Seasonal affective disorders are characterized by affective episodes (depression, hypomania, or mania) recurring regularly during certain seasons. Patients with this form of affective disorder were probably first recognized by Esquirol1 and later by Kraepelin.2 It is only in the past decade, however, that they have become the focus of systematic study. Rosenthal et al described a form of the condition where fallwinter depressions alternate with !!undepressed periods in the spring and summer,5 a condition we will refer to in this paper as seasonal affective disorders (SAD). More recently, Wehr et al have described a variant with an opposite pattern, namely depressions in the summer and nondepressed periods in fall and winter, a condition they have called reverse-SAD.4 A similar syndrome has been recognized by Dr. Philip Boyce (personal communication. 1987). Other patterns of regularly occurring episodes of mania and depression certainly occur; for example, regular spring depressions, spring and fall depressions, and fall hypomanias. However, there has been little systematic work on these conditions. Most studies to date have been performed on SAD. which will therefore be the major focus in this paper.
CLINICAL AND DEMOGRAPHIC FEATURES
We have used the following operational criteria for a diagnosis of seasonal affective disorder:
* a history of at least one major depressive episode, according to the Research Diagnostic Criteria (RDC)5;
* regularly occurring fall-winter depressions (at least two occurring during consecutive winters) alternating with nondepressed periods during spring and summer;
* no other major psychiatric disorder; and
* no psychosocial variables accounting for the regular changes in mood.
Other researchers have used similar criteria in their studies.
Most patients (83%) with SAD are women (Table 1), whose onset of illness typically occurs in the third decade. In the Washington, DC area depressions last, on the average. 5.1 months, generally beginning in October or November and subsiding in March or April. Most patients (83%) report that they have suffered hypomania at some time in the past, generally in the spring and summer, and therefore qualify for a lifetime diagnosis of major depression with hypomania (bipolar II); a smaller percentage have a lifetime psychiatric diagnosis of unipolar or bipolar I affective disorder. These features are similar to those reported elsewhere, especially the preponderance of women.6,7 However. Yerevanian et al reported that their patients with "seasonal major depression" most commonly had a history of recurrent unipolar depressions, a finding which may reflect the climate in Rochester, New York, where there are relatively few sunny days throughout the year.7 The presence of sunny summers may be a prerequisite for the emergence of hypomanic symptoms. The majority of patients have at least one first degree relative with a psychiatric condition, most frequently major affective disorder or alcohol abuse (Table 1).
Depressions are frequently mild to moderate but may be severe. However, hospitalization has been required for only 6% of our population and ECT for only l%. Vegetative symptoms are usually of the atypical variety (increased eating, carbohydrate craving, weight gain, and oversleeping) although some patients report more typical vegetative symptoms (Table 2). Social withdrawal and impaired work functioning are common. The single outstanding characteristic of this depression is reactivity to changes in environmental lighting and climate. Those patients who have lived at different latitudes frequently note that the farther north they have lived, the longer and deeper their winter depressions have been. Sometimes they have never been depressed until moving north. Most patients who have travelled south during the winter months report improvement in mood a few days after their arrival and relapse a few days after returning to the north. In addition, reduction in environmental light for any reason, for example a spell of cloudy weather in the summer, or a move from a well-lit to a poorly-lit office, is frequently followed by deterioration in mood. Over half the women studied report premenstrual mood changes, which are often worse in the winter.
Clinical and Demographic Features of SAD (N=220)
Because of the atypical vegetative symptoms, which frequently dominate the clinical picture and often precede the affective symptoms, patients with SAD often consult general medical practitioners. Differential diagnoses include other conditions that produce sustained decreases in energy and increased intake of carbohydrates. Thus patients have frequently been worked up for hypothyroidism, hypoglycemia, and infectious mononucleosis, generally with negative results.
SAD has been described in children,8 in whom the syndrome does not generally present as a predominantly affective picture but rather with fatigue, irritability, difficulty getting out of bed in the morning, and school problems. However, sadness, crying spells, and carbohydrate craving may also occur. Children usually do not experience the condition as an internal change, but as a change in the external world. Thus they often feel as though parents or teachers are being unfairly and excessively demanding. One way to distinguish SAD from school phobia is by determining when the problems begin. SAD in children rarely begins before December, at least in the Washington, DC area, whereas school phobia is more likely to start at the beginning of the fall semester.
