The Severity Of Individual Breathing Cessation Events In Diagnostics Of Obstructive Sleep Apnea

Obstructive sleep apnea (OSA) is a prevalent nocturnal breathing disorder characterized by complete (apnea) and partial (hypopnea) breathing cessations during sleep 1. OSA is connected to cardiovascular disease, increased mortality rate and impairment of quality of life 2-4. Its prevalence has been estimated to be 9-27 % among middle-aged population and the prevalence is expected to increase in the future 5,6. Currently, the most commonly used diagnostic parameter for OSA is apnea-hypopnea index (AHI) based on number of apneas and hypopneas during sleep, neglecting their durations and also durations and depths of related desaturation events. In adults OSA is diagnosed if patient has 1) AHI≥5 events/h with associated symptoms (e.g. excessive daytime sleepiness (EDS)) or medical or psychiatric disorder (e.g. hypertension or mood disorder) or 2) AHI≥15 events/h 7. Previously, the investigators have introduced novel parameters called as obstruction severity, obstruction duration, desaturation severity, desaturation duration, and adjusted-AHI incorporating number, duration and morphology of individual obstruction events (i.e. apneas, hypopneas, and oxygen desaturations) 8,9.we have shown that these parameters can enhance the severity estimation of OSA and that the individual breathing cessation event severity is more strongly connected to increased mortality and cardiovascular morbidities than conventional AHI8,10,11. The potential of these parameters to enhance the assessment of severity of OSA has been previously investigated only using ambulatory polygraphic recordings without EEG registration. Therefore, the parameters do not take into account hypopneas followed by arousal (but not desaturation). In addition, parameters do not take into account apneas not followed by desaturation and novel parameters need to be refined to take into account also such events in order to further increase their prognostic value. It is known that age and body mass index (BMI) are positively related with the prevalence of OSA12,13. Furthermore, AHI is higher and apneas longer in supine position compared to lateral position14-16. However, it is not thoroughly explored whether the severity of individual obstruction events progress over time and which factors affect to this progression. EDS is a common symptom of OSA and it can be measured in various methods. The most widely used methods are Epworth Sleepiness Scale (ESS) questionnaire which measures subjective sleepiness and Multiple Sleep Latency Test (MSLT) and Maintenance of Wakefulness Test (MWT) which measure sleepiness objectively. However, the correlation between ESS score and the severity of OSA (i.e. number of obstruction events) have been shown to be poor 17. Furthermore, albeit MSLT and MWT have been shown to have better correlations with daytime functions than ESS their correlations with AHI and arousal index still remainsweak18. Therefore, it needs to be investigated whether the severity of individual breathing cessation events is correlated better with MSLT and MWT and thus, explaining with more accuracy the daytime symptoms. It has been shown that between patients with identical AHI values, durations of individual apneas and hypopneas and durations, depths, and areas of individual desaturations can differ significantly 19. Longer and deeper obstruction events could be considered to be more harmful than shorter and shallower events as they are linked to increased mortality rate in patients with moderate or severe OSA 8. Nonetheless, it has not been thoroughly investigated whether in severe OSA patients with identical AHI values sleep efficiency or hypertension is related to the severity of individual breathing cessation events and if this relationship is modulated by sleeping position. Positional therapy (PT), the avoidance of the supine posture during sleep is the treatment of choice for Positional Patients (PP) who have only or most of the apneas/hypopneas while sleeping in the supine posture 22. Non positional patients (NPP) have many apneas/hypopneas while sleeping in both, the supine and the lateral posture. For these patients Continuous Positive Airway Pressure (CPAP) is the treatment of choice, however, many patients have strong difficulties to adhere to CPAP and thus, the investigators would like to assess the therapeutic value of PT for these NNP. Since apneas/hypopneas are more severe while sleeping in the supine than in the lateral posture.15, 16 PT for these patients could represent a valuable treatment. This is an issue that the investigators also want to assess in this project. In this research project the investigators aim to tackle these shortcomings in the diagnostics of OSA. The main aim is to improve the severity estimation of OSA by further refining the novel parameters for diagnostics of OSA. These improvements to the parameters will be done in order to prevent the harmful consequences of the disease and to be able to target the limited treatment resources to those in the greatest need. 2\. HYPOTHESES AND AIMS In this research project the investigators tackle the limitations of AHI and refine the novel parameters introduced in our earlier papers. To achieve this, four aims will be undertaken. The first aim is to explore, whether the severity of individual breathing cessations progresses over time and how different confounding factors (e.g. BMI, sleeping position, and gender) affect this progression. The investigators believe that individual event characteristics have a major role in the overall severity of OSA. The current knowledge on the progression of OSA is limited to number of events while the detailed changes in the individual event characteristics has not been thoroughly explored before. This is a major shortcoming as event severity can be considered clinically significant and have been associated with the risk of OSA related mortality and cardiovascular morbidity. The second aim is to improve our recently developed novel parameters to provide even more accurate severity estimation by including more detailed analysis of hypopnea events followed by arousal and apnea events that do not induce desaturations. This more detailed analysis can enhance the estimation of OSA related mortality and incidence of comorbidities and improve the discrimination of patients needed to be treated but it can also help clinicians to choose best individualized treatment method for each patient. The third aim of the research project is to evaluate the correlation of excessive daytime sleepiness (EDS) with the individual breathing cessation event severity, sleep structure, and frequency and occurrence of cortical arousals. As EDS is a major problem reducing the quality of life, finding effective measures to recognize factors behind EDS could aid to enhance the treatment of these patients and improvement of their life quality. The fourth aim is to explore whether the percentage time of disturbed breathing from total sleep time is related to sleep efficiency or hypertension in severe OSA patients having similar AHI values and whether this percentage alters depending on confounding factors (e.g. BMI, hypertension, and proportion of supine sleep). More detailed characterization of severe OSA patients could aid the clinicians to recognize the patients with most urgent need for effective treatment. The fifth aim is to assess the therapeutic value of positional therapy (PT) for severe NPP. This assessment is of interest since many severe OSA patients that should be treated with CPAP are not been treated due to the difficulties to adhere to CPAP treatment. This behavioral treatment could represent a partial solution until these patients lose weight, consider surgery (ENT or bariatric) or the use of dental appliance or give to CPAP another chance. It is important to note that by losing weight NPP may become PP 23 and thus, PT could represent no a partial but a more radical solution. 