- Email: [email protected]
Real-time attended home-polysomnography with telematic data transmission
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
journal homepage: www.ijmijournal.com
Real-time attended home-polysomnography with telematic data transmission Marie Bruyneel ∗ , Sandra Van den Broecke, Walter Libert, Vincent Ninane Chest Service, Saint-Pierre University Hospital, Brussels, Belgium
a r t i c l e
i n f o
a b s t r a c t
Article history:
Purpose: Home-polysomnography (HPSG) has been proposed as a cost-effective alternative
Received 3 September 2012
for obstructive sleep apnea (OSA) diagnosis.
Received in revised form 26 February 2013 Accepted 27 February 2013
We assessed, in a feasibility study, whether telematic transmission using the Dream® and Sleepbox® technologies was associated with low HPSG failure rate. Methods: Patients referred by chest physicians for clinical suspicion of OSA underwent one HPSG, using Dream® and Sleepbox® (Medatec, Belgium), which is a wireless system able to
Keywords:
communicate with Dream® , and with Internet through a wi-fi/3G interface. It is equipped
Polysomnography
with a digital infrared camera, and with a speaker/microphone system for bidirectional
Obstructive sleep apnea
audio/video communication via Skype® .
Diagnosis
The Sleep Lab nurse performed a remote discontinuous monitoring of the PSG. In case of
Telemedicine
sensor loss, she called the patient who had been previously educated to replace the sensors.
Home sleep studies
Results: Twenty-one patients have been studied. 90% of the recordings were of excellent
Telemonitoring
quality. We observed a 10% PSG failure rate: one failure of the Dream® , and one recording of poor quality. There were 2 successful Skype® interventions resulting in readjustment of the defective probes (nasal cannula and EEG). PSG signal visualization was possible in 90% of cases but Skype® connection was problematic in 19% of cases. However, patients could be reached by phone to solve the problem. Conclusions: Real-time attended HPSG through telematic data transmission is feasible and could be an interesting perspective to decrease the failure rate of home sleep studies, even if some technical aspects need to be improved. © 2013 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
The use of assistive technology and telemedicine is likely to increase in developed countries [1], mainly to enhance monitoring in nursing homes, in hospital wards, and to allow data transmission from the patient’s home, but also to face the increasing demand for medical investigations related with the aging of the population, the growing complexity of medical
technology or to avoid patients displacement, especially when patients live far away from the hospital [2,3]. Obstructive sleep apnea syndrome (OSA) is a growing health problem, which is now recognized as an independent risk factor for hypertension, coronary heart disease, stroke and motor vehicle accidents [4–7]. The Wisconsin Sleep Cohort Study [8] found that OSA affected 2% of women and 4% of men but its prevalence, nowadays, is probably higher [9]. There is an increasing need for sleep study recordings and, despite the
∗ Corresponding author at: Chest Service, Saint-Pierre University Hospital, Rue Haute, 322, 1000 Brussels, Belgium. Tel.: +32 2 5354219; fax: +32 2 5354174. E-mail address: Marie [email protected] (M. Bruyneel). 1386-5056/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmedinf.2013.02.008
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
development of sleep medicine and sleep labs, the availability of polysomnography (PSG) remains limited in many countries [10] that face very long waiting lists. Home sleep studies with portable monitoring (PM) devices have been proposed to decrease costs and facilitate the diagnostic process [11]. We have recently shown that home-PSG (Type 2 PM [12]) is a cost-effective alternative for the diagnosis of OSA, with a good diagnostic accuracy. It is more comfortable for the patients, whose sleep efficiency is better than in the hospital [13]. Others studies, performed with simplified portable devices, have also shown an interest in these methods for screening/diagnosis with variable results [11]. The major problem encountered with these devices is the potential loss of data, observed with polysomnographic procedures (4.7–20%) [13] as well as with polygraphic procedures (up to 24%) [14], leading to less cost-savings than expected. The purpose of the study was to assess, in a feasibility study, whether telematic transmission using Dream® (portable polysomnographic device) and Sleepbox® technologies, with possible remote intervention, decreases home-PSG failure rate.
2.
Materials and methods
2.1.
Patients
Twenty-one consecutive patients referred by chest physicians to our sleep laboratory for clinical suspicion of OSA have been studied. Exclusion criteria included age under 18, restrictive respiratory disorders, other suspected sleep disorders and distance home-hospital >30 km. They all accepted that a video system would be functioning in their own sleep room and that they could be contacted directly to solve potential problems related to recording, and had to sign written informed consent. The study was approved by the local Human Studies Committee.
2.2.
Polysomnographies
2.2.1.
Devices
Recordings have been performed with a portable polysomnograph (Dream® , Medatec, Belgium), working on batteries, which allows the recording of thoracic and abdominal movements using piezoelectric sensors, airflow by nasal prongs, pulse oximetry (Nonin® , Minneapolis; 4-beat averaged values), electroencephalograms (EEG) using 5 channels (C3/A1, C4/A2, FP2/A2, FP1/A1, O1/A1), right and left electrooculograms (EOG), submental electromyogram, one anterior tibialis electromyogram (EMG) and ECG. Tracheal sounds were recorded via a microphone and body position was assessed using a built-in position sensor (mercury gauge) with 4 different levels. The sampling rate was 400 Hz. It is similar to attended in-hospital sleep studies, so we assumed that the recording’s quality was comparable. The Sleepbox® is a prototype developed by Medatec, Belgium (Fig. 1), which is not yet commercialized. It is a dual port wireless system, enabling on one side a short-range
697
Fig. 1 – The Sleepbox® is a dual port wireless system, enabling on one side a short-range communication with the wireless sleep monitor (Dream® ), and on the other side an Internet communication through a wi-fi/3G interface. It is equipped with a digital day/night camera with infrared leds, and with a speaker/microphone system for bidirectional audio/video communication using the Skype® computer program.
communication in the 2.4 GHz band with the low power wireless sleep monitor (Dream® ), and on the other side an Internet communication through a wi-fi/3G interface. It is equipped with a digital day/night camera with infrared leds, and with a speaker/microphone system for bidirectional audio/video communication using the Skype® computer program. Through highly secured communication protocols, the Sleepbox® allows real-time monitoring of the sleep signals (Dream® must be at a maximum distance of 30 m from the Sleepbox® ), together with video and audio telemonitoring, from any authorized hospital. A central secured server allows authorized people only to log on one particular Sleepbox® , and all communications happen through encrypted channels. Medatec Company has chosen to work with 3G network rather than with wired network to avoid computer connection problems related to the patient’s Internet network and use. The other reason was that we weren’t sure that all patients were equipped with an Internet connection at home.
