FEASIBILITY OF NON-LABORATORY ELECTROPHYSIOLOGICAL RECORDING OF SMALL VIGILANCE FLUCTUATIONS

Regina Conradt, Ulrich Brandenburg, Thomas Penzel, Jörg Hermann Peter
Schlafmedizinisches Labor, Innere Medizin der Universität Marburg, Germany

INTRODUCTION Daytime Sleepiness and impaired vigilance are among the most important symptoms of Obstructive Sleep Apnea (OSA). Nasal CPAP is a common therapy for OSA. Polygraphy, including electroencephalogram (EEG), electrooculogram (EOG) and electromyogram (EMG) is considered to be one of the most reliable methods of assessing whether a person is alert, drowsy or sleepy. The sleep stage classification of Rechtschaffen and Kales is almost exclusively used for sleep stage scoring. It does not provide substages for the description of drowsiness with rapid electrophysiological changes. Moreover, the epoch length of 30 sec which is used for whole- night records is too long for the description of short micro-sleep episodes or fast vigilance fluctuations. When dealing with electrophysiologic activity with regard to vigilance and reactivity, not only more vigilance substages (here 10) and shorter scoring epochs are useful but also an adaptive segmentation. That means, when a change takes place at the electrophysiological activity the current epoch is ended and a new one started.

METHOD In 6 patients diagnosed with OSA vigilance was tested using a 90 minute, four choice reaction time test (RTT) before and with nCPAP-therapy. Reaction times (RT), electrophysiological signals (4 EEGs, 2 EOGs, EMG chin and ECG) and oronasal airflow were recorded digitally by EMBLATN (Flaga hf. Medical Devices, Island), an ambulatory data recorder and stored on a Power Macintosh 7200. Each of the available 16 channels has the same physical properties and the configuration can be altered to meet different data acquisition requirements for different signals and sensors. The gain of the amplifier is kept at a low value while the necessary accuracy is digitally achieved. All the 16 analog inputs are simultaneously sampled and depend on the input signal sampling rates of 200Hz (ECG), 100Hz (EEG, EOG, EMG), 10 Hz (Airflow, RT-signal), or 50, 20 and 1 Hz are possible. All data were recorded in real time, and were continuously stored in a data file on computer. For ambulatory monitoring, a PCMCIA slot for PC Cards in EMBLATN is available. Rechargeable Lithium-Ion batteries give a recording time of around 24 hours, varying somewhat depending on the sampling rate and number of channels being recorded.

RESULTS With nCPAP therapy the mean RT decreased significantly to 1.03 (SD±0.44) sec compared to the pretreatment reaction time of 2.30(±0.30) sec. In contrast, each patient showed a marked increased vigilance with nCPAP therapy compared to the pretreatment investigation. Percentages of drowsy substages decreased from 41.5 (33.4) % in the pretreatment group to 8.76 (±21.43) % with nCPAP therapy.

CONCLUSION Ambulatory monitoring systems with electrophysiological parameter become more importance. The systems have to be easy to handle and the quality of recording must be of a high standard. This study shows the feasibility of recording very small vigilance fluctuations which need high recording quality for every electrophysiological signal. Our study suggests that the EMBLA would be a useful tool and its advantages include a lower cost of recording and more patients` comfort.