BACKGROUND: The qualitative presence of microsleep during the multiple sleep latency test (MSLT) has been shown to correlate with an increased incidence of subjective complaints of sleepiness, tiredness, accidents/near accidents, and gap driving. However, there is no data on how to quantify microsleep and effectively incorporate it as a diagnostic tool in the measurement of sleepiness. The purpose of this study was to integrate microsleep with the MSLT score and determine if it improved the correlation between the MSLT and symptomatic sleepiness. METHODS: The charts of 54 patients who had an MSLT score of greater than 5min and the presence of microsleep on at least one nap were reviewed. If microsleep was present in a given nap it was used as a surrogate for sleep onset. This MSLT plus microsleep score (MSL-M) was then averaged into the total sleep latency and compared to the MSLT score to determine if it improves correlation with the Epworth sleepiness scale (ESS). A microsleep nap percentage (MNP) was also obtained and correlated with ESS to determine if a better association could be derived. RESULTS: Using the Spearman correlation the MSL-M improved the correlation with the ESS when compared to MSLT (r=0.106 versus r=0.063), but the results were not statistically significant. Of note, both the MSLT and MSL-M were only weakly correlated to the ESS. The MNP also did not have a good correlation with ESS (r=-0.099). CONCLUSIONS: The addition of microsleep onset to the MSLT score as a quantitative assessment tool failed to significantly enhance the correlation between subjective and objective accounts of sleepiness, beyond the improvement seen in the MSLT value by the simple presence of microsleep alone.
BACKGROUND: The qualitative presence of microsleep during the multiple sleep latency test (MSLT) has been shown to correlate with an increased incidence of subjective complaints of sleepiness, tiredness, accidents/near accidents, and gap driving. However, there is no data on how to quantify microsleep and effectively incorporate it as a diagnostic tool in the measurement of sleepiness. The purpose of this study was to integrate microsleep with the MSLT score and determine if it improved the correlation between the MSLT and symptomatic sleepiness. METHODS: The charts of 54 patients who had an MSLT score of greater than 5min and the presence of microsleep on at least one nap were reviewed. If microsleep was present in a given nap it was used as a surrogate for sleep onset. This MSLT plus microsleep score (MSL-M) was then averaged into the total sleep latency and compared to the MSLT score to determine if it improves correlation with the Epworth sleepiness scale (ESS). A microsleep nap percentage (MNP) was also obtained and correlated with ESS to determine if a better association could be derived. RESULTS: Using the Spearman correlation the MSL-M improved the correlation with the ESS when compared to MSLT (r=0.106 versus r=0.063), but the results were not statistically significant. Of note, both the MSLT and MSL-M were only weakly correlated to the ESS. The MNP also did not have a good correlation with ESS (r=-0.099). CONCLUSIONS: The addition of microsleep onset to the MSLT score as a quantitative assessment tool failed to significantly enhance the correlation between subjective and objective accounts of sleepiness, beyond the improvement seen in the MSLT value by the simple presence of microsleep alone.