2.2.2. Protocol 2.2.2.1. At home. All patients underwent one home PSG. A standard sleep questionnaire had been administered. Epworth Sleepiness Scale (ESS) score was recorded, as well as Body Mass Index (BMI), neck circumference and medical history/current medications. The morning after, subjective sleep quality and quantity have been evaluated by a questionnaire used in previous studies [13–15], and patients had also to answer few questions about the technical aspects of telemonitoring (Table 1). All home polysomnographic hook-ups have been performed by the same trained sleep technician, at home, around 6:00 PM.
698
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Table 1 – Questionnaire about the technical aspects of polysomnographic telemonitoring. Questions Have you call the nurse at bedtime? Yes – No Has the nurse called you during the night? Yes – No If Yes, how many times? 1 – 2 – 3 – >3 Did you understand what she asked you? Yes – No Was the task difficult? Yes – No Could you perform what she asked you? Yes – No Do you find the telemonitoring system Reasuring – Stressful?
The Sleepbox® was placed in the patient’s room, on the night table, in order to communicate with the patient. They were told that they will be distantly observed through a webcam, the first time when going to bed, to test the system, and then only in case of signal loss. Patients were carefully educated about the different recorded parameters and the right way to put probes/electrodes, in order to be able to replace them if necessary. The patient was asked to call (by phone) the nurse when going to bed. The technician came back the next morning to remove the monitor.
2.2.2.2. Telemonitoring from the Sleep Lab. The nurse working during the night shift in the Sleep Lab, was asked to perform a discontinuous monitoring of the recording performed at home. She currently attends 2 polysomnographies, but the maximum authorized by the Belgian regulation is 4, such that watching 3 recordings remains a reasonable workload. The visualization of the recording is a bit less performing than for attended PSG, because the transmission via 3G network results in a jerky course (at least one screen change/s) compared to continuous course. Immediately after the call of the patient, when he decided to go to bed, she had to connect to the Medatec SleepWeb program to control that she could watch the Dream’s recorded parameters on her own computer. To communicate with the patient during the night, in case of defective signal, she had to connect to Skype® and wear a microphone, in order to wake the patient by speaking him loudly through the Sleepbox® speaker. Fig. 2 summarizes the working scheme.
Fig. 2 – This scheme summarizes the polysomnographic telemonitoring protocol.
For the first check, she was instructed to check both systems. Afterwards, she controlled the sleep parameters each hour, until 6:00, when Sleepbox® was programmed to stop transmitting data. In case of defective signal of oxymetry, nasal pressure, ground electrode or when less than 2 EEG recordings were readable, she had to call and watch the patient through Skype® and ask him to replace the lost probes/electrodes. In case of impossibility to connect to Medatec SleepWeb program, it was considered as a 3G network signal strength problem and she did not disturb the patient during sleep.
2.2.3.
Analysis
Manual scorage of PSG, according to Rechtschaffen and Kales’s rules [16] and to national Belgian guidelines [17], has been applied. Respiratory events are expressed as Apnea–Hypopnea Index (AHI, mean number of events per hour of sleep). Apnea is defined as a complete cessation of airflow for at least 10 s and hypopnea as a 50% decrease in airflow or thoracoabdominal movement, for at least 10 s, or a smaller decrease provided it is accompanied by oxygen desaturation >3% or by an arousal. Arousals are also expressed as the mean number of occurrences per hour of sleep (ArI). Arousals are scored in agreement with the AASM rules [18]. Quality of recordings has been graded according to Redline et al. [19] and is summarized in Table 2. PSG failure was
Table 2 – Classification of the recording’s quality. Respiratory channels Unsatisfactory Poor Fair Good Very good Excellent
No usable data (artifact, software dysfunction, etc.) Respiratory channels (airflow and both bands) or oxymetry good for <4 h of data Respiratory channel (airflow or either bands) and oximetry good for >4 h and <5 h Respiratory channel (airflow or either bands), and oximetry good for >5 h At least oximetry, airflow and either band good for >5 h At least oximetry, airflow and both bands good for >5 h
Neurologic parameters Or EEG channels good for <4 h of data And one EEG signal good for >4 h and <5 h And one EEG signal good for >5 h And at least one EEG channel good for >5 h And at least one EEG channel, one EOG channel, chin EMG good for >5 h
699
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
patients were able to follow the indications of the nurse to replace correctly the probes. In patient 14, there was a loss of EEG on 3:18 AM, which was detected on 3:30. The nurse took contact with the patient via Skype® to enhance hook-up. On 4:30 and 5:30 the same problem occurred but Skype® was unavailable and the problem could not be resolved. She did not call the patient per phone. In patient 19, a defective nasal pressure signal was detected (after a 22 min period), and the nurse took contact with the patient by phone (because Skype® was not available) to ask him to replace the nasal cannula. The remote detected problems are resumed in Table 4.
Table 3 – Characteristics of the patients. N = 21 Male, female (%) Age (y) Mean ± SD BMI (kg/m2 ) Mean ± SD Epworth Sleepiness Scale Mean ± SD Neck circumference (cm) Mean ± SD AHI Median (min–max)
71/29 50 ± 11 30 ± 6 7±4 41 ± 3 223 (2–114)
considered if the recording’s “unsatisafactory” or “poor”.
3.
quality
was
graded
as
Results
Between January and February 2012, 21 patients were recorded. No patient refused home recording. Characteristics of the patients are resumed in Table 3. In all patients, 3G network quality was optimal. The nurses were able to communicate with all patients at bedtime (100% successful SleepWeb and Skype® checks). They performed a total of 164 visualizations of the PSG via the Medatec SleepWeb program (mean of 8 checks/patient). Sixteen times (9.75% of checks), the connection could not be established due to an “overflow” problem (overload of computer capacity for data transmission). They experienced much more Skype® connection problems. Indeed, in 4 patients, during the night, despite normal transmission of PSG recording, it was not possible to use Skype® properly. The total number of Skype® connections was 32 (1.52 checks/patient) and the Skype® connection failure rate reached 18.75%. The nurse detected 4 losses of signal (2 different patients). Intervention, when possible, was each time successful to resolve the sensor loss (100% success rate). The mean duration between detection and intervention was 17 min. The 2
3.1.
Polysomnographies
3.1.1.
