Tongue-Stabilising Devices: Real Effect, Brutal Dropout
A randomised crossover trial put a tongue-retaining device against CPAP. It cut AHI from 38.7 to 12.7 and improved quality of life just as much. At thirty months, 15% of users were still wearing it.
A tongue-stabilising device is a soft silicone bulb that sits outside your lips and holds the tip of your tongue forward by gentle suction. Unlike a mandibular advancement device it does not touch your teeth or move your jaw — which makes it the obvious candidate for people with few teeth, crowns, bridges or jaw joint problems.
It is also the device category where the gap between "does it work" and "will you use it" is widest.
Against CPAP, head to head
A randomised crossover trial enrolled 36 patients, randomised them to three weeks of a tongue-retaining device or CPAP, washed out, then switched them to the other. Twenty-seven completed.
| Mean AHI (events/hour) | |
|---|---|
| Baseline | 38.7 ± 24.0 |
| CPAP | 2.5 ± 0.5 |
| Tongue-retaining device | 12.7 ± 2.6 |
The difference between treatments was significant (95% CI of mean differences 4.65–15.62, p<0.001). CPAP won clearly on the sleep study.
But there was no significant difference in Epworth sleepiness or Functional Outcomes of Sleep Questionnaire scores between the two treatments. People felt about the same on either.
That split is worth sitting with. On the number that defines the disease, CPAP is far better. On how patients felt and functioned over three weeks, the two were indistinguishable. The authors' conclusion was measured: CPAP superior for polysomnographic parameters, both similarly improved quality of life and daytime sleepiness, and a TRD "might be considered a short-term alternative."
Note that word — short-term. The next study explains why.
The dropout problem
A long-term follow-up tracked 39 patients at 12 and 30 months.
At 12 months, 35.9% were still using the device. At 30 months, 15.4%.
Six in seven had abandoned it inside two and a half years. Among those still going at 30 months, Epworth scores had improved by an average of 2.0 ± 2.8 points — a real benefit for a small surviving minority. Six patients showed an average increase in blood pressure. The most frequently reported side effects were mouth dryness and excessive salivation.
The crossover trial found the same side-effect pattern over three weeks: drooling, tongue numbness and tongue pain. Those are tolerable for a fortnight and apparently not for years.
Evidence tier 2. The efficacy signal is real and replicated — a roughly two-thirds AHI reduction in the crossover trial. Long-term adherence is poor enough that the realistic expected benefit for any individual starting one is much lower than the efficacy figure suggests.
One safety signal worth knowing
There is a published case report of tongue-stabilising-device-emergent central sleep apnea — central events appearing after starting the device.
A single case report is the weakest form of evidence and it is not a reason to avoid the category. It is a reason for follow-up testing rather than assuming a device that quietens you has fixed you, which is the standing rule for every appliance on this site.
Who it is actually for
People for whom a mandibular advancement device is not an option. This is the real niche: few or no teeth, extensive crowns or bridges, significant periodontal disease, active temporomandibular joint disorder, or very limited jaw protrusion. All of those rule out an appliance that grips teeth and advances the jaw. A TSD grips neither. Who is a poor candidate for a mandibular device.
People trialling the concept cheaply. If holding the tongue forward helps you, that is diagnostic information worth having, and a TSD is a low-commitment way to find out.
Not people with severe apnea who can tolerate CPAP. An AHI of 12.7 against 2.5 is not a close call when the disease is severe.
If you try one
Expect the first fortnight to be unpleasant — drooling and tongue soreness are near-universal early and are the reason most people stop. Knowing that in advance is the difference between pushing through an adjustment period and concluding the thing is broken.
Get retested with it in. This applies to every oral device and matters more here given the central-apnea case report. A device that reduces noise without opening the airway removes your warning signal. How oral appliances should be verified.
Decide at three months, not three nights. And if you are in the 85% who stop, that is a known outcome rather than a personal failure — the question then becomes which of the alternatives fits. CPAP versus an oral appliance, and the devices compared.
And no oral device treats apnea you have not had diagnosed. Witnessed breathing pauses, gasping arousals, waking unrefreshed, or sleepiness more sleep does not fix point at a sleep study first. Snoring versus apnea.