JDSM

O RI G I N AL A RT IC L E S http://dx.doi.org/10.15331/jdsm.6742

A Fully Digital Workflow and Device Manufacturing for Mandibular

Repositioning Devices for the Treatment of Obstructive Sleep
Apnea: A Feasibility Study
Shouresh Charkhandeh, DDS1,2; David Kuhns, PhD3; Sung Kim, BS3
The Snore Centre, Calgary, Canada; 2Zephyr Sleep Technologies, Calgary, Canada; 3ProSomnus Sleep Technologies, Dublin, California
1

Study Objectives: The objective of this study was to examine the feasibility and patient preference of mandibular repositioning
devices (MRDs) made from (1) a fully digital workflow including patient scanning, bite registration, and device manufacturing using
an open platform intraoral scanner (iTero; Align Technology Inc., San Jose, California, United States) and a computer aided design/
computer aided manufacturing (CAD/CAM) MRD (MicrO2 Sleep Appliance; ProSomnus Sleep Technologies, Pleasanton, California,
United States), and (2) a conventional workflow using polyvinyl siloxane (PVS) impressions as the source of patient data. The fit
accuracy and comfort level of devices from each workflow were compared. A method for capturing a digital scan using a stable open
bite technique is tested for accuracy and ease of implementation.
Methods: This prospective feasibility study reports on selected patients (n = 5) recruited from a dental practice (ages 34 to 51 years;
apnea-hypopnea index 1–32 events/h). Each patient received two MRDs. The first MRD was manufactured using a conventional
workflow model (ie, PVS impressions, PVS bite registration, CAD/CAM of the MRD with quality control on physical stone models).
The second MRD was manufactured using a digital workflow model (ie, digital intraoral scanning using an open platform intraoral
scanner, digital open bite registration, and CAD/CAM of the MRD). All patients were then followed up using our standard clinical
protocol set based on the American Academy of Dental Sleep Medicine guidelines. The dental fit (ie, the accuracy of the fit against the
teeth), the occlusal fit (ie, the accuracy of the upper and lower splint fitting together), and the patient preference (ie, which appliance
they preferred using) was recorded.
Results: In the conventional workflow 2 appliances required minor dental adjustments and 3 appliances required occlusal adjustments.
In the digital workflow, no dental or occlusal adjustments were required. All patients preferred the digital MRDs in terms of comfort.
Conclusions: It is feasible to utilize a fully digital workflow for manufacturing of MRDs for treatment of obstructive sleep apnea with
predictable outcomes. Devices from a full digital workflow are preferred by patients compared to the standard process. The full digital
workflow using the CAD/CAM MRD required fewer adjustments and patient satisfaction with the device and the process could make
for a better experience for the patient and the doctor. Follow-up studies are required using larger sample size and different intraoral
scanners and MRDs.
Keywords: bite registration, device manufacturing, digital workflow, intraoral scanning, mandibular repositioning devices,
obstructive sleep apnea, oral appliance therapy
Citation: Charkhandeh S, Kuhns D, Kim S. A fully digital workflow and device manufacturing for mandibular repositioning devices
for the treatment of obstructive sleep apnea: a feasibility study. Journal of Dental Sleep Medicine. 2017;4(4):97–102.

INTRODUCTION American Academy of Sleep Medicine guidelines, oral appli-

ances are recommended as an alternative for patients with
Obstructive sleep apnea (OSA) is a very common chronic mild to moderate OSA and patients who cannot tolerate CPAP
disease with many adverse clinical consequences, affecting or are not adherent to the therapy.11 There has been tremen-
an estimated 10% to 20% of the United States population.1,2 dous growth in this area of dentistry over the past decade and
The condition remains undiagnosed in most of these patients, many patients are now seeking this treatment modality. It is
and studies show the percentage of undiagnosed OSA could estimated that approximately 200,000 appliances are being
be as high as 80% to 90%.3,4 Untreated OSA is associated made every year in North America and by the year 2023 this
with higher risk of fatal cardiovascular and cerebrovascular number will reach over 1 million. Although the growth seems
events,5-8 hence, the importance of long-term treatment of to be very fast, it is worth noting that currently fewer than 10%
patients. Currently, the most commonly prescribed treatment of patients who could respond to oral appliance therapy (OAT)
for patients with OSA is continuous positive airway pressure receive a mandibular repositioning device (MRD).12 Based on
(CPAP) therapy. Although this treatment is efficacious and meta-analysis, 50% to 60% of all patients with OSA would
relatively safe with minor side effects, patient adherence is low respond to OAT,11 but MRDs are only prescribed in 5% to 10%
and many patients refuse to start CPAP treatment. It is reported of all diagnosed patients.
that patient adherence to CPAP could be as low as 50%; there- The number of patients prescribed an MRD could increase
fore, an alternative therapy is required.9,10 Based on published because of many factors, such as the general public’s raised

