Download A Comparison of Three Different Sleep Schedules for Reducing

SLEEPINESS
A Comparison of Three Different Sleep Schedules for Reducing Daytime
Sleepiness in Narcolepsy
Ann E. Rogers PhD, RN,1 Michael S. Aldrich MD,2 and Xihong Lin PhD3
1School
of Nursing, University of Pennsylvania
of Neurology, University of Michigan
3School of Public Health, University of Michigan
2Department
Study Objective: To determine if the combination of scheduled sleep
periods and stimulant medications were more effective than stimulant
medications alone in controlling the excessive daytime sleepiness experienced by narcoleptic patients.
Design: Twenty-nine treated narcoleptic subjects were randomly
assigned to one of three treatment groups: 1) two 15-minute naps per day;
2) a regular schedule for nocturnal sleep; or 3) a combination of scheduled naps and regular bedtimes. Measures of symptom severity and
unscheduled daytime were obtained at baseline and at the end of the twoweek treatment period, using the Narcolepsy Symptom Status
Questionnaire (NSSQ) and 24-hour ambulatory polysomnographic monitoring. No alterations were made in stimulant medications during the
study period.
Setting: N/A
Patients or Participants: N/A
Interventions: N/A
Measurements and Results: The addition of two-15 minute naps did not
alter either symptom severity or the duration of unscheduled daytime
sleep. Regular times for nocturnal sleep reduced perceived symptom
severity, but did not reduce the amount of unscheduled daytime sleep.
Only the combination of scheduled naps and regular nocturnal sleep
times, significantly reduced both symptom severity and the amount of
unscheduled daytime sleep in treated narcoleptic subjects. The type of
sleep schedule prescribed, however, was less important than the severity
of the patients’ pre-treatment daytime sleepiness. Subjects with severe
daytime sleepiness benefited from the addition of scheduled sleep periods, while those who were only moderately sleepy or able to maintain
alertness did not benefit from scheduled sleep periods.
Conclusions: Scheduled sleep periods are helpful for only those patients
who remain profoundly sleepy despite stimulant medications and should
not be prescribed for all patients with narcolepsy.
Key words: Narcolepsy; treatment; behavioral interventions
INTRODUCTION
increased only 3.6 minutes, resulting in a mean sleep latency of
7.8 minutes on the MSLT (Multiple Sleep Latency Test).
During our pilot study,8 16 narcoleptic subjects, serving as
their own controls, took three 15-minute naps per day, but did not
alter their (stimulant) medication regime or usual sleeping habits
at night. One month later, despite statistically significant increases in mean sleep latencies on the MWT (Maintenance of
Wakefulness Test), there was no change in the number of successful trials (20 minutes without sleep). Finally, there was no
change in the number of sleep attacks recorded in the subjects’
diaries, and no significant improvements in symptom severity as
measured by the Narcolepsy Symptom Status Questionnaire
(NSSQ).
It is possible that discrepancy between the objective (MWT)
and subjective measures of alertness (sleep diaries and NSSQ
scores) was due to differences in the setting for data collection.
Objective data about the subjects’ ability to remain alert was
recorded in the laboratory, while subjective data reflected the
subjects’ experiences outside the laboratory. Since it is probably
inappropriate to assume that laboratory measures of alertness
will accurately reflect subjects’ ability to remain alert in other circumstances, we next designed a study where both objective and
subjective measurements of alertness would reflect the subjects’
experiences while carrying out their usual activities. It was
hypothesized that the combination of scheduled sleep periods
and stimulant medications would be more effective than stimulant medications alone.
FREQUENT, REGULARLY SCHEDULED NAPS,1-4 AND
REGULAR TIMES FOR ARISING AND RETIRING EACH
NIGHT,4,5 are widely recommended as an adjuvant therapy for
narcolepsy. Patients are advised to take three to six 15-minute
naps per day,6 or two longer naps (30—60 minutes each).7 Sleep
loss is to be avoided and regular bedtimes are encouraged.4,6
Although these prescriptions require substantial adjustment in
patient lifestyle, there has been only a limited evaluation of their
efficacy.
With one exception,8 all previous studies testing the efficacy
of sleep satiation via ad libitum sleep,9,10 daytime naps,11-14 and
pharmacologically mediated consolidation of nocturnal sleep15,16
have involved untreated narcoleptic subjects. Although napping
can produce a transient improvement in alertness and performance,12-14,17 it does not reduce the number or duration of
unscheduled daytime naps. Nor does reducing or eliminating
nocturnal sleep disruptions with benzodiazepines or gammahydroxybutyrate9,15,16 have any effect on daytime sleepiness.
