Predictors for postoperative oxygen therapy and postoperative mechanical ventilation need in children who underwent non-cardiac surgery under general anesthesia
Siriwimol ratprasert, MD, Maliwan Oofuvong, MD, PhD, Thavat Chanchayanon, MD
Department of Anesthesia, Prince of Songkla University, Hat Yai, Songkhla, 90110, Thailand
Background: Postoperative oxygen therapy in children is not only related to postoperative hypoxemia but also related to other factors during anesthesia which has never been investigated. To identify predictors for postoperative oxygen therapy might reduce short term consequences of postoperative oxygen therapy such as length of hospital stay. The objectives of this study were to determine predictors for the need for postoperative oxygen devices and postoperative mechanical ventilation use among children undergoing non-cardiac surgery under general anesthesia.
Methods: This was a cross sectional study of a prospective study of Oofuvong’s regarding perioperative respiratory event in children. We recruited children aged < 15 years, ASA physical status I-III who scheduled for non-cardiac surgery and underwent general anesthesia between November 2012 and December 2013 at Songklanagarind Hospital. Demographic, surgical-related and anesthesia-related data were recorded. The association between oxygen devices and postoperative mechanical ventilation need and potential predictors were determined by Wilcoxon rank sum test and Chi-square test. The multivariate polytomous logistic regression analysis was performed to identify independent predictors of oxygen devices and mechanical ventilation need postoperatively. The relative risk ratios and 95% confidence interval (CI) were displayed and considered significantly if p-value were < 0.05.
Results: We included 1334 children in the study. The incidences of postoperative oxygen devices and mechanical ventilation were 4.05% and 2.32%. Age group distributions were < 1 year (10.64%), 1-6 years (57.80%) and > 6 years (31.56%). The incidences of intraoperative and PACU respiratory events were 9.3% and 8.62%. Predictors for postoperative oxygen devices and mechanical ventilation needed were similar which were ASA physical status 3 (p < 0.01), low BMI (< 15 kg/m2, p < 0.05), thoracic surgery compared to eye surgery (p < 0.01), long anesthetic time (p < 0.05) and having intraoperative and PACU respiratory events (p < 0.001). Another predictor of oxygen devices was endotracheal intubation compared to face mask or LMA used (p < 0.01). Postoperative oxygen devices need was significantly associated with longer PACU time (median and IQR; 60 (50,80) in OD group vs 45 (35,60) in NO group, p < 0.001), longer days of hospitalization (median and IQR; 4 (2.3,9) in OD group vs  2 (1,3) in NO group, p < 0.001) and higher hospital cost (median and IQR; 22602.5 (15758,40978.5) in OD group vs 13861 (9924,23329) in NO group, p < 0.001).
Conclusions: The common predictors of postoperative oxygen therapy needed were ASA physical status, longer anesthetic time and having perioperative respiratory events. To reduce perioperative respiratory events by closed monitoring and increased vigilance of anesthesiologist might decrease days of hospitalization and hospital charge related to postoperative oxygen therapy in pediatric anesthesia.

Predictors of postoperative oxygen therapy in children who underwent non-cardiac surgery under general anesthesia