Embolic Stroke of Undetermined Source and Patent Foramen Ovale Risk of Paradoxical Embolism Score Validation and Atrial Fibrillation Prediction

Background and Purpose: The Risk of Paradoxical Embolism (RoPE) score stratifies patients with stroke according to the probability of having a patent foramen ovale (PFO), which (through Bayes theorem and simple assumptions) can be used to estimate the probability that a PFO is pathogenic in a given subgroup of patients with specific features (ie, a given RoPE score value): a higher PFO prevalence corresponds to a higher probability that a PFO is pathogenic. Among alternative mechanisms in embolic stroke of undetermined source (ESUS), the actual stroke cause may be covert atrial fibrillation. We aimed to validate the RoPE score in a large ESUS population and investigate the rate of stroke recurrence and new incident atrial fibrillation during follow-up according to PFO status and RoPE score. Methods: We pooled data of consecutive patients with ESUS from 3 prospective stroke registries. We assessed RoPE score's calibration and discrimination for the presence of PFO (and consequently for the probability that it is pathogenic). Multivariate logistic regression analysis was performed to identify factors associated with PFO. Results: Among 455 patients with ESUS (median age 59 years), 184 (40%) had PFO. The RoPE score's area under the receiver operating characteristic curve was 0.75. In addition to RoPE score variables, absence of left ventricular hypertrophy, absence of atherosclerosis, and infratentorial lesions were independently associated with PFO. In patients with PFO and RoPE 7 to 10, PFO and RoPE 0 to 6, and without PFO, new incident atrial fibrillation rate was 3.1%, 20.5%, and 31.8%, respectively (log-rank test=6.28, P=0.04). Stroke recurrences in patients with likely pathogenic PFO were not statistically different from other patients. Conclusions: This multicenter study validates the RoPE score to predict the presence/absence of PFO in patients with ESUS, which strongly suggests that RoPE score is helpful in identifying patients with ESUS with pathogenic versus incidental PFOs. Left ventricular hypertrophy, atherosclerosis, and infratentorial stroke may further improve the score. Low RoPE scores were associated with more incidental atrial fibrillation during 10-year follow-up.