Comparison of a behavioural test to assess daytime sleepiness with the traditional MWT
Standard objective tests of daytime sleepiness include the MSLT and maintenance of wakefulness test (MWT). These require continual EEG monitoring and are cumbersome and expensive. We have devised a simple test of 'wakefulness' based on a behavioural response to an intermittently illuminating light emitting diode (LED). This study reports this tests ability to discriminate subjects with severe symptomatic OSA from normals and compares this result with the traditional EEG based MWT. 10 subjects (7M 3F) with severe OSA (>4% SaO2 dip rate 33.5 (SD 19.7)) and symptoms of daytime sleepiness (Epworth Sleepiness Score (ESS) 17 (5.1)) and 10 normal subjects (4M 6F, ESS 4.4(2.5)) were studied. The EEG MWT and the behavioural LED test were performed on each subject in random order on 2 separate days. Both tests included 4x40 minute sleep resistance challenges at 2 hourly intervals. Tests were performed under the same conditions while sound isolated in a darkened room. During the behavioural test, instead of EEG monitoring, subjects were asked to press a switch in response to a LED regularly illuminating for 1 second in 3. When there was no response for 21 seconds the test was terminated. Both tests effectively discriminated the normals from the sleepy OSA subjects. The mean sleep latency during the behavioural test was longer than for the EEG MWT and this difference was significant in the OSA group but not in the normal group (see below). Retrospective analysis showed there was no difference in these results if sleep onset was defined as no response for 15 seconds instead of 21. Mean Sleep latency in minutes (SD) Difference between EEG MWT and behavioural test EEG MWT LED test Mean(SD) Paired t test Normal subjects 38.1 (2.8) 39.8 (0.6) 1.7 (3.0) p > 0.1 OSA patients 7.3 (3.7) 10.5 (3.9) 3.2 (4.2) p < 0.05 The behavioural test discriminated normal subjects from sleepy patients as well as the MWT and was simpler to administer. It has the advantage that sleep onset is defined objectively as a failure to respond rather than from subjective EEG interpretation. This technique may provide a simple method of objectively quantifying daytime sleepiness.