Original Research
Is the initial diagnostic impression of “noncardiac chest pain” adequate to exclude cardiac disease?

https://doi.org/10.1016/j.annemergmed.2004.03.021Get rights and content

Study objective

In patients presenting to the emergency department (ED) with an initial diagnostic impression of noncardiac chest pain, we determine the 30-day incidence of adverse cardiac events and characteristics associated with those events.

Methods

The multicenter, prospectively collected Internet Tracking Registry for Acute Coronary Syndromes (itrACS) registry of patients with chest pain enrolled from June 1, 1999, to August 1, 2001, was reviewed. We included patients if the physician's initial diagnostic impression was noncardiac chest pain after the medical history, physical examination, and initial 12-lead ECG. ED records, inpatient records, and follow-up results were reviewed for evidence of an adverse cardiac event defined as ST-segment or non–ST-segment elevation myocardial infarction, unstable angina, revascularization, or cardiac death within 30 days.

Results

Of 17,737 patients enrolled in itrACS, 2,992 had an initial emergency physician impression of noncardiac chest pain. Of these, 85 (2.8%) patients had definite evidence for an adverse cardiac event. The adverse cardiac event group was older (61.2 versus 47.9 years), more likely to be men (58.6% versus 38.7%), and had a higher Acute Cardiac Ischemia–Time Insensitive Predictive Instrument score (26.1 versus 15.6). Factors associated with adverse cardiac events included hypercholesterolemia, diabetes, history of coronary artery disease, and history of congestive heart failure.

Conclusion

When the initial impression is noncardiac chest pain, high-risk features such as traditional cardiovascular risk factors or a history of coronary artery disease are associated with adverse cardiac events. In the absence of well-defined criteria, treating physicians should consider further evaluation before diagnosing patients with noncardiac chest pain if these features are present.

Introduction

Heart disease is the leading cause of mortality among the US population, causing more than 710,000 deaths per year.1 Furthermore, the rate of missed acute myocardial infarction in the emergency department (ED) is 2.1%,2 representing a significant source of patient morbidity and mortality, as well as insurance losses because of malpractice.3 The consequence has been a conservative strategy by emergency physicians in the evaluation of chest pain, exemplified by the fact that more than two thirds of patients who have chest pain and are admitted to the hospital do not have an acute ischemic syndrome. However, with increased examination of health care spending, the cost of noncardiac chest pain admissions has placed emergency physicians under pressure to limit chest pain admissions to high-risk patients. Simultaneously, the fear of an unexpected bad outcome has limited their willingness to send undiagnosed chest pain patients home from the ED.

Because of these concerns, risk stratification of patients with chest pain has become common practice. Although many risk stratification schemes have been proposed for patients with chest pain,4., 5., 6., 7., 8., 9. most focus on patients who are likely to have myocardial ischemia as a cause of their pain. Even with these risk stratification schemes, the most widely adopted and earliest risk stratification tool used in practice is the initial impression. When evaluating patients experiencing chest pain in the ED, clinicians must form an initial impression based on immediately available data, such as medical history, physical examination findings, and often ECG findings. In some patients, the initial impression does not suggest a cardiac cause of the patient's chest pain. It has been suggested these patients do not need any further cardiac evaluation performed.9

Although this approach is intuitive and practical, patient outcomes have not been described. Furthermore, early studies demonstrated that patients labeled as having noncardiac chest pain have an incidence of coronary artery disease ranging from 0.8% to 28.1%,10 making this a subset of patients whose outcomes should be determined. The purpose of this study was to evaluate patients presenting to the ED with chest pain whose treating physicians' initial impression was noncardiac chest pain. We reviewed the prospectively collected Internet Tracking Registry for Acute Coronary Syndromes (itrACS) database of undifferentiated patients with chest pain to characterize this group of patients, identify patients with adverse cardiac events, and describe their associated risk factors.

Section snippets

Materials and methods

We performed secondary analysis of the prospectively collected itrACS chest pain registry.

Results

Of the 17,737 total itrACS patients, 570 (3.2%) patients had an initial diagnostic impression of acute myocardial infarction, 1,536 (8.7%) patients had unstable angina or non–Q-wave myocardial infarction, 4,699 (26.5%) patients had high-risk chest pain, 6,954 (39.2%) patients had low-risk chest pain, and 3,696 (20.8%) patients had noncardiac chest pain. The extent of evaluation increased with severity of the initial diagnostic impression, with 53.2% of patients with low-risk chest pain having

Limitations

There are several limitations to our study that should be recognized. The itrACS registry includes mostly patients with chest pain. Although patients with syncope, shortness of breath, or “anginal equivalents” are included in the database, there was a questionnaire-driven bias toward including patients with chest pain, which was evident throughout the study.

The initial impression was based on the initial history, physical examination, and 12-lead ECG. Serum cardiac biomarkers were not included

Discussion

The ED evaluation of the patient with chest pain remains a difficult task. Conservative approaches that include admission and provocative testing of low-risk patients with chest pain are cost-prohibitive. Liberal approaches to these patients are fraught with difficulties surrounding missed acute coronary syndromes and potential adverse outcomes. Unfortunately, a single objective test to identify these patients prospectively is not yet available. Despite our best efforts using clinical risk

References (21)

There are more references available in the full text version of this article.

Cited by (0)

Dr. Miller and Dr. Hoekstra were formerly affiliated with The Ohio State University Department of Emergency Medicine, Columbus, OH.

The abstract was presented in part in poster format at the Society for Academic Emergency Medicine annual meeting, Atlanta, GA, May 2001.

The itrACS registry was supported by an unrestricted educational grant from Schering-Plough Pharmaceuticals.

Author contributions: The itrACS registry, patient enrollment, and data collection were supervised by JWH, CVP, JEH, and BRT. All authors contributed to the research question and concept of the data analysis. CDM, CJL, and SK performed the chart review. Statistical advice and data analysis were performed by CJL. CDM, SK, and JWH drafted the Introduction and Discussion. Materials and Methods and Results were drafted by CJL and CDM. All authors contributed to manuscript revision. CDM takes responsibility for the paper as a whole.

Reprints not available from the authors.

View full text