1
Christian Emil FaberPh.D. thesis
Flextube reflectometry – a new method for determination of sites of upper airway narrowing during sleep
Ph. D. thesis
Christian Emil Faber
Faculty of Health Sciences
University of Aarhus
Department of Otorhinolaryngology
1
Christian Emil FaberPh.D. thesis
Preface
This Ph.D. thesis is a result of my Ph.D. study as a clinical assistant at the ENT Department, Aarhus University Hospital and as a student at the Faculty of Health sciences, University of Aarhus from 1999 to 2002. The Ph.D. thesis is based on the following 4 papers, which will be referred to by their roman numerals and an overview.
IFaber CE, Grymer L, Norregaard O, Hilberg O. Flextube reflectometry for localization of upper airway narrowing - a preliminary study in models and awake subjects. Respir Med 95(8) 631-38, 2001
IIFaber CE, Hilberg O, Jensen FT, Norregaard O, Grymer L. Flextube reflectometry for determination of sites of upper airway narrowing in sleeping obstructive sleep apnoea patients. Respir Med 95(8) 639-48, 2001
IIIFaber CE, Grymer L, Hilberg O, Norregaard O. Flextube reflectometry and pressure-recordings for level diagnosis in obstructive sleep apnoea. Rhinology 40(4) 203-10, 2002
IVFaber CE, Hilberg O, Grymer L. Flextube reflectometry for level diagnosis in patients with obstructive sleep apnoea and snoring. Rhinology 40 (3) 122-28, 2002
My interest in the field of sleep research was initiated during clinical work with OSAHS patients and snorers at the ENT departments at Aarhus University Hospital, Vejle Hospital and Odense University Hospital.
Many individuals have contributed to make this Ph.D. study process possible, for which I am most grateful. First and foremost I would like to express my gratitude to my 4 supervisors Luisa Grymer MD, Ole Norregaard MD, Christian Brahe Pedersen professor, MD and Ph.D. and Ole Hilberg MD and PhD for their guidance and support.
The idea for this study sprang from a new application of the acoustic reflection method invented by Steen Rasmussen. He is the founder and managing director ofRhinoMetrics A/S to whom I owe a special thank for developing the flextube reflectometry system on which this thesis is based and for constructive advice and many fruitful discussions.
I am particularly grateful for the help from secretary Jette Breiner and Marianne Gadegaard and the bioanalyticists Annette Tillebech and Kirsten B. Andersen. I owe my thanks to physicist Finn T. Jensen for advice and counselling during the MRI studies and to Niels Trolle, statistician, for statistical guidance. I also wish to acknowledge the participating patients and healthy subjects who served as volunteers and the staff at the ENT department and Respirationscenter Vest at Aarhus University Hospital and the staff at RhinoMetrics A/S especially Jan Stavngaard.
I am grateful for the financial assistance provided by: Erhvervsfremme Styrelsen, Aarhus Kommunehospital´s Forskningsfond, Aarhus Amt, Oticon Fonden, and A.P.Møllers Fond.
My most sincere and deepest gratitude goes to my family, especially to my wife Annette.
Aarhus, 2003
Christian E. Faber
1
Christian Emil FaberPh.D. thesis
DEFINITIONS AND ABBREVIATIONS
AHI Apnoea-hypopnoea index = number of apnoeas and hypopnoeas per hour of sleep
AIApnoea index = number of apnoeas per hour of sleep
Apnoea Cessation of airflow for 10 seconds or more.
BMI Body mass index (=Weight in kilograms/(height in meters) 2)
Central apnoea Cessation of airflow and respiratory muscle activity with a duration of 10 seconds or more.
Cephalogram Standardised lateral radiographic projection of the head and neck
CI Confidence interval
Confounding Key problem in epidemiological studies when the observed association is not causal but mediated through a third independent factor
Compliance Expression of the ability of yielding to pressure without disruption of an air or fluid filled organ – in terms of unit of volume change per unit of pressure change.
CSA Cross sectional area
CT Computerised tomography
dB Decibel
DSP Digital signal processor
EDSExcessive daytime sleepiness
EEG Electroencephalogram
EEG arousal An abrupt shift in EEG frequency during sleep, 3 seconds or longer in duration. May occur without increase in EMG amplitude during NREM sleep, but must be accompanied by concurrent increases in EMG in REM sleep.
EMG Electromyogram
EOG Electrooculogram
ESS Epworth Sleepiness Scale
FNMM Fiberoptic nasopharyngoscopy with the Muller manoeuvre
HypopnoeaReduced airflow for at least 10 seconds. Can be defined as a 30% drop in the thermistor signal associated with a 3% desaturation. These criteria vary in different studies
MCAMinimal cross-sectional area
Mixed apnoea Cessation of airflow with a duration of 10 seconds or more. The event starts with a cessation of respiratory muscle activity followed by respiratory effort.