Quality
Globally, the distant watching PSG quality results were excellent (19/21 patients, 90%) We experienced only 1 recording of poor quality (only 3h36 of EEG recording) and 1 failure with a recording of unsatisfactory quality due to defective batteries in the Dream® ). Among the 20 analyzable recordings, we obtained satisfactory signals 98.5% of recording time for thoracic band, 98% for abdominal band, 98.6% for oxymetry, 98.9% for nasal pressure, 96.2% for EOG and 96.2% for EEG. The mean duration of signal loss was 16 ± 8 min for thoracic band, 19 ± 8 min for abdominal band, 35 ± 24 min for oxymetry, 45 ± 28 min for nasal pressure, 76 ± 118 min for EOG, and finally 11 ± 26 min for EEG. Detailed quality data are resumed in Table 5.
3.1.2.
Polysomnographic diagnosis
Seventy five percent of the patients suffered from OSA (median AHI was 23). Among them, 20% exhibited associated periodic leg movement during sleep. Epilepsy was observed in one patient. Complete polysomnographic parameters are shown in Table 6.
Table 4 – Problems detected during the telemonitoring of home-PSG in the 21 patients. No. of patients
Detected problems
Cause
Intervention
9 14 14 19 13
SleepWeb OK No PSG Lost EEG (3:18 AM) Lost EEG (5:30 AM) Lost nasal cannula SleepWeb OK No PSG
Patient in bathroom
Dream® blocked
– Replaced via Skype® call Not replaced: Skype® out, No phone call Replaced via phone call, Skype® out Impossible
Table 5 – Detailed individual polysomnographic quality data.
Thoracic band Abdominal band Oxymetry Nasal pressure EOG EEG
Good quality signal (% of recording time)
Number of signal losses
98.5 98 98.6 98.9 96.2 96.2
9 10 3 3 5 34
Mean duration of signal loss (min) (mean ± SD) 16 19 35 45 76 11
± ± ± ± ± ±
8 8 24 28 118 26
700
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Table 6 – Complete polysomnographic parameters of the 21 recordings. Parameters
Results, median (min–max)
Total sleep time (min) Sleep efficiency (%) Sleep latency (min) Apnea–Hypopnea Index Arousal index Slow wave sleep (%) Rapid eye movement sleep (%) Desaturation time <90% (min)
3.1.3.
396 (199–520) 79 (45–92) 22 (6–84) 23 (2–114) 22 (8–73) 28 (13–47) 17 (0–31) 2 (0–211)
Subjective evaluation of sleep quality
Respectively 57 and 43% of the patients described sleep amount as habitual or reduced, and 52 and 43% expressed their sleep quality as usual or worse during PSG.
3.1.4.
Subjective evaluation of distant-watching system
All patients but one (20/21, 95%) preferred to have the PSG at home rather than in the hospital. Eighty-eight percent (15/17) of them appreciated the distant watching and found it reassuring, whereas only 2 patients (12%) considered it as stressful.
4.
Discussion
The main finding of this feasibility study is the demonstration that remote discontinuous attended home-PSG through telematic data transmission is feasible and could be an interesting perspective to decrease the failure rate of home sleep studies. To the best of our knowledge, this is also the first study using a real-time surveillance to enhance the quality of complete home sleep studies. Ninety percent of the home-PSG has been graded as very good or excellent. Only one PSG had to be repeated, for complete blocking of the recording device, a problem that, in all cases, cannot be distantly resolved. The other missed PSG is a poor quality trace, containing only 3h36 of EEG recording, but as respiratory traces were excellent and the diagnosis obvious (AHI = 41), we decided not to repeat the PSG. Despite these problems, the failure rate remains however low, comparable to those described in previous studies performed with unattended Type 2 PM (portable full polysomnograph) [13]. A larger series with the benefit of experience should result in a reduced number of missed PSG. The use of the system was simple, easy, comfortable and also reassuring for the patient. When the nurse’s intervention was required, the patient did wake up and was able to replace the defective sensor, and the quality of the recording was immediately improved. As expected, as they checked hourly the recording, the time between lacking data and intervention was always lower than 1 h. Few technical problems have been encountered, as expected with a recently developed device that was tested in real-life conditions for the first time. Some problems were expected, such as connection troubles; but for unexpected reasons, no problems were met with 3G network. On the opposite, problems occurred frequently with the Skype® connection,
and that leads to the conclusion that Skype® is probably not the best interface to communicate with the patient. However, the Skype® connection problems could easily be circled by taking contact with the patient via his/her cell-phone. The video communication was then lost but simple instructions could be given easily by phone. It must be stressed however that the nurse was not instructed to act so when she first met this problem and she did not call the patient by phone (no. 14), leading to poor quality PSG. Afterward, when this problem occurred for the second time (no. 19), she called the patient and could resolve the problems by phone. In the same way, overflow problems with Medatec SleepWeb were unexpected. These problems reflect the “infancy diseases” of an innovative technologic interface. The Medatec Company is currently working to face and improve the Sleepbox® System. The potential advantages of home-PSG with telemonitoring are multiple and include cost saving, by avoiding hospitalization for PSG and by preventing repeated sleep studies for failure. Campos et al. described similar findings using remote video-EEG consultation, allowing important cost savings by avoiding travel and lost of time related to consultations [20]. The use of complete PSG (Type 2 PM [12]) is also full of interest because diagnosis of other sleep disorders than OSA is possible, in the usual environment of the patient. And indeed, in addition to OSA diagnosis, we were able to diagnose epilepsy and periodic leg movements in our small population, what should not be the case when using Type 3 PM (unattended portable multi-channel sleep testing device) [12]. Despite the very exciting aspect of this study, we have to emphasize some limitations: the small size of the study (not surprising for a pilot study) leads to the need for further larger investigations to confirm the efficiency of the device, after technical improvements. In the future, we have also to compare telemonitored home-PSG with a control group (unattended home-PSG or attended in lab-PSG), in order to quantify quality improvement and cost-savings. Regarding costs, we can suppose, however, that the additional cost of 3G network connection, technician visits to patient’s home for hook-up and education will remain lower than the costs related to hospitalization. Regarding nurses’ extra-work, telemonitoring did not result in additional costs in our current work scheme but this element has also to be considered. Another possible problem, even if it was not remarkable in the present study, is the potential inability of the patient to replace correctly the probes when a problem is detected. Adequate patient education is necessary to minimize this problem. As telemedicine and assistive technology, including telemonitoring, teleassistance, and teleconsultation [21–23], is a part of the medicine of the future, it is really exciting to use novel technologies to enhance diagnostic means, diagnostic tools in order to contribute to increase healthcare accessibility.
