Objective To evaluate the value of stress perfusion cardiovascular magnetic resonance in the diagnosis of coronary ar-tery disease.Methods Relevant literatures about stress perfusion cardiovascular magnetic resonance in the diagnosis of coronary ar-tery disease were searched.According to the recommended contents of the Cochrane Collaboration,we set inclusion criteria and screen literatures.Using software STATA 12.0 and Meta-disc 1.4,we analyzed the data ,including the heterogeneity,publication bias,subgroup analysis,Meta-regression,clinical application,pooled sensitivity(SEN),specificity(SPE),positive likelihood ratio (PLR),negative likelihood ratio(NLR),diagnostic odds ratio(DOR)and 95%confidence interval(95%CI),we drew the summary receiver operating characteristic curve(SROC)and calculated the area under the curve(AUC).Results Twenty-nine literatures were included,the pooled SEN,SPE,PLR,NLR,DOR and 95%CI were 0.90(0.89~0.92)、0.78(0.75~0.80)、4.03(3.32~4.91)、0.14 (0.11~0.18)、32.54(22.25~47.60)respectively.The AUC of SROC was 0.907 8.Conclusion The current evidence indicates that stress perfusion cardiovascular magnetic resonance has a high diagnostic accuracy for detection coronary artery disease.However, the relatively high false positive rate was found due to the medium specificity.