Oral Appliance Therapy; The Definition of Effectiveness

A debate rages. Neither side is willing to listen to the other or objectively evaluate the data. The polarization intensifies and each side is becoming more emboldened and increasingly impassioned in their beliefs. It’s ubiquitous in the U.S., but sadly, also common in Canada and many other developed countries.

Of course, I’m referring to the debate regarding the effectiveness of Oral Appliance Therapy (OAT) compared to Positive Airway Pressure (PAP). Efficacy refers to a therapy’s performance under “ideal” and controlled circumstances. Effectiveness refers to an evaluation of its performance under ‘real-world’ conditions.¹ Research data tells us what efficacy is. As clinicians, effectiveness is what we strive to optimize for every patient we treat; first by meeting the published minimum efficacy outcomes and then by attempting to further ameliorate treatment outcomes via modifications for specific patients based on their particular presentations.

A recent systematic review and meta-analysis regarding the effectiveness of PAP and OAT in the treatment of Obstructive Sleep Apnea (OSA) clearly demonstrated that PAP is far superior to OAT in the reduction of Apnea Hypopnea Index (AHI).² AHI is the metric currently used to establish sleep apnea severity, and physicians typically refer to PAP criteria of normalizing the AHI to below 5 to establish a successful outcome. This is problematic because while more than 1/3 of OAT cases will meet this criterion, another 1/3 will experience a clinically important reduction of >50% but remain above 5 AHI, and the final ~1/3 will not achieve this >50% reduction.³ These findings further bolster the position that PAP is Gold Standard Therapy for Sleep Apnea.

In a recent paper describing effective Oral Appliance Therapy (OAT), the American Academy of Dental Sleep Medicine (AADSM) states,⁴ “A properly fitted oral appliance worn nightly will decrease the frequency and/or duration of apneas, hypopneas, respiratory effort related arousals (RERAs) and/or snoring events. Oral appliances have been demonstrated to improve nocturnal oxygenation as well as the adverse health and social consequences of OSA and snoring.” Note that the expectation is to “decrease” and “improve.” There is no mention of normalizing AHI.

However, there is a very large elephant in the room wearing a gigantic sign that says, “I can’t wear PAP!” This elephant is routinely ignored by physicians with very sound reason; physicians are accustomed to patients not complying with prescriptions. For example, approximately 50% of hypertension medications are taken as prescribed,⁵ as few as 12% of asthma sufferers take their asthma meds as prescribed,⁶ and the National Association of Chain Drug Stores documents that only 25-30% of patients actually fill and take medications as prescribed.⁷ In contrast, the 2015 Frost and Sullivan report⁸ documents that 60% of Americans diagnosed with Sleep Apnea remain compliant with PAP; suggesting that sleep medicine is faring very well regarding PAP compliance.

From a physician’s perspective, medicine is managing approximately 2/3 of the patients; 1/3 optimally with all-night wear and 1/3 sub-optimally with partial night wear. Viewed from a different vantage point, medicine is mismanaging 2/3 of the patients; 1/3 are sub-optimally managed, wearing PAP only part of the night, and 1/3 are unable to tolerate PAP therapy at all (Figure 1).

PAP and OAT Effectiveness – Adherence is Key

A recent comprehensive, systematic literature review involving 82 papers, spanning 1994-2015 found that, “CPAP adherence remains persistently low over twenty years’ worth of reported data”⁹ It is interesting that the most current PAP guidelines published in 2019¹⁰ make no mention of these findings and simply continue to refer to PAP as gold standard therapy. The guidance provided for patients having difficulty with PAP remains unchanged in the latest PAP guidelines; try a different mask interface, reduce the pressure, add humidification, or try AutoPAP or BiPAP. This does not foster much hope for improvement in PAP adherence. Alternative therapies are referenced, and OAT is briefly mentioned, bundled into a paragraph on alternative therapies. There is no mention that OAT is currently indicated as a “Standard of Care” for any severity of OSA if the patient cannot tolerate PAP or prefers an oral appliance to PAP. Without any guidance for the sleep physicians, how can one expect them to know how, when, and for whom OAT is appropriate? In contrast, the current OAT guidelines explicitly state that PAP is the “Gold Standard” and first line therapy for OSA, providing clear guidance to dentists about the use of PAP.¹¹