PHOTOTHERAPY FOR SAD
* Review of Research
Seasonal rhythms of behavior are common among lower animals and their timing is often influenced by environmental changes of which changes in daylength (photoperiod) is the most important.8 In animals the effects of photoperiod are mediated in many cases by the pattern of nocturnal melatonin secretion by the pineal gland. The timing and pattern of melatonin secretion is influenced by the timing of exposure to environmental light.10 Lewy and colleagues have shown that human melatonin secretion can be suppressed by bright artificial light but not by light of regular indoor intensity.11 This finding stimulated attempts to modify winter depressive symptoms by extending the day with bright artificial light. After this proved successful in a single patient,12 studies in four separate groups ot patients at the NIMH showed that extending the day with five to six hours of bright, full-spectrum light had significant antidepressant effects whereas dimmer, control light treatments did not.3,13-15 This finding has been replicated recently by other groups.10
Symptoms of Depression in SAD (N=220)
Because so many groups have replicated the basic finding that bright artificial light has antidepressant effects in SAD, and there have not to date been any studies to the contrary, it is now generally accepted that phototherapy is a viable treatment for the condition.6,7,16-20 However, those properties of phototherapy that are necessary or optimal for achieving an antidepressant effect continue to be a focus of interest and research. It appears that two hours of light treatment in the morning may be more effective than two hours in the evening.17,20 However, other studies have shown that light treatment may be quite effective if administered only in the evening14,14,21 or even during the day.1,22,25 Although most studies have used full-spectrum light, there is evidence that it is not necessary to achieve an antidepressant effect.7 and we do not know at this time which spectral properties of light are necessary for its antidepressant effect. Although early studies used five to six hours of light per day, more recent studies suggest that two hours of treatment per day may be effective.17,20 However, there does seem to be some relationship between duration and efficacy, with greater effects occurring as duration is increased.0 20 Despite its high success rate in the treatment of SAD. phototherapy has been less successful in the treatment of nonseasonal depressions.7,25
* Practical Considerations
This section describes how phototherapy has usually been administered in the six years since it was first used. We recognize that this is not the only possible way to administer it. nor even necessarily the best. However, it is safe, well tested, and has proven to be effective in approximately 80% of the 112 patients treated at the NIMH and in a high percentage of eases treated at other centers as well.
The light source most frequently used has been full-spectrum fluorescent light (Vitalit®). Six or eight 40-watt tubes have been inserted into a rectangular metal fixture, 2 by 4 feet, with a reflecting surface behind them and a plastic diffusing screen in front. Patients have been asked to place the light box at eye level, either horizontally on a desk or table, or vertically on the floor. They have been asked to sit approximately 3 feet away from the lights and to stare at them for a few seconds every minute. The intensity of light measured at 3 feet from this light source is 2500 lux, the amount of light to which one would be exposed by looking out of a window on a spring day. This is five to ten times brighter than ordinary room lighting. We do not know whether it is necessary for patients to glance directly at the lights intermittently, as we have asked them to do, since we do not know where the relevant photoreceptors are located in the retina. If they are the rods on the periphery of the retina, it may be unnecessary for patients to glance at the lights at all. Indeed, only recently has it been shown that the effects of phototherapy in most cases are probably mediated by the eye and not the skin.21
In treating a patient with SAD, the initial suggestion for timing and duration of treatment is a matter of clinical judgment; the convenience and time availability of the patient are relevant considerations in this regard. Since it does not appear that timing is crucial, it seems reasonable initially to recommend times that are convenient for the patient. It is advantageous to establish in the first week whether the patient is responsive to phototherapy. Therefore we generally begin by recommending at least four hours of treatment per day. If this is successful, the patient can reduce the duration later. An alternative approach would be to begin with two hours of treatment in the morning, a regimen found to be effective in a high percentage of patients.17,20 When a response occurs, it is almost always apparent within the first four days.15 If no response occurs within the first week, one can either increase the duration or alter the timing. Treatment may be given in conjunction with antidepressant medications, an approach which may enable the patient to be treated with a lower dosage of medication.
Although side effects are uncommon, patients sometimes complain of irritability (of the kind seen in hypomania). eyestrain, headaches, or insomnia. The latter is most likely to occur when patients use lights late at night. Side effects can generally be reversed easily by decreasing the duration of treatments or suggesting that patients sit further from the light source. In one case, however, we had to discontinue treatment altogether because of severe eye irritation. Although we have seen hypomanic responses in several cases, we have seen no cases of florid mania following phototherapy in typical SAD patients. We did observe mania in one atypical patient, who had a history of both anorexia and bulimia. Fortunately, we have not to date encountered any evidence of longterm adverse effects of phototherapy.
