Data Loading...
Obstructive Sleep Apnoea - University of Sydney Flipbook PDF
Obstructive Sleep Apnoea and Driver Performance: Prevalence, Correlates, and Implications for Driver Fatigue Anup Desai
140 Views
7 Downloads
FLIP PDF 262.58KB
Obstructive Sleep Apnoea and Driver Performance: Prevalence, Correlates, and Implications for Driver Fatigue
Anup Desai M.B., B.S. FRACP
A thesis submitted to the Faculty of Medicine, University of Sydney, in fulfillment of the requirements for the degree of Doctor of Philosophy
Woolcock Institute of Medical Research Royal Prince Alfred Hospital
December 2002
University of Sydney
Statement
The work described in this thesis was carried out in the Woolcock Institute of Medical Research and Royal Prince Alfred Hospital under the supervision of A/Professors R. Grunstein and G. Marks. Unless otherwise stated, it is the original work of the author and has neither been presented nor is it currently being presented for any other degree.
Anup Desai
Acknowledgements and Dedication
Acknowledgements Firstly, I would like to thank Associate Professor Ron Grunstein for his constant assistance and support with this huge undertaking. From the early days when we struggled with recruitment issues, to the last months of repeated thesis revisions, he has always provided good advice, ample help, and an impressive knowledge of the area. In addition to being involved in the minutia of every day protocols and activities, he has always also held a broader perspective of my work, which has helped to guide this thesis to its timely completion.
I would also like to thank Associate Professor Guy Marks, my co-supervisor, who in particular provided excellent guidance in statistical analysis and interpretation. Through his direction, I feel I have also learnt a lot about scientific writing and methodology.
I would like to thank the late Professor Ann Woolcock, whose input and encouragement at the very start of this thesis I now appreciate even more. It was her intervention in particular, that helped me understand the importance of undertaking and formalizing research of this nature into the body of work that it is now. I have known Ann Woolcock since my childhood, and I have always been aware of how important research is to her. I am sure she would be very happy and hopefully also very proud to see this thesis in its final form. In addition, I am grateful to her for introducing me to the Woolcock Institute of Medical Research (formally known as the Institute of Respiratory Medicine), a world
class research facility that has provided me with a supportive and friendly working environment from the start. Particular thanks to Victoria Keena, information manager, Anthony Williams, research officer, Courtney Rogers, librarian, and Susan ForrestBlythe, research assistant.
There are so many others to thank in a work of this size. I would like to thank Dr. Mark Howard, for his assistance with the design and running of the Commercial Drivers Study, and Dr. David Joffe, for his assistance with recruitment, access to Royal North Shore Hospital, and advice on neurocognitive testing. Dr. David Jankelson and Dr. David Barnes also assisted with recruitment of research subjects and I thank them for this. Dr. Delwyn Bartlett has helped me throughout this work with advice, knowledge, and assistance with testing, and for all this I am very grateful to her. Dr. Michael Barnett generously agreed to proof read part of this work, and I thank him for his excellent advice.
Gunnar Ungar’s assistance with computer and data management has been invaluable. Benjamin Harris and Jonathon Williamson assisted tremendously with testing at Royal North Shore Hospital, while Brad Wilsmore, Qiao Yang and James Newcombe all assisted with testing at Royal Prince Alfred Hospital. I thank Brad Wilsmore in particular for his good cheer when assisting with difficult statistics. I would like to thank my scorers, Jane Downey and Adele Bailey for their hard work, my night sleep technician Andrew Dong for his constant work and dependability, the staff at Royal Prince Alfred Hospital and Royal North Shore Hospital for their assistance with the research protocols,
all my research subjects for their time and patience, and finally, the NSW Transport Workers Union, for their help with the Commercial Drivers Study.
On a more personal note, I would like to thank my parents, for their encouragement and confidence in my abilities, which has helped to empower me with the commitment to finish such a substantive body of work, but most of all, I would like to thank my wife, Ankita, who has supported me unconditionally throughout the long process of this thesis. In particular, she has had to share the first year of our marriage with the rigorous demands of this thesis. This thesis is as much hers, as it is mine.
