Response to Comments: Hypoglossal Nerve Stimulation for Obstructive Sleep Apnea

An evaluation of anatomy is an integral part of the otolaryngologists assessment but is not a contraindication to hypoglossal nerve stimulation. Moreover, there is no current evidence that suggested limitations like 3+ tonsils are a contraindication to effective treatment with the device. We recommend that this section be removed.

Thank you for your comment. Pronounced anatomical abnormalities were used as exclusion criteria on the STAR trial. Tonsil size of 3 or 4 (tonsils visible beyond the pillars or extending midline) was presented as such an example. No peer review data was submitted to validate this claim and was not recommended by any of our Subject Matter Experts (SMEs) during our evidentiary CAC. The reconsideration process is always open for appropriate data to be presented after the LCD is finalized.

We would agree that the documentation of a sleep study (within 24 months of the initial consultation as well as documentation of PAP failure or intolerance (as proposed) are reasonable to establish candidacy and preserve patient access. However, since many patients are self-referred to an implanting otolaryngologist, often after years without treatment, we feel that documentation of these items in the records of the implanting physician should be sufficient without a requirement for a direct referred from a sleep medicine physician. The requirement that there has to be a formal referral from a sleep medicine physician would only add additional cost and time to the process.

We respectfully disagree. The documentation that the patient is CPAP intolerant is an obvious point that should be in the implanting physician’s documentation. If he just references the referring sleep physician’s referral and progress notes that contain that documentation will be considered adequate without having to restate this, but the information shall be in the implanting physician’s chart.

In addition, the decision to undergo an elective surgery always requires a conversation of the risks, benefits, alternatives and complications that can result from the surgery. We recommend that this should be documented in the medical record of the implanting otolaryngologist in lieu of a requirement for documentation of shared decision making by multiple providers. Sleep medicine specialists will appropriately remain integral to this process as implanting surgeons who are not board certified or board eligible sleep medicine specialists themselves will need to refer these patients to a sleep physician for device titration and management.

The shared decision-making requirement has been clarified. The requirement intends to require counseling of Beneficiaries by physicians with expertise and training in sleep medicine. This may or may not involve multiple providers.

…With regards to Limitations, the language is far too vague regarding several terms including:

• Neuromuscular disease which can include conditions such as foot drop or sciatica which are not contraindications to HGN stimulation. Similarly, active psychiatric diseases such as depression and anxiety may actually be improved with control of obstructive sleep apnea and should not be considered limitations.
• Uncontrolled hypertension is also a condition that may be improved with control of obstructive sleep apnea and while it will need to be managed to optimize safety around the time of surgery, should not be a contraindication.
• There is no data suggesting that this device is unsafe in patients who plan to become pregnant and we recommend that this be removed.
• In addition, there are individuals with decreased neurologic tone (i.e. those with Down syndrome) or increased neurologic tone (cerebral palsy) who can benefit greatly from this device, but could be excluded with the limitation to those with any condition or procedure that has compromised neurological control of the upper airway and thus we recommend that this be clarified or excluded.

Thank you for the comments. The mental illness limitation has been clarified to indicate active serious mental illness that reduces the ability to carry out Activities of Daily Living (ADLs) and would interfere with the patient’s ability to operate the HNS and report problems to the attending provider. Palmetto expects that additional literature regarding the safety and efficacy of this service will be forthcoming. Until additional literature is received, no further changes will be made to the policy as the language is consistent with the STAR criteria and the FDA approval.

Endoscopy, both during sleep and wake, is a standard component of an otolaryngologist’s training and should not require further certification by a third party including an FDA approved manufacturer. We recommend that this section be removed completely.

Thank you for the comment. No peer reviewed articles were submitted with this request. There was a consensus among the subject matter experts (SMEs) that the second party review is appropriate, and the 15 procedures was a conservative but good number, allowing for greater success with implantation. Should data be forthcoming and demonstrates that similar results are being obtained on the first 5-10, it may be submitted through the reconsideration process.

The insertion of the device should be carried out by a board certified or board eligible otolaryngologist to parallel the language for the sleep medicine physician throughout the document. Young attendings are not able to be board certified until they have been in practice for almost a year and this would restrict access to some of the most recently trained implanting otolaryngologists. In addition, fellowship is not required to perform hypoglossal nerve implantation.

