Postoperative delirium prediction model for elderly patients undergoing orthopedic surgery / 中华创伤骨科杂志

ABSTRACT

Objective To establish a preliminary model which can effectively predict the risk for postoperative delirium (POD) in elderly orthopedic patients and verify its effectiveness.Methods This prospective study involved 2 cohorts.For an analysis cohort,the assessment data of 148 elderly orthopedic patients were collected who had been treated at Department of Orthopaedics,the First Affiliated Hospital of Zhengzhou University from June to October 2017.The relevant risk factors for POD (gender,age,BMI,schooling ＜ 9 years,history of smoking and alcohol drinking,concomitant diseases and perioperative factors) were screened after comparing POD and non-POD patients.All the risk factors were analyzed and a predictive model was established after valuation of independent risk factors.A predictive cohort of 66 patients was included according to the same inclusion and exclusion criteria as the analysis cohort out of the patients who had been treated at our hospital from November to December 2017.The 2 cohorts were scored by the predictive model to verify the validity of the model.Results A total of 18 risk factors were identified in this study.In the analysis cohort,age (P =0.006),schooling ＜ 9 years (P =0.043),cerebrovascular disease or mental illness (P =0.004),preoperative albumin (P =0.038) and intraoperative infusion of allogeneic blood (P =0.019) were risk factors for POD.Of them,age (P =0.037),schooling ＜ 9 years (P =0.003) and intraoperative infusion of allogeneic blood (P =0.042) were independent ones.In the predictive model,age ＞ 75 years was assigned 3 points,schooling ＜ 9 years 2 points and intraoperative infusion of allogeneic blood 5 points.The validity of the ROC curve was verified for the predictive model.According to the ROC curve,the analysis cohort had AUC =0.66 and the predictive cohort AUC =0.75,indicating a certain predictive value of the model.Conclusion Our predictive model can be used to effectively screen out those with a high risk for postoperative delirium from elderly orthopedic patients.