Light mask 500 lux treatment for delayed sleep phase syndrome

Pq.

Nem-Psychopharmacol.

&Bid

Psychiat

Copydght Printed

1999. Vol. 23, pp. 15-24 8 1999 Elsevler Science Inc.

in the USA.

All rights reserved

027%5846/99/$-see

ELSEVIER

front

matter

PI1 !30278-6846(98)00088-8

LIGHT MASK 500 LUX TREATMENT FOR DELiWED SLEEP PHASE SYNDROME KATSUHISA ANDO’, 3, DANIEL F. KRIPKEl, ROGER J. COLE2 and JEFFREY A. ELLIOTT’

l Department

of Psychiatry, University of California, San Diego, San Diego, California, USA 2 Synchrony Applied Health Sciences, San Diego, California, USA

(Final form, November

1998)

Abstract Ando, Katsuhisa, Daniel F. Kripke, Roger J. Cole and Jeffrey A. Elliott: Light Mask 500 Lux Prog. Neuro-Psychopharmacol. & Biol. Treatment for Delayed Sleep Phase Syndrome. Psychiat. 1999. a, pp. 15-24. 01999 Elsevier Science Inc.

Bright light exposure has been demonstrated as an effective treatment for circadian rhythm sleep disorders. Recent studies suggest that more moderate intensities of light A light mask treatment, using light applied through might affect endogenous rhythms. eyelids during sleep, was tested for Delayed Sleep Phase Syndrome. The active light group (n=5) received 500 lux light for 3 hours prior to awakening for 12 days. The placebo light group (n=5) received 0.1 lux light with the same timing. Circadian rhythm phase was assessed from core body temperature and urinary 6 The SIGH-SAD-SR mood scale was administered to sulfatoxymelatonin measurements. assess mood. There were slight trends toward a phase advance of the body temperature rhythm and a phase delay of the melatonin rhythm, and both groups reported anti-depressant benefits. However, no statistically significant effects of 500 lux light mask treatment were demonstrated compared with the placebo-light treatment. More extensive studies will be required to clarify the factors of dose-response and phaseresponse.

circadian, disorder, temperature

Kevwords:

delayed

sleep phase

delayed sleep phase (aMTGs), sleep disorder, temperature

syndrome,

syndrome

3 Present address: Department of Psychiatry of Medicine, Hamamatsu.

light,

(DSPS),

6sulfatoxymelatonin,

urinary

6sulfatoxymelatonin

and Neurology, Hamamatsu Shizuoka, Japan

15

sleep

University

School

K.Ando et al.

16

Introduction

Delayed Sleep Phase Syndrome (DSPS) is a circadian rhythm sleep disorder in which the major sleep episode is delayed in relation to the desired sleep time.

Clinically, the patients

complain of chronic difficulty in falling asleep and difficulty awakening.

Their efforts to

advance sleep timing have little success, and hypnotics give little or no benefit (Weitzman et al., 1981; ASDA Diagnostic Classification

Steering Committee, 1990).

Because light with 3000 lux or more is thought to be effective to shift biological rhythms in humans,

early

Nevertheless,

morning

light treatment

recent studies

endogenous rhythms.

suggest

of several

that much

thousand

lux is applied

lower intensities

of light

for DSPS.

might

affect

Some studies have shown that a few hundred lux or less can suppress

nocturnal plasma melatonin levels (Lewy et al., 1985; Bojkowski et al., 1987; Brainard et al., 1988).

One study demonstrated that 300-500 lux light shifted rectal temperature and plasma

melatonin

rhythms

(Honma et al., 1995).

indicated that phase-shifts

of body temperature

through the eyelids during sleep. simulating

A preliminary

study (Cole and Kripke,

1991)

could be achieved with light masks shining

Moreover, rather dim light administered

late in sleep,

dawn light, might tend to correct circadian phase among patients with phase

delays related to seasonal affective disorder (Terman and Schlager, 1990; Avery et al., 1992; Avery et al., 1993; Avery et al., 1994).

In light mask and dawn simulation studies, the light

reaching the retina was truly dim, because light intensities were reduced to a few percent or less by closed eyelids (Moseley et al., 1988; Robinson et al., 1991; Ando and Kripke, 1996).

