Total
Sleep Deprivation Endogenous Depression Willem van den Burg, Rutger H.
van
den
Hoofdakker, MD,
PhD
\s=b\ Ten endogenous depressive patients were deprived of sleep for two whole nights according to the following schedule:
sleep/sleep deprivation/sleep/sleep deprivation/sleep. No drugs were administered. Experimental conditions were as neutral as possible. Blind and nonblind ratings were taken. The patients were generally rated as improved after sleep deprivation, but a substantial effect, though temporary with rapid relapse, occurred in only two cases. After subsequent sleep, relapse followed as a rule. The net antidepressive effect of the total procedure was slightly more than nil. (Arch Gen Psychiatry 32:1121-1125, 1975) first
the
beneficial
reports Thesleepknow,deprivation (SD) depression accidental on
on
effects
concern,
as
on
of far
a few observations made by Schulte.1·2 Some of his depressive patients had noticed that a night of complete wakefulness resulted in a subse¬ quent decrease in depressiveness and an increase in their capability to work. They deliberately made use of this expe¬ rience. These observations prompted Pflug and Tolle3 to investigate the therapeutic value of SD in hospitalized de¬ pressive patients; later Pflug4 also studied the effects in outpatients. Their results indicated a great antidepressive potency of SD, that is in endogenous depression. In this diagnostic category, it would provide "a push towards re¬ covery."5 One night of SD nearly always resulted in a considerable improvement the next day, especially on the following items: depressive mood, suicidality, and psy¬ chomotor inhibition. Although after the next night of sleep some patients relapsed (to variable extents), re¬ peated SD generally accomplished a lasting improvement. Neurotic depressions, on the other hand, were reported not to benefit much by SD. In both groups, SD was nearly as we
Accepted for publication Feb 24, 1975. From the Department of Biological Psychiatry, Psychiatric University Clinic, State University, Groningen, Netherlands. Reprint requests to the Department of Biological Psychiatry, University Hospital, Oostersingel 59, Groningen, Netherlands (Mr. van den Burg).
always combined with the administration of anti¬ depressants. This measure was considered to prevent re¬ lapses. Little is known about variables other than SD and anti-
depressive drug treatment, which might have been impor¬ tant in establishing these results. Years of research with volunteered subjects have shown that any dramatic spe¬ cific effects of SD, except for an increasing inclination to go to sleep, are highly dependent on other factors than SD per se. Such important factors are the subject's mental status, his intentions and expectations, and the psycholog¬ ical characteristics of the experimental setting, including
the attitude and biases of his observers.6 However, from the reports of Pflug and Tolle, the following can be gath¬ ered: (1) SD took place in a setting where this, as a neces¬ sary preliminary to special neurological examinations, was a normal routine. The hospitalized patients were in¬ formed that their staying awake served this diagnostic purpose. The outpatients were told that SD would benefit their mental health. Thus, it seems likely that both pa¬ tients and nurses had a positive attitude toward the proce¬ dure. (2) The ratings were taken by one rater. No informa¬ tion is given that he scored "blindly." Furthermore, it is not clear whether or not the diagnosis of endogenous or neurotic depression was made by an independent judge or by the rater himself. The aim of the present study was to measure the effects of SD in patients with the symptomatology of endogenous depression as "clean" as possible. To this end, the follow¬ ing measures were taken: (1) no antidepressive drugs were given; (2) care was taken to exclude possible extra psychological factors, as far as possible; and (3) use was made of both blind and nonblind raters. METHODS Ten patients, natives of the rural province of Groningen in the Netherlands, were successively admitted to the study. They were accepted if the following was present (1) abnormal, persistent de-
Downloaded From: http://archpsyc.jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/12264/ by a University of California - San Diego User on 01/17/2017
Fig 1 .—Sums of ranks assigned for each patient to his depression ratings on days after sleep deprivation and after nights of sleep. The depressed a patient, the higher the rank.