While at the beginning of the introduction of this technology, cadaver-based courses have been carried out to train new implanting surgeons; the integration of this into otolaryngology training now allows residents to learn this procedure in real-time on live patients as they learn all the other surgeries that they perform. There should not be a requirement that cadaver courses need to be carried out when residency programs can certify competency for this procedure for their graduating residents as they do for all operative privileges.

As implanting otolaryngologists and sleep surgeons, we are trained regarding the evaluation of patients with sleep apnea and the criteria that would deem a patient a candidate for the hypoglossal nerve stimulator. In addition, the diagnosis and management of obstructive sleep apnea is an integral part of otolaryngology training and our board examination as it is in many other specialties. In light of this, either an otolaryngologist or sleep physician should be able to evaluate and refer a patient to an implanting physician. In addition, the ideal relationship for the patient would be a collaborative one between an otolaryngologist and the sleep physician. This is important as some of the post-insertion programming such as activation and initially adjusting the implant may be best carried out in the otolaryngologists office at the same time as postoperative care is being carried out. In addition, patients with suboptimal outcomes with titration may need awake endoscopy with adjustment of the implant configuration in order to maximize their results. The language would more accurately reflect this relationship if it said that evaluation, referral and post implant evaluation of the hypoglossal nerve stimulator, but not including expected post-op care by the inserting physician, should be performed by a board eligible or certified sleep physician or otolaryngologist.

Thank you for the comment. There is evidence to support the training and there was no additional evidence presented to the contrary. There is no evidence that DISE training is part of a general training program for Otolaryngologists. SME’s, including otolaryngologists, agreed with the Company approved training.

As a taxpayer and a sleep apnea treatment provider, I am recommending that the criteria for receiving Medicare benefits for hypoglossal nerve stimulation surgery include the failure of CPAP and Oral Appliance Therapy with a Mandibular Advancement Device. Both of these therapies have a great deal of studies showing greater efficacy (a return to normal breathing of an AHI<5) at a significantly initial and lifetime reduced cost. In addition, because they are both non-surgical therapies they are associated with fewer treatment morbidities and mortalities.

We agree. Sleep specialists may determine the appropriate alternative treatments for OSA in the event CPAP is not a viable option for their patient. This policy is intended to establish the requirements for coverage of the HSN device. You may submit a request for a new LCD for non-surgical therapies for OSA, if you so choose, to address these types of interventions.

I am writing to you on behalf of the American Academy of Dental Sleep Medicine (AADSM) to request that the proposed LCD be revised so that patients are required to try and fail oral appliance therapy (OAT), in addition to continuous positive airway pressure (CPAP), prior to Hypoglossal Nerve Stimulator (HSN) surgery for obstructive sleep apnea (OSA) treatment.

OAT is an effective OSA treatment which reduces apnea-hypopnea index (AHI) and respiratory disturbance index in patients. Studies have demonstrated that OAT and CPAP are similarly effective treatments, even when patients have moderate-to-severe OSA.

OAT is a much more affordable therapy for OSA. HSN typically costs $30,000 - $40,000 to place the device, including surgery expenses. Battery replacement for the device can cost an additional $17,000, bringing the total cost to approximately $47,000 - $57,000.

The proposed LCD requires patients to try CPAP and will cover HSN surgery if CPAP fails to be an effective treatment. The AADSM supports this provision but recommends that patients are required to try and fail OAT, in addition to CPAP, before providing coverage for HSN surgery. Including OAT increases patients’ choice of treatment options and encourages them to start with a more cost-effective option. Requiring patients to start with an effective treatment that is affordable benefits both patients and taxpayers as it reduces the overall cost of care.

We disagree. This policy is intended to establish the requirements for coverage of the HSN device when the gold standard of treatment, CPAP, is not a viable option for their patient. You may submit a request for a new LCD for non-surgical therapies for OSA, if you so choose, to address these types of interventions.

LCD Language of concern: The inserting provider shall be certified by the FDA-approved manufacturer’s second opinion service of validation via video clip submissions of at least 80% agreement in at least 15 consecutive studies. Documentation of this proficiency must be available to submit for contractor review.

Proposed LCD Language:

The inserting provider shall be certified by the FDA-approved manufacturer to perform DISE. This certification shall include a written/video based test where 10 pre-recorded DISE video clips are assessed for patient candidacy, and the submission of 5 DISE performed by the inserting provider and/or the clinician in the inserting provider’s organization who routinely performs DISE, to the FDA approved manufacturer’s physician Training & Certification team for agreement on patient candidacy for HSN implant (i.e., lack of full concentric collapse of the upper airway). Those physicians who were certified by the FDA-approved manufacturer prior to the date of this LCD becoming ‘Final/Active’ shall be grandfathered, and will not need to repeat DISE training and certification. 