Dedication I would like to dedicate this thesis to my beautiful wife, Ankita
Table of Contents STATEMENT ................................................................................................................... 2 ACKNOWLEDGEMENTS AND DEDICATION......................................................... 3 TABLE OF CONTENTS ................................................................................................. 6 INDEX OF TABLES AND FIGURES.......................................................................... 11 ABBREVIATIONS AND DEFINITIONS.................................................................... 15 SUMMARY ..................................................................................................................... 19 AIMS ................................................................................................................................ 22 CHAPTER 1: BACKGROUND .................................................................................... 24 1.1 DRIVER FATIGUE ...................................................................................................... 24 1.1.1 Importance of Driver Fatigue ........................................................................... 24 1.1.2 Economic Costs of Fatigue Related Accidents.................................................. 26 1.2 FACTORS INFLUENCING DRIVER FATIGUE ................................................................ 31 1.2.1 Sleep Loss or Sleep Deprivation ....................................................................... 32 1.2.2 Circadian Timing of Drive ................................................................................ 41 1.2.3 Sleep Disorders: Obstructive Sleep Apnoea ..................................................... 47 1.2.4 Length of Driving without Rest ......................................................................... 76 1.2.5 Alcohol Consumption ........................................................................................ 77 1.3 ADDITIVE FATIGUE EFFECTS .................................................................................... 79 1.3.1 Additive Effects of Mild OSA and Other Fatigue Promoting Factors .............. 80
1.3.2 The Effect of Sleep Deprivation on Obstructive Sleep Apnoea......................... 81 CHAPTER 2: METHODS ............................................................................................. 84 2.1 COMMERCIAL DRIVERS STUDY ................................................................................ 84 2.1.1 Project Design................................................................................................... 84 2.1.2 Subjects.............................................................................................................. 85 2.1.3 Protocol............................................................................................................. 88 2.1.4 Data Collection Instruments ............................................................................. 90 2.1.5 Statistical Analysis .......................................................................................... 108 2.2 MILD OBSTRUCTIVE SLEEP APNOEA STUDY .......................................................... 111 2.2.1 Project Design................................................................................................. 111 2.2.2 Subjects............................................................................................................ 112 2.2.3 Protocol........................................................................................................... 113 2.2.4 Data Collection Instruments ........................................................................... 117 2.2.5 Statistical Analysis .......................................................................................... 128 2.3 MEDICO-LEGAL CASE SERIES ................................................................................ 130 2.3.1 Polysomnography............................................................................................ 130 2.3.2 Objective Tests of Daytime Sleepiness ............................................................ 130 CHAPTER 3: RESULTS: PREVALENCE OF OSA................................................ 132 3.1 RESPONSE RATES .................................................................................................... 132 3.1.1 Drivers surveyed in the field ........................................................................... 132 3.1.2 Drivers undergoing sleep studies.................................................................... 134 3.2 DEMOGRAPHIC AND PHYSICAL CHARACTERISTICS ................................................. 135 3.2.1 Physical Characteristics ................................................................................. 135
3.2.2 Driving Characteristics 1................................................................................ 136 3.2.3 Driving Characteristics 2................................................................................ 137 3.2.4 Medical Health................................................................................................ 138 3.2.5 Accident rates.................................................................................................. 139 3.3 PREVALENCE OF DAYTIME SLEEPINESS .................................................................. 140 3.4 PREVALENCE OF OBSTRUCTIVE SLEEP APNOEA ..................................................... 142 3.4.1 Prevalence of OSA and Sleep Apnoea Syndrome............................................ 142 3.4.2 Analysis of Accuracy of Maislin Prediction of OSA ....................................... 144 CHAPTER 4: RESULTS: CORRELATES OF OSA................................................ 147 4.1 RELATIONSHIP OF OBSTRUCTIVE SLEEP APNOEA TO DRIVER SLEEPINESS ............. 147 4.2 RELATIONSHIP OF OSA TO SELF-REPORTED ACCIDENT RATES ............................. 148 4.2.1 Drivers Surveyed in the Field.......................................................................... 148 4.2.2 Drivers Studied in the Sleep Laboratory......................................................... 150 4.3 RELATIONSHIP OF OSA TO DRIVING IMPAIRMENT (DRIVING SIMULATOR) ............ 151 4.3.1 Obstructive Sleep Apnoea ............................................................................... 152 4.3.2 Sleep Apnoea Syndrome .................................................................................. 154 4.4 REGRESSION MODELS............................................................................................. 159 4.4.1 Factors Predicting Driver Sleepiness ............................................................. 159 4.4.2 Factors Predicting Reported Accidents .......................................................... 161 4.4.3 Factors Predicting Driving Impairment (Driving Simulator)......................... 162 4.4.4 Factors Predicting Obstructive Sleep Apnoea ................................................ 163 4.5 EFFECT OF OSA/SLEEPINESS ON VIGILANCE (PSYCHOMOTOR VIGILANCE TASK).. 164 4.6 EFFECT OF OSA/SLEEPINESS ON NEUROPSYCHOLOGICAL MEASURES ..................... 166
4.7 SUMMARY OF RESULTS FOR CHAPTERS 3 AND 4..................................................... 167 CHAPTER 5: RESULTS: INTERACTIVE EFFECTS OF MILD OSA ................ 172 5.1 DEMOGRAPHIC AND PHYSICAL MEASURES............................................................. 172 5.2 LIFESTYLE DETAILS................................................................................................ 172 5.3 DRIVER DETAILS .................................................................................................... 174 5.4 BASELINE POLYSOMNOGRAPHY ............................................................................. 174 5.5 DAYTIME DRIVING PERFORMANCE AND COGNITIVE FUNCTION OUTCOMES .......... 175 5.5.1 Effect of Time of Day and Sleep Deprivation.................................................. 175 5.5.2 Effect of Obstructive Sleep Apnoea Status ...................................................... 181 5.5.3 Impact of Sleep Deprivation on Diurnal Variation Differed with OSA Status 181 5.6 NIGHT TIME DRIVING PERFORMANCE AND COGNITIVE FUNCTION OUTCOMES ...... 183 5.6.1 Effect of Time of Day....................................................................................... 183 5.6.2 Effect of OSA Status ........................................................................................ 186 5.7 EFFECT OF SLEEP DEPRIVATION ON POLYSOMNOGRAPHIC MEASURES ON OSA .... 186 CHAPTER 6: RESULTS: CASE SERIES ................................................................. 188 6.1 CASES ..................................................................................................................... 188 Case A: under-treated OSA and undiagnosed Periodic Limb Movement Disorder 188 Case B: undiagnosed mild OSA and upper airway resistance syndrome ................ 189 Case C: “sleep attack” in a private driver .............................................................. 189 Case D: undiagnosed OSA in a truck driver............................................................ 190 Case E: undiagnosed idiopathic hypersomnolence in a truck driver ...................... 190 Case F: undiagnosed OSA in a truck driver ............................................................ 191 Case G: untreated OSA in a truck driver................................................................. 192
CHAPTER 7: DISCUSSION ....................................................................................... 194 7.1 COMMERCIAL DRIVERS STUDY .............................................................................. 194 7.2 MILD OBSTRUCTIVE SLEEP APNOEA STUDY .......................................................... 203 7.3 MEDICO-LEGAL CASE SERIES ................................................................................ 210 7.4 FINAL SUMMARY AND IMPLICATIONS ..................................................................... 215 CHAPTER 8: PUBLICATIONS ................................................................................. 217 8.1 PAPERS ................................................................................................................... 217 8.2 ABSTRACTS ............................................................................................................ 217 REFERENCES.............................................................................................................. 219 ATTACHMENTS ......................................................................................................... 244
