Background and purpose: Previous studies showed that insular strokes are associated with electrocardiographic (ECG)- changes.However, they did not take into account the 1st ECG recorded at admission, but continuous ECG recorded up to 72 hours after onset. Whether these changes are the consequence of the infarct, or are associated with a cardiac source of cerebral ischemia, remains unsettled. If ECG changes are the consequence of insular infarcts, they should not have developed by the time of admission. The aim of this study was to test the hypothesis that ECG changes in patients with insular infarcts are not present at admission. Methods: We recruited consecutive patients admitted within 48 hours (median 3 hours) after the onset of symptoms of acute hemispheric cerebral ischemia. We compared ECG variables between patients with and without insular infarcts, and with left and right insular infarcts. Results: The study population consisted of 208 patients (94 men; median age: 69 years). Seventy patients had a recent insular infarct (right in 33). ECG variables did not significantly differ between patients with and without insular infarcts, and with left and right insular infarcts. These results were not explained by a lack of statistical power (1- β ≥ 0.90). Conclusion: The lack of statistical link between insular infarcts and ECG changes at admission, suggests that ECG changes are not associated with the cause of insular infarcts, but are their consequence. Background and purpose: Previous studies showed that insular strokes are associated with electrocardiographic (ECG) -change. However, they did not take into account the 1st ECG recorded at admission, but continuous ECG recorded up to 72 hours after onset. Whether these changes are the consequence of the infarct, or are associated with a cardiac source of cerebral ischemia, remains unsettled. If ECG changes are the consequence of insular infarcts, they should not have developed had the time of admission. The aim of this study was to test the hypothesis that ECG changes in patients with insular infarcts are not present at admission. Methods: We recruited consecutive patients admitted within 48 hours (median 3 hours) after the onset of symptoms of acute hemispheric cerebral ischemia. Weighed ECG variables between patients with and without insular infarcts, and with left and right insular infarcts. Results: The study population consisted of 208 patients (94 men; median age: 69 years). S These variables were not significantly differ between patients with and without insular infarcts, and with left and right insular infarcts. These results were not explained by a lack of statistical power (1- β ≥ 0.90). Conclusion: The lack of statistical link between insular infarcts and ECG changes at admission, suggests that ECG changes are not associated with the cause of insular infarcts, but are their consequence.