ABSTRACT
OBJECTIVE: To examine the use of cardiac troponin (cTn) testing for acute coronary syndrome (ACS) diagnosis in primary care. DESIGN AND SETTING: Prospective cohort study; general practitioner-initiated cTn tests conducted from 24 September 2009 to 3 September 2010 in Perth, Western Australia. Patient outcomes were obtained from linked data sources for up to 12 months after the final test. Clinical information and outcomes were compared with data from emergency department patients with ACS symptoms. PARTICIPANTS: 369 patients with samples collected at community laboratories. Requesting GPs provided the clinical context for testing. MAIN OUTCOME MEASURES: Cardiovascular risk status, symptoms prompting cTn testing; estimated ACS likelihood and referral decision before and after testing; result turnaround time; hospital presentations, procedures and mortality. RESULTS: Of the 328 GPs who received a survey request, 124 (37.8%) responded. 122 of 124 test results (98.4%) were negative. Based on clinical risk factors, 71 of 104 patients (68.2%) were at high or intermediate risk of ACS. 69 of 124 patients (55.6%) had typical ischaemic pain and 62 of 124 patients (50.0%) were tested within 48 hours of symptom onset (23.4% within 12 hours, with no serial testing). Test results affected GPs' estimation of ACS likelihood (P < 0.01) but not their referral decisions (P = 0.23). 94 of 355 patients (26.5%) presented to hospital with cardiovascular symptoms or diagnoses during follow-up; 27 of 355 patients (7.6%) had at least one ACS, 13 of 255 (3.7%) within 30 days of testing. CONCLUSIONS: GP-initiated cTn testing involves patients at high risk of ACS. ACS and associated adverse outcomes can occur in patients undergoing testing, even when the cTn test result is negative. Potential gaps exist in physicians' understanding of the limitations of cTn testing, and cTn test results have minimal influence on their management of patients. GPs may benefit from guidance about ordering cTn testing.
