Lewy et al. 10.1073/pnas.0602425103. |
Supporting Materials and Methods
Supporting Figure 7
Fig. 7. Study schedule.
Supporting Materials and Methods
Recruitment and Screening.
After the study was approved by the Oregon Health & Science University (OHSU) Institutional Review Board, informed and written consent was obtained from all subjects interested in participating. Subjects were recruited by using newspaper advertisements, which appeared in the late fall. Each year, at least 27 subjects met screening criteria, which included the following: (i) having a Structured Interview Guide for the Hamilton Depression Rating Scale - Seasonal Affective Disorder Version (SIGH-SAD) (1) score of ³19 with an atypical score of ³6 (2); (ii) reporting a depression that developed during the fall or winter and remitted the following spring for at least the two preceding years; and (iii) abstaining from treatment for the prior 4 weeks.Protocol Design.
The 4-week double-blind placebo-controlled protocol was conducted in Portland, OR (latitude 45°N). Each year, subjects were divided into three cohorts of nine subjects. The baseline week for the first cohort followed the first or second Thursday of January, to avoid psychosocial and operational interference from holidays and weekends, which was considered to be more of a potential confound than the expected gradual improvement as the photoperiod lengthened following the winter solstice; the second and third cohorts started 2 and 4 weeks later, respectively. In each cohort, three subjects were randomly assigned to each of the three treatment groups: PM melatonin, AM melatonin, or placebo (see below); assignment was then adjusted slightly (if necessary) to ensure that there were no significant differences in screening SIGH-SAD scores and waketimes between the treatment groups.For the duration of the 4-week protocol (Fig. 7), subjects maintained constant sleep schedules individualized to their weekday habits, getting into bed at their chosen time with the lights out, and staying there until their chosen waketime (naps were discouraged as well). They recorded their sleep and bedtimes in daily diaries. For years 2-4, each subject wore a wrist activity monitor on the nondominant arm (activity monitors were not available for the first winter of the study, 1997-1998). An "event marker" button on the activity monitor was pressed when getting into and out of bed.
Baseline Week.
On the Thursday at the end of the first (baseline) week, subjects were admitted to the OHSU General Clinical Research Center (GCRC) for depression severity (SIGH-SAD) assessments by at least one blinded rater, and dim light melatonin onsets (DLMOs) were assessed by using the procedure described below. At the end of the evening, subjects were given capsules to be taken during the first treatment week. Most subjects collected urine for one 24-hour period at home during this week for analysis of 6-hydroxymelatonin by RIA (3); as a validation analysis, when DLMO data were corrected for the differences in amplitude of total overnight melatonin production (4), the findings were consistent with those presented below.Treatment Weeks.
On the next two consecutive Thursdays, subjects returned to the OHSU GCRC for SIGH-SAD assessments, after which they received another week’s supply of capsules and exchanged completed diaries for new ones. On the fourth Thursday after the first GCRC admission, posttreatment SIGH-SAD assessments and DLMOs were obtained, and study materials were collected. Subjects were given a bright-light fixture to take home at the end of the evening and were instructed on how to do light treatment, as compensation for participation in the study, which they had been told during the consenting procedure was the treatment of choice for SAD. Subjects also were told that each capsule would contain either melatonin or placebo and that their contents could differ within or between each day or week. Subjects were further told that the study was using melatonin administration to understand how some aspects of bright light worked; they also were told that some may experience little or no overall change, that some symptoms might improve and that other symptoms could possibly worsen (and that any worsening of symptoms would be carefully monitored and that they could discontinue the study at any time to immediately begin light treatment).Melatonin Administration.