Index of Tables and Figures
Table 1 Site visits for drivers surveyed in the field ........................................................ 133 Table 2 Age and obesity: comparison between subjects in the field study and laboratory study........................................................................................................................ 135 Table 3 Age & obesity in laboratory study: comparison of participants and nonparticipants.............................................................................................................. 136 Table 4 Driving characteristics of participants in the field study and the laboratory study ................................................................................................................................. 137 Table 5 Sleep and other habits of drivers in the field study and laboratory study.......... 138 Table 6 Medical health of drivers in the field study and laboratory study ..................... 139 Table 7 Self-reported accidents for field and laboratory drivers .................................... 140 Table 8 Relation between OSA and pathological sleepiness in drivers in the field study ................................................................................................................................. 142 Table 9 Prevalence of OSA for drivers studied in laboratory......................................... 143 Table 10 Prevalence of sleep apnoea syndrome for drivers studied in laboratory ......... 144 Table 11 Likelihood ratios for predicting OSA for Maislin score quartiles ................... 146 Table 12 Prevalence of sleep apnoea syndrome for drivers in the field and laboratory. 147 Table 13 Comparison of accident rates between drivers with and without sleep apnoea syndrome................................................................................................................. 149 Table 14 OSA (RDI) in drivers with and without a history of accidents ....................... 150 Table 15 AusEd mean reaction time by OSA status....................................................... 153 Table 16 Effect of sleep apnoea syndrome on driving simulator performance .............. 155
11
Table 17 Predictor variables for driver sleepiness.......................................................... 159 Table 18 Predictors of excessive driver sleepiness......................................................... 160 Table 19 Driving simulator variables partially explained by RDI.................................. 162 Table 20 Driving simulator variables explained by sleep apnoea syndrome.................. 163 Table 21 Predictor variables of OSA .............................................................................. 164 Table 22 Effect of sleep apnoea syndrome on PVT performance, adjusted for diurnal variation .................................................................................................................. 166 Table 23 Demographic & physical measures: comparison between OSA & control groups ................................................................................................................................. 172 Table 24 Lifestyle details: comparison between OSA and control groups..................... 173 Table 25 Caffeine intake and sleep hours immediately prior to testing weekends......... 173 Table 26: Driver details: comparison between OSA and control groups ....................... 174 Table 27 Baseline polysomnography: comparison between controls and OSA subjects 175 Table 28 Effect of sleep deprivation on outcomes (adjusted for diurnal variation) ....... 176 Table 29 Effect of time of day on outcomes (adjusted for intervention)........................ 179 Table 30 Impact of sleep deprivation on diurnal variation differed with OSA status .... 182 Table 31 Effect of time of day on outcomes in the sleep deprived state (adjusted for OSA status) ...................................................................................................................... 185 Table 32 Effect of sleep deprivation on PSG measures (adjusted for group)................. 187 Table 33 Case details of fall asleep road fatalities.......................................................... 192
12
