| Background:With the development of precision medicine and the diagnosis and treatment of gastric cancer,the comprehensive treatment mode of gastric cancer has gradually formed which is mainly based on surgical treatment.Postoperative pulmonary complications(PPCs)are one of the most common complications following surgery,which will significantly delay the postoperative recovery of patients,prolong the length of hospital stay,affect the prognosis and quality of life of patients,even endanger the patients’ life.Objectives:The objectives of this paper were to analyze the clinical high-risk factors for PPCs after radical gastrectomy and establish an individualized prediction model for clinical prevention and treatments guidance of PPCs.Methods:This retrospective study was based on gastric cancer(GC)patients who underwent radical gastrectomy in the Gastrointestinal Surgery Department of Qilu Hospital of Shandong University from January 2017 to December 2018.Risk factors that might be related to the occurrence of PPCs were identified by clinical observation and literature analysis,such as age,gender,BMI,smoking history,chronic diseases(including heart disease,diabetes,and high blood pressure),pulmonary function test(PFT),the American society of anesthesiologists(ASA)score,operation method,selection of gastrectomy,and operation time,etc.Then the univariate and multivariate logistic analyses were used to determine the high-risk factors.Those gastric cancer patients were defined as the modeling group,and the predicted factors were selected by the optimal subset regression analysis.Then by using the multivariate logistic regression analysis to develop the nomogram of the clinical risk factor prediction model.In addition,the validation group was based on GC patients who received the same treatments at the same institution from January 2019 to June 2020 to evaluate the model’s accuracy and applicability.The evaluating methods included the coordination index(C-index),the receiver operating characteristic(ROC)curve,and the calibration curve.Results:Among 485 patients who received radical gastrectomy in the modeling group,101 patients(20.8%)developed PPCs,while 44(25.4%)of 173 patients developed PPCs in the validation group.Univariate logistic analysis showed that male,age>50 years old,preoperative chronic underlying diseases(including heart disease,diabetes,hypertension),ASA score Ⅲ,moderate/severe pulmonary ventilation dysfunction,total laparoscopic surgery,total gastrectomy and operation time>5 hours were significantly correlated with PPCs(P<0.05).After adjusting the confounding factors,multivariate logistic analysis showed that age>50 years old,moderate/severe pulmonary ventilation dysfunction,and operation time>4 hours were independent risk factors for PPCs.After the optimal subset regression analysis,the prediction model was successfully developed by eight clinical factors including sex,age,body mass index(BMI),heart disease,American Society of Anesthesiologists(ASA)score,pulmonary function test(PFT),resection range of gastrectomy and operation time.The C-index of this prediction model of the modeling group and the validation group was 0.767 and 0.718,respectively,which combining the ROC curves proved that the model had good predictive ability and accuracy.The calibration curves also showed the satisfactory forecasting abilities of PPCs’ incidence in actual observation.Conclusions:Gender,age,BMI,heart disease,ASA score,PFT,selection of gastrectomy and operation time were reliable predictors of PPCs for gastric cancer.Among the factors,age>50 years,moderate/severe pulmonary ventilation dysfunction and operation time>4 hours were the independent risk factors of PPCs.The prediction model based on above high-risk factors could accurately assess the risk of PPCs in gastric cancer patients,and is beneficial to the prevention and treatment of PPCs following radical gastrectomy in clinical practice. |
PDF Full Text Request