The aforementioned meta-analysis demonstrating superiority of PAP for normalizing AHI also documented the clear superiority of OAT regarding treatment adherence.² Patient preference heavily favoring OAT was one rationale for the 2015 OAT guidelines indicating OAT be a “Standard of Care” for all severities of OSA.¹¹ Other studies also document higher patient preference and adherence for OAT. A study comparing a titrated oral appliance with PAP found that side effects existed with both therapies, that they were of similar intensity and frequency, and that OAT was preferred by >70% of the patients.¹² Another recent cross-over study comparing OAT and PAP (both utilized adherence tracking) found that patients preferred OAT to PAP in all variables investigated, with the average rating across all variables favoring OAT (80.3%) vs. PAP (19.7%).¹³ A number of other investigators have similarly concluded that patients prefer OAT to PAP.^14,15,16,17

Over the past decade a number of publications have documented the similarity of PAP and OAT outcomes. Objectively measured health outcomes such as psychomotor speed, driving simulator performance, cardiovascular system outcomes and mortality have been demonstrated to be similar for both PAP and OAT.¹⁸ Functional outcomes such as snoring, sleepiness, neurocognitive function, quality of life, and mood have also been found to be similar with both therapies. PAP has been shown to be slightly better at snoring reduction while OAT scores slightly higher in the improved Quality-of-Life category.¹³ It is important to note that blood pressure outcomes¹⁸ and measurements of cardiovascular mortality¹⁹ have both been demonstrated to be similar with PAP and OAT.

How does one reconcile that PAP is superior to OAT in normalizing AHI, yet health outcomes are similar? Let’s circle back to the elephant in the room that too many choose to ignore; poor patient adherence associated with PAP. Let’s start by refining our understanding of PAP adherence, which when using a commonly cited benchmark of 4 hours use/night, 5 nights/week, is approximately 50% at 6 months and 17% at 5 years.^20,21,22,23 A recent publication documented that the likely PAP adherence profile for a particular patient is strongly associated with their initial experience with it.  This study, which evaluated predictors of PAP adherence found, “the strongest predictors of adherence at 24 months were early adherence measures. Adherence in the first month was the most predictive of adherence at 24 months.”²⁰ Patients that do adhere to PAP, actually wear it varying number of hours nightly – 4 hours is deemed necessary to demonstrate adherence.

PAP adherent patients often do not demonstrate superior outcomes to OAT patients, even when the OAT does not fully normalize AHI. This may be explained by the “Dose Dependent Effect.” The literature documents that PAP health benefits are related to usage; for instance, 6 hours nightly for objective sleepiness to be resolved, greater than 5.6 hours nightly for objective hypertension to be resolved and more than 6 hours nightly is associated with the greatest mortality risk reduction.²³ Navarro-Soriano et al. recently documented that both cardiovascular events and development of hypertension had a dose dependent relationship with hours of PAP use;²⁴ patients using PAP < 4 hours per night were 5.1 times more likely to develop severe hypertension. A dose dependent effect for PAP use on sleepiness has also been shown.²⁵ When McEvoy et al.²⁶ evaluated the incidence of cardiovascular events with PAP wear compared to controls, they found no difference. On the other hand, Anandam et al.¹⁹ demonstrated a reduction in events over controls; patients wore PAP an average of 3.3 hours in the McEvoy study vs. 5.8 hours in the Anandam study, once again, suggestive of a dose dependent effect.

In contrast, OAT adherence is very high, documented to be approximately 90% measured subjectively,²⁷ and 85.9% measured objectively.²⁸ When measured objectively, nightly use has been shown to be approximately 7.5 hours.²⁸ A long-term follow-up study documenting use after 5.7 +/- 3.5 years found that, among adherent patients, 93.7% used the oral appliance more than 4 nights per week, 100% wore it more than half of each night, and 95% were satisfied with the treatment.²⁹

In 2013, Vanderveken et al³⁰ documented the safe use of an intra-oral OAT objective adherence monitor. This enabled for the first time, the calculation of OAT Mean Disease Alleviation (MDA), defined as a combined function of efficacy and adherence, and representing a measure of the overall therapeutic effectiveness. Over the following 2 years, this concept evolved further^23,31 resulting in the development of the SARAH Index, (Sleep Adjusted Residual AHI), defined as [(AHI treated X Hours treated) + (AHI untreated X Hours untreated)] divided by Total Sleep Time (See Figure 2).

When either the MDA or Sarah Index are calculated, OAT is found to be similar to PAP, with overall effectiveness hovering around 50% for both.^23,30 This is thought to be the reason for the similar health benefits OAT and PAP deliver. The common medical consensus has been that high PAP efficacy would transfer into “real-life” circumstances, and that PAP effectiveness would equal its efficacy. PAP has fallen short of this mark and demonstrates an effectiveness similar to OAT in “real-life” circumstances (See Figure 3).