This study tested a light mask treatment for DSPS, using 500 lux applied to the eyelids late in sleep.

A low illumination

level, hypothetically

optimal

exposure

time, and a special

illumination pattern were combined.

Methods

The study protocol was approved by an Institutional

Review Board of the University

of

Light mask 500 Iux treatment California,

Diagnoses of delayed sleep phase syndrome were made according to

San Diego.

the criteria

of the

Classification

Steering Committee, 1990).

International

Classification

of Sleep

Disorders

Paid volunteers were recruited by newspaper advertisements interviews with responders (N=llS),

drug users were excluded.

An oral explanation

(ASDA

and fliers.

Diagnostic

After telephone

those who described delayed sleep-wake patterns were Shift workers or psychotropic

asked to record sleep logs for two weeks at home for screening.

obtained.

17

for DSPS

All selected volunteers were in good general health.

of the procedure was provided

and written informed

consent was

To balance severity of syndromes. volunteers were stratified into two groups by

bedtime before or after OZOOh,based on the Z-week screening sleep log data.

Then, subjects

were randomly

within strata.

assigned

to the active light or the placebo

Studies were performed throughout the year. experimental

instructions,

and general compliance

light condition

Volunteers were well-motivated

to follow the

because they had been suffering severely from sleep disturbance, was excellent.

However,

one volunteer

influenza, and one was dropped due to insufficient compliance.

dropped out because

of

Of 12 volunteers who met

DSPS criteria, data from 10 volunteers (5 active and 5 placebo) were available.

EcWment

A fiber-optic system was integrated into sleep eyeshades to deliver cool and diffuse light through closed eyelids.

A white halogen light, a power control and a timing unit were

mounted in a remote box to control illumination standardize illumination

intensities.

The system was designed to

striking the eyelids independent of body position in bed.

Study Design

The

study

randomization

employed

a

baseline-vs-treatment

within-subject

between active and placebo light. treatments.

design

with

parallel

Studies were performed in the

homes of volunteers between October, 1994 and October, 1995.

The start days of the study

schedule were fixed on weekends. Subjects’ retiring times and awakening times were not fixed, and their activity and light exposures

during the daytime were not restricted,

to permit

K.Ando et al.

18

customary activities,

For one week prior to light treatment, data were collected while volunteers were encouraged to maintain

their typical sleep/wake

schedule (day(D) -7 to D 0; baseline).

baseline, volunteers received 12 nights light treatment (D 1 to D 12). when going to bed, and was removed after awakening. hours of 500 lux light during late sleep. linear pattern over 30 minutes,

starting

Illuminance

Following the

The mask was put on

The active light group received 3

was gradually increased in an almost

3.5 hours before projected

wake-up

placebo light group received 0.1 lux light for 3 hours with the same timing.

time.

The

Light intensities

were calibrated prior to each light treatment period, by placing the photo sensor of a lux meter where the eyelids overlying the cornea would be positioned when the mask is worn.

Data Collection

Core Body Tern::

For rectal temperature

probes (Yellow Springs Instrument

recording,

thin flexible

temperature

4400 series) were inserted into the rectum for about 48

hours at the end of baseline (D -2 to D 0) and after the treatment periods (D 12 to D 14). probes were connected to a wrist-mounted every minute.

The

recording device, which stored core temperature

The time of the minimum temperature (BTmin) was defined as the time when

the minimal temperature value of the 24-hour period was recorded. nights were calculated for baseline and post-treatment,

6Sulfatoxvmelatonin:

The averages of the 2

respectively.

Urine specimens were collected for two periods of

two nights each during body temperature monitoring (D -2 to D 0 and D 12 to D 14).

For each

night, urines were collected at one-hour intervals from 2 hours before projected bed time to 2 hours after awakening. awakening.

The sleep interval was represented by a single urine collection at

When urination was necessary during sleep, the extra specimens were collected.

Five or more specimens were sampled per overnight. each sample were recorded.

The time of collection and volume of

A 2-ml aliquot from each sample was frozen.

