November 8, 2015 – Finding better ways to accurately diagnose acute coronary syndrome in patients with chest pain could save time and money.
Certain risk scores are more effective than initial patient histories, physical examinations, or electrocardiograms alone in diagnosing acute coronary syndrome (ACS), according to a new study by DCRI and Duke researchers.
The study appears in the current issue of JAMA. The DCRI’s Kristin Newby, MD, MHS, was the study’s senior author.
Only about 10 percent of patients who present with acute chest pain are ultimately diagnosed with ACS. However, arriving at that diagnosis can often be an expensive, time-intensive process. Early, accurate estimation of the probability of ACS in patients with acute chest pain could prevent many hospital admissions among low-risk patients and ensure that high-risk patients are treated promptly.
Unlike heart attacks, there are no consensus guidelines for diagnosing ACS in a patient. Consequently, doctors often rely upon a variety of tools, techniques, and risk factors to diagnose the condition, including initial patient histories, physical examinations, and electrocardiograms. There are also risk scoring systems such as History, ECG, Age, Risk Factors, Troponin (HEART) and Thrombolysis in Myocardial Infarction (TIMI) for categorizing patients who present with chest pain.
In this study, the researchers conducted a review of the MEDLINE and EMBASE databases for articles published between 1995 and July 2015 related to prospective studies of diagnostic test accuracy among emergency patients with symptoms suggesting ACS. Of 2,992 unique citations identified by the researchers, 58 were included in their final analysis.
The researchers found that a prior abnormal stress test, the presence of peripheral arterial disease, and pain in both arms were the clinical findings and risk factors most suggestive of ACS. ST-segment depression and any ischemia on electrocardiogram (ECG) were the most useful ECG findings. Both the HEART and TIMI risk scores also performed well in diagnosing ACS.
The accuracy of individual risk factors and symptoms was generally poor. An overall diagnostic process that incorporated all elements of the patient’s history and a physical examination was more useful, but the best diagnostic tests were clinical prediction tools such as TIMI and HEART that incorporated historical elements along with the initial ECG and cardiac troponin results.
These prediction tools, the researchers concluded, are effective because they combine the patient’s clinical context, ECG data, and cardiac biomarkers into a quantitative assessment. They cautioned, however, that the clinician must still analyze each component of the tool independent of the others.
In addition to Newby, Duke authors included Alexander C. Fanaroff, MD; Jennifer A. Rymer, MD, MBA; and Sarah A. Goldstein, MD.