Study forges a path toward polygenic risk scoring and personalized treatment.

A new study published in Circulation: Genomic and Precision Medicine analyzed 1,047 patients to identify polygenic risk factors for postoperative atrial fibrillation (PoAF) and assess their predictive values for developing the condition after cardiac surgery. Primary investigator and first author Miklos Kertai, M.D., a professor of cardiothoracic anesthesiology, and co-investigator and senior author Lisa Bastarache, M.S., a research assistant professor of biomedical informatics at Vanderbilt University Medical Center, found that atrial fibrillation risk rose 63 percent for every standard deviation increase in polygenic risk score.

These findings have significant potential for advancing personalized treatment strategies to prevent many PoAF events. New-onset PoAF impacts about 35 percent of patients undergoing cardiac surgery. It is associated with adverse neurological events, congestive heart failure, myocardial infarction, perioperative mortality, prolonged hospital length of stay and increased hospital costs.

Until now, studies generally have been limited to evaluating how individual genes predispose a person to PoAF, Kertai says. Yet, no one genetic variant or small group of variants has risen to the level of being consistently significant as a biomarker. “If further studies bear out our results, and with clinical genotyping seemingly more within reach every day, our findings bode well for prediction and prevention of this dangerous post-op complication,” he said.

From SNPs to Risk Score

Bastarache said the ultimate aim is personalized PoAF prevention. “We asked the question, ‘can genetics improve clinical predictions of who is at risk for atrial fibrillation?’ Because PoAF is such a common problem, a little bit more predictive power could potentially help a lot of people,” she said.

To define a robust study population, the team utilized deidentified EHR data and biorepository samples to retrospectively study 1,047 patients of European ancestry who underwent coronary artery bypass graft and/or valve surgery in the past two decades. None of the patients had a preoperative history of atrial fibrillation, yet 25 percent of them developed PoAF.

“If further studies bear out our results, our findings bode well for prediction and prevention of this dangerous post-op complication.”

The researchers examined 129 single nucleotide polymorphisms (SNPs) and 2,746 common genetic variants previously associated with atrial fibrillation. When they aggregated significant SNPs associated with PoAF to form a polygenic risk score, the score demonstrated improved discrimination, calibration and risk reclassification compared to conventional clinical predictors.

“Prior research has shown that polygenic risk scores are fairly predictive of atrial fibrillation in the general population,” Bastarache said. “In our study, we found that the stress of surgery may unmask a person’s genetic risk factors. This is intuitive, but very nice to see in real data.”

Clinical Ramifications

In the new study, adding the polygenic risk score to conventional clinical predictors increased the proportion of patients categorized as high risk from 18 to 29 percent. Among patients without PoAF, the proportion categorized as low risk increased from 48 to 54 percent. In the intermediate risk categories, the polygenic risk score improved reclassification by increasing risk estimates among patients with PoAF and decreasing risk estimates among those without PoAF.

This improved risk identification is important for two main reasons, Kertai said. “First, it can facilitate adherence to evidence-based recommendations for the prevention of PoAF. Second, it can highlight the opportunity to selectively apply preventative measures.” This could include a combination of prophylactic perioperative beta-blocker and amiodarone for patients at high risk for PoAF.

Even with the combination of these medications, Kertai says there is still a subset of patients who will develop PoAF. A polygenic risk score may help these patients as well because it could identify genetic variations associated with other diseases. “Some medications and treatment strategies used for those conditions could in an indirect way modify the risk of atrial fibrillation after cardiac surgery,” Kertai said.


Miklos Kertai, M.D.

Miklos Kertai, M.D., Ph.D., is a professor of cardiothoracic anesthesiology at Vanderbilt University Medical Center. His research interests focus largely on precision perioperative medicine and the significance of genomics in informing the future of drug trials and clinical practice.

Lisa Bastarache, M.S.

Lisa Bastarache, M.S., is a research assistant professor of biomedical informatics and scientific director for PheWAS Core at Vanderbilt University Medical Center, where she is responsible for the development and implementation of new phenotyping methods.