Sleep Disturbance in Schizophrenia A Revisit Louis Reich, MD; Brian L. Weiss, MD; Patricia Coble, RN; Richard McPartland, MEE; David J. Kupfer, MD
Rapid eye movement (REM) sleep mechanisms may play a role in the pathophysiology of schizophrenia, but results have been inconclusive and studies of sleep in schizophrenia have been deficient in identifying, comparing, and differentiating between subcategories of schizophrenia. Twenty-nine hospitalized, drug-free schizophrenics were divided into three subgroups\p=m-\acute,latent, and schizoaffective. The REM intensity measures and REM latency were found to differentiate significantly the schizoaffective group. Sleep-continuity indexes separated the latent and acute groups. Seven patients who later required treatment with tricyclic antidepressants had base line REM latencies significantly lower and hospitalizations significantly more prolonged than the patients who did not require antidepressants. Sleep measurements may thus identify diagnostic subgroups of schizophrenia as well as predict which schizophrenic patients will have either a prolonged postpsychotic depressive syndrome or a concurrent affective syndrome.
Since
clinical psychiatrists have often suggested that troubled sleep represents a nonspecific symptom of a troubled mind,1·2 and that dream material and psychotic thinking are related, a number of investigators have de¬ voted considerable energy to studying the electroencephalographic patterns of schizophrenic patients.2 4 In par¬ ticular, it has been hypothesized that the physiological processes associated with the dreams of sleep are related to "daytime" hallucinations, and indeed many earlier studies in actively ill or remitted chronic schizophrenics attempted to find a relationship between "nighttime" dreams and "daytime" hallucinations. Several recent re¬ views of sleep and mental illness point out that studies of sleep in schizophrenia are deficient with respect to defin¬ ing the patients' clinical states namely differentiating Accepted for publication Sept 6, 1974. From the Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine. Reprint requests to the Department of Psychiatry, University of Pittsburgh School of Medicine and Western Psychiatric Institute and Clinic, 3811 O'Hara St, Pittsburgh, PA 15261 (Dr. Kupfer).
among the
subcategories of schizophrenia and identifying the stage, length, and intensity of the illness.2·' Despite current views that abnormalities of the EEG sleep pattern are not definitely related to the behavioral components of the schizophrenic syndrome,2 4 previous studies have sug¬ gested a number of promising leads meriting further in¬ vestigation. In this report, we have studied possible links between the sleep of schizophrenics and a variety of clini¬ cal symptoms, and have examined whether or not base line sleep measurements in untreated schizophrenic pa¬ tients can be used to predict subsequent clinical response to psychotropic medication. METHOD
Twenty-nine hospitalized schizophrenic patients (12 men and 17 with a mean age of 24.0 ± 1.8 years) were studied on the clinical research unit. Demographic and clinical data, consisting of sex, age, number of hospitalizations, presence of hallucinations or delusions on admission, and length of hospital stay are shown in Table 1. A structured diagnostic interview conducted by two sen¬ ior psychiatrists, a self-rating symptom form, and a clinical be¬ havior symptom form were used to establish the diagnosis in each case.6 On the basis of the clinical assessment and ratings, patients were divided into diagnostic subgroups. Fourteen patients were classified acute schizophrenics; this group was characterized by actively psychotic cognitive disturbances and bizarre or inappro¬ priate behavior, highlighted by an acute onset and the presence of either hallucinations or delusions in the absence of any organic cause.7 " Nine patients were grouped as latent schizophrenics (bor¬ derline); they had episodes of dissociative phenomena and contin¬ uous schizophreniform thought disorder (or history of brief psy¬ chotic [minipsychotic] episodes), which for practical purposes remained more or less "covered up" most of the time.'-9 Six pa¬ tients were considered schizoaffective; this group demonstrated a fairly uniform proportion of both schizophrenic and affective symptoms on admission and similar symptomatology was present in their previous episodes.""11 Five patients had actually been hospitalized for depressive episodes, while a sixth had received outpatient treatment for chronic somatic and depressive prob¬ women
lems.
