- Email: [email protected]
Cognitive and overall functioning in institutionalised patients taking typical versus atypical antipsychotics
172. Psychotic disorders and antipsychotics
$260
References [1] Lopes-Machado EZ, Crippa JAS, Hallak JEC, Guimaraes FS, Zuardi AW. "Electrodermallynonresponsiveschizophrenicsmake more errors in the stroop color word test, indicating selective attention deficit". Schizophrenia Bulletin (in press).
~ A
pharmacoepidemiological study of diabetes mellitus and antipsychotic treatment in the United States
E Cavazzoni 1, K. Hornbuckle 1, D. Hutchins2, L. Jovanovic3, A. Breier 1, J. Buse 4. 1Lilly Research Laboratories, Indianapolis, U.S.A.; 2Advance PCS, Scottsdale, U.S.A.; 3Sansum Medical Research Institute, Santa Barbara, U.S.A.; 4Diabetes Care Center, Chapel Hill U.S.A. A comparison of the odds of developing diabetes mellitus (DM) during exposure to various antipsychotics. Cases of treatmentemergent DM have been reported for both conventional and atypical antipsychotics. However, large epidemiological studies are needed to determine the existence and extent of any association between development of DM and antipsychotic medications and to resolve whether there is a substantial difference in the risk of developing DM among different antipsychotic therapies. This is a retrospective cohort study based on the prescription claim data of AdvancePCS Inc. in the US covering the period of 12/01/1997 to 8/31/2000. Prescription claims for antipsychotics were used to identify and select adult subjects that began antipsychotic monotherapy between 12/1/1998 and 2/29/2000. The antipsychotic cohorts studied were combined conventional antipsychotics (N = 19,782), haloperidol (N= 8,476), thioridazine (N=3,133), combined atypical antipsychotics (N= 38,969), olanzapine (N=13,863), risperidone (N=20,633), quetiapine (N=4,196), and clozapine (N=277). The AdvancePCS general patient population (N=5,816,473) included all adult AdvancePCS members who had made at least one prescription claim (other than an antipsychotic) between January 1, 2000 and February 29, 2000. Prescription claims for diabetes medications were used to identify subjects with DM. Those with antipsychotic prescription claim(s) 6 months prior to starting anti-diabetic therapy or those with anti-diabetic prescription claims any time prior to starting antipsychotic therapy were excluded. Cox Proportional Hazards regression was used to adjust for differences in age, gender, and duration of antipsychotic exposure between cohorts in the estimation of risk of developing diabetes. Hazard ratios (HRs) of DM in antipsychotic cohorts, were determined relative to the AdvancePCS general patient population and between selected antipsychotic cohorts. The relative risk of developing DM during treatment with conventional and atypical antipsychotics cohorts compared to the AdvancePCS general population were HR=3.5; CI: 3.1-3.9; p<.001 and HR=3.1; CI: 2.9-3.4; p<.001, respectively. The relative risk of developing DM during treatment with individual antipsychotics compared to the AdvancePCS general population were haloperidol HR=3.1; CI: 2.6-3.7; p<.001, thioridazine HR=4.2; CI: 3.2-5.5; p<.001, olanzapine HR=3.0; CI: 2.6-3.5; p<.001, risperidone HR=3.4; CI: 3.1-3.8; p<.001, quetiapine HR=I.7; CI: 1.2-2.4; p=.002, and clozapine HR=3.3; CI: 1.4-8.0; p=.007. The combined conventional and atypical cohorts were compared and there was no significant difference in risk of developing DM (HR=0.97; CI: 0.8-1.1; p=0.6). Compared to the haloperidol cohort, a statistically significant increase in risk of developing DM was observed in the
risperidone cohort HR=I.2; CI: 1.0-1.5; p= 0.04. There was no statistically significant difference between the risk of developing DM in the olanzapine and risperidone cohorts HR=0.9; CI: 0.81.1; p= 0.2. An increased risk of developing diabetes compared to a general reference population was observed in the AdvancePCS prescription-database cohorts during treatment with either conventional or atypical antipsychotics. Though the risk of developing diabetes was significantly greater for patients in the risperidone cohort than in the haloperidol cohort, this analysis did not demonstrate a generally elevated risk between the atypical and conventional antipsychotic cohorts. It remains unclear whether the observed increases are related to factors intrinsic or extrinsic to those psychiatric conditions commonly treated with antipsychotic drugs.
•
Cognitive and overall functioning in insUtutionalised patients taking typical versus atypical antipsychotics
J. Cervilla 1, R. Ana2, G. Hoyos 3, L. Minguez3, E. Gonzalez de Pablos 3. 1Sant Joan de D~u - Serveis de Salut Mental, Servei Salut Mental Garraf Vilanova i la Geltr{t, Barcelona, Spain; 2Hospital Santa Clotilde, Santander, Spain; 3Complejo Hospitalario San Luis, Palencia, Spain Atypical antipsychotics have been postulated as having general functioning and cognitive advantages over typical antipsychotics. Objective: The objective of this study was to investigate such notion using a sample of 149 (n=149) patients. Sample: 141 participants out of the 149 were women,the overall mean age 64.5 years) and two thirds had a diagnosis of schizofrenia. They were all long-term in-patients living in a psychitric institution (Complejo Hospitalario San Luis, in Palencia, Spain). Methods: A trained clinical psychologist and a geriatritian administrated a psychopathological scale (BPRS), a cognitive screening scale (the Spanish version of the Minimental State Examination validated by Lobo et al.), a functionality scale (Barthel index) and a behavioural disorders scale (The Spanish version of Cohen-Mansfield's Agitation inventory for the Elderly).We compared using non-parametric univariate statistics differences of scores in all such scales among those subjects taking an atipical antipsychotic (72 patiensts who took riperidone or olanzapine) and typical ones (haloperidol or chlorpromazine). Results: We did not find significant differences amongst the two groups for neither the psychopathological nor the behavioural/agitation scