Sleep, 6(4):319-325 1983 Raven Press, New York Excessive Daytime Sleepiness Associated with Insufficient Sleep T. Roehrs, F. Zorick, J. Sicklesteel, R. Wittig, and T. Roth Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, Michigan, U,S.A. Summary: Chronic insufficient sleep as an identifiable cause of excessive daytime sleepiness was investigated post hoc by comparing a series of patients with this diagnosis with patients with narcolepsy. Among the prominent features differentiating patients with insufficient sleep from patients with narcolepsy was the report, obtained on the sleep history, of a disparity between the reported amount of sleep obtained on weekdays versus weekends. On evaluation in the laboratory, patients with insufficient sleep showed atypically high sleep efficiency at night and a prolonged sleep time (longer than they report sleeping on a weekday night at home). Compared with patients with narcolepsy, they show a somewhat elevated percentage of stage 3-4 and REM sleep, although this is probably not higher than that of age-matched controls. On the Multiple Sleep Latency Test they displayed moderate sleepiness and no sleep onset REM periods. A mental status examination and Minnesota Multiphasic Personality Inventory did not suggest a primary psychiatric disorder. Key Words: Daytime sleepiness-chronic insufficient sleep-multiple Sleep Latency Test-Narcolepsy. Excessive daytime sleepiness (EDS) is the most common complaint of patients seen at sleep disorders centers (1). Most often, on polysomnographic evaluation a sleeprelated physiological abnormality is found as the cause for the EDS. In the national cooperative study (1), which included about 5000 cases from 11 sleep disorders centers, the most common diagnoses of EDS were sleep apnea (43%) and narcolepsy (25%). These disorders can usually be identified and differentiated from other disorders of excessive somnolence (DOES) by characteristic polysomnographic features (2). In some cases establishing a cause of the patient's EDS may be difficult. This may occur when the nocturnal polysomnogram reveals the absence of sleep-related physiological abnormalities (e.g., apnea, hypopnea, periodic leg movements, or sleep onset REM periods), a normal or even prolonged sleep time, and a normal distribution of sleep stages (i.e., absence of shortened REM latency and high REM density). Yet, the Multiple Sleep Latency Test (MSLT) indicates excessive sleepiness and fails to show sleep onset REM periods indicative of narcolepsy. Often such patients receive a diagnosis of NREM narcolepsy or idiopathic hypersomnolence. In the case series, the Accepted for publication August 1983. Address correspondence and reprint requests to Timothy Roehrs, Ph.D., Sleep Disorders Center, 2921 W. Grand Blvd., Detroit, MI 48202, U.S.A.
320 T. ROEHRS ET AL. third most frequent diagnosis of EDS was idiopathic hypersomnolence, representing 9% of EDS patients (1). Although EDS is usually associated with physiological abnormalities, some cases of EDS may result from a behavioral cause. A persistent failure to obtain sufficient sleep to maintain waking alertness can be overlooked as a cause of EDS. The daily sleep loss may be quite mild, voluntary, and, therefore, not perceived as the cause. However, when the sleep of healthy, normal individuals was restricted by as little as 2 h for 7 consecutive nights, they showed excessive sleepiness on the MSLT (3). It is not yet clear whether chronic insufficient sleep is frequently the cause of EDS and whether it can be differentiated clearly from other DOES conditions. This report presents the results of the post hoc evaluation of 59 consecutive patients at the Henry Ford Hospital Sleep Disorders Center who received the diagnosis of DOES associated with insufficient sleep. To identify the signs and symptoms that differentiate this diagnosis, these patients were then compared with patients who had received a diagnosis of narcolepsy. Patients with narcolepsy were chosen as the comparison group because they can be pathologically sleepy without disturbed nocturnal sleep. METHODS Subjects The subjects were patients referred to the Sleep Disorders Center of Henry Ford Hospital for evaluation of their persistent complaint of EDS. This series included 59 consecutive, unselected patients receiving an insufficient sleep diagnosis and 66 selected patients with narcolepsy. Patients with narcolepsy often show fragmented sleep associated with alpha intrusions, leg movements, and apneas. Since the purposes of this study required a comparison group of patients with pathological sleepiness in the absence of disturbed nocturnal sleep, narcolepsy patients with any of these signs were excluded from the series. Procedure The evaluation consisted of a medical, sleep, and psychiatric history, a physical examination, and a mental status examination. The patients also completed the Henry Ford Hospital Sleep Disorders Questionaire, the Cornell Medical Index, and the Minnesota Multiphasic Personality Inventory (MMPI), and maintained a 2-week diary of their sleep. Then each patient underwent a nocturnal polysomnogram and a MSLT on the following day. The polysomnogram included the standard central (C3) and occipital (Oz) EEGs, EOGs, submental EMG, anterior tibialis EMG, ECG recorded with a V5 lead, and nasal/oral airflow recorded with a thermistor. All patients were required to stay in bed at least 8 h with the restriction that their arising time be no later than 0800 h the next morning. Patients were required to awaken at 0800 h to avoid interfering with the MSLT. For the MSLT, patients were instructed at 1000, 1200, 1400, and 1600 h to lie down on a bed in a dark, quiet room, and try to fall asleep. Standard EEGs (always including an Oz), EMG, and EOGs were recorded during naps. Each nap test lasted 20 min if sleep did not occur. If sleep occurred, patients were allowed to sleep for 15 min to test for the appearance of REM sleep. The patients were observed between the scheduled naps and were not allowed to lie down or to fall asleep at other than scheduled times. All recordings were scored for sleep stages according to the Sleep, Vol, 6, No, 4, 1983
INSUFFICIENT SLEEP 321 standards of Rechtschaffen and Kales (4). Respiration and tibialis EMG recordings were evaluated by a clinical polysomnographer. On the basis of the entire clinical evaluation consisting of interviews, questionnaries, and polysomnography, each patient in this series had received a single diagnosis derived from the Association of Sleep Disorders Centers diagnostic classification (2). There were no patients in this series with clinically significant apnea or periodic leg movements, which would warrant a secondary diagnosis, for the reasons previously described (2). The diagnosis was the consensus of two clinical polysomnographers. This report is a post hoc effort to identify consistent features which differentiate the two patient groups.. RESULTS Demographics and sleep history The demographic characteristics of the patients and details of their sleep histories are presented in Table 1. The two groups of patients did not differ appreciably in their mean age at the time of diagnosis. There also was no significant difference between groups in the number of men and women within each group. Almost all of the insufficient sleep patients, but only two-thirds of the narcopletic patients, were employed (X 2 = 12.70, p < 0.000. Among those employed in each group, a majority (96% of insufficient sleep patients and 86% of narcolepsy patients) worked the day shift. All patients had a primary complaint of EDS. Patients who received an insufficient sleep diagnosis on the average dated the onset of their sleep-wake problem to the late 20s, while those with narcolepsy reported an average onset age of 20 (t = 3.77, p < 0.001). A similar percentage (77% and 74%) of patients in each group had previously consulted a physician regarding their complaint. About half of the patients in each group who had consulted a physician were prescribed stimulants to treat their daytime sleepiness. The insufficient sleep patients reported different sleep habits on weekdays and weekends in comparison with narcoleptics (see Fig. O. On weekends there was no difference between the two patient groups, with both averaging 8-8.5 h of sleep. However, on weekdays the insufficient sleep patients had shorter sleep times (6.4 h) than the narcolepsy patients (7.5 h) (t = 5.32, p < 0.000. In other words, comparing within a group, the insufficient sleep group slept 2 h longer on weekends than on TABLE 1. Demographic characteristics and sleep history Number of women Number of men Age Employed Day shift Onset of sleep-wake problem Previous physician consultation Use of stimulants History of snoring a p < 0.001. b P < 0.005. Insufficient sleep 22 37 40.8 ± 12.2 93% 96% 27.8 ± 13.5 77% 44% 79% Narcolepsy 32 34 45.4 ± 12.1 67%a 86% 19.9 ± 9.45 a 74% 52% 51%a.b Sleep, Vol. 6, No.4, 1983
322 T. ROEHRS ET AL. H R S 0 F S L E E p 9 8 7 6 5 4....} 2 1 0 IS NAR FIG. 1. Reported hours of sleep on weekdays (WD) and weekends (WE) for insufficient sleep (IS) and narcolepsy (NAR) patients. weekdays (t ::= 1O.S9, p < 0.001). In contrast, patients with narcolepsy reported similar sleep times on weekends and weekdays. Finally, a significantly greater percentage of patients with insufficient sleep reported that they snored compared with patients with narcolepsy (X 2 = 10.42, p < 0.005). Polysomnographic evaluation The results of the polysomnographic evaluation are presented in Table 2. The two patient groups spent similar amounts of time in bed when in the laboratory. The patients with insufficient sleep had a greater total sleep time (8.6 h) than the patients with narcolepsy (7.6 h) (I = S.IS, P < 0.001). This difference is not reflective of poor sleep in the narcolepsy patients, but of excessive sleep in the insufficient sleep patients. In the first place, patients with narcolepsy slept in the laboratory as they reported they slept at home and similarly to healthy normals sleeping their first night in the laboratory (S). Secondly, the insufficient sleep patients slept 2.2 h longer in the laboratory than they report sleeping on weekdays at home and longer than normal volunteers do on their first night in the laboratory (S). The additional sleep of the insufficient sleep patients in the laboratory is not the result of increased time in bed compared with patients with narcolepsy, but of more efficient sleep. The sleep efficiency (ratio of time in bed to sleep time), shown in Fig 2, of the patients with insufficient sleep (93%) was better than that (83%) of the patients with narcolepsy (t = 6.75, p < 0.001). Patients with narcolepsy showed efficiencies similar to those reported for normals on the first night in this laboratory (S). In addition, the insufficient sleep patients showed a distri- TABLE 2. Nocturnal polysomnographic parameters Time in bed (min) Total sleep time (min) Stage 1 (%) Stage 3/4 (%) REM (%) a p < 0.02; b P < 0.001. Insufficient sleep 542.8 ± 55.9 501.8 ± 55.8 12.9 ± 6.6 7.1 ± 6.6 21.0 ± 5.7 Narcolepsy 531.4 ± 53.0 442.2 ± 64.9 b 24.7 ± 9.4 b 4.0 ± 6.0" 17.6 ± 6.0 b Sleep. Vol. 6, No.4, 1983
INS UFFICIENT SLEEP 323 100 FIG. 2. Sleep efficiency (TSTrrIB x 100) on / the nocturnal polysomnogram for insufficient 90 T S T 813 T sleep (IS) and narcolepsy (NAR) patients. I 70 B 60 50'---- IS HAR bution of sleep stages different from that of patients with narcolepsy. They had less stage 1 sleep (t = 7.56, p < 0.001) and more stage 3-4 sleep (t = 2.49, p < 0.02). Insufficient sleep patients also had more REM sleep (t = 3.09, p < 0.001). As previously noted, there were no other significant findings on the nocturnal polysomnograms of the patients in either group (i.e., sleep apnea episodes or periodic leg movements). Sleep latencies at night and during the following day (MSLT) and number of sleep onset REM periods on the MSLT are found in Table 3. The latencies at night did not differ between groups; but then patients with insufficient sleep showed much greater variability. On the MSLT the next day, patients with insufficient sleep showed moderate sleepiness, consistently falling asleep within 5 to 7 min. Healthy normals and patients with an EDS complaint but no objective findings usually fall asleep in 12 to 15 min (6). The patients with insufficient sleep were less sleepy than the narcoleptics. On each nap they stayed awake longer (nap 1, t = 4.1, P < 0.001); nap 2, t = 3.2, p < 0.02; nap 3, t = 3.60, p < 0.001; nap 4, t = 4.39, p < 0.001). The pattern of sleepiness also differed between the groups. The insufficient sleep group showed a biphasic pattern of sleep tendency, minimal in the morning and evening and maximal at midday, as is typically seen in healthy normals (7). This was not the case for patients TABLE 3. Multiple Sleep Latency Test results Insufficient sleep Narcolepsy Night (bedtime) min to stage 1 7.5 ± 11.6 4.8 ± 4.7 Nap 1 (1000 h) min to stage 1 6.5 ± 5.1 3.5 ± 2.5 b Nap 2 (1200 h) min to stage 1 5.7 ± 3.4 3.7 ± 3.2a Nap 3 (1400 h) min to stage 1 5.3 ± 3.4 3.2 ± 2.8 b Nap 4 (1600 h) min to stage 1 5.6 ± 3.7 3.1 ± 2.5 b Number REM naps 0.4 ± 0.8 3.1 ± 0.9 b a p < 0.02; b P < 0.001. Sleep, Vol. 6, No.4, /983
324 T. ROEHRS ET AL. with narcolepsy, who showed no variation in sleep tendency. Finally, as would be expected, the number of sleep onset REM periods differed between the groups: narcoleptics had significantly more REM onsets (t = 15.55, p < 0.001). Psychological evaluation The mean number of MMPI elevations (T score greater than 70 on any clinical scale except Mt) for the insufficient sleep group and the narcolepsy group was not different. The mean number of elevations per patient was 1.5 ± 2.2 among insufficient sleepers and 1.2 ± 1.7 among narcoleptics. Sixty-two percent of insufficient sleep patients had one elevation or no elevations; thus, the insufficient sleep patients exhibit little psychopathology. DISCUSSION The results of this case series indicate that there are systematic differences between patients with chronically insufficient sleep and patients with narcolepsy. Among the differentiating features found was a history revealing a significant disparity between the reported amount of sleep obtained on weekdays versus weekends in the insufficient sleep group. The sleep laboratory evaluation of insufficient sleep patients showed relatively high sleep efficiencies at night and typically longer sleep than they report for a weekday night at home. Their nocturnal polysomnogram also revealed the absence of physiological abnormalities. On the MSLT they displayed moderate sleepiness and no sleep onset REM periods. A mental status examination and MMPI did not suggest a primary psychiatric disorder. Previously, Reynolds et al. (8) reported MSL T characteristics of depressed patients. Their depressed patients were generally less sleepy (9.9 min to stage 1 sleep) than the insufficient sleepers in the present study (5.8 min to stage 1 sleep). However, a comparison of depressed