Downloaded From: http://archpsyc.jamanetwork.com/ by a University of Iowa User on 12/16/2015
Patients were studied by continuous nightly recordings of their EEG, horizontal electrooculogram (EOG), and submental electromyogram (EMG) in their own rooms on the clinical research unit for four consecutive nights. No patient received any medica¬ tion prior to or during the period of EEG sleep recordings. In those cases in which medication had been taken prior to hospitalization, sleep studies were not begun until at least two drug-free
Table 1.—Demographic and Clinical Data Total Schizo-
Women Men Mean age, yr First hospitalization Second hospitalization
Multiple (3rd or more hospitalization) Average length current hospitalization, days
Schizo¬
awakening
affective,
Latent, N
=
6
8
20.8
19.0
38.8
78.1
59.2
139.3
84.9
Sleep Parameters*
Schizo-
14
9
6
As shown in Table 2, the most significant features of in the various schizophrenic subgroups can be summarized as follows: acute schizophrenics take a long time to get to sleep (80 minutes) and sleep poorly (43 minutes awake after sleep onset; only 305 minutes sleep time); schizoaffective patients take a long time to get to sleep (102 minutes) but sleep somewhat better thereafter (24 minutes awake after sleep onset; 346 minutes sleep time); in contrast to the relatively low sleep efficiency (TSA/TRP) of 73% in the acute schizophrenic and schiz¬ oaffective groups, borderline patients get to sleep rela¬ tively easily (43 minutes sleep latency) and thereafter sleep reasonably well (87%-TSA/TRP) (12 minutes awake after sleep onset; 375 minutes sleep time). With regard to the other non-REM sleep parameters, the only distin¬ guishing feature was delta sleep. The latent group showed significantly more delta sleep time as compared with the schizoaffective group.
Table 2.— Non-REM
=
=
sleep continuity
ies on three-point scales for the clinical presence of hallucinations and delusions. A score of 2 indicated the definite presence of hallu¬ cinations or delusions; 1 represented their equivocal presence; and a zero score denoted their absence. The individual scores were then correlated with selected sleep measurements for the entire group of 29 schizophrenics and also for the group of 14 acute schiz¬ ophrenics. During the time sleep was studied, a ward physician completed a symptom discrimination checklist, KDS-7,16 on 24 of the patients (13 acute, 7 latent, and 4 schizoaffective). The psy-
Acute,
14
Latent, affective,
RESULTS
trifluoperazine hydrochloride) chlorpromazine hydrochloride during hospitalization, while one was treated exclusively with an¬ tidepressants and two required no psychotropic medication. For purposes of statistical analysis, we used a standardized equivalent dose range,71115 so that 8 mg of trifluoperazine was considered equivalent to 100 mg of chlorpromazine. Medications were given in divided dosages, with a daily range of 400 to 1,000 mg of chlor¬ promazine (32 to 80 mg of trifluoperazine). In addition, it was noted which patients were treated with and discharged on anti¬ depressant medication (amitriptyline hydrochloride). All patients were rated by two clinicians prior to the sleep stud¬
=29
N
=
chosis and depression scores obtained from the KDS-7 were com¬ pared with the mean sleep values. Non-paired t tests (two-tailed) and Pearsonian correlations were performed to determine statis¬ tical confidence levels (P
Rapid eye movement (REM) sleep mechanisms may play a role in the pathophysiology of schizophrenia, but results have been inconclusive and studies of sleep in schizophrenia have been deficient in identifying, comparing, and differentiating between subcategories of schizophrenia. Twenty-nine hospitalized, drug-free schizophrenics were divided into three subgroups\p=m-\acute,latent, and schizoaffective. The REM intensity measures and REM latency were found to differentiate significantly the schizoaffective group. Sleep-continuity indexes separated the latent and acute groups. Seven patients who later required treatment with tricyclic antidepressants had base line REM latencies significantly lower and hospitalizations significantly more prolonged than the patients who did not require antidepressants. Sleep measurements may thus identify diagnostic subgroups of schizophrenia as well as predict which schizophrenic patients will have either a prolonged postpsychotic depressive syndrome or a concurrent affective syndrome.
Since
clinical psychiatrists have often suggested that troubled sleep represents a nonspecific symptom of a troubled mind,1·2 and that dream material and psychotic thinking are related, a number of investigators have de¬ voted considerable energy to studying the electroencephalographic patterns of schizophrenic patients.2 4 In par¬ ticular, it has been hypothesized that the physiological processes associated with the dreams of sleep are related to "daytime" hallucinations, and indeed many earlier studies in actively ill or remitted chronic schizophrenics attempted to find a relationship between "nighttime" dreams and "daytime" hallucinations. Several recent re¬ views of sleep and mental illness point out that studies of sleep in schizophrenia are deficient with respect to defin¬ ing the patients' clinical states namely differentiating Accepted for publication Sept 6, 1974. From the Department of Psychiatry, Western Psychiatric Institute and Clinic, University of Pittsburgh School of Medicine. Reprint requests to the Department of Psychiatry, University of Pittsburgh School of Medicine and Western Psychiatric Institute and Clinic, 3811 O'Hara St, Pittsburgh, PA 15261 (Dr. Kupfer).