MRI Magnetic resonance imaging
MSLT Multiple Sleep Latency Test. The mean time required from lights out to the first epoch of any sleep stage determined for each of 4 or 5 daytime ”nap opportunities”. Method for objective assessment of sleepiness.
Muller manoeuvre Manoeuvre where the patient is asked to close the mouth and inhale vigorously while the examiner occludes the nostrils. The degree of collapse at the retropalatal and retrolingual levels is recorded by inspection with a fiberoptic endoscope
MWTThe maintenance of wakefulness test (MWT). The patients are asked to remain awake in a quiet darkened room and they are then monitored for electroencephalographic sleep onset.
NCPAP Nasal continuous positive airway pressure. A mechanical device for treatment of OSAHS by the application of positive air pressure to the pharynx via a nose-mask
Non-REM sleep Sleep without rapid eye movements. Non-REM sleep is classified into stages 1-4 with increasing depth
Obstructive apnoea Cessation of airflow lasting 10 seconds or more while the respiratory muscles are still working.
OSA Obstructive sleep apnoea
OSAHS Obstructive sleep apnoea hypopnoea syndrome
PcritCritical pressure. Objective measure of collapsibility of the pharynx. If the nasal pressure drops below this pressure, the upper airway collapses, and airflow ceases.
PSGPolysomnography
RDI Respiratory disturbance index (apnoeas and hypopnoeas per hour recording)
REM sleep Sleep with rapid eye movements
ROC curveReceiver operator characteristic curve. A graph of the pairs of true positive rates (sensitivity) and false positive rates (100% - specificity) that correspond to each possible cut-off for the diagnostic test result. The diagnostic test whose ROC curve encloses the largest area (below and to the right of the curve) is the most accurate one.
SaO2Arterial oxyhemoglobin saturation
SD Standard deviation
Somnofluoroscopy Lateral fluoroscopic examination of the upper airway with synchronous PSG.
SPL Sound pressure level
UARS Upper Airway Resistance Syndrome. Chronic daytime sleepiness and fragmentation of sleep by EEG arousals related to an abnormal increase in respiratory effort during sleep due to increased upper airway resistance.
UPPP Uvulopalatopharyngoplasty
zCharacteristic impedance. The unit is Newton-sec/m3
Contents
Preface......
DEFINITIONS AND ABBREVIATIONS......
Contents......
List of figures......
Introduction......
Aims of the thesis......
Background......
Sleep related breathing disorders - definitions......
Snoring......
Apnoeas and hypopnoeas......
The obstructive sleep apnoea-hypopnoea syndrome......
The Upper Airway Resistance Syndrome......
Pathophysiology......
Critical pressure......
Contributing factors to nocturnal upper airway narrowing......
Anatomic narrowing of the upper airway......
Increased collapsibility of the tissues......
Muscular imbalance......
Sleep position and sleep stages......
Gender......
Age......
Obesity......
Family history......
Alcohol......
Benzodiazepines......
Smoking......
Other contributing factors......
Effects of upper airway narrowing......
Hypertension......
Mortality......
Coronary disease and stroke......
Diabetes mellitus......
Road traffic accidents......
Cognitive impairment......
Reduced quality of life......
Prevalence of OSAHS and snoring......
Denmark......
Internationally......
Diagnostic methods......
Measurement of daytime sleepiness......
Polysomnography......
Measurement of snoring......
Level diagnosis......
Visual inspection and indirect laryngoscopy......
Fiberoptic nasopharyngoscopy......
Upper airway pressure recordings......
Cephalometric radiograms......
Fluoroscopy......
Computerised tomography......
MR scanning......
Acoustic reflections principle......
Acoustic reflections using a mouthpiece
Acoustic reflections using a nosepiece......
Treatment......
Nasal continuous positive airway pressure......
Surgery......
Oral appliances......
Weight reduction......
Behaviour modification......
Medication......
Cardiac Pacing......
Materials and methods......
Design......
Reliability in vitro and comparison with endoscopy during wakefulness......
The frequency of respiratory disturbances and reproducibility......
Comparison with other methods for determination of obstructive level......
The influence of the flextube on sleep and the inconvenience of the flextube......
The average level distribution and the association with BMI and RDI......
Study population......
Study I......
Study II......
Study III......
Study IV......
Methods......
Flextube reflectometry......
Acoustic rhinometry
Questionnaires......
Polysomnography......
Autoset......
Upper airway pressure recordings by micro transducers......
