The Vivos DNA Appliance is now FDA Approved for the treatment of mild-moderate obstructive sleep apnea. It is considered a medical device because it is used to treat a medical condition, so it is subject to regulation. Medical device regulators like the U.S. Food and Drug Administration are tasked with protecting consumers by ensuring the safety and effectiveness of medical devices on the market.

Generally, it would be reasonable for a consumer to assume that having FDA approval for a medical device means that the device is safe and effective. Given that my treatment with the Vivos mRNA Appliance failed and left me with tipped teeth, I gave the approval further scrutiny and found that all is not what it seems.

My Background

Prior to having my career destroyed by depression and debilitating fatigue due to undiagnosed obstructive sleep apnea, I was a successful biomedical engineer specialising in design reviews for medical devices. While I’m located in Australia and my local medical device regulator is the Therapeutic Goods Administration, the U.S. is an important market for my previous employer and clients, and smaller regulators like the TGA often take cues from the FDA, so I’m reasonably familiar with their approval process.

The Medical Device Approval Process

The regulatory approval process for a medical device is typically long and complex, involving proving to the regulator that the device has been designed according to quality management standards such as ISO 13485 to ensure that it is safe, and supplying clinical trial data to prove that it is effective. The FDA has two main programs for approving innovative medical devices that are unlike existing products on the market, called the Premarket Approval process and the De Novo process, which impose slightly different regulatory requirements.

In order to reduce both the regulator’s workload and the regulatory burden on companies marketing medical devices, the FDA also has a program called the Premarket Notification (510k) program for devices that are similar to existing products which are already approved. Under this program, the applicant only has to argue that the medical device they wish to market is substantially equivalent to a “predicate device” that has already been approved. They don’t have to demonstrate that the new device is safe and effective, because the FDA assumes that if it’s substantially equivalent to an existing device that has already been proven to be safe and effective, then the new device should be safe and effective too.

This is like the way generic drug manufacturers can get approval for drugs with identical ingredients to existing approved drugs once their patents expire without having to conduct clinical trials.

The mRNA Appliance Approval

Back in 2014 when Dr Singh was marketing the DNA Appliance under the umbrella of BioModelling Solutions, they were granted approval for the mRNA Appliance via the 510(k) program by arguing that it was substantially equivalent to existing mandibular advancement devices on the market, including the SomnoMed MAS RxA, DynaFlex, and Adjustable Dorsal Intraoral Device.

The 510(k) submission for the mRNA Appliance says: “As with the predicate devices, the mRNA device functions by preventing the retruding of the mandible during sleep to obstruct the airway”, referring to the mandibular advancement function of the mRNA Appliance. The submission glosses over the maxillary and mandibular expansion functions, saying: “The only significant difference is to place expansion screw and beta-titanium 3D-axial springs as required for adjustability and the optional occlusal table”.

This is in stark contrast to U.S. patent 7357635 on the DNA Appliance from 10 years earlier which claims to be able to remodel adult jaws and facial bones. That’s certainly a significant difference compared to other mandibular advancement devices; in fact, it’s the basis of Vivos’s main marketing claim that DNA Appliances can permanently cure obstructive sleep apnea.

It’s not unusual for one regulatory body to ignore the databases of other regulatory bodies when making their decisions. Examiners under time pressure to process approvals can rely on the courts to arbitrate over potential mistakes later in the minority of cases that end up being contested commercially. The FDA bought the argument that the mRNA Appliance is just another mandibular advancement device and granted BioModelling Solutions 510(k) approval to market it.

The FDA wrote the company a warning letter critical of the company’s design control procedures, which noted: “In addition, we reviewed your Daytime Nighttime Appliance (DNA) and mandibular Repositioning Nighttime Appliance (mRNA) devices. We have identified several concerns with these devices as they relate, but not limited, to the following: promotional and advertising claims; intended use and device modifications; 510(k); and registration/listing.”

In November 2019, I went to a talk in Sydney presented by Vivos to recruit dentists into their training program where Dr Singh mentioned that the mRNA Appliance was FDA approved and could permanently cure mild to moderate obstructive sleep apnea, which seemed too good to be true. The audience were led to believe that the FDA had assessed the latter claim, which is not the case: They only approved it based on its mandibular advancement function.

Section 510(k) approval letters from the FDA typically note that the Federal Food, Drug, and Cosmetic Act prohibits misbranding, and that device labelling must be truthful and not misleading. It’s not clear to me if Dr Singh addressed the concerns from the FDA’s warning letter, particularly the promotional and advertising claims. I believe that by conflating the 510(k) approval of the mRNA Appliance with Vivos’s claims that it can expand jaws via epigenetics, Dr Singh is indeed misleading potential customers about what the FDA actually examined in the approval process.

DNA Appliances without the mandibular advancement function were not included in the original 510(k) approval because they aren’t mandibular advancement devices. This is why up until recently Vivos has only been claiming FDA approval for the mRNA Appliance.

