Predicting Coronary Heart Disease in Remote Settings: A Prospective, Cross-sectional Observational Study

Abstract

'Background' Coronary heart disease (CHD) places a major burden on the Australian health care system. Determining the likelihood of CHD in a patient presenting with chest pain can be particularly difficult in a remote setting where access to transportation and specialised investigations including myocardial stress studies and coronary angiography can be difficult and delayed. The objective is to develop a predictive model for determining the risk of CHD, including the value of high sensitivity C-reactive protein (hsCRP), in patients presenting with chest pain with a particular emphasis on resources and information likely to be available in a remote primary health care setting. 'Methods' A prospective, cross-sectional observational study of patients with no prior diagnosis of CHD presenting to a specialist chest pain assessment clinic at Cairns Hospital from November 2012 to May 2013. 'Results' Out of the 163 participants included in the study analyses, a total of 38 were classified as CHD likely (23.3% (95% CI 17.1-30.6)). Logistic regression modelling identified two factors that were independently associated with likely CHD, namely the presence of typical chest pain (OR 83.7 (95% CI 21.7-322.1)) and an abnormal baseline ECG (OR 12.8 (95% CI 1.9-86.0)). 'Conclusion' In this study, it was demonstrated that the presence of typical chest pain and an abnormal resting ECG, remain the cornerstone of predicting a subsequent diagnosis of CHD. This information is easily accessible in remote primary health care and should be utilised to expedite assessment in patients presenting with symptoms suggestive of CHD.