Proposed Article Structure
This topic is best addressed through a clinically oriented structure that moves from rationale to trial design, then to efficacy interpretation and implications for surgical decision-making. The article therefore follows these sections: Highlights; Clinical Background and Unmet Need; Study Design and Methods; Key Results; Clinical Interpretation and Expert Commentary; Practice Implications; Limitations and Future Research; Conclusion; Funding, Registration, and Citation.
Highlights
In the OPINION randomized trial, adding surgical left atrial appendage occlusion to valve surgery in patients without pre-operative atrial fibrillation did not significantly reduce the 1-year composite of ischaemic stroke, transient ischaemic attack, or cardiovascular death.
The trial enrolled a relatively young valvular heart disease population with elevated thrombo-embolic risk by CHA₂DS₂-VASc score but no documented atrial fibrillation before surgery.
The neutral primary result contrasts with the proven benefit of left atrial appendage occlusion in patients with established atrial fibrillation, underscoring that benefit in one risk population should not be assumed in another.
Current practice should remain selective rather than routine: prophylactic surgical appendage closure in non-AF valvular surgery patients cannot be justified solely on the basis of the OPINION data.
Clinical Background and Unmet Need
The left atrial appendage is the dominant site of thrombus formation in atrial fibrillation, and surgical exclusion of this structure has become an increasingly important adjunctive strategy during cardiac surgery. The strongest evidence comes from patients with known AF. In the LAAOS III trial, surgical left atrial appendage occlusion performed during cardiac surgery reduced ischaemic stroke or systemic embolism in patients with AF receiving anticoagulation, firmly establishing the procedure in an appropriate high-risk population.
The more difficult question is whether the same logic should be extended to patients who do not have AF before surgery. Valve surgery patients occupy a grey zone. They often have enlarged atria, abnormal atrial substrate, haemodynamic stress, and a substantial risk of post-operative AF, reported across studies in roughly 15% to more than 50% depending on the population, rhythm monitoring intensity, and procedure type. Because thrombo-embolic risk may rise after surgery, and because the left atrial appendage is a plausible source of future emboli, some surgeons have considered prophylactic surgical left atrial appendage occlusion even in the absence of documented pre-operative AF.
However, biological plausibility is not the same as demonstrated clinical benefit. Patients without AF may have different stroke mechanisms, different baseline event rates, and a lower proportion of appendage-mediated embolism. In addition, any preventive procedure added to valve surgery must show meaningful outcome benefit to justify added operative steps and potential procedural trade-offs. The OPINION trial was designed to address this unresolved question directly.
Study Design and Methods
OPINION was a multicentre, open-label, randomized, superiority trial conducted at three cardiac surgery centres in China. The investigators enrolled patients undergoing mitral or aortic valve repair or replacement who had no documented pre-operative atrial fibrillation but had a CHA₂DS₂-VASc score of at least 2, reflecting an elevated thrombo-embolic risk profile.
Participants were randomly assigned in a 1:1 ratio to undergo surgical left atrial appendage occlusion during the index valve operation or to receive no appendage occlusion. Randomization occurred in the operative setting, and the trial used an intention-to-treat framework for the primary analysis. This design is appropriate for preserving the benefits of random allocation and reflects real-world treatment assignment.
The primary endpoint was clinically important and pragmatic: a composite of ischaemic stroke, transient ischaemic attack, or cardiovascular mortality at 1 year. The choice of a composite endpoint increased efficiency, but it also introduces interpretive challenges because cardiovascular death is not necessarily mechanistically linked to appendage closure in the same way that thrombo-embolic events are.
Between April 2021 and June 2024, 2157 patients were enrolled and randomized. After 39 withdrew informed consent, 2118 participants remained in the intention-to-treat population: 1062 in the surgical left atrial appendage occlusion group and 1056 in the control group.
Baseline characteristics were well balanced. Mean age was 55.5 years in the intervention group and 55.6 years in the control group. Women represented about one-third of the cohort. Mean CHA₂DS₂-VASc score was approximately 2.9 in both groups, and median EuroSCORE II was low at about 1.6%, indicating an overall population at modest operative risk despite elevated thrombo-embolic risk by score.
Key Results
The primary composite endpoint occurred in 73 patients in the surgical left atrial appendage occlusion arm and 87 patients in the control arm at 1 year, corresponding to event rates of 6.9% and 8.2%, respectively. The hazard ratio was 0.83 with a 95% confidence interval of 0.61 to 1.14, and the P value was .25.
Numerically, the event rate favored appendage occlusion, but the difference was not statistically significant. The confidence interval is informative. It includes the possibility of modest benefit, but it also includes no effect and excludes a large treatment advantage. Put simply, the trial does not support a conclusion that routine prophylactic appendage closure meaningfully reduces 1-year stroke, TIA, or cardiovascular death in this specific population.
Several features of the result deserve emphasis. First, the absolute risk difference was modest, about 1.3 percentage points. Second, because the confidence interval crossed unity, the observed numerical reduction may reflect chance. Third, the composite includes outcomes with different causal pathways; if the benefit of appendage closure is concentrated mainly on cardioembolic stroke, inclusion of cardiovascular death may dilute a true but small signal.
The abstract does not provide a detailed breakdown of the individual endpoint components, secondary outcomes, post-operative AF incidence, anticoagulation patterns, imaging confirmation of appendage closure completeness, or perioperative safety events. These details are important for full interpretation. For example, if post-operative AF developed frequently but was short-lived, or if anticoagulation practices differed after surgery, the relationship between appendage management and thrombo-embolic risk might be altered. Similarly, incomplete appendage exclusion can reduce efficacy, and without procedural adjudication data it is difficult to know whether technical success influenced outcomes.