The mmRNA Appliance Approval

More recently, Vivos have released the mmRNA Appliance, which is like the mRNA Appliance in that it is a DNA Appliance with mandibular advancement function. The difference is in the way the upper and lower appliances connect to advance the mandible.

The Vivos mmRNA Appliance was also approved under 510(k) by arguing substantial equivalence to the mRNA Appliance, saying that “The device is customized to each patient and features an adjustment mechanism (expansion screw) to allow it to be further customized to each patient.” They haven’t mentioned that the purpose of the expansion screw is actually to expand the jaws via epigenetic bone remodelling like they do in their marketing. That’s not how a mandibular advancement device works. They proceed to state that: “The mechanism of action is mandibular advancement, the same mechanism of action is used in the primary predicate device” (i.e. the mRNA Appliance). Again, this is not what Vivos claim in their subsequent marketing when they say that the appliance is FDA approved.

Under “Device Function”, the mmRNA Appliance submission states:

“The mmRNA appliance is a customized oral device featuring a lower tray and upper tray. These trays put gentle pressure on the tissue at the back of the throat to prevent the airway from collapsing during sleep.”

Mandibular advancement devices, like the predicate devices cited in the submission, do not function by “putting gentle pressure on the tissue at the back of the throat”; they function by advancing the mandible to open the airway behind the tongue.

It goes on to cite Hoekema’s 300-page PhD thesis from 2007 titled Oral appliance therapy in Obstructive Sleep Apnea-Hypopnea syndrome – A clinical study on therapeutic outcomeswhich compares the effectiveness of oral appliances to CPAP for treating obstructive sleep apnea in a number of studies.

Vivos is often criticised for its lack of published research in mainstream sleep medicine journals, and their ability to provide relevant citations isn’t much better: A PhD theses is a poor substitute for a peer-reviewed publication in a major sleep medicine journal. The thesis doesn’t cover expansion appliances like the mRNA, much less the mmRNA which hadn’t been invented in 2007, and at 300 pages I doubt the examiner would have had the time to read it. There are much more focused and recent meta-analyses available in the peer reviewed literature, such as The Comparison of CPAP and Oral Appliances in Treatment of Patients With OSA: A Systematic Review and Meta-analysis, which took me a full 30 seconds to find on Google Scholar.

The FDA approved the mmRNA Appliance 510(k) submission, which isn’t surprising because they had already approved the mRNA Appliance by 510(k) and the mmRNA Appliance really is very similar to the mRNA Appliance predicate device and they both do mandibular advancement.

What A Tangled Web We Weave

It is common for devices approved under the 510(k) program to reference other predicate devices which were also approved via 510(k), leading to a chain (or more accurately, a tree, since you can cite more than one predicate device) of approvals before you arrive at a device that actually had a full Premarket Approval process complete with design controls and a clinical trial to provide safety and effectiveness data. The list of devices in the mRNA Appliance predicate device tree include:

• mRNA Appliance
  • SomnoMed MAS RxA
    • Bird MSDA
      • TAP Anti-Snoring Device
        • H&M Anti-snoring device
        • Distar, Inc – TheraSnore
  • DynaFlex
    • SomnoMed MAS RxA (see above)
    • Acrylic Herbst Splint Appliance
      • Removable Acrylic Herbst
        • Snore Guard
          • Snore Guard Advance
            • Hays & Meade Anti-Snoring Device
          • SleepRight Original
            • SleepRight Low Profile
              • EZ Splint
                • Dr Hays Bite Guard
                • NTI Tension Suppression System
              • Doctor’s NightGuard
                • Dental Concepts BruxGuard
                • Hollywood Products Mouth Peace
                • GEM Scientific Products, Inc Tension Reliever
    • TAP T
      • TAP Anti-Snoring Device (see above)
      • Thornton Oral Appliance
        • TAP Anti-Snoring Device (see above)
        • SnoreFree – OSAP
        • Distar, Inc – TheraSnore
      • TAP II
        • TAP Anti-Snoring Device (see above)
        • Thornton Oral Appliance (see above)
        • OASYS Oral Airway System
          • Dr Keith Thornton – TOA
            • TAP Anti-Snoring Device (see above)
            • SnoreFree – OSAP (see above)
            • Distar, Inc – TheraSnore (see above)
          • Snore TEC
          • TAP Anti-Snoring Device (see above)
  • Adjustable Dorsal Intraoral Device
    • DynaFlex (see above)

The products at the leaf nodes of the tree also all claimed substantial equivalence to an existing product, but I either couldn’t find their database entries, or the summaries are missing so I can’t tell what the product was that originally underwent Premarket Approval that led to all these mandibular advancement devices being considered safe and effective. Because of the nature of the 510(k) approval process, none of manufacturers for the products listed above were required to provide safety or effectiveness data or to conduct clinical trials. That’s just how the process works.

