In-laboratory polysomnography (PSG) is the gold standard for the diagnosis of obstructive sleep apnea (OSA).1 However, this technique is expensive and time-consuming.2 This condition added to the increasing prevalence of the disease explains the development of new systems for the diagnosis of OSA.3,4

Sleepwise (SW) is a non-invasive technology based on video image processing, by recording the patient's respiratory movements to transform them into a breathing signal that permits to determine episodes of hypopnea and apnea. Its technology is based on the principle that the volume of air that circulates into the lungs is proportional to the thoracic movement that a subject presents while breathing. Furthermore, according to the analysis of the body movements SW can also infer sleep/awake periods.

In a previous publication our group compared the diagnostic accuracy of SW and the in-laboratory polysomnography, concluding that Sleepwise determined the diagnosis and the severity of OSA with high reliability.5 We hypothesized that SW testing could also be performed at home. Therefore, the main objective of this study was to evaluate the diagnostic accuracy of SW compared with a home sleep apnea test (HSAT). As secondary objectives we checked SW reproducibility in different nights and also rated the easiness and comfort of its use.

Thirty-eight patients were included in the study and their clinical data are shown in Tables 1 and 2. Agreements between procedures are shown in Table 3. The graphical representation of this agreement, described with the Bland–Altman method can be seen in Figs. 1–3.

Fig. 1.

Bland–Altman plot for polysomnography (PSG) and Sleepwise (SW) apnea–hypopnea index (AHI). Differences between PSG and SW seem to be larger in patients with AHI over 60.

(0.07MB).

Fig. 2.

Bland–Altman plot for home sleep apnea test (HSAT) and Sleepwise (SW) apnea–hypopnea index (AHI).

(0.06MB).

Fig. 3.

Bland–Altman plot for two different Sleepwise (SW) procedures apnea–hypopnea index (AHI).

(0.06MB).

There was almost perfect agreement between PSG and SW, with a Lin's concordance correlation coefficient of 0.933 and a κ of 0.930 (95% CI 0.867–0.993).

Between HSAT and SW, agreement was lower, with a Lin's concordance correlation coefficient of 0.917 and a κ of 0.793 (95% CI 0.713–0.872).

Agreement between two different SW procedures was again almost perfect, with a Lin's concordance correlation coefficient of 0.923 and a κ of 0.837 (95% CI 0.729–0.944).

None of the participants reported difficulties in understanding the functioning of SW. The device was considered easy-to-use and comfortable. All the recordings had an adequate quality for their proper analysis.

In this present study we found a very high correlation between SW and the gold-standard methods for OSA diagnosis both in-laboratory and in-home, as well as a robust reproducibility of the test. These findings are in line with our previous study, in which we described a good agreement between PSG and SW.5 Therefore, we propose SW as an innovative and reliable system for the in-home diagnosis of OSA.

In-laboratory, technician-attended PSG monitoring at least sleep stages and respiration, is accepted as the gold standard for the diagnosis of OSA.8 However, this procedure requires technical expertise, is labor-intensive and time-consuming and together with the high prevalence of OSA and the great demand of examinations results in a problem of long waiting lists. Reliable diagnostic systems that short cut the conventional test are thus a must. Portable monitors are alternative approaches to diagnosis providing an equivalent diagnosis to in-laboratory PSG.3 SW is not just a portable in-home device but more importantly it is also a non-invasive system. SW does not use any sensor attached to patient's body which implies several advantages compared to other systems for the diagnosis of OSA. First, it is more comfortable for the subject as SW records patient's sleep without any nuisance and then the quality of sleep is therefore probably improved. SW can detect respiratory movements independently of the position while sleeping or even if the patient is covered by a blanket. Second, it is easier to use as patients can be able to set up the system themselves at home with brief instructions. All participants in our study rated the instructions to set up SW as easy or very easy and expressed their satisfaction to the system comfort. The fact we excluded participants presenting difficulties for the understanding of the set-up of the system could be a selection bias. However, these patients would not have likely been candidates for a HSAT either.

One of the limitations of HSAT is the absence of sleep staging evaluation that provides a denominator for the AHI. This type of portable monitor provides breathing events quantified per hour of monitoring time as a respiratory disturbance index (RDI) that can underestimate the severity of OSA. Regarding this point, SW offers an undeniable advantage over the rest of the currently available portable diagnostic systems. One SW's limitation is the absence of cardiorespiratory data registration, such as oxygen saturation, pulse or cardiac rhythm that helps physicians to determine the severity of the apneas and hypopneas and their systemic repercussion. It is technically possible to add the information in question, but that would lead to a drawback of the system ceasing to be noninvasive. However, future SW versions would incorporate these variables obtained in a non-invasive and friendly system, such as it has been agreed with experts as fundamental for the evaluation of OSA.

Although it is difficult to compare two different procedures by simply comparing the kappa index, our findings suggest that SW could be more reliable than HSAT to define OSA's severity. The explanation for this fact is that SW calculates breathing events based on the time of sleep and not the monitoring time.

