a reversible disease, and a multidisciplinary approach incorporating dietary and lifestyle interventions should by instituted prior to ablation,” Dr. Donnellan and colleagues concluded.

They noted that serial abdominal imaging to assess for preablation changes in NAFLD was limited in patients and that only 56% of control subjects underwent dedicated abdominal imaging to rule out hepatic steatosis. Also, the heterogeneity of imaging modalities used to diagnose NAFLD may have influenced the results and the study’s single-center, retrospective design limits their generalizability.

The authors reported having no relevant financial relationships.

Help your patients better understand their risk of NASH and NAFLD by sharing AGA patient education content at

A version of this article originally appeared on

Next Article: