The prevalence of multiple sleep-onset REM periods in a population-based sample

The presence of 2 or more sleep-onset rapid eye movement periods (SOREMPs) on a Multiple Sleep Latency Test (MSLT) has been used as 1 of the criteria for the diagnosis of narcolepsy and is thought to be specific to this disorder. However, previous studies have shown the prevalence of SOREMPS in healthy volunteers and apneic patients to be higher than expected. The present study determined the prevalence of 2 or more SOREMPs in a representative sample of the population from southeast Michigan and investigated potential associations with other sleep-related variables.

Cross-sectional laboratory-based analysis.

Sleep disorders clinic.

Population-based sample.

N/A.

A population-based sample of 333 subjects was assessed by nocturnal polysomnography and daytime MSLT (5 naps), and an additional 206 subjectively sleepy people were also assessed (total = 539). Sample demographics were comparable to the 2000 census. Epworth Sleepiness Scale scores were also determined. Groups were formed based on a median split of each sleep variable (Epworth Sleepiness Scale, MSLT, total sleep time from nocturnal polysomnography) for comparisons of SOREMPs in each group.

The prevalence of 2 or more SOREMPs was 3.9%. Only mean sleep latency on the MSLT was a discriminator for the presence of 2 or more SOREMPs (short latency = 6.3%, long latency = 1.9%, p < .05). Among the subjects who had an MSLT of 5 minutes or less (an indicator of a pathologic level of sleepiness), 9.5% had 2 or more SOREMPS.

The overall prevalence of 2 or more SOREMPs in our sample is 3.9%. Interestingly, of the variables assessed (MSLT, Epworth Sleepiness Scale, and total sleep time from nocturnal polysomnography), objective sleepiness, as determined by the MSLT, was the only measure significantly associated with 2 or more SOREMPs. Therefore, subpopulations with excessive sleepiness (eg, shift workers, young adults, patients with apnea) are likely to have a greater prevalence of SOREMPs.