Extending the nocturnal sleep period to 12 hours, reduced subjective sleepiness and significantly improved mean sleep latencies the following day.10 However, the mean sleep latency
Accepted for publication January 2001
Address correspondence to: Ann E. Rogers PhD, RN, Associate Professor,
School of Nursing, University of Pennsylvania, 420 Guardian Drive,
Philadelphia PA 19104-6096; Tel: 215-573-7512; Fax: 215-573-7492;
E-mail: [email protected]
SLEEP, Vol. 24, No. 4, 2001
385
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
METHODS
attacks, cataplexy, sleep paralysis, and hypnagogic hallucinations) during 12 typical daily situations and mood states (e.g.,
working, driving, sitting, watching TV, exercising, anger, sadness, tension, boredom). Although this questionnaire is not often
used, it is both reliable and sensitive. In order to test the reliability of this questionnaire, 20 stable treated narcoleptic patients
completed the clinical questionnaire once a week for two
weeks.19 Test-retest reliability for the entire questionnaire was
quite good. The particular sub-scales of interest for this study,
sleepiness, and sleep attacks, and cataplexy were also quite reliable (r =. 91, , and r=.83, and r=.68 ). Furthermore, when Mitler
and his colleagues,20 evaluated the effectiveness of low, moderate, and high doses of methylphenidate, pemoline, protriptyline,
and viloxazine in controlling sleepiness, performance, and cataplexy, they found that decreases in scores on the clinical status
questionnaire (improvement) were highly correlated with
improvements in the ability to stay awake during the
Maintenance of Wakefulness Test. Drugs that produced no
improvements in objective alertness (protriptyline, and viloxazine) as measured by the MWT did not alter the scores on the
NSSQ.
Twenty-four-hour ambulatory polysomnographic recordings
were made before starting the assigned sleep schedule and at the
end of the two-week treatment period using the Oxford Medilog
9000 recorder. On the day of the recording, subjects came to the
sleep laboratory at 17:00h to discuss the study, give informed
consent, complete two questionnaires and to review the sleep
diaries that they would be completing during the study period.
Subjects were instructed to push the event marker on the side of
the recorder just before lights out at night and again in the morning (lights on). The following montage was used for all 24-hour
recordings; electroencephalogram (C4-A1, C3-A2), electrooculogram (ROC/A1, LOC/A2), and submental electromyogram
(EMG). All recordings were obtained while subjects were carrying out their usual activities; no recordings were obtained during
weekends. Recordings started at 18:00h and ended at 1800h the
following day.
The cassette tapes were played back through the Medilog
9000 Reply and the SS90-III Sleep Stager System for automatic
epoch-by-epoch scoring. All tapes were then replayed a second
time and the scoring of the entire record checked by an experienced polysomnographic technologist. Thirty-second epochs
were used and visual scoring was done using standardized criteria for staging sleep.21
The 24-hour recordings were separated into nighttime and
daytime periods based on the event marker for bedtime and
awakening. If the subject failed to push the event marker, information about lights out and lights on was obtained from a sleep
diary.22 The following information was calculated from the nocturnal recordings: total time in bed, time asleep, sleep latency,
REM sleep latency, wake past sleep onset (WASO), duration and
percentage of each sleep stage, number and duration of awakenings, and sleep efficiency. The following information was
obtained from the daytime portion of the recordings: total duration of daytime sleep, number and duration of each nap, type and
duration of each sleep stage, and percentage of total sleep in 24
hours that occurred during the waking period. The number and
duration of daytime sleep periods was determined using the
method described by Broughton and Mamelak.23 To count as a
nap, each daytime sleep period had to be at least three minutes
In order to determine if the combination of scheduled sleep
periods and stimulant medications were more effective than stimulant medications alone, we first obtained baseline information,
then randomly assigned narcoleptic subjects to one of three treatment protocols: 1) two regularly scheduled 15-minute naps per
day; 2) a regular schedule for arising and retiring each day; and
3) a combination of scheduled naps and regular bedtimes.
Subjects who were assigned to take two regularly scheduled 15minute naps per day (Groups 1 and 3) were encouraged to take
the first nap in the morning and the second in the afternoon. We
did not attempt to designate specific times for these naps, but
encouraged patients to choose times that were compatible with
their work and/or school schedules. After they chose their nap
times, we programmed an inexpensive wristwatch to indicate
when the subject was to start and end their scheduled naps, and
instructed the subject to wear that watch during the study period.
