Non-invasive assessment of low risk acute chest pain in the emergency department: A comparative meta-analysis of prospective studies

Abstract Background The aim of this meta-analysis was to compare the diagnostic accuracy of cardiac computed tomographic angiography (CCTA), stress echocardiography (SE) and radionuclide single photon emission computed tomography (SPECT) for the assessment of chest pain in emergency department (ED) setting. Methods A systematic review of Medline, Cochrane and Embase was undertaken for prospective clinical studies assessing the diagnostic efficacy of CCTA, SE or SPECT, as compared to intracoronary angiography (ICA) or the later presence of major adverse clinical outcomes (MACE), in patients presenting to the ED with chest pain. Standard approach and bivariate analysis were performed. Results Thirty-seven studies (15 CCTA, 9 SE, 13 SPECT) comprising a total of 7800 patients fulfilled inclusion criteria. The respective weighted mean sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and total diagnostic accuracy for CCTA were: 95%, 99%, 84%, 100% and 99%, for SE were: 84%, 94%, 73%, 96% and 96%, and for SPECT were: 85%, 86%, 57%, 95% and 88%. There was no significant difference between modalities in terms of NPV. Bivariate analysis revealed that CCTA had statistically greater sensitivity, specificity, PPV and overall diagnostic accuracy when compared to SE and SPECT. Conclusions All three modalities, when employed by an experienced clinician, are highly accurate. Each has its own strengths and limitations making each well suited for different patient groups. CCTA has higher accuracy than SE and SPECT, but it has many drawbacks, most importantly its lack of physiologic data.