Article Text

Download PDFPDF

P055 Comparison of WatchPat to limited and full polysomnography studies in adults with suspected sleep-breathing disorders
  1. Alain Fildes and
  2. Adrian Kendrick
  1. University Hospitals, Bristol, Bristol Royal Infirmary, Bristol, UK


Background WatchPAT is a portable diagnostic device for studies of suspected Obstructive Sleep Apnoea (OSA) in adults. It is worn as a pulse oximeter and uses Peripheral Arterial Tone (PAT), which reflects sympathetic nervous system activation. PAT allows for complex signal analysis providing detailed information on sympathetic nervous system changes associated with Sleep Disordered Breathing (SDB) events and specific ‘signatures’ of sleep stages.

Methods Patients were admitted on clinical grounds for overnight sleep studies – either for full (FP: EEG, Respiratory, leg movements and ECG) or limited (LP: no EEG) polysomnography. Patients verbally consented to use the WatchPat device. Simultaneous recordings were obtained from WatchPat and SOMNOscreen PSG (S-Med, UK). This study had internal Trust approval. Studies were manually scored for SpO2 mean, SpO2 >4%, SpO2 time <90%, AHI, Sleep Onset Latency (SOL), Rem Onset Latency (ROL), REM% and Deep Sleep/N3(%), and compared using Bland-Altman analysis. Data are given mean±SD.

Results 39 studies were completed of which 13 were FP. The results are summarized in the table 1. For AHI, 25/39 (64%) of the studies were correctly classified by WatchPat as positive (mild/mod/severe) compared to SOMNO, giving a sensitivity of 75% and specificity of 80%. Of the studies that were mild (SOMNO), 10/17 (59%) were correctly identified, with 4/17 classified as normal and 3/17 as mod/severe. Data from the sleep data showed similarities between the data, although the bias for ROL was wide.

Abstract P055 Table 1

Summary of data comparing WatchPat to the in-lab SomnoMedics device. Data for SOL, ROL, REM% and N3% are based on n=13. Data for SPO2 and AHI are based on n = 39. Data are given as mean ± SD. For Bland-Altman, data is mean difference between devices ± bias of X1 SD.

Discussion This study has demonstrated that there is clinically acceptable differences between an in-lab study and WatchPAT for SpO2 and AHI and therefore WatchPat could be used as a clinically acceptable screening tool for suspected OSA. In terms of sleep staging, accepting that the sample size is small (n=13), WatchPat should be used with caution where sleep disorders other than OSA are suspected.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.