Recurrent fall/winter major depression (known as Seasonal Affective Disorder (SAD)) is a
prevalent and disruptive disorder whose pathophysiological basis is unknown, but several
hypotheses attribute a causal role to the circadian timing system. Bright white light
exposure via the retina has been shown to reverse the symptoms of SAD. Recent physiological
studies demonstrated the existence of retinal ganglion cells capable of transducing light
input to the retinohypothalamic tract, the primary circadian afferent in humans. This
retinohypothalamic system appears to be maximally sensitive to light in the 446-477nm
Using light-emitting diode (LED) technology, light of narrow bandwidths now can be delivered
from a safe, relatively inexpensive device. We propose to contrast in SAD patients the
efficacy and tolerability of 468 nm LED light from a portable 11cm x 6cm
commercially-available device (GoLITEÔ) to a broader 400-700 nm wavelength LED-generated
light housed in an identical device. The broad wavelength (white) light from our LED device
is similar to that from cool-white fluorescent 10,000 lux devices currently the standard for
treatment of SAD (see e.g., Lam & Levitt, 1999).
Twenty-four depressed SAD outpatients will be randomized to a 3-week trial of light therapy
using either the narrow 468 nm LED source or the broader 400-700 nm LED source, each housed
in a GoLITEÔ device. Subjects will be given devices and written instruction for
administering daily treatments at home, 45min every (q) a.m. The devices will be described
to subjects in terms of wavelength but not specifically described as "blue" or "white."
Weekly depression ratings and assessments of adverse effects will be obtained by a trained
rater blind to the treatment condition. Depressive symptoms will be rated weekly by the same
The following hypotheses will be evaluated:
- H1-- Depressed SAD patients will demonstrate greater antidepressant therapeutic benefit
from the narrow-wavelength (blue) source than from the broad-wavelength (white) source.
- H2-- Depressed SAD patients will manifest fewer adverse effects during treatment with
the narrow-wavelength (blue) source than with the broad-wavelength (white) source.
BACKGROUND AND SIGNIFICANCE
Discovery in the late 1970s that bright light affects neuroendocrine rhythms such as
melatonin secretion, and can reset the circadian pacemaker in humans, paved the way for the
discovery that seasonally recurrent fall/winter depressions--christened Seasonal Affective
Disorder (SAD)--can be treated with bright light exposure (Lewy et al., 1980; Rosenthal et
al., 1985). It was reasoned that seasonal rhythms of mood in humans, like other seasonal
physiological cycles in mammals, could be regulated by the biological "clock." A series of
studies has demonstrated that bright light exposure does produce therapeutic effects in SAD
patients beyond the level achieved by placebos (Eastman et al., 1998; Lewy et al., 1998;
Terman et al., 1998) and that retinal exposure is required for efficacious treatment (Wehr
et al., 1987; Koorengevel, 2001).
Attempts to construct an action spectrum for suppression of melatonin secretion and for
resetting the circadian pacemaker suggest that 446-477 nm wavelengths are most potent
(Brainard et al., 2001; Lockley et al., 2003), which is inconsistent with rod or cone
mediation of the responses (Brainard, 2004). Discovery that a melanopsin-containing system
of retinal ganglion cells serves as a primary afferent to the hypothalamic circadian
pacemaker (Provencio et al., 2000; Gooley et al, 2001; Hattar et al, 2002) suggests a
possible basis for the potency of violet/blue light on these neuroendocrine circadian
endpoints and raises the possibility that light of this wavelength also might be uniquely
potent for treatment of SAD (Brainard, 2004).