If for any reason light treatment is judged to be unsuitable for a patient with SAD, antidepressant medications can often be effective. There have not to our knowledge been any controlled trials in which different antidepressants have been compared for efficacy in the treatment of SAD. We have observed successful outcomes with tricyclics, lithium, and monoamine oxidase inhibitors. However, often the antidepressant effects are incomplete or side effects are unacceptable, in which case supplementation with light may improve the outcome. We have treated three SAD patients successfully with atenolol, 100 mg per day, although for most SAD patients this drug was no better than placebo.26
MECHANISMS OF ACTION OF PHOTOTHERAPY
We do not know at this time how phototherapy exerts its antidepressant effects. Two main theories have been suggested: first, that phototherapy acts by modifying melatonin secretion27; and second, that it acts by correcting abnormally timed circadian rhythms.17 There are several lines of research that argue against both of these hypotheses. A major problem with the melatonin hypothesis is that the effects of phototherapy do not seem to depend on extending the normally illuminated hours of the day, nor on the capacity of light to suppress melatonin secretion.22 A major problem with the circadian phase correcting hypothesis is that phototherapy has been shown to be effective when administered in the morning, the evening, or during the day. On the basis of the animal literature, one would predict that these interventions would have different effects on the timing of circadian rhythms.
We are thus left without any plausible comprehensive theories of the mechanism of phototherapy. However, it seems likely that light, entering via the eye, modifies brain chemistry and physiology in such a way as to correct the abnormalities resulting from light deficiency in vulnerable individuals. This modification seems reversible and when light is withdrawn, relapse occurs. It appears that this photochemical effect of light requires that the light be of a sufficient intensity and administered for a sufficient duration. There also seems to be a circadian variation in the sensitivity of the reaction, with morning probably being a more sensitive time than evening. The hypothalamus seems a likely site of action because it mediates many of the vegetative functions that are disturbed in SAD and reversed by phototherapy.
However, despite the absence of a cohesive theory explaining the seasonal changes in SAD and the antidepressant effects of phototherapy, research in this area is particularly promising for several reasons. First, these patients become depressed in a regular and predictable way; they can be followed longitudinally and studied prospectively in their different phases. Second, the depressions can be reversed by a nonpharmacological manipulation (phototherapy), and this effect can be reversed by discontinuing light treatment. Third, there are many animal models on which we can draw for inspiration. Of these models, those involving seasonal rhythms of energy conservation, such as weight gain, daily torpor, hibernation, and estivation seem particularly relevant because many symptoms of SAD could be viewed as energy conserving behaviors.27
REVERSE-SAD (SUMMER DEPRESSIONS) AND PHOTOTHERAPY
Although patients with regularly occurring summer depressions have been documented since the time of Esquirol, it is only very recently that they have become the focus of systematic study. Wehr et al described twelve patients with the condition,4 all of whom had a lifetime diagnosis of bipolar affective disorder. Depressions usually began between March and fune and ended between August and October. Manias or hypomanias generally began between September and October and ended between March and May. The symptoms of depression and hypomania were similar to those found in the winter variety of SAD. Boyce has described a similar group of summer depressives in the southern hemisphere (personal communication, 1987).
As with the winter SAD patients, several summer SAD patients noted a relationship between latitude or climate and severity of illness. Some patients reported that they did not have symptoms when they lived further north; others reported that when they travelled south in the spring, their summer depressions began earlier in the year. Conversely, summer vacations in the north were accompanied by clinical improvements, and in some cases improvement occurred when patients bathed in cold lakes.
The striking resemblance between the clinical picture and environmental sensitivity of summer and winter SAD patients, albeit to different environmental stimuli, prompted Wehr and colleagues to explore the effects of having a single patient with summer SAD stay in an air conditioned apartment for several days during the summer and periodically expose herself to very cold temperatures. After five days the patient showed a marked antidepressant response and she relapsed nine days after this treatment was discontinued.
Thus we are at this time in the same situation with summer SAD as we were with winter SAD six years ago. The syndrome has been described and an individual patient has been treated with an environmental manipulation. Controlled studies are now required to explore the clinical usefulness of temperature modifications in this group of patients.
It has been useful to categorize patients with recurrent affective disorders according to the seasonal pattern of their episodes. This pattern has provided information that may be relevant to etiology and has led to the development of specific treatments. So far these treatments have included manipulation of environmental light and temperature. While the efficacy of the former treatment is by now well established, the latter treatment requires further study before its usefulness can be determined.
Besides having treatment implications, manipulation of environmental variables to which individuals are sensitive may help clarify the nature of their underlying vulnerability, which is not well understood at present. Although we were not aware of the early European pioneers in the field of environmental psychiatry when we first started our studies, it has since become apparent that we are following in their tradition, which began long before effective psychotropic medications were developed. While the psychotropic medications continue to be a mainstay in the treatment of affective disorders, they have certain problems and limitations. It is therefore useful to have at our disposal effective nonpharmacological tools as an alternative or supplement to existing treatments.
