Is the 2013 American Thoracic Society CPAP-tracking system algorithm useful for managing non-adherence in long-term CPAP-treated patients?

Background Whereas telemedicine usage is growing, the only clinical algorithm for Continuous Positive Airway Pressure (CPAP) adherence management is that stipulated by the 2013 American Thoracic Society (ATS). The capacity of the latter to predict non-adherence in long-term CPAP-treated patients has not been validated. Methods Patients from the prospective real-life InterfaceVent study (NCT03013283, study conducted in an adult cohort undergoing at least 3 months of CPAP) and eligible for ATS algorithm usage were analysed. The residual device Apnea–Hypopnea-Index (AHI flow ) and High Large Leak (HLL) thresholds proposed in the ATS algorithm were evaluated for predicting adherence (i.e. AHI flow > 10/h, HLLs 95th > 24 L/min for ResMed® devices and ResMed® nasal mask, HLLs 95th > 36 l/min for ResMed® devices and ResMed® oronasal masks, HLLs > 1 h for Philips® devices and HHLs > 60 l/min for Fisher & Paykel® devices). Adherence was defined according to the 2013 ATS algorithm (i.e. CPAP use > 4 h/j for at least 70% of days). Results 650/1484 patients eligible for ATS algorithm usage were analysed (15.38% non-adherent, 74% male with a median (IQ 25–75 ) age of 68 (61–77) years, a body mass index of 30.8 (27.7–34.5) kg/m 2 , an initial AHI of 39 (31–55) events/h, and CPAP-treatment-duration of 5.1 (2.2–7.8) years). Logistic regression analysis demonstrated no significant relationship between the ATS proposed AHI flow or HLL thresholds and non-adherence. Complementary ROC curve analysis failed to determine satisfactory AHI flow and HLL thresholds. Conclusion When managing non-adherence in long-term CPAP-treated patients, our data do not validate absolute AHI flow or HLL thresholds in general. Trial registration The INTERFACE-VENT study is registered on ClinicalTrials.gov (Identifier: study ( NCT03013283 ).

Abstract licence: CC BY-NC-ND

12 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.