Diagnostic performance of diffusion-weighted magnetic resonance imaging in differentiating human renal lesions (benignity or malignancy): a meta-analysis

Purpose This study aims to quantitatively evaluate the potential of diffusion-weighted magnetic resonance imaging (DW-MRI) for differentiating malignant and benign human renal lesions. Materials and methods A systematic literature was performed to identify previous research related to the diagnostic performance of DW-MRI for determining whether human renal lesions were benign or malignant. ADC values were extracted from normal renal tissue and different lesion types. Data were extracted to assess the diagnostic performance of DW-MRI for differentiating malignant and benign human renal lesions, as well as running threshold effect and heterogeneity. Results Nine publications with 11 subsets were eligible for data extraction and diagnostic performance calculation. A total of 988 apparent diffusion coefficient (ADC) measurements were included. The differences in ADC values between benign lesions (2.47 ± 0.81 × 10 −3 mm 2 /s) and malignant lesions (1.81 ± 0.41 × 10 −3 mm 2 /s) were statistically significant ( P < 0.001). The diagnostic odds ratio, the overall positive, negative likelihood ratios, pooled weighted sensitivity and specificity with 95% CI were 20.05 (95% CI 12.56–32.02), 3.32 (95% CI 2.13–5.18), 0.20 (95% CI 0.15–0.27), 88% (95% CI 0.84–0.91) and 72% (95% CI 0.67–0.76), respectively. The area under the curve of the summary receiver operating characteristic was 0.90. Conclusions This meta-analysis indicated that DW-MRI had a relatively good diagnostic accuracy in differentiating malignant and benign human renal lesions. We preliminarily recommend that DW-MRI is performed with a maximum b value ranging from 800 to 1000 s/mm 2 at 3.0 T for imaging protocol, and that DW-MRI should be used with caution when the study population includes children.