RESUMO
Objective:To analyze the MRI features of medulloblastoma (MB) in children, and screen out the key signs that can predict the risk of MB before surgery.Methods:Clinical and radiological data of 62 children with MB confirmed by pathology in Shenzhen Children′s Hospital from December 2012 to December 2021 were retrospectively analyzed. According to the diagnosis and treatment guidelines for children with MB (2021 edition), the patients were divided into standard risk group (43 cases) and high risk group (19 cases). MRI features of MB were observed and recorded, including tumor site, location of tumor center, tumor morphology, signal intensity of T 1WI, T 2WI and diffusion weighted imaging (DWI), enhancement pattern, cystic lesion size, location and number, peritumoral edema and hydrocephalus, and the maximum diameter of tumor was measured. The χ 2 test or Fisher exact probability method was used to compare the differences in age, gender and MRI signs between the two groups. The t test of two independent samples was used to compare the differences in the maximum diameter of tumors between the two groups. The indicators with statistically significant differences were included in binary logistic regression analysis to obtain independent influencing factors associated with the risk groups. The receiver operation characteristic curve was used to evaluate the diagnostic efficacy. Results:There were significant differences in age ( P=0.026), enhancement pattern ( P=0.018), cystic lesion size ( P=0.005), location ( P=0.011) and number ( P=0.003) between standard risk group and high risk group. There were no significant differences in gender, tumor site, location of tumor center, tumor morphology, signal intensity of T 1WI, T 2WI and DWI, peritumoral edema, hydrocephalus and maximum diameter of tumor between the two groups ( P>0.05). Binary logistic regression results showed the age (OR=0.207, 95%CI 0.040-0.983, P=0.042) and the number of cystic lesions (OR=0.215, 95%CI 0.073-0.630, P=0.005) were the protective factors for MB in high risk group, the enhancement pattern Ⅲ (OR=5.226, 95%CI 1.516-52.920, P=0.048) was the dangerous factor for MB in high risk group. The area under the curve of the combined diagnosis of high risk MB was 0.845 (95%CI 0.741-0.949). Conclusions:The age and MRI signs the pattern of tumor enhancement Ⅲ and the number of cystic lesion can be used to predict the risk grouping of MB preoperatively. When the child is younger and MB enhancement pattern is mainly peripheral enhancement without obvious cystic change, it may indicate high risk MB.