Rationale: Obstructive sleep apnea (OSA) is an independent risk factor for stroke, but little is known about the role of continuous positive airway pressure (CPAP) on mortality in stroke patients. Objective: To analyze the independent impact of long-term CPAP treatment on mortality in ischemic stroke patients. Measurements and Main Results: Of 223 patients consecutively admitted for stroke, a sleep study was performed on 166 of them (two months after the acute event). Thirty-one had an apnea-hypopnea index (AHI)<10; 39 had an AHI between 10-19 and 96 had an AHI20. CPAP treatment was offered when AHI 20. Patients were followed up in our outpatient clinic at 1, 3 and 6 months, and for every 6 months thereafter for 5 years (prospective observational study). Mortality data were recorded from our computer database and official death certificates. The mean age of subjects was 73.3 ± 11 years (59% males); Mean AHI was 26 (for all patients with a predominance of obstructive events). Patients with an AHI20 who did not tolerate CPAP (n=68) showed an increase adjusted risk of mortality (HR 2.69 [95%CI: 1.32-5.61] compared to patients with IAH<20 (n=70) and an increased adjusted risk of mortality (HR 1.58 [95%CI:1.01-2.49]; p=0.04) compared to patients with moderate-severe OSA who tolerated CPAP (n=28); There were no differences in mortality between patients without OSA, patients with mild disease and patients who tolerated CPAP. Conclusions: Our results suggest that long-term CPAP treatment in moderate-severe OSA and ischemic stroke is associated with a reduction in the excess risk of mortality seen in these patients.