Fiberoptic nasopharyngoscopy with the Muller manoeuvre......
Fiberoptic nasopharyngoscopy without the Muller Manoeuvre......
MR scanning......
Statistical methods......
Study I......
Study II......
Study III......
Study IV......
results......
Reliability in vitro and comparison with endoscopy during wakefulness......
The frequency of respiratory disturbances and reproducibility......
Comparison with other methods for determination of the obstructive level......
The influence of the flextube on sleep and the inconvenience of the flextube......
The average level distribution and the association with BMI and RDI......
Discussion of the materials and methods......
Design......
Study population......
Flextube reflectometry......
Polysomnography......
AutoSet®
Upper airway pressure recordings by micro transducers......
Fiberoptic nasopharyngoscopy......
MR scanning......
Data collection and data checking......
Discussion of the results......
Reliability in vitro and comparison with endoscopy during wakefulness......
The frequency of respiratory disturbances and reproducibility......
Comparison with other methods for determination of obstructive level......
The influence of the flextube on sleep and the inconvenience from the flextube......
The average level distribution and the association with BMI and RDI......
Suggestions for future research......
Larger validation studies......
Technical development......
Titration of NCPAP......
Characteristics of respiratory disturbances......
Level of obstructions and severity of OSAHS......
Upper airway resistance syndrome......
Infant studies......
Determination of usefulness......
Summary......
Summary in danish......
References......
List of figures
Figure 1: The Epworth Sleepiness Scale.......
Figure 2: A graphic illustration of the basic principle of acoustic sound reflections.......
Figure 3:The flextube reflectometry system.......
Figure 4: The average flextube narrowing in an OSAHS patient.......
Figure 5: The frequency of flextube narrowings and respiratory disturbances.......
Figure 6: Reproducibility of the level distribution determined by flextube reflectometry.......
Figure 7:Level distribution determined by flextube reflectometry compared with MRI.......
Figure 8: Level distribution by the flextube reflectometry method and the ApneaGraph®......
Figure 9: Frequency of respiratory disturbanceswith and without a flextube inserted.......
1
Christian Emil FaberPh.D. thesis
Introduction
The obstructive sleep apnoea hypopnoea syndrome (OSAHS) is characterised by periodic reduction (hypopnoea) or cessation (apnoea) of airflow during sleep. It results from upper airway narrowing or collapse. The syndrome is associated with disrupted sleep and snoring. Patients are often referred to health care professionals through the concern of bed partners or parents who have witnessed apnoeas followed by choking and loud snoring.
Prevalence studies estimate that 4% of middle-aged men and 2% of middle-aged women are affected by OSAHS (1,2).
A link between OSAHS and cardiac arrhythmias, ischaemic heart disease, cardiac failure, hypertension, stroke, daytime sleepiness, and road traffic accidents has been suggested (3).
Current diagnostic testing with polysomnography (PSG) provides information regarding the number of apnoeas and hypopnoeas per hour sleep, but PSG does not provide any information regarding the level of pharyngeal narrowing or collapse. A number of investigative methods have been used in order to determine the level of obstructive predominance in OSAHS: lateralcephalograms, awake endoscopy, awake endoscopy with the Muller manoeuvre, endoscopy during sleep, endoscopy with NCPAP during sleep, fluoroscopy, CT scanning, MR scanning, and manometry. Data from different studies using various methods suggest that different patients have different patterns of narrowing or collapse of the pharynx. Some studies have suggested that surgery following identification of the specific area of obstruction is more likely to be successful than surgery carried out without knowledge of the obstructive site (4-6).
A recent Cochrane Library Document suggested that further research should be undertaken to identify and standardise techniques to identify the obstructive sites in the upper airway of patients with OSAHS. This would allow evaluation of selected surgical procedures through randomised controlled trials (7).