The Real-World Effectiveness Equation

SomnoMed, a leading OAT manufacturer, has taken these concepts and created a software algorithm that can be used to determine the effectiveness of PAP and OAT for a particular patient. It takes into consideration baseline AHI, post-therapeutic PAP and OAT AHI, and the length of use for both therapies. The results are quite surprising, even when one is already aware of the MDA concept.

Case I (See Figure 4) involves a patient with a baseline AHI of 99. Post-therapeutic PAP AHI was 3.9, but OAT fell short with an AHI of 27.9. In most offices, this oral appliance effort would be dismissed as ineffective and the patient would be encouraged to continue with PAP. This particular patient could tolerate PAP for 5.5 hours per night and would likely be told they are adherent, as the minimum expectation is 4 hours, 5 nights/week. Any attempts on the patient’s part of expressing the opinion that their OA provides them a similar benefit to their PAP would be dismissed due to the OA’s inability to reduce the AHI below 5. The patient would be told that their oral appliance is ineffective.

The SomnoMed Effectiveness Equation (Somnomed, Plano, TX) demonstrates the total number of events on PAP was 268 and on OAT was 258. So, the number of arousals from sleep, and the number of desaturations occurring each night were similar and in fact, marginally less with OAT. Once again, this helps to explain why health outcomes of both therapies are comparable and why patients often claim that they feel similarly with both therapies even though they have residual apnea with the oral appliance.

Case II (see Figure 5) involves a patient with baseline AHI of 27.6. Post-therapeutic PAP AHI was 4.5, but OAT fell short with an AHI of 12.5. Once again, in most offices, this appliance effort would be dismissed as ineffective and the patient would be encouraged to continue with PAP. In this particular case, the patient could tolerate PAP for 5 hours per night, surpassing minimum expectations and thus considered adherent. Once again, any attempts on the patient’s part of expressing the opinion that their OA provides them a similar benefit to their PAP would be dismissed.

What does the SomnoMed Effectiveness Equation tell us? The total number of events on PAP was 105 and on OAT was 107. The number of arousals from sleep, and the number of desaturations occurring each night were similar with both therapies. Nevertheless, this patient would be encouraged to stay with PAP and discouraged from wearing their oral appliance.

We must keep in mind that the 1/3 of patients that happily wear PAP throughout their sleep period are being optimally managed. It is the next 1/3 that wear PAP only part of the night and are sub-optimally managed and the last 1/3 that cannot tolerate wearing PAP at all that can benefit greatly from OAT. Patient preference is clearly related to adherence, and adherence is clearly related to outcomes. However, there are also potential confounders to the MDA and SARAH Index approach. These include what we don’t know regarding continued therapeutic effectiveness once PAP and or OAT are removed each night, whether routinely not wearing PAP later in sleep period holds implications due to being exposed to more events during REM sleep and the lack of any accounting for what position and in which sleep stage the events are occurring.

The Achilles heel of PAP therapy is adherence, and the current PAP guidelines provide guidance regarding the management of patients that have difficulties with PAP adherence.¹⁰ The Achilles heel for OAT is residual AHI. Fortunately, there is an abundance of evidence-based guidance available to offer these patients as well (See Figure 6).

When OAT Outcomes Fall Short

The MDA concept is not a pass to accept a sub-optimal outcome. When OAT falls short of the mark in normalizing AHI, then the first thing to do is ensure the appliance is optimally calibrated. For some patients this may mean further advancement, and it is important to remember that the patient’s ability to tolerate a certain level of advancement is an adaptive capacity that generally increases with time. So, what is uncomfortable today, will not necessarily be uncomfortable tomorrow. Next, consider altering vertical if the appliance is adaptable. To avoid making changes to vertical after the fact, consider building your original appliances using a vertical based on phenotype developed by Levendowski that was created when he developed the ApneaGuard.³² Generally, this involves using more Vertical Opening Dimension (VOD) for heavier males and less VOD for petite females. Finally, if the appliance does not restrain the lower component from dropping open in supine sleep, apply elastics and determine if preventing the mandible from dropping during supine sleep helps to normalize any residual supine related apnea. Milano et al. demonstrated an improvement in outcomes with elastic use for patients with supine related AHI.³³ See Figure 7.