Figure 1 Temporal profile of mean (standard error of mean) PVT lapses after baseline nights (B1, B2), after each of seven consecutive sleep restriction nights (P1-P7), and after a recovery night of sleep (R1) (Dinges et al. 1997) ......................................... 36 Figure 2 Performance in the sustained wakefulness condition expressed as mean relative performance and the percentage blood alcohol concentration equivalent (Dawson and Reid 1997).......................................................................................................... 38 Figure 3 Standard deviation of speed across the day during control and sleep deprivation conditions (+/- SE) (Lenne et al. 1998) .................................................................... 39 Figure 4 Driving simulator reaction time (RT) across time of day (Lenne et al. 1997)... 43 Figure 5 The risk of heavy lorry accidents in Sweden (Kecklund and Akerstedt 1994) .. 47 Figure 6 Prevalence of OSA and sleep apnoea syndrome in middle-aged men (Bearpark et al. 1995;Young et al. 1993) .................................................................................. 50 Figure 7 Effect of CPAP treatment on accidents (George 2001)...................................... 66 Figure 8 Distribution of accidents in patients with OSA (George 2001) ......................... 66 Figure 9 AusEdTM driving simulator view........................................................................ 95 Figure 10 The road layout of AusEdTM............................................................................. 98 Figure 11: Flow chart of study protocol ......................................................................... 115 Figure 12 Accident rate of drivers in the 2 study groups................................................ 140 Figure 13 Driver sleepiness: drivers surveyed in the field ............................................. 141 Figure 14 Driver sleepiness: drivers studied in the laboratory ....................................... 141 Figure 15 ROC curve: Maislin prediction of OSA ......................................................... 146 Figure 16 Correlation between driver sleepiness and OSA ............................................ 148
13
Figure 17 Accidents compared to RDI for drivers studied in the laboratory................. 151 Figure 18 AusEd mean reaction time by OSA status ..................................................... 153 Figure 19 AusEd mean reaction times for drivers with and without sleep apnoea syndrome................................................................................................................. 157 Figure 20 Tracking ability in drivers with and without sleep apnoea syndrome............ 158 Figure 21 Impact of sleep deprivation on diurnal variation differed with OSA status... 183
14
Abbreviations and Definitions
General AI:
arousal index (see section 2.1.4.5.2)
AHI:
apnoea hypopnoea index (see section 1.2.3.1)
CI:
confidence interval
CPAP:
continuous positive airway pressure (see section 1.2.3.7)
EDS:
excessive daytime sleepiness
ESS:
Epworth sleepiness scale (see section 2.1.4.1.2)
Effect Size:
describes how many standard deviation units the experimental group was different from the control group; for reference, an effect size of 0.20 is considered small, 0.50 is considered medium, and 0.80 or greater is considered to be large (Hedges and Olkin 1985).
MSLT:
multiple sleep latency test, a polysomnographic test of sleep tendency, that requires subjects to try to fall asleep during four or five daytime nap opportunities (see section 2.3.2). (Association of Sleep Disorders Centers Task Force on Daytime Sleepiness 1986)
MVA:
motor vehicle accident
MWT:
maintenance of wakefulness test, a polysomnographic test of sleep tendency, where subjects are asked to stay awake under soporific conditions several times during the day for 20 or 40 minutes at a time (see section 2.3.2) (Mitler et al. 1982)
NAB:
neurobehavioural assessment battery (see section 2.2.4.4)
15
OSA:
obstructive sleep apnoea (see section 1.2.3.1)
OSLER:
Oxford sleep resistance test (see section 2.2.4.6)
PLMs:
periodic limb movements, where subjects show repeated stereotyped movements, usually of the lower limbs, in certain stages of their sleep (O'Keeffe 1996).
PSG:
polysomnography (see section 2.1.4.5)
PVT:
psychomotor vigilance task (see section 2.1.4.3)
RDI:
respiratory disturbance index (see section 1.2.3.1)
REM:
rapid eye movement sleep
SAS:
sleep apnoea syndrome, clinical definition requiring the presence of both OSA on polysomnography, together with excessive daytime sleepiness (see section 1.2.3.1)
SDB:
sleep disordered breathing
SD:
sleep deprivation
AusEdTM - Australia Edinburgh Driving Simulator Steering Variables: Median lane position (cm):
median lane position adopted on the road for the duration of the simulator drive
Sd lane position median (cm):
standard deviation (sd) of lane position from the median lane position
Area steer dev median (cm ms):
area of steering deviation (steer dev) from the median lane position
16
Av steer dev median (cm):
average (av) steering deviation (steer dev) from the median lane position
Area dev steering centre (cm ms): area of steering deviation (dev) from the centre of the left hand lane Av dev steering centre (cm):
average (av) steering deviation (dev) from the centre of the left hand lane
Sd of lane position (cm):
standard deviation (sd) of mean lane position
Reaction Time Variables: RT mean (ms):
mean reaction time (RT) for braking after the sudden appearance of trucks
RT sd mean (ms):
standard deviation (sd) from the mean reaction time (RT) for braking
RT median (ms):
median reaction time (RT) for braking
RT sd median (ms):
standard deviation (sd) from the median reaction time (RT) for braking
RT 95% percentile (ms):
ninety fifth percentile of reaction times (RT) for braking
Speed Variables: Area speed dev zone (km/h ms):
area of speed deviation (dev) outside the 60-80 km/hr zone
