External auditory canal and middle ear tumors: characterization by morphology and diffusion features on CT and MRI
Peng Wang,Heng Zhang,Jing Zhao,Hongyan Qiao,Yuxi Ge,Qingjun You,Shudong Hu
DOI: https://doi.org/10.1007/s00405-022-07509-1
2022-07-17
European Archives of Oto-Rhino-Laryngology
Abstract:PurposeTo explore the value of morphology and diffusion features on CT and MRI in the characterization of external auditory canal and middle ear tumors (EAMETs).MethodsForty-seven patients with histologically proved EAMETs (23 benign and 24 malignant) who underwent CT and MRI were retrospectively analyzed in this study. CT and MRI characteristics (including size, shape, signal intensity, border, enhancement degree, and bone changes) and apparent diffusion coefficient (ADC) value were analyzed and compared between benign and malignant EAMETs. Logistic regression, receiver operating characteristic (ROC) curve, and Delong test were performed to assess the diagnostic performance.ResultsCompared with benign tumors, the malignant EAMETs are characterized by irregular shape, ill-defined border, invasive bone destruction, and intense enhancement (all p < 0.05). There were no significant differences on the size and signal intensity between benign and malignant tumors. The ADC value of malignant tumors were (879.96 ± 201.15) × 10–6 mm2/s, which was significantly lower than benign ones (p < 0.05). Logistic regression demonstrates the presence of ill-defined margin, invasive bone destruction, and low ADC value (≤ 920.33 × 10–6 mm2/s) have significant relationship with malignant EAMETs. The combination of characterization by morphology and diffusion features on CT and MRI can further improve the diagnostic efficiency when compared with morphology and diffusion features alone (both p < 0.05).ConclusionSome CT and MRI characteristics are helpful in identifying malignant EAMETs from benign ones (especially ill-defined margin, invasive bone destruction, and low ADC value), and the combination of morphology and diffusion features on CT and MRI has best diagnostic efficiency for discriminating these two entities.
otorhinolaryngology