Magnetic resonance myocardial perfusion imaging in the diagnosis of functionally significant obstructive coronary artery disease: a systematic review protocol

Cardiac magnetic resonance (CMR) myocardial perfusion imaging has been suggested as a non-invasive alternative to pressure wire guided fractional flow reserve (FFR) in detecting haemodynamically significant obstructive coronary artery disease (CAD). The objective of this systematic review is to determine the diagnostic accuracy of CMR and to compare it to FFR. A systematic review of diagnostic test accuracy of CMR and FFR will be conducted. Relevant English-language articles published before November 2013 in Medline, PubMed, EMBASE, Google scholar, Scopus and Cochrane databases will be identified. Relevant referenced articles from those selected will also be analysed. Articles describing diagnostic studies that compared CMR to FFR in patients with known or suspected coronary artery disease will be included. Two investigators will independently screen, assess quality and extract data from the selected articles. The Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool will be used to assess methodological quality. STATA 13 (the xtmelogit command) software will be used to calculate bivariate random effects models and estimate pooled sensitivity and specificity with 95% confidence intervals. Forests plots and summary receiver operating characteristics curves will also be generated. Sub-group pooled analyses using image quality of CMR (in terms of magnetic flux density - Tesla) and basis of analyses (coronary arterial territory vs. patients basis) at different FFR cutoffs (≤0.75 and ≤0.8) will also be performed. This systematic review will help to determine if CMR is an adequate alternative to FFR in the diagnosis of significant and obstructive CAD. We will also be able to assess diagnostic accuracy of specific types of CMR scan at different FFR cutoffs. This systematic review had been registered at PROSPERO and the registration number is CRD42013006180.