Orbital Indeterminate Lesions in Adults: Combined Magnetic Resonance Morphometry and Histogram Analysis of Apparent Diffusion Coefficient Maps for Predicting Malignancy.

ABSTRACT

RATIONALE AND OBJECTIVES: The aim of this study was to evaluate the added value of histogram analysis of apparent diffusion coefficient (ADC) maps in differentiating indeterminate orbital malignant tumors from benign tumors, compared to using magnetic resonance (MR) morphological features alone. MATERIALS AND METHODS: We retrospectively evaluated 54 patients with orbital tumors from March 2013 to February 2015. All the patients were assessed by both routine MR and diffusion-weighted imaging, and divided into benign group and malignant group. Routine MR imaging features and histogram parameters derived from ADC maps, including mean ADC (ADCmean), median ADC (ADCmedian), standard deviation, skewness, kurtosis, and 10th and 90th percentiles of ADC (ADC10 and ADC90), were compared between two groups. Univariate and multivariate logistic regression analyses were used to identify the most valuable variables in predicting malignancy. Receiver operating characteristic (ROC) curve analysis was used to determine the diagnostic value of significant variables. RESULTS: Multivariate logistic regression analysis indicated that two or more quadrants involved, iso-intense on T2-weighted imaging (T2WI), and ADC10 were significant predictors for orbital malignancy. By using model 2 (iso-intense on T2WI + two or more quadrants involved + ADC10 < 0.990) as the criterion, higher AUC and specificity could be achieved than by using model 1 (iso-intense on T2WI + two or more quadrants involved) alone, (model 2 vs model 1; area under curve (AUC), 0.827 vs 0.793; sensitivity, 65.4% vs 69.2%; specificity, 100% vs 89.3%). CONCLUSIONS: Iso-intense on T2WI, two or more quadrants involved, and ADC10 are risk factors for orbital malignancy. Histogram analysis of ADC map might provide added value in predicting orbital malignancy.