Av speed dev zone (km/h):
average (av) speed deviation (dev) outside the 6080 km/hr zone
17
Crash Variable: Crashes:
number of crashes of all types during the duration of the simulator drive
Psychomotor Vigilance Task (PVT) PVT median (ms):
median reaction time for the trial
Lapses:
number of reaction times at least 500ms in duration. The lapse count can be transformed (square root (x) + square root (x+1)) for analysis to remove zero scores and normalise the distribution
Fast 10% (ms):
mean optimum response domain for the trial (mean of fastest 10% of responses)
1/RT slow 10% (1/ms):
this represents the worst (slowest) response domain for the trial (reciprocal of mean of slowest 10% of responses)
Errors:
number of false-starts on depressing the button, the use of the 'wrong' button, or holding the button down for more than three seconds
18
Summary
Obstructive sleep apnoea (OSA) is characterised by repetitive reductions or pauses in breathing during sleep due to upper airway narrowing or closure. Due to disruption to normal sleep patterns, many patients with OSA suffer from increased daytime sleepiness. Epidemiological studies have established a link between OSA and driver fatigue and accidents, generally showing a two to seven times increased risk of road traffic accidents in non-commercial drivers with OSA. There is emerging evidence that commercial drivers have a higher prevalence of OSA than the general population, being predominately male, middle-aged and overweight, three important risk factors for OSA. However, little is known about the relationship between OSA and driver sleepiness in commercial drivers, whether road accidents are increased in commercial drivers with OSA, and whether OSA interacts with other fatigue promoting factors, such as sleep deprivation, to further escalate road accident risk.
One thousand randomly selected commercial drivers were surveyed in the field. In addition, 61 randomly selected NSW commercial drivers had in hospital sleep studies and daytime performance testing, including a PC based driving simulator task. The prevalence of OSA, defined as Respiratory Disturbance Index (RDI) > 10, was approximately 50% in NSW commercial drivers. Approximately one quarter of the drivers reported pathological daytime sleepiness, and 12-14% had both OSA and pathological daytime sleepiness. A diagnosis of OSA was the most important factor predicting excessive daytime sleepiness in these drivers: OSA was more important than
19
15 other work-related, lifestyle and medical factors that could be expected to promote, or be associated with, daytime sleepiness. Drivers with sleep apnoea syndrome (both OSA and pathological daytime sleepiness) had an increased driving accident risk, using driving simulator and daytime performance testing as proxy measures for accident risk. These results demonstrate the importance of OSA as a cause of driver fatigue in commercial drivers and suggest that all commercial drivers should be screened for the presence of sleep apnoea syndrome in order to potentially reduce road accident risk through treatment.
A separate, but related body of work examined the combined effects of mild OSA and other fatigue promoting factors (sleep deprivation and circadian influences) on driving performance. Twenty nine subjects, consisting of a group with mild OSA and a group of non-OSA controls, were tested on several occasions throughout the night and day using an intensive performance battery, under both baseline conditions and after a period of 36 hours of total sleep deprivation. The results suggest that drivers with mild OSA are not different to the control group in their response to sleep deprivation or time of day influences. However, the subjects with mild OSA were less aware of their impairment due to sleep deprivation, which is of concern if drivers with OSA are relying on their subjective awareness of fatigue to make decisions about when to stop driving.
A final perspective on OSA and driver fatigue is provided through a clinical case series of seven fall-asleep fatality associated MVA’s associated with unrecognised or undertreated sleep disorders. As well as demonstrating the day to day potential for devastating
20
road accidents due, at least in part, to un-recognised or untreated sleep disorders, these cases also serve to highlight some of the current medico-legal controversies and difficulties in this area of driver fatigue.
In conclusion, this body of work has provided novel information about the epidemiology and implications of OSA in commercial drivers, and about how OSA interacts with other fatigue promoting factors. Finally, it has explored some of the medico-legal issues that relate to sleep disorders and driver fatigue. As well as providing much needed information in the area of driver fatigue, at the same time this work raises many more questions and suggests areas of future research. For instance, such research should examine the relationship between objective accident rates and OSA/sleep apnoea syndrome in commercial drivers, the interaction between mild sleep apnoea syndrome and other fatigue risk factors, and driver perception of sleepiness prior to sleep onset in drivers with sleep disorders.
21