Performance of MRI protocols for diagnosis of benign and malignant ovarian tumors: a systematic review and meta-analysis.
Pin Yang, Fengxian Fan, Tingli Yang et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis assessed MRI protocols for distinguishing benign from malignant ovarian tumors. It included 81 studies, with 12,594 patients and 14,023 lesions. Across lesion-based analyses, MRI had pooled sensitivity of 0.92 (95% CI: 0.91-0.94) and specificity of 0.88 (95% CI: 0.86-0.90); the pooled area under the SROC curve was 0.96 (95% CI: 0.94-0.97). Diffusion-weighted imaging showed lower pooled sensitivity and specificity than MRI overall, and the authors describe it as an adjunct to conventional MRI.
Key findings
- Across MRI studies, pooled sensitivity was 0.92 (95% CI: 0.91-0.94) and specificity was 0.88 (95% CI: 0.86-0.90).
- The pooled area under the SROC curve for MRI was 0.96 (95% CI: 0.94-0.97).
- For DWI alone, pooled sensitivity was 0.88 (95% CI: 0.83-0.92) and specificity was 0.79 (95% CI: 0.72-0.84).
- MR scoring systems had pooled sensitivity of 0.94 (95% CI: 0.92-0.96) and specificity of 0.90 (95% CI: 0.87-0.93).
Why it matters
Distinguishing benign from malignant ovarian tumors informs subsequent treatment, and these pooled results indicate that MRI protocols can provide useful diagnostic discrimination. The findings also clarify that DWI alone performed less well than MRI overall, supporting its interpretation as an adjunct rather than a replacement for conventional MRI. Differences between protocol subgroups may help frame diagnostic assessment.
What to keep in mind
The abstract does not report a pooled prospective evaluation; the evidence combines 81 included studies and reports diagnostic performance across differing MRI protocols.