Background: Although several authors have studied the association between patent foramen ovale (PFO) and ischaemic stroke, the matter is still controversial; few have suggested an association between cryptogenetic stroke and PFO, while others have denied this association. The aim of this study was to evaluate PFO prevalence in the whole ischaemic stroke population, independently from age and stroke subtypes and to identify the characteristics associated with the presence of PFO. Methods: SISIFO study was a multicenter, prospective, single-wave, cross-sectional survey conducted on consecutive patients with acute ischemic stroke admitted to selected clinical centres. Data regarding vascular risk factors were registered for each patient; all patients underwent computed tomography scan and/or magnetic resonance imaging of the brain; an electrocardiogram and standard laboratory blood tests were performed. A Doppler ultrasound study of extra-cranial arteries was performed too. The cases were classified according to TOAST and OCSP criteria. Each patient underwent transcranial Doppler or transcranial color-coded duplex sonography with bubble test as diagnostic tool for right-to-left-shunt. Where right-to-left shunt was detected, PFO presence was confirmed by echocardiography. Findings: 1,130 consecutive patients were included. We found a PFO in 247 (21.9%; 95% CI, 19.5-24.3%) patients; PFO was present in 23.5% of patients with cryptogenic stroke and in 21.3% of patients with stroke of known causes; this difference was not statistically significant. At the univariate analysis, decreasing age, hypertension, diabetes mellitus, and atrial fibrillation, and stroke characteristics such as NIHSS, OCSP and TOAST were predictors of PFO presence. At the multivariate analysis, we found a significant interaction between age and OCSP syndrome. Being LACI the reference category, the prevalence of PFO in PACI and POCI decreased significantly along with age, whereas there was no change in TACI. Conclusion: If any relationship exists between stroke and PFO, this is more likely in PACI and POCI at a younger age. Our results are consistent with recent findings that underline PFO alone must not be considered a significant independent predictor for stroke; so the presence of PFO alone doesn't permit rushed causal correlations or ‘therapeutic aggressiveness'.