Lalmohamed et al1 used epidemiologic analysis to test the association between total hip replacement (THR) or total knee replacement (TKR) and acute myocardial infarction (AMI). Not surprisingly, during the first 2 postoperative weeks, the risk of AMI was elevated in both populations of patients undergoing THR or TKR. The risk was elevated for 6 weeks in patients undergoing THR but only for 2 weeks in those undergoing TKR. It has been previously established that patients undergoing surgical procedures have an increased risk of MI.1 The risk factors for perioperative cardiac morbidity and mortality have been established for many years, and although different studies2- 3 find slightly different risk factors, there is remarkable consistency over time: age older than 60 years, coronary artery disease, peripheral vascular disease, congestive heart failure, recent MI, and the standard risk factors for coronary artery disease, including diabetes mellitus, hypertension, smoking, and hyperlipidemia. Occasionally, an investigator will suggest that one risk factor or another is no longer important, such as MI in the last 30 days, but subsequent studies will identify once again that recent MI, MI in the last 6 months, or MI in the last year remains a risk factor for subsequent MI. Epidemiologic studies are limited by the population of patients in the database. If no one performs elective surgery on a patient within 30 days of an AMI, then that variable will not be significant in epidemiologic analysis. Recent MI is still a risk factor for cardiac morbidity; it simply is not a significant risk factor identified in the study because there are no patients with that risk profile in the database. Failure to demonstrate that a risk factor is significant does not imply the risk factor is not still a clinical issue; it simply implies one could not demonstrate the effect with the database. Infrequently, a new perioperative risk factor is identified, such as erectile dysfunction.4 It is highly likely that these “new” risk factors are caused by peripheral vascular disease, which is highly associated with coronary artery disease rather than being a new independent perioperative risk factor.