Diagnostic Accuracy of MRI for Orbital and Intracranial Invasion of Sinonasal Malignancies: A Systematic Review and Meta-Analysis.

: In this study, we review the diagnostic accuracy of magnetic resonance imaging (MRI) in detecting orbital and intracranial invasion of sinonasal malignancies (SNMs) using histopathological and surgical evidence as the reference standard. : A systematic search of studies in English was conducted in MEDLINE and Embase, limited to articles published since 1990. We included studies using preoperative MRI to detect the intracranial and orbital invasion of SNMs, with histological or surgical confirmation as the reference standard, and reported patient numbers in each class as required to assess diagnostic accuracy. The outcome measures were sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Heterogeneity was assessed with the Higgins inconsistency test (). : Seven original articles with 546 subjects were included in the review, with six included in the meta-analysis. The pooled overall accuracy for orbital invasion was higher at 0.88 (95% CI, 0.75-0.94) than that for intracranial invasion at 0.80 (95% CI, 0.76-0.83). The meta-analytic estimates and their 95% confidence intervals were as follows for intracranial/orbital invasion: sensitivity 0.77 (0.69-0.83)/0.71 (0.40-0.90); specificity 0.79 (0.74-0.83)/0.91 (0.78-0.97); PPV 0.76 (0.64-0.85)/0.78 (0.61-0.88); and NPV 0.82 (0.72-0.89)/0.90 (0.63-0.98). Substantial heterogeneity was observed in the Higgins inconsistency test () for orbital invasion (84%, 83%, and 93% for sensitivity, specificity, and NPV, respectively). : MRI yielded moderate-to-high diagnostic accuracy for intracranial and orbital invasion, despite some limitations leading to false diagnoses. Loss of the hypointense zone on postcontrast MRI was found to predict dural invasion. Infiltration of the extraconal fat beyond the periorbita was found to be an MRI feature of orbital invasion.