During the period

of urine collection, between 2 hours before and 2 hours after sleep, subjects stayed indoors and wore dark goggles if the light indoors or outdoors was bright.

Urinary aMT6s concentrations

were determined

by radioimmunoassay

(Aldous, Arendt,

Light mask 500 Iux treatment for DSPS 1988) with reagents supplied by Stockgrand LTD, Guildford,

19

UK.

The amount of urinary

aMT6s per hour (hourly excretion rate) was calculated from the concentration

Data points were set at the

volume of each voiding and the elapsed time between samples. mid time of each urine sampling period. the interpolated

of aMTGs, the

The aMT6s excretion mid-rise (MT50) was defined as

time of rising to 50% of the maximum excretion value of the night.

Single-

night data were used for each estimation, and results of two nights were averaged for baseline and post-treatment,

SleeD

respectively.

During the baseline period (D -7 to D 0) and after the treatment period (D 12 to

D 20), sleep logs were completed to indicate bedtimes and out-of-bed times.

Mood S&:

The SIGH-SAD-SR

treatment, and immediately

(Williams et al., 1990) was completed at baseline, during

after treatment (D -2, D 5 and D 12).

It is a self-administration

form including the standard Hamilton Depression Rating items plus 8 items associated with Seasonal Affective Disorder.

The comparisons for variables between active and placebo light treatments within baseline and post-treatment

were statistically

analyzed by ANCOVA

with the baseline data as the

covariate.

In the active light group, the mean age was 34.41S.D. 13.8 years (4 males and 1 female). the placebo light group, the mean age was 32.6hS.D. 8.1 years (3 males and 2 females). bedtime

based on the sleep log during screening

minutes,

respectively.

bedtime

after 0200h.

was 0239h*60

In the active group 3 and in the placebo No significant

found in age or in screening bedtime.

differences

between

minutes,

In

Mean

and 0207hh43

group 4 volunteers

the two treatment

had

groups were

K. Ando et al.

20

Four pairs of rectal temperature data were available in active and placebo light conditions, because two post-treatment

recordings were not completed.

For urine melatonin,

sleep log,

and mood scale, 5 subjects were available in each treatment condition.

The time of the minimum

temperature

(BTmin) advanced by 54iS.D.

063711 (baseline) to 054311 (post-treatment)

69 minutes

from

in the active light group, and delayed by 1sS.D.

171 minutes from 072311 to 074211 in the placebo light group.

ANCOVA with the baseline

BTmin as the covariate showed no significant treatment effect.

(Table 1)

Urinary aMT6s excretion

was analyzed

considering

(Table 1) suggested that the timing of melatonin receiving min.).

excretion

mid-rise

@ITso).

Results

secretion delayed more among volunteers

500 lux (86rtS.D. 117 min.) than among those receiving placebo light (5hS.D. 66

ANCOVA with the baseline MT50 as the covariate was not significant.

Table 1 Circadian Phase of Core Temperature

Core body temperature (BTmin) Baseline Post-treatment. Urinary 6-sulfatoxymelatonin @ITso) Baseline Post-treatment

and Urinary Melatonin

Active Light Group

Placebo Light Group

0637h f 34.9min 0543h f 43.4min

0723h f 46.2min 0742h f 117.8min

0242h f 77.6min 0408h f 15l.lmin

0350h f 96.9min 0355h f 61.5min

Both into-bed time and out-of-bed time on sleep log advanced in active light group (by 67 minutes and 71 minutes respectively), placebo

light

ANCOVA.

group.

However,

whereas only several minute changes were shown in

analyses

showed

no significant

treatment

effect

with

Total sleep period showed no significant change by treatment (Table 2).

The mean 29-item SIGH-SAD

scores declined

from 21.4 to 15.2 from baseline

to post-

treatment among subjects receiving 500 lux, and from 14.2 to 8.6 among subjects receiving placebo.

ANCOVA showed no significant treatment effects in total scores (Table 2).

