John W Pickering1, Joanna M Young2, Peter George3, Sally Aldous4, Louise Cullen5, Jaimi H Greenslade5, A Mark Richards6, Richard Troughton7, Michael Ardagh1, Christopher M Frampton6, Martin P Than8. 1. Emergency Department, Christchurch Hospital, Christchurch, New Zealand; University of Otago, Christchurch, New Zealand. 2. Lipid and Diabetes Research Group, Christchurch, New Zealand. 3. Canterbury Health Laboratories, Christchurch, New Zealand; University of Otago, Christchurch, New Zealand. 4. Cardiology Department, Christchurch Hospital, Christchurch, New Zealand. 5. Department of Emergency Medicine, Royal Brisbane and Women's Hospital, Brisbane, Australia; School of Medicine, The University of Queensland, Brisbane, Australia; School of Public Health, The Queensland University of Technology, Brisbane, Australia. 6. University of Otago, Christchurch, New Zealand. 7. University of Otago, Christchurch, New Zealand; Cardiology Department, Christchurch Hospital, Christchurch, New Zealand. 8. Emergency Department, Christchurch Hospital, Christchurch, New Zealand. Electronic address: martin.than@xtra.co.nz.
Abstract
OBJECTIVES: International guidance recommends that early serial sampling of high sensitivity troponin be used to accurately identify acute myocardial infarction (AMI) in chest pain patients. The background evidence for this approach is limited. We evaluated whether on presentation and 4-hour high-sensitivity troponin I (hs-cTnI) could be used to accurately rule-out AMI. DESIGN AND METHODS: hs-cTnI was measured on presentation and at 4-hours in adult patients attending an emergency department with possible acute coronary syndrome. We determined the sensitivity for AMI for at least one hs-cTnI above the 99th percentile for a healthy population or alone or in combination with new ischemic ECG changes. Both overall and sex-specific 99th percentiles were assessed. Patients with negative tests were designated low-risk. RESULTS: 63 (17.1%) of 368 patients had AMI. The median (interquartile range) time from symptom onset to first blood sampling was 4.8h (2.8-8.6). The sensitivity of the presentation and 4h hs-cTnI using the overall 99th percentile was 92.1% (95% CI 82.4% to 97.4%) and negative predictive value 95.4% (92.3% to 97.4%) with 78.3% low-risk. Applying the sex-specific 99th percentile did not change the sensitivity. The addition of ECG did not change the sensitivity. CONCLUSION: Hs-cTnI >99th percentile thresholds measured on presentation and at 4-hours was not a safe strategy to rule-out AMI in this clinical setting irrespective of whether sex-specific 99th percentiles were used, or whether hs-cTnI was combined with ECG results.
OBJECTIVES: International guidance recommends that early serial sampling of high sensitivity troponin be used to accurately identify acute myocardial infarction (AMI) in chest painpatients. The background evidence for this approach is limited. We evaluated whether on presentation and 4-hour high-sensitivity troponin I (hs-cTnI) could be used to accurately rule-out AMI. DESIGN AND METHODS: hs-cTnI was measured on presentation and at 4-hours in adult patients attending an emergency department with possible acute coronary syndrome. We determined the sensitivity for AMI for at least one hs-cTnI above the 99th percentile for a healthy population or alone or in combination with new ischemic ECG changes. Both overall and sex-specific 99th percentiles were assessed. Patients with negative tests were designated low-risk. RESULTS: 63 (17.1%) of 368 patients had AMI. The median (interquartile range) time from symptom onset to first blood sampling was 4.8h (2.8-8.6). The sensitivity of the presentation and 4h hs-cTnI using the overall 99th percentile was 92.1% (95% CI 82.4% to 97.4%) and negative predictive value 95.4% (92.3% to 97.4%) with 78.3% low-risk. Applying the sex-specific 99th percentile did not change the sensitivity. The addition of ECG did not change the sensitivity. CONCLUSION: Hs-cTnI >99th percentile thresholds measured on presentation and at 4-hours was not a safe strategy to rule-out AMI in this clinical setting irrespective of whether sex-specific 99th percentiles were used, or whether hs-cTnI was combined with ECG results.