Abstract

Abstract Introduction Positive airway pressure (PAP) therapy has been used for obstructive sleep apnea (OSA) since 1980 and remains the treatment of choice for moderate to severe OSA. Proper mask fitting for patient comfort is essential to continued success on PAP therapy, as early adherence rates are strong predictors of future use. Mask selection continues to be challenging given the significant heterogeneity in craniofacial phenotypes and the rapidly growing mask options. Artificial intelligence software has combined survey data and digital photography to algorithmically predict PAP masks for best fit. This study aimed to describe mask failure rates in the Pittsburgh Veteran population with OSA following mask selection utilizing this specialized software. Methods Retrospective chart review was performed on 124 patients who underwent mask fitting with AI software May through November 2021. The primary outcome was to determine the rate of mask success within 1-3 months of fitting as defined by absence of mask switching. Mask switch was identified when patients underwent re-evaluation for new mask selection either by the AI software or had evidence of new mask selection from supply orders. Results Of the 124 patients who underwent evaluation with AI mask fit software, 96% were male with an average AHI 35/hr, BMI 34 and average age of 58 years. Of the 65 patients who were new to PAP therapy, 14 patients required refitting, yielding an initial mask acceptance rate of 78%. Of the 59 patients established on PAP, 8 required subsequent mask switches, yielding a refit mask acceptance rate of 86%. Overall mask acceptance rate after first exposure to fitting software across both groups was 82%. Conclusion AI software appears to result in successful mask fitting in a high proportion of VA patients. Higher mask acceptance rate was noted in the refit group, suggesting a particular use-case for this software. Most importantly, this software allows for remote mask fitting. This is ideal given recent telehealth growth and can save time and visits, leading to expedited therapy and increased access. Support (If Any)

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call