Background: In patients with cryptogenic stroke (CS), a patent foramen ovale (PFO) can be incidental or pathogenic. The Risk of Paradoxical Embolism (RoPE) score has been developed to determine the likelihood that a PFO is pathogenic or incidental using clinical variables. We hypothesize that echocardiographic features and conditions promoting paradoxical embolism differ between patients with pathogenic and incidental PFOs.
Methods: The International PFO Consortium collects clinical, radiological and echocardiographic data of patients with CS and PFO. In the original RoPE score, a value of 0-6 was classified as a low RoPE score and 7-10 as a high score. Since information on cortical versus deep stroke location (one of the items on the RoPE score) was not available, we used two alternative approaches to stratify for PFO pathogenicity. In a first approach, we used a 9-point score and lowered the cut-off for dichotomization by 1 point (RoPE score 0-5 vs 6-9). In a second approach, patients with a RoPE score of 6 were excluded since they could either be classified as low or high RoPE score depending on stroke location. The associations between RoPE stratum and echocardiographic features (atrial septal aneurysm (ASA), right-to-left shunt (RLS) at rest and large RLS) as well as conditions promoting paradoxical embolism (deep vein thrombosis (DVT), pulmonary embolism (PE) and Valsalva maneuver (VM) were studied.
Results: We analyzed 1044 CS patients with a PFO. Average age was 55 (SD 16) and 635 patients (61%) were male. Preceding VM was more frequent in patients with a high vs low RoPE score in both analyses: 11% vs 5% (OR: 2.1 95%CI 1.3-4.3) and 10% vs 5% (OR: 2.0 95%CI 1.2-3.6). The distribution of ASA (35% vs 34% and 32% vs 34%, in the first and the second analysis respectively), RLS at rest (28% vs 28% and 29% vs 28%), large RLS (67% vs 66% and 65% vs 66%), PE (2% vs 2% and 1% vs 2%), and DVT (4% vs 4% and 3% vs 4%) did not differ by RoPE stratum.
Conclusion: In patients with CS, preceding VM was significantly associated with pathogenic PFO, while echocardiographic features or conditions promoting paradoxical embolism were not. The formation of a significant right-to-left pressure gradient at the atrial septum level appears to play a substantial role in the pathogenicity of PFO.