Estimates of SAD prevalence range from 0.4 - 9.7% of the general population in the United
States, depending on the survey methods and precise criteria used (Rosen et al., 1990;
Blazer et al., 1998). In addition to those with recurrent episodes of frank Major
Depression, patients with less severe "sub-syndromal" winter depression and those with
annual exacerbation of a year-round mood disorder also have been shown to benefit from light
treatment in fall/winter months (Kasper et al., 1989). Since the 1990s, cool-white
fluorescent sources capable of yielding 10,000 lux polychromatic white light have been the
treatment standard (Lewy et al., 1998; Terman et al., 1998; Desan et al., 2001; also see
e.g., Lam & Levitt, 1999). Although well-tolerated, some transient adverse effects of the
10,000 lux white light have been reported, such as agitation or feeling "wired," insomnia,
headache, eye or vision problems, nausea, sedation, and chest tightness (Labbate et al.,
1994; Kogan et al., 1998; Terman & Terman, 1999). The more common of these
complaints--headache, eye or vision problems, and insomnia--remit rapidly after
discontinuation of light exposure (Oren et al., 1991), however, narrower bandwidth light if
more potent in antidepressant effectiveness might be administered at lower intensities and
thereby further reduce adverse effects and increase tolerability of the treatment (Brainard,
A between-subjects double-blind comparison of 3-weeks treatment using narrow-band LED panels
at blue (468 nm at 500 mW/cm2) vs red (700 nm at 15 mW/cm2) wavelengths in 24 SAD patients
demonstrated both greater reductions in depressive symptoms and also remission rates (54.5%
vs 30.8%) in favor of the blue light source (Byrne et al., 2004). Comparisons of this narrow
bandwidth blue LED light with white light would be of even greater clinical interest since
white fluorescent sources currently are the industry standard.
At the irradiance levels emitted by commonly utilized light therapy devices, dermatologic
safety concerns are minimal. Similarly, thermal damage to cornea, lens or retina requires
milliwatt-to-watt exposure, far in excess of that emitted from the therapeutic devices.
Ocular safety for 10,000 lux white fluorescent sources has been assessed and comprehensive
ophthalmologic examinations of individuals with healthy eyes who used white-light therapy
daily during fall/winter months for up to 5 years did not reveal adverse effects (Gallin et
al., 1995; Gorman et al., 1993). Shorter wavelengths of light are of greater concern due to
photo-keratitis of the cornea and cataract of the lens from 180-400nm ultraviolet light, and
photochemical injury to the retina at 310-550 nm with a peak near 440 nm (Sliney, 2004). The
blue-light LED source to be utilized in the present research (Go-LITE™) was determined to
have an averaged radiance well below the 10-mW/cm2*sr safety limit for continuous viewing.
We will recruit adults, age 18-64 years, who meet diagnostic criteria for seasonal affective
disorder by means of public service announcements in the community. The approved
announcement will be released via the Partners weekly research announcement system, may be
disseminated to press outlets, and will be used on the voice mail menu of Dr. Anderson’s SAD
clinical services at BWH. An informational webpage linked to the BWH website
(http://www.brighamandwomens.org/psychiatry/Research/sadresearch.asp) will provide more
detailed information and a mechanism for prospective subjects to request further
information. Interested individuals will be directed to a research assistant at the BWH
Clinical Trials Center who will answer questions regarding the study and will conduct a
telephone screening to assess eligibility. Those subjects found eligible and interested will
be referred by the research assistant for a psychological screening interview.
Each subject will be enrolled by his or her written informed consent obtained by members of
the study staff. Following an initial telephone screening, each prospective subject will
undergo an in-depth psychological evaluation at BWH by the principal investigator, Janis L.
Anderson, PhD., or at McLean Hospital by Carol Glod, RN, Ph.D. Prior to the psychological
screening, Dr. Anderson or Dr. Glod will review the Screening Consent Form with the subject.
Dr. Anderson is an experienced clinical psychologist and Dr. Glod is an experienced doctor
of nursing; each is trained to consider both cognitive and emotional factors that affect the
ability to consent. If written informed consent is obtained on the Screening Consent Form
then the psychological evaluation will take place. In that evaluation meeting, Dr. Anderson
or Dr. Glod also will present the subject with a copy of the Study Consent Form to review
and take home. Subjects will be encouraged to inform their medical providers and therapists
about their interest in participating in the study. The meetings with Dr. Anderson or Dr.
Glod will be scheduled in advance; nothing in the circumstances of this protocol would
require a hasty decision on the part of the potential subject. Subsequent to the
psychological screening for prospective subjects who meet diagnostic criteria, a medical
examination will be performed by a doctor of Ophthalmology. If the subject is found to meet
all inclusion criteria for the study, doctor of Ophthalmology, a licensed physician, will
review the Study Consent Form and obtain written informed consent for the study procedures
themselves. Upon each outpatient visit for the study procedures, the subject’s mental status
will be reassessed. However, due to the nature of SAD symptoms and the inclusion/exclusion
criteria for the present study, we do not anticipate enrolling subjects whose ability to
consent would be transitory.
