Abstract DP121: PFO Closure Decision-Making in Stroke Prevention: Detailed Analysis of the PASCAL POSSIBLE Category

Background: The PFO-Associated Stroke Causal Likelihood (PASCAL) classification is a prospectively validated decision-making guide regarding PFO closure to prevent recurrent stroke. The algorithm classifies patients into three categories: PROBABLE, who have dramatic recurrent stroke relative risk reduction; POSSIBLE, who have a moderate relative stroke risk reduction; and UNLIKELY, who have no stroke risk reduction and magnified atrial fibrillation risk. The intermediate POSSIBLE category has two patient subsets: 1) high RoPE score risk but low risk PFO characteristics risk (atrial septal aneurysm or large shunt); and 2) low RoPE score risk but high risk PFO characteristics. Whether these two patient subsets within the POSSIBLE category have differential benefit-harm with closure has not been previously investigated. Methods: Secondary analysis of individual participant-level data from the Systematic, Collaborative, PFO Closure Evaluation (SCOPE) consortium meta-analysis, including all six randomized trials of transcatheter PFO closure vs antithrombotic therapy alone. Results: Among 3740 patients, 37.0% were in the PASCAL PROBABLE category, 48.4% in POSSIBLE, and 14.6% in UNLIKELY. Among 1811 patients in the POSSIBLE group, 954 (52.7%) had high RoPE score but no high-risk PFO and 857 (47.3%) a low RoPE score but a high-risk PFO characteristic. Patient characteristics within each POSSIBLE subgroup are shown in Table 1. Patients with a high risk PFO but low RoPE score, compared with patients with a high RoPE score but without high risk PFO, were older, more often male, more frequently with vascular risk factors of hypertension, hyperlipidemia, diabetes, prior stroke and smoking, and less often had superficial infarct on imaging. Rates of the primary composite outcome of recurrent stroke, vascular death, or early all-cause mortality and of recurrent stroke alone are shown in Table 2 and Figure 1. Higher recurrent event rates were observed in the High risk PFO/Low RoPE subgroup than the High RoPE/Low risk PFO groups in both the device closure and medical management groups. However, both subgroups showed beneficial reduction in recurrent stroke with PFO closure. Conclusion: Within the PASCAL POSSIBLE category, High risk PFO/Low RoPE patients and High RoPE/Low risk PFO patients are equally frequent, differ in several characteristics, and the High risk PFO/Low RoPE patients have higher recurrent stroke risk. However, both POSSIBLE subgroups benefit from PFO closure.