ADC Thresholds in the Discrimination of Indolent and Aggressive Orbital Adnexal Lymphoma

Abstract

Aim: This study aims to determine whether ADC measurements allow discrimination of indolent (group A) from aggressive orbital lymphoma (group B) and how much ADC value can predict lymphoma subtype. Methods: A retrospective two-armed cohort study including 32 Orbital Adnexal Lymphoma (OAL) lesions evaluated by conventional magnetic resonance image (MRI) and Diffusion Weighted Image (DWI) examination preceding histopathological lesion confirmation. DWI using single-shot echo-planar imaging with b factors of 0,400 and 800 sec/mm2 were performed on 3 Tesla MRI unit. Lymphomas were grouped into indolent or aggressive histopathological subtypes according to the Revised European-American Classification of Lymphoid Neoplasms (REAL). The groups minimum, maximum and percentiles of ADC values of indolent were compared to those of aggressive lymphoma subtype. Multiple receiver operating characteristic curve (ROC) analysis was performed for predicting aggressive OAL. Results: Of the 32 (OAL) cases, 23 (71.9%) were indolent and 9 (28.1%) were aggressive OAL subtypes. In general, ADC values were significantly higher in the indolent compared to the aggressive OL subtypes. (P<0.001). Extra-ocular muscles (P = 0.69) and globe involvement (P = 0.9) were not significant confounders for association of ADC to histological type. The area under curve (AUC) for ADC minimum was 94.4% and for ADC maximum was 81.6%. ADC minimum showed higher sensitivity and specificity than ADC maximum in revealing (OAL) subtypes. Conclusions: ADC measurements are higher in indolent than in aggressive (OAL). ADC minimum has higher sensitivity and specificity than ADC maximum in (OAL) lesions. These concepts confirm and expand the knowledge on (OAL), revealing the importance of ADC minimum in discriminate indolent from aggressive OAL. DWI can be used to characterize (OAL) based on ADC values, having higher sensitivity and specificity to point the (OAL) subtypes.