Computed Tomography Imaging Findings for Predicting Histological Subtypes and Clinical Outcomes in Patients With Head and Neck Nodal Involvement of Diffuse Large B-Cell Lymphoma and Follicular Lymphoma

OBJECTIVE This study aimed to assess computed tomography (CT) imaging findings for predicting the histological subtypes and clinical outcomes in patients with head and neck nodal involvement of diffuse large B-cell lymphoma (DLBCL) and follicular lymphoma (FL). METHODS This retrospective study included 64 patients with histologically confirmed head and neck nodal lymphoma (43 with DLBCL and 21 with FL) who underwent pretreatment CT examinations. The CT imaging findings were retrospectively assessed and compared according to the 2 pathologies and their clinical outcomes. RESULTS Multiplicity (86% vs 57%, P < 0.05), necrosis (44% vs 5%, P < 0.01), ill-demarcated margin (33% vs 0%, P < 0.01), and surrounding fat stranding (56% vs 14%, P < 0.01) were significantly more frequent in DLBCLs than in FLs. Multivariate logistic regression analysis revealed that necrosis was a significant factor for predicting the diagnosis of DLBCL (P < 0.01). Multiplicity (100% vs 67%, P < 0.01), bilaterality (44% vs 13%, P < 0.05), and surrounding fat stranding (69% vs 28%, P < 0.01) were significantly more frequent in the poor outcome group than in the good outcome group. Multivariate logistic regression analysis revealed that bilaterality and surrounding fat stranding were significant factors for predicting poor outcomes (P < 0.05). CONCLUSIONS In patients with head and neck nodal lymphoma, necrosis was useful for predicting the diagnosis of DLBCL, whereas bilaterality and surrounding fat stranding were useful for predicting poor outcomes.