Evaluation of Chest Pain in the Emergency Department

IN RESPONSE:

We are bemused by the number and wide range of responses generated by our editorial. At one end of the spectrum is the letter from Dr. Weissman and colleagues, who state that one of the approaches we suggested has “indeed improved efficiency and reduced cost” at the William Beaumont Hospital in Royal Oak, Michigan. They state that the use of single-photon emission computed tomographic imaging with Tc-99m sestamibi has changed the management decision in two thirds of their patients with unexplained chest pain. No patient discharged on the basis of the results of this imaging study, had an adverse cardiac event on follow-up. The cost savings was $1771 per patient, and 69 hospital days were avoided in the first 50 patients seen.

At the other end of the spectrum is the letter from Dr. Kellerman, who casts aspersions on our reasons for writing the editorial. He also states that we support the use of cardiac imaging because we specialize in noninvasive imaging and that we have quoted only our own work to make this claim. We emphasize that we had no ulterior motives for writing this editorial. We simply responded to the invitation of the editors of Annals. We mentioned noninvasive imaging as only one of many tools that could currently be used to diagnose acute ischemic syndromes in selected patients with chest pain. Our experience with noninvasive imaging in the emergency department [1] is shared by many institutions [2-5], including those of Drs. Weissman and O'Neill, who, like us, have shown that the judicious use of modern imaging technology can reduce unnecessary hospitalizations and costs.

Other interesting responses to our letter came predominantly from irate emergency department physicians. Dr. Aghababian, in his role as the President of the American College of Emergency Physicians, takes exception to some of our comments. He expressed concern at our statement that “emergency department physicians, by the nature of their training are generalists, and not experts in cardiovascular medicine.” We did not mean by this statement to belittle emergency department physicians. They are certainly more conversant than cardiologists in the management of many other conditions such as trauma or poisoning. We would not feel slighted if it were stated somewhere that emergency department physicians have greater expertise in these conditions than do cardiologists! That cardiologists are more conversant with acute ischemic syndromes and the interpretation of electrocardiograms should not be unexpected, given that they spend 36 to 48 months in training in cardiovascular medicine. Similar concerns were raised by Drs. Kellerman, Gibler, and Finkelstein.

Dr. Finkelstein states that “detailed guidelines developed by expert panels generally lie well behind the state of the art and often bring comfort to plaintiff's attorneys.” We believe, however, that guidelines for standards of care are necessary. When physicians perform within these guidelines, they should not be held culpable for any unfortunate occurrence. In the managed care environment, if the primary care physician must be consulted for every patient whose chest pain is of unclear cause, as asserted by Dr. Bloom, valuable time could be lost in consultation with someone who may not have any more experience with acute ischemic syndromes than the emergency physician.

Dr. Cheng suggests that comparing current and previous electrocardiograms can be useful when the cause of chest pain is unclear. In our experience, a previous electrocardiogram is primarily useful when it is normal. In the setting of a previous abnormal electrocardiogram, new changes are less likely to be helpful in separating acute myocardial infarction from other cardiac abnormalities that may not require hospital admission. We believe that the special life-support stations described by Dr. Napadano are similar to the chest pain unit described by us and that the potential conflicts mentioned by him are as relevant today as they were 20 years ago.

In our editorial, we did not imply a rejection of the probabilistic approach as alleged by Dr. Sturman. Physicians use such approaches, consciously or unconsciously, in everyday practice. We merely discussed the reasons why physicians do not routinely use such approaches when they are formalized as diagnostic instruments while evaluating individual patients, despite the encouraging results reported in several studies. We can understand why physicians cringe from making a “yes” or “no” decision solely on the basis of a number generated by a computer. The same applies to the use of artificial neural networks mentioned by Dr. Jones. Although such approaches are helpful in the emergency department, they should be considered no more than aids in the diagnosis of acute myocardial infarction and other ischemic syndromes.

We agree with Gibler that unless cardiologists are available on a 24-hour basis in the emergency department, emergency department physicians are best suited to care for patients with chest pain. We did not mean to minimize the role of emergency department physicians in the evaluation of such patients. However, ample evidence suggests that the current approaches used in the emergency departments in many hospitals in the United States are inefficient and costly. It is not that emergency department physicians send home an excess of patients who ultimately develop an infarction, but that they tend to admit too many patients to the hospital who have no business being there. A careful evaluation of such patients in the emergency department with the judicious use of modern technology, be it rapid serum enzyme assays or imaging of regional myocardial function [1, 2] or perfusion [3-5], could help reduce unnecessary hospital admissions and cost. We believe that chest pain in the emergency department is a major public health problem and that new cost-effective and efficient approaches, such as those described in our editorial, should be evaluated to address this problem. In this regard we believe that a prospective multicenter study using currently available technologies for the assessment of myocardial ischemic syndromes is in order. We could not agree more with Drs. Aghababian and Gibler that a team approach is required for the assessment of patients presenting to the emergency department with chest pain. We believe, however, that the members of the team must be chosen after great deliberation.

• Copyright ©2004 by the American College of Physicians