Adjunctive Therapies to Optimize Outcomes

For those patients continuing to experience inadequate outcomes, even after the oral appliance is optimally calibrated, there are a number of adjunctive therapies that can further improve outcomes. Often referred to as combination or hybrid therapy, combining PAP and OAT has been shown to increase PAP compliance³⁴ and improve overall outcomes.³⁵ When a patient has access to both therapies, it provides them options; they can wear both all night long, they can start with both and take their PAP off when they are done with it and continue with their OAT, they can wear their PAP as long as possible, and when they remove it they can switch to their oral appliance that is kept on their night stand, and finally, they can start with their oral appliance and move to their PAP part way through the night. The patient can experiment and find what works for them. See Figure 8.

For those patients carrying excess weight, weight reduction can help the cause dramatically. A 10% loss in weight has been associated with a 26% decrease in AHI.³⁶ It is also important to note that aerobic exercise results in a, “reduction in disease severity and in daytime sleepiness, as well as an increase in sleep efficiency and in peak oxygen consumption, regardless of weight loss.”³⁷

For patients experiencing residual supine AHI, all that may be required is to add positional therapy to the regimen; adjunctive use of positional therapy has been demonstrated to improve both tongue-retaining device³⁸ and OAT³⁹ outcomes. Positional therapy combined with OAT have actually been demonstrated to be as effective as PAP for positional OSA.⁴⁰ A number of positional aides are currently available to help facilitate positional therapy; for example, vibratory devices that vibrate when the patient is in supine position, encouraging a change to non-supine position and strap-on devices that make supine sleep uncomfortable thus encouraging non-supine sleep. Further on position, wedges that elevate the head position by as little as 15 cm have been demonstrated to reduce AHI in both supine and non-supine position.⁴¹ See Figure 9.

Nasal patency is another area to investigate. Higher nasal resistance has been demonstrated to be predictive for OAT outcomes.⁴² Once again, a number of over-the-counter aides are available to improve nasal patency making it simple and inexpensive to evaluate the potential for benefit.

Currently, palatal surgery is rarely considered primary therapy for OSA, however, combining palatal surgery with OAT has resulted in favorable outcomes.⁴³

A systematic review and meta-analysis of the Myofunctional therapy literature demonstrated that it reduces AHI by approximately 50% and also decreases oxygen desaturations, snoring and sleepiness in adults. An even larger AHI decrease of 62% has been demonstrated for children.⁴⁴ Used adjunctively, Myofunctional therapy could help resolve residual AHI. Further along these lines, a novel technology has been introduced recently that utilizes neuromuscular stimulation of the genioglossus muscle during day hours to improved muscle physiology, demonstrated to be effective for mild OSA⁴⁵ it is currently being investigated for its clinical utility for moderate to severe OSA. And finally, some additional adjunctive efforts include improving sleep hygiene,⁴⁶ and a variety of lifestyle modifications such as reducing smoking⁴⁷ and alcohol consumption⁴⁸ and reducing use of central nervous system depressants.⁴⁹

Once all of the above have been exhausted, you can confidently accept that both your oral appliance and your oral appliance treatment effort have been optimized. In this case, should the result be unacceptable, there are a number of surgical options available, such as Maxillary-Mandibular Advancement surgery⁵⁰ and implanted tongue neuromuscular stimulation devices.⁵¹

The continued use of AHI as a metric to evaluate OSA severity is currently being challenged,^52,53,54,55 with the literature suggesting a “reconsideration of the role of AHI as the prime diagnostic metric.”⁵² It is interesting that shortly before coining the term “Apnea Index”,⁵⁶ the authors preferred to report the total number, or duration, of nocturnal apneas,^57,58 rather than the number per hour of sleep. Based on what the literature currently documents regarding adherence, health outcomes, and MDA, had this alternative definition been established in 1978, we would have had a different perspective on the therapeutic effectiveness of OAT all these years.

In summary, OAT effectiveness has been demonstrated to be similar to PAP when viewed through the lens of MDA; a conclusion, supported by an abundance of literature evidence. Thus, rather than focusing on the AHI exclusively, we have discussed the importance of also considering adherence levels when comparing PAP and OAT outcomes. All stakeholders – physicians, dentists, technologists, and auxiliaries should be familiar with these concepts and they should be discussed openly when evaluating and comparing outcomes achieved by PAP and OAT. Our patients can be best served when we communicate accurately and consistently regarding their care. Finally, OAT providers should be careful to ensure that every effort has been made to optimize treatment outcomes by first optimizing the actual oral appliance, and then utilizing patient appropriate adjunctive therapies in an effort to further improve outcomes. Patients can be best helped by physicians and dentists working collaboratively to provide the best solution for that particular patient.