RESEARCH DESIGN AND METHODS
A between-subjects double-blind trial will compare daily outpatient treatment using a narrow
(468 nm, 27 nm half-peak bandwidth) LED light source (GoLITE™) to a broad 400-700 nm
wavelength LED-generated light housed in an identical device in 24 seasonally-depressed
adults. Dependent variables will be severity of depression and extent of adverse effects.
Research procedures will not begin in the Autumn/Winter until the subject has a depression
rating of at least 20 on the SIGH-SAD (Structured Interview Guide for the Hamilton
Depression Rating Scale-Seasonal Affective Disorders Version; Williams et al., 1988). The
SIGH-SAD is a structured clinical interview for rating not only the traditional Hamilton
Depression Rating Scale items but also the "atypical" vegetative symptoms characteristic of
SAD. The validity of and reliability of the SIGH-SAD is well-established, and it is the
preferred instrument for assessing symptom severity in SAD studies (e.g., see Lam & Levitt,
1999). A companion instrument, the Hypomania Interview Guide for Seasonal Affective Disorder
(HIGH-SAD) also will be administered. A rater, who will be blind to the treatment condition
of the subjects, will administer the SIGH-SAD and HIGH-SAD prior to and for each week during
the 3-wk trial.
Subjects will begin treatment when post-treatment SIGH-SAD depression score ≥ 20. Dr.
Anderson or Dr. Glod will meet with each subject to provide the light device and instruction
for its outpatient use as detailed below. The subject will be asked to sign a written
agreement for return of the light device. Subjects will use the light device and return
weekly for symptom ratings; each will be seen by Dr. Anderson or Dr. Glod as well. At the
conclusion of the 3-wk trial, Dr. Anderson or Dr. Glod will meet with the subject to discuss
their response to light treatment and the indications for continuing light treatment through
the end of that winter. Subjects will be offered the opportunity to keep the same device or
try the other device until the end of the fall/winter season and will be followed monthly if
they choose to do so. Alternative treatment modalities also will be discussed and referrals
for follow-up care provided as indicated. All forms of SAD treatment will cease by June 1.
EQUITABLE SELECTION OF SUBJECTS
SAD is diagnosed on the basis of history and a definitive diagnosis is difficult to make in
adolescents. In addition, their symptom profile differs from that seen in adults which would
make assessment using any single measure difficult and would require a substantially larger
sample size. Therefore in this preliminary investigation we propose to recruit individuals
between the ages of 18-64. Because the safety of light treatment in pregnancy has not been
established, and because mood effects of pregnancy and recent delivery would complicate
assessment of treatment response in this 3-wk trial, only nonpregnant, non-lactating women
will be included.
Recruitment will be by public means and no one will be excluded on the basis of race,
gender, ethnic background, or other demographic factors. Dr. Anderson is responsible for
outpatient monitoring of light treatment and is only fluent in English. Therefore, if
potential subjects do not understand English it will be necessary to have available a
bilingual physician in the community who would be able to work with Dr. Anderson in
providing outpatient care.
PRIVACY AND CONFIDENTIALITY
Disclosure of Information Outside Partners:
Identifying information will be recorded only on screening contact sheets (which will be
maintained under secure conditions in the Clinical Trials Center and later stored in locking
file cabinets in a locked research room at 221 Longwood Ave. under the control of Dr.
Anderson), on the signed consent forms (stored in locking file cabinets in a locked research
room at 221 Longwood Ave. under the control of Dr. Anderson or at McLean Hospital under the
control of Dr. Glod), and in emails sent within the Partners system to staff members who
schedule the appointments, obtain laboratory specimen; or conduct symptom ratings.
All data will be stored in computer files with code number for the subjects; no identifiers
other than the code number will be kept in the data files nor will they be used in any
reports or publications concerning this research. The key for the code will be kept in
locking file cabinets in a locked research room at 221 Longwood Ave. under the control of
Dr. Anderson or at McLean Hospital under the control of Dr. Glod.
Any records that may be of use to other treatment providers for purposes of continuity of
care, whether at Partners institutions or elsewhere, will be released upon written
authorization of the subject or their legal representative in accordance with current
applicable laws and regulations.
