Breast Lesions: Diagnosis Using Diffusion Weighted Imaging at 1.5T and 3.0T—Systematic Review and Meta-analysis

Abstract Introduction We compared the diagnostic performance of diffusion weighted imaging (DWI) acquired with 1.5T and 3.0T magnetic resonance (MR) units in differentiating malignant breast lesions from benign ones. Materials and Methods A comprehensive search of the PubMed and Embase databases was performed for studies reported from January 1, 2000 to February 19, 2016. The quality of the included studies was assessed. Statistical analysis included pooling of diagnostic sensitivity and specificity and assessing data inhomogeneity and publication bias. Results A total of 61 studies were included after a full-text review. These included 4778 patients and 5205 breast lesions. The overall sensitivity and specificity were 90% (95% confidence interval [CI], 88%-92%) and 86% (95% CI, 82%-89%), respectively. The pooled diagnostic odds ratio was 53 (95% CI, 37-74). For breast cancer versus benign lesions, the area under the curve was 0.94 (95% CI, 0.92-0.96). For the 44 studies that used a 1.5T MR unit, the pooled sensitivity and specificity were 91% (95% CI, 89%-92%) and 86% (95% CI, 81%-90%), respectively. For the 17 studies that used a 3.0T MR unit, the pooled sensitivity and specificity were 88% (95% CI, 83%-91%) and 84% (95% CI, 0.78-0.89), respectively. Publication bias and significant heterogeneity were observed; however, no threshold was found among the 61 studies. No significant difference was found in the sensitivity or specificity between the subgroups. Conclusion The results of the comparison between the subgroups that had used either a 1.5T or 3.0T MR unit suggest that the diagnostic accuracy for breast cancer compared with benign lesions is not significantly different.