- Email: [email protected]
A quantitative sleep-EEG study on the effects of benzodiazepine and zopiclone in schizophrenic patients
SCHIZOPHRENIA RESEARCH ELSEVIER
Schizophrenia Research 15 ( 1995) 303-312
A quantitative sleep-EEG study on the effects of benzodiazepine and zopiclone in schizophrenic patients Naofumi Kajimura a,c,,, Masaaki Kato a, Teruo Okuma b, Masanori Sekimoto a, Tsuyoshi Watanabe ~, Kiyohisa Takahashi a a National Center Hospital for Mental, Nervous and Muscular Disorders, National Center of Neurology and Psychiatry, Tokyo, Japan b National Center of Neurology and Psychiatry, Tokyo, Japan ~ Clinical Psychobiology Branch, National Institute of Mental Health, Room 4s-239 Building 10, 9000 Rockville Pike, Bethesda, MD 20892, USA
Received 29 December 1993; revision received 15 July 1994; accepted 29 July 1994
Abstract
Polysomnographic examinations (PSG) were performed on 6 male schizophrenic outpatients who were being treated with benzodiazepine (BZD) hypnotics in combination with neuroleptics and 6 healthy male volunteers. In schizophrenic subjects, zopiclone (ZPC), 15mg/day, was substituted for the BZD hypnotics, and PSGs were recorded again during ZPC therapy. All-night sleep stage scoring was carried out by visual analysis, and computerized period-amplitude analysis of sleep EEG was also performed. The schizophrenics showed marked reduction in the amount of slow-wave sleep (SWS) and in the number of delta half-waves during all-night sleep, especially those with higher amplitude, as compared to the normals. The number of delta half-waves in the patients was markedly reduced during the first sleep cycle. The average amplitude of delta half-waves during all-night sleep in the schizophrenics was significantly lower than that in the normals. The half-wave count of total delta waves in the schizophrenics was higher during treatment with ZPC than with BZDs, although no significant differences were observed in the amount of SWS between the two treatments. Soundness of sleep in the subjective sleep assessment was better evaluated during treatment with ZPC than BZDs. These results suggest that reduction of SWS in schizophrenia may be attributable mainly to the decrease in the number of delta waves with higher amplitude and that ZPC may induce deeper sleep in schizophrenics than BZDs. Kew~,ords: Polysomnogram; Period-amplitude analysis; Benzodiazepine; Zopiclone; Delta wave ; (Schizophrenia)
I. Introduction
It is well known that schizophrenic patients often suffer from insomnia in addition to various psychiatric symptoms. Polysomnographic sleep studies on schizophrenics have revealed a number of abnormalities, including decreased total sleep time and sleep efficiency, reduced slow-wave sleep * Corresponding author. Phone: 301-496-2141. Fax: 301-4965439. 0920-9964/95/$9.50 © 1995 ElsevierScience B.V. All rights reserved SSDI 0920-9964(94)00054-9
(SWS), prolonged sleep latency and shortened R E M latency (Hiatt et al., 1985; Zarcone et al., 1987; Keshavan et al., 1990). Feinberg et al. (1969) found a deficit in SWS mainly in stage 4 sleep of schizophrenics, while Tandon et al. (1992) indicated that shorter R E M latency was observed in schizophrenic patients when compared with normal controls but SWS did not differ between the schizophrenics and the controls. On the other hand, Ganguli et al. (1987) reported, based on quantitative E E G analysis, that the SWS percent
304
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
in schizophrenic patients was similar to that seen in healthy volunteers, but the total delta wave count was significantly less in the schizophrenics than in the controls. Thus, the reduced SWS or the decreased number of delta waves during sleep has not necessarily been recognized as a common characteristic in schizophrenia. In addition, no adequately detailed computerized analyses of delta and other waves during all-night sleep have been performed in schizophrenic subjects. Different types of hypnotic agents, such as barbiturates, antihistamines, sedating antidepressants, sedating neuroleptics and benzodiazepine (BZD) derivatives, have been administered as an adjunct to antipsychotic medication for the treatment of insomnia in schizophrenics (Mendelson, 1987; Morin, 1993). Although these drugs have usually been selected according to the types and severity of insomnia or the accompanied psychiatric symptoms, BZD hypnotics are the most frequently used members, which possess sufficient efficacy, of this drug class (Roth et al., 1988; Plasky, 1991; Morin, 1993). Most of the BZDs, however, inhibit SWS (Pakes et al., 1981). Thus, it may not be appropriate to use BZD hypnotics to treat the insomnia of schizophrenics, which is frequently accompanied with reduced SWS. Zopiclone (ZPC), a member of the cyclopyrrolone compounds, is a new hypnotic structurally unrelated to the BZDs (Julou et al., 1985). It is reported that ZPC acts on the gammaaminobutyric acid (GABA) receptors but at a site distinct from that of the BZDs (Blanchard et al., 1983; Julou et al., 1985). The effects of ZPC on sleep generally resemble those of BZDs, however, several authors reported that, in contrast to the BZDs, ZPC enhanced SWS in normal subjects (Nicholson and Stone, 1983; Billiard et al., 1987). In the present study, all-night polysomnographic examinations (PSG) were performed on 6 healthy male volunteers and 6 male schizophrenic outpatients, and the results of visual sleep-stage scoring and computerized period-amplitude analysis of EEG were compared in the schizophrenics and normals. In the schizophrenic subjects, the findings of EEG analysis and subjective sleep assessment during BZD and ZPC therapy were also compared and reported.
2. Materials and methods
2.1. Subjects Six healthy male volunteers aged 22-38 years (mean age: 27.3 years) and 6 male schizophrenic outpatients aged 22-39 years (mean age: 31.8 years) served as subjects of this study. The duration of illness of the schizophrenic patients was more than 4 years (mean duration: 10.3 years). To the best of our knowledge, there has been no report which indicates significant effects of duration of schizophrenia on the sleep architecture, although some correlations between the psychiatric symptoms and the sleep abnormalities have been suggested in this disease (Ganguli et al., 1987; Tandon et al., 1992). However, an acute phase of schizophrenia may not be appropriate in investigating EEG sleep due to its clinical instability, which may affect sleep architecture. The diagnosis of schizophrenia was based on DSM-lU-R (APA, 1987) criteria and the patients were classified as the disorganized type in 4 and the undifferentiated type in 2 patients. All of the patients were being treated with neuroleptics and anticholinergic drugs and also with BZD hypnotics for more than 8 weeks before the start of this study. The background of the 6 patients, including the doses of neuroleptics, previous BZD hypnotics and ZPC, is summarized in Table 1. Subjects who had a history of drug or alcohol abuse or of significant medical or neurological problems were excluded from the study. Informed consent was obtained from all of the subjects prior to the study.
2.2. Polygraphic recording Polysomnograms were recorded according to the method described in the standardized sleep manual of Rechtschaffen and Kales (1968). EEGs were recorded from disc electrodes placed at Fpl, Fp2, F3, F4, C3, C4, P3, P4, O1, 02, F7, F8, T5, T6, Fz, and Pz (10-20 electrode system) using A1 + A2 for reference. These 16 electrode locations were adopted for further topographic analysis (the results of the topograms will not be presented in this paper). EOGs were recorded monopolarly from both canthi, and EMGs bipolarly from the
305
N. Kajimura et al./Sehizophrenia Research 15 (1995) 303-312
Table 1 Background of the schizophrenics Case
Age
Age at onset
Type
Neuroleptlcs
Previous hypnotics
Zopiclone
1 2 3 4 5 6
27 39 37 22 37 29
13 23 27 18 27 21
Undifferentiated Undifferentiated Disorganized Disorganized Disorganized Disorganized
105 451 72 736 88 296
Nitrazepam Diazepam Triazolam Haloxazolam Nimetazepam Flunitrazepam
15-.22.5 15 15 15 15 15~ 7.5
5 5 0.5 10 5 4
The diagnosis of schizophrenia and type of schizophrenia were made according to DSM-111-R criteria. The drug values represent the daily dose (mg/day). The doses of neuroleptics have been converted to chlorpromazineaccording to the method described by Davis (1080). chin. The EEGs and EOGs were recorded with a time constant of 0.3 s, sensitivity of 10/~V/mm and high-cut filter of 120 Hz, while the E M G conditions were 0.003 s, 3.5 gV/mm and 500 Hz, respectively. Polysomnograms were not only recorded on paper but also recorded simultaneously by an analog tape recorder (XR-7000L) for further computerized analysis. In accordance with the Rechtschaffen and Kales manual (1968), all-night sleep stage scoring was carried out on the C3 E E G for each 20-second time-code-delimited epoch from the polysomnograms by visual analysis. The sleep variables monitored in this study were total sleep time, excluding periods of awakening and movement during the night, sleep latency (latency of the initial stage 2), R E M latency, and percentage of each sleep stage relative to total sleep time. Period-amplitude analysis was performed by the zero-crossing method on the C3 E E G using the Medilog Sleep Analyzing Computer (SAC: DEE-1100) (Smith et al., 1977, 1978), and the delta (0.33 3.0Hz), theta (2.5 7.1Hz), alpha (8.0-12.0Hz), beta (16.0-30.0Hz) and sigma ( 11.4 16.7 Hz) wave counts during all-night sleep were obtained. Concerning the delta waves, the counts during first, second and third sleep cycles (time interval from the end of R E M to the end of next R E M stage) were also measured, respectively. The C3 EEG was used for the computerized analysis in the present study, because the sleep stage scoring in recent sleep studies, including this study, has ordinarily been performed on this lead (Rechtschaffen and Kales, 1968).