Light mask 500 Iw treatment for DSPS

21

Table 2 Sleep Log and Mood Scale

Sleep log Into-bed time Baseline Post-treatment Out-of-bed time Baseline Post-treatment Total Sleep Period Baseline Post-treatment SAD 29-item Baseline Post-treatment

Active Light Group

Placebo Light Group

0233h f 85.3min 0126h f 886min

0145h f 25.0min 0155h f 43.lmin

095411 f 89.2min 0843h f 100.5min

0927h f 70.7min 0928h f 14.6min

442min f 14min 437min f 34min

462min f 66min 453min f 39min

21.4 f 14.3 15.2 f 15.0

14.2 f 5.6 8.6 f 5.2

Discussion

There was a slight trend for temperature

phases in the 500-1~~ treated group to advance

during treatment, as compared to placebo, whereas the phase of melatonin rise was delayed in the treated group. Sleep log data suggested that, by the volunteers’ perception, light treatment was effective in advancing sleep periods. benefit.

Both groups reported considerable anti-depressant

However, no statistically significant benefits of 500 lux light mask treatment were

demonstrated as contrasted to the 0.1 lux placebo-treated

group in this small sample.

Even with a small number of subjects in each group, the experimental had sufficient power to demonstrate

design would have

large treatment benefits of clinical value, but it would

appear that any advantage of active treatment on circadian phase or sleep timing was not large for most volunteers.

It is possible that a longer duration of treatment would have been

more effective, although we had expected 12 days of treatment to produce much of the benefit which might be obtained with long-term experimental

illumination

pattern

treatment.

more closely

There is the possibility

approximated

the rising

that had the light of dawn

(Terman and Schlager, 1990; Avery et al., 1992), that a greater effect might have been derived, though we know of no specific data demonstrating

that dawn simulation patterns are more

K.Ando et al.

22

effective in humans than continuous light at the end of sleep.

Brighter illumination

would

possibly be more effective, as suggested by a subsequent study with 3000 lux performed with similar

light masks

designed

without

(Cole et al., 1996).

first evaluating

sleep/wake rhythm.

Light treatment

endogenous

phase,

schedules

have usually

using the timing

Because, in DSPS patients, the endogenous

reference

been of the

circadian system may be

delayed in reference to the sleep/wake rhythm (Morris et al., 1990) light might not have been presented at the optimal phase of the circadian night (Minors et al., 1991). studies

will

be required

to separate

the factors

of dose-response

More extensive

and phase-response

empirically.

Although no volunteers were suffering from major depression, their self-reported depression scores were above average, and their scores improved quite dramatically The initial SIGH-SAD scores and the degree of improvement

during treatment.

noted in the volunteers receiving

active treatment were similar to what has been reported with light treatment of seasonal affective disorders (Avery et al., 1992; Avery et al., 1993; Avery et al., 1994), however, the problem of distinguishing

active treatment from placebo effects was also similar (Eastman et

al., 1992).

Although this study suggested that moderate light treatment with 500 lux was not effective enough, active and placebo benefits might be distinguished with sufficient numbers of subjects, with more attention

to suggestion

effects in the procedures

(e.g., elimination

of placebo-

responders in a prospective placebo phase), or with lengthier durations of treatment.

AcknowleW

This research was supported by AG123641 and MH001171.

Katharine M. Rex, William J.

Mason, Shawn D. Youngstedt, Melville R. Klauber and Joseph D. Assmus assisted this study.

23

Light mask 500 lux treatment for DSPS

Radioimmunoassay ALDOUS, M. and ARENDT, J. (1988) an iodinated tracer. Ann Clin Biochem 25: 298303. ANDO, K. and KRIPKE, Psychiatry 29: 22-25.

D.F. (1996)

Light attenuation

of 6sulphatoxy

melatonin using

by the human

eyelid.