Aims of the thesis
Four studies (study I, II, III, and IV) were carried out. The aims of the present thesis were to
Assess the reliability in vitro of acoustic reflectometry measurements using a flextube and compare flextube measurement with endoscopy during wakefulnessby
Determination of the maximum difference between actual and acoustically measured duration of occlusions of the flextube in vitro. Determination of the maximum difference between actual and acoustically measured distances from the 0 point of occlusions of the flextube in vitro. (Study I)
Comparing the actual level (retropalatal or retrolingual or both) of narrowings made in a model with assessments made by blinded observers who studied the resulting acoustic graphs. (Study I)
Finding the correlation between the degree of flextube narrowing, and the degree of cross-sectional area decrease assessed by endoscopy during the Muller manoeuvre, and the correlation between the degree of flextube narrowing, and the magnitude of the median pressure difference (Po-Patm) during the Muller manoeuvres. (Study I)
Compare the flextube device with other sleep recording devices for determination of the frequency of respiratory disturbances, and estimate the reproducibility of flextube measurements by
Determination of the correlation between the number of flextube narrowings per hour of sleep and the number of obstructive apnoeas and hypopnoeas per hour of sleep found using PSG in subjects investigated simultaneously by both methods, and by assessing whether the mean difference between the 2 data sets was statistically significantly different from 0. (Study II)
Calculating the correlation between the number of flextube narrowings per hour recording and the respiratory disturbance index (RDI = apnoeas and hypopnoeas per hour recording) determined by the upper airway pressure-recording device (ApneaGraph®), in subjects investigated on 2 different nights, and by assessing whether the mean difference between the 2 data sets was statistically significantly different from 0. (Study III)
Calculating the correlation between the number of flextube narrowings per hour recording and the RDIs per hour recording determined by AutoSet®, and by assessing whether the mean difference between the 2 data sets was statistically significantly different from 0. (Study IV)
Assessing whether the mean difference between the percentages of retropalatal flextube narrowing on 2 different nights in the same patients was statistically significantly different from 0. (Study II)
Compare the predominant obstructive level found using flextube reflectometry with the predominant obstructive level found using other methodsby
Assessing whether the mean difference between the percentage of retropalatal flextube narrowing of the total narrowing (retropalatal and retrolingual) and the percentage of retropalatal obstructions found by MRI on 2 different nights was statistically significantly different from 0. (Study II)
Calculating the correlation between the percentage of retropalatal narrowing found using flextube reflectometry and the percentage of retropalatal obstructions found using pressure recordings on 2 different nights. (Study III)
Assessing the agreement between the levels of maximum narrowing found using fiberoptic endoscopy and using flextube reflectometry. (Study IV)
Determine the influence of the flextube on sleep quality and on the frequency of respiratory disturbances and the inconvenience associated with use of the flextube by
Determination of the magnitude of the influence of the flextube on the results of PSG: total sleep time, sleep efficiency, obstructive apnoeas per hour of sleep, central apnoeas per hour of sleep, mixed apnoeas per hour of sleep, hypopnoeas per hour of sleep, AHI, EEG arousals per hour of sleep, REM latency, sleep stage distribution, and minimum SaO2. (Study II)
Calculating the correlation between the RDIs found using the AutoSet® system at baseline (home recordings without flextube or pressure catheter inserted) and the RDIs found by the AutoSet® system in hospital with a flextube inserted and by assessing whether the mean difference between the 2 data sets was statistically significantly different from 0. (Study III and IV)
Assessing the inconvenience associated with the flextube reflectometry procedures and the frequency of complications, including bleeding or mucosal tears. (Study IV)
Find the average level distribution in patients referred for OSAHS and/or snoring and calculate the association between level distribution and BMI and RDIby
Calculating the mean (standard deviation = SD) percentage of retropalatal flextube narrowing of the total narrowing (retropalatal and retrolingual) during obstructiveevents for consecutive patients referred for snoring and/or OSAHS and by determination of the proportion of referred patients who had predominantly “retropalatal” and “retrolingual” flextube narrowings and “no predominant obstructive level”. (Study IV)
Assessing whether the percentage of retrolingual flextube narrowings was related to BMI and RDI measured by AutoSet® at home without a flextube inserted. (Study IV)
Background
The following review will describe the existing knowledge and some of the unsolved problems regarding the definitions and the pathophysiology of snoring and OSAHS, the contributing factors to nocturnal upper airway narrowing, the effects of upper airway narrowing, the diagnostic methods used in OSAHS, methods used for diagnosis of obstructive levels in the pharynx, the prevalence of OSAHS and snoring, and possible treatments.
Sleep related breathing disorders - definitions
Snoring
Snoring is an inspiratory sound generated by vibrations of partially collapsed pharyngeal airway walls and of the soft palate as demonstrated by lateral fluoroscopy (8). “Simple snorers” are subjects with intermittent partial upper airway obstruction giving rise to snoring without sleep fragmentation and without daytime symptoms.
Apnoeas and hypopnoeas
An apnoea is defined as a cessation of airflow for at least 10 seconds. The definition of hypopnoeas varies. It is often defined as a reduction (from discernible to > 50 %) from baseline in the amplitude of airflow, and there is often a requirement forsimultaneous oxygen desaturation and/or arousal to fulfil the criteria. The event is required to last 10 seconds or longer (9). Theapnoea-hypopnoea index (AHI) is the average number of apnoeas plus hypopnoeas observed per hour of sleep. The respiratory disturbance index (RDI) is the average number of apnoeas and hypopnoeas per hour recording.