Subjects assigned to a regular schedule for arising and retiring
(Groups 2 and 3) were also allowed to select times for nocturnal
sleep that were compatible with their lifestyles. We made no
attempt to increase the duration of their nocturnal sleep. Subjects
were requested to follow their assigned sleep schedule for two
weeks but make no other alterations in their medications or usual
daily activities. At the end of the two-week period, a second 24hour ambulatory polysomnographic recording was obtained.
Subjects
Twenty-nine treated narcoleptic subjects were obtained from
three sites; Evanston Hospital (Evanston, IL), Ingham Medical
Center (Lansing, MI), and the University of Michigan Medical
Center (Ann Arbor, MI). All met the following inclusion criteria:
age 18—65; a history of recurrent daytime naps occurring almost
daily for at least three months; a mean sleep latency of less than
five minutes and two or more sleep onset rapid eye movements
periods (SOREMS) on MSLT; an absence of other sleep disorders as determined by a clinical evaluation and nocturnal
polysomnography; no other neurological disorder, mental illness,
or alcoholism; HLA-DR15 and DQw6 positive; no concurrent
use of analeptic medications other than those prescribed for narcolepsy; and major sleep period at night. Written, informed consent was obtained from all subjects, and all subjects were paid for
their participation.
Seventeen subjects (58.6%) were female and 12 subjects
(41.4%) were male. Ages ranged from 18—64 years, with a
mean age of 43.7 years (SD=13.9). All narcoleptic subjects were
taking stimulant medications; 21 subjects (72.4%) had prescriptions for methylphenidate, five subjects (17.2%) had prescriptions for dextroamphetamine, and the remaining three subjects
(14.3%) had prescriptions for pemoline. Almost half of the subjects (44.8%) were also taking medications to control cataplexy.
Procedure
The severity of five common symptoms of narcolepsy was
assessed pre and post treatment using the Narcolepsy Symptom
Status Questionnaire (NSSQ). This questionnaire, developed by
Mitler and his colleagues,18 asks subjects to indicate on a sevenpoint scale the severity of their symptoms (sleepiness, sleep
SLEEP, Vol. 24, No. 4, 2001
386
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
Table 1—Comparisons of the three treatment groups at baseline
Group 1
(n=10)
39.0 ± 14.7
Group 2
(n=10)
43.2 ±12.4
Group 3
(n=9)
49.6 ±13.3
7 females
3 males
5 females
5 males
5 females
4 males
0.642
39.2±41.9
27.8±31.8
76.4±42.5
0.044*
124.3±22.4
127.3± 33.2
121.3±24.4
0.969
Nocturnal Sleep
Time in bed (mins)
Sleep Duration (mins)
WASO (mins)
Sleep Efficiency (percent)
454.0±104.9
403.1±114.8
39.0 ±33.2
88.2±8.0
434.3±103.0
354.2±6.8
67.1±42.7
81.0±11.4
463.2± 89.7
349.7±81.9
101.0±63.2
76.0 ± 11.6
0.513
0.302
0.063
0.071
Total Sleep during recording
period (mins)
442.2±109.1
382.0±92.4
437.9±111.4
0.035*
Mean Age (yrs)
Gender
Symptom Severity
Mean Duration
Daytime Sleep (mins)
Total Score on
NSSQ (pretest)
p-value**
0.236
* p<0.05
** the p-values were calculated using the Kruskal-Wallis test for continuous measures and the chi-square test used for gender
Table 2—Effects of scheduled sleep periods
Group 1 (n=10)
Mean Duration
Sleep Daytime (mins)
Total Score on
NSSQ
Pretreatment
Post Treatment
Mean difference
T
p-value*
43.5 ± 41.5
30.4 ± 24.5
-13.100
1.451
0.185
124.3±22.4
124.4± 21.7
0.1111
0.0113
0.9912
41.0 ± 25.0
9.300
0.800
0.450
108.5± 26.0
-22.200
-5.268
0.002**
29.3 ± 25.3
-54.4
-3.64
0.011**
19.5625
-3.3196 0.013**
Group 2 (n=10)
Mean Duration
31.7 ± 39.1
Sleep Daytime (minutes)
Total Score on
NSSQ
130.7 ± 34.3
Group 3 (n=9)
Mean Duration
76.4 ± 42.5
Sleep Daytime (minutes)
Total Score on
NSSQ
121.3 ± 24.4
101.8 ±22.0
-
* four influential outliers removed from analysis, and two-sample t-tests were used to calculate the p-value
**p<0.05
oped to determine which of the three treatment protocols was
most effective for reducing unscheduled daytime napping and
subjective symptom severity. We fit a linear regression model of
the change in unscheduled daytime napping on treatment indicators and pretreatment daytime napping to determine which of the
three treatments was more effective in reducing unscheduled
daytime napping. This model allowed us to study the treatment
effect on reduction of daytime sleep duration after adjusting for
differences in day sleep duration among the three groups at baseline. Next, we fit a linear regression model of change of NSSQ
scores before and after treatment on treatment indicators and pretreatment NSSQ scores. For each analysis, we also fit a model
long and consist of stages 2, 3/4 or REM sleep and be preceded
and followed by at least 15 minutes of wakefulness or stage 1
sleep.