Patients with seasonal affective disorder may experience relief from symptoms with light
therapy during this study. If the treatment is effective, they may continue it with
equipment from and supervision by Dr. Anderson or Dr. Glod through the end of the winter
season. If the treatment is not effective, referrals for additional treatment will be made.
However, relief from depression is in no way guaranteed.
FORESEEABLE RISKS AND DISCOMFORTS
1. Blood drawing. There may be some discomfort or bruising on initial insertion into a
vein. There is a rare possibility of developing a small blood clot, inflammation, or
local infection. Fainting may rarely occur in individuals very uncomfortable with blood
2. Light treatment. Safety of equipment. Light therapy with bright fluorescent light has
been tested in many individuals and serious irreversible adverse effects have not been
reported. The most common side effect of light therapy is irritability or agitation
that remits after termination of the light or decrease of the daily duration of
exposure. Participants will be carefully monitored for such side effects and light
therapy exposure will be decreased or eliminated if these adverse effects occur.
Although light treatment has been tested in pregnant women in a small number of
studies, the safety of light treatment in pregnant women has not been conclusively
established. Female subjects will be told that participation in this study may be
hazardous to an embryo or fetus and therefore that it is essential that adequate birth
control be used while participating. If they should become pregnant while in the study,
they should notify the investigators immediately.
MINIMIZATION OF RISKS AND SAFETY MONITORING
Subjects will be instructed to begin daily 45-min sessions of exposure to the light each
morning shortly after they awaken (provided it is after 6 a.m.). Written instructions will
be given, potential adverse effects will be discussed, and subjects will be instructed to
telephone the supervising clinician (either Dr. Anderson or Dr. Glod) with any questions or
should any adverse effects occur. A telephone appointment will be scheduled for 2-3 days
after the first use of the light device. Patients who complain of feelings of jitteriness
after using the light device will be advised to reduce their duration of daily exposure to
30 min per day. Patients who complain of headache after using the light device each day for
2-3 days will be advised to return for a visit to review the procedures they are using. An
appointment will be scheduled for 7 days after the first use of the light device. Any
patient who continues to complain of headache after the first week of light exposure will be
withdrawn from the study. Patients who complain of feelings of jitteriness after using the
light device 20 min per day will be withdrawn from the study.
An appointment will be scheduled for 14 days and 21 days after the first use of the light
device. In general, the criteria for withdrawing patients from the study are based on Axis
V of the DSM-IV, Global Assessment of Functioning (GAF). No patient with a GAF less than 50
will be enrolled in this study, and any patient whose GAF decreases to less than 50 during
the course of the study will be withdrawn. This specifically includes, but is not limited
to, any patient reporting suicidal ideation, symptoms of psychosis, or symptoms of mania at
any point during the study will be withdrawn from the study and evaluated immediately by Dr.
Anderson or Dr. Glod for treatment.
Any adverse events will be reported promptly to the Human Research Committee.
- History of recurrent major depressive episodes with winter-type seasonal pattern by
Diagnostic and Statistical Manual of the American Psychiatric Association, 4th Ed.
(DSM-IV) criteria (American Psychiatric Association, 1990), based on diagnostic
interview utilizing the Structured Clinical Interview for DSM-IV Axis I Disorders
(SCID) and graphic diagnostic tool
- Free of medical illness, not pregnant, as determined by detailed history and physical
examination including blood and urine chemistries and thyroid function tests
- History of concurrent psychiatric illness that would preclude compliance with the
protocol and ability to complete the study safely
- Active suicidal or homicidal ideation or plan
- Variable psychiatric symptoms such as rapid cycling or severe premenstrual syndrome
that could interfere with accurate assessment of the treatment effect
- History of substance abuse/dependence with less than one year remission
- GAF < 50
- Light treatment in the previous month
- Pregnant or lactating
- Antidepressant medications in the previous month
- Nightwork or other habitual alteration of sleep/wake cycle
- Medical conditions that affect mood or produce hallmark symptoms of mood disorder
- Use of photosensitizing medications (amiodarone, benoxaprofen, chlorpromazine,
demeclocycline, fleroxacin, nalidixic acid, ofloxacin, piroxicam, porfimer,
psoralens, quinidine, temoporfin) or remedies (St. John’s wort)
- Macular degeneration or cataract