Reviewed for medical accuracy by Dr Evelyne Jonniaux, Medical Advisor
Last updated 17 September 2026
Grommets improve hearing by 12 dB at 3 months, 4 dB at 6 to 9 months, and by nothing at all at 12 and 18 months. On language development and school learning, no difference has ever been found, including in a cohort followed to age 11. The question is not whether grommets work, but on what.
The questions we get asked
How much hearing do grommets actually restore?
The 2010 Cochrane review pooled 10 trials and 1,728 participants. In trials randomising children, the mean hearing gain against control is 12 dB at 3 months, 4 dB between 6 and 9 months, and no difference at 12 and then 18 months. The authors are explicit about why: natural resolution also improves hearing in the children who were not operated on. The gap closes because the control group catches up, not because the tubes stop working.
Do grommets protect language development?
This is the most widespread belief and the one that holds up least well. The Pittsburgh series followed 6,350 infants and randomised 429 of them to immediate or delayed insertion. At 3 years, no difference in vocabulary or cognitive score. At 6 years, no difference across 30 measures. At 9 to 11 years, no difference across 48 measures, including passage comprehension, spelling and calculation.
One important caveat: none of these trials included children who already had a language or learning disorder. Nothing can be concluded for that subgroup, and the Cochrane review says so explicitly.
What about tonsillectomy for recurrent throat infection?
In children, the 2014 Cochrane review reports 3.0 episodes of sore throat in the first postoperative year against 3.6 in controls. One of the three episodes in the surgical group is the predictable postoperative one. In less severely affected children the result reverses: 1.2 moderate or severe episodes in operated children against 0.4 in controls.
The 2019 US guideline sets a numeric threshold: watchful waiting below 7 episodes in the past year, 5 per year over two years, or 3 per year over three years.
And for paediatric sleep apnoea?
The CHAT trial randomised 464 children aged 5 to 9. On the primary endpoint, attention and executive function, the difference is not significant. On secondary endpoints, behaviour, quality of life and polysomnography, surgery does better. Worth noting as well: 46 percent of the children who were not operated on normalised spontaneously within 7 months.
What the figures say
| Finding | Figure | Source |
|---|---|---|
| Hearing gain from grommets at 3 months | 12 dB | Cochrane, 2010 |
| Hearing gain at 6 to 9 months, then at 12 months | 4 dB, then no difference | Cochrane, 2010 |
| Language and cognition difference at 9 to 11 years | none across 48 measures | Paradise et al., N Engl J Med, 2007 |
| Tympanosclerosis after grommets | about one third of ears | Cochrane, 2010 |
| Haemorrhage after tonsillectomy | 3.5 percent | National Prospective Tonsillectomy Audit, 2007 |
| Tonsillectomy mortality | 7 per 100,000 (US) and 1 in 41,263 (Sweden) | Edmonson, JAMA, 2022 and Ostvoll, 2015 |
One figure we could not source: the widely repeated tonsillectomy mortality of “1 in 10,000” traces back to no identifiable primary publication. The two registries named above are the only verifiable data we found.
What Noria Health does, and does not do
We organise a paediatric ENT consultation with a specialist team near Brussels, audiometry and tympanometry included, and confirm feasibility within 24 to 48 hours. Follow-up is documented at D14, D30 and D90.
We do not set a surgical indication remotely, we do not contradict the team following the child, and we do not recommend postponing an operation already scheduled.
Who reviewed this page
Noria Health editorial team.
See also: all our articles on paediatrics.
Have your case reviewed. Noria Health arranges a specialist consultation near Brussels and confirms feasibility within 24 to 48 hours. Open a case.