Résumé : Background: The left atrial septal pouch (LASP) is formed by the caudal fusion of the area of overlap of the septum primum and the septum secundum, leaving an opening toward the left atrium. The association between LASP and stroke has not been validated by the previous studies. Methods: The prevalence of the LASP was determined in 223 ischemic stroke patients and 223 control subjects with other cardiac pathologies, in a monocentric retrospective case–control study design. Stroke subtypes were defined according to the modified TOAST criteria. Results: The mean age was 66 ± 15, 54% males, with a high prevalence of cardiovascular risk factors. The prevalence of the LASP was 81 (18%), irrespective of age or pathology. The number of LASP was similar in the stroke and control groups (18% vs. 19%, P = 0.7), as well as in the cryptogenic stroke subgroup (16%, P = 0.6). LASP was not associated with ischemic stroke on univariate (OR = 1.095; 95% CI = 0.676–1.772; P = 0.7) and multivariate logistic regression analysis (OR = 1.004; 95%CI = 0.574–1.758, P = 0.98). There was no statistical association between LASP and cryptogenic stroke on univariate (OR = 1.26; 95%CI = 0.526–3.016; P = 0.6) or multivariate analysis (OR = 0.705; 95%CI = 0.193–2.577, P = 0.6). The association of LASP to AF, left ventricular dysfunction, and thrombophilia did not lead to a higher incidence of stroke (OR = 0.99; 95%CI = 0.37–2.66; P = 0.99). Conclusion: Our study did not show any association between LASP and ischemic stroke. A septal pouch was present in 18% of the population. Other associated risk factors need to be considered to incriminate the septal pouch as the etiology of a stroke.