News · Sleep
Tonsil surgery helped snoring children, but not predictably
A SLEEP analysis of 363 children in a randomized trial found 65% improved after early tonsil surgery versus 39% who waited, yet almost nothing measured beforehand predicted who would benefit.
Based on a peer-reviewed secondary analysis of a randomized trial in SLEEP
- This is a secondary analysis of the Pediatric Adenotonsillectomy Trial for Snoring, published in the journal SLEEP.
- 363 children aged 3 to 12 with mild sleep-disordered breathing, meaning snoring with little or no obstructive sleep apnea, were randomized to early surgery or watchful waiting with supportive care.
- Benefit was defined strictly: clinically meaningful improvement in at least 2 of 4 domains covering behavior, quality of life, daytime sleepiness and blood pressure.
- 65% of the surgery group hit that bar versus 39% of those who waited.
- Of every baseline characteristic tested, including sleep-study data, only a high Mallampati score predicted who would benefit (OR, 4.16; 95% CI, 1.22-15.6).
- The Mallampati score is a simple look at how much of the throat is visible when a child opens their mouth.
- 39% of children who had no surgery improved anyway, which is why the decision is genuinely difficult.
Tonsil removal is one of the most common operations performed on children, and one of the least settled. For a child who merely snores, without real apnea, the honest answer has long been that nobody knows. A study in the journal SLEEP narrows that down, and then widens it again.
Surgery helped: clinically-meaningful positive outcomes occur more often after early adenotonsillectomy versus watchful waiting for children with mild sleep-disordered breathing. Working out in advance which child would benefit proved almost impossible.
The question left over from the trial
The parent study, the Pediatric Adenotonsillectomy Trial for Snoring, had already shown group-level benefit across several domains. What it had not answered is the question a parent in a consulting room actually asks: will it help my child?
So this analysis went looking for predictors, aiming to identify baseline demographic, symptom, physical exam, historical, or polysomnographic features that may predict such benefit.
A deliberately strict definition of “better”
Rather than count any improvement, the researchers set a composite bar. Meaningful benefit was defined as a composite outcome of clinically-significant change in at least 2 of 4 domains: behavior, quality of life, subjective sleepiness, and blood pressure.
Each of those four had shown greater improvement 12 months after surgery than watchful waiting in the original trial. Requiring two at once is a harder test than most surgical studies apply.
The analysis covered participants aged 3.0-12.9 years in the 12-month multi-center randomized clinical trial. Among 363 participants, the mean age was 6.6 years, and 185, or 51%, were male.
The result, and the gap in it
A total of 115, or 65%, in the surgery arm experienced clinically-meaningful improvement on 2 or more domains, versus 73, or 39%, in the watchful waiting arm.
That is a real and clinically useful difference. It is also a reminder that 39% of children got meaningfully better with supportive care alone, which is why the operation is not automatic.
Then the search for predictors came up almost empty. Among variables explored for effect modification, including demographics, health factors, symptoms and signs, and polysomnographic data, only a high Mallampati score of III or IV versus I or II showed predictive value.
The cheapest test won
The surviving predictor is worth dwelling on, because it costs nothing. The Mallampati score is what you get by asking a child to open their mouth and stick out their tongue, then grading how much of the throat you can see.
A crowded view scored high, and that grade carried roughly four times the odds of meaningful benefit.
Meanwhile, the overnight sleep study, the expensive and disruptive test that currently anchors this decision, predicted nothing in this group. For mild cases, the wide-open-mouth look outperformed the laboratory.
Where the uncertainty remains
This is a secondary analysis, meaning the predictor hunt was not what the trial was designed and powered to do. Subgroup findings from a secondary analysis are the least reliable class of trial result, and the Mallampati finding carries a very wide range of uncertainty, running from just above 1 to over 15. It needs replication before it changes practice.
Follow-up was also 12 months, which cannot address whether benefits or harms persist, and the study covers mild cases only.
The authors state the limitation themselves: many clinical features accessible prior to surgery, with the intriguing exception of a high Mallampati score, may not help to identify which children stand to benefit.
Which leaves parents roughly where they were, but better informed about the odds: two in three helped by surgery, two in five helped by waiting, and one cheap look at the back of the throat that might tilt the decision.
People also ask
What is mild sleep-disordered breathing?
It means habitual snoring with little or no measurable obstructive sleep apnea on a sleep study. It sits in an awkward middle ground: severe apnea has a clear surgical case, and occasional snoring has none. Children in this category are common, and until recently there was little trial evidence on whether removing tonsils and adenoids helped them.
How much better did surgery do?
115 of the surgery group, or 65%, achieved clinically meaningful improvement in at least two of the four domains, against 73, or 39%, in the watchful waiting group. That is a substantial difference. It also means roughly four in ten children improved without any operation, and roughly a third who had surgery did not clear the bar.
What is a Mallampati score?
A quick clinical look at the back of the throat when someone opens their mouth and sticks out their tongue, graded I to IV by how much of the soft palate and throat is visible. Higher grades mean a more crowded airway. Anesthetists have used it for decades to anticipate difficult intubation. Here, a score of III or IV was the only baseline feature that predicted who benefited from surgery.
Why could sleep studies not predict who would benefit?
That is the study's uncomfortable finding. Among variables explored, including demographics, health factors, symptoms and signs, and polysomnographic data, only the Mallampati score showed predictive value. Overnight sleep studies are expensive and burdensome for children, and in this mild group they did not identify who stood to gain. That challenges how these children are currently assessed.
Should my child have their tonsils out?
This is general information and not advice about your child; that decision belongs with an ENT specialist or pediatrician who can examine them. What this analysis adds is that meaningful benefit is more likely with surgery than without, that a substantial minority improve regardless, and that a simple throat examination may be more informative than a sleep study for this particular question.