With regard to the delta wave, the SAC system measured the amplitude and period of each halfwave, and the half-wave with amplitude of more than 5 #V and duration between 0.167 and 1.52 s (since the duration was for half-cycle waveform) was detected as the delta wave. The theta wave was counted as one theta wave once 2 out of 3 consecutive waveforms met the theta criteria and the counting ended once 2 or more waves no longer met the criteria. For the alpha wave, the system counted the beginning of 6 consecutive waves, all of which met the alpha criteria, as one alpha wave and finished the counting once 2 of 6 waves no longer met the alpha criteria. The beta and sigma waves qualified in the same manner as the alpha wave. Delta waves were analyzed in detail by classifying the delta half-wave according to amplitude and frequency. The number of delta half-waves with frequency of 0.5-2.0 Hz and amplitude of > 31 I(V, which were used as a criterion for the scoring of SWS in the sleep staging algorithm of the SAC, was also determined and analyzed. In addition, average amplitude and frequency of delta halfwaves during all-night sleep were obtained. 2.3. Subjective sleep assessment
A rating scale devised by us was used to score subjective assessment of sleep in the schizophrenic subjects. It consisted of 5 items, falling asleep, soundness of sleep, dreams, physical condition and mood in the morning. Soundness of sleep and dreams were evaluated using a 5-point (1-5) scale
306
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
and 3-point (1-3) scale, respectively. The remaining 3 items were assessed on 4-point (1-4) scales. The higher the score was for each item, the worse the evaluation.
neuroleptics or anticholinergic agents should also be considered, further examination is required to clarify this issue.
2.5. Statistical analysis 2.4. Experimental procedure PSGs were performed on 6 healthy male volunteers, and 6 male schizophrenic patients undergoing treatment with neuroleptics, anticholinergic drugs and BZD hypnotics. On the schizophrenic subjects, BZD hypnotics were replaced with ZPC 15 mg/day, and PSGs were recorded again at the end of 8 weeks of ZPC therapy. Subjective sleep assessment was performed in the patients during treatment with BZDs and ZPC. The doses of neuroleptics and anticholinergic drugs remained fixed throughout the study, whereas the subsequent doses of ZPC depended on the judgment of the attending psychiatrist. To exclude the laboratory effect (Agnew et al., 1966), each subject spent an adaptation night in the laboratory one week before each PSG. Disc electrodes were attached to the subjects at 20.00 h. Polygraphic recording was started at 22.00 h and completed at 7.00 h the next morning. Oral BZD hypnotics or ZPC were administered to the schizophrenics at 21.30 h. The rating scale for subjective sleep assessment was filled out after awakening. It is reported that ZPC is rapidly absorbed, achieving peak concentration within 2 h, and has a half-life of 3.5 to 6.5 h in healthy volunteers (Houghton et al., 1985; Julou et al., 1985). On the other hand, the schizophrenic subjects in this study had been chronically taking various BZDs, including nitrazepam, diazepam, triazolam, haloxazolam, nimetazepam or flunitrazepam, before ZPC treatment. Since these BZDs have different potency or half-life, they may possess some different effects on sleep, respectively. However, most of the BZDs have been reported to have basically similar effects on sleep EEG in clinical use, especially during chronic treatment (Pakes et al., 1981). Thus, in the present study, the effect of chronic administration of the BZDs on sleep EEG was summarized and compared with that of ZPC. Although the drug interactions between these hypnotics and
Student's t-test (two-tailed) was used for the statistical analysis of the polysomnogram results, and Wilcoxon's signed-rank test was used for that of the subjective sleep assessment. One way analysis of variance (ANOVA) with repeated measures was also employed for the statistical analysis of the delta half-wave result during the individual sleep cycles, and differences between sleep cycles were explored with post hoc Fisher's least significant difference (LSD) test.
3. Results
During the present study, one schizophrenic patient developed insomnia after the start of ZPC therapy, and the dose of ZPC was increased from 15 to 22.5mg/day. Another patient developed daytime drowsiness with malaise in the second week of ZPC therapy, and the dose of ZPC was decreased from 15 to 7.5 mg/day. These regimens relieved the symptoms, and the two patients were able to continue taking their respective doses of ZPC throughout the remainder of the study. Though bitter taste in the mouth also occurred as an adverse effect in one patient being treated with ZPC, it was mild enough to continue the treatment.
3.1. Sleep variables The sleep variables of the healthy volunteers and the schizophrenic patients are shown in Table 2. The schizophrenics exhibited longer sleep latency and a marked decrease in the amount of SWS in comparison with the normal subjects. The amounts of stage 1 sleep during BZD therapy and stage 2 sleep during ZPC therapy of the schizophrenics were greater than those of the normal subjects. There were no significant differences in the number of awakenings and the total awake time between the normals and the schizophrenics, although these results are not shown in Table 2. However, the
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
307
Table 2 Sleep variables in normal volunteers and schizophrenics Sleep variables Total sleep time Sleep latency REM latency % Stage t % Stage 2 % Stages 3 and 4 % Stage REM
(min) (rain) (min)
N
S (BZDs)
S (ZPC)
485.1 _+11.1 12.0+_3.7 86.1 -+ 12.3 10.0+2.1 56.4-+2.3 13.3+_2.2 20.4±2.0
411.5 ± 36.0* 53.8 _+16.7'* 111.7 _+33.0 16.2 -+ 1.8" 62.9+_3.5 3.1 -+ 1.3"** 17.8-+2.4
436.5 ± 37.2 59.9 ± 19.3'* 45.4 + 20.8 8.3 + 1.8 68.6_+2.2*** 4.0± 1.1 *** 19.0+_2.0
D A
Each value represents the mean ±S.E.M. of the values in 6 subjects. N, normal volunteers; S: schizophrenics. Polysomnographic examination of schizophrenics was performed during benzodiazepine (BZD) hypnotic treatment and at the end of 8 weeks of zopiclone (ZPC) therapy. The existence of statistically significant differences between the normal volunteers and the schizophrenics is indicated by the asterisks after the values for the schizophrenics (*p <0.1; **p <0.05; ***p < 0.01 ). The presence of statistically significant differences between BZD and ZPC therapy in the schizophrenics is indicated by the letters after the values at the end of ZPC treatment (A, p<0.1; D, p<0.001 ).
sleep efficiency (%) of the normal subjects was higher, though not significant, than that of the schizophrenic patients (91.0 ± 1.2 for the normals; 81.5 ± 7.4 for the schizophrenics during BZD therapy; 83.0 ± 7.0 for the schizophrenics during ZPC therapy ). In the schizophrenic patients, the percentage of stage 1 sleep was lower and that of stage 2 was higher during treatment with ZPC than with BZDs. There were no significant differences in sleep latency and in the amounts of SWS and REM sleep between BZD and ZPC treatments.
The schizophrenics showed a marked decrease in the number of delta half-waves of 0.5-2.0 Hz, >__31/~V in comparison with the normals. There were no significant differences in the number of waves of other frequency bands between the patients and the normals. In the schizophrenics, no differences were observed in the number of waves of individual frequency bands between BZD and ZPC treatments.
3.2. Analysis of the activity of individualfrequency bands
Table 4 shows the delta half-wave counts per hour during all-night sleep classified according to the amplitude and frequency. The half-wave count of total delta waves (0.33-3.0 Hz, >_5 ltV) in the schizophrenics was significantly lower than that in
The delta, theta, alpha, beta and sigma wave counts during all-night sleep are shown in Table 3.
3.3. Delta half-wave analysis
Table 3 Delta, theta, alpha, beta and sigma wave counts during all-night sleep Waveforms (Hz)
N
S (BZDs)
S (ZPC)
Delta Theta Alpha Beta Sigma
6162.7_+632.0 3360.2_+1312.1 1184.2_+253.4 819.3-+409.7 1597.2-+342.3
2311.5+634.4"** 1806.5+1115.2 1262.7+324.8 1821.3-+546.0 2451.8_+603.4
2578.2_+551.1"** 1397.2_+832.8 1077.3_+263.7 1369.7_+473.1 2324.5_+355.1
(0.5 2.0) (2.5 7.1) (8.0-12.0) (16.0-30.0) (11.4-16.7)
The delta, theta, alpha, beta and sigma wave counts during all-night sleep are expressed as means+S.E.M, of the values in 6 subjects. The delta wave count is shown as the half-wave count used for slow-wave sleep in the sleep staging algorithm of the SAC (0.5 2.0 Hz, > 31 laV). The alpha, beta and sigma wave counts represent 6 or more consecutive waves, respectively, while the theta wave count is based on 3 or more consecutive waves. Details are the same as in Table 2. ***p <0.01.
