Postoperative AF incidence following mitral repair exceeds 30%, with an annual stroke rate reaching up to 1%. In non-valvular AF cases, 90% of left atrial thrombi were located in the LAA, compared to 57% in valvular AF. The LAAOS III trial, published in 2021, demonstrated a statistically significant reduction in ischemic stroke and systemic embolism with LAA closure in patients with pre-existing AF and a CHADS2-VASc score >2, compared to those without closure.
The present study addresses the potential benefits of LAA closure in patients without AF or recent AF episodes.
A cohort of 1036 patients undergoing robotic mitral repair through right thoracotomy between 2005 and 2020 at a single institution was selected. Exclusion criteria included AF episodes within 30 days preoperatively, presence of a transcatheter LAA closure device, prior embolic events, and active endocarditis, resulting in a sample size of n=764. LAA closure was achieved through double-layer continuous sutures via left atriotomy post-mitral valve repair. Postoperative anticoagulation was not standard except for indications or persistent AF. Data on post-discharge embolic events and AF episodes were collected using California State emergency and hospitalization records.
The primary objective of this study was to compare long-term stroke/transient ischemic attack (TIA) risk in mitral repair patients based on whether or not LAA was closed.
A change in surgical practice occurred in the center after 2014, with LAA closure becoming routine in mitral repair patients without AF indications. Consequently, LAA was closed in 15 out of 284 patients (5.3%) before 2014 and in 416 out of 480 (86.7%) afterward. Both groups shared similar baseline surgical indications; however, the LAA closure group included older patients. Preoperative variables, such as age, gender, and comorbidities, showed no significant differences.
Clamp and pump times were shorter in the LAA closure group (p<.0001), as were reoperations for bleeding (p=.02). This may be attributed to greater surgical experience with robotic mitral repair post-2014.
Despite higher postoperative AF incidence in the LAA closure group (31.8% vs. 25.2%; p=.047), patients on warfarin at discharge (7.4% vs. 3.6%; p=.02), and those on antiarrhythmics (not beta-blockers) (32.9% vs. 18%; p<.001), the LAA closure group showed lower postoperative stroke/TIA incidence (2 vs. 7 cases). The cumulative 8-year stroke/TIA incidence was 2% in the LAA closure group compared to 6.3% in controls (HR, 0.26; 95% CI 0.09-0.78; p=.02).
This study suggests that routine LAA closure is safe in patients without prior AF episodes and may reduce late stroke/TIA incidence, as indicated by the LAAOS III trial (in anticoagulated patients with established AF).
COMMENTARY:
Postoperative AF following mitral repair, observed in 31.8% of the LAA closure group, is a common issue in ICUs and cardiac surgery units. Thus, it is logical to consider addressing the ultimate risk posed by AF—embolisms. Indeed, a previous blog entry discussed the synergistic protective effect of LAA closure.
There is no doubt that anticoagulation in high-risk patients is first-line treatment, as emphasized in the LAAOS III trial, which clarifies that LAA closure offers additional embolic protection when a patient is correctly anticoagulated. In the LAAOS III subgroup analysis, results were not significant in non-AF patients (HR, 0.76; 95% CI 0.5-1.1), motivating the authors to conduct this study.
One major limitation of this study was tracking new AF episodes in discharged patients, as incidence might be underestimated due to lack of monitoring. Only emergency or hospitalization data were recorded. Additionally, anticoagulation duration was not documented, an essential factor in interpreting these results.
Another limitation involves LAA closure technique; only continuous double sutures were used, excluding clip or amputation methods. Postoperative echocardiography did not confirm complete LAA closure, despite literature questioning the full efficacy of the double-suture method, with incomplete closure rates of up to 30%.
A notable finding is the statistically significant association between double-suture LAA closure and increased postoperative AF episodes. Consequently, the reduced stroke incidence in this subgroup may reflect increased use of antiarrhythmics and anticoagulation at discharge.
The authors suggest that the shorter pump and clamp times may contribute to better outcomes in the LAA closure group. However, this impact is unlikely, as LAA closure success depends on complete occlusion, irrespective of time.
The question arises: could prophylactic LAA closure increase postoperative AF risk? If so, which technique is the least arrhythmogenic?
In conclusion, AF is common (up to 30%) post-mitral repair. Studies show that LAA closure, in anticoagulated patients with established AF, significantly reduces embolic events. In patients undergoing mitral repair without prior AF episodes, LAA closure has proven safe and may reduce stroke/TIA incidence.
Given this study’s limitations, a prospective, multicenter, randomized study that includes various LAA closure techniques would be ideal for assessing efficacy in reducing potential post-AF stroke/TIA risks.
This article raises critical questions about “preventing” risks via surgical procedures when this is not the primary indication. The most notable question from this study’s results: are we inducing AF by closing the LAA in previously AF-free patients? Is there a less arrhythmogenic method?
REFERENCE:
Chikwe J, Roach A, Emerson D, Chen Q, Rowe G, Gill G, et al. Left atrial appendage closure during mitral repair in patients without atrial fibrillation. J Thorac Cardiovasc Surg. 2024 Jul;168(1):86-93.e5. doi: 10.1016/j.jtcvs.2023.02.030.
