Development and Validation of a Risk Nomogram Model for Perioperative Respiratory Adverse Events in Children Undergoing Airway Surgery: An Observational Prospective Cohort Study

Abstract

The aim of this study was to explore the associated risk factors of perioperative respiratory adverse events (PRAEs) in children undergoing airway surgery and establish and validate a nomogram prediction model for PRAEs. Patients and Methods: This study involved 709 children undergoing airway surgery between November 2020 and July 2021, aged ≤18 years in the affiliated hospital of Xuzhou Medical University. They were divided into training (70%; n = 496) and validation (30%; n = 213) cohorts. The least absolute shrinkage and selection operator (LASSO) was used to develop a risk nomogram model. Concordance index values, calibration plot, decision curve analysis, and the area under the curve (AUC) were examined. Results: PRAEs were found in 226 of 496 patients (45.6%) and 88 of 213 patients (41.3%) in the training and validation cohorts, respectively. The perioperative risk factors associated with PRAEs were age, obesity, degree of upper respiratory tract infection, premedication, and passive smoking. The risk nomogram model showed good discrimination power, and the AUC generated to predict survival in the training cohort was 0.760 (95% confidence interval, 0.695-0.875). In the validation cohort, the AUC of survival predictions was 0.802 (95% confidence interval, 0.797-0.895). Calibration plots and decision curve analysis showed good model performance in both datasets. The sensitivity and specificity of the risk nomogram model were calculated, and the result showed the sensitivity of 69.5% and 64.8% and specificity of 73.3% and 81.6% for the training and validation cohorts, respectively. Conclusion:The present study showed the proposed nomogram achieved an optimal prediction of PRAEs in patients undergoing airway surgery, which can provide a certain reference value for predicting the high-risk population of perioperative respiratory adverse events and can lead to reasonable preventive and treatment measures.

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 , …”

A novel nomogram for predicting respiratory adverse events during transport after interventional cardiac catheterization in children

“…Bias-corrected calibration curve showed that the model could accurately predict the occurrence of RAEs during transport when the observed probability of RAEs during transport was less than 40%. Inversely, a few existing models that cannot accurately and objectively predict RAEs during anesthesia induction or perioperative period, and do not include a special type of surgery such as interventional cardiac catheterization (8,(30)(31)(32). Based on the clinical data of 19,095 pediatric patients undergoing elective ambulatory anesthesia for surgery and radiology, Subramanyam et al developed and validated a risk prediction model for the occurrence of RAEs from the onset of anesthesia induction until discharge from the PACU, with a Cstatistic of 0.64 (8).…”

Advances in pediatric perioperative care using artificial intelligence

“…Other studies looked at predictive models to prevent desaturation <95%, laryngospasm, bronchospasm, prolonged cough, airway obstruction, stridor, pneumonia, atelectasis, pneumothorax and pyrexia have been performed as well [17][18][19]. Similarly, decision trees and K-means classifiers were used to develop an Automated Unsupervised Respiratory Event Analysis (AUREA) model to classify breathing patterns obtained through dual belt respiratory inductance plethysmography (RIP) in infants [20].…”