Authors’ contribution Marie Bruyneel and Vincent Ninane wrote the scientific project. Marie Bruyneel and Sandra Van den Broecke collected and analyzed the data. Walter Libert collected the data. All the authors participated to manuscript redaction.
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Summary points What was already known on the topic: [7]
• Home-polysomnography (HPSG) is useful for OSA diagnosis but the failure rate of the recordings remains problematic. • When the quality of the recordings is bad, it leads to less cost-savings than expected. • Unattended polysomnography is better than polygraphy because it allows a wide range of diagnosis in sleep disorders.
[8]
[9]
[10]
What this study added to our knowledge: [11]
• Real-time telemonitored HPSG is feasible. • It could be an interesting perspective to decrease the failure rate of home PSG. • As it is an innovatice device, we faced technical problems that we need to improve. • However, 90% of the recordings were of excellent quality.
[12]
[13]
[14]
Conflict of interest None declared.
[15]
Acknowledgements We thank warmly Mr Driessens, Mr Karmoun and all the staff of MEDATEC Company, Brussels, Belgium, for their precious and enthusiastic collaboration. We also thank l’ “Association André Vésale-Aide à la recherche médicale” for their financial backing.
references
[1] A.G. Ekeland, A. Bowes, S. Flottorp, Effectiveness of telemedicine: a systematic review of reviews, Int. J. Med. Inform. 79 (2010) 736–771. [2] V. Nangalia, D.R. Prytherch, G.B. Smith, Health technology assessment review: remote monitoring of vital signs-current status and future challenges, Crit. Care 14 (2010) 233. [3] T. Raza, M. Joshi, R.M. Schapira, Z. Agha, Pulmonary telemedicine – a model to access the subspecialist services in undeserved rural areas, Int. J. Med. Inform. 78 (2009) 53–59. [4] M.E. Dyken, V.K. Somers, T. Yamada, Z.Y. Ren, M.B. Zimmerman, Investigating the relationship between stroke and obstructive sleep apnea, Stroke 27 (1996) 401–407. [5] L.J. Findley, M.E. Unverzagt, P.M. Suratt, Automobile accidents involving patients with obstructive sleep apnea, Am. Rev. Respir. Dis. 138 (1988) 337–340. [6] J.M. Marin, S.J. Carrizo, E. Vicente, A.G. Agusti, Long-term cardiovascular outcomes in men with obstructive sleep
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
701
apnoea–hypopnoea with or without treatment with continuous positive airway pressure: an observational study, Lancet 365 (2005) 1046–1053. J. Hung, E.G. Whitford, R.W. Parsons, D.R. Hillman, Association of sleep apnoea with myocardial infarction in men, Lancet 336 (1990) 261–264. T. Young, M. Palta, J. Dempsey, J. Skatrud, S. Weber, S. Badr, The occurrence of sleep-disordered breathing among middle-aged adults, N. Engl. J. Med. 328 (1993) 1230–1235. R.L. Riha, W.T. McNicholas, The genetic cardiovascular aspects of obstructive sleep apnoea/hypopnea syndrome, J. Eur. Respir. 33 (2009) 233–236. J.F. Masa-Jimenez, F. Barbé Illa, F. Capote Gil, E. Chiner Vives, Resources and delays in the diagnosis of sleep apnea–hypopnea syndrome, Arch. Bronconeumol. 43 (2007) 188–198. N.A. Collop, Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients, J. Clin. Sleep Med. 3 (2007) 737–747. Standards of Practice Committee of the American Sleep Disorders Association, Practice parameters for the use of portable recording in the assessment of obstructive sleep apnea, Sleep 26 (1994) 907–913. M. Bruyneel, C. Sanida, G. Art, W. Libert, L. Cuvelier, M. Paesmans, et al., Sleep efficiency during sleep studies: results of a prospective study comparing home-based and in-hospital polysomnography, J. Sleep Res. 20 (2011) 201–206. J.F. Masa, J. Corral, R. Pereira, J. Duran-Cantolla, M. Cabello, L. Hernández-Blasco, et al., Effectiveness of home respiratory polygraphy for the diagnosis of sleep apnoea and hypopnoea syndrome, Thorax 66 (2011) 567–573. F. Portier, A. Portmann, P. Czernichow, L. Vascaut, E. Devin, D. Benhamou, et al., Evaluation of home versus laboratory polysomnography in the diagnosis of sleep apnea syndrome, Am. J. Respir. Crit. Care Med. 162 (2000) 814–818. A. Rechtschaffen, A. Kales, A Manual of Standardised Terminology, Techniques and Scoring for Sleep Stages in Human Subjects, Brain Information Service, Los Angeles, 1968. INAMI/RIZIV, Convention relative au diagnostic et au traitement du syndrome des apnées obstructives du sommeil. www.inami.fgov.be (accessed 20.02.12). M. Bonnet, D. Carley, M. Carskadon, P. Easton, C. Guilleminault, R. Harper, et al., EEG arousals: scoring rules and examples: a preliminary report from the sleep disorders atlas task force of the American Sleep Disorders Association, Sleep 15 (1992) 173–184. S. Redline, M.H. Sanders, B.K. Lind, S.F. Quan, C. Iber, D.J. Gottlieb, et al., Methods for obtaining and analyzing unattended polysomnography data for a multicenter study, Sleep 21 (1998) 759–767. C. Campos, E. Caudevilla, A. Alesanco, N. Lasierra, O. Martinez, J. Fernandez, et al., Setting up a telemedicine service for remote real-time video-EEG consultation in La Rioja (Spain), Int. J. Med. Inform. 81 (2012) 404–414. M. Vitacca, L. Bianchi, A. Guerra, C. Fracchia, A. Spanevello, B. Balbi, et al., Tele-assistance in chronic respiratory failure patients: a randomised clinical trial, Eur. Respir. J. 33 (2009) 411–418. E. Biratti, A. Roth, Telecardiology, Isr. Med. Assoc. J. 13 (2011) 498–503. M. Bruyneel, W. Libert, V. Ninane, Detection of bed-exit events using a new wireless bed monitoring assistance, Int. J. Med. Inform. 80 (2011) 127–132.
journal homepage: www.ijmijournal.com
Real-time attended home-polysomnography with telematic data transmission Marie Bruyneel ∗ , Sandra Van den Broecke, Walter Libert, Vincent Ninane Chest Service, Saint-Pierre University Hospital, Brussels, Belgium
a r t i c l e
i n f o
a b s t r a c t
Article history:
Purpose: Home-polysomnography (HPSG) has been proposed as a cost-effective alternative
Received 3 September 2012
for obstructive sleep apnea (OSA) diagnosis.
