O011 Tolerance and validity of WatchPAT for paediatric sleep assessment

Abstract Introduction The WatchPAT (Itamar Medical) sleep assessment device uses a one-size finger probe with an inflatable cushion, and has been validated in adults. We assessed its tolerability and validity in children under and over 12 years of age compared to polysomnography (PSG). Methods Children (8-18years) underwent in-laboratory PSG with simultaneous WatchPAT-300. WatchPAT-automated total sleep time (TST; minutes), wake after sleep onset (WASO; minutes), %REM, %light and %deep sleep were compared to PSG. Good quality WatchPAT data was defined as not having data loss on the automated report. Results Data were collected in 64 patients with mean(±SD) age 11.9 ± 2.6y and BMI z-score 1.0 ± 1.2; 39% female, 61% <12y. All tolerated PSG; WatchPAT was aborted in 6 (10% of <12y, 8% of ≥12y; p>.05) before sleep onset due to finger discomfort. 44 (69%) had good quality WatchPAT recording, similar in both <12y (69%) and ≥ 12y (68%; p>.05), with mean PSG-TST being 409.6 ± 56.7 minutes compared to WP-TST 426.9 ± 52.7 (p=.03). Mean difference from PSG for WatchPAT was %REM +7.3% (p<.001), WASO +14.6min (p=.04), %deep sleep +0.5% (p>.05), and %light sleep -5.2% (p=.01). PSG- and WatchPAT-calculated TST and WASO were moderately correlated (r = 0.58, p<.001; r = 0.59, p<.001 respectively); %deep sleep was weakly correlated (r = 0.39, p=.01). A similar pattern of mean difference was seen in both age groups. Discussion Two-thirds of children in both age groups tolerated and had good quality WatchPAT data. Compared to PSG, WatchPAT distinguishes deep sleep but overestimates WASO and TST; the latter may affect accuracy in assessment of OSA severity in this population.