Risk over time and risk factors of intraoperative respiratory events: a historical cohort study of 14,153 children

BACKGROUND: The variation in the rate of intraoperative respiratory events (IRE) over time under anesthesia and the influence of anesthesia-related factors have not yet been described. The objectives of this study were to describe the risk over time and the risk factors for IRE in children at a tertiary care hospital in southern Thailand. METHODS: The surveillance anesthetic database and chart review of IRE of 14,153 children who received surgery at Songklanagarind Hospital during January 2005 to December 2011 were used to obtain demographic, surgical and anesthesia-related data. Incidence density of IRE per person-time was determined by a Poisson modelling. Risk of IRE over time was displayed using Kaplan Meier survival and Nelson-Aalen curves. Multivariate Cox regression was employed to identify independent predictors for IRE. Adjusted hazard ratios (HR) and their 95% confidence intervals (CI) were obtained from the final Cox model. RESULTS: Overall, IRE occurred in 315 out of 14,153 children. The number (%) of desaturation, wheezing or bronchospasm, laryngospasm, reintubation and upper airway obstruction were 235 (54%), 101 (23%), 75 (17%), 21 (5%) and 4 (1%) out of 315 IRE, respectively. The incidence density per 100,000 person-minutes of IRE at the induction period (61.3) was higher than that in the maintenance (13.7) and emergence periods (16.5) (p < 0.001). The risk of desaturation, wheezing and laryngospasm was highest during the first 15, 20 and 30 minutes of anesthesia, respectively. After adjusting for age, history of respiratory disease and American Society of Anesthesiologist (ASA) classification, anesthesia-related risk factors for laryngospasm were assisted ventilation via facemask (HR: 18.1, 95% CI: 6.4-51.4) or laryngeal mask airway (HR: 12.5, 95% CI: 4.6-33.9) compared to controlled ventilation via endotracheal tube (p < 0.001), and desflurane (HR: 11.0, 95% CI: 5.1-23.9) compared to sevoflurane anesthesia (p < 0.001). CONCLUSIONS: IRE risk was highest in the induction and early maintenance period. Assisted ventilation via facemask or LMA and desflurane anesthesia were anesthesia-related risk factors for laryngospasm. Therefore, anesthesiologists should pay more attention during the induction and early maintenance period especially when certain airway devices incorporated with assisted ventilation or desflurane are used.