308
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
Table 4 Delta half-wave count per hour during all-night sleep classified according to amplitude and frequency N
S (BZDs)
S (ZPC)
Amplitude (~tV) 5-15 15-25 25-35 35-45 45 55 55 <
3940.9+ 173.0 1870.7 + 93.6 611.1 + 47.0 286.4+30.2 147.6+20.2 201.7 + 40.2
4395.7___226.0 1354.4 + 231.0" 381.8 + 67.3"* 139.9+31.1"** 56.2+ 16.0"** 37.3 + 12.5"**
4481.1 _+199.2' 1478.3 + 169.4* 418.0 + 69.6** 154.9+29.5"* 63.6_+ 14.2"** 48.0 _+11.9***
Frequency (Hz) 0.33-0.5 0.5 - 1.0 1.0 -1.5 1.5 -2.1 2.1 -2.6 2.6 -3.0
134.4 + 20.7 1415.7 ___146.3 1739.0+72.7 1449.2___39.5 1166.9 -+ 81.3 1154.2+ 110.3
178.3 _+35.0 1476.3 + 145.2 1565.8+46.7" 1269.0+ 106.1 1007.2 _+131.2 868.1 4-153.5
177.9 ___25.4 1584.2 -+95.7 1647.8-+50.4 1337.4-+68.6 1034.7 _+83.0 863.24-95.5*
Total
7059.5 + 29.6
6366.1 4- 253.5**
6645.1 _+173.2**
A
The delta half-wave count per hour during all-night sleep for each amplitude and frequency band is shown as the mean + S.E.M. of the values in 6 subjects. The delta wave frequency is defined as lying between 0.33 and 3.0 Hz, while the minimum delta wave amplitude is defined as 5 ~tV. Details are the same as in Table 2. *p<0.1; **p<0.05; ***p <0.01; A, p<0.1; B, p <0.05.
the normal subjects. The schizophrenics showed a markedly reduced number of delta half-waves with higher amplitude than the normals. There were no distinctive differences in the half-wave counts of delta waves classified according to frequency between the normal subjects and the patients. In the schizophrenics, the number of total delta half-waves was higher during treatment with ZPC than with BZDs. ZPC therapy was accompanied by higher half-wave counts of delta waves with amplitude above 55/2V and with frequency of 1.0-1.5 Hz than BZD therapy. Table 5 shows the half-wave counts of delta waves (0.33-3.0 Hz, >5/~V) per hour during the first, second and third sleep cycles. The delta halfwave counts during the first and second sleep cycles of the schizophrenics were significantly lower than those of the normal subjects. There was no significant difference, however, in the delta half-wave count during the third sleep cycle between the patients and the normals. In the normal subjects, an ANOVA revealed the statistically significant effect of sleep cycle on the delta half-wave count, and post hoc Fisher's LSD test indicated that, during the first sleep cycle as compared with the second and third cycles, there
were significantly more delta half-waves per hour (p<0.05). In the schizophrenics, however, ANOVAs revealed no statistically significant effect of sleep cycle on the delta half-wave counts per hour during BZD and ZPC treatments.
3.4. Average amplitude and frequency of delta halfwave Table 6 shows the average amplitude and frequency of delta half-waves during all-night sleep. The average amplitude of delta half-waves in the schizophrenic subjects was significantly lower than that in the normals, while there was no difference in the average frequency between the two groups. In the schizophrenics, no differences were observed in the average amplitude and frequency between BZD and ZPC treatments.
3.5. Subjective sleep assessment The score for soundness of sleep in the subjective sleep assessment was lower (evaluated as a good condition) during treatment with ZPC than BZDs (Table 7). There were no significant differences in
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
309
Table 5 Delta half-wave count per hour in first, second and third sleep cycles
Sleep cycle First Second Third ANOVA
N
S (BZDs)
S (ZPC)
7515.8 ___109.2 7229.8 +49.5 7048.0 + 81.2
6656.5 + 180.7"** 6431.3 +255.5** 6355.6 + 357.9"
6759.7 + 190.8"** 6746.4 + 169.8"* A 6770.0 +_192.6
F = 8.953 p=0.0059
F = 1.263 n.s.
F = 0.029 n.s.
The half-wave counts of delta waves (0.33 3.0 Hz, >5 IxV) per hour in the first, second and third sleep cycles are shown as the mean + S.E.M. of the values in 6 subjects. An ANOVA was used to evaluate the effect of sleep cycle on the delta half-wave counts. Other details are the same as in Table 2. *p<0.1; **p<0.05; ***p<0.01; A, p<0.1. Table 6 Average amplitude and frequency of delta half-waves during all-night sleep
Amplitude (IxV) Frequency (Hz)
N
S (BZDs)
S (ZPC)
17.7+0.7 1.09 + 0.05
13.9+0.8"** 0.97 + 0.08
14.3+0.7"** 0.97 + 0.05
The average amplitude and frequency of delta half-waves during all-night sleep are shown as means_+S.E.M, of the values in 6 subjects. Details are the same as in Table2. ***p <0 01. Table 7 Subjective sleep assessment in schizophrenics Items
BZDs
ZPC
Falling asleep Soundness of sleep Dreams Physical condition in the morning Mood in the morning
2.33 _%0.61 3.67 +0.33 1.67 + 0.33 2.50+0.22 2.33+0.21
2.00 + 0.63 2.67___0.49"* 1.50+0.34 2.50+0.22 2.00+0.37
Each score indicates the mean+S.E.M, **p<0.05.
of 6 patients.
regard to other subjective assessment items between the two treatments.
4. Discussion
In the visually scored sleep parameters, the chronic schizophrenic patients in this study showed a prolongation of sleep latency, fairly large amounts of stages 1 and 2 sleep, and very little
SWS when compared to the normal subjects. These findings essentially coincide with the results of previous EEG studies in drug-naive schizophrenics (Feinberg et al., 1969; Keshavan et al., 1990). However, Tandon et al. (1992) found shortened REM latency, rather than reduced SWS, in nonmedicated or drug-free schizophrenic patients. The schizophrenic patients in the present study had been chronically taking neuroleptics and anticholinergic drugs together with BZD hypnotics or ZPC throughout the study. It is reported that both neuroleptics and anticholinergic drugs possess significant inhibitory effects on REM sleep but have no such effects on NREM sleep including SWS (Gillin et al., 1978; Taylor et al., 1991). Thus, the chronic schizophrenics undergoing chronic treatment with the psychotropic drugs may also show NREM abnormalities, including the reduced SWS, of visually scored sleep parameters similar to those of non-medicated or drug-free schizophrenics. It may be difficult, however, to evaluate REM sleep abnormalities, including the shortened REM latency, in the schizophrenic patients with antipsychotic medication. In the period-amplitude analysis, the schizophrenic subjects showed marked reduction in the number of 0.5-2.0 Hz, >31/~V delta half-waves as well as in the number of total delta band half-waves (0.33-3.0 Hz, >5 pV) during sleep in comparison with the normal subjects. In addition, the patients exhibited markedly decreased halfwave count of delta waves with higher amplitude. However, no distinctive difference was found in
310
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
the distribution of delta wave frequency between the schizophrenics and the normals. The finding that the number of total delta waves during sleep is markedly decreased in the schizophrenics is consistent with the previous quantitative sleep EEG study in schizophrenia (Ganguli et al., 1987). However, the detailed delta wave analysis in the present study revealed that the number of delta waves with higher-amplitude was particularly decreased in the schizophrenics. This is also supported by the present finding that the average amplitude of delta waves during all-night sleep in the schizophrenics was significantly lower than that in the normal subjects but no difference was found in the average frequency between the schizophrenics and the normals. Feinberg (1989) suggested the positive correlation between the delta wave amplitude during sleep and the metabolic rate of the cerebral cortex in humans. Thus, the reduced number of delta waves with higheramplitude in the schizophrenics may be involved in the hypoactivity of the prefrontal cortex, which has recently been reported to be associated with negative symptoms in this disease (Andreasen, 1989). These results suggest that the reduction in SWS in schizophrenia may be attributable mainly to the decrease in the number of delta waves with higher amplitude rather than with lower frequency. The schizophrenic subjects, when compared to the normal subjects, showed a decreased number of delta half-waves during the first and second sleep cycles but exhibited no obvious difference in the number of the waves during the third cycle. In addition, the number of delta half-waves per hour in the normal subjects was significantly greater during the first sleep cycle than during other sleep cycles, whereas, in the schizophrenic subjects, there were no significant differences in the delta halfwave counts between the sleep cycles. Thus, the reduced number of delta waves during the first sleep cycle may be a remarkable characteristic of the schizophrenics in this study. This is supported by the finding of Hiatt et al. (1985) that schizophrenic patients showed an abnormally low level of delta count in the first N R E M period. The schizophrenic subjects in this study had continuously been taking neuroleptics and anticholinergic drugs together with BZD hypnotics or
ZPC throughout the present study. It is well known that neuroleptics, anticholinergic agents and the hypnotics possess various effects on delta power in previous spectral EEG analysis (Itil, 1974). Thus, further discussion on sleep EEG of schizophrenia may not be appropriate, and the following is focused on the comparison of sleep EEG between BZD and ZPC treatments in the schizophrenics. In the visually scored sleep parameters, there was less stage 1 and greater stage 2 sleep in the schizophrenics during treatment with ZPC than with BZDs, but there was no obvious difference in the amount of SWS between the treatment with BZDs and ZPC. It is well known that BZD hypnotics increase stage 2 and decrease SWS (Pakes et al., 1981). Billiard et al. (1987) reported that in normal humans ZPC significantly increased the duration of SWS. Nicholson and Stone (1983) also found that stage 2 showed a significant increase and SWS tended to be extended with ZPC treatment in healthy adults. Thus, the result in the present study that no difference was found in the amount of SWS between the BZD and ZPC treatments suggests that it is difficult to induce an increase of SWS in schizophrenia by ZPC therapy because the mechanism of producing delta waves might be deficient in schizophrenic subjects. However, the finding that there was less stage 1 and greater stage 2 sleep during treatment with ZPC than with BZDs suggests that ZPC may induce deeper sleep in schizophrenics than BZDs. This is supported by the finding that soundness of sleep in the subjective sleep assessment was better evaluated during treatment with ZPC than BZDs. It was found in the period-amplitude analysis in the present study that the schizophrenics showed a higher half-wave count of total delta waves during treatment with ZPC than with BZDs. Furthermore, the count of delta half-waves with amplitude of over 55 #V was higher in ZPC therapy than in BZD therapy, although no difference was observed in the average amplitude of delta half-waves between the two treatments. Based on spectral analysis of sleep EEG, Trachsel et al. (1990) found that both ZPC and BZD hypnotics depressed delta power density during sleep. Johnson et al. (1979) also reported that BZD
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
hypnotics decrease delta power during sleep by reducing the amplitude and, to a lesser extent, the number of delta waves, while Wright et al. (1986) claimed that ZPC decreases higher-amplitude delta waves but increases the total number of delta waves. Thus, in normal humans, BZDs and ZPC possess inhibitory effect on the amplitude of delta waves during sleep, but may possess differential effects on the occurrence of delta waves. These results suggest that, in schizophrenics who have fewer delta waves during sleep, inhibitory effect of ZPC on the amplitude of delta waves may be weaker than that of BZDs. It can not be concluded from the present study whether ZPC enhances the occurrence of delta waves during sleep in schizophrenics, however, it may be convincing that BZDs suppress the occurrence of delta waves during sleep as compared with ZPC. No difference was observed in the number of 0.5-2.0 Hz, >__31/~Vdelta half-waves between the treatment with BZDs and ZPC. Thus, delta waves which are used as a criterion for the scoring of SWS, ie., delta waves with higher amplitude and with lower frequency, may not easily be increased in schizophrenia. The schizophrenic subjects showed no differences in the numbers of sigma and beta waves during sleep between BZD and ZPC treatments. It is well known that BZD derivatives increase the amount of sigma waves during sleep and enhance the beta activity (Itil, 1974; Hirshkowitz et al., 1982). It is also reported that ZPC possesses facilitatory effects on sigma and beta waves (Saletu et al., 1987; Iwata et al., 1989). Quadens et al. (1983) stated that ZPC increases the duration of stage 2 sleep, which is characterized by sigma waves, as comparable to flurazepam. Thus, both the BZD and ZPC therapy in the schizophrenics may be accompanied by equivalent increases of sigma and beta waves during sleep, resulting in no differences of these waves between the two treatments. This may be supported by the finding that the schizophrenics showed considerable increases, though not significant, in the numbers of sigma and beta waves during BZD and ZPC treatments in comparison with the normal subjects who received no medication. Although a more definitive study without other
311
concurrent medications and with a larger population and a double-blind crossover study will be necessary, the present findings suggest that lowered amplitude of delta waves during sleep may be involved in the decreased amount of SWS in schizophrenia and that both reduced SWS and decreased count of delta waves with higheramplitude and lower-frequency may not easily be enhanced in schizophrenics. Moreover, ZPC may induce deeper sleep in schizophrenics than BZDs by decreasing stage 1 and increasing stage 2 sleep and by increasing the number of total delta waves during sleep, resulting in better evaluation of soundness of sleep in subjective sleep assessment.
