I disagree with Dr. Anthony Papadatos' assertion that you should never (or almost never?) send a patient home after a single negative troponin. (EMN 2005;27:32.)
Miller et al (Ann Emerg Med 2004;44:565) showed that even today emergency physicians discharge to home many chest pain patients who have acute myocardial infarction. In this study of 17,000 patients with chest pain, 2,992 were thought to have noncardiac chest pain, and 75 percent of them were discharged.
Physicians were blinded to troponin results, and did not use them in their decision regarding “noncardiac chest pain.” Of these 2,992 patients, 1,592 (53.2%) had cardiac biomarkers drawn, with 45 (52%) positive. Thus, 2.8 percent of cardiac biomarkers in patients with supposed “noncardiac chest pain” were positive. These patients would have been changed to “cardiac” CP if the physician had been aware of the troponin. These data represent an underestimation because there were a variety of troponin assays used in the multiple different institutions that took part in the study, but they only counted the troponin as positive if it equaled 1.0 ng/ml. For most of the assays, levels from 0.1 ng/ml to 0.99 ng/ml were in the range of positive or at least not negative.
I published our experience at Hennepin County Medical Center where under certain conditions we discharge patients with a single negative troponin drawn at least six hours after symptom onset. (J Emerg Med 2004;26:401.) Of 390 patients with chest pain and a negative troponin, only two (0.34%) had an adverse 30-day event; both were small troponin leaks and both were also protocol violations. This management was only for selected patients. During the study period, 52 percent of chest pain patients were admitted, and 48 percent were discharged home while only seven percent (15% of discharges) had a troponin determination before being sent home.
Would you rather miss many MIs (harming patients) in the belief that you are safe from the lawyers who accuse physicians of relying on a single negative troponin? Or would you rather detect those patients, also knowing that some of those you send home with a negative troponin will probably also have MI? I would choose the latter option. Ultimately there is no evidence to support that such a management strategy puts the physician at any higher legal risk.
Stephen W. Smith, MD