2\. RESEARCH PLAN AND METHODS The planned research includes five work packages designed to enable reaching the project aims. Work package I: Progression of individual obstruction events over time - retrospective follow-up study In this package,we will explore how the duration of individual apnea and hypopnea events and duration, depth and area of individual desaturation events progress over time. It will be investigated how different confounding factors (e.g. age, BMI, sleeping position and gender) affect the progression of the individual breathing cessation event severity. This package is based on polysomnographic (PSG) recordings of about 400 patients with suspected OSA recorded with Embla (Natus Medical Incorporated, Pleasanton, CA, USA) or Rembrandt (Natus Medical Incorporated, Pleasanton, CA, USA) devices and analyzed using RemLogic software at the Sleep Disorders Unit, Lowenstein Hospital - Rehabilitation Center, Raanana, Israel. For all patients, two - three PSG recordings were conducted at different time points, at baseline, and after 3, 5, 10 years or more. These polygraphic recordings were analyzed in conformity with the standard respiratory rules specified by AASM and based on clinical practice in Lowenstein Hospital at the time of analysis. Based on the analyzed recordings, the severity of individual obstruction events will be calculated using custom-made MATLAB (Matlab R2014b, MathWorks Inc., Natick, MA, USA) functions. Severity of individual obstruction events will be compared between the different time points for each patient and in a group level. Grouping of the patients will be done based on the confounding factors (e.g. patients' BMI, gender, sleeping position, AHI, and age). Work package II: Optimization of the diagnostic parameters of OSA which incorporates the severity of individual breathing cessation events In this package, the novel parameters will be further refined by adding features to also include apneas not followed by desaturations and hypopneas followed by arousal (but not desaturation). Furthermore, the aim is to evaluate whether apneas should be given more weight than hypopneas while estimating the overall severity of OSA. In this package, in-laboratory PSG recordings (n=7540) including EEG conducted and analyzed at the Sleep Disorders Unit, Lowenstein Hospital - Rehabilitation Center for the patients with clinical suspicion of OSA will be studied. The basic anthropometric data and information on mortality and co-morbidities will be also collected. All recordings were analyzed according the standard respiratory rules defined by AASM and based on clinical practice in Lowenstein Hospital. Novel parameters and severity of individual obstruction events will be calculated using custom-made MATLAB functions. The calculated severity indexes based on enhanced and the original obstruction severity parameters will be compared with incidence of OSA related mortality and co-morbidities. The information on mortality, cause of death and co-morbidities hopefully will be obtained from the Ministry of Health. Work package III: Effect of severity of individual obstruction events on excessive daytime sleepiness In this package, the investigators investigate the correlation between the severity of individual breathing cessation events and excessive daytime sleepiness measured with objective means (i.e. Multiple Sleep Latency Test (MSLT) and Maintenance of Wakefulness Test (MWT)). Furthermore, it will be studied which information obtained from PSG recordings (e.g. frequency and duration of cortical arousal, sleep structure or severity of individual obstruction events) is best correlated with excessive daytime sleepiness and therefore, would be beneficial to take into account while estimating the severity of OSA. Clinical in-laboratory PSG recordings (n=3000) of patients with suspected OSA, conducted and analyzed at the Sleep Disorders Unit, Lowenstein Hospital - Rehabilitation Center with the Embla devices and RemLogic software, will be used. The values of the novel parameters and the severity of individual obstruction events will be calculated using custom-made MATLAB functions and their correlation on results of objective tests measuring excessive daytime sleepiness (i.e. MSLT and MWT) will be investigated. Work package IV: Differences in severity of individual obstruction events between patients having severe OSA In this package, the severe OSA patients will be matched by AHI and the percentage time of disturbed breathing from total sleep time will be compared between these patients. The patient population consists of 463 patients diagnosed to have severe OSA after full overnight PSG recording conducted at the Sleep Disorders Unit, Lowenstein Hospital - Rehabilitation Center. It will be studied whether the percentage time of disturbed breathing from total sleep time differs between the patients having similar AHI and whether it is related to different outcome measures (e.g. hypertension and sleep efficiency). Furthermore, it will be investigated whether this percentage is modulated by different confounding factors (e.g. gender, BMI, and sleeping position). Work package V: The assessment of the therapeutic value of Positional Therapy for severe Non Positional Patients (NPP) In this package the investigators we will assess the therapeutic value of Positional Therapy (PT) i.e., the avoidance of the supine posture during sleep, as a treatment alternative for severe Non Positional Patients (NPP) who failed to adhere to Continuous Positive Airway Pressure (CPAP). The patient population consists of the same 463 patients diagnosed to have severe OSA after full overnight PSG recording conducted at the Sleep Disorders Unit, Loewenstein Hospital - Rehabilitation Center. In this group of patients the investigators will assess the effect of sleeping only in the lateral posture (PT) and the assessment of the severity of the disease will be calculated according to conventional and new methods developed previously by our research group.