Acknowledgments The authors would like to thank to Dr. Thomas A. Wehr, Chief, Clinical Psychobiology Branch, National Institute of Mental Health, for his useful comments and suggestions.
References Agnew, H.W.Jr., Webb, W.B. and Williams, R.L. (1966) The first night effect: an EEG study of sleep. Psychopharmacologia 2, 263-266. American Psychiatric Association (1987) Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revised. American Psychiatric Association Press, Washington, DC. Andreasen, N.C. (1989) Neural mechanisms of negative symptoms. Br. J. Psychiatry 155 Suppl. 7, 93 98. Blanchard, J.C., Zundel, J.L. and Julou, L. (1983) Differences between cyclopyrrolones (suriclone and zopiclone) and benzodiazepines binding to rat hippocampus photolabelled membranes. Biochem. Pharmacol. 32, 3651-3653. Billiard, M., Besset, A., de Lustrac, C. and Brissaud, L. (1987) Dose-response effects of zopiclone on night sleep and on nighttime and daytime functioning. Sleep 10 Suppl. 1, 27 34. Davis, J.M. (1980) Organic therapies: antipsychotic drugs. In: A.M. Freedman, H.I. Kaplan and B.J. Sadock (Eds.), Comprehensive Textbook of Psychiatry. Williams Wilkins, Baltimore, pp. 2257 2289. Feinberg, I. (1989) Effects of maturation and aging on slowwave sleep in man: implications for neurobiology. In: A. Wauquier, C. Dugovic and M. Radulovacki (Eds.), Slow Wave Sleep: Physiological, Pathophysiological, and Functional Aspects. Raven Press, New York, pp. 31-48. Feinberg, I., Braun, M., Koresko, R.L. and Gottlieb F. (1969)
312
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
Stage 4 sleep in schizophrenia. Arch. Gen. Psychiatry 21,262-266. Gangnli, R., Reynolds, C.F.V. and Kupfer, D.J. (1987) Electroencephalographic sleep in young, never-medicated schizophrenics: a comparison with delusional and nondelusional depressives and with healthy controls. Arch. Gen. Psychiatry 44, 36-44. Gillin, J.C., van Kammen, D.P. and Bunney, W.E. (1978) Pimozide attenuate d-amphetamine-induced sleep changes in man. Life Sci. 22, 1805-1810. Hiatt, J.F., Floyd, T.C., Katz, P.H. and Feinberg, I. (1985) Further evidence of abnormal non-REM sleep in schizophrenia. Arch. Gen. Psychiatry 42, 797-802. Hirshkowitz, M., Thornby, J.I. and Karakan, I. (1982) Sleep spindles: pharmacological effects in humans. Sleep 5, 85-94. Houghton, G., Dennis, M., Templeton, R. and Martin, B. (1985) A repeated dose pharmacokinetic study of a new hypnotic agent, zopiclone (Imovane). Int. J. Clin. Pharmacol. Ther. Toxicol. 23, 97-100. Itil, T.M. (1974) Quantitative pharmaco-electroencephalography. In: T.M. Itil (Ed.), Modern Problems in Pharmacopsychiatry. Karger, Basel, pp. 43-75. Iwata, T., Hayakawa, T., Terashima, M., Ohta, T., Okada T., Ueda, Y. and Ishii, N. (1989) Effects of zopiclone on sleep spindles studied with an improved waveform recognition method. Jpn. J. Psychiat. Neurol. 43, 669-674. Johnson, L.C., Seales, D.M., Naitoh, P., Church, M.W. and Sinclair, M. (1979) The effects of flurazepam hydrochloride on brain electrical activity during sleep. Electroenceph. Clin. Neurophysiol. 47, 309-321. Julou, L., Blanchard, J.C. and Dreyfus, J.F. (1985) Pharmacological and clinical studies of cyclopyrrolones: zopiclone and suriclone. Pharmacol. Biochem. Behav. 23, 653-659. Keshavan, M.S., Reynolds, C.F. and Kupfer, D.J. (1990) Electroencephalographic sleep in schizophrenia: a critical review. Compr. Psychiatry 30, 34-47. Mendelson, W.B. (Ed.) (1987) Human Sleep: Research and Clinical Care. Plenum, New York. Morin, C.M. (Ed.) (1993) Insomnia: Psychological Assessment and Management. Guilford, New York. Nicholson, A.N. and Stone, B.M. (1983) Zopiclone: sleep and performance studies in healthy man. Pharmacology 27 Suppl. 2, 92-97.
Pakes, G.E., Brogden, R.N., Heel, R.C., Speight, T.M. and Avery, G.S. ( 1981) Triazolam: a review of its pharmacological properties and therapeutic efficacy in patients with insomnia. Drugs 22, 81-110. Plasky, P. (1991) Antidepressant usage in schizophrenia. Schizophr. Bull. 17, 649-656. Quadens, O.P., Hoffman, G. and Buytaert, G. (1983) Effects of zopiclone as compared to flurazepam on sleep in women over 40 years of age. Pharmacology 27 Suppl. 2, 146-155. Rechtschaffen, A. and Kales, A. (1968) A manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. US Government Printing Office, Washington, DC. Roth, T., Roehrs. T. and Zorick F (1988) Pharmacological treatment of sleep disorders. In: R.L. Williams, I. Karacan and Moore C.A. (Eds.), Sleep Disorders: Diagnosis and Treatment. John Wiley and Sons, New York, pp. 373-395. Saletu, B., Anderer, P., Kinsperger, K. and Grunberger, J. (1987) Topographic brain mapping of EEG in neuropsychopharmacology-part U: clinical applications (pharmaco-EEG imaging). Meth. Find. Exp. Clin. Pharmacol. 9, 385-408. Smith, J.R., Karakan, I. and Yang, M. (1977) Ontogeny of delta activity during human sleep. Electroenceph. Clin. Neurophysiol. 43, 229-237. Smith, J.R., Karacan, I. and Yang, M. (1978) Automated analysis of the human sleep EEG. Waking Sleeping 2, 75 82. Tandon R., Shipley, J.E., Taylor, S., Greden, J.F., Eiser, A., DeQuardo, J and Goodson, J. (1992) Electroencephalographic sleep abnormalities in schizophrenia: relationship to positive/negative symptoms and prior neuroleptic treatment. Arch. Gen. Psychiatry 49, 185-194. Taylor, S.F., Tandon R., Shipley, J.E. and Eiser, A.S. (1991) Effect of neuroleptic treatment on polysomnographic measures in schizophrenia. Biol. Psychiatry 30, 904-912. Trachsel, L., Dijk, D.J., Brunner, D.P., Klene, C. and Borbely, A.A. (1990) Effect of zopiclone and midazolam on sleep and EEG spectra in a phase-advanced sleep schedule. Neuropsychopharmacology 3, 11-18. Wright, N.A., Belyavin, A., Borland, R.G. and Nicholson, • A.N. (1986) Modulation of delta activity by hypnotics in middle-aged subjects: studies with a benzodiazepine (flurazepam) and a cyclopyrrolone (zopiclone). Sleep 9, 348-352. Zarcone, V.P., Benson, K.L. and Berger, P.A. (1987) Abnormal rapid eye movement latencies in schizophrenia. Arch. Gen. Psychiatry 44, 45-48.
