| Objective : By comparing the different immune inflammation and nutritional indicators that affect the postoperative complications of Pancreaticoduodenectomy(PD),to find the best evaluation index for the postoperative complications of PD;to select the appropriate prognostic indicators,combined with the influencing factors of postoperative complications of PD,a predictive model for postoperative complications of PD was constructed and verified.Methods : The clinical data of PD patients treated in the First Hospital of Lanzhou University from January 2017 to December 2021 were retrospectively collected.Propensity Score Matching(PSM)was used to balance the clinical baseline characteristics between groups.Logistic univariate and multivariate analysis and receiver operating characteristic curve(ROC)were used to evaluate the relationship between different immune inflammation and nutritional indicators and severe complications after PD.The total sample was divided into training group and validation group by computer random number method.Logistic regression analysis was used to screen the risk factors affecting the occurrence of severe complications in patients after PD,and a nomogram prediction model was constructed and evaluated.Results :1.To investigate the effect of inflammatory nutritional indicators on severe complications after PD : 236 patients were included in this study.The optimal cutoff values of neutrophil-to-white blood cell ratio(NWR),neutrophil-to-lymphocyte ratio(NLR),Prognostic Nutritional Index(PNI)and Advanced Lung Cancer Inflammation Index(ALI)were calculated to be 0.62,1.71,45.4 and 200.17,respectively.Univariate analysis of the total sample showed that anemia,CA199 ≥35U / ml,TNM stage,T stage,N stage,ALB,LYM,NWR > 0.62,NLR > 1.71,PNI <45.4,ALI < 200.17,and Controlling Nutritional Status(CONUT)score ≥ 2 were risk factors for severe complications after PD.The distribution of each group was balanced by PSM method.NWR,NLR,PNI,ALI and CONUT scores were put into logistic multivariate regression analysis respectively.The results showed that NLR(P= 0.012),PNI(P = 0.046)and ALI(P = 0.026)were independent risk factors for severe complications after PD.ROC curve was drawn to compare the predictive efficacy of preoperative NLR,PNI or ALI.PNI had higher efficacy in predicting severe complications after PD,with sensitivity and specificity above 60 %.NLR has a higher sensitivity,but the specificity is not high.2.Construction and evaluation of nomogram model : The total samples were divided into training group(n = 189)and validation group(n = 47)by computer random number method.Univariate and multivariate analysis of the training group showed that TNM stage,CA199 > 35.00 U / ml,anemia and NLR > 1.71 were independent risk factors for severe complications in patients after PD.Age and PNI were added to the model and a nomogram model was constructed.The prediction performance of the evaluation model showed that the consistency index(C-index)of the training group and the validation group were 0.756 and 0.705,respectively,and the AUC values were 0.756 and 0.631,respectively,indicating that the model had good discrimination.The calibration curve indicated that the prediction accuracy of the model was good,but the generalization ability was insufficient.The clinical influence curve and clinical decision curve show that the model has good clinical value.Conclusion : NLR ≥ 1.71,PNI < 45.4 and ALI < 200.17 were independent risk factors for severe complications in patients after PD.PNI has higher efficacy in predicting severe complications after PD,and its sensitivity and specificity are both above 60 %.The logistic nomogram model combined with NLR,PNI,anemia,age,CA199 and TNM staging can make a more accurate prediction of serious complications in patients after PD,and the model has good discrimination,accuracy,clinical practicability and stability. |
PDF Full Text Request