To the Editor The article by Juo et al1 addresses the important issue of temporal trends in perioperative myocardial infarction after high-risk vascular surgery. We congratulate the authors on their contribution; however, we have some reservations concerning the methods of this study.
Definitions of myocardial infarction used in the database are out of date in light of recent developments in perioperative cardiac monitoring. The diagnostic threshold for myocardial infarction in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) cohort was defined as new elevation in troponin values greater than 3 times the upper level of the reference range. Furthermore, troponin levels were not measured routinely but only when the clinicians suspected ischemia. The problem with this approach is that 65% of perioperative myocardial infarctions are asymptomatic.2