Determinants and Outcomes of Invasive Coronary Angiography in Unselected Patients Presenting With Chest Pain to Emergency Departments in Western Australian Teaching Hospitals

Background Chest pain is a common cause of presentation to the emergency department (ED), and its outcomes are affected by various diagnostic tests and timely management. Our aim was to identify determinants of invasive coronary angiography (ICA) in a chest pain cohort following ED presentation, and to compare outcomes by time to ICA.Methods We identified all adults aged >= 20 years presenting with chest pain to public teaching hospital EDs in Perth, Western Australia, from 1 January 2016 to 31 March 2017. Data were obtained from linked administrative ED, hospitalisation, pathology, and death records. Multivariable logistic regression was used to identify determinants of ICA within 90 days of ED presentation. Restricted cubic splines fitted in multivariable Cox regression models showed the relationship between time to ICA and outcomes of death and major cardiovascular events.Results The cohort comprised 16,974 people, with 8,609 (51%) male patients and a mean age of 56 years; 986 (5.8%) patients had ICA within 1 day of presentation, 394 (2.3%) within 2-3 days, 543 (3.2%) within 4-90 days, and 15,051 (88.7%) had no ICA or had ICA after 90 days. Age, sex, residential area, triage code, troponin classification, ED diagnosis, and medical history were all significantly associated with 90-day ICA. The adjusted odds ratio of women receiving ICA was 0.53 (95% confidence interval 0.47-0.61) in comparison with men. Restricted cubic splines showed a biphasic relationship between time to ICA and death.Conclusions The determinants of ICA reflect clinical practice guidelines, although ICA remains less likely in women than in men. Early ICA is associated with lower risk of death but may not affect myocardial infarction or composite outcomes after adjusting for cofactors.