If you’ve been diagnosed with mild obstructive sleep apnea — typically defined as an apnea-hypopnea index (AHI) of 5-15 events per hour — your sleep physician has probably discussed two main treatment options: continuous positive airway pressure (CPAP) and a mandibular advancement device (MAD). The question of which one to choose isn’t as straightforward as “CPAP is best for everyone,” despite what some sleep clinics still suggest.

The evidence on this comparison has matured considerably over the past five years, and the picture that emerges is more nuanced than either device’s advocates typically present.

How Each Device Works

CPAP delivers pressurised air through a mask worn over the nose, mouth, or both. The air pressure acts as a pneumatic splint, holding the upper airway open and preventing the tissue collapse that causes apneas and hypopneas. It’s effective — when used correctly, CPAP eliminates apneas almost completely.

Mandibular advancement devices (also called mandibular advancement splints or oral appliances) are custom-fitted dental devices worn in the mouth during sleep. They work by holding the lower jaw (mandible) in a slightly forward position, which increases the space behind the tongue and tightens the soft tissue structures of the upper airway, reducing the likelihood of collapse.

The mechanisms are fundamentally different. CPAP applies external force to keep the airway open. MADs reposition internal anatomy to make collapse less likely. Both approaches reduce AHI, but through different physiological pathways.

What the Research Shows

The most comprehensive head-to-head comparison is the SARAH trial, a randomised controlled trial comparing MADs and CPAP in patients with mild-to-moderate OSA. The trial found that CPAP was superior to MADs in reducing AHI — a finding consistent with virtually all comparative studies. CPAP typically reduces AHI to below 5 events per hour (essentially normalising it), while MADs typically reduce AHI by 50-70%, which may or may not bring it below the clinical threshold.

But here’s the critical detail that changes the clinical picture: adherence.

CPAP adherence rates are notoriously poor. Data from Medicare’s compliance requirements show that roughly 50% of CPAP users don’t meet the minimum usage threshold of 4 hours per night for 70% of nights within the first 90 days. Real-world adherence beyond the compliance monitoring window drops further.

MAD adherence is substantially higher. Studies consistently report 70-80% nightly usage rates, with average nightly wear times of 6-7 hours — considerably more than the average CPAP usage of 4-5 hours.

When you calculate “effective AHI reduction” — accounting for the hours the device is actually worn — MADs often match or exceed CPAP for mild OSA patients. A device that’s 80% effective and worn 7 hours per night delivers more therapeutic benefit than a device that’s 95% effective and worn 4 hours per night.

Clinical Outcomes Beyond AHI

AHI reduction is a surrogate measure. What patients actually care about is how they feel during the day and whether their cardiovascular risk is reduced.

On daytime sleepiness, measured by the Epworth Sleepiness Scale, the difference between CPAP and MADs in mild OSA is generally not statistically significant. Both devices improve subjective sleepiness to similar degrees, likely because the marginal AHI difference between the two in mild cases doesn’t translate into a meaningful difference in sleep fragmentation.

On cardiovascular outcomes, the evidence is still developing. The SAVE trial and subsequent studies have questioned whether treating mild-to-moderate OSA with CPAP reduces cardiovascular events at all — a finding that applies equally to MADs. For severe OSA, the cardiovascular benefit of treatment is clearer.

Blood pressure is one area where both devices show measurable benefit. A Cochrane review of oral appliances for OSA found that MADs produced a mean reduction in blood pressure comparable to some antihypertensive medications, with the effect mediated through improved sleep quality and reduced sympathetic activation during sleep.

Who’s a Good Candidate for a MAD?

Not everyone with mild OSA is suitable for a mandibular advancement device. Good candidates typically share several characteristics.

Adequate dental health. MADs require enough healthy teeth to anchor the device — generally at least 8-10 teeth per arch. Patients with extensive dental work, loose teeth, or advanced periodontal disease may not be suitable.

BMI below 35. MADs tend to be less effective in patients with obesity, particularly those with significant fat deposition around the neck. The anatomical repositioning provided by a MAD is less able to overcome the mass effect of excess tissue in these patients.

Predominantly supine-dependent OSA. Patients whose apneas occur mainly when sleeping on their back often respond well to MADs, as the forward jaw positioning counteracts the gravitational effect of supine sleep on tongue and soft palate position.

CPAP intolerance. Patients who have tried CPAP and can’t tolerate it — due to mask discomfort, claustrophobia, aerophagia, or simple non-adherence — are often excellent MAD candidates. The Australasian Sleep Association guidelines recognise MADs as first-line therapy for mild OSA and as second-line for moderate-to-severe OSA when CPAP is not tolerated.

Practical Considerations

Cost. A custom-fitted MAD from a qualified dental sleep medicine practitioner costs $1,500-3,000 in Australia. CPAP machines cost $1,000-3,000 for the unit, plus ongoing costs for masks ($100-250 every 6-12 months), filters, and tubing. Over a five-year period, total costs are roughly comparable.

Fitting process. A properly fitted MAD requires dental impressions or digital scans, jaw relationship records, and typically 2-3 fitting appointments to optimise jaw advancement. The advancement is usually titrated gradually — starting at 50-60% of maximum protrusion and increasing until symptoms resolve or the maximum comfortable advancement is reached.

Side effects. The most common MAD side effects are jaw discomfort (particularly in the first few weeks), excessive salivation, and temporary bite changes in the morning. Long-term use can cause minor dental movement in some patients, which should be monitored with annual dental check-ups.

Follow-up. Both CPAP and MAD therapy require follow-up sleep studies to verify treatment efficacy. A MAD that feels comfortable but isn’t adequately reducing AHI isn’t doing its job. Post-titration polysomnography or home sleep study should be standard practice — though, unfortunately, not all providers insist on it.

The decision between CPAP and a MAD for mild sleep apnea should be a shared one between the patient, their sleep physician, and their dentist. The “best” treatment is the one the patient will actually use consistently. For many people with mild OSA, that turns out to be a well-fitted oral appliance rather than the theoretically superior CPAP machine gathering dust on the nightstand.