Analysis
Demographic variables and pre-treatment scores were examined prior to testing our hypothesis. To determine if the three
treatments were more effective in reducing unscheduled daytime
napping than the use of stimulants alone we compared scores
obtained at baseline to scores obtained after two weeks of treatment using Students t-Tests. Regression models were then develSLEEP, Vol. 24, No. 4, 2001
387
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
with interaction between treatment and baseline data and found
that the interaction was not significant. We also fit a model controlling for other baseline sleep-related variables besides length
of daytime sleep, the NSSQ score, and demographic variables.
None of these variables were significant. Regression diagnostic
techniques, such as residual plots and Cook distances, were used
to identify outlying observations and influential observations.
One highly influential outlier in Group 2 (regular bedtimes
group) was found and removed from the analysis of daytime
sleep duration. For the NSSQ analysis, one highly influential
outlier in each of the three treatment groups was found and
removed from the analysis.
Table 4—Regression coefficient estimates for the change of
NSSQ scores
Variable
Intercept
Group2
Group3
Pre NSSQ
R-square=0.46
Unscheduled daytime napping Our linear regression analysis
suggested that, after adjusting for the effect of pre-treatment daytime sleep, treatment Groups 1 (scheduled naps) and 2 (regular
bedtimes) had almost identical reductions in daytime sleep, and
treatment Group 3 (combination therapy) had much more reduction in daytime sleep duration than Groups 1 and 2 (See Table 3).
Specifically, the mean reduction in unscheduled daytime sleep
for treatment Group 2 (regular bedtimes) decreased 0.06 minutes
(SE= 9.75 minutes) more than subjects in treatment Group 1, a
difference between the two groups that was not significant (pvalue=0.99). In contrast, subjects in treatment Group 3 reduced
their daytime sleep on average by 16.5 minutes (SE=10.8 minutes) more than subjects in treatment Group 1, but the differences
were not statistically significant (p-value=0.14).
We also found that longer pre-treatment daytime sleep durations were strongly associated with greater reductions in
unscheduled daytime sleep durations at the end of the two-week
treatment protocol. Specifically, for every 10-minute increase in
pre-treatment daytime sleep duration, the mean reduction in
unscheduled daytime sleep duration after treatment increased by
six minutes. This effect occurred in all three treatment groups
and represented a significant change (p-value=0.0001, (See Table
3).
As shown in Table 2, only one treatment, the combination of
two 15-minute naps and regular bedtimes, significantly reduced
the amount of unscheduled daytime sleep in treated narcoleptic
subjects (p= 0.011). Subjects assigned to treatment Group 3
(combination therapy) averaged 76.4 minutes of unscheduled
daytime sleep when treated with stimulant medications compared
to 29.3 minutes of unscheduled daytime sleep when treated with
stimulant medications and scheduled sleep periods. The amount
of unscheduled daytime sleep was not significantly reduced by
the addition of two 15-minute naps per day (p=0.185) or regular
times for sleep at night (p=0.450).
Although there was no change in the amount of unscheduled
daytime sleep recorded during 24-hour ambulatory polysomnographic recordings, subjects assigned to treatment group two
(regular bedtimes) reported less severe symptoms at the end of
the two-week treatment period (p=0.002). Subjects assigned to
the combination treatment group (Group 3), also reported
Subjective Symptom Severity
Our second linear regression analysis suggests that, after
adjusting for the effect of pre-treatment NSSQ scores, treatment
groups 2 and 3 (regular bedtimes and combination therapy) had
similar improvement in NSSQ scores and both groups had significantly more improvement in their NSSQ scores compared to
group 1 (naps) (See Table 4). Specifically, the mean improvement in NSSQ scores for treatment Group 2 was 19.3 points
(SE=9.2) more than that of treatment Group 1, and the difference
between the two groups was statistically significant (p-value
=0.047). The mean improvement in NSSQ scores in treatment
group 3 (combination treatment) was 21.0 points (SE= 8.8) more
than treatment Group 1 (naps) and the difference was statistically significant (p-value = 0.02). There were no differences
between Groups 2 and 3 in their improvement in NSSQ scores (pvalue =0.87).
