The purpose of this study is to compare the effectiveness of a novel personalized surgical approach to the standard AT in children with small tonsils (ST). This will be accomplished by randomizing children with ST and OSA to one of these two treatments and comparing outcomes after 6 months. It is the investigators' central hypothesis that a personalized drug-induced sleep endoscopy (DISE)-directed surgical approach that uses existing procedures to address the specific fixed and dynamic anatomic features causing obstruction (ie, anatomic endotypes) in each child with ST will perform better than the currently recommended standard first line approach of AT. This novel approach may improve OSA outcomes and reduce the burden of unnecessary AT or secondary surgery for persistent OSA after an ineffective AT. To test this hypothesis, the investigators propose to study children aged 2-17 years with small tonsils and OSA.

Obstructive sleep apnea (OSA) is common in children with an estimated prevalence of 1-6%. Untreated pediatric OSA is associated with hypertension, autonomic dysfunction, attention-deficit / hyperactivity disorder, neurocognitive deficits, poor school performance, poor quality of life, and a >200% increase in health care utilization compared to controls. Adenotonsillar hypertrophy is the primary risk factor for pediatric OSA, and adenotonsillectomy (AT) is the recommended first line treatment. However, the prevalence of persistent OSA after AT has been reported to be at least 20% among children with large tonsils, indicating that AT may not be the optimal intervention in all cases. Although the prevalence of small tonsils in children with OSA symptoms is up to 70%, there is currently no clear evidence on the outcome of AT in children with small tonsils. It has been shown that clinic assessments of tonsil size do not correlate with OSA severity or response to AT, leaving confusion about how best to treat OSA in children with small tonsils. This knowledge gap represents an opportunity to apply a personalized treatment approach and improve outcomes.

DISE entails passage of a flexible endoscope through the nose into the pharynx and enables direct observation of the sites and patterns of upper airway obstruction during sedated sleep. DISE was developed to guide surgical decisions in adult OSA, and in recent years has also been used to design personalized surgical interventions in children. To help standardize DISE assessments, the investigators previously developed and validated the DISE Rating Scale in children based on ordinal ratings of maximal airway obstruction (none, partial, complete) at six anatomic sites from the nose to the larynx.

The investigators also demonstrated that DISE ratings of adenotonsillar obstruction during sedated sleep are strongly associated with both OSA severity and response to AT, unlike clinic assessments of tonsil size. The investigators' preliminary data and other published studies have identified a high prevalence of non-adenotonsillar obstruction that can occur at the nose, palate, base of tongue, and larynx. Surgery that is tailored to the specific anatomic sites of obstruction observed during DISE may improve pediatric OSA outcomes, but existing studies are small, uncontrolled case series in heterogeneous study populations. There are no trials directly comparing AT and DISE-directed surgery in the common scenario of surgically naïve children with small tonsils.

The purpose of this study is to compare the effectiveness of a novel personalized surgical approach to the standard AT in children with small tonsils (ST). This will be accomplished by randomizing children with ST and OSA to one of these two treatments and comparing outcomes after 6 months. It is the investigators' central hypothesis that a personalized drug-induced sleep endoscopy (DISE)-directed surgical approach that uses existing procedures to address the specific fixed and dynamic anatomic features causing obstruction (i.e., anatomic endotypes) in each child with ST will perform better than the currently recommended standard first line approach of AT. This novel approach may improve OSA outcomes and reduce the burden of unnecessary AT or secondary surgery for persistent OSA after an ineffective AT. To test this hypothesis, the investigators propose to study children aged 2-17 years with small tonsils and OSA.

To be in the study:

• Child has a diagnosis of moderate to severe OSA diagnosed by PSG (oAHI ≥ 5).
• Child age is 2.00 to 17.99 years of age.
• Child has small tonsils (Brodsky 1+ or 2+) noted during routine physical exam.
• Caregiver can provide signed and dated consent and is 18 years of age or older at the time of consent.
• Caregiver can speak, read, and write in English or Spanish.
• Caregiver is primary caretaker of the child.
• Child is not expecting their own child.
• Child is eligible for surgical treatment

Cannot be in the study if:

• Child has history of previous tonsillectomy, tonsillotomy, or partial tonsillectomy.
• Child has any contraindication to surgery (e.g. bleeding disorders).
• Child has significant cardiopulmonary comorbidity besides OSA requiring supplemental oxygen, subglottic or tracheal stenosis, tracheostomy dependence.
• Child has a genetic abnormality, Down syndrome, neuromuscular disorder, craniofacial anomaly.
• Caregiver is unwilling or unable to comply with study procedures.
• Child is or plans to become a parent themselves.

IPD will not be made available to researchers who are not primary researchers on this study.

It is not practical or feasible to maintain blinding among the surgeons performing the procedures or the clinical staff caring for patients in the perioperative setting. Given this, it is unlikely that blinding could be maintained among parents or caregivers of patients, therefore there will be no attempt to blind caregivers to the treatment assignment. However, the sleep medicine collaborators who will be reviewing and scoring postoperative PSGs will be instructed not to review the medical record so as to ensure blinding to treatment assignment.