Rationale:

[We] understand the importance of ensuring that HSN systems are implanted only in anatomically qualified patient candidates, and that the best current predictor of candidacy is Drug Induced Sleep Endoscopy (DISE) results that show an absence of Complete Concentric Collapse (CCC) of the upper airway. We agree with [the] objective of ensuring that every implanting physician be fully trained and certified to perform and read DISE, such that the patient has a clinical probability of therapeutic success based on the upper airway anatomy. 

In order to support the MACs desire to enhance DISE training, and have robust documentation of each physician’s DISE certification in an FDA monitored Quality System, we recommend a modification to the criteria laid out in the draft LCD. We agree that second party review of 15 DISE is reasonable, but suggest that the first 10 be incorporated into our existing Physician Training program, in the third Module (DISE Training) in the form of a test. We propose that 10 pre-recorded DISE video clips, where the manufacturer’s second party review Physician panel align on the correct interpretation of whether CCC is present or not, be used in the form of a documented test, even before the physician proceeds with the cadaver training module. We then suggest that a second element of DISE training be implemented – where the future implanting physician submit 5 DISE done by himself/herself, along with the future implanter’s assessment, for second opinion review. If there is discrepancy about patient candidacy between the future implanter and the second opinion expert reviewer, the future implanter will continue to submit additional DISE until the expert reviewer is prepared to certify the future implanter. 

As a final point, we request that physicians who have been performing DISE and HSN implants prior to the go-live date of this LCD be grandfathered. Most patients with Inspire devices are included in an ongoing real-world registry (ADHERE). Results of ADHERE are published periodically, with the most recent paper including 1000 patients enrolled. The ADHERE results show that patients in the real-world setting have outcomes that are even better clinically than the original FDA PMA clinical study, STAR. Therefore, the current DISE training, leading to patient selection, is working, and we suggest that the group of implanters trained in the ‘old’ DISE methodology are successfully capable of performing the diagnostic and selecting patients who are anatomically suited for the therapy. These physicians can be ‘grandfathered’, and going forward, the new DISE training can be implemented for new prospective implanters.

Thank you for the comment. No peer reviewed articles were submitted with this request. There was a consensus among the subject matter experts (SMEs) that the second party review is appropriate, and the 15 procedures was a conservative but good number, allowing for greater success with implantation. Should data be forthcoming and demonstrates that similar results are being obtained on the first 5-10, it may be submitted through the reconsideration process. Providers that underwent FDA approved device manufacturer DISE training prior to the date of this LCD, shall be deemed to meet the criteria for satisfactory performance of DISE without further documentation. All such providers shall maintain certification of completion of this training, supply proof of training by manufacturer, and DISE results should be made available upon request.

We propose clarifying that LCD L36953 does allow for Home Sleep Tests for Sleep Apnea patients.

We agree. The requirement for sleep testing has been clarified to indicate the need for sleep testing by an accredited sleep facility.

I have been the sole provider responsible for evaluating and approving or denying the implant here…. During that time, I saw over 400 consults for Obstructive Sleep Apnea and consideration of the hypoglossal nerve stimulator. After each consult, the patient was sent to [a doctor] for a drug induced sleep endoscopy to determine final candidacy based on anatomical findings, as noted in the protocol set forth by Inspire and approved by the FDA. This resulted in implantation of 40 patients with a success rate equal to the national average of almost 90%. Neither myself NOR [the doctor] are board certified in sleep medicine, yet we were able to run one of the most robust and successful Inspire programs in the country.

You can see why it would be particularly concerning for me to have a "No PA/NP's allowed" in the shared decision making wording, given that ZERO of these implants would have taken place without my assessment and workup of these patients. Is an APP not capable of determining whether or a patient has demonstrated CPAP failure or intolerance? Particularly an APP who has undergone specific training for this device? This is within our scope of practice. Specifically, APP's working in the Otolaryngology or Sleep medicine specialty should be particularly versed in this when performing consults to determine eligibility for this device. During my time at [our institution] I met with multiple providers in the Sleep Medicine department on no less than 3 occasions, in person, in an attempt to garner interest in participation in the Inspire program. Every single time, the answer was a resounding "No thanks! Not interested". It is unclear why.

