Diagnostic performance of diffusion-weighted imaging for differentiating benign and malignant gallbladder lesions: A systematic review and meta-analysis.

Abstract

Although diffusion-weighted imaging (DWI) has been characterized as an alternative imaging modality for gallbladder (GB) lesions, it has not been routinely used in clinical practice because of relatively low signal-to-noise ratio. To assess the sensitivity and specificity of the diagnostic performance of DWI for differentiating benign and malignant GB lesions. Meta-analysis. Patients with GB lesions. DWI at 3.0T or 1.5T. PubMed and EMBASE were searched following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy guidelines. Bivariate modeling and hierarchical summary receiver operating characteristic (HSROC) modeling were performed to compare the overall diagnostic performance of DWI. Subgroup analyses were performed for qualitative and quantitative assessment of the DWI. Meta-regression analyses were performed according to the characteristics of the patients, study, and MRI. Eight studies (including 557 patients) were included. The DWI exhibited a pooled sensitivity of 91%, a pooled specificity of 87%, and HSROC of 0.95. In subgroup analyses, qualitative assessment (sensitivity, 90%; specificity, 87%; HSROC, 0.94) was more accurate than quantitative assessment (sensitivity, 82%; specificity, 86 %; HSROC, 0.88). On meta-regression analysis, studies that used 3.0T field strength and thinner slices (≤5 mm) reported a significantly higher sensitivity (P ≤ 0.02) than those using only 1.5T field strength and thicker slices (>5 mm). DWI can discriminate malignant from benign GB lesions with excellent diagnostic performance in both qualitative and quantitative assessments. To enhance the diagnostic ability of DWI, images obtained with thinner slices (≤5 mm) with 3T field strength and qualitative assessment are recommended. 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:1375-1388.