What decides the suspicion of acute coronary syndrome in acute chest pain patients?

BACKGROUND: Physicians assessing chest pain patients in the emergency department (ED) base the likelihood of acute coronary syndrome (ACS) mainly on ECG, symptom history and blood markers of myocardial injury. Among these, the ECG has been stated to be the most important diagnostic tool. We aimed to analyze the relative contributions of these three diagnostic modalities to the ED physicians’ evaluation of ACS likelihood in clinical practice. METHODS: 1151 consecutive ED chest pain patients were prospectively included. The ED physician’s subjective assessment of the patient’s likelihood of ACS (obvious ACS, strong, vague or no suspicion of ACS), the symptoms and the ECG were recorded on a special form. The ED TnT value was retrieved from the medical records. Frequency tables and logistic regression models were used to evaluate the contributions of the diagnostic tests to the level of ACS suspicion. RESULTS: Symptoms determined whether the physician had any suspicion of ACS (odds ratio, OR 526 for symptoms typical compared to not suspicious of ACS) since neither ECG nor TnT contributed significantly (ORs not significantly different from 1) to this assessment. ACS was suspected in only one in ten patients with symptoms not suspicious of ACS. Symptoms were also more important (OR 620 for typical symptoms) than ECG (OR 31 for ischemic ECG) and TnT (OR 3.4 for a positive TnT) for the assessment of obvious ACS/strong suspicion versus vague/no suspicion. Of the patients with ST-elevation on ECG, 71% were considered to have an obvious ACS, as opposed to only 6% of those with symptoms typical of ACS and 10% of those with a positive TnT. CONCLUSION: The ED physicians used symptoms as the most important assessment tool and applied primarily the symptoms to determine the level of ACS suspicion and to rule out ACS. The ECG was primarily used to rule in ACS. The TnT level played a minor role for the assessment of ACS likelihood. Further studies regarding ACS prediction based on symptoms may help improve decision-making in ED patients with possible ACS.