Journal of Dental Sleep Medicine 97 Vol. 4, No. 4, 2017

 Digital Workflow and Device Manufacturing for Mandibular Repositioning Devices—Charkhandeh et al.

Table 1—Patient profiles.

Baseline AHI, Follow-Up
Patient No. Age, years Sex, M/F events/h AHI, events/h Height, m Weight, kg BMI, kg/m2
1 51 M 32 8 1.78 109 34.4
2 34 M 1 1 1.88 93 26.3
3 35 M 3 2 1.85 90 26.3
4 34 M 5 2.2 1.88 92 26.0
5 40 M 28 4 1.75 86 28.1

AHI = apnea-hypopnea index, BMI = body mass index.

awareness of the importance of proper sleep, positive health and cone beam computed tomography). Proper medical
outcomes associated with OAT, and prospective selection of history, dental history, and consent forms were signed. All
patients who would respond to OAT.12,13 Although this is great patients were also examined by a dentist to determine the suit-
news in terms of growth in the field of dental sleep medicine, ability for an MRD.
the number of patients will increase our current clinical work- Each patient went through both MRD manufacturing work-
flow models and device manufacturing may not be able to flow models. They first went through the conventional workflow
keep up with the increased demand. Therefore, having more (Figure 1) using a George Gauge (Great Lakes Orthodontics,
efficient workflow systems and device manufacturing become New York, United States) to obtain a PVS bite registration.
crucial in delivery of care. All patients then went through the digital workflow model
The use of intraoral scanners for digital impressions of the (Figure 2). The digital workflow started with taking upper and
teeth and a fully digital workflow system, including the device lower intraoral scans using the Ortho Software Module on the
manufacturing, could be a way to more efficiently produce iTero scanner (Align Technology Inc., San Jose, California,
MRDs. The use of digital scanners in dentistry is expected to United States). The bite registration was taken using the initial
grow by 13.9% (compound annual growth rate) from 2015 to bite and protrusion level as a reference. In order to stabilize
2022.14 Digital scanners have been used widely in the dental the bite at the protruded position and enabling a scan of the
industry over the past decade and there seems to be a lot of “maxillomandibular” relation, the dentist segmented each bite
inconsistent results in terms of being able to register the into 3 pieces, using a surgical blade and a heat torch. The blade
maxillomandibular occlusal relationship, when the teeth are was used to cut through the PVS material, to minimize the
not fully occluding (ie, “digital open bite registration”). This chipping of the material and distortion of the bite. The heat
has been the main hurdle for many practitioners to make torch was used to enable cutting through the plastic fork of
use of a fully digital workflow, and therefore, they still use a the George Gauge without creating any distortion in the bite
conventional “bite registration.” This creates inefficiencies in (Figure 2). The bite was then stabilized by asking the patient to
the workflow, and it has the potential to introduce inaccura- bite into 2 out of the 3 pieces for each bite scan. When scanning
cies in device manufacturing. The difference in the degree of the left side, the anterior and the right segments were seated
accuracy between conventional bite registration using poly- (Figure 3). When scanning the right side, the anterior and the
vinyl siloxane (PVS) and digital bite registration could result left segments were seated in patients’ mouths. This stabiliza-
in improper articulation of the models. An additional benefit tion method was used to minimize the amount of distortion
of a fully digital manufacturing process is the reduced need for and inaccuracies in the bite registration by minimizing the
physical models. possible cantilever effect and compression of retrodiscal tissue
If a reliable, fully digital workflow can be established, it in the temporomandibular joint when no posterior support
could improve patient care by providing more efficient and is present or there is movement of the mandible with respect
accurate care to the patient. The objective of the current study to the maxilla. All scans were sent to the appliance manufac-
was to examine the feasibility and patient preference of MRDs turer (ProSomnus Sleep Technologies, Pleasanton, California,
made from a fully digital workflow versus the conventional United States).
workflow. Upon receipt by the manufacturer, each physical impression
was poured and articulated with the bite. The bite was then
METHODS AND RESULTS inspected to ensure enough vertical gap (> 3 mm) to enable
manufacture of the MicrO2 Sleep Appliance (ProSomnus Sleep
Five patients were selected to participate in this feasibility Technologies, Pleasanton, California, United States). A techni-
study. The patients’ characteristics and data are summarized in cian measured the vertical gap between the upper and lower
Table 1. They were all tested for sleep apnea using an at-home arches, and determined if any anatomical features are closer to
level 3 sleep recorder. The patients’ baseline apnea-hypopnea each other than the required specification of 3 mm as described
indexes are summarized in Table 1. After the baseline sleep in the manufacturing technical work instructions. This could
study, all patients received full orthodontic records (ie, upper be the vertical gap between 2 opposing cusps, for example.
and lower PVS impressions, intraoral/extraoral photographs, All technicians were trained to the same work instructions

