Construction and validation of a risk prediction model for delayed emergence from anesthesia in orthopedic surgery patients

Liu Pei; Qi Chun; Huang Li; Wang Ting; Wei Xi; Kuang Yuan; Tang Xiaojuan

doi:10.11986/j.issn.1673-873X.2025.07.10

Abstract：Objective To establish a risk prediction model for delayed emergence from anesthesia in orthopedic surgery patients,and to validate and evaluate its predictive performance,with a view to providing a reference for clinical decision-making.Meth-ods A total of 357 patients who underwent elective orthopedic surgery at a hospital in Wuhan from July 2021 to July 2024 were enrolled and randomly assigned,in a 7∶3 ratio,to a modeling(training)group(n=249)and a validation group(n=108).Perioperative clinical data were collected,and occurrences of delayed emergence were recorded.Multivariable logistic re-gression was used to identify independent risk factors for delayed emergence,and a nomogram risk-prediction model was con-structed based on the regression results;the model's predictive performance was then assessed.Results There were no signifi-cant differences between the modeling and validation groups in age,sex distribution,anesthetic technique or other baseline clin-ical characteristics(P＞0.05).Within the modeling group,compared with patients without delayed emergence,those with de-layed emergence differed significantly in age,body mass index(BMI),intraoperative fluid volume,blood loss,hemoglobin value,pain score,incidence of intraoperative hypothermia,and use of sedative medications(all P＜0.05).Stepwise regression analysis identified age,BMI,pain score,intraoperative fluid volume,and use of sedative medications as independent factors as-sociated with delayed emergence(P＜0.05).A nomogram based on these factors was constructed,and its discriminatory abili-ty was evaluated by receiver operating characteristic(ROC)analysis:the area under the curve(AUC)was 0.95(95％CI:0.92-0.98)in the modeling group and 0.72(95％CI:0.61-0.83)in the validation group,indicating good discrimination in both cohorts.The Hosmer-Lemeshow goodness-of-fit test showed x2=3.03,P=0.930 for the modeling group and x2=5.00,P=0.760 for the validation group,and the calibration curves' slopes were close to 1.Decision curve analysis(DCA)demonstrated that the model provides net clinical benefit for identifying high-risk patients,particularly at lower high-risk thresholds,indicating good clinical utility.Conclusion Age,BMI,pain score,intraoperative fluid volume,and use of sedative medications are independent risk factors for delayed emergence from anesthesia in orthopedic surgery patients.The nomogram risk-prediction model constructed from these factors shows good discrimination,accuracy,and clinical usefulness.