among the
subcategories of schizophrenia and identifying the stage, length, and intensity of the illness.2·' Despite current views that abnormalities of the EEG sleep pattern are not definitely related to the behavioral components of the schizophrenic syndrome,2 4 previous studies have sug¬ gested a number of promising leads meriting further in¬ vestigation. In this report, we have studied possible links between the sleep of schizophrenics and a variety of clini¬ cal symptoms, and have examined whether or not base line sleep measurements in untreated schizophrenic pa¬ tients can be used to predict subsequent clinical response to psychotropic medication. METHOD
Twenty-nine hospitalized schizophrenic patients (12 men and 17 with a mean age of 24.0 ± 1.8 years) were studied on the clinical research unit. Demographic and clinical data, consisting of sex, age, number of hospitalizations, presence of hallucinations or delusions on admission, and length of hospital stay are shown in Table 1. A structured diagnostic interview conducted by two sen¬ ior psychiatrists, a self-rating symptom form, and a clinical be¬ havior symptom form were used to establish the diagnosis in each case.6 On the basis of the clinical assessment and ratings, patients were divided into diagnostic subgroups. Fourteen patients were classified acute schizophrenics; this group was characterized by actively psychotic cognitive disturbances and bizarre or inappro¬ priate behavior, highlighted by an acute onset and the presence of either hallucinations or delusions in the absence of any organic cause.7 " Nine patients were grouped as latent schizophrenics (bor¬ derline); they had episodes of dissociative phenomena and contin¬ uous schizophreniform thought disorder (or history of brief psy¬ chotic [minipsychotic] episodes), which for practical purposes remained more or less "covered up" most of the time.'-9 Six pa¬ tients were considered schizoaffective; this group demonstrated a fairly uniform proportion of both schizophrenic and affective symptoms on admission and similar symptomatology was present in their previous episodes.""11 Five patients had actually been hospitalized for depressive episodes, while a sixth had received outpatient treatment for chronic somatic and depressive prob¬ women
lems.
Downloaded From: http://archpsyc.jamanetwork.com/ by a University of Iowa User on 12/16/2015
Patients were studied by continuous nightly recordings of their EEG, horizontal electrooculogram (EOG), and submental electromyogram (EMG) in their own rooms on the clinical research unit for four consecutive nights. No patient received any medica¬ tion prior to or during the period of EEG sleep recordings. In those cases in which medication had been taken prior to hospitalization, sleep studies were not begun until at least two drug-free
Table 1.—Demographic and Clinical Data Total Schizo-
Women Men Mean age, yr First hospitalization Second hospitalization
Multiple (3rd or more hospitalization) Average length current hospitalization, days
Schizo¬
awakening
affective,
Latent, N
=
6
8
20.8
19.0
38.8
78.1
59.2
139.3
84.9
Sleep Parameters*
Schizo-
14
9
6
As shown in Table 2, the most significant features of in the various schizophrenic subgroups can be summarized as follows: acute schizophrenics take a long time to get to sleep (80 minutes) and sleep poorly (43 minutes awake after sleep onset; only 305 minutes sleep time); schizoaffective patients take a long time to get to sleep (102 minutes) but sleep somewhat better thereafter (24 minutes awake after sleep onset; 346 minutes sleep time); in contrast to the relatively low sleep efficiency (TSA/TRP) of 73% in the acute schizophrenic and schiz¬ oaffective groups, borderline patients get to sleep rela¬ tively easily (43 minutes sleep latency) and thereafter sleep reasonably well (87%-TSA/TRP) (12 minutes awake after sleep onset; 375 minutes sleep time). With regard to the other non-REM sleep parameters, the only distin¬ guishing feature was delta sleep. The latent group showed significantly more delta sleep time as compared with the schizoaffective group.
Table 2.— Non-REM
=
=
sleep continuity
ies on three-point scales for the clinical presence of hallucinations and delusions. A score of 2 indicated the definite presence of hallu¬ cinations or delusions; 1 represented their equivocal presence; and a zero score denoted their absence. The individual scores were then correlated with selected sleep measurements for the entire group of 29 schizophrenics and also for the group of 14 acute schiz¬ ophrenics. During the time sleep was studied, a ward physician completed a symptom discrimination checklist, KDS-7,16 on 24 of the patients (13 acute, 7 latent, and 4 schizoaffective). The psy-
Acute,
14
Latent, affective,
RESULTS
trifluoperazine hydrochloride) chlorpromazine hydrochloride during hospitalization, while one was treated exclusively with an¬ tidepressants and two required no psychotropic medication. For purposes of statistical analysis, we used a standardized equivalent dose range,71115 so that 8 mg of trifluoperazine was considered equivalent to 100 mg of chlorpromazine. Medications were given in divided dosages, with a daily range of 400 to 1,000 mg of chlor¬ promazine (32 to 80 mg of trifluoperazine). In addition, it was noted which patients were treated with and discharged on anti¬ depressant medication (amitriptyline hydrochloride). All patients were rated by two clinicians prior to the sleep stud¬
=29
N
=
chosis and depression scores obtained from the KDS-7 were com¬ pared with the mean sleep values. Non-paired t tests (two-tailed) and Pearsonian correlations were performed to determine statis¬ tical confidence levels (P