Schizophrenia Research 15 ( 1995) 303-312
A quantitative sleep-EEG study on the effects of benzodiazepine and zopiclone in schizophrenic patients Naofumi Kajimura a,c,,, Masaaki Kato a, Teruo Okuma b, Masanori Sekimoto a, Tsuyoshi Watanabe ~, Kiyohisa Takahashi a a National Center Hospital for Mental, Nervous and Muscular Disorders, National Center of Neurology and Psychiatry, Tokyo, Japan b National Center of Neurology and Psychiatry, Tokyo, Japan ~ Clinical Psychobiology Branch, National Institute of Mental Health, Room 4s-239 Building 10, 9000 Rockville Pike, Bethesda, MD 20892, USA
Received 29 December 1993; revision received 15 July 1994; accepted 29 July 1994
Abstract
Polysomnographic examinations (PSG) were performed on 6 male schizophrenic outpatients who were being treated with benzodiazepine (BZD) hypnotics in combination with neuroleptics and 6 healthy male volunteers. In schizophrenic subjects, zopiclone (ZPC), 15mg/day, was substituted for the BZD hypnotics, and PSGs were recorded again during ZPC therapy. All-night sleep stage scoring was carried out by visual analysis, and computerized period-amplitude analysis of sleep EEG was also performed. The schizophrenics showed marked reduction in the amount of slow-wave sleep (SWS) and in the number of delta half-waves during all-night sleep, especially those with higher amplitude, as compared to the normals. The number of delta half-waves in the patients was markedly reduced during the first sleep cycle. The average amplitude of delta half-waves during all-night sleep in the schizophrenics was significantly lower than that in the normals. The half-wave count of total delta waves in the schizophrenics was higher during treatment with ZPC than with BZDs, although no significant differences were observed in the amount of SWS between the two treatments. Soundness of sleep in the subjective sleep assessment was better evaluated during treatment with ZPC than BZDs. These results suggest that reduction of SWS in schizophrenia may be attributable mainly to the decrease in the number of delta waves with higher amplitude and that ZPC may induce deeper sleep in schizophrenics than BZDs. Kew~,ords: Polysomnogram; Period-amplitude analysis; Benzodiazepine; Zopiclone; Delta wave ; (Schizophrenia)
I. Introduction
It is well known that schizophrenic patients often suffer from insomnia in addition to various psychiatric symptoms. Polysomnographic sleep studies on schizophrenics have revealed a number of abnormalities, including decreased total sleep time and sleep efficiency, reduced slow-wave sleep * Corresponding author. Phone: 301-496-2141. Fax: 301-4965439. 0920-9964/95/$9.50 © 1995 ElsevierScience B.V. All rights reserved SSDI 0920-9964(94)00054-9
(SWS), prolonged sleep latency and shortened R E M latency (Hiatt et al., 1985; Zarcone et al., 1987; Keshavan et al., 1990). Feinberg et al. (1969) found a deficit in SWS mainly in stage 4 sleep of schizophrenics, while Tandon et al. (1992) indicated that shorter R E M latency was observed in schizophrenic patients when compared with normal controls but SWS did not differ between the schizophrenics and the controls. On the other hand, Ganguli et al. (1987) reported, based on quantitative E E G analysis, that the SWS percent
304
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
in schizophrenic patients was similar to that seen in healthy volunteers, but the total delta wave count was significantly less in the schizophrenics than in the controls. Thus, the reduced SWS or the decreased number of delta waves during sleep has not necessarily been recognized as a common characteristic in schizophrenia. In addition, no adequately detailed computerized analyses of delta and other waves during all-night sleep have been performed in schizophrenic subjects. Different types of hypnotic agents, such as barbiturates, antihistamines, sedating antidepressants, sedating neuroleptics and benzodiazepine (BZD) derivatives, have been administered as an adjunct to antipsychotic medication for the treatment of insomnia in schizophrenics (Mendelson, 1987; Morin, 1993). Although these drugs have usually been selected according to the types and severity of insomnia or the accompanied psychiatric symptoms, BZD hypnotics are the most frequently used members, which possess sufficient efficacy, of this drug class (Roth et al., 1988; Plasky, 1991; Morin, 1993). Most of the BZDs, however, inhibit SWS (Pakes et al., 1981). Thus, it may not be appropriate to use BZD hypnotics to treat the insomnia of schizophrenics, which is frequently accompanied with reduced SWS. Zopiclone (ZPC), a member of the cyclopyrrolone compounds, is a new hypnotic structurally unrelated to the BZDs (Julou et al., 1985). It is reported that ZPC acts on the gammaaminobutyric acid (GABA) receptors but at a site distinct from that of the BZDs (Blanchard et al., 1983; Julou et al., 1985). The effects of ZPC on sleep generally resemble those of BZDs, however, several authors reported that, in contrast to the BZDs, ZPC enhanced SWS in normal subjects (Nicholson and Stone, 1983; Billiard et al., 1987). In the present study, all-night polysomnographic examinations (PSG) were performed on 6 healthy male volunteers and 6 male schizophrenic outpatients, and the results of visual sleep-stage scoring and computerized period-amplitude analysis of EEG were compared in the schizophrenics and normals. In the schizophrenic subjects, the findings of EEG analysis and subjective sleep assessment during BZD and ZPC therapy were also compared and reported.
2. Materials and methods
2.1. Subjects Six healthy male volunteers aged 22-38 years (mean age: 27.3 years) and 6 male schizophrenic outpatients aged 22-39 years (mean age: 31.8 years) served as subjects of this study. The duration of illness of the schizophrenic patients was more than 4 years (mean duration: 10.3 years). To the best of our knowledge, there has been no report which indicates significant effects of duration of schizophrenia on the sleep architecture, although some correlations between the psychiatric symptoms and the sleep abnormalities have been suggested in this disease (Ganguli et al., 1987; Tandon et al., 1992). However, an acute phase of schizophrenia may not be appropriate in investigating EEG sleep due to its clinical instability, which may affect sleep architecture. The diagnosis of schizophrenia was based on DSM-lU-R (APA, 1987) criteria and the patients were classified as the disorganized type in 4 and the undifferentiated type in 2 patients. All of the patients were being treated with neuroleptics and anticholinergic drugs and also with BZD hypnotics for more than 8 weeks before the start of this study. The background of the 6 patients, including the doses of neuroleptics, previous BZD hypnotics and ZPC, is summarized in Table 1. Subjects who had a history of drug or alcohol abuse or of significant medical or neurological problems were excluded from the study. Informed consent was obtained from all of the subjects prior to the study.
2.2. Polygraphic recording Polysomnograms were recorded according to the method described in the standardized sleep manual of Rechtschaffen and Kales (1968). EEGs were recorded from disc electrodes placed at Fpl, Fp2, F3, F4, C3, C4, P3, P4, O1, 02, F7, F8, T5, T6, Fz, and Pz (10-20 electrode system) using A1 + A2 for reference. These 16 electrode locations were adopted for further topographic analysis (the results of the topograms will not be presented in this paper). EOGs were recorded monopolarly from both canthi, and EMGs bipolarly from the
305
N. Kajimura et al./Sehizophrenia Research 15 (1995) 303-312
Table 1 Background of the schizophrenics Case
Age
Age at onset
Type
Neuroleptlcs
Previous hypnotics
Zopiclone
1 2 3 4 5 6
27 39 37 22 37 29
13 23 27 18 27 21
Undifferentiated Undifferentiated Disorganized Disorganized Disorganized Disorganized
105 451 72 736 88 296
Nitrazepam Diazepam Triazolam Haloxazolam Nimetazepam Flunitrazepam
15-.22.5 15 15 15 15 15~ 7.5
5 5 0.5 10 5 4
The diagnosis of schizophrenia and type of schizophrenia were made according to DSM-111-R criteria. The drug values represent the daily dose (mg/day). The doses of neuroleptics have been converted to chlorpromazineaccording to the method described by Davis (1080). chin. The EEGs and EOGs were recorded with a time constant of 0.3 s, sensitivity of 10/~V/mm and high-cut filter of 120 Hz, while the E M G conditions were 0.003 s, 3.5 gV/mm and 500 Hz, respectively. Polysomnograms were not only recorded on paper but also recorded simultaneously by an analog tape recorder (XR-7000L) for further computerized analysis. In accordance with the Rechtschaffen and Kales manual (1968), all-night sleep stage scoring was carried out on the C3 E E G for each 20-second time-code-delimited epoch from the polysomnograms by visual analysis. The sleep variables monitored in this study were total sleep time, excluding periods of awakening and movement during the night, sleep latency (latency of the initial stage 2), R E M latency, and percentage of each sleep stage relative to total sleep time. Period-amplitude analysis was performed by the zero-crossing method on the C3 E E G using the Medilog Sleep Analyzing Computer (SAC: DEE-1100) (Smith et al., 1977, 1978), and the delta (0.33 3.0Hz), theta (2.5 7.1Hz), alpha (8.0-12.0Hz), beta (16.0-30.0Hz) and sigma ( 11.4 16.7 Hz) wave counts during all-night sleep were obtained. Concerning the delta waves, the counts during first, second and third sleep cycles (time interval from the end of R E M to the end of next R E M stage) were also measured, respectively. The C3 EEG was used for the computerized analysis in the present study, because the sleep stage scoring in recent sleep studies, including this study, has ordinarily been performed on this lead (Rechtschaffen and Kales, 1968).
