Predictive value of CT imaging features in preoperative high-risk group of childhood hepatoblastoma / 中华放射学杂志

ABSTRACT

Objective:To investigate the value of CT findings of childhood hepatoblastoma (HB) in predicting preoperative tumor risk stratification.Methods:Totally 46 children with HB confirmed by surgery and pathology were retrospectively enrolled from October 2010 to October 2019 in Shenzhen Children′s Hospital and Xuzhou Children′s Hospital. The preoperative abdominal plain CT and three-phasic contrast-enhanced CT with complete clinical files were evaluated. According to the clinical risk stratification established by the multidisciplinary diagnosis and treatment consensus for children with HB, the HB children were divided into high-risk group and non-high-risk group with 16 and 30 cases respectively. The maximum diameter of tumor, relative tumor volume index, cystic change or necrosis, bleeding, calcification, fibrous septations, tumor rupture, liver capsule retraction and subcapsular effusion were evaluated. Enhancement percentage and enhancement index on arterial, venous and delayed phases of each tumor were measured and calculated. Pearson′s χ 2 test or Fisher′s exact test were used to compare the differences in gender and lesion morphological characteristics between the high-risk group and the non-high-risk group. Two independent sample t test or Mann-Whitney U test were used to compare the differences in age, gestational age, birth weight, α-fetoprotein, platelets, maximum diameter of tumor, relative tumor volume index and CT parameters of the lesion between the two groups. Statistically significant features were included in the binary logistic regression analysis and independent predictors related to high-risk group were obtained. The ROC curve was used to determine the critical value of the high-risk group. Results:There were statistically significant differences in age, maximum diameter of tumor, relative tumor volume index and tumor rupture between the high-risk group and the non-high-risk group (all P<0.05). The logistic regression analysis showed that the maximum diameter of tumor (OR=1.906, P=0.004) and tumor rupture (OR=16.558, P=0.005) were risk factors of the high-risk group. Based on ROC curve, the optimum cut-off point of maximum diameter of tumor to predict high-risk group was 10.5 cm. Tumor rupture, maximum diameter of tumor and maximum diameter of tumor combined with tumor rupture for predicting the incidence of high-risk group resulted in the area under the curve of 0.744, 0.807 and 0.879, respectively. The sensitivity and specificity of maximum diameter of tumor combined with tumor rupture were 75.0% and 96.7%, respectively. Conclusion:The age of onset in high-risk group is relatively older. The maximum diameter of tumor greater than 10.5 cm accompanied by tumor rupture can be regarded as a high-risk sign.