Journal of Dental Sleep Medicine 98 Vol. 4, No. 4, 2017

 Digital Workflow and Device Manufacturing for Mandibular Repositioning Devices—Charkhandeh et al.

Figure 1—PVS and digital patient data capture.

Figure 2—Bite preparation for digital scan.

A B C

(A) Sectioning the bite-registration. (B) Three segments after sectioning. (C) Check for any distortion.

and specifications. If the specification was not met, the data the distance difference between each surface point for the
would be reviewed by quality personnel to determine if a new 2 files. Figure 4 shows the upper arch overlay on the left and
bite would be required. The models were scanned separately the lower arch overlay on the right, both with a scale of ± 150
and together with the bite and the files were imported into microns. The gray reference portions of the overlay repre-
the design software. Digital arches and bites were received sent the scans of the poured impressions; the colorized map
through the vendor’s portal and imported into the design soft- shows the delta intersection of the layered digital scan, with
ware. Once uploaded, the quality of the scans and bites were red above the reference surface and blue below the reference
inspected digitally. The scan quality specification included surface. The green areas describe areas of agreement between
evaluating that the digital mesh represents the physical model the scanned stone model and the iTero scan directly from the
without holes in the mesh or missing data that would prohibit patient. However, many areas of deviation are denoted by blues
the manufacture of the device. and reds in the 2 images.
To complete the comparison of the physical impressions and Each patient then received both appliances and wore each
digital impressions, both were loaded into the design software appliance for 3 weeks. The dentist also recorded the amount
and set in place using the same coordinate system. With the of adjustment needed for each device in terms of adjustment
models layered (Figure 4), a color map was generated showing made to the inside surface (to fit the teeth) and adjustment

Journal of Dental Sleep Medicine 99 Vol. 4, No. 4, 2017

 Digital Workflow and Device Manufacturing for Mandibular Repositioning Devices—Charkhandeh et al.

Figure 3—Bite set for iTero Scan.

A B

(A) Placing sectioned bite registration. (B) Digital bite registration.

Figure 4—Overlay of impression and digital data.

AB

(A) Maxillary Arch. (B) Mandibular Arch.

Table 2—Fit adjustments and patient preference for each process.

Device Interior Adjustments Device Occlusal Adjustments Patient Preference
Patient No. Conventional Digital Conventional Digital Conventional Digital
1 x x
2 x x
3 x
4 x x x
5 x x

AHI = apnea-hypopnea index, BMI = body mass index.

Journal of Dental Sleep Medicine 100 Vol. 4, No. 4, 2017

 Digital Workflow and Device Manufacturing for Mandibular Repositioning Devices—Charkhandeh et al.