The DNA Appliance Approval

Things get really weird with the 510(k) approval for the DNA Appliance though: It’s summary argues substantial equivalence to the mRNA Appliance, despite the lack of mandibular advancement function which was the whole basis of the mRNA Appliance’s 510(k) approval.

This time around, under “Device Function”, the DNA Appliance submission states: “The DNA appliance is a customized oral device featuring a lower tray, upper tray, or both, depending on patient need. These trays put gentle pressure on the tissue at the back of the throat to prevent the airway from collapsing during sleep.”

This is obviously cut-and-pasted with a little editing from the mmRNA Appliance submission, again using the inaccurate “gentle pressure” line.

The next paragraph is also almost identical, again citing Hoekema’s PhD thesis on oral appliance therapy even though the oral appliances he examined were all “mandibular repositioning appliances”, another name for mandibular advancement devices, which the DNA Appliance is not.

If it isn’t obvious that mandibular advancement devices function by advancing the mandible, Hoekema notes that: “Studies on the outcomes of the variability in mandibular advancement and bite-opening suggest that a mandibular repositioning appliance (MRA) derives its therapeutic effect mainly from the amount of mandibular advancement imposed by the appliance.” (p237)

The DNA Appliance does not advance the mandible at all, so Hoekema’s research is not relevant here.

They then state a claim about the DNA Appliance consistent with Vivos marketing: “The DNA Appliance aims to expand the nasal airway through jaw expansion and mid-facial redevelopment. In doing so, an oral device may be able to permanently improve the oropharyngeal airway. Studies have shown that the DNA appliance can increase nasal cavity volume and reduce the incidence of apnea-hypopnea episodes.”

The submission does not cite any such studies, and this is not how the predicate device was claimed to work, which was by mandibular advancement. Because of the nature of the 510(k) program, there was no requirement for Vivos to provide evidence to the FDA that the DNA Appliance can “permanently improve the oropharyngeal airway.” All they had to do was argue substantial equivalence to the predicate device.

The cynical part of me thinks that Vivos have been gradually shifting the goal posts to get the FDA to accept a 510(k) submission for the DNA Appliance, and cited an irrelevant 300-page PhD thesis to distract the examiner from the fact that the DNA Appliance is not substantially equivalent to the predicate device, nor any of the mandibular advancement devices in its ancestral approval tree since it does not advance the mandible.

The DNA Appliance simply does not work like the predicate devices upon which the mRNA Appliance was approved because it doesn’t advance the mandible.

No wonder the FDA rejected the DNA Appliance application first time around, requesting real world data to prove that the device is effective. This seems unusual for a 510(k) approval process, so I’m not sure what the FDA were thinking except that it seems in line with their new emphasis on real-world evidence.

This posting about the approval on MedtechDive includes this hilariously vacuous analysis, apparently without realising the irony of it: “The DNA device is identical to the already-marketed mandibular [advancement] appliances except the top and bottom trays do not interlock to advance the jaw.”

The FDA agreed with Vivos’s claim that the DNA appliance is substantially equivalent to a mandibular advancement device, even though it does not advance the mandible.

As far as the real-world data goes, with a claimed 27,000 customers, I would have thought Vivos could supply data on more than just 74 cases. There is no evidence that these cases represent a random sample and are obviously not the entire patient population. With a cure rate of only 27% I’d hardly consider the DNA Appliance very effective at treating obstructive sleep apnea.

The introduction to the real-world data supplied by Vivos is also hilarious: “When conducting maxillary expansion for other indications, several dentists noticed their patients also noticed significant improvements in snoring, sleep apnea, and other symptoms of sleep disordered breathing.”

… as if dentists discovered by accident that the DNA Appliance is supposed to improve breathing.

I also haven’t seen any long-term studies on relapse rates with the DNA Appliance, which is likely to be a problem given that muscle tone in the airway continues to degrade as we age. We don’t just want to expand the airway enough to prevent our current apneas, we also need some leeway for the future collapse of muscle tone, or we’ll be back to having OSA again as we age; possibly when we’re too old for maxillomandibular advancement surgery.

Perhaps all this doesn’t matter given that dentists have been supplying the DNA Appliance to patients with obstructive sleep apnea anyway, despite Vivos not having an investigational device exemption for it. This appears to be legal since the appliance is registered with the FDA as a Class I dental device, and dentists can use it off-label to treat obstructive sleep apnea without FDA approval.

Conclusion

I believe that the FDA have made an obvious logical error in approving the DNA Appliance under the 510(k) program, since it clearly isn’t substantially equivalent to the predicate device. If you’re a potential Vivos customer relying on FDA approval of the DNA Appliance as an indication that it is the permanent cure for obstructive sleep apnea that Vivos claims, note that aside from a cursory look at some real-world data, the FDA appears not to have examined the safety and effectiveness of the DNA Appliance during the approval process.

Thank you to Rich Barton for feedback which helped me improve this article.