Abstract
This cross-sectional study evaluated the role of magnetic resonance imaging (MRI), including diffusion-weighted imaging (DWI), in differentiating benign from malignant sinonasal lesions. Sinonasal neoplasms are uncommon, comprising approximately three percent of head and neck cancers and one percent of all malignancies. Accurate diagnosis is challenging due to the complex anatomy and overlapping imaging features. Despite advancements in endoscopic imaging and surgery, distinguishing benign from malignant lesions remains difficult. This study aimed to assess whether diffusion-weighted sequences and apparent diffusion coefficient (ADC) values improve the diagnostic accuracy of MRI and to establish an optimal ADC cutoff for malignancy detection. Seventy-nine patients with histopathologically confirmed sinonasal masses were prospectively evaluated at a tertiary center in India between August 2022 and January 2024. All patients underwent standardized head and neck MRI sequences, including T1-weighted, T2-weighted, short tau inversion recovery (STIR), post-contrast T1, and DWI (b-values 0 and 1000 s/mm²). Mean ADC values were calculated from regions of interest in solid tumor components. Statistical analyses (SPSS version 20, IBM Corp., Armonk, NY) determined diagnostic performance and receiver operating characteristic (ROC) thresholds using histopathology as the reference standard. An ADC cutoff of 1.37 × 10⁻³ mm²/s provided 97.5% sensitivity and 97.3% specificity, demonstrating near-perfect agreement with histopathology. Among conventional sequences, T2-weighted imaging achieved the highest predictive accuracy, while post-contrast T1 had lower sensitivity. Combined conventional MRI achieved an overall accuracy of 86.8%. Integrating DWI and ADC mapping into sinonasal MRI protocols significantly enhances diagnostic precision by providing a reliable, noninvasive differentiation between benign and malignant lesions. The established ADC threshold (1.37 × 10⁻³ mm²/s) serves as a robust imaging biomarker that, when used with conventional MRI, supports improved preoperative planning and clinical decision-making for sinonasal tumors.