Table 3—Regression coefficient estimates for the change of daytime sleep
T
1.60
0.01
-1.53
-5.27
P-value
0.12
0.99
0.14
0.0001
Note that Group 2 is the dummy variable for group 2 and Group3 is the
dummy variable for Group 3 with Group 1 as the reference group.
SLEEP, Vol. 24, No. 4, 2001
p-value
0.01
0.047
0.027
0.006
Regression Models
Stimulants Alone Versus the Combination of Stimulants and
Scheduled Sleep Periods
SE
7.54
9.75
10.80
0.11
T
2.90
-2.12
-2.28
-3.04
decreased symptom severity at the end of the two week treatment
period (p=0.013). NSSQ scores remained unchanged for subjects
taking two 15-minute naps per day (Group 1).
Although subjects were randomized to treatment groups, subjects assigned to the combination therapy (Group 3) slept longer
during the day at baseline, and subjects assigned to the regular
bedtime group (Group 2) slept less during the 24-hour than the
other two groups of subjects (See Table 1). Our regression analysis adjusted for the difference of the baseline daytime sleep
duration when comparing treatment effects. No other differences
were found in demographic characteristics, NSSQ scores, or
polysomnographic data
Estimate
12.10
0.06
-16.49
-0.59
SE
19.25
9.19
8.82
0.15
Note that Group 2 is the dummy variable for Group 2 and Group3 is the
dummy variable for Group 3 with Group 1 as the reference group.
RESULTS
Variable
Intercept
Group2
Group3
Pre Daytime
Sleep
R-square=0.74
Estimate
55.75
-19.25
-21.00
-0.45
388
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
Table 5—Comparisons of baseline measures by severity of daytime sleepiness
Least Sleepy
(n=8)
42.1 ± 16.0
6 females
2 males
Moderately Sleepy
(n=7)
47.0 ±14.9
7 females
0 males
Profoundly Sleepy
(n=10)
47.0 ±11.9
7 females
3 males
1.6±3.1
33.7±15.6
97.7±16.1
0.000*
113.6±18.4
131.1±29.1
129.5±29.8
0.350
Nocturnal Sleep
Time in bed (mins)
Sleep Duration (mins)
WASO (mins)
Sleep Efficiency (%)
426.1±101.5
389.9±107.2
25.7±14.2
91.0±4.7
453.4±91.9
364.9±99.2
77.6±42.2
79.5±11.8
455.7±106.9
349.0±111.2
95.0±64.6
76.4±12.1
0.584
0.313
0.020*
0.012*
Total Sleep during
recording period (mins)
391.5 ±105.8
398.6 ±95.4
446.7±109.7
0.300
Mean Age (yrs)
Gender
Symptom Severity
Mean Duration
Daytime Sleep (mins)
Total Score on
NSSQ (pretest)
p-value**
0.914
1.422
*
p< 0.05
** The Kruskal-Wallis Test was used to calculate the p-values.
Our linear regression analysis, shown in Table 4, also suggests
that higher pre-treatment NSSQ scores were strongly associated
with greater improvements on NSSQ after treatment. Subjects
with more severe symptoms, as measured by the NSSQ, benefited more from the combination of stimulant medications and
scheduled sleep than subjects with less severe symptoms.
Specifically, for every 10-point increase in pre-treatment NSSQ
score, the improvements in NSSQ score increased by 4.5 points
(p-value=0.006). This effect occurred in all three treatment
groups (See Table 4).
the addition of scheduled sleep periods (Students t-test, p=0.028).
Mean amounts of unscheduled daytime sleep dropped from 97.7
minutes to 62.9 minutes.
DISCUSSION
This study demonstrated that the addition of scheduled sleep
periods may benefit some treated narcoleptic patients. Subjects
assigned to the combination therapy group had less daytime sleep
and reported less severe symptoms at the end of the two-week
treatment period. Although subjects assigned to sleep regular
hours at night reported less severe symptoms at the end of the
study, their objective level of daytime sleepiness did not change.
