| Objective This study was designed to explore the independent risk factors of lymph node metastasis by analyzing multiple preoperative clinicopathological data of colon cancer patients,and to construct and internally validate an online version of a dynamic nomogram to assess the risk of LNM in patients with CC individually.Methods Clinicopathological data,including demographics,signs or symptoms,laboratory tests,tumor markers,enhanced CT abdomen,colon endoscopy and pathological biopsies,were retrospectively collected from January 2016 to June 2022 from 300 CC patients who visited the Department of Gastrointestinal Surgery at Bayannuo Hospital of Inner Mongolia Medical University and underwent radical surgery for CC.The patients were divided into the LNM-positive group(LNM group)and the LNM-negative group(non-LNM group)according to the postoperative pathological findings.SPSS 25.0 and R software were used for the statistical analysis of the data.Between-group differences between the LNM and non-LNM groups were compared using a one-way analysis of differences,in which two independent samples t-test was used to compare the differences between groups for quantitative data and the Pearson chi-square test was used to compare the differences in the distribution of qualitative data between the two groups.The univariate analysis combined with LASSO regression was used to screen clinical characteristics for multivariate logistic regression analysis.Finally,the independent influencing factors of LNM were screened by multivariate logistic regression analysis.Using independent influencing factors as predictors,we construct a dynamic nomogram and developed an online version of the interactive dynamic nomogram based on the Shiny website.The bootstrap method(1000 times)was used for internal validation of the model.Receiver operating characteristic curves(ROC)and calibration curves were used to assess the discrimination and consistency assessment of the nomogram,respectively;decision curve analysis(DCA)and clinical impact curve(CIC)were used to assess the clinical application value of the nomogram.Compared with preoperative abdominal contrast-enhanced CT,net reclassification index(NRI)and integrated discrimination improvement(IDI)were used to evaluate the improvement of accuracy of nomograms in predicting LNM,and the DCA curve was used to evaluate the improvement of nomograms in clinical application.Results Multivariate logistic regression analysis showed that intestinal obstruction(OR=6.71,95%CI 3.41-13.88,P<0.001),anemia(OR=2.14,95%CI 1.06-4.40,P=0.035),O blood group(OR=1.28,95%CI 1.04-13.46,P=0.048),tumor surface ulcer(OR=2.28,95%CI 1.26-4.18,P=0.007),and positive LNM reported on CT examination(OR=6.71,95%CI 3.41-13.88,P<0.001)were independent risk factors for CC-LNM,while the percentage of neutrophils >75%(OR=0.47,95%CI 0.22-0.97,P=0.044),the increase of tumor diameter(OR=0.64,95%CI 0.49-0.82,P<0.001),moderate differentiation(OR=0.31,95%CI 0.11-0.81,P=0.020),and well differentiation(OR=0.16,95%CI 0.06-0.39,P<0.001),were the independent protective factors of CC-LNM.A nomogram prediction model was constructed from independent influencing factors and was shown as an online version of an interactable dynamic nomogram.The model had the good discrimination(AUC=0.834,95%CI 0.755-0.855)and consistency(calibration curves showed that the predicted results of the nomogram were highly consistent with the true results).Internal validation showed a corrected C statistic of 0.810.The DCA curve indicated the net benefit of the model >0 over a threshold probability range of 2%-99%.The CIC curve showed that the number of positive LNM cases predicted by the nomogram was highly consistent with the actual number of positive patients when the threshold probability was ≥ 70%.the DCA curve,NRI,and IDI showed that the accuracy and clinical application of the nomogram in predicting LNM was significantly superior to that of preoperative abdominal contrast-enhanced CT.Conclusion 1.Intestinal obstruction,anemia,blood group O,neutrophil percentage >75%,small tumor diameter,hypofractionation,surface ulceration,and positive LN status on CT examination are independent risk factors for LNM in patients with CC.2.This study is the first to construct a web-based interactive dynamic nomogram for the preoperative prediction of CC-LNM.The nomogram uses eight common preoperative clinical characteristics as predictors,which can predict the risk of CC-LNM more accurately and help in preoperative disease accounting and guiding postoperative follow-up treatment,and has potential clinical application value.3.Compared with abdominal enhancement CT scan,nomogram has higher accuracy and better clinical application value in predicting LNM and can be used as a powerful supplement to preoperative abdominal enhancement CT examination. |
PDF Full Text Request