Received in revised form 26 February 2013 Accepted 27 February 2013
We assessed, in a feasibility study, whether telematic transmission using the Dream® and Sleepbox® technologies was associated with low HPSG failure rate. Methods: Patients referred by chest physicians for clinical suspicion of OSA underwent one HPSG, using Dream® and Sleepbox® (Medatec, Belgium), which is a wireless system able to
Keywords:
communicate with Dream® , and with Internet through a wi-fi/3G interface. It is equipped
Polysomnography
with a digital infrared camera, and with a speaker/microphone system for bidirectional
Obstructive sleep apnea
audio/video communication via Skype® .
Diagnosis
The Sleep Lab nurse performed a remote discontinuous monitoring of the PSG. In case of
Telemedicine
sensor loss, she called the patient who had been previously educated to replace the sensors.
Home sleep studies
Results: Twenty-one patients have been studied. 90% of the recordings were of excellent
Telemonitoring
quality. We observed a 10% PSG failure rate: one failure of the Dream® , and one recording of poor quality. There were 2 successful Skype® interventions resulting in readjustment of the defective probes (nasal cannula and EEG). PSG signal visualization was possible in 90% of cases but Skype® connection was problematic in 19% of cases. However, patients could be reached by phone to solve the problem. Conclusions: Real-time attended HPSG through telematic data transmission is feasible and could be an interesting perspective to decrease the failure rate of home sleep studies, even if some technical aspects need to be improved. © 2013 Elsevier Ireland Ltd. All rights reserved.
1.
Introduction
The use of assistive technology and telemedicine is likely to increase in developed countries [1], mainly to enhance monitoring in nursing homes, in hospital wards, and to allow data transmission from the patient’s home, but also to face the increasing demand for medical investigations related with the aging of the population, the growing complexity of medical
technology or to avoid patients displacement, especially when patients live far away from the hospital [2,3]. Obstructive sleep apnea syndrome (OSA) is a growing health problem, which is now recognized as an independent risk factor for hypertension, coronary heart disease, stroke and motor vehicle accidents [4–7]. The Wisconsin Sleep Cohort Study [8] found that OSA affected 2% of women and 4% of men but its prevalence, nowadays, is probably higher [9]. There is an increasing need for sleep study recordings and, despite the
∗ Corresponding author at: Chest Service, Saint-Pierre University Hospital, Rue Haute, 322, 1000 Brussels, Belgium. Tel.: +32 2 5354219; fax: +32 2 5354174. E-mail address: Marie [email protected] (M. Bruyneel). 1386-5056/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijmedinf.2013.02.008
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
development of sleep medicine and sleep labs, the availability of polysomnography (PSG) remains limited in many countries [10] that face very long waiting lists. Home sleep studies with portable monitoring (PM) devices have been proposed to decrease costs and facilitate the diagnostic process [11]. We have recently shown that home-PSG (Type 2 PM [12]) is a cost-effective alternative for the diagnosis of OSA, with a good diagnostic accuracy. It is more comfortable for the patients, whose sleep efficiency is better than in the hospital [13]. Others studies, performed with simplified portable devices, have also shown an interest in these methods for screening/diagnosis with variable results [11]. The major problem encountered with these devices is the potential loss of data, observed with polysomnographic procedures (4.7–20%) [13] as well as with polygraphic procedures (up to 24%) [14], leading to less cost-savings than expected. The purpose of the study was to assess, in a feasibility study, whether telematic transmission using Dream® (portable polysomnographic device) and Sleepbox® technologies, with possible remote intervention, decreases home-PSG failure rate.
2.
Materials and methods
2.1.
Patients
Twenty-one consecutive patients referred by chest physicians to our sleep laboratory for clinical suspicion of OSA have been studied. Exclusion criteria included age under 18, restrictive respiratory disorders, other suspected sleep disorders and distance home-hospital >30 km. They all accepted that a video system would be functioning in their own sleep room and that they could be contacted directly to solve potential problems related to recording, and had to sign written informed consent. The study was approved by the local Human Studies Committee.
2.2.
Polysomnographies
2.2.1.
Devices
Recordings have been performed with a portable polysomnograph (Dream® , Medatec, Belgium), working on batteries, which allows the recording of thoracic and abdominal movements using piezoelectric sensors, airflow by nasal prongs, pulse oximetry (Nonin® , Minneapolis; 4-beat averaged values), electroencephalograms (EEG) using 5 channels (C3/A1, C4/A2, FP2/A2, FP1/A1, O1/A1), right and left electrooculograms (EOG), submental electromyogram, one anterior tibialis electromyogram (EMG) and ECG. Tracheal sounds were recorded via a microphone and body position was assessed using a built-in position sensor (mercury gauge) with 4 different levels. The sampling rate was 400 Hz. It is similar to attended in-hospital sleep studies, so we assumed that the recording’s quality was comparable. The Sleepbox® is a prototype developed by Medatec, Belgium (Fig. 1), which is not yet commercialized. It is a dual port wireless system, enabling on one side a short-range
697
Fig. 1 – The Sleepbox® is a dual port wireless system, enabling on one side a short-range communication with the wireless sleep monitor (Dream® ), and on the other side an Internet communication through a wi-fi/3G interface. It is equipped with a digital day/night camera with infrared leds, and with a speaker/microphone system for bidirectional audio/video communication using the Skype® computer program.
communication in the 2.4 GHz band with the low power wireless sleep monitor (Dream® ), and on the other side an Internet communication through a wi-fi/3G interface. It is equipped with a digital day/night camera with infrared leds, and with a speaker/microphone system for bidirectional audio/video communication using the Skype® computer program. Through highly secured communication protocols, the Sleepbox® allows real-time monitoring of the sleep signals (Dream® must be at a maximum distance of 30 m from the Sleepbox® ), together with video and audio telemonitoring, from any authorized hospital. A central secured server allows authorized people only to log on one particular Sleepbox® , and all communications happen through encrypted channels. Medatec Company has chosen to work with 3G network rather than with wired network to avoid computer connection problems related to the patient’s Internet network and use. The other reason was that we weren’t sure that all patients were equipped with an Internet connection at home.