As the only active program in the state of South Carolina, this extremely discriminatory policy would deny these patients access to care. Advanced practice providers can freely practice every single subspecialty - Oncology, Cardiology, Pain Management, Organ Transplant, Nephrology, Otolaryngology, and so on - but here you are saying we do not have the capability to simply determine CPAP failure, and recommend a bronchoscopy?

Along the lines of my above comments, it is discriminatory and inappropriate to determine that an advanced practice provider is incapable of performing an activation or titration. Again, ZERO of the 44 activations and titrations done over the last 3 years at [our institution] had ANY physician involvement. Preventing APP's from participating in fact in this case would present a huge barrier to access to care, given the backlog of appointments available for visits with a sleep medicine physician. At [our institution] specifically, it can take 3 months to get an appointment, thus pushing the patient out of the protocol window for activation. The sleep medicine physicians would need to be in constant contact with the implanting surgeon to know when patients are appropriate for activation. Why? Because the patient should not be activated until satisfactory levels of postop swelling have resolved. Since none of our sleep medicine physicians have ever seen a postoperative implant patient, how can they be expected to assess this? The activation date is determined AT THE POSTOP APPOINTMENT based on issues like possible nerve praxis, swelling, pain, et cetera. As is commonplace across ALL surgical subspecialties, APP's see the bulk of postoperative patients to reduce burden on the surgeon.

We respectfully disagree. The Subject Matter Experts (SMEs) that provided input for this service agreed with the provider specialty limitation. They agreed that all providers performing this service and the follow-up care should meet specified educational requirements. In the original Star trial and follow-up through five years, all the implanting surgeons were ENT, and approved per the FDA criteria. There was no documentation or literature submitted to support the commenters’ recommendation for inclusion of additional specialty providers. The limitation will remain as written.

…while I do not agree with the policy plan for “Shared Decision Making”, I can at least understand the idea behind it. However, this mandate that the implanting surgical team cannot participate in the activation and titration of a surgical device that they implanted is beyond my comprehension. Nowhere that I know of in the practice of medicine, is this done. In the surgical world this would be the equivalent of not allowing a heart surgeon to take care of the post-op treatment of his patients; in the implant world, this would be the equivalent of not allowing an interventional cardiologist to titrate a pacemaker he/she implanted; or in the OSA world, this would be the equivalent of a policy that allowed Sleep Medicine specialists to prescribe CPAP, but then mandated that only a surgical specialist be allowed to titrate the device.

Most importantly, this policy would absolutely be against the best interest of patients. First, a Sleep medicine specialist is not as well versed with the anatomy of the airway as the Otolaryngologist. Second, a sleep physician who has not participated in the implantation, would have no idea what nuances of the patients anatomy or the surgery itself might be in play as the device is later titrated. Lastly, there is no need for this mandate as it goes against the incredibly successful practices at most major implant centers, including our own (Figure 1). Our center alone has implanted nearly 100 hypoglossal nerve stimulators and later activated them without incident or complication. To change from this practice could only have a negative impact on patient care and would be going against the medical standards for this device as it is practiced around the world. There is literally no need to create this policy that requires a change from a successful current medical practice.

In addition, as seen with the previous comment, no other insurers to date have mandated that the implanting surgeon cannot activate and titrate the hypoglossal nerve stimulator.

I would therefore ask that this statement excluding the implanting physician from activation and titration be removed as well as the requirement that only a Sleep medicine physician being allowed to perform this service.

The Subject Matter Experts (SMEs) that provided input for this service agreed with the provider specialty limitations. They agreed that all providers performing this service and the follow-up care should meet specified educational requirements, which included formal training and expertise in sleep medicine. However, we have added language to include the initial empiric programming of the device as part of the routine post-op care provided by the implanting surgeon.

Medicare beneficiaries are particularly likely to have sleep apnea driven largely by ventilatory control instability. This is because by some estimates approximately 50% of patients with a history of atrial fibrillation or other arrhythmia, significant CAD (recent or nonrecent), CHF, CVA, or renal disease with sleep-disordered breathing have sleep apnea driven mainly by loop gain rather than obstruction. I note that the LCD has excluded patients with severe CHF. Do we know if data has been published regarding use of HNS in a significant number of patients with these other comorbidities? These comorbidities are likely to be severely overrepresented in the Medicare population and underrepresented in investigational study populations. In the two studies I have seen which specifically looked at the efficacy of HNS in patients age 65 and older, one reports a rate of the above comorbidities in 1% of the study population while the other does not report specific values (https://www.ncbi.nlm.nih.gov/pubmed/31084350, https://www.ncbi.nlm.nih.gov/pubmed/29957285). The STAR trial reports a rate of 2% (2 patients). Particularly in patients with atrial fibrillation, in whom ventilatory control instability is exceedingly common, it may be very helpful to ask Inspire Medical Systems to share any data they might have in this regard.

