Validation of the Vancouver Chest Pain Rule using troponin as the only biomarker: a prospective cohort study.
Am J Emerg Med. 2013 Jul;31(7):1103-7
Authors: Greenslade JH, Cullen L, Than M, Aldous S, Chu K, Brown AF, Richards AM, Pemberton CJ, George P, Parsonage WA
OBJECTIVES: The objective of this study is to evaluate the accuracy of the Vancouver Chest Pain Rule using troponin as the only biomarker in an emergency department (ED) setting.
METHODS: This is an analysis of prospectively collected data from 2 EDs in Australia and New Zealand. Trained research nurses collected clinical data using a customised case report form. Based on a modified Vancouver Chest Pain Rule using troponin as the only biomarker, low-risk patients were identified using clinical history, age, pain characteristics, electrocardiography, and troponin results at 0 and 2 hours after presentation. The primary outcome was diagnosis of acute coronary syndrome (ACS) within 30 days of presentation as adjudicated by 2 independent cardiologists. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated to assess the accuracy of the rule.
RESULTS: There were 1635 patients, and 20.4% had ACS. One hundred twenty-one patients (7.4%) were assigned to the low-risk group on presentation, and a further 418 (25.6%) were assigned low risk after 2-hour electrocardiography and troponin testing. Of the 539 patients (33%) who were eligible for early discharge, 30 (5.6%) had ACS. Sensitivity was 91.0% (95% confidence interval [CI], 85.7%-93.6%), negative predictive value was 94.4% (95% CI, 92.2-96.1), specificity was 39.1% (95% CI, 36.5-41.8), and positive predictive value was 27.7% (95% CI, 25.2-30.5).
CONCLUSIONS: The Vancouver Chest Pain Rule with troponin as the only biomarker identified a sizable low-risk cohort. However, sensitivity was lower than that identified in the original derivation study and was considered insufficient to enable safe early discharge. Modifications to the tool would be required if troponin was incorporated as the only biomarker.
PMID: 23702078 [PubMed - indexed for MEDLINE]