2.2.2. Protocol 2.2.2.1. At home. All patients underwent one home PSG. A standard sleep questionnaire had been administered. Epworth Sleepiness Scale (ESS) score was recorded, as well as Body Mass Index (BMI), neck circumference and medical history/current medications. The morning after, subjective sleep quality and quantity have been evaluated by a questionnaire used in previous studies [13–15], and patients had also to answer few questions about the technical aspects of telemonitoring (Table 1). All home polysomnographic hook-ups have been performed by the same trained sleep technician, at home, around 6:00 PM.
698
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Table 1 – Questionnaire about the technical aspects of polysomnographic telemonitoring. Questions Have you call the nurse at bedtime? Yes – No Has the nurse called you during the night? Yes – No If Yes, how many times? 1 – 2 – 3 – >3 Did you understand what she asked you? Yes – No Was the task difficult? Yes – No Could you perform what she asked you? Yes – No Do you find the telemonitoring system Reasuring – Stressful?
The Sleepbox® was placed in the patient’s room, on the night table, in order to communicate with the patient. They were told that they will be distantly observed through a webcam, the first time when going to bed, to test the system, and then only in case of signal loss. Patients were carefully educated about the different recorded parameters and the right way to put probes/electrodes, in order to be able to replace them if necessary. The patient was asked to call (by phone) the nurse when going to bed. The technician came back the next morning to remove the monitor.
2.2.2.2. Telemonitoring from the Sleep Lab. The nurse working during the night shift in the Sleep Lab, was asked to perform a discontinuous monitoring of the recording performed at home. She currently attends 2 polysomnographies, but the maximum authorized by the Belgian regulation is 4, such that watching 3 recordings remains a reasonable workload. The visualization of the recording is a bit less performing than for attended PSG, because the transmission via 3G network results in a jerky course (at least one screen change/s) compared to continuous course. Immediately after the call of the patient, when he decided to go to bed, she had to connect to the Medatec SleepWeb program to control that she could watch the Dream’s recorded parameters on her own computer. To communicate with the patient during the night, in case of defective signal, she had to connect to Skype® and wear a microphone, in order to wake the patient by speaking him loudly through the Sleepbox® speaker. Fig. 2 summarizes the working scheme.
Fig. 2 – This scheme summarizes the polysomnographic telemonitoring protocol.
For the first check, she was instructed to check both systems. Afterwards, she controlled the sleep parameters each hour, until 6:00, when Sleepbox® was programmed to stop transmitting data. In case of defective signal of oxymetry, nasal pressure, ground electrode or when less than 2 EEG recordings were readable, she had to call and watch the patient through Skype® and ask him to replace the lost probes/electrodes. In case of impossibility to connect to Medatec SleepWeb program, it was considered as a 3G network signal strength problem and she did not disturb the patient during sleep.
2.2.3.
Analysis
Manual scorage of PSG, according to Rechtschaffen and Kales’s rules [16] and to national Belgian guidelines [17], has been applied. Respiratory events are expressed as Apnea–Hypopnea Index (AHI, mean number of events per hour of sleep). Apnea is defined as a complete cessation of airflow for at least 10 s and hypopnea as a 50% decrease in airflow or thoracoabdominal movement, for at least 10 s, or a smaller decrease provided it is accompanied by oxygen desaturation >3% or by an arousal. Arousals are also expressed as the mean number of occurrences per hour of sleep (ArI). Arousals are scored in agreement with the AASM rules [18]. Quality of recordings has been graded according to Redline et al. [19] and is summarized in Table 2. PSG failure was
Table 2 – Classification of the recording’s quality. Respiratory channels Unsatisfactory Poor Fair Good Very good Excellent
No usable data (artifact, software dysfunction, etc.) Respiratory channels (airflow and both bands) or oxymetry good for <4 h of data Respiratory channel (airflow or either bands) and oximetry good for >4 h and <5 h Respiratory channel (airflow or either bands), and oximetry good for >5 h At least oximetry, airflow and either band good for >5 h At least oximetry, airflow and both bands good for >5 h
Neurologic parameters Or EEG channels good for <4 h of data And one EEG signal good for >4 h and <5 h And one EEG signal good for >5 h And at least one EEG channel good for >5 h And at least one EEG channel, one EOG channel, chin EMG good for >5 h
699
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
patients were able to follow the indications of the nurse to replace correctly the probes. In patient 14, there was a loss of EEG on 3:18 AM, which was detected on 3:30. The nurse took contact with the patient via Skype® to enhance hook-up. On 4:30 and 5:30 the same problem occurred but Skype® was unavailable and the problem could not be resolved. She did not call the patient per phone. In patient 19, a defective nasal pressure signal was detected (after a 22 min period), and the nurse took contact with the patient by phone (because Skype® was not available) to ask him to replace the nasal cannula. The remote detected problems are resumed in Table 4.
Table 3 – Characteristics of the patients. N = 21 Male, female (%) Age (y) Mean ± SD BMI (kg/m2 ) Mean ± SD Epworth Sleepiness Scale Mean ± SD Neck circumference (cm) Mean ± SD AHI Median (min–max)
71/29 50 ± 11 30 ± 6 7±4 41 ± 3 223 (2–114)
considered if the recording’s “unsatisafactory” or “poor”.
3.
quality
was
graded
as
Results
Between January and February 2012, 21 patients were recorded. No patient refused home recording. Characteristics of the patients are resumed in Table 3. In all patients, 3G network quality was optimal. The nurses were able to communicate with all patients at bedtime (100% successful SleepWeb and Skype® checks). They performed a total of 164 visualizations of the PSG via the Medatec SleepWeb program (mean of 8 checks/patient). Sixteen times (9.75% of checks), the connection could not be established due to an “overflow” problem (overload of computer capacity for data transmission). They experienced much more Skype® connection problems. Indeed, in 4 patients, during the night, despite normal transmission of PSG recording, it was not possible to use Skype® properly. The total number of Skype® connections was 32 (1.52 checks/patient) and the Skype® connection failure rate reached 18.75%. The nurse detected 4 losses of signal (2 different patients). Intervention, when possible, was each time successful to resolve the sensor loss (100% success rate). The mean duration between detection and intervention was 17 min. The 2
3.1.
Polysomnographies
3.1.1.
