The search for electrophysiological biomarkers of vigilance failure in OSA

Obstructive sleep apnoea (OSA) is rising in prevalence and is linked
with driving impairment and elevated risk motor vehicle accidents
(MVA). However, daytime sleepiness and neurobehavioral/vigilance
impairment varies markedly between OSA patients. Accident risk may
be confined to a subset of patients who are more susceptible to vigilance
failure. Identification of OSA patients at risk of vigilance failure and
MVAs is critical but currently challenging since routine clinical metrics
of OSA severity and sleepiness have poor predictive value. New
neurobehavioral and electrophysiological tests/measures are needed to
more accurately identify the “sleepy” OSA phenotype. Ideally these tests
need to be simple, inexpensive and clinically deployable.
Currently the EEG and ECG signals routinely collected during overnight
sleep studies are only superficially analysed by manual scoring,
potentially missing a lot of disease specific effects on the brain. Quantitative
power spectral analysis automatically quantifies these signals at
much higher resolution allowing for detection of specific spectral frequency
signatures that may help to explain the variation in
neurobehavioral function and the risk of vigilance failure on OSA.
Previous small studies from our group suggest spectral EEG markers
derived from sleep and wake EEG, as well as evoked related potentials
(ERPs) correlate with driving simulator impairment in OSA patients
following extended wakefulness. There also appears to be a phenotypic
difference in heart rate variability (HRV) in healthy subjects who are
vulnerable vs resistant to sleep loss. These findings will be presented and discussed and future OSA neurobehavioral and electrophysiological
phenotyping approaches discussed. Future laboratory based and large
accident risk validation studies will be crucial before useful markers of
vigilance failure are translated into routine clinical practice.