Diffusion weighted imaging for the differential diagnosis of benign vs. malignant ovarian neoplasms

In order to assess the diagnostic accuracy of diffusion weighted imaging (DWI) in differentiating between benign and malignant ovarian neoplasms, a systemic meta-analysis was conducted. Relevant studies were retrieved from scientific literature databases, including the PubMed, Wiley, EBSCO, Ovid, Web of Science, Wanfang, China National Knowledge Infrastructure and VIP databases. Following a multi-step screening and study selection process, the relevant data was extracted for use in the present study. Statistical analyses were performed using Meta-disc software version 1.4 and STATA statistical software version 12.0. A total of 285 articles were retrieved from the database searches. Following a careful screening process, 10 case-control studies were selected for the present meta-analysis. The 10 studies investigated the efficacy of DWI in diagnosing ovarian neoplasms, and included a combined total of 1,159 subjects, of which 559 patients had malignant lesions and 600 had benign lesions. The results showed that the pooled sensitivity, pooled specificity, pooled positive likelihood ratio, pooled negative likelihood ratio, pooled diagnostic odds ratio (DOR) and area under the curve of the summary receiver operating characteristics curve of DWI for differentiating between benign and malignant ovarian neoplasms were 0.93, 0.89, 7.58, 0.10, 85.33 and 0.95, respectively. A subgroup analysis based on ethnicity revealed no significant difference between Asians and Caucasians. Another subgroup analysis by magnetic resonance imaging (MRI) type showed that the DORs for GE Healthcare Life Sciences and Siemens AG machines were 100.76 [95% confidence interval (CI), 65.28-155.53] and 30.85 (95% CI, 10.40-91.53), respectively; this indicates that the diagnostic efficiency of the GE Healthcare Life Sciences MRI is superior compared with the Siemens AG MRI. The DWI demonstrated an excellent diagnostic performance in discriminating between benign and malignant ovarian neoplasms, and predicted the surgical outcome in ovarian neoplasms.