patients and insufficient sleepers in the same study by specific measures of depression would help in identifying MSL T differences between these two patient groups. Chronic, voluntary sleep restriction is the presumed cause of the EDS of these patients. Evidence suggesting insufficient sleep as the cause is the disparity between weekday and weekend sleeping habits. The difference in sleep time was 2 h on the average. Experimental work with gradual sleep reduction for prolonged periods has demonstrated that as little as 30 to 60 min of sleep loss per night produced daytime decrements (3,9). Most of these patients consider their weekday sleep loss trivial and assume that it is recovered on weekends. However, while recovery from a single episode of experimental sleep restriction occurs in 1 or 2 nights, it is not clear how repeated episodes of sleep restriction affect daytime sleepiness. It may be that repeated episodes of sleep restriction make one more sensitive to subsequent sleep loss and less able to recover within 1 or 2 nights. Since this is a post hoc study, it is not possible to evaluate the diagnostic sensitivity and specificity of any given result. The difference between weekend and weekday sleep habits, however, is a strong finding, and a prospective study using this parameter as a diagnostic indicator would be important. Other evidence pointing to a sleep insufficiency as the cause of these patients' EDS is the high sleep efficiency these patients display in the laboratory. Typically, normals and patients sleep poorly the first night in the laboratory (10). However, these patients on their first night in the laboratory slept with efficiencies of better than 90%. Likewise, Sleep, Vol. 6, No.4, 1983
INSUFFICIENT SLEEP 325 their sleep in the laboratory on a weekday night was longer than they report sleeping at home on weekdays. In fact, it was similar to their reported weekend sleep times (8.4 h). An interesting question remains, and that is whether these patients are long sleepers or normal sleepers. Until these patients are studied on an ad libitum sleep schedule for a prolonged period, an answer to this question is not possible. However, regardless of whether they are normal or long sleepers, for the above reasons, they appear to be sleep-deprived. Confirmation of insufficient sleep as the cause of a patient's EDS requires, of course, that increased sleep and regular sleep habits (similar weekday and weekend sleep) alleviate the problem. Unfortunately, this is often difficult to achieve, because the life style (perceived social and economic constraints) of these patients requires shortened sleep times during the week. Most of these patients were employed and the majority worked a day shift. Many are factory workers with very early work report times. To obtain 7 or 8 h of sleep would require an early evening bedtime, which is resisted since that would restrict the individual's social and family life. This case series demonstrates that insufficient sleep as a possible cause of EDS is identifiable. As well, such an identification is critically important to establishing an appropriate and effective treatment. Almost 50% of the patients in this series with insufficient sleep had received stimulant medications for their EDS. The use of stimulants, while appropriate for narcoleptics, is questionable as a treatment for insufficient sleep. REFERENCES 1. Coleman M, Roffwarg H, Kennedy S, et al. Sleep-wake disorders based on a polysomnographic diagnosis-a national cooperative study. JAMA 1982;247:997-1003. 2. Association of Sleep Disorder Centers. Diagnostic classification of sleep and arousal disorders, first edition, prepared by the Sleep Disorders Classification Committee, HP Roffwarg, Chairman. Sleep 1979;2:1-137. 3. Carskadon MA, Dement WC. Cumulative effects of sleep restriction on daytime sleepiness. Psychophysiology 1981;18:107-13. 4. Rechtschaffen A, Kales A, eds. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Brain Information Service/Brain Research Institute, University of California at Los Angeles, 1968. 5. Zorick F, Roth T, Hartse K, Piccione P, Stepanski E. Evaluation and diagnosis of persistent insomnia. Am J Psychiatry 1981 ;138:769-73. 6. Zorick F, Roehrs T, Koshorek G, et al. Patterns of sleepiness in various disorders of excessive daytime somnolence. Sleep 1982;5:9165-74. 7. Richardson GS, Carskadon MA, Orav EJ, Dement WC. Circadian variation of sleep tendency in elderly and young adult subjects. Sleep 1982;5:982-94. 8. Reynolds CF, Coble PA, Kupfer DJ, Holzer BC. Application of the multiple sleep latency test in disorders of excessive sleepiness. Electroencephalogr Clin Neurophysiol 1982;53:443-52. 9. Friedman J, Globus G, Huntley A, Mullaney D, Naitoh P, Johnson L. Performance and mood during and after gradual sleep reduction. Psychophysiology 1977;14:245-50. 10. Stepanski E, Roehrs T, Saab P, Zorick F, Roth T. Readaptation to the laboratory in long-term sleep studies. Bull Psychonomic Soc 1981 ;17:224-6. Sleep, Vol. 6, No.4, 1983