To enhance the phase-shifting effects of exogenous melatonin, it is our thinking that plasma melatonin levels should be continuously elevated for a certain number of hours; for example, to produce the largest possible phase advance, plasma melatonin levels should be elevated throughout the afternoon, extending into the evening, and overlapping with the endogenous melatonin profile, thereby creating a continuous elevation in melatonin levels with an earlier rise (5, 6). Although a large dose would be able to elevate plasma melatonin for the desired duration, it also would cause a large spike in levels within 15-60 min of ingestion that might make some patients sleepy. Therefore, in lieu of a low-dose, sustained-release formulation, we devised a regimen consisting of the administration of several low doses of melatonin at 2-h intervals (5, 6), which was changed slightly each year to attempt to minimize the soporific side effect while maximizing phase-shifting effects.In all four years, starting on the Saturday after the baseline assessments, and ending on the Wednesday before the final Thursday (before phase assessments), subjects took one capsule every 2 h beginning at waketime. In the first 2 years of the study, subjects took a total of seven capsules per day. In the first year of the study, subjects in the PM-melatonin group took four placebo capsules beginning at waketime and three 0.1-mg doses thereafter; subjects in the AM-melatonin group took three 0.1-mg doses beginning at waketime and four placebo capsules thereafter; and subjects in the placebo group took seven placebo capsules beginning at waketime. The second year of the study was identical to the first year, except the melatonin dose was decreased to 0.075 mg. In the third year of the study, an eighth capsule was added 16 h after waketime, so that a fourth melatonin dose of 0.075 mg would be taken, for a total of eight capsules per day: subjects in the PM-melatonin group took four placebo capsules beginning at waketime and four 0.075-mg doses of melatonin thereafter; subjects in the AM-melatonin group took four 0.075-mg doses of melatonin beginning at waketime and four placebo capsules thereafter; and subjects in the placebo group took eight placebo capsules beginning at waketime. During the fourth year of the study, the first week of treatment was identical to that of the third year, but during the last two treatment weeks the time of the melatonin doses for the PM-melatonin and the AM-melatonin groups was advanced or delayed by 2 h, respectively: subjects in the PM-melatonin group took three placebo capsules beginning at waketime, then 0.075-mg doses of melatonin in the next four capsules, followed by placebo in the last capsule; subjects in the AM-melatonin group took a placebo capsule beginning at waketime, then 0.075-mg doses in the next four capsules, followed by placebo in the last three capsules. Subjects were told that each capsule would contain either melatonin or placebo and that their contents could differ within or between each day or week. The total daily dose of melatonin was either 0.225 mg (year 2) or 0.3 mg (years 1, 3, and 4). Melatonin was obtained from Regis Chemical (Morton Grove, IL), and all capsules were formulated at Oregon State University and given under IND 26,318.
Actigraphic Sleep Times.
Onsets and offsets were calculated automatically by using the ACTIWARE software provided by Mini-Mitter (Sun River, OR), after the entry of initial estimated onset and offsets by a research associate blind to treatment. Actiwatches record motor activity. The data are expressed as activity counts per 1-min epoch and are stored in on-board memory. Downloaded data are expressed graphically as actograms. An event marker is used to record each subject’s in-bed and out-of-bed times to improve computation of sleep onset and sleep offset times. In the case of an occasional missing or anomalous event marker, the rater referred to sleep diaries for these estimates. Sleep onsets and offsets for the 3 days before the day of DLMO assessments were averaged.DLMOs.
Blood was drawn through an indwelling, saline-lock, venous catheter inserted at » 17:00, at which time ambient light intensity was reduced to <30 lux. Samples were collected every 30 min from 18:00 until at least 23:00 (4, 7). Plasma was separated by centrifugation, then poured or pipetted into clean polypropylene tubes and frozen (–20° C) until assayed. Plasma melatonin levels were measured by using the Buhlmann RIA kit supplied by ALPCO Diagnostics (Windham, NH), which employs the antibody raised by Kennaway and colleagues (8, 9), validated by GCMS (10, *). The lower sensitivity limit of this assay is 0.5 pg/ml.*Lewy, A. J., Sack, R. L., Boney, R. S., Clemons, A. A., Anderson, N. R., Pen, S. D., Bauer, V. K., Cutler, N. L. & Harker, C. T. (1997) Sleep Res. 26, 733 (abstr.).
1. Williams, J. B. W., Link, M. J., Rosenthal, N. E. & Terman, M. (1988) Structured Interview Guide for the Hamilton Depression Scale—Seasonal Affective Disorder Version (SIGH-SAD) (New York State Psychiatric Institute, New York).
2. Terman, M., Terman, J. S. & Rafferty, B. (1990) Psychopharmacol. Bull. 26, 505–510.
3. Arendt, J., Bojkowski, C., Franey, C., Wright, J. & Marks, V. (1985) J. Clin. Endocrinol. Metab. 60, 1166–1173.
4. Lewy, A. J., Cutler, N. L. & Sack, R. L. (1999) J. Biol. Rhythms 14, 227–236.
5. Lewy, A. J. & Sack, R. L. (1997) J. Biol. Rhythms 12, 595–603.
6. Lewy, A. J., Bauer, V. K., Cutler, N. L. & Sack, R. L. (1998) Psychiatry Res. 77, 57–61.
7. Lewy, A. J. & Sack, R. L. (1989) Chronobiol. Int. 6, 93–102.
8. Earl, C., O’Occhio, M., Kennaway, D. & Seamark, R. (1985) Endocrinology 117, 226–230.
9. Voultsios, A., Kennaway, D. J. & Dawson, D. (1997) J. Biol. Rhythms 12, 457–466.
10. Lewy, A. J. & Markey, S. P. (1978) Science 201, 741–743.