more
affect was the central feature of their disorder; (2) this feature was accompanied by some of the following symptoms: morbid guilt, hypochondriasis, anorexia, loss of libido, sleep dis¬ turbance, retardation, and agitation; and (3) no precipitating life stress could provide an adequate explanation for the present de¬ pression. Patients satisfying these criteria were considered as having conditions of symptomatologically endogenous ("vital") depressions. The patient group (nine women, one man; aged be¬ tween 31 and 65 years; mean age, 53) was rather seriously de¬ pressed; the mean Hamilton' rating scale score assessed before the experimental period was 25.2 (standard deviation 6.6). The SD took place for two nights according to the following schedule: sleep/SD/sleep/SD/sleep. If a patient was taking any medication, it had been stopped at least three days before (no ef¬ fects of withdrawal were noted). The patients were not greatly motivated; they were simply told that this procedure might help their conditions, which seemed rather contradictory to them. Of course, consent was obtained for each patient. Also, SD did not fit into the therapeutic or diagnostic "culture" of the ward. The atti¬ tude of the nurses could be described as neutral, though sceptical, as a predominant value system dictated that sleep is good and that a lack of sleep is bad for patients. During the SD nights, the patients sat in the common room of the ward under the supervision of night staff. Notes were taken on their clinical condition. In order to avoid, as far as possible, extra psychological influences, no special attention was given to them. Daily and nightly naps were prevented. During the days, the clinical condition was rated by both a nonblind rater—a research nurse, who scored three times daily; morn¬ ing, noon, and late afternoon; and a blind rater—a resident psychi¬ atrist, not connected with the research ward, who scored twice a day, morning and mid-afternoon. Seven blind raters were em¬ ployed, the same one for each patient. Each rater scored no more than two patients in order to prevent the possibility of the rater becoming familiar with the SD schedule. Ratings were taken on symptom lists (7-point scales) especially designed for each individual patient and constructed in coopera¬ tion with the particular raters concerned. The aim was to make
pressed
=
very short interviews possible and to make scoring easier and more reliable. Furthermore, it was hoped that this shortness of the blind raters' interviews might diminish the risk of their being informed about the SD schedule. Patients and raters were asked not to talk about sleep. Blind and nonblind ratings were taken at different times. For self-ratings, a visual analogue method8'9 was used.
RESULTS General Effects of Sleep
Deprivation
first analysis of the data, for each patient the five daily scores of the raters were ranked from 1 (least depressive) to 5 (most depressive). The sums of these ranks per day indicate on which days the patients, on the whole, were rated to be more or less depressive. Figure 1 gives the results. The pattern is as follows: improvement after the first SD night, a relapse after the following night of sleep, again an improvement after the second SD night, and again a relapse after the next night of sleep. A Freedman test10 against the hypothesis of non-random¬ ness of these data was significant for the blind raters' data (P
Sleep Deprivation Endogenous Depression Willem van den Burg, Rutger H.
van
den
Hoofdakker, MD,
PhD
\s=b\ Ten endogenous depressive patients were deprived of sleep for two whole nights according to the following schedule:
sleep/sleep deprivation/sleep/sleep deprivation/sleep. No drugs were administered. Experimental conditions were as neutral as possible. Blind and nonblind ratings were taken. The patients were generally rated as improved after sleep deprivation, but a substantial effect, though temporary with rapid relapse, occurred in only two cases. After subsequent sleep, relapse followed as a rule. The net antidepressive effect of the total procedure was slightly more than nil. (Arch Gen Psychiatry 32:1121-1125, 1975) first
the
beneficial
reports Thesleepknow,deprivation (SD) depression accidental on
on
effects
concern,
as
on
of far
a few observations made by Schulte.1·2 Some of his depressive patients had noticed that a night of complete wakefulness resulted in a subse¬ quent decrease in depressiveness and an increase in their capability to work. They deliberately made use of this expe¬ rience. These observations prompted Pflug and Tolle3 to investigate the therapeutic value of SD in hospitalized de¬ pressive patients; later Pflug4 also studied the effects in outpatients. Their results indicated a great antidepressive potency of SD, that is in endogenous depression. In this diagnostic category, it would provide "a push towards re¬ covery."5 One night of SD nearly always resulted in a considerable improvement the next day, especially on the following items: depressive mood, suicidality, and psy¬ chomotor inhibition. Although after the next night of sleep some patients relapsed (to variable extents), re¬ peated SD generally accomplished a lasting improvement. Neurotic depressions, on the other hand, were reported not to benefit much by SD. In both groups, SD was nearly as we
Accepted for publication Feb 24, 1975. From the Department of Biological Psychiatry, Psychiatric University Clinic, State University, Groningen, Netherlands. Reprint requests to the Department of Biological Psychiatry, University Hospital, Oostersingel 59, Groningen, Netherlands (Mr. van den Burg).