Figure 5—Comparison of steps in a digital workflow compared to a conventional process.

to the occlusal surfaces (to achieve properly balanced occlu- Also,device manufacturing companies prefer the digital work-
sion). The results are summarized in Table 2. After wearing flow due to the consistency,efficiency and in-office inspection
each device for 3 weeks, each patient was asked which appli- of scan quality. Figure 5 illustrates the reduction in manual
ance they preferred in terms of comfort. The conventional steps in a digital workflow. Each manual step adds error due to
device was worn first, followed by the digital device. There was human touch that varies by technician. Additionally,manually
no washout period between the 2 appliances. The results are designing and building up the device via the “salt and pepper”
shown in Table 2. A follow-up sleep study was performed to method also adds technique and material variance. The
evaluate the efficacy of the treatment, and the patients’ follow- monomer/polymer mix, cure time, and resultant properties
up apnea-hypopnea indexes are summarized in Table 1. will also be varied. Milling from a controlled cured material
reduces these errors, allowing better translation of data from
DISCUSSION the patient to the finished device. This results in less time being
spent both in terms of the dental clinic and in manufacture
The results of the study suggest that it is feasible to implement a of the device. A digital workflow model may also help provide
fully digital workflow using the existing technology in a dental faster turnaround time for the appliance delivery and hence
practice and device manufacturing facilities, with no need for provide more timely care for patients in need of an MRD.
physical dental models. The digitally designed and manufac- Currently,time to efficacious treatment for CPAP therapy can
tured devices had better accuracy in terms of dental fit and be very quick,whereas that for OAT appears to take more time.
occlusal/bite fit as demonstrated by the digital file overlay and This can be caused by many factors such as lack of proper
patient preference. All patients preferred the digital devices in patient selection,subjective titration,and conventional device
terms of comfort. Although the sample size in this study was manufacturing. However,with better methods for patient
very small and further controlled double-blinded studies are selection for OAT and efficacious target protrusion,this time
required to reproduce the results, the possibility of having a can be reduced significantly.
fully digital workflow in dental sleep medicine is potentially The addition of fully digital device manufacturing could be a
beneficial for both the patient and the clinician. Digital scans significant addition to this model by further reducing the time
seem to be preferred by most patients in terms of comfort.15 to efficacious treatment. Also, incorporating digital workflow
Dental clinicians using such technologies will minimize in dental sleep medicine would allow better long-term moni-
the chances of distortion and inaccuracies,and in case of a toring of occlusal changes for patients with an MRD, as the
distorted scan,a full new impression will not be required. digital scans can be readily stored, retrieved, and compared. As
The distorted segment can easily be erased and re-scanned. the demand for OAT and MRDs increases, we as clinicians and

Journal of Dental Sleep Medicine 101 Vol. 4, No. 4, 2017

 Digital Workflow and Device Manufacturing for Mandibular Repositioning Devices—Charkhandeh et al.

