There are a few published series20,34–36 reporting the prevalence of OSA in patients with acute cardiovascular events. However, with the exception of one study,20 all34–36 included a heterogeneous patient population such as stable angina, unstable angina, myocardial infarction, and heart failure. Our study focused on a homogeneous group of patients who had experienced a first acute myocardial infarction. Using an AHI of 15 events per hour as a cutoff, we found a relatively high prevalence of OSA (65.7%). Because patients with known OSA were excluded from this study, it is conceivable that we might have underestimated the actual prevalence among all patients admitted to the hospital with acute myocardial infarction. Nonetheless, the prevalence is higher than that reported in previously published series. In a study of Japanese patients admitted to the hospital with acute myocardial infarction, Nakashima and associates20 reported a prevalence of 43%. However, in their study, the sleep study was done 14 to 21 days after hospital admission, whereas in ours, it was done 2 to 5 days after hospital admission. Skinner and associates34 reported an OSA prevalence of 46% in 26 patients admitted with acute coronary syndrome. In another two studies35,36 using a lower AHI cutoff of 10 (instead of 15), the prevalence of OSA in patients with acute coronary syndrome ranged from 57 to 66.4%. There are a number of possible reasons for the difference in the prevalence of OSA between studies. Differences in the timing of sleep studies, diagnostic criteria applied, diagnostic device used, and characteristics of the study population can all affect the measurement prevalence. It was suggested that performing sleep studies during acute cardiovascular events might result in a high rate of false-positive diagnoses due to detection of transient abnormalities.34 However, an alternative explanation to their findings was that these patients were not really “false positives” but demonstrated true sleep-disordered breathing, albeit transient, during an acute cardiac illness.37 Furthermore, it could be argued that identifying and treating OSA, a risk factor known to influence long-term outcomes in patients with cardiovascular disease, should be done as early as possible when patients are in the hospital where facilities for prompt investigation are more likely to be available. The issue on the ideal timing of investigation for OSA after an acute cardiovascular event remains unresolved. Another possible explanation for the higher prevalence observed in our series is selection bias. Patients who suspected they might have OSA might have been more likely to participate in the study. In fact, 45.5% of the male patients screened (117 of 257 patients) underwent the overnight sleep study, whereas only 9.0% of the female patients screened (3 of 33 patients) underwent the test. Besides, acute myocardial infarction represents the highest risk category among patients with unstable coronary syndrome. It is possible that the prevalence of OSA increases along the spectrum of severity of coronary artery disease.