Featured Article: Ohman EM, Armstrong PW, Christenson RH, Granger CB, Katus HA, Hamm CW, et al. Cardiac troponin T levels for risk stratification in acute myocardial ischemia. N Engl J Med 1996;335,1333-41. 2 Sir William Osler taught us that observations from patients could inform us. In my case, an event during my internship in 1981 led me to question how accurate the diagnosis of myocardial infarction (MI) 3 was. A 28-yearold man was admitted after he collapsed at the end of the Dublin

more »

... n race. His electrocardiogram (ECG) was very abnormal. Myocardial biomarkers were increased, with high total creatine kinase (CK) and CK-MB activities. Later coronary angiograms demonstrated normal coronary arteries. I followed this up by a formal study showing that the prior "gold standard" markers of MI, CK, and CK-MB were frequently increased in marathon runners, while infarct-avid scanning failed to detect any MI (1 ). At the time, troponin testing was in its infancy and a standardized test was not available. This interest in biomarkers continued after I came to Duke University in 1987, but focused more on the kinetics and the relationship to the ECG. Collaboratively with Galen Wagner and Rob Christensen, we performed several studies on a small cohort of patients with established MI who received fibrinolytic therapy to establish reperfusion (2 ). I enjoyed digesting the data together and exploring it in several ways. It became clear that small studies, although very carefully performed, were not going to move the biomarker field forward in any substantial way. On rounds in the Duke Coronary Care Unit, Rob Califf also famously announced that he was not sure why we were doing all the CK and CK-MB testing, as "it does not change what we do to our patients." This stimulated me to think of ways we could change this, and the need for large cohorts of patients to have an impact.