Acute chest pain - a stepwise approach, the challenge of the correct clinical diagnosis

Study objective: To assess the safety and the accuracy of a 4 h stepwise diagnostic approach relying on clinical judgement in unselected patients with acute chest pain. Design: Prospective cohort study. Setting: Emergency department (ED) of a tertiary care university hospital. Patients: 1288 unselected patients presenting with acute chest pain. Interventions: After history and physical examination, clinical judgement (step 1), governed the need for further patient evaluation: baseline 12 lead electrocardiogramm, (ECG) and laboratory examinations (step 11), serial 12 lead ECG and laboratory examinations after 4 h (step 111), and 4 h troponin T measurement (step IV) to exclude or to confirm a coronary origin of chest pain. Patients were followed clinically for 6 months for future occurrence of cardiac events (myocardial infarction, percutaneous transluminal coronary angioplasty (PTCA), CABG, cardiac death), any death and for accuracy of the ED diagnosis in non-coronary chest pain patients. Measurements and results: Chest pain was diagnosed to be coronary in origin in 381 and non-coronary in 907 patients, respectively. Cardiac events occurred during follow up in 240 (19%) of 1288 patients, in 233 of 381 (61%) with presumed coronary and seven of 907 (1%) with presumed non-coronary chest pain. Sensitivity, specificity, positive predictive value and negative predictive value for correct detection of coronary chest pain were 97, 86, 61 and 99%, respectively. In non-coronary chest pain patients the agreement between the ED diagnosis and the final diagnosis was good (kappa = 0.71, 95% confidence interval (CI) 0.67-0.75). Conclusions: The 4 h stepwise approach guided by clinical judgement was safe for ruling out impending cardiac events in unselected patients with acute chest pain. However, more extensive evaluation is necessary for accurate rule-in of coronary chest pain. (C) 2002 Elsevier Science Ireland Ltd. All rights reserved.