OBJECTIVE To explore the influencing factors of postoperative delirium (POD)in Parkinson's disease (PD)patients over 50 years old after deep brain electrical stimulation(DBS), and to construct an nomogram model to predict… Click to show full abstract
OBJECTIVE To explore the influencing factors of postoperative delirium (POD)in Parkinson's disease (PD)patients over 50 years old after deep brain electrical stimulation(DBS), and to construct an nomogram model to predict the risk of POD. METHODS A prospective cohort study was conducted on 229 PD patients over 50 years old hospitalized from January 2015 to August 2019. All subjects received deep brain stimulation during hospitalization.The incidence of POD of patients and clinical data such as gender, age, educational level and duration of disease were collected, the influencing factors of POD of Parkinson patients were analyzed, and an nomogram model for predicting the risk of POD was established. RESULTS POD occurred in 47 of 229 patients with PD, with an incidence rate of 20.52%. Gender, age, PDSS, preoperative cerebral ischemia, preoperative pulmonary inflammation and preoperative length of stay are the influencing factors of POD in Parkinson patients over 50 years old (P<0.05).After deviation correction, the area under receiver operating characteristic (ROC)curve of nomogram model for predicting POD risk is 0.755, the sensitivity is 74.47%, and the specificity is 73.08%. Meanwhile, The mean absolute error between the prediction of POD risk by nomogram model and actual delirium risk is 0.024, suggesting that nomogram model has good prediction efficiency. CONCLUSION Based on the influencing factors of delirium after DBS for PD patients over 50 years old, the nomogram model for predicting the risk of POD has been established in this study. The prediction efficiency is good, which can provide reference for early clinical identification of high-risk patients and formulate intervention countermeasures as soon as possible.
               
Click one of the above tabs to view related content.