Search

Background: Non-invasive ischaemia tests and biomarkers are widely adopted to rule out acute coronary syndrome in the emergency department. Their diagnostic accuracy has yet to be precisely defined.Methods: Medline, Cochrane Library CENTRAL, EMBASE and Biomed Central were systematically screened (start date 1 September 2016, end date 1 December 2016). Prospective studies (observational or randomised controlled trial) comparing functional/imaging or biochemical tests for patients presenting with chest pain to the emergency department were included.Results: Overall, 77 studies were included, for a total of 49,541 patients (mean age 59.9 years). Fast and six-hour highly sensitive troponin T protocols did not show significant differences in their ability to detect acute coronary syndromes, as they reported a sensitivity and specificity of 0.89 (95% confidence interval 0.79–0.94) and 0.84 (0.74–0.9) vs 0.89 (0.78–0.94) and 0.83 (0.70–0.92), respectively. The addition of copeptin to troponin increased sensitivity and reduced specificity, without improving diagnostic accuracy. The diagnostic value of non-invasive tests for patients without troponin increase was tested. Coronary computed tomography showed the highest level of diagnostic accuracy (sensitivity 0.93 (0.81–0.98) and specificity 0.90 (0.93–0.94)), along with myocardial perfusion scintigraphy (sensitivity 0.85 (0.77–0.91) and specificity 0.92 (0.83–0.96)). Stress echography was inferior to coronary computed tomography but non-inferior to myocardial perfusion scintigraphy, while exercise testing showed the lower level of diagnostic accuracy.Conclusions: Fast and six-hour highly sensitive troponin T protocols provide an overall similar level of diagnostic accuracy to detect acute coronary syndrome. Among the non-invasive ischaemia tests for patients without troponin increase, coronary computed tomography and myocardial perfusion scintigraphy showed the highest sensitivity and specificity.