The addition of two 15-minute naps per day did not alter either
the amount of unscheduled daytime sleep or subjective symptom
severity.
Although combination therapy appears more effective than the
other two treatment approaches tested, the apparent superiority of
combination therapy may be more a reflection of the subjects’
baseline levels of sleepiness than inherent superiority of this
approach. Subjects assigned to the combination therapy group
were significantly sleepier at baseline than the other two groups
of subjects, sleeping on average 76 minutes per day (compared to
28 minutes and 39 minutes). All three sleep schedules produced
significant reductions in the amount of unscheduled daytime
sleep, if pretreatment levels of daytime sleepiness were high.
Subjects with severe daytime sleepiness benefited from the addition of scheduled sleep periods, while those who were only moderately sleepy or able to maintain alertness did not benefit from
scheduled sleep periods. Thus, it appears that the efficacy of
scheduled sleep periods is highly associated with pre-treatment
levels of daytime sleepiness.
Several factors should be considered when recommending
scheduled sleep periods as an adjunct to stimulant medications.
Additional Findings—Effects of Pretreatment Levels of
Daytime Sleepiness
To further examine the relationship between the severity of
daytime sleepiness and the efficacy of scheduled sleep periods,
we divided the subjects into three groups according to the severity of their daytime sleepiness during the baseline recordings.
The first group of subjects, the least sleepy group, included six
subjects who were able to remain awake for the entire day and
two other subjects who had day sleep durations of less than 10
minutes. Subjects in the moderately sleepy group (n=7) had day
sleep durations between 10 minutes and 60 minutes. Subjects in
the profoundly sleepy group (n=10), those with day sleep durations of over 60 minutes, averaged 97.7 minutes of unscheduled
daytime sleep (See Table 5). Although subjects spent approximately the same time in bed at night, subjects who were the least
sleepy during the daytime, had more consolidated sleep at night,
as evidenced by a fewer minutes awake after sleep onset
(WASO), and a higher sleep efficiency.
The addition of scheduled sleep periods did not benefit treated subjects who were alert or only moderately sleepy during the
daytime (See Table 6). Only those subjects who were profoundly sleepy during the baseline recording appeared to benefit from
SLEEP, Vol. 24, No. 4, 2001
389
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
Table 6—Effects of scheduled sleep periods by severity of daytime sleepiness
Least Sleepy
Mean Duration
Sleep Daytime (mins)
Total Score on
NSSQ
Pretreatment
Post Treatment
Mean difference
T
p-value
1.6±3.1
10.1±12.3
11.6
2.081
0.076
113.6±18.4
102.9±18.0
-10.7
-1.477
0.190
28.1±17.0
-5.6
-0.652
0.539
109.6±22.4
-21.4
-3.205
0.018*
62.9±39.0
-34.8
-2.671
0.028*
128.5±29.8
-5.6
-0.586
0.574
Moderately Sleepy
Mean Duration
33.7±15.6
Sleep Daytime (minutes)
Total Score on
NSSQ
131.1±29.1
Profoundly Sleepy
Mean Duration
97.7±16.1
Sleep Daytime (minutes)
Total Score on
NSSQ
129.5±29.1
* p<0.05
First, is the patient likely to benefit from the addition of scheduled sleep periods? Our data suggest that blanket statements
regarding the necessity of scheduled sleep periods for all patients
with narcolepsy should be reconsidered. Although stimulant
medications do not “normalize” patient scores on laboratory tests
of daytime sleepiness,24 roughly 30%—40% of the treated narcoleptic patients we have studied using 24-hour ambulatory
polysomnographic recordings are able to remain awake throughout the day when engaged in their usual daily activities.25
Scheduled sleep periods are not necessary for these patients, and
should only be recommended if the patient is sleeping for more
than 60 minutes during the daytime. Since narcoleptic patients
often underestimate the severity their daytime sleepiness,22,25,26 a
sleep log and/or actigraphy should be used to determine how long
a patient is sleeping during the day.
