Diagnostic performance of stress myocardial perfusion imaging for coronary artery disease: a systematic review and meta-analysis

OBJECTIVES: To determine and compare the diagnostic performance of stress myocardial perfusion imaging (MPI) for the diagnosis of obstructive coronary artery disease (CAD), using conventional coronary angiography (CCA) as the reference standard. METHODS: We searched Medline and Embase for literature that evaluated stress MPI for the diagnosis of obstructive CAD using magnetic resonance imaging (MRI), contrast-enhanced echocardiography (ECHO), single-photon emission computed tomography (SPECT) and positron emission tomography (PET). RESULTS: All pooled analyses were based on random effects models. Articles on MRI yielded a total of 2,970 patients from 28 studies, articles on ECHO yielded a sample size of 795 from 10 studies, articles on SPECT yielded 1,323 from 13 studies. For CAD defined as either at least 50 %, at least 70 % or at least 75 % lumen diameter reduction on CCA, the natural logarithms of the diagnostic odds ratio (lnDOR) for MRI (3.63; 95 % CI 3.26–4.00) was significantly higher compared to that of SPECT (2.76; 95 % CI 2.28–3.25; P = 0.006) and that of ECHO (2.83; 95 % CI 2.29–3.37; P = 0.02). There was no significant difference between the lnDOR of SPECT and ECHO (P = 0.52). CONCLUSION: Our results suggest that MRI is superior for the diagnosis of obstructive CAD compared with ECHO and SPECT. ECHO and SPECT demonstrated similar diagnostic performance. KEY POINTS: • MRI can assess myocardial perfusion. • MR perfusion diagnoses coronary artery disease better than echocardiography or SPECT. • Echocardiography and SPECT have similar diagnostic performance. • MRI can save coronary artery disease patients from more invasive tests. • MRI and SPECT show evidence of publication bias, implying possible overestimation. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s00330-012-2434-1) contains supplementary material, which is available to authorized users.