Overview
Tonsillectomy is one of the most common operations performed in children, often recommended for recurrent throat infections or, more commonly in this setting, obstructive sleep apnea (OSA). In children with severe OSA and obesity, current practice guidelines usually favor overnight hospital admission after surgery because these children are considered at higher risk for breathing problems after anesthesia and airway surgery. However, most children still recover without major complications. This study asked a practical question: among children who are already considered high risk, can we identify a subgroup that is still safe for same-day discharge?
Why this matters
OSA occurs when the upper airway repeatedly narrows or collapses during sleep, causing breathing pauses, low oxygen levels, and fragmented sleep. In children, enlarged tonsils and adenoids are common contributors. Obesity can worsen airway narrowing and make postoperative breathing issues more likely. For that reason, children with severe OSA and obesity are often admitted overnight even if they appear well immediately after surgery. While this cautious approach can improve safety, it also increases hospital use, family burden, and cost. A better risk-stratification tool could help clinicians personalize admission decisions instead of using one rule for every child.
Study design
Researchers performed a retrospective cohort study at a tertiary pediatric hospital from 2021 to 2024. They reviewed children aged 2 to 18 years who had both obesity, defined as a body mass index at or above the 95th percentile, and severe OSA before tonsillectomy. Severe OSA was defined by polysomnography findings of an apnea-hypopnea index, or AHI, of at least 10 events per hour and/or an oxygen saturation nadir below 80%.
The main outcome was a severe perioperative event, which included admission to the intensive care unit, a prolonged hospital stay of more than 48 hours, or the need for advanced respiratory support after surgery. The investigators used Bayesian logistic regression with informative priors to identify which preoperative features best predicted risk. They then tested model-based risk grouping and simpler clinical rules to see how well they could identify children with very low risk of severe complications.
Key findings
A total of 304 children met the study criteria. Of these, 36 children, or 11.8%, experienced a severe perioperative event. That means most children did not have major postoperative problems, even though they were in a group generally considered high risk.
The strongest predictor of a severe event was the lowest oxygen saturation during the sleep study, called the SpO2 nadir. Children with a lower oxygen nadir were more likely to have complications. Class III obesity was also associated with higher risk, and AHI contributed additional predictive value. In practical terms, the sleep study results were more informative than age or other commonly considered factors.
The model-based risk stratification identified about one-third of patients as having a predicted risk of less than 5% for a severe perioperative event. In other words, a meaningful subset of children with obesity and severe OSA still appeared to have a very low complication risk based on their preoperative sleep study results.
The researchers also tested a simpler clinical rule: AHI less than 25 events per hour combined with an oxygen saturation nadir greater than 85%. This rule showed a sensitivity of 85.7% and a negative predictive value of 99.5%. The observed rate of severe events in the low-risk group was under 5%, suggesting that this rule may be useful for identifying children who could potentially be considered for same-day discharge planning.
What the results mean
This study supports the idea that not all children with obesity and severe OSA carry the same postoperative risk. Even within a population traditionally managed with automatic overnight observation, many children may be safe for discharge on the day of surgery if their polysomnography results suggest lower risk.
The most important signal came from the oxygen saturation nadir. This makes clinical sense: a child who drops to very low oxygen levels during sleep may have less reserve if airway swelling, pain medications, or residual anesthesia temporarily worsen breathing after surgery. By contrast, a child whose oxygen levels remain relatively stable during sleep may tolerate recovery better.
Class III obesity also mattered, likely because more severe obesity is linked to more difficult airway mechanics, reduced respiratory reserve, and greater vulnerability to postoperative obstruction. AHI, which reflects the frequency of breathing interruptions, added value as well, but it was not as strong a predictor as the oxygen nadir.
Clinical implications
These findings suggest a more individualized approach to postoperative admission after tonsillectomy. Instead of admitting every child with obesity and severe OSA, clinicians may be able to use sleep study data to separate truly high-risk patients from those who may be low risk despite fitting broad guideline categories.
A potential benefit of this approach is reduced unnecessary hospitalization. Same-day discharge can lower costs, minimize disruption to family routines, and reduce the chance of hospital-associated complications. It may also improve bed availability in children’s hospitals. At the same time, this should not be interpreted as a reason to routinely send all such children home. The authors emphasize that these results need prospective validation before being used widely in practice.
In real-world decision-making, discharge after tonsillectomy must also consider factors beyond polysomnography. These include the child’s age, postoperative pain control, baseline medical conditions, distance from emergency care, family reliability, and the presence of any intraoperative or immediate recovery-room concerns. Children who receive opioids, have difficult airway management, or show oxygen desaturation in the recovery unit may still warrant admission even if their sleep study suggests low risk.
Strengths and limitations
A strength of this study is that it focused on a clinically important and common dilemma in pediatric otolaryngology: how to safely manage children who are considered high risk by current guidelines. Another strength is the use of both a statistical model and a simplified rule, making the findings potentially useful for both researchers and clinicians.
There are also important limitations. The study was retrospective, which means it relied on previously collected data and may be affected by missing information or selection bias. It was conducted at a single tertiary children’s hospital, so the results may not apply equally to all settings. In addition, the definition of severe perioperative events combined several outcomes of different severity, and postoperative management practices may have influenced the results. Finally, Bayesian modeling can improve prediction, but it still requires external validation before being adopted in routine care.
Bottom line
Among children with obesity and severe OSA undergoing tonsillectomy, most did not experience severe perioperative complications. Oxygen saturation nadir on preoperative sleep study was the strongest predictor of risk, and a simple rule using AHI below 25 events per hour together with oxygen nadir above 85% identified a very low-risk subgroup.
The study suggests that some children currently admitted overnight may be candidates for same-day discharge planning, but these findings should be confirmed prospectively before changing standard practice. For now, the main message is that postoperative admission decisions after tonsillectomy may be better guided by individualized risk assessment rather than by obesity and OSA severity alone.