Secondly, if scheduled sleep periods are recommended, is the
patient likely to comply with your recommendations? Or are
there other approaches that may be equally efficacious and/or
more acceptable to the patient? It may be quite difficult for some
patients to adhere to a regular sleep schedule and/or take two 15minute naps a day.8 Finding the time to nap during the workday
may be especially difficult,27,28 especially if it involves disclosing
their illness. Finally, many patients may resist altering their
lifestyles to accommodate their sleepiness.25,27,29
Compliance is an issue with all patient populations,30,31 and
patients with narcolepsy are no exception.32 Just a little over onethird of the patients in this study (36%) took all their prescribed
stimulant medications. Fifty-seven percent of the least sleepy
subjects took all of their prescribed medications, compared to
29% of the moderately sleepy group and 25% of the profoundly
sleepy subjects. The least sleepy subjects also took a greater percentage of their prescribed stimulant medications than members
of the other two groups, 85% compared to 65% of the dosage
taken by members of the moderately sleepy group, and 51%
taken by members of the profoundly sleepy group. Profoundly
sleepy subjects were also more likely to decrease their dosage by
larger amounts than the least sleepy subjects (they reduced their
SLEEP, Vol. 24, No. 4, 2001
dosages by an average of 41% compared to an average dosage
reduction of 15% for the least sleepy subjects). Moderately
sleepy subjects reduced their dosages on average by 35%. Twothirds of the profoundly sleepy subjects in this study failed to take
their prescribed dosage of stimulant medications, and half of the
moderately sleepy subjects had not taken the full amount of their
prescribed stimulants. Perhaps if they these subject had taken
their prescribed stimulant medications, they might not have such
severe daytime symptoms. Subjects who were compliant with
their prescribed stimulant regimen, but remained moderately or
profoundly sleepy (25% of the subjects in these two groups),
might benefit from increasing the dosage of their current stimulant or changing to a more potent stimulant.
However, if stimulant doses are already high, and the patient
is taking their medications as prescribed, scheduled sleep periods
should be considered as an adjunct therapy for those patients
whose have significant amounts of unplanned daytime sleep.
They should also be considered if a patient is reluctant to take
higher doses of stimulants, experiences adverse side effects with
higher doses, or has another illness that rules out the use of higher doses or the use of certain stimulants.
At present it is difficult to recommend one type of sleep
schedule over another. Since all three treatments produced significant reductions in daytime sleep durations, when pretreatment
levels of sleepiness were high, patient preferences should be considered. For example, patients who are reluctant to request
accommodations from their employer may prefer to avoid scheduled daytime naps. Other patients may prefer scheduled daytime
naps, or the combination of scheduled daytime naps and regular
hours for nighttime sleep. If scheduled daytime naps are used,
patients need to know that a nap will increase their alertness for
only a short period of time (usually less than two hours).33
Additional studies are needed to test the efficacy of the three different sleep schedules in a group of profoundly sleepy subjects.
390
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al
ACKNOWLEDGMENTS
23. Broughton R, Mamelak M. Effects of nocturnal gamma-hydroxybutyrate on sleep/waking patterns in narcolepsy-cataplexy. Canadian j of
neurol sciences. 1980; 7:23-24.
24. Mitler MM, Hajdukovic RM. Relative efficacy of drugs for the
treatment of narcolepsy. Sleep 1991;14:218-220.
25. Rogers AE, Aldrich MS, Caruso CC. Patterns of sleep and wakefulness in treated narcoleptic subjects. Sleep 1994;17:590-597.
26. Bastuji H, Jouvet M. Interet de l’agenda de sommeil pour l’etude
des troubles de la vigilance. Electroencephalogr clin neurophysio
1985;60:299-305.
27. Goswami M, Pollak C. Narcolepsy: a disorder of excessive daytime
sleepiness. Progress in Clinical Neurosciences 1990;6(2):143-156.
28. Pollak CP. Chronobiological findings in narcolepsy and their implications for treatment and psychosocial adjustment. In: Goswami M,
Pollak CP, Cohen FL, Thorpy MJ, Kavey NB, eds. Psychosocial aspects
of narcolepsy. Binghamton, NY: The Hayworth Press, Inc., 1992:147154.
29. Piscopo JA. Controversies in the treatment of narcolepsy. In:
Controversies in the treatment of narcolepsy 7th Annual APSS Meeting,
Rochester, MN: American Association of Professional Sleep Societies,
1993:1-5.
30. Vander Stichele R. Measurement of patient compliance and the
interpretation of randomized clinical trials. Eur J clin pharmacol
1991;41:27-35.
31. Simons MR. Nursing interventions related to compliance. Nurs clin
north am 1992;27: 477-485.
32. Rogers AE, Aldrich MS, Berrios AM, Rosenberg RS. Compliance
with stimulant medications in narcolepsy. Sleep. 1997; 20:28-33.
33. Scharf MB, Brown D, Woods M, Brown L, Hirschowitz J. The
effects and effectiveness of gamma-hydroxybutyrate in patients with
narcolepsy. J Clin Psychiatry 1985;46:222-225.