Quality
Globally, the distant watching PSG quality results were excellent (19/21 patients, 90%) We experienced only 1 recording of poor quality (only 3h36 of EEG recording) and 1 failure with a recording of unsatisfactory quality due to defective batteries in the Dream® ). Among the 20 analyzable recordings, we obtained satisfactory signals 98.5% of recording time for thoracic band, 98% for abdominal band, 98.6% for oxymetry, 98.9% for nasal pressure, 96.2% for EOG and 96.2% for EEG. The mean duration of signal loss was 16 ± 8 min for thoracic band, 19 ± 8 min for abdominal band, 35 ± 24 min for oxymetry, 45 ± 28 min for nasal pressure, 76 ± 118 min for EOG, and finally 11 ± 26 min for EEG. Detailed quality data are resumed in Table 5.
3.1.2.
Polysomnographic diagnosis
Seventy five percent of the patients suffered from OSA (median AHI was 23). Among them, 20% exhibited associated periodic leg movement during sleep. Epilepsy was observed in one patient. Complete polysomnographic parameters are shown in Table 6.
Table 4 – Problems detected during the telemonitoring of home-PSG in the 21 patients. No. of patients
Detected problems
Cause
Intervention
9 14 14 19 13
SleepWeb OK No PSG Lost EEG (3:18 AM) Lost EEG (5:30 AM) Lost nasal cannula SleepWeb OK No PSG
Patient in bathroom
Dream® blocked
– Replaced via Skype® call Not replaced: Skype® out, No phone call Replaced via phone call, Skype® out Impossible
Table 5 – Detailed individual polysomnographic quality data.
Thoracic band Abdominal band Oxymetry Nasal pressure EOG EEG
Good quality signal (% of recording time)
Number of signal losses
98.5 98 98.6 98.9 96.2 96.2
9 10 3 3 5 34
Mean duration of signal loss (min) (mean ± SD) 16 19 35 45 76 11
± ± ± ± ± ±
8 8 24 28 118 26
700
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Table 6 – Complete polysomnographic parameters of the 21 recordings. Parameters
Results, median (min–max)
Total sleep time (min) Sleep efficiency (%) Sleep latency (min) Apnea–Hypopnea Index Arousal index Slow wave sleep (%) Rapid eye movement sleep (%) Desaturation time <90% (min)
3.1.3.
396 (199–520) 79 (45–92) 22 (6–84) 23 (2–114) 22 (8–73) 28 (13–47) 17 (0–31) 2 (0–211)
Subjective evaluation of sleep quality
Respectively 57 and 43% of the patients described sleep amount as habitual or reduced, and 52 and 43% expressed their sleep quality as usual or worse during PSG.
3.1.4.
Subjective evaluation of distant-watching system
All patients but one (20/21, 95%) preferred to have the PSG at home rather than in the hospital. Eighty-eight percent (15/17) of them appreciated the distant watching and found it reassuring, whereas only 2 patients (12%) considered it as stressful.
4.
Discussion
The main finding of this feasibility study is the demonstration that remote discontinuous attended home-PSG through telematic data transmission is feasible and could be an interesting perspective to decrease the failure rate of home sleep studies. To the best of our knowledge, this is also the first study using a real-time surveillance to enhance the quality of complete home sleep studies. Ninety percent of the home-PSG has been graded as very good or excellent. Only one PSG had to be repeated, for complete blocking of the recording device, a problem that, in all cases, cannot be distantly resolved. The other missed PSG is a poor quality trace, containing only 3h36 of EEG recording, but as respiratory traces were excellent and the diagnosis obvious (AHI = 41), we decided not to repeat the PSG. Despite these problems, the failure rate remains however low, comparable to those described in previous studies performed with unattended Type 2 PM (portable full polysomnograph) [13]. A larger series with the benefit of experience should result in a reduced number of missed PSG. The use of the system was simple, easy, comfortable and also reassuring for the patient. When the nurse’s intervention was required, the patient did wake up and was able to replace the defective sensor, and the quality of the recording was immediately improved. As expected, as they checked hourly the recording, the time between lacking data and intervention was always lower than 1 h. Few technical problems have been encountered, as expected with a recently developed device that was tested in real-life conditions for the first time. Some problems were expected, such as connection troubles; but for unexpected reasons, no problems were met with 3G network. On the opposite, problems occurred frequently with the Skype® connection,
and that leads to the conclusion that Skype® is probably not the best interface to communicate with the patient. However, the Skype® connection problems could easily be circled by taking contact with the patient via his/her cell-phone. The video communication was then lost but simple instructions could be given easily by phone. It must be stressed however that the nurse was not instructed to act so when she first met this problem and she did not call the patient by phone (no. 14), leading to poor quality PSG. Afterward, when this problem occurred for the second time (no. 19), she called the patient and could resolve the problems by phone. In the same way, overflow problems with Medatec SleepWeb were unexpected. These problems reflect the “infancy diseases” of an innovative technologic interface. The Medatec Company is currently working to face and improve the Sleepbox® System. The potential advantages of home-PSG with telemonitoring are multiple and include cost saving, by avoiding hospitalization for PSG and by preventing repeated sleep studies for failure. Campos et al. described similar findings using remote video-EEG consultation, allowing important cost savings by avoiding travel and lost of time related to consultations [20]. The use of complete PSG (Type 2 PM [12]) is also full of interest because diagnosis of other sleep disorders than OSA is possible, in the usual environment of the patient. And indeed, in addition to OSA diagnosis, we were able to diagnose epilepsy and periodic leg movements in our small population, what should not be the case when using Type 3 PM (unattended portable multi-channel sleep testing device) [12]. Despite the very exciting aspect of this study, we have to emphasize some limitations: the small size of the study (not surprising for a pilot study) leads to the need for further larger investigations to confirm the efficiency of the device, after technical improvements. In the future, we have also to compare telemonitored home-PSG with a control group (unattended home-PSG or attended in lab-PSG), in order to quantify quality improvement and cost-savings. Regarding costs, we can suppose, however, that the additional cost of 3G network connection, technician visits to patient’s home for hook-up and education will remain lower than the costs related to hospitalization. Regarding nurses’ extra-work, telemonitoring did not result in additional costs in our current work scheme but this element has also to be considered. Another possible problem, even if it was not remarkable in the present study, is the potential inability of the patient to replace correctly the probes when a problem is detected. Adequate patient education is necessary to minimize this problem. As telemedicine and assistive technology, including telemonitoring, teleassistance, and teleconsultation [21–23], is a part of the medicine of the future, it is really exciting to use novel technologies to enhance diagnostic means, diagnostic tools in order to contribute to increase healthcare accessibility.
