Diagnostic value of spiral CT energy spectrum imaging in pathological differentiation of colorectal cancer

Abstract

Abstract Objective: To investigate the accuracy of CT energy spectrum imaging in predicting preoperative pathological differentiation of colorectal cancer patients. Methods: From 2019 September to 2021 November, a retrospective study was performed for the eighty-two patients with colorectal cancer through preoperative colonoscopy or surgical pathology confirmed in our hospital. According to the pathological results, the patients were divided into middle and high differentiation group (62 cases) and low differentiation group (20 cases). GE Revolution CT scanner was used to scan the patients with energy spectrum imaging, in the arterial and venous phase, it measured and recorded the single energy CT values from 40kev to 140kev and various energy spectrum parameters of lymph nodes around the lesions. And statistically analyze the above indices. Result: In arterial phase: the single energy CT values from 40kev to 140kev in the low differentiation group were higher than those in the medium and high differentiation group (all P <0.05); IC(Iodine concentration), NIC(Normalized iodine concentration) and λ(the slope of the energy spectrum curve) of colorectal cancer in low differentiation group were higher than that in medium and high differentiation group(all P<0.05). The ROC (receiver operating curve) showed that the AUC (area under the curve) of the single energy CT values from 40kev to 140kev were less than 0.800. The AUC of IC and NIC were 0.800 and 0.828 respectively; the highest AUC of the combination of IC and NIC was 0.850. In venous phase: the single energy CT values from 40kev to 130kev in the low differentiation group were higher than those in the medium and high differentiation group(all P<0.05), and IC, NIC, λ and Eff-Z(Effective-Z) of colorectal cancer in low differentiation group were higher than that in the medium and high differentiation group(all<0.05). The ROC curve showed that the AUC of the single energy CT values from 40kev to 90kev were higher than 0.800. The AUC of IC, NIC, λ and Eff-Z were 0.927, 0.882, 0.859 and 0.863 respectively; the parameters were combined in pairs, IC and λ achieved the highest AUC, and the highest AUC was 0.934. Three parameters were combined, IC, NIC and λ achieved the highest AUC, and the highest AUC was 0.942, the combined AUC of the four was 0.943. Compared with the arterial phase, the single energy CT values from 40kev to 90kev and energy spectrum parameters in venous phase had better diagnostic efficacy in predicting different degrees of differentiation of colorectal cancer. Conclusions: Parameters of energy spectrum CT imaging can better predict the differentiation grade of colorectal cancer before operation and provide imaging basis for clinicians.