CT features in predicting the pathological classification and prognosis in gastroenteric neuroendocrine neoplasm / 实用放射学杂志

ABSTRACT

Objective To analyze the correlations between the CT features and pathological classifications of gastroentric neuroendocrineneoplasm(GE-NEN) and to evaluate the association between prognostic value and CT features.Methods 45 cases pathologically and immunohistochemically proved GE-NEN were studied retrospectively.The correlation between the CT features (tumor's volume,shape,enhancement patterns and TNM stage) and pathological classifications was analyzed.Meanwhile,the association between the CT features and patients' overall survival was evaluated by Kaplan Meier method using Log-rank test and multivariate analysis in Cox proportional hazard model.Results The CT features were proved to be associated statistically with pathological classifications including the tumor's volume,shape,with or without areas of cystic change,necrosis or ulceration,T stage and lymphadenopathy (P ＜ 0.05).On the other side,the patients' overall survivals were associated with age,pathological classifications,volume of tumors,with or without areas of cystic change,necrosis or ulceration and TNM stage.Age,areas of cystic change,necrosis or ulceration and metastases were the independent prognostic factors.Conclusion Contrast-enhanced CT can be useful in the classification of GE-NEN and predicting the patient's survival.