Performance of a nomogram model established based on clinical indices and magnetic resonance imaging signs in the diagnosis of traditional Chinese medicine syndrome types of primary liver cancer / 临床肝胆病杂志

ABSTRACT

ObjectiveTo investigate the performance of a nomogram model established based on clinical indices and magnetic resonance imaging （MRI） signs in determining the traditional Chinese medicine （TCM） syndrome types of primary liver cancer. MethodsA retrospective analysis was performed for the clinical data of 138 patients with primary liver cancer who were hospitalized in The Affiliated Hospital of Shaanxi University of Chinese Medicine from September 2018 to July 2023， and the patients were divided into excess syndrome group with 84 patients and deficiency syndrome group with 54 patients. All patients underwent Gd-EOB-DTPA contrast-enhanced MRI scan before treatment. The independent-samples t test was used for comparison of continuous data between two groups， and the chi-square or the Fisher’s exact test was used for comparison of categorical data between groups. A Logistic regression analysis was used to investigate the independent predictive factors for the TCM syndrome type of primary liver cancer， and a nomogram model was established. The patients were randomly divided into training group with 110 patients and validation group with 28 patients at a ratio of 8∶2， and the calibration curve， the receiver operating characteristic （ROC） curve， and the decision curve were used to evaluate the clinical performance of this model. ResultsThere were significant differences between the excess syndrome group and the deficiency syndrome group in neutrophils， lymphocyte count （LYM）， platelet count， albumin （Alb）， neutrophil-lymphocyte ratio （NLR）， prothrombin time （PT）， alpha-fetoprotein （AFP）， direct bilirubin （DBil）， indirect bilirubin， total bilirubin， presence or absence of portal vein invasion， number of tumors， hepatobiliary tumor signal， and apparent diffusion coefficient （ADC） （all P<0.05）. The Logistic regression analysis showed that AFP （odds ratio ［OR］=0.003， 95% confidence interval ［CI］： 0.000‍ ‍—‍ ‍0.052， P<0.001）， PT （OR=0.032， 95%CI： 0.004‍ ‍—‍ ‍0.286， P=0.002）， LYM （OR=0.032， 95%CI： 0.004‍ ‍—‍ ‍0.286， P=0.002）， Alb （OR=0.009， 95%CI： 0.001‍ ‍—‍ ‍0.163， P=0.001）， NLR （OR=0.040， 95%CI： 0.003‍ ‍—‍ ‍0.457， P=0.010）， DBil （OR=0.014， 95%CI： 0.001‍ ‍—‍ ‍0.198， P=0.002）， portal vein cancer thrombus （OR=0.005， 95%CI： 0.000‍ ‍—‍ ‍0.115， P=0.001）， number of tumors （OR=12.740， 95%CI： 1.212‍ ‍—‍ ‍133.937， P=0.034）， and ADC （OR=19.269， 95%CI： 3.163‍ ‍—‍ ‍117.387， P=0.001） were independent predictive factors for TCM syndrome types of primary liver cancer. In the training group， the model had an area under the ROC curve （AUC） of 0.962， a sensitivity of 84.1%， a specificity of 92.4%， and an accuracy of 89.1%， and in the validation group， the model had an AUC of 0.848， a sensitivity of 63.6%， a specificity of 100.0%， and an accuracy of 85.7%. The calibration curve showed that the nomogram model had good consistency between predicted syndrome types and actual syndrome types in the training group and the validation group， and the decision curve showed that the nomogram model had good net benefits within a relatively wide range of threshold probability. ConclusionThe nomogram model based on clinical indices and MRI signs has good clinical efficacy and value in judging the TCM syndrome type of primary liver cancer.