Scoring criteria for portable monitor recordings: a comparison of four hypopnoea definitions in a population-based cohort. Issue 11 (20th August 2015)

• Record Type: Journal Article
• Authors: Vat, Sopharat; Haba-Rubio, Jose; Tafti, Mehdi; Tobback, Nadia; Andries, Daniela; Heinzer, Raphael
• Is Part Of: Thorax. Volume 70:Issue 11(2015)
• Page Start: 1047
• Page End: 1053
• Publication Date: 2015-08-20
• Subjects: Sleep apnoea -- Equipment Evaluations -- Respiratory Measurement; Chest -- Diseases -- Periodicals; Thorax; Chest -- Diseases; Periodicals; Periodicals; 617.54
• DOI: 10.1136/thoraxjnl-2014-205982
• Languages: English
• ISSNs: 0040-6376

Abstract

Abstract : Rationale: Limited-channel portable monitors (PMs) are increasingly used as an alternative to polysomnography (PSG) for the diagnosis of obstructive sleep apnoea (OSA). However, recommendations for the scoring of PM recordings are still lacking. Pulse-wave amplitude (PWA) drops, considered as surrogates for EEG arousals, may increase the detection sensitivity for respiratory events in PM recordings. Objectives: To investigate the performance of four different hypopnoea scoring criteria, using 3% or 4% oxygen desaturation levels, including or not PWA drops as surrogates for EEG arousals, and to determine the impact of measured versus reported sleep time on OSA diagnosis. Methods: Subjects drawn from a population-based cohort underwent a complete home PSG. The PSG recordings were scored using the 2012 American Academy of Sleep Medicine criteria to determine the apnoea–hypopnoea index (AHI). Recordings were then rescored using only parameters available on type 3 PM devices according to different hypopnoea criteria and patients-reported sleep duration to determine the 'portable monitor AHIs' (PM-AHIs). Main results: 312 subjects were included. Overall, PM-AHIs showed a good concordance with the PSG-based AHI although it tended to slightly underestimate it. The PM-AHI using 3% desaturation without PWA drops showed the best diagnostic accuracy for AHI thresholds of ≥5/h and ≥15/h (correctly classifying 94.55% and 93.27% of subjects, respectively, vs 80.13% and 87.50% with PWA drops). There was a significant but modest correlation between PWA drops and EEG arousals (r=0.20, p=0.0004). Conclusion: Interpretation of PM recordings using hypopnoea criteria which include 3% desaturation without PWA drops as EEG arousal surrogate showed the best diagnosis accuracy compared with full PSG.