Anatomical features of PFO in patients with cryptogenic stroke in comparison to autopsy results

Abstract Background In young patients suffering a stroke, well over 30% is considered cryptogenic stroke and may be associated with a PFO. The anatomical characteristics of stroke-related PFOs that allow right-to-left shunting can be assessed with TEE. The prevalence of a PFO in autopsy studies is approximately 25%, suggesting that the a priori chance of finding a possible "bystander-PFO" in young stroke victims is substantial. Purpose To assess the anatomical characteristics of PFO in patients with cryptogenic stroke and compare these with previously published results from autopsy in the general population. Methods Data on all consecutive patients that underwent PFO closure from 2016 to 2022 were collected. Anatomical PFO characteristics, including PFO diameter, tunnel length and the presence of atrial septal aneurysm, were measured with TEE during the closure procedure and analyzed by two cardiologists. Results In total 228 patients were enrolled. The mean age was 43±11 years, 116 patients (50.9%) were male, the mean RoPE-score was 7.0±1.4. The most frequent shunt grade measured with contrast bubble study during Valsalva maneuver was 3 (>25 bubbles, without opacification). Patients were divided in two groups depending on how the PFO diameter was measured, unstretched (n=119) and stretched (n=109). The mean PFO diameter was 5.4±2.6mm in the unstretched group and 10.9±3.6mm in the stretched group compared to 4.9±2.6mm in previous autopsy (p=0.192 and p<0.001, respectively). The percentage of patients with a PFO size larger than 10mm was 2% in autopsy compared to 6% in the unstretched group (p=0.109) and 52% in the stretched group (p<0.001). The mean PFO tunnel length in our combined cohort was 7.8±2.9mm. Conclusion The difference in PFO diameter found in this study indicates that PFOs found during published autopsy series, which are present in approximately 25% of the population, are significantly smaller than those related to a cryptogenic stroke. The exact prevalence of PFOs in the population that are comparable to those in cryptogenic stroke patients remains thus unknown. Future studies should define which characteristics could classify or rule out a PFO prone for associated clinical conditions, such as cryptogenic stroke.Distribution of PFO size