A new study has found that postoperative atrial fibrillation (POAF) lasting two days or more is associated with worse long-term survival after cardiac surgery. Even after close matching on operative variables and comorbidity burden, patients with prolonged atrial fibrillation had significantly worse survival compared with patients who experienced POAF less than two days. According to researchers, the results highlight the importance of careful monitoring of this population.
“I think that the message is to follow these patients long term,” said Martin I. Sigurdsson, MD, PhD, of the Department of Anesthesiology, Perioperative and Pain Medicine at Brigham and Women’s Heart & Vascular Center, in Boston. “Usually, at four to six weeks post-surgery, clinicians stop thinking about atrial fibrillation as a problem, but perhaps we should be following this population a little more carefully and potentially consider additional treatment.”
As Dr. Sigurdsson reported at the 2016 annual meeting of the Society of Cardiovascular Anesthesiologists (abstract 180), POAF after cardiac surgery has been associated with decreased survival, partially explained by the higher comorbidity burden of these patients. It is unclear, however, whether mortality is associated with a longer duration of POAF.
Controls Matched Carefully
In this prospectively collected, single-center study, long-term survival was compared between patients with POAF and propensity score–matched patients without POAF.
All patients underwent cardiac surgery and were matched by age, sex, prior atrial fibrillation, operative characteristics and comorbidity burden assessed by the Elixhauser Comorbidity Index, Dr. Sigurdsson noted. Researchers then compared survival for two subsets of patients with POAF and their propensity score–matched controls: those with POAF duration less than two days and patients with prolonged POAF of two days or more.
A total of 513 of 1,709 patients (30%) with available data had POAF.
Patients with POAF were older, had more comorbidities, had experienced longer cardiopulmonary bypass time and were more likely to have had valve surgery. There was also a slightly higher incidence of prior diagnosis of atrial fibrillation, Dr. Sigurdsson reported.
After propensity matching, however, there was no difference in any of the matching values between patients with POAF and propensity score–matched controls.
“After thorough matching on multiple operative variables and the overall burden of comorbidities in these groups, we were able to find a very well-controlled group of patients for every single patient who had postoperative atrial fibrillation,” said Dr. Sigurdsson.
The researchers compared survival between patients with POAF and their propensity score–matched controls. Long-term survival was significantly worse for patients with POAF than for control patients (hazard ratio [HR], 1.43; 95% CI, 1.11-1.86; P=0.006).
According to Dr. Sigurdsson, the most interesting—and novel—element of the study was the separation of patients into roughly two equal-sized groups based on duration of POAF: less than two days or two or more days, which included those who were discharged with atrial fibrillation.
The researchers found that patients with shorter-duration POAF had identical survival as their matched controls (HR, 0.91; 95% CI, 0.60-1.39; P=0.67). The survival of patients with atrial fibrillation longer than two days, however, was substantially much worse than their matched controls (HR, 1.43; 95% CI, 1.37-2.80; P<0.001).
“Prolonged postoperative atrial fibrillation is actually driving the signal for increased mortality for the overall group,” Dr. Sigurdsson observed.
As for patient characteristics, the biggest difference between duration-stratified groups was a history of atrial fibrillation. Dr. Sigurdsson noted that 14% of patients with sustained POAF and their controls had a history of atrial fibrillation versus 6.5% of those experiencing shorter atrial fibrillation and matched controls.
“These groups do differ,” he explained, “but perhaps this is just an indication that they have an overall higher burden of atrial fibrillation, so they are having shorter or longer bursts of atrial fibrillation.”
The biggest takeaway for Dr. Sigurdsson: “Follow these patients longer and consider treating them,” he said.
“These things are happening already, of course, but not in a standardized fashion. A specific protocol needs to be developed.”