1. Esquirol E: Mental Maladies: Treatise on Insanity, Hunt EK (ed). Philadelphia. Lea & Blanchard. 1845. pp 275-315.
2. Kraepelin E: Manic-Depressiv Illness and Paranoia. Barclay RM (trans), Robertson GM. Livingstone E, Livingstone M (eds). Edinburgh. E & S Livingstone. 1921.
3. Rosenthal NE. Sack DA. Gillin |C, et al: Seasonal affective disorder: A description of the syndrome and preliminary findings with light therapy. Arch Gen Psychiatry 1984: 41:72-80.
4. Wehr TA. Sack DA, Rosenthal NE: Reverse seasonal affective disorder with spring-summer depression and fallwinter hypomania: Possible relationship to environmental temperature. Submitted for publication. 1986.
5. Spitzer RL. Endicott J, Robins E: Research diagnostic criteria: Rationale and reliability. Arch Gen Psychiatry 1978: 35:773-782.
6. Wirz-Justice A. Buchelli C. Graw P, et al: Light treatment of seasonal affective disorder in Switzerland. Acta Psychiatr Scand 1986; 74:1 93-204.
7. Yerevanian BI. Anderson IL. Grota L|. et al: Effects of bright incandescent light on seasonal and non-seasonal major depressive disorder. Psychiatry Res 1986; 18:355-364.
8. Rosenthal NE. Carpenter C), |ames SP. et al: Seasonal affective disorder in children and adolescents. Am J Psychiatry 1986; 143:356-358.
9. Gwinner E: Circannual systems, in Aschoff J (ed): Handbook of Behavioral Neurobiology. New York. Plenum Press, 1981. vol 4, pp 382-389.
10. Tamarkin L, Baird CJ, Almeida OFX: Melatonin: A coordinating signal for mammalian reproduction? Science 1985; 227:714-720.
11. Lewy AJ. Wehr TA, Goodwin FK, et al: Light suppresses melatonin secretion in humans. Science 1980; 210:1267-1269.
12. Lewy AJ. Kern HA, Rosenthal NE, et al: Bright artificial light treatment of a manic-depressive patient with a seasonal mood cycle. Am f Psychiatry 1982: 139:1496-1498.
13. Rosenthal NE. Sack DA. Carpenter CJ. et al: Antidepressant effects of light in seasonal affective disorder. Am / Psychiatry 1985; 142:606-608.
14. James SP. Wehr TA. Sack DA. et al: Treatment of seasonal affective disorder with light in the evening. Br I Psychiatry 1985; 147:424-428.
15. Rosenthal NE. Sack DA, James SP, et al: Seasonal affective disorder and phototherapy. Ann NY Acad Set 1985; 453:260-269.
16. Checkley S. Winton F. Francy C. et al: Antidepressant effects oflight in seasonal affective disorder. Presented at the Royal College of Psychiatry. Southampton. England, lune 1986.
17. Lewy AJ. Sack RL. Miller S. et al: Antidepressant and circadian phase-shifting effects of light. Science 1987; 235: 352-354.
18. Thompson C. Isaacs G. Miles A: Seasonal affective disorder, phototherapy and salivary melatonin. Proceedings of the Centennial Celebration. Psychiatrische Universitatsklinik, Basel. Abstract 136b, lune 1986.
19. Hellekson CJ. Kline JA, Rosenthal NE: Phototherapy for seasonal affective disorder in Alaska. Am I Psychiatry 1986; 143:1055-1037.
20. Terman M. Quitkin FM.Terman JS: Light therapy for SAD: Dosing regimens. New Research Abstracts. Annual Meeting of the American Psychiatric Association, 1986; 121:77.
21. Wehr TA. Skwerer RG. Iacobsen FM. et al: Eye- versus skin-phototherapy of seasonal affective disorder. Am J Psychiatry 1987; 144:753-766.
22. Wehr TA. Iacobsen FM, Sack DA, et al: Phototherapy in seasonal affective disorder: Time of day and suppression of melatonin are not critical for antidepressant effects. Arch Gen Psychiatry 1986; 43:870-875.
23. Jacobsen FM, Wehr TA. Skwerer RA. et al: Morning versus midday phototherapy of seasonal affective disorder. 1986. in press.
24. Wirz-Justice A. Buchelli C. Graw P. et al: How much light is antidepressant? Psychiatry Res 1986; 17 : 75-77.
25. Kripke DF: Therapeutic effects of bright light in depression. Ann NY Acad Sci 1985; 453:270-281.
26. Rosenthal NE. Sack DA. Iacobsen FM. et al: The role of melatonin in seasonal affective disorder (SAD) and phototherapy, in Wurtman RJ. Waldhauser F: Melatonin in Humans. Cambridge. Massachusetts. Center for Brain Sciences and Metabolism Charitable Trust. 1985.
Clinical and Demographic Features of SAD (N=220)
Symptoms of Depression in SAD (N=220)