Can pediatric sleep questions be incorporated into a risk model to predict respiratory complications following adenotonsillectomy?

Adenotonsillectomy, one of the most frequent surgical procedures in children, is usually performed for sleep-disordered breathing, a disease spectrum from primary snoring to obstructive sleep apnea. Children undergoing an adenotonsillectomy may be at risk for perioperative respiratory complications, necessitating intervention or escalation of care. However, there is no effective preoperative screening or risk-stratification model for perioperative respiratory complications that incorporates not only clinical history and physical examination but also sleep question responses for children as there is for adults. The aim of this prospective observational study was to develop a risk-stratification model for perioperative respiratory complications in children undergoing an adenotonsillectomy incorporating not only clinical history and physical examination but also sleep question responses. A 25-question sleep questionnaire was prospectively administered preoperatively for 1895 children undergoing an adenotonsillectomy from November 2015 to December 2017. The primary outcome measure was overall perioperative respiratory complications, collected prospectively and defined as having at least one major or minor complication intraoperatively or postoperatively. The incidence of overall perioperative respiratory complications was 20.4%. Preoperative factors associated with perioperative respiratory complications in the multiple regression model were age, race, preoperative tonsil size, the presence of a syndrome, and the presence of a pulmonary disease. None of the sleep questionnaire responses remained in the multivariable analysis. The area under the ROC curve for the risk stratification model incorporating sleep question responses was only 0.6114% (95% CI: 0.60, 0.67). Preoperative sleep question responses may be unable to predict overall perioperative respiratory complications in children undergoing an adenotonsillectomy. A robust risk stratification model incorporating sleep question responses with clinical history and physical examination was unable to discriminate or predict perioperative respiratory complications in our population undergoing an adenotonsillectomy. •A model incorporating sleep question responses with clinical history and physical examination does not discriminate or predict perioperative respiratory complications in children undergoing an adenotonsillectomy.•Preoperative sleep question responses are unable to predict overall perioperative respi