Clearly it is important to recognize patients with high loop gain-drive sleep apnea before directing them towards HNS. Fortunately, this phenotype of sleep apnea has readily distinguishable features on polysomnography. Namely, events occur in periodic breathing pattern (caused by cyclical ventilatory overshoot and ventilatory undershoot) rather than the familiar pattern of progressively diminishing flow-limited breaths followed by large recovery breaths as seen in anatomic obstruction. I don't know if you feel this would be practical but one thought is that going forward any sleep study being used to qualify a patient for HNS could include a statement by the reading physician to the effect that "I have viewed this study epoch-by-epoch to verify the obstructive hypopneas scored on this study do not primarily occur in a periodic breathing pattern." Easy tips for recognizing periodic breathing include a crescendo-decrescendo waveform pattern, self-similar events occurring at a fixed time interval (events look alike and occur in a cyclical pattern), and improvement during REM sleep.

Thank you for the comments. There was no evidence submitted with the comment to support the commenters’ recommendation.

The American Academy of Sleep Medicine (AASM) appreciates the opportunity to comment on the Proposed Local Coverage Determination (LCD) for Hypoglossal Nerve Stimulation for the Treatment of Obstructive Sleep Apnea (OSA), as the proposed LCD will directly impact AASM member reimbursements. The comments included in this response reflect the opinions of our over 9,000 individual members and 2,500 accredited sleep centers, providing sleep medicine services to the Medicare population.

Covered Indications, Limitations, and/or Medical Necessity

CPAP failure or intolerance

AASM urges Palmetto GBA to include CPAP refusal or non-acceptance as a possible indication, provided that consultation was performed with a sleep expert AND the surgeon qualified for implantation (if they are not the same). There are patients who will agree to undergo PAP home treatment post lab titration or who refuse PAP treatment despite urging by the provider. As long as a conversation about RBAs and limitations of hypoglossal nerve stimulation takes place with a sleep expert and surgeon, and that in their judgement the therapy is appropriate, it should also be allowed.

Uncontrolled hypertension

While we agree and acknowledge that hypertension be assessed and controlled as best as possible for implantation procedure safety, exclusion of this group may forgo treatment for those whose sleep apnea needs management most. AASM, therefore, recommends that hypertension be removed from the list of exclusions.

Drug Induced Sleep Endoscopy

DISE interpretation for complete circular collapse is still a bit more subjective and the second opinion service’s opinion may not necessarily be “more valid” unless the second opinion is provided by a clinician that is able to demonstrate CME or other course /education in sleep endoscopy.

Credentialing and Accreditation Standards

'Routine post-op care' is rather vague language. Post-implant device programming may occasionally require otolaryngologist evaluation using data outside of polysomnography metrics.

An example would be programming changes made based on findings from an awake endoscopy exam where impacts of device adjustment on anatomy are assessed. An otolaryngologist should be permitted to make these post-operative device adjustments when indicated as part of shared decision making with the managing sleep physician. This is a new and novel therapy for treatment of OSA, and otolaryngologists are as qualified as sleep medicine physicians to make these initial empiric programming changes to the devices if it works for their practice models. Beyond this stage, post-implant evaluation of efficacy and further adjustments should occur under the guidance of a sleep medicine physician.

Thank you for the comments. CPAP is the gold standard for treatment of OSA and it is reasonable for Beneficiaries to be required to at least attempt CPAP use under the observation and management of a sleep specialist. Conservative therapy should be tried and failed and/or not tolerated, prior to a surgical approach. This requirement will remain as written.

The DISE interpretation proficiency must be certified by the FDA-approved manufacture’s second opinion service of validation. We believe this is sufficient evidence of expertise. The standard will remain as written.

Control of hypertension is necessary to optimize surgical outcomes. This limitation will remain as written.

We agree with the assessment that the language was vague regarding ‘routine post-op care’. The sentence is revised to include the initial empiric programming as representative of routine post-op care.