Biol

ASDA DIAGNOSTIC CLASSIFICATION STEERING COMMITTEE (1990) The International Classification of Sleep Disorders: Diagnostic and Coding Manual, American Sleep Disorders Association, Rochester, pp l-396. Gradual versus AVERY, D.H., BOLTE, M.A.P., COHEN, S.A. and MILLET, M.S. (1992) J Clin Psychiatry a: 359-363. rapid dawn simulation treatment of winter depression. AVERY, D.H., BOLTE, M.A., DAGER, S.R., WILSON, L.G., WEYER, M., COX, G.B., and DUNNER, D.L. (1993) Dawn simulation treatment of winter depression: A controlled study. Am J Psychiatry m: 113-117. AVERY, D.H., BOLTE, M.A., WOLFSON, J.K. and KAZARAS, A.L. (1994) Dawn simulation Biol Psychiatry x6: compared with a dim red signal in the treatment of winter depression. 181-188. BOJKOWSKI, C.J., ALDHOUS, M.E., ENGLISH, J., FRANEY, C., POULTGN, A.L., SKENE, D.J. and ARENDT, J. (1987) Suppression of nocturnal plasma melatonin and 6sulphatoxymelatonin by bright and dim light in man. Horm Metab Res l_$!:437-440. BRAINARD, G.C., LEWY, A.J., MENAKER, M., FREDRICKSON, R.H., MILLER, L.S., WELEBER, R.G., CASSONE, V.M. and HUDSON, D.J. (1988) Dose-response relationship between light irradiance and the suppression of plasma melatonin in human volunteers. Brain Res a: 212-218. COLE, R.J. and KRIPKE, D.F. (1991) Light exposure during sleep with a bright light mask: Sleep Res 2Q: 450. effects on human body temperature rhythm. COLE, R.J., COLE, L.M., KRIPKE, D.F., ANDO, K. and ELLIOTT, J.A. (1996) Light mask 3000 lux treatment for delayed sleep phase syndrome. Sot Light Treatment Biol Rhythms 8: 20. EASTMAN, C.I., LAHMEYER, H.W., WATELL, L.G., GOOD, G.D. and YOUNG, M.A. (1992) A placebo-controlled trial of light treatment for winter depression. J Affect Disord 26: 211-221. HONMA, K., HONMA, S., NAKAMURA, K., SASAKI, M., ENDO, T. and TAKAHASHI, T. (1995) Differential effects of bright light and social cues on reentrainment of human circadian rhythms. Am J Physiol268: R528535. LEWY, A.J., NURNBERGER, J.I., WEHR, T.A., PACK, D., BECKER, L.E., POWELL, R. and NEWSOME, D.A. (1985) Supersensitivity to light: possible trait marker for manicdepressive illness. Am J Psychiatry &Q: 725-728.

24

K. Ando et al.

MINORS, D.S., WATERHOUSE, J.M. and WIRZ-JUSTICE, response curve to light. Neurosci L&t m: 36-40. MORRIS, M., LACK, L. and DAWSON, temperature rhythms. Sleep 13: 1-14.

D. (1990)

A. (1991)

Sleep-onset

A human

insomniacs

phase-

have delayed

MOSELEY, M.J., BAYLISS, S.C. and FIELDER, A.R. (1988) Light transmission through the human eyelid: in vivo measurement. Ophthalmic and Physiological Optics 8: 229-230. ROBINSON, J., BAYLISS, S.C. and FIELDER, A.R. (1991) Transmission adult and neonatal eyelid in vivo. Vision Res a: X337-1840.

of light across the

TERMAN, M. and SCHLAGER, D.S. (1990) Twilight therapeutics, winter depression, melatonin, and sleep. In: Sleep and Biological Rhythms, J. Montplaisir and R. Godbout (Eds), pp 113-128, Oxford University Press, New York. WEITZMAN, E.D., CZEISLER, C.A., COLEMAN, R.M., SPIELMAN, A.J., ZIMMERMAN, J.C. Arch Gen Psychiatry 3: and DEMENT, W.C. (1981) Delayed sleep phase syndrome. 737-746. Seasonal WILLIAMS, J.B.W., LINK, M.J., ROSENTHAL, N.E. and TERMAN, M. (1990) affective disorder assessment tools packet. In: SLBTR 1988-1990: The Complete Works, M. Terman (Ed), pp 207-242, SLBTR, New York.

Inquiries and reprint requests should be addressed to:

Daniel F. Kripke, M.D. Department of Psychiatry University of California, San Diego, 0667 9500 Gilman Drive, La Jolla, CA 92093-0667 U.S.A.