always combined with the administration of anti¬ depressants. This measure was considered to prevent re¬ lapses. Little is known about variables other than SD and anti-
depressive drug treatment, which might have been impor¬ tant in establishing these results. Years of research with volunteered subjects have shown that any dramatic spe¬ cific effects of SD, except for an increasing inclination to go to sleep, are highly dependent on other factors than SD per se. Such important factors are the subject's mental status, his intentions and expectations, and the psycholog¬ ical characteristics of the experimental setting, including
the attitude and biases of his observers.6 However, from the reports of Pflug and Tolle, the following can be gath¬ ered: (1) SD took place in a setting where this, as a neces¬ sary preliminary to special neurological examinations, was a normal routine. The hospitalized patients were in¬ formed that their staying awake served this diagnostic purpose. The outpatients were told that SD would benefit their mental health. Thus, it seems likely that both pa¬ tients and nurses had a positive attitude toward the proce¬ dure. (2) The ratings were taken by one rater. No informa¬ tion is given that he scored "blindly." Furthermore, it is not clear whether or not the diagnosis of endogenous or neurotic depression was made by an independent judge or by the rater himself. The aim of the present study was to measure the effects of SD in patients with the symptomatology of endogenous depression as "clean" as possible. To this end, the follow¬ ing measures were taken: (1) no antidepressive drugs were given; (2) care was taken to exclude possible extra psychological factors, as far as possible; and (3) use was made of both blind and nonblind raters. METHODS Ten patients, natives of the rural province of Groningen in the Netherlands, were successively admitted to the study. They were accepted if the following was present (1) abnormal, persistent de-
Downloaded From: http://archpsyc.jamanetwork.com/pdfaccess.ashx?url=/data/journals/psych/12264/ by a University of California - San Diego User on 01/17/2017
Fig 1 .—Sums of ranks assigned for each patient to his depression ratings on days after sleep deprivation and after nights of sleep. The depressed a patient, the higher the rank.
more
affect was the central feature of their disorder; (2) this feature was accompanied by some of the following symptoms: morbid guilt, hypochondriasis, anorexia, loss of libido, sleep dis¬ turbance, retardation, and agitation; and (3) no precipitating life stress could provide an adequate explanation for the present de¬ pression. Patients satisfying these criteria were considered as having conditions of symptomatologically endogenous ("vital") depressions. The patient group (nine women, one man; aged be¬ tween 31 and 65 years; mean age, 53) was rather seriously de¬ pressed; the mean Hamilton' rating scale score assessed before the experimental period was 25.2 (standard deviation 6.6). The SD took place for two nights according to the following schedule: sleep/SD/sleep/SD/sleep. If a patient was taking any medication, it had been stopped at least three days before (no ef¬ fects of withdrawal were noted). The patients were not greatly motivated; they were simply told that this procedure might help their conditions, which seemed rather contradictory to them. Of course, consent was obtained for each patient. Also, SD did not fit into the therapeutic or diagnostic "culture" of the ward. The atti¬ tude of the nurses could be described as neutral, though sceptical, as a predominant value system dictated that sleep is good and that a lack of sleep is bad for patients. During the SD nights, the patients sat in the common room of the ward under the supervision of night staff. Notes were taken on their clinical condition. In order to avoid, as far as possible, extra psychological influences, no special attention was given to them. Daily and nightly naps were prevented. During the days, the clinical condition was rated by both a nonblind rater—a research nurse, who scored three times daily; morn¬ ing, noon, and late afternoon; and a blind rater—a resident psychi¬ atrist, not connected with the research ward, who scored twice a day, morning and mid-afternoon. Seven blind raters were em¬ ployed, the same one for each patient. Each rater scored no more than two patients in order to prevent the possibility of the rater becoming familiar with the SD schedule. Ratings were taken on symptom lists (7-point scales) especially designed for each individual patient and constructed in coopera¬ tion with the particular raters concerned. The aim was to make
pressed
=
very short interviews possible and to make scoring easier and more reliable. Furthermore, it was hoped that this shortness of the blind raters' interviews might diminish the risk of their being informed about the SD schedule. Patients and raters were asked not to talk about sleep. Blind and nonblind ratings were taken at different times. For self-ratings, a visual analogue method8'9 was used.
RESULTS General Effects of Sleep
Deprivation
first analysis of the data, for each patient the five daily scores of the raters were ranked from 1 (least depressive) to 5 (most depressive). The sums of these ranks per day indicate on which days the patients, on the whole, were rated to be more or less depressive. Figure 1 gives the results. The pattern is as follows: improvement after the first SD night, a relapse after the following night of sleep, again an improvement after the second SD night, and again a relapse after the next night of sleep. A Freedman test10 against the hypothesis of non-random¬ ness of these data was significant for the blind raters' data (P