34. Johns MW. Reliability and factor analysis of the Epworth Sleepiness
Scale. Sleep 1992;15:376-381.
35. Johns MW. A new method for measuring daytime sleepiness: the
Epworth Sleepiness Scale. Sleep 1991;14:540-545.
36. US Modafinil in narcolepsy multicenter study group . Randomized
trial of modafinil for the treatment of pathological somnolence in narcolepsy. Annals neurol 1998;43:88-97.
This study was supported a grant (R29 NS29378) from the
National Institute of Neurological Diseases and Stroke (NIH),
and assistance from Oxford Medilog Inc.
REFERENCES
1. Mitler MM, Gujavarty KS. Narcolepsy: when to suspect it and how
to help. Consultant: the journal of medical consultation. 1982;4:215224.
2. Parkes JD, Dahlitz M. Amphetamine prescription. Sleep
1993;16:201-203.
3. Regestein QR, Reich P, Mufson MJ. Narcolepsy: an initial clinical
approach. J Clin Psychiatry 1983;44:166-171.
4. Roth B. Narcolepsy and hypersomnia. Basal, Switzerland: S. Karger,
1980.
5. McMahon BT, Walsh JK. Narcolepsy: the not so gentle tyrant.
Journal of applied rehabilitation counseling 1982;11:80-83.
6. Kales A, Cadieux RJ, Soldatos CR, et al. Narcolepsy-cataplexy I,
clinical and electrophysiologic characteristics. Archives neurol 1982;
39:164-168.
7. Parkes JD. Sleep and its disorders. London, UK: W.B. Saunders Co.,
1985.
8. Rogers AE, Aldrich MS. Do regularly scheduled naps reduce sleep
attacks and excessive daytime sleepiness associated with narcolepsy?
Nurs Res 1993;42:111-117.
9. Montplaisir J, Godbout R. Nocturnal sleep of narcoleptics: revisited.
Sleep 1986;9:159-161.
10. Yuchiyama M, Mayer G, Meirer-Ewart K. Differential effects of
extended sleep in narcoleptic patients. Electroencephalogr clin neurophysio 1994;91:212-218.
11. Dement WC, Baird WP. Narcolepsy: care and treatment for the primary care physician whose patient is afflicted with narcolepsy. Stanford,
CA: American Narcolepsy Association, 1976.
12. Godbout R, Montplaisir J. All-day performance variations in normal
and narcoleptic subjects. Sleep 1986;9:200-204.
13. Mullington J, Broughton R. Scheduled naps in the management of
daytime sleepiness in narcolepsy-cataplexy. Sleep 1993;16:444-456.
14. Roehrs R, Zorick F, Wittig R, Paxton C, Sicklesteel J, Roth T.
Alerting effects of naps in patients with narcolepsy. Sleep 1986;9:191194.
15. Scrima L, Hartman PG, Johnson FRjr, Hiller FC. Efficacy of
gamma-hydroxybutyrate versus placebo in treating narcolepsy-cataplexy: double blind subjective measures. Biol Psychiatry 1985;26:331341.
16. Thorpy MJ, Snyder M, Aloe FS, Ledereich PS, Starz KE. Short-term
triazolam use improves nocturnal sleep of narcoleptics. Sleep
1992;15:212-216.
17. Helmus T, Rosenthal L, Bishop C, Roehrs T, Syron ML, Roth T. The
alerting effects of short and long naps in narcoleptic, sleep-deprived, and
alert individuals. Sleep 1997;20:251-257.
18. Mitler MM, Gujavarty KS, Browman CP. Maintenance of wakefulness test: a polysomnographic technique for evaluating treatment efficacy in patients with excessive somnolence. Electroencephalogr clin
neurophysio 1982;53:658-661.
19. Rogers AE. Reliability testing of the narcolepsy symptom status
questionnaire. Unpublished data.
20. Mitler MM, Hajdukovic R, Erman M, Koziol JA. Narcolepsy. J clin
neurophysiology 1990;7:93-118.
21. Rechtschaffen A, Kales A. A manual of standardized terminology,
techniques, and scoring system for sleep stages of human subjects.
Washington DC: U.S. Department of Health, Education, and Welfare,
1968.
22. Rogers AE, Caruso C, Aldrich MS. Reliability of sleep diaries for
assessment of sleep/wake patterns. Nurs Res 1993;42:368-371.
SLEEP, Vol. 24, No. 4, 2001
391
Reducing Daytime Sleepiness in Narcolepsy—Rogers et al