Authors’ contribution Marie Bruyneel and Vincent Ninane wrote the scientific project. Marie Bruyneel and Sandra Van den Broecke collected and analyzed the data. Walter Libert collected the data. All the authors participated to manuscript redaction.
i n t e r n a t i o n a l j o u r n a l o f m e d i c a l i n f o r m a t i c s 8 2 ( 2 0 1 3 ) 696–701
Summary points What was already known on the topic: [7]
• Home-polysomnography (HPSG) is useful for OSA diagnosis but the failure rate of the recordings remains problematic. • When the quality of the recordings is bad, it leads to less cost-savings than expected. • Unattended polysomnography is better than polygraphy because it allows a wide range of diagnosis in sleep disorders.
[8]
[9]
[10]
What this study added to our knowledge: [11]
• Real-time telemonitored HPSG is feasible. • It could be an interesting perspective to decrease the failure rate of home PSG. • As it is an innovatice device, we faced technical problems that we need to improve. • However, 90% of the recordings were of excellent quality.
[12]
[13]
[14]
Conflict of interest None declared.
[15]
Acknowledgements We thank warmly Mr Driessens, Mr Karmoun and all the staff of MEDATEC Company, Brussels, Belgium, for their precious and enthusiastic collaboration. We also thank l’ “Association André Vésale-Aide à la recherche médicale” for their financial backing.
references
[1] A.G. Ekeland, A. Bowes, S. Flottorp, Effectiveness of telemedicine: a systematic review of reviews, Int. J. Med. Inform. 79 (2010) 736–771. [2] V. Nangalia, D.R. Prytherch, G.B. Smith, Health technology assessment review: remote monitoring of vital signs-current status and future challenges, Crit. Care 14 (2010) 233. [3] T. Raza, M. Joshi, R.M. Schapira, Z. Agha, Pulmonary telemedicine – a model to access the subspecialist services in undeserved rural areas, Int. J. Med. Inform. 78 (2009) 53–59. [4] M.E. Dyken, V.K. Somers, T. Yamada, Z.Y. Ren, M.B. Zimmerman, Investigating the relationship between stroke and obstructive sleep apnea, Stroke 27 (1996) 401–407. [5] L.J. Findley, M.E. Unverzagt, P.M. Suratt, Automobile accidents involving patients with obstructive sleep apnea, Am. Rev. Respir. Dis. 138 (1988) 337–340. [6] J.M. Marin, S.J. Carrizo, E. Vicente, A.G. Agusti, Long-term cardiovascular outcomes in men with obstructive sleep
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
701
apnoea–hypopnoea with or without treatment with continuous positive airway pressure: an observational study, Lancet 365 (2005) 1046–1053. J. Hung, E.G. Whitford, R.W. Parsons, D.R. Hillman, Association of sleep apnoea with myocardial infarction in men, Lancet 336 (1990) 261–264. T. Young, M. Palta, J. Dempsey, J. Skatrud, S. Weber, S. Badr, The occurrence of sleep-disordered breathing among middle-aged adults, N. Engl. J. Med. 328 (1993) 1230–1235. R.L. Riha, W.T. McNicholas, The genetic cardiovascular aspects of obstructive sleep apnoea/hypopnea syndrome, J. Eur. Respir. 33 (2009) 233–236. J.F. Masa-Jimenez, F. Barbé Illa, F. Capote Gil, E. Chiner Vives, Resources and delays in the diagnosis of sleep apnea–hypopnea syndrome, Arch. Bronconeumol. 43 (2007) 188–198. N.A. Collop, Clinical guidelines for the use of unattended portable monitors in the diagnosis of obstructive sleep apnea in adult patients, J. Clin. Sleep Med. 3 (2007) 737–747. Standards of Practice Committee of the American Sleep Disorders Association, Practice parameters for the use of portable recording in the assessment of obstructive sleep apnea, Sleep 26 (1994) 907–913. M. Bruyneel, C. Sanida, G. Art, W. Libert, L. Cuvelier, M. Paesmans, et al., Sleep efficiency during sleep studies: results of a prospective study comparing home-based and in-hospital polysomnography, J. Sleep Res. 20 (2011) 201–206. J.F. Masa, J. Corral, R. Pereira, J. Duran-Cantolla, M. Cabello, L. Hernández-Blasco, et al., Effectiveness of home respiratory polygraphy for the diagnosis of sleep apnoea and hypopnoea syndrome, Thorax 66 (2011) 567–573. F. Portier, A. Portmann, P. Czernichow, L. Vascaut, E. Devin, D. Benhamou, et al., Evaluation of home versus laboratory polysomnography in the diagnosis of sleep apnea syndrome, Am. J. Respir. Crit. Care Med. 162 (2000) 814–818. A. Rechtschaffen, A. Kales, A Manual of Standardised Terminology, Techniques and Scoring for Sleep Stages in Human Subjects, Brain Information Service, Los Angeles, 1968. INAMI/RIZIV, Convention relative au diagnostic et au traitement du syndrome des apnées obstructives du sommeil. www.inami.fgov.be (accessed 20.02.12). M. Bonnet, D. Carley, M. Carskadon, P. Easton, C. Guilleminault, R. Harper, et al., EEG arousals: scoring rules and examples: a preliminary report from the sleep disorders atlas task force of the American Sleep Disorders Association, Sleep 15 (1992) 173–184. S. Redline, M.H. Sanders, B.K. Lind, S.F. Quan, C. Iber, D.J. Gottlieb, et al., Methods for obtaining and analyzing unattended polysomnography data for a multicenter study, Sleep 21 (1998) 759–767. C. Campos, E. Caudevilla, A. Alesanco, N. Lasierra, O. Martinez, J. Fernandez, et al., Setting up a telemedicine service for remote real-time video-EEG consultation in La Rioja (Spain), Int. J. Med. Inform. 81 (2012) 404–414. M. Vitacca, L. Bianchi, A. Guerra, C. Fracchia, A. Spanevello, B. Balbi, et al., Tele-assistance in chronic respiratory failure patients: a randomised clinical trial, Eur. Respir. J. 33 (2009) 411–418. E. Biratti, A. Roth, Telecardiology, Isr. Med. Assoc. J. 13 (2011) 498–503. M. Bruyneel, W. Libert, V. Ninane, Detection of bed-exit events using a new wireless bed monitoring assistance, Int. J. Med. Inform. 80 (2011) 127–132.