Morphological distribution and internal enhancement architecture of contrast-enhanced magnetic resonance imaging in the diagnosis of non-mass-like breast lesions: a meta-analysis.

The aim of this study is to assess the diagnostic performance of contrast-enhanced magnetic resonance (MR) imaging in patients with non-mass-like breast lesions in a meta-analysis. Literature study was performed on PubMed data base on the diagnostic performance of MR imaging in patients with non-mass-like breast lesions. Details of the relevant studies were reviewed and a meta-analysis was performed to estimate the overall sensitivity and specificity of contrast-enhanced MR imaging of non-mass-like breast lesions. A summary receiver operating characteristic curve (sROC) was developed to explore the threshold effect by ROC space. Spearman correlation coefficient was calculated using Meta-Disc version 1.4 to analyze the heterogeneity between studies. A total of 858 non-mass-like lesions from 15 studies were included in the meta-analysis (sample size range: 27-131). Pooled weighted estimates of sensitivity and specificity were 50% (95% CI: 46%, 53%) and 80% (95% CI: 77%, 83%), respectively. The heterogeneity among studies was caused by other factors other than threshold effect. The findings were influenced by cancer prevalence (p = 0.0359). Subgroup analyses indicated that the sensitivity and specificity in studies with combined diagnostic criterion was higher than that in studies with single diagnostic criterion. In evaluation of non-mass-like breast lesions, contrast-enhanced MRI has high specificity and relatively lower sensitivity.