Coronary CT angiography for acute chest pain.

Each year, more than 6 million people in the United States go to emergency departments because of acute chest pain. Some of them have coronary artery disease, but most have diseases that are unrelated to the heart or they have no discernible physiological conditions. The vast majority will undergo various diagnostic tests, and many will be held for observation and may be admitted to the hospital. This approach, however, is guaranteed to lead to unnecessary stays in the emergency department and to overtreatment. More disturbingly, it is unclear whether this approach actually results in better outcomes than a more conservative one. In this issue of the Journal, Hoffmann et al.1 describe an important comparative-effectiveness study of two diagnostic strategies for patients presenting to the emergency department with chest pain. They investigate the effect of using coronary computed tomographic angiography (CCTA) to evaluate patients with symptoms suggestive of acute coronary syndromes to determine whether CCTA can safely reduce the time such patients spend in the hospital. In their study, Rule Out Myocardial Infarction Using Computer Assisted Tomography II (ROMICATII), 1000 such patients were randomly assigned to either CCTA or standard diagnostic procedures, which were performed at the discretion of the physicians in the emergency department. The authors found that the length of stay in the hospital in the CCTA group was 7.6 hours shorter than the length of stay in the standardevaluation group, and a follow-up evaluation 28 days later showed no overlooked cases of acute coronary syndromes in either group — a fact consistent with the low-to-intermediate-risk status of the study patients. They also found that the CCTA group incurred only slightly higher costs, as compared with the standard-evaluation group. The authors’ findings build on similar data from another study by Litt et al.2 that also randomly assigned patients with suspected acute coronary syndromes to CCTA or traditional care. Both studies confirm the somewhat unremarkable fact that CCTA provides faster diagnostic results than standard evaluation (which meant some type of stress test in 74% of the patients in the study by Hoffmann et al. and 64% of the patients in the study by Litt et al.). It should be noted that ROMICAT-II enrolled patients only during “weekday daytime hours” at sites where a “dedicated accelerated diagnostic protocol” was performed; the costs (and wait time) increase when any service is provided on nights and weekends. Although shorter lengths of stay in the hospital are highly desirable, especially from the patient’s point of view, the ROMICAT-II study reveals a deeper flaw in the approach to chest pain in the emergency department. The underlying assumption of the studies by Hoffmann et al. and Litt et al. is that some diagnostic test must be performed before discharging these low-to-intermediate-risk patients from the emergency department. This assumption is unproven and probably unwarranted. The rationale for any test, as compared with no testing, should be that it will lead to an improved outcome, and here there is no evidence that the tests performed led to improved outcomes. Indeed, event rates for major adverse cardiac events among all patients in the studies by Hoffmann et al. and Litt et al. (whether the patients underwent CCTA, stress testing, or no testing at all) were so low — less than 1% had a myocardial infarction and no patients died — that it is impossible to know whether the CCTA