However, we found 3 cases with substantial differences between AHI value determined by PSG and SW. These concrete patients were severe OSA with a BMI over 30 and this findings concur with our previous study.5 SW and PSG presented larger differences in the AHI value in obese patients with severe disease although patients were properly classified as severe OSA by both techniques, thus without implications in the therapeutic decision in any case.

HSAT use is accepted for the diagnosis of OSA in uncomplicated adult patients presenting with signs and symptoms that indicate an increased risk of moderate to severe OSA. On the contrary, polysomnography should be elected in patients with significant cardiorespiratory disease, neuromuscular disease with respiratory impairment, suspicion of hypoventilation, opioid medication use, history of stroke, or severe insomnia.4 Participants of this study did not presented any major comorbidity to contraindicate a HSAT. In the future it would be interesting to compare SW agreement with PSG in complicated adult patients.

The AASM recommends that the raw data from the HSAT devices must be reviewed and interpreted by a physician who is either board certified in sleep medicine or overseen by a board certified sleep medicine physician.9 Although in this study we provided the results of automatic analysis by SW, the system also allows to review the raw data manually.

Even though reproducibility of the two SW procedures was considered to be almost perfect, kappa values were slightly lower than that of the comparison between PSG and SW. The reason for this could be the fact that those procedures were performed in two different nights, with presumably different AHI due to a night-to-night variability. This error could be solved in the future by performing a study comparing two simultaneous SW recordings overnight on the same individual. This would be interesting to evaluate the system repeatability which is currently assumed to be high according to our results.

Other authors have developed various techniques based on motion analysis for diagnosis of OSA, but there are significant differences when compared to SW from the video analysis of breathing to the detection of respiratory events.10–13 All this studies were performed in an experimental, restricted environment with control subjects, but never tested in real life or compared to other diagnostic techniques. A strength of our study lies in demonstrating the utility and applicability of SW as an in-home diagnosis system of OSAS while previous studies have always been developed in a laboratory environment.

The benefits presented by SW lead us to hypothesize that it could be a useful tool to study pediatric patients. We have analyzed some pediatric patients but preliminary results have not reached the same level of perfect agreement found in the study with adults. The most probable explanation is that children's respiratory mechanics and body size are different from adults and thus the same algorithm cannot be applied. Further studies are needed to adapt the algorithm and improve the effectiveness of the device in this concrete population group.

Economic analyses have compared the cost-effectiveness of management pathways that incorporate diagnostic strategies using HSAT or PSG.14–16 Contrary to what one might think, all have concluded that PSG is the preferred diagnostic strategy from an economic perspective for adults suspected to have moderate to severe OSA. The reason for that is the favorable cost-effectiveness of OSA treatment in this group of patients particularly when longer time horizons are considered. HSAT could lead to increase false negatives, and so leave patients untreated, or increase false positives and consequently treat unnecessarily patients. The impact of these errors can be magnified when extrapolated over long time horizons. The HomePAP study concluded that for payers, a home-based diagnostic pathway for OSA with robust patient support incurs fewer costs than a laboratory-based pathway, but for providers, costs are comparable if not higher, resulting in a negative operating margin.17 In that sense, our previous results comparing SW to PSG demonstrated a sensitivity of a 100%, specificity of 87%, a positive predictive value of 97% and a negative predictive value of 100% for the diagnosis of OSA.5 According to that, we believe SW would be a more cost-effective system compared to available HSAT adding also that not consumable material is required and all the automatic analysis is performed digitally. However, this is a hypothesis to be confirmed in further specific cost-effective studies.

In conclusion, SW was found to be highly accurate for non-invasive and automatic diagnosis of OSA in outpatients when compared to standard methods for OSA diagnosis both in-laboratory and in-home. SW proved to be a technique with repeatable and concordant results on different nights for the same patient and resulted easy to set and very comfortable for the patients. According to these results, we conclude SW is a remarkable non-invasive, easy-to-use, portable, effective and highly accurate option for OSA diagnosis at patients’ home.

The research reported in this publication was supported by a grant from the Spanish Society of Pulmonary Medicine and Thoracic Surgery (SEPAR) with code 030/2015. Esteve-Teijin Healthcare, S.L collaborated disinterestedly with our project instructing the participants in the set-up of both the respiratory polygraph and the camera for the Sleepwise recording for the in-home tests.

All authors have contributed to the conception and design of the study; analysis and interpretation of data; and revising the article critically for important intellectual content. IAV and CF collected the data. IGO and JA performed the statistical analysis and interpreted the results. AMF wrote the manuscript. All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors read and approved the final manuscript.

MAC is Chief Executive Officer (CEO) of Tarsio Medics, S.L, Malgrat de Mar (Barcelona), Spain. AMF, JRM and JA are minority shareholders of Tarsio Medics, S.L, Malgrat de Mar (Barcelona), Spain since July 2017 and serve as clinical advisors for this company. The terms of this arrangement have been reviewed and approved by the Fundació Institut d’Investigació en Ciències de la Salut Germans Trias i Pujol (IGTP) in accordance with its policy on objectivity in research. IGO, IV, CF and AR declare that there is no conflict of interests regarding the publication of this paper.