- © 2004 Canadian Medical Association or its licensors
In a well-structured study on the contentious subject of emergency department assessment of chest discomfort, Jim Christenson and associates1 conclude that the “miss rate” for acute coronary syndromes (ACS) was 5.3% at 2 Vancouver hospitals, more than 2 1/2 times that of our US counterparts.2
At first glance, this is devastating news. However, certain points might benefit from clarification. For example, of the 21 patients with ACS who were discharged from the emergency department, most had a diagnosis of chest pain not yet determined or atypical chest pain. But how many of these were discharged with empiric antianginal treatment (e.g., acetylsalicylic acid or nitroglycerin as required) and had definitive follow-up? Would such treatment of this subset of patients, if at low risk for ACS, not meet the standard of practice?3,4
In the ACS patients whose condition was truly missed, what were the negative consequences of not being admitted from the emergency department? In the one case of death, what was the temporal relationship between death and the ACS diagnosis? In other words, is it likely that the outcome would have been prevented by admission on the index visit? Also, were there other adverse events in this subset of patients and, if so, is it likely that admission would have averted their occurrence?
The authors refer to the study by Pope and colleagues2 as the standard. However, I believe there are significant differences in the design of that study that would reduce the number of missed diagnoses. For example, those authors included all critically ill patients, whereas Christenson and associates, because of consent issues, did not. Pope and colleagues used only creatine kinase data in making the diagnosis, whereas Christenson and associates also used the data from the more sensitive and specific troponin assay. Pope and colleagues did not employ diagnostic data such as an outpatient stress test or angiogram results to capture more patients, but Christenson and associates did. I believe that if the design of the US study were to be used on the study population examined by Christenson and associates, the percentage of missed diagnoses would be considerably lower than 5.3%.
Once it is clarified whether appropriate treatment was rendered to the “missed” group after discharge and whether admission would have prevented any adverse events, then the significance of the 5.3% figure, in isolation, can be gauged.
Brian D. Steinhart Emergency Physician St. Michael's Hospital Toronto, Ont.
Footnotes
-
Competing interests: None declared.
References
- 1.
- 2.
- 3.
- 4.