Risks Associated With Blood Transfusion After Knee Arthroplasty And The Predictive Model Is Established

Objective: Blood transfusion is one of the adverse events after knee replacement,the aim of this study was to investigate the risk factors of blood transfusion for knee arthroplasty patients after knee replacement and establish the nomogram to predict the probability of the postoperative blood transfusion.This study would reduce the complications related to blood transfusion and decrease financial losses.Methods: We collected and retrospectively analyzed the data from the patients with knee arthroplasty in our hospital from 2018.01 to 2021.12 via the case system and telephone line follow-up.The patients were divided into the case group and the control group based on blood transfusion.Collected data includes:(1)the general information of the objects;(2)basic disease history;(3)preoperative hemoglobin,red blood cells deposited preoperatively,smoking history,the types of knee arthritis;(4)ASA grading,anesthesia method,intraoperative use of tranexamic acid,intraoperative use of a tourniquet time,operation time,unilateral or bilateral,total knee surgery arthroplasty(TKA),or unicompartmental knee arthroplasty(UKA)if admitted to the hospital a staging joint replacement will be subject to the first joint replacement;(5)Intraoperative and postoperative blood transfusion or blood transfusion.SPSS statistical software was used to collect information for statistical analysis for single factors and multiple factors and select independent risk factors.Using R language on the independent risk factors of the regression coefficient weighted analysis,and build the Nomogram prediction model.Results: The data from a total of 597 patients who met inclusion and exclusion criteria with knee arthroplasty were collected,in which case group was 50 cases consisting of 7 cases of male and female 43 patients.The control group was 547 cases consisting of 105 cases of male and females and 442 cases.The statistical analysis showed that the weight,preoperative hemoglobin,digestive system diseases,intraoperative use of tranexamic acid,and operation time were independent risk factors for knee joint replacement patients with postoperative blood transfusion.The nomogram was successfully established using the R language,and the evaluation,C-the index of 0.832(95% CI: 0.777 ～ 0.887),with 1000 times bootstrapping from internal validation sampling,to draw a forecasting model C-the index was 0.811,area under the receiver-operating characteristic curve(ROC)was 0.832,the prediction model of Calibration curve(Calibration curve)overall slope was near 1.The results showed that the model predicts the knee joint replacement had good accuracy of perioperative blood transfusion.The Clinical decision curve indicated better clinical practicability.Conclusion: The preoperative hypohemoglobin,low weight,digestive system disease,intraoperative unused tranexamic acid,and long operation time in the knee joint replacement patients were independent risk factors for postoperative blood transfusion.The combination of the above six predictors of nomogram accurately predicted the risk of blood transfusion.After total knee arthroplasty,the Nomogram prediction model can be individualized knee arthroplasty after blood transfusion,which benefited the early detection and intervention of knee arthroplasty after blood transfusion.