With regard to the delta wave, the SAC system measured the amplitude and period of each halfwave, and the half-wave with amplitude of more than 5 #V and duration between 0.167 and 1.52 s (since the duration was for half-cycle waveform) was detected as the delta wave. The theta wave was counted as one theta wave once 2 out of 3 consecutive waveforms met the theta criteria and the counting ended once 2 or more waves no longer met the criteria. For the alpha wave, the system counted the beginning of 6 consecutive waves, all of which met the alpha criteria, as one alpha wave and finished the counting once 2 of 6 waves no longer met the alpha criteria. The beta and sigma waves qualified in the same manner as the alpha wave. Delta waves were analyzed in detail by classifying the delta half-wave according to amplitude and frequency. The number of delta half-waves with frequency of 0.5-2.0 Hz and amplitude of > 31 I(V, which were used as a criterion for the scoring of SWS in the sleep staging algorithm of the SAC, was also determined and analyzed. In addition, average amplitude and frequency of delta halfwaves during all-night sleep were obtained. 2.3. Subjective sleep assessment
A rating scale devised by us was used to score subjective assessment of sleep in the schizophrenic subjects. It consisted of 5 items, falling asleep, soundness of sleep, dreams, physical condition and mood in the morning. Soundness of sleep and dreams were evaluated using a 5-point (1-5) scale
306
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
and 3-point (1-3) scale, respectively. The remaining 3 items were assessed on 4-point (1-4) scales. The higher the score was for each item, the worse the evaluation.
neuroleptics or anticholinergic agents should also be considered, further examination is required to clarify this issue.
2.5. Statistical analysis 2.4. Experimental procedure PSGs were performed on 6 healthy male volunteers, and 6 male schizophrenic patients undergoing treatment with neuroleptics, anticholinergic drugs and BZD hypnotics. On the schizophrenic subjects, BZD hypnotics were replaced with ZPC 15 mg/day, and PSGs were recorded again at the end of 8 weeks of ZPC therapy. Subjective sleep assessment was performed in the patients during treatment with BZDs and ZPC. The doses of neuroleptics and anticholinergic drugs remained fixed throughout the study, whereas the subsequent doses of ZPC depended on the judgment of the attending psychiatrist. To exclude the laboratory effect (Agnew et al., 1966), each subject spent an adaptation night in the laboratory one week before each PSG. Disc electrodes were attached to the subjects at 20.00 h. Polygraphic recording was started at 22.00 h and completed at 7.00 h the next morning. Oral BZD hypnotics or ZPC were administered to the schizophrenics at 21.30 h. The rating scale for subjective sleep assessment was filled out after awakening. It is reported that ZPC is rapidly absorbed, achieving peak concentration within 2 h, and has a half-life of 3.5 to 6.5 h in healthy volunteers (Houghton et al., 1985; Julou et al., 1985). On the other hand, the schizophrenic subjects in this study had been chronically taking various BZDs, including nitrazepam, diazepam, triazolam, haloxazolam, nimetazepam or flunitrazepam, before ZPC treatment. Since these BZDs have different potency or half-life, they may possess some different effects on sleep, respectively. However, most of the BZDs have been reported to have basically similar effects on sleep EEG in clinical use, especially during chronic treatment (Pakes et al., 1981). Thus, in the present study, the effect of chronic administration of the BZDs on sleep EEG was summarized and compared with that of ZPC. Although the drug interactions between these hypnotics and
Student's t-test (two-tailed) was used for the statistical analysis of the polysomnogram results, and Wilcoxon's signed-rank test was used for that of the subjective sleep assessment. One way analysis of variance (ANOVA) with repeated measures was also employed for the statistical analysis of the delta half-wave result during the individual sleep cycles, and differences between sleep cycles were explored with post hoc Fisher's least significant difference (LSD) test.
3. Results
During the present study, one schizophrenic patient developed insomnia after the start of ZPC therapy, and the dose of ZPC was increased from 15 to 22.5mg/day. Another patient developed daytime drowsiness with malaise in the second week of ZPC therapy, and the dose of ZPC was decreased from 15 to 7.5 mg/day. These regimens relieved the symptoms, and the two patients were able to continue taking their respective doses of ZPC throughout the remainder of the study. Though bitter taste in the mouth also occurred as an adverse effect in one patient being treated with ZPC, it was mild enough to continue the treatment.
3.1. Sleep variables The sleep variables of the healthy volunteers and the schizophrenic patients are shown in Table 2. The schizophrenics exhibited longer sleep latency and a marked decrease in the amount of SWS in comparison with the normal subjects. The amounts of stage 1 sleep during BZD therapy and stage 2 sleep during ZPC therapy of the schizophrenics were greater than those of the normal subjects. There were no significant differences in the number of awakenings and the total awake time between the normals and the schizophrenics, although these results are not shown in Table 2. However, the
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
307
Table 2 Sleep variables in normal volunteers and schizophrenics Sleep variables Total sleep time Sleep latency REM latency % Stage t % Stage 2 % Stages 3 and 4 % Stage REM
(min) (rain) (min)
N
S (BZDs)
S (ZPC)
485.1 _+11.1 12.0+_3.7 86.1 -+ 12.3 10.0+2.1 56.4-+2.3 13.3+_2.2 20.4±2.0
411.5 ± 36.0* 53.8 _+16.7'* 111.7 _+33.0 16.2 -+ 1.8" 62.9+_3.5 3.1 -+ 1.3"** 17.8-+2.4
436.5 ± 37.2 59.9 ± 19.3'* 45.4 + 20.8 8.3 + 1.8 68.6_+2.2*** 4.0± 1.1 *** 19.0+_2.0
D A
Each value represents the mean ±S.E.M. of the values in 6 subjects. N, normal volunteers; S: schizophrenics. Polysomnographic examination of schizophrenics was performed during benzodiazepine (BZD) hypnotic treatment and at the end of 8 weeks of zopiclone (ZPC) therapy. The existence of statistically significant differences between the normal volunteers and the schizophrenics is indicated by the asterisks after the values for the schizophrenics (*p <0.1; **p <0.05; ***p < 0.01 ). The presence of statistically significant differences between BZD and ZPC therapy in the schizophrenics is indicated by the letters after the values at the end of ZPC treatment (A, p<0.1; D, p<0.001 ).
sleep efficiency (%) of the normal subjects was higher, though not significant, than that of the schizophrenic patients (91.0 ± 1.2 for the normals; 81.5 ± 7.4 for the schizophrenics during BZD therapy; 83.0 ± 7.0 for the schizophrenics during ZPC therapy ). In the schizophrenic patients, the percentage of stage 1 sleep was lower and that of stage 2 was higher during treatment with ZPC than with BZDs. There were no significant differences in sleep latency and in the amounts of SWS and REM sleep between BZD and ZPC treatments.
The schizophrenics showed a marked decrease in the number of delta half-waves of 0.5-2.0 Hz, >__31/~V in comparison with the normals. There were no significant differences in the number of waves of other frequency bands between the patients and the normals. In the schizophrenics, no differences were observed in the number of waves of individual frequency bands between BZD and ZPC treatments.
3.2. Analysis of the activity of individualfrequency bands
Table 4 shows the delta half-wave counts per hour during all-night sleep classified according to the amplitude and frequency. The half-wave count of total delta waves (0.33-3.0 Hz, >_5 ltV) in the schizophrenics was significantly lower than that in
The delta, theta, alpha, beta and sigma wave counts during all-night sleep are shown in Table 3.
3.3. Delta half-wave analysis
Table 3 Delta, theta, alpha, beta and sigma wave counts during all-night sleep Waveforms (Hz)
N
S (BZDs)
S (ZPC)
Delta Theta Alpha Beta Sigma
6162.7_+632.0 3360.2_+1312.1 1184.2_+253.4 819.3-+409.7 1597.2-+342.3
2311.5+634.4"** 1806.5+1115.2 1262.7+324.8 1821.3-+546.0 2451.8_+603.4
2578.2_+551.1"** 1397.2_+832.8 1077.3_+263.7 1369.7_+473.1 2324.5_+355.1
(0.5 2.0) (2.5 7.1) (8.0-12.0) (16.0-30.0) (11.4-16.7)
The delta, theta, alpha, beta and sigma wave counts during all-night sleep are expressed as means+S.E.M, of the values in 6 subjects. The delta wave count is shown as the half-wave count used for slow-wave sleep in the sleep staging algorithm of the SAC (0.5 2.0 Hz, > 31 laV). The alpha, beta and sigma wave counts represent 6 or more consecutive waves, respectively, while the theta wave count is based on 3 or more consecutive waves. Details are the same as in Table 2. ***p <0.01.