Clinical Interpretation and Expert Commentary
The central message from OPINION is straightforward: evidence supporting surgical left atrial appendage occlusion in patients with known AF should not be extrapolated automatically to patients without AF undergoing valve surgery.
This distinction matters because the mechanistic foundation differs. In established AF, blood stasis and altered appendage flow create a clear substrate for thrombus formation within the left atrial appendage. By contrast, in patients without AF, stroke risk may arise from a broader mixture of mechanisms, including aortic atheroembolism, ventricular dysfunction, vascular disease, perioperative haemodynamic disturbances, or later incident arrhythmia. If only a subset of events originate in the appendage, then routine occlusion would be expected to yield a smaller overall effect.
The trial also raises questions about patient selection. The mean age of approximately 55 years is younger than many Western valve surgery cohorts and younger than typical AF stroke-prevention populations. Although the participants were high risk by CHA₂DS₂-VASc threshold, the absolute event burden may still have been too low, or the follow-up too short, to reveal a clear preventive effect. A 1-year horizon may be particularly limiting for a prophylactic structural intervention intended to reduce future embolic events, especially if some benefit would accrue only after late development of AF.
Another important issue is whether CHA₂DS₂-VASc is the right enrichment tool in patients without AF. The score was developed and validated primarily in AF populations for stroke risk estimation. Using it to select non-AF surgical patients is clinically sensible but not fully validated for this procedural decision. Better phenotyping may be needed, perhaps incorporating left atrial size, appendage morphology, rhythm-monitoring burden, biomarkers, or history of atrial cardiomyopathy.
From a guideline perspective, current recommendations strongly support surgical left atrial appendage exclusion in patients with AF undergoing cardiac surgery when technically feasible, but they do not clearly endorse routine prophylactic closure in patients without AF. OPINION reinforces that restraint. Until additional evidence emerges, surgeons and heart teams should avoid assuming that a low incremental procedural burden necessarily translates into measurable net clinical benefit.
Practice Implications
For clinicians, the practical takeaway is that routine appendage occlusion during mitral or aortic valve surgery should not become standard for patients who have no documented pre-operative AF solely because they have a CHA₂DS₂-VASc score of 2 or greater.
This does not mean appendage closure should never be considered in non-AF patients. Individualized decisions may still be reasonable in selected circumstances, such as strong suspicion of occult AF, marked atrial enlargement, prior embolic phenomena of uncertain source, or high anticipated likelihood of recurrent atrial arrhythmia. But those decisions should be framed as case-by-case judgments rather than evidence-based routine care.
The trial also highlights the importance of rigorous post-operative rhythm surveillance. If a substantial proportion of valve surgery patients develop early or late AF, systematic ECG monitoring and appropriate anticoagulation pathways may be more impactful than empirical appendage closure alone. Future integrated strategies may need to combine rhythm detection, anticoagulation algorithms, and selective structural prevention.
Limitations and Future Research
Several limitations temper interpretation. The trial was open label, although this is often unavoidable in surgical trials. Follow-up was limited to 1 year, which may be too short for a prophylactic intervention aimed partly at future AF-related embolism. The abstract does not detail secondary efficacy outcomes, perioperative complications, completeness of appendage occlusion, or subgroup analyses by valve type, atrial size, or post-operative AF. Generalizability outside the studied centres and population also requires caution.
Future studies should address whether longer-term follow-up changes the estimate of benefit, whether certain phenotypes derive greater value, and whether appendage closure is more effective when paired with systematic long-term rhythm monitoring. Trials could also examine whether patients with atrial cardiomyopathy but no overt AF represent a more biologically coherent target population than those selected by CHA₂DS₂-VASc score alone.
Conclusion
The OPINION trial provides important randomized evidence in a previously understudied area of cardiac surgery. In patients undergoing mitral or aortic valve surgery who had no pre-operative atrial fibrillation but had CHA₂DS₂-VASc scores of at least 2, routine prophylactic surgical left atrial appendage occlusion did not significantly reduce the 1-year risk of ischaemic stroke, transient ischaemic attack, or cardiovascular death.
Clinically, the study argues against routine adoption of this strategy in non-AF valvular surgery patients and supports a more selective, individualized approach. Scientifically, it reminds us that structural stroke-prevention strategies must be matched to the biology of the population being treated. The next phase of research should focus on better risk stratification, longer follow-up, and integrated approaches to post-operative AF detection and embolic prevention.
Funding, Registration, and Citation
Trial registration: ChiCTR.org registry ChiCTR2100042238.
Publication: Yuan X, Ju F, Wu H, Zhao Y, Wang X, Liu S, Wang X, Zhu J, Su P, Xu F, Feng W, Yang Y, Wang Y, Sun H. Surgical left atrial appendage occlusion in valvular heart disease without atrial fibrillation: the OPINION trial. European Heart Journal. 2026 May 21;47(20):2440-2448. PMID: 40888584.
The abstract provided does not specify funding details; readers should consult the full-text article for the complete funding statement and disclosures.
References
1. Yuan X, Ju F, Wu H, Zhao Y, Wang X, Liu S, Wang X, Zhu J, Su P, Xu F, Feng W, Yang Y, Wang Y, Sun H. Surgical left atrial appendage occlusion in valvular heart disease without atrial fibrillation: the OPINION trial. European Heart Journal. 2026;47(20):2440-2448. PMID: 40888584.
2. Whitlock RP, Belley-Cote EP, Paparella D, et al. Left Atrial Appendage Occlusion during Cardiac Surgery to Prevent Stroke. New England Journal of Medicine. 2021;384(22):2081-2091. PMID: 33999547.
3. Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the EACTS. European Heart Journal. 2021;42(5):373-498. PMID: 32860505.
4. January CT, Wann LS, Calkins H, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation. Circulation. 2024;149(1):e1-e156. PMID: 38033089.