3D-Printed CPAP Masks for Children With Obstructive Sleep Apnea

Obstructive sleep apnea (OSA) is a common problem in the general pediatric population, generally cited as between 1-5%, with morbidity ranging from daytime behavioral problems and inattention to cardiopulmonary effects including hypertension and ventricular hypertrophy. OSA is dramatically more prevalent among children with certain craniofacial anomalies and syndromes (e.g. Pierre-Robin, Treacher Collins, etc.), generally because of small, short jaws, midface hypoplasia, and/or disproportionally large tongues . While tonsillectomy and adenoidectomy is considered first line therapy for OSA in the general pediatric population, children with OSA due to craniofacial anomalies frequently require more aggressive intervention to improve their breathing at night, which often includes continuous positive airway pressure ventilation (CPAP). Most children who require CPAP therapy are able to find a mask that will adequately seal while providing acceptable comfort, however a small percentage of children encounter significant difficulty finding a functional CPAP interface, most often because of dysmorphic facial features. This can prove a significant barrier to effective CPAP therapy and lead to frustration on the part of patients' caregivers and providers, as well as the associated morbidity of untreated severe OSA. The purpose of this feasibility study is to investigate the use of patient-specific computational design and three-dimensional (3D) printing to produce personalized CPAP masks for children intolerant of standard CPAP masks due to poor fit secondary to craniofacial anomalies who encounter significant difficulty using CPAP because of poorly fitting masks despite exhausting available commercial mask options.

Study Design

This will be a prospective case study examining the feasibility of using patient-specific CAD and 3D printing technology to produce personalized CPAP masks for children intolerant of commercially available masks due to poor fit. Patients will be recruited from the pediatric otolaryngology, pediatric oral maxillofacial surgery, and pediatric sleep clinics at the University of Michigan Medical Center, with a target cohort of five.

Only patients of study team members will be included in the study, and no recruitment will take place outside the clinics of participating study team members. We anticipate the study period to last 3 years, though it may terminate sooner if recruitment targets are reached expeditiously.

After recruitment and consent of a potential study subject, each subject will undergo an initial mask-design evaluation with members of the research team. At this initial visit, the following will be performed:

1. A head \& neck physical exam, which a focus on characteristics important for treatment of OSA (e.g. jaw size and position, tongue size and position, characteristics of palate, presence or absence of cleft, external nasal deformity, external or internal nasal valvular collapse)
2. Review of patient's photographs
3. Generation of a topographic model of the patient's facial anatomy utilizing the 3dMDface system (specifics below) which will be export as a Stereolithography (.STL) file.

The patient's facial .STL file will then be imported into a computer-aided design (CAD) modeling program (MimicsTM or MagicsTM, Materialise, Belgium). Contact points along the topographic model are drawn out and utilized for modeling of the custom mask. The custom mask model is then exported in .STL format for import into the 3D printer for fabrication using a fused-deposition modeling (FDM) method (Object Pro, Stratasys, Israel).

Once the mask has been fabricated, patients will undergo a mask-fit evaluation. If a mask appears to fit well, the patient will use it at home with home CPAP for one month of consistent use. They will then return for debriefing and re-evaluation, with subsequent iterations of mask design as needed. If the mask functions for the patient better than any other alternatives, the patient will be allowed to continue to use the mask for up to one year, with ongoing surveillance by the research team, assessing for mask durability and ongoing usage/compliance.

Data collection will include:

• Objective data regarding the nature of each patient's obstructive sleep apnea and CPAP pressure requirements sleep-disordered breathing from the most recent available polysomnograms (sleep study, abbreviated PSG)
• Information regarding all prior CPAP masks attempted
• One month compliance data downloads from CPAP machines - with both the prior "best alternative" if usable, and with the customized mask, including percentage of days used, average hours usage, residual apnea-hypopnea index (AHI) and time spent in large leak per night.
• Standardized interview with patients and their parents regarding other masks used and experience with the customized mask(s).
• Descriptive anatomic information regarding facial anomalies contributing to poor mask fit, obtained from physical exam, photographs, and imaging studies.
• Validated quality of life measures taken before and at completion of trial period with customized CPAP mask, including the Pediatric Sleep Questionnaire (PSQ) and the OSA-18 , to assess the impact of successful CPAP use for our patients.

Analysis will involve basic descriptive statistics to describe both the objective outcomes (e.g. compliance data from CPAP machines) and quality of life measures (PSQ and OSA-18 surveys) as well as information on individual experiences collected during interviews.

Additional Data on 3D Photography System

Patient facial modeling information will be obtained utilizing a 3dMDface system (3dMD, Atlanta, GA). The 3dMD system is a three-dimensional photography system which generates a three-dimensional model of the patient's face utilizing multiple convergent cameras (see Figure 1 below). The three-dimensional model is generated utilizing hybrid stereophotogrammetry, with software algorithms using both projected random patterns and texture of the skin (pores, freckles, etc.) to stereo-triangulate and generate a 3D surface image. The 3D model of the patient's face is created within the 3dMDvultus software system, which can then export the model in .STL format. No patient identifying information is stored within the .STL file.

All 3D photography sessions with study subjects will be performed or supervised by one of the members of the study team. There will be no cost associated with using the 3dMDface system. Utilizing 3D photography allows us to obtain the most time-accurate topographic information of the patient's face while avoiding the cost and risk associated with conventional CT or MRI imaging.

Data Included in Registry:

• Pre- and post- intervention OSA-18 and PSQ questionnaire aggregates. Data to be recorded includes: sum score and mean score for each questionnaire.
• Pre-intervention AHI and oxygen saturation (SpO2) nadir based on most recent sleep PSG.
• Pre-intervention CPAP requirements based on most recent CPAP titration PSG.
• Pre-intervention compliance data downloads from subject's CPAP machines from prior 1 month's use. Data to be recorded includes: percentage of days used, average hours usage, residual apnea-hypopnea index (AHI), size of leak in liters per minute, and time spent in large leak per night.
• Post-intervention compliance data downloads from subject's CPAP machines from prior 1 month's use. Data to be recorded includes: percentage of days used, average hours usage, residual apnea-hypopnea index (AHI), size of leak in liters per minute, and time spent in large leak per night.

Data Registry QA:

• Physical and electronic PDF versions of each data registry primary source (questionnaires, polysomnograms, and CPAP compliance reports) will be kept for the 5 years past the duration of the study. Data uploaded to the patient registry will be verified with the institutional IRB.