308
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
Table 4 Delta half-wave count per hour during all-night sleep classified according to amplitude and frequency N
S (BZDs)
S (ZPC)
Amplitude (~tV) 5-15 15-25 25-35 35-45 45 55 55 <
3940.9+ 173.0 1870.7 + 93.6 611.1 + 47.0 286.4+30.2 147.6+20.2 201.7 + 40.2
4395.7___226.0 1354.4 + 231.0" 381.8 + 67.3"* 139.9+31.1"** 56.2+ 16.0"** 37.3 + 12.5"**
4481.1 _+199.2' 1478.3 + 169.4* 418.0 + 69.6** 154.9+29.5"* 63.6_+ 14.2"** 48.0 _+11.9***
Frequency (Hz) 0.33-0.5 0.5 - 1.0 1.0 -1.5 1.5 -2.1 2.1 -2.6 2.6 -3.0
134.4 + 20.7 1415.7 ___146.3 1739.0+72.7 1449.2___39.5 1166.9 -+ 81.3 1154.2+ 110.3
178.3 _+35.0 1476.3 + 145.2 1565.8+46.7" 1269.0+ 106.1 1007.2 _+131.2 868.1 4-153.5
177.9 ___25.4 1584.2 -+95.7 1647.8-+50.4 1337.4-+68.6 1034.7 _+83.0 863.24-95.5*
Total
7059.5 + 29.6
6366.1 4- 253.5**
6645.1 _+173.2**
A
The delta half-wave count per hour during all-night sleep for each amplitude and frequency band is shown as the mean + S.E.M. of the values in 6 subjects. The delta wave frequency is defined as lying between 0.33 and 3.0 Hz, while the minimum delta wave amplitude is defined as 5 ~tV. Details are the same as in Table 2. *p<0.1; **p<0.05; ***p <0.01; A, p<0.1; B, p <0.05.
the normal subjects. The schizophrenics showed a markedly reduced number of delta half-waves with higher amplitude than the normals. There were no distinctive differences in the half-wave counts of delta waves classified according to frequency between the normal subjects and the patients. In the schizophrenics, the number of total delta half-waves was higher during treatment with ZPC than with BZDs. ZPC therapy was accompanied by higher half-wave counts of delta waves with amplitude above 55/2V and with frequency of 1.0-1.5 Hz than BZD therapy. Table 5 shows the half-wave counts of delta waves (0.33-3.0 Hz, >5/~V) per hour during the first, second and third sleep cycles. The delta halfwave counts during the first and second sleep cycles of the schizophrenics were significantly lower than those of the normal subjects. There was no significant difference, however, in the delta half-wave count during the third sleep cycle between the patients and the normals. In the normal subjects, an ANOVA revealed the statistically significant effect of sleep cycle on the delta half-wave count, and post hoc Fisher's LSD test indicated that, during the first sleep cycle as compared with the second and third cycles, there
were significantly more delta half-waves per hour (p<0.05). In the schizophrenics, however, ANOVAs revealed no statistically significant effect of sleep cycle on the delta half-wave counts per hour during BZD and ZPC treatments.
3.4. Average amplitude and frequency of delta halfwave Table 6 shows the average amplitude and frequency of delta half-waves during all-night sleep. The average amplitude of delta half-waves in the schizophrenic subjects was significantly lower than that in the normals, while there was no difference in the average frequency between the two groups. In the schizophrenics, no differences were observed in the average amplitude and frequency between BZD and ZPC treatments.
3.5. Subjective sleep assessment The score for soundness of sleep in the subjective sleep assessment was lower (evaluated as a good condition) during treatment with ZPC than BZDs (Table 7). There were no significant differences in
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
309
Table 5 Delta half-wave count per hour in first, second and third sleep cycles
Sleep cycle First Second Third ANOVA
N
S (BZDs)
S (ZPC)
7515.8 ___109.2 7229.8 +49.5 7048.0 + 81.2
6656.5 + 180.7"** 6431.3 +255.5** 6355.6 + 357.9"
6759.7 + 190.8"** 6746.4 + 169.8"* A 6770.0 +_192.6
F = 8.953 p=0.0059
F = 1.263 n.s.
F = 0.029 n.s.
The half-wave counts of delta waves (0.33 3.0 Hz, >5 IxV) per hour in the first, second and third sleep cycles are shown as the mean + S.E.M. of the values in 6 subjects. An ANOVA was used to evaluate the effect of sleep cycle on the delta half-wave counts. Other details are the same as in Table 2. *p<0.1; **p<0.05; ***p<0.01; A, p<0.1. Table 6 Average amplitude and frequency of delta half-waves during all-night sleep
Amplitude (IxV) Frequency (Hz)
N
S (BZDs)
S (ZPC)
17.7+0.7 1.09 + 0.05
13.9+0.8"** 0.97 + 0.08
14.3+0.7"** 0.97 + 0.05
The average amplitude and frequency of delta half-waves during all-night sleep are shown as means_+S.E.M, of the values in 6 subjects. Details are the same as in Table2. ***p <0 01. Table 7 Subjective sleep assessment in schizophrenics Items
BZDs
ZPC
Falling asleep Soundness of sleep Dreams Physical condition in the morning Mood in the morning
2.33 _%0.61 3.67 +0.33 1.67 + 0.33 2.50+0.22 2.33+0.21
2.00 + 0.63 2.67___0.49"* 1.50+0.34 2.50+0.22 2.00+0.37
Each score indicates the mean+S.E.M, **p<0.05.
of 6 patients.
regard to other subjective assessment items between the two treatments.
4. Discussion
In the visually scored sleep parameters, the chronic schizophrenic patients in this study showed a prolongation of sleep latency, fairly large amounts of stages 1 and 2 sleep, and very little
SWS when compared to the normal subjects. These findings essentially coincide with the results of previous EEG studies in drug-naive schizophrenics (Feinberg et al., 1969; Keshavan et al., 1990). However, Tandon et al. (1992) found shortened REM latency, rather than reduced SWS, in nonmedicated or drug-free schizophrenic patients. The schizophrenic patients in the present study had been chronically taking neuroleptics and anticholinergic drugs together with BZD hypnotics or ZPC throughout the study. It is reported that both neuroleptics and anticholinergic drugs possess significant inhibitory effects on REM sleep but have no such effects on NREM sleep including SWS (Gillin et al., 1978; Taylor et al., 1991). Thus, the chronic schizophrenics undergoing chronic treatment with the psychotropic drugs may also show NREM abnormalities, including the reduced SWS, of visually scored sleep parameters similar to those of non-medicated or drug-free schizophrenics. It may be difficult, however, to evaluate REM sleep abnormalities, including the shortened REM latency, in the schizophrenic patients with antipsychotic medication. In the period-amplitude analysis, the schizophrenic subjects showed marked reduction in the number of 0.5-2.0 Hz, >31/~V delta half-waves as well as in the number of total delta band half-waves (0.33-3.0 Hz, >5 pV) during sleep in comparison with the normal subjects. In addition, the patients exhibited markedly decreased halfwave count of delta waves with higher amplitude. However, no distinctive difference was found in
310
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
the distribution of delta wave frequency between the schizophrenics and the normals. The finding that the number of total delta waves during sleep is markedly decreased in the schizophrenics is consistent with the previous quantitative sleep EEG study in schizophrenia (Ganguli et al., 1987). However, the detailed delta wave analysis in the present study revealed that the number of delta waves with higher-amplitude was particularly decreased in the schizophrenics. This is also supported by the present finding that the average amplitude of delta waves during all-night sleep in the schizophrenics was significantly lower than that in the normal subjects but no difference was found in the average frequency between the schizophrenics and the normals. Feinberg (1989) suggested the positive correlation between the delta wave amplitude during sleep and the metabolic rate of the cerebral cortex in humans. Thus, the reduced number of delta waves with higheramplitude in the schizophrenics may be involved in the hypoactivity of the prefrontal cortex, which has recently been reported to be associated with negative symptoms in this disease (Andreasen, 1989). These results suggest that the reduction in SWS in schizophrenia may be attributable mainly to the decrease in the number of delta waves with higher amplitude rather than with lower frequency. The schizophrenic subjects, when compared to the normal subjects, showed a decreased number of delta half-waves during the first and second sleep cycles but exhibited no obvious difference in the number of the waves during the third cycle. In addition, the number of delta half-waves per hour in the normal subjects was significantly greater during the first sleep cycle than during other sleep cycles, whereas, in the schizophrenic subjects, there were no significant differences in the delta halfwave counts between the sleep cycles. Thus, the reduced number of delta waves during the first sleep cycle may be a remarkable characteristic of the schizophrenics in this study. This is supported by the finding of Hiatt et al. (1985) that schizophrenic patients showed an abnormally low level of delta count in the first N R E M period. The schizophrenic subjects in this study had continuously been taking neuroleptics and anticholinergic drugs together with BZD hypnotics or
ZPC throughout the present study. It is well known that neuroleptics, anticholinergic agents and the hypnotics possess various effects on delta power in previous spectral EEG analysis (Itil, 1974). Thus, further discussion on sleep EEG of schizophrenia may not be appropriate, and the following is focused on the comparison of sleep EEG between BZD and ZPC treatments in the schizophrenics. In the visually scored sleep parameters, there was less stage 1 and greater stage 2 sleep in the schizophrenics during treatment with ZPC than with BZDs, but there was no obvious difference in the amount of SWS between the treatment with BZDs and ZPC. It is well known that BZD hypnotics increase stage 2 and decrease SWS (Pakes et al., 1981). Billiard et al. (1987) reported that in normal humans ZPC significantly increased the duration of SWS. Nicholson and Stone (1983) also found that stage 2 showed a significant increase and SWS tended to be extended with ZPC treatment in healthy adults. Thus, the result in the present study that no difference was found in the amount of SWS between the BZD and ZPC treatments suggests that it is difficult to induce an increase of SWS in schizophrenia by ZPC therapy because the mechanism of producing delta waves might be deficient in schizophrenic subjects. However, the finding that there was less stage 1 and greater stage 2 sleep during treatment with ZPC than with BZDs suggests that ZPC may induce deeper sleep in schizophrenics than BZDs. This is supported by the finding that soundness of sleep in the subjective sleep assessment was better evaluated during treatment with ZPC than BZDs. It was found in the period-amplitude analysis in the present study that the schizophrenics showed a higher half-wave count of total delta waves during treatment with ZPC than with BZDs. Furthermore, the count of delta half-waves with amplitude of over 55 #V was higher in ZPC therapy than in BZD therapy, although no difference was observed in the average amplitude of delta half-waves between the two treatments. Based on spectral analysis of sleep EEG, Trachsel et al. (1990) found that both ZPC and BZD hypnotics depressed delta power density during sleep. Johnson et al. (1979) also reported that BZD
N. Kajimura et al./Schizophrenia Research 15 (1995) 303 312
hypnotics decrease delta power during sleep by reducing the amplitude and, to a lesser extent, the number of delta waves, while Wright et al. (1986) claimed that ZPC decreases higher-amplitude delta waves but increases the total number of delta waves. Thus, in normal humans, BZDs and ZPC possess inhibitory effect on the amplitude of delta waves during sleep, but may possess differential effects on the occurrence of delta waves. These results suggest that, in schizophrenics who have fewer delta waves during sleep, inhibitory effect of ZPC on the amplitude of delta waves may be weaker than that of BZDs. It can not be concluded from the present study whether ZPC enhances the occurrence of delta waves during sleep in schizophrenics, however, it may be convincing that BZDs suppress the occurrence of delta waves during sleep as compared with ZPC. No difference was observed in the number of 0.5-2.0 Hz, >__31/~Vdelta half-waves between the treatment with BZDs and ZPC. Thus, delta waves which are used as a criterion for the scoring of SWS, ie., delta waves with higher amplitude and with lower frequency, may not easily be increased in schizophrenia. The schizophrenic subjects showed no differences in the numbers of sigma and beta waves during sleep between BZD and ZPC treatments. It is well known that BZD derivatives increase the amount of sigma waves during sleep and enhance the beta activity (Itil, 1974; Hirshkowitz et al., 1982). It is also reported that ZPC possesses facilitatory effects on sigma and beta waves (Saletu et al., 1987; Iwata et al., 1989). Quadens et al. (1983) stated that ZPC increases the duration of stage 2 sleep, which is characterized by sigma waves, as comparable to flurazepam. Thus, both the BZD and ZPC therapy in the schizophrenics may be accompanied by equivalent increases of sigma and beta waves during sleep, resulting in no differences of these waves between the two treatments. This may be supported by the finding that the schizophrenics showed considerable increases, though not significant, in the numbers of sigma and beta waves during BZD and ZPC treatments in comparison with the normal subjects who received no medication. Although a more definitive study without other
311
concurrent medications and with a larger population and a double-blind crossover study will be necessary, the present findings suggest that lowered amplitude of delta waves during sleep may be involved in the decreased amount of SWS in schizophrenia and that both reduced SWS and decreased count of delta waves with higheramplitude and lower-frequency may not easily be enhanced in schizophrenics. Moreover, ZPC may induce deeper sleep in schizophrenics than BZDs by decreasing stage 1 and increasing stage 2 sleep and by increasing the number of total delta waves during sleep, resulting in better evaluation of soundness of sleep in subjective sleep assessment.
