Development and validation of a postoperative delirium prediction model for pediatric patients

Abstract

Postoperative delirium is a serious complication that relates to poor outcomes. A risk prediction model could help the staff screen for children at high risk for postoperative delirium. Our study aimed to establish a postoperative delirium prediction model for pediatric patients and to verify the sensitivity and specificity of this model. Data were collected from a total of 1134 children (0–16yr) after major elective surgery between February 2020 to June 2020. Demographic and clinical data were collected to explore the risk factors. Multivariate logistic regression analysis was used to develop the model, and we assessed the predictive ability of the model by using the area under the receiver operating characteristics curve (AUROC). Further data were collected from another 100 patients in October 2020 to validate the model. Prevalence of postoperative delirium in this sample was 11.1%. The model consisted of 5 predictors, namely, age, developmental delay, type of surgery, pain, and dexmedetomidine. The AUROC was 0.889 ( P < .001, 95% confidence interval (CI):0.857–0.921), with sensitivity and specificity of 0.754 and 0.867, and the Youden of 0.621. The model verification results showed the sensitivity of 0.667, the specificity of 0.955. Children undergoing surgery are at risk for developing delirium during the postoperative period, young age, developmental delay, otorhinolaryngology surgery, pain, and exposure to dexmedetomidine were associated with increased odds of delirium. Our study established a postoperative delirium prediction model for pediatric patients, which may be a base for development of strategies to prevent and treat postoperative delirium in children.

Use of artificial intelligence in paediatric anaesthesia: a systematic review

“…Machine learning was the most represented with 38 included studies describing its use, 13 23 28 29 31 32 , 34 , 35 , 36 , 38 , 39 , 40 , 42 44 46 47 , 49 , 50 , 51 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 70 whereas two applications incorporated the use of fuzzy logic 21 , 22 and only one study described the use of each of natural language processing 23 and computer vision. 13 Many branches of machine learning were discussed, including regression models, 29 45 46 49 50 51 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 decision trees, 23 , 33 , 34 , 35 expert systems, …”

“… 13 Many branches of machine learning were discussed, including regression models, 29 45 46 49 50 51 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 decision trees, 23 , 33 , 34 , 35 expert systems, 59 K-means classifiers, 35 K-nearest neighbours, 29 , 44 Bayesian approaches, 29 neural networks, 23 , 25 , 34 , 45 random forest models, 23 29 31 , 45 , 46 , 47 and support vector machines. 34 , 45 , 70 Regression models were the most commonly described, with one study describing a linear regression model, 50 23 studies describing logistic regression models, 29 36 46 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , …”

“…33 Shim and colleagues 45 Jiao and colleagues 23 Chiesa and colleagues 46 Safranek and colleagues 47 Hu and colleagues 29 Ward and colleagues 31 Regression: Linear, Logistic, Elastic Net, LASSO Characterising the strength of the relationship between a dependent variable and one or more explanatory variables. 48 Shim and colleagues 45 Zhang and colleagues 49 Guevara and colleagues 50 Agostoni and colleagues 51 Alassaf and Reitsma 52 Ali and colleagues 53 Ammer and colleagues 54 Ariza and colleagues 55 Cheon and colleagues 56 Chiesa and colleagues 46 Fishman and colleagues 57 Hino and colleagues 58 Hu and colleagues 29 Kawaguchi and colleagues 59 Kim and colleagues 60 Lee and colleagues 61 Lin and colleagues 62 Nafiu and colleagues 63 Nasr and colleagues 64 Packiasabapathy and colleagues 65 Spencer and colleagues 66 Tao and colleagues 67 Vlasov 68 Ward and colleagues 31 Support vector machines Classifies data by creating a decision boundary, known as the hyperplane, that is orientated as far as possible from the closest data points from each observed class of data. 69 Galvez and colleagues 34 Shim and colleagues 45 Chini and colleagues 70 …”

Advances in pediatric perioperative care using artificial intelligence

“…A model for prediction of postoperative delirium in pediatric patients who had major elective surgery demonstrated good predictability [23]. Similarly, a predictive model for emergence agitation in children receiving sevoflurane anesthesia based upon age, anesthesia time, Pediatric Anesthesia Behavior score (PAB) and operative procedure was associated with good predictive values [24].…”

Perioperative neurocognitive disorders: A narrative review focusing on diagnosis, prevention, and treatment

“…The development of perioperative NCDs results from the interaction of multiple factors. Many prediction models have been developed to predict the risk of perioperative NCDs 41,91–96 . However, none of these models have been externally validated or widely accepted for clinical use.…”