Polysomnography variables associated with postoperative respiratory issues in children <3 Years of age undergoing adenotonsillectomy for obstructive sleep apnea

Objectives: To determine the polysomnogram (PSG) variables associated with increased incidence of postoperative respiratory complications in children <3 years of age undergoing adenotonsillectomy (T&A) for the management of obstructive sleep apnea (OSA). Methods: Retrospective case series of children <3 years of age who underwent T&A for the management of OSA at a tertiary care children's hospital from 1/1/08–6/1/2018. Postoperative hospital courses were analyzed to determine if certain clinical or PSG variables might have predicted an increased rate of respiratory complications. Results: A total of 195 children <3 years of age with OSA on preoperative PSGs underwent T&A. The mean age of patients was 25.8 months (range 9–35 months). Seventeen patients (16.2%) had mild OSA, 35 (17.9%) had moderate, and 141 (72.3%) had severe OSA. Most patients (n = 155, 79.5%) required no respiratory intervention postoperatively. Those with severe OSA were more likely to require oxygen support (n = 25, 17.7%) when compared to those with mild-moderate OSA (n = 3, 5.6%). Patients admitted directly to the pediatric intensive care unit (PICU) from the sleep lab (n = 7) had a mean apnea-hypopnea index (AHI) of 63.6 events/hour (range 23–146/hr.), and a mean SpO₂ nadir of 56.3% (range 46–68%). Four of these patients required CPAP after T&A. Of the 7 patients (3.6%) who required intubation after their procedure, only 1 did not have severe OSA, and 2 were intubated to manage post-obstructive pulmonary edema (POPE). Only 4 patients (4.2%) had unplanned PICU admissions. An AHI >12.7 events/hour, >18.5 obstructive apneas, and SpO2 nadir <72.5% were associated with a greater likelihood of requiring postoperative respiratory interventions. Conclusions: Most children <3 years of age, even with severe OSA, had no respiratory issues postoperatively. Those with severe OSA and hypoxemia admitted directly from the sleep lab were more likely to require CPAP postoperatively. All but one patient who developed POPE or who required intubation had severe OSA with associated hypoxemia on their preoperative PSGs.