Acknowledgments The authors would like to thank to Dr. Thomas A. Wehr, Chief, Clinical Psychobiology Branch, National Institute of Mental Health, for his useful comments and suggestions.
References Agnew, H.W.Jr., Webb, W.B. and Williams, R.L. (1966) The first night effect: an EEG study of sleep. Psychopharmacologia 2, 263-266. American Psychiatric Association (1987) Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revised. American Psychiatric Association Press, Washington, DC. Andreasen, N.C. (1989) Neural mechanisms of negative symptoms. Br. J. Psychiatry 155 Suppl. 7, 93 98. Blanchard, J.C., Zundel, J.L. and Julou, L. (1983) Differences between cyclopyrrolones (suriclone and zopiclone) and benzodiazepines binding to rat hippocampus photolabelled membranes. Biochem. Pharmacol. 32, 3651-3653. Billiard, M., Besset, A., de Lustrac, C. and Brissaud, L. (1987) Dose-response effects of zopiclone on night sleep and on nighttime and daytime functioning. Sleep 10 Suppl. 1, 27 34. Davis, J.M. (1980) Organic therapies: antipsychotic drugs. In: A.M. Freedman, H.I. Kaplan and B.J. Sadock (Eds.), Comprehensive Textbook of Psychiatry. Williams Wilkins, Baltimore, pp. 2257 2289. Feinberg, I. (1989) Effects of maturation and aging on slowwave sleep in man: implications for neurobiology. In: A. Wauquier, C. Dugovic and M. Radulovacki (Eds.), Slow Wave Sleep: Physiological, Pathophysiological, and Functional Aspects. Raven Press, New York, pp. 31-48. Feinberg, I., Braun, M., Koresko, R.L. and Gottlieb F. (1969)
312
N. Kajimura et al./Schizophrenia Research 15 (1995) 303-312
Stage 4 sleep in schizophrenia. Arch. Gen. Psychiatry 21,262-266. Gangnli, R., Reynolds, C.F.V. and Kupfer, D.J. (1987) Electroencephalographic sleep in young, never-medicated schizophrenics: a comparison with delusional and nondelusional depressives and with healthy controls. Arch. Gen. Psychiatry 44, 36-44. Gillin, J.C., van Kammen, D.P. and Bunney, W.E. (1978) Pimozide attenuate d-amphetamine-induced sleep changes in man. Life Sci. 22, 1805-1810. Hiatt, J.F., Floyd, T.C., Katz, P.H. and Feinberg, I. (1985) Further evidence of abnormal non-REM sleep in schizophrenia. Arch. Gen. Psychiatry 42, 797-802. Hirshkowitz, M., Thornby, J.I. and Karakan, I. (1982) Sleep spindles: pharmacological effects in humans. Sleep 5, 85-94. Houghton, G., Dennis, M., Templeton, R. and Martin, B. (1985) A repeated dose pharmacokinetic study of a new hypnotic agent, zopiclone (Imovane). Int. J. Clin. Pharmacol. Ther. Toxicol. 23, 97-100. Itil, T.M. (1974) Quantitative pharmaco-electroencephalography. In: T.M. Itil (Ed.), Modern Problems in Pharmacopsychiatry. Karger, Basel, pp. 43-75. Iwata, T., Hayakawa, T., Terashima, M., Ohta, T., Okada T., Ueda, Y. and Ishii, N. (1989) Effects of zopiclone on sleep spindles studied with an improved waveform recognition method. Jpn. J. Psychiat. Neurol. 43, 669-674. Johnson, L.C., Seales, D.M., Naitoh, P., Church, M.W. and Sinclair, M. (1979) The effects of flurazepam hydrochloride on brain electrical activity during sleep. Electroenceph. Clin. Neurophysiol. 47, 309-321. Julou, L., Blanchard, J.C. and Dreyfus, J.F. (1985) Pharmacological and clinical studies of cyclopyrrolones: zopiclone and suriclone. Pharmacol. Biochem. Behav. 23, 653-659. Keshavan, M.S., Reynolds, C.F. and Kupfer, D.J. (1990) Electroencephalographic sleep in schizophrenia: a critical review. Compr. Psychiatry 30, 34-47. Mendelson, W.B. (Ed.) (1987) Human Sleep: Research and Clinical Care. Plenum, New York. Morin, C.M. (Ed.) (1993) Insomnia: Psychological Assessment and Management. Guilford, New York. Nicholson, A.N. and Stone, B.M. (1983) Zopiclone: sleep and performance studies in healthy man. Pharmacology 27 Suppl. 2, 92-97.
Pakes, G.E., Brogden, R.N., Heel, R.C., Speight, T.M. and Avery, G.S. ( 1981) Triazolam: a review of its pharmacological properties and therapeutic efficacy in patients with insomnia. Drugs 22, 81-110. Plasky, P. (1991) Antidepressant usage in schizophrenia. Schizophr. Bull. 17, 649-656. Quadens, O.P., Hoffman, G. and Buytaert, G. (1983) Effects of zopiclone as compared to flurazepam on sleep in women over 40 years of age. Pharmacology 27 Suppl. 2, 146-155. Rechtschaffen, A. and Kales, A. (1968) A manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. US Government Printing Office, Washington, DC. Roth, T., Roehrs. T. and Zorick F (1988) Pharmacological treatment of sleep disorders. In: R.L. Williams, I. Karacan and Moore C.A. (Eds.), Sleep Disorders: Diagnosis and Treatment. John Wiley and Sons, New York, pp. 373-395. Saletu, B., Anderer, P., Kinsperger, K. and Grunberger, J. (1987) Topographic brain mapping of EEG in neuropsychopharmacology-part U: clinical applications (pharmaco-EEG imaging). Meth. Find. Exp. Clin. Pharmacol. 9, 385-408. Smith, J.R., Karakan, I. and Yang, M. (1977) Ontogeny of delta activity during human sleep. Electroenceph. Clin. Neurophysiol. 43, 229-237. Smith, J.R., Karacan, I. and Yang, M. (1978) Automated analysis of the human sleep EEG. Waking Sleeping 2, 75 82. Tandon R., Shipley, J.E., Taylor, S., Greden, J.F., Eiser, A., DeQuardo, J and Goodson, J. (1992) Electroencephalographic sleep abnormalities in schizophrenia: relationship to positive/negative symptoms and prior neuroleptic treatment. Arch. Gen. Psychiatry 49, 185-194. Taylor, S.F., Tandon R., Shipley, J.E. and Eiser, A.S. (1991) Effect of neuroleptic treatment on polysomnographic measures in schizophrenia. Biol. Psychiatry 30, 904-912. Trachsel, L., Dijk, D.J., Brunner, D.P., Klene, C. and Borbely, A.A. (1990) Effect of zopiclone and midazolam on sleep and EEG spectra in a phase-advanced sleep schedule. Neuropsychopharmacology 3, 11-18. Wright, N.A., Belyavin, A., Borland, R.G. and Nicholson, • A.N. (1986) Modulation of delta activity by hypnotics in middle-aged subjects: studies with a benzodiazepine (flurazepam) and a cyclopyrrolone (zopiclone). Sleep 9, 348-352. Zarcone, V.P., Benson, K.L. and Berger, P.A. (1987) Abnormal rapid eye movement latencies in schizophrenia. Arch. Gen. Psychiatry 44, 45-48.