ObjectiveTo establish the risk prediction models for delayed emergence in elderly patients undergoing non-cardiac surgery and evaluate the predictive efficiency.

MethodsA total of 778 elderly patients(both sexes, age≥65 years) of ASA grade Ⅱ-Ⅲ, who undergoing non-cardiac surgery under general anesthesia were enrolled.The perioperative clinical data of patients were collected and the occurrence of delayed emergence was recorded.The ROC curve analysis was carried out for meaningful measurement date, and the best cut-off value was calculated to be transformed into classified data.Multivariate logistic regression analysis was performed on all characteristic variables to investigate the independent risk factors of delayed emergence, and nomogram prediction model was drew.Discrimination, calibration and clinical usefulness of the predicting model were assessed using the C-index, calibration plot and decision curve analysis.Internal validation was assessed using the bootstrapping validation to calculate the corrected C-index.

ResultsA total of 718 elderly patients were enrolled in this study, 66 patients developed delayed emergence, and the incidence was 9.2%.Predictors contained in the prediction nomogram included age >74 years, preoperative MMSE score≤25 points, preoperative Fried′s phenotype score≥3 points, complication with chronic obstructive pulmonary disease, intraoperative hypotension, intraoperative bispectral index < 45(P < 0.01).The model displayed the good discrimination with a C-index of 0.750 (95%CI: 0.679-0.821) and good calibration.Corrected C-index value of 0.743 was reached in the interval validation.Decision curve analysis showed that the nomogram model predicting the risk threshold of delayed emergence was 2.0%-81.5%, which increased the clinical benefit.

ConclusionsThis novel nomogram incorporating age, preoperative MMSE, preoperative Fried′s phenotype score, history of chronic obstructive pulmonary disease, intraoperative hypotension and intraoperative bispectral index can be conveniently used to facilitate the delayed emergence risk prediction in elderly patients undergoing non-cardiac surgery, which has the good predictive efficiency.