practice managers will require workflows and manufacturing 4. Young T, Skatrud J, Peppard PE. Risk factors for obstructive sleep
technologies that are easily scalable. The full digital workflow apnea in adults. JAMA. 2004;291(16):2013–2016.
model is aligned with the future need to treat the many patients 5. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the
association between sleep-disordered breathing and hypertension.
who will soon receive a diagnosis. N Engl J Med. 2000;342(19):1378–1384.
In terms of the bite registration and the translation of a 6. Peker Y, Carlson J, Hedner J. Increased incidence of coronary
conventional bite to a digital bite registration, the investiga- artery disease in sleep apnoea: a long-term follow-up. Eur Respir J.
tors found that this procedure is very technique sensitive at 2006;28(3):596–602.
this stage and depends heavily on the expertise of the clinician 7. Punjabi NM, Polotsky VY. Disorders of glucose metabolism in sleep
apnea. J Appl Physiol. 2005;99(5):1998–2007.
and their level of experience with digital scanners. An impor-
8. Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term
tant aspect is to have proper stabilization of the protrusive bite cardiovascular outcomes in men with obstructive sleep apnoea-
to prevent any unwanted movement of the mandible and the hypopnoea with or without treatment with continuous positive airway
dentition during the scanning. Another very important aspect pressure: an observational study. Lancet. 2005;365(9464):1046–1053.
is to maintain the proper vertical dimension both in anterior 9. Engleman HM, Wild MR. Improving CPAP use by patients with
the sleep apnoea/hypopnoea syndrome (SAHS). Sleep Med Rev.
and posterior region. Failure to provide such posterior support 2003;7(1):81-89.
may result in unwanted compression of the soft tissue around 10. McArdle N, Devereux G, Heidarnejad H, Mackay T, Douglas N.
the temporomandibular joint, thus resulting in an inaccurate Long term use of CPAP therapy for sleep apnea/hypopnea syndrome.
bite registration. Keeping this in mind, a properly designed Am J Respir Crit Care Med. 1999;159(4 Pt 1):1108-1114.
gauge specifically designed for digital open bite registration 11. Ramar K, Dort LC, Katz SG, et al. Clinical practice guideline for the
treatment of obstructive sleep apnea and snoring with oral appliance
could be very helpful in overcoming discrepancies. therapy: an update for 2015. J Clin Sleep Med. 2015;11(7):773–827.
The sample size was very small for this study and the process 12. Frost & Sullivan. Hidden health crisis costing America billions.
was limited to one type of scanner and device type. Investi- Underdiagnosing and undertreating obstructive sleep apnea draining
gators believe more testing with higher numbers of patients healthcare system. American Academy of Sleep Medicine website.
http://www.aasmnet.org/sleep-apnea-economic-impact.aspx.
and utilizing different scanners and appliance designs would Published 2016. Accessed August 24, 2017.
be beneficial. Because of the small number of participants, the 13. Remmers J, Charkhandeh S, Grosse J, et al. Remotely controlled
statistical significance of the results is unknown and a follow- mandibular protrusion during sleep predicts therapeutic success
up randomized study with a larger sample size would be neces- with oral appliances in patients with obstructive sleep apnea. Sleep.
2013;36(10):1517-1525.
sary for better statistical analysis of the results. In addition,
14. News & Trends/Industry News. Dentistry Today. 2016;35(9):48.
randomizing the sequence of conventional versus digital MRDs 15. Yuzbasioglu E, Kurt H, Turunc R, Bilir H. Comparison of digital
with a washout period added in between would be important and conventional impression techniques: evaluation of patients’
for a more accurate comparison. Also, an easier and less tech- perception, treatment comfort, effectiveness and clinical outcomes.
BMC Oral Health. 2014;14(10):7.
nique-sensitive bite registration process needs to be established
to avoid segmenting the bite and creating possible issues. With
the increasing number of patients requiring MRD, such an ACKNOWLEDGMENTS
approach could result in more efficient workflow models with The authors thank the clinical staff at The Snore Centre (Amanda Duffy
faster turnaround time, and more accurate and more cost- and Ava McLellan) and the iTero Regional Manager, John Martin for
effective and scalable workflows. More follow-up studies would their clinical support. We also thank the ProSomnus Sleep Technologies
be useful to understand how such digital workflow, in combi- team for their support.
nation with other available technologies for patient selection,
would affect the general efficiency of clinical care delivery for SUBMISSION & CORRESPONDENCE
OAT, including time to efficacious treatment, avoidance of INFORMATION
OAT nonresponders, number of appointments required for
Submitted for publication February 2, 2017
OAT, cost of total treatment, and patient satisfaction. Submitted in final revised form June 24, 2017
Accepted for publication July 12, 2017
REFERENCES Address correspondence to: Shouresh Charkhandeh, #201 – 5540
Windermere Blvd, NW, Edmonton, AB, T6W 2Z8; Tel: +1-780-437-0377;
1. Punjabi NM. The epidemiology of adult obstructive sleep apnea. Email: [email protected]
Proc Am Thorac Soc. 2008;5(2):136-143.
2. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The DISCLOSURE STATEMENT
occurrence of sleep-disordered breathing among middle-aged adults.
N Engl J Med. 1993;328(17):1230–1235. This study was funded by ProSomnus Sleep Technologies. Dr. Kuhns is
VP of Technology and Sung Kim is VP of Engineering at ProSomnus.
3. Kapur V, Strohl KP, Redline S, Iber C, O’Connor G, Nieto J.
Underdiagnosis of sleep apnea syndrome in U.S. communities. Dr. Charkhandeh is the Clinical Director at The Snore Centre and the
Sleep Breath. 2002;6(2):49–54. Chief Dental Officer at Zephyr Sleep Technologies.

Journal of Dental Sleep Medicine 102 Vol. 4, No. 4, 2017