Risk factors for desaturation after tonsillectomy

Analysis of 4092 consecutive pediatric cases


  • Presented at the Triological Society 115th Annual Meeting at COSM, San Diego, California, U.S.A., April 20, 2012.

  • The authors have no funding, financial relationships, or conflicts of interest to disclose.

Send correspondence to David W. Roberson, MD, Department of Otolaryngology, Children's Hospital Boston, 300 Longwood Avenue, Boston, MA 02115. E-mail: david.roberson@childrens.harvard.edu



To identify clinical risk factors for oxygen desaturation in the first 24 hours post-tonsillectomy, thus permitting the identification of those patients who warrant inpatient monitoring.

Study Design

A retrospective analysis of 4092 consecutive patients undergoing tonsillectomy over a two-year period.


Detailed clinical data were recorded for all patients who desaturated in the postoperative period (n = 294) and randomly selected controls (n = 368). Univariate and Multivariate analysis was performed in order to identify independent risk factors for desaturation.


There were 294/4092 patients (7.2%) who experienced desaturations (defined as sustained saturations <90%) in the first 24 hours postoperatively (mean nadir, 78.7%). Multivariate analysis identified seven independent clinical risk factors for desaturation in the initial 24 hours post-tonsillectomy: trisomy 21, weight, coexistent cardiac disease, a coexistent syndromic diagnosis, a clinical diagnosis of obstructive sleep apnea (OSA), a coexistent neurologic diagnosis, and a prior diagnosis of pulmonary disease. A policy that admits all patients exhibiting any one of these risk factors except OSA would have identified 92% of the patients who subsequently desaturated. However, such a policy would also have required admission of 60% of the patients in our control group.


These findings are generally consistent with the Clinical Practice Guidelines recently published by the American Academy of Otolaryngology. In a tertiary care center, it may not be possible to identify an algorithm that admits all children at risk of desaturation while permitting the discharge of a high percentage of patients.

Level of Evidence

3b. Laryngoscope, 123:2554–2560, 2013