Flip of the Coin for New-Onset AFib With TAVR or SAVR

Patrice Wendling

June 10, 2019

One in every two patients undergoing transcatheter or surgical aortic valve replacement (TAVR/SAVR) will develop new-onset atrial fibrillation (AF) and as a result face higher odds of in-hospital mortality, a large nationwide study suggests.

Using data from the National Inpatient Sample (NIS), investigators found the incidence of new-onset AF was 50.4% among TAVR hospitalizations and 50.1% among SAVR hospitalizations.

The odds of in-hospital mortality with new-onset AF were 57% higher with TAVR (odds ratio [OR], 1.57; 95% CI, 1.21 - 2.04) and 36% higher with SAVR (OR, 1.36; 95% CI, 1.08 - 1.70) after full multivariate adjustment.

Both groups had longer lengths of stay, but the TAVR group also had higher odds of in-hospital stroke (OR, 1.10; 95% CI, 1.02 - 1.16), according to the study, published online June 3 in JAMA Internal Medicine.

An association between new-onset AF and valve replacement isn't new, but estimates are based on randomized controlled trials or smaller observational studies and range from 2% to 100%, explained senior author Pankaj Arora, MD, University of Alabama at Birmingham.

"In our clinics, when we're counseling patients whether we should put in a valve, we don't really incorporate atrial fibrillation into the clinical decision-making, so this really was the motivation to understand the population burden of atrial fibrillation," he told theheart.org | Medscape Cardiology.

"You can argue that administrative database studies are often limited by granularity and correctly identifying the condition — whether it's truly new-onset or prevalent. But recognizing all these issues, we wanted to see, in the biggest cohort we could look at, what these numbers look like."

Based on International Classification of Disease (ICD)-9 codes in the NIS database, the investigators identified 48,715 TAVR hospitalizations (47.4% women; mean age, 81.3 years; 82.3% white) and 122,765 SAVR hospitalizations (39% women; mean age, 67.8 years; 78% white) from January 2012 to September 2015.

The incidence of new-onset AF was determined using the ICD-9 code 427.3 if it appeared in any of the secondary discharge fields.

Validation Cohort

To address the question of whether the AF diagnosis in the NIS sample represents new-onset AF, the investigators turned to the New York State Inpatient database, which has a unique present-on-admission identifier indicating whether each diagnosis is present at the time of admission.

This cohort consisted of 1736 patient-linked hospitalizations for TAVR and 5141 for SAVR between January 2012 and December 2014.

The incidence of new-onset AF was two times higher in patients undergoing SAVR than in those undergoing TAVR (30.6% vs 14.1%), but overall still lower than observed in the NIS cohort.

The fully adjusted odds of in-hospital mortality were 50% higher for SAVR, although this difference failed to reach statistical significance (OR, 1.50; 95% CI, 0.97 - 2.30) and 76% higher for TAVR (OR, 1.76; 95% CI, 1.17 - 2.65), the authors report.

Consistent with the NIS sample, both groups had significantly longer lengths of stay. In-hospital stroke was more common in the TAVR group with new-onset AF vs no AF (8.5% vs 6.1%) but not significantly higher in the fully adjusted model.

"The thing I'm not sure about is the temporal sequence of atrial fibrillation and stroke," Arora said. "Could stroke have happened and then that may have brought the AFib out? Sure. We see strokes post-TAVR all the time. Or was the AF there first? We can't say from these data."

Postprocedure Management

Commenting to theheart.org | Medscape Cardiology, Amit N. Vora, MD, UPMC Pinnacle, Harrisburg, Pennsylvania, said overall, it was a well-done paper using a widely used dataset but that the 50% rates of new-onset AF likely represent the total burden of AF in this population.

"The results that they captured from the validation cohort are more in-line with previously published studies than a 50% rate of new-onset atrial fibrillation," he said. "That just seems pretty high."

Last year, Vora and colleagues published an analysis from the Society of Thoracic Surgeons (STS)/American College of Cardiology (ACC) Transcatheter Valve Therapy (TVT) registry, in which 8.4% of patients developed new-onset AF. In-hospital mortality and stroke rates in these patients were 7.8% and 4.7%, respectively.

The risk for AF is important to discuss with patients consenting for valve replacement, Vora agreed, noting that postprocedure management in these patients can be challenging because the optimal antithrombotic strategy is not clear. The GALILEO trial was stopped early last year after a preliminary analysis showed that patients treated post-TAVR with rivaroxaban (Xarelto, Bayer/Janssen) vs an antiplatelet therapy had higher rates of bleeding, thromboembolic events, and death.

The ongoing POPULAR TAVI and ATLANTIS trials in patients undergoing TAVR should provide new insights in the coming months, Vora said.

AF should be a part of the evaluation of patients undergoing valve replacement in clinical practice and routinely, although not historically, captured in clinical trials, Arora said.

"The association between AF and hard end points is so strong, it makes perfect sense to include rates of atrial fibrillation at 30 days, 1 year, or whatever follow-up the trial has," he said. "The ideal comparison would be to do pre and post, to put your patient on a monitor for 14 days, and then see what develops. That would be ideal."

None of the authors report having relevant conflicts of interest.

JAMA Intern Med. Published online June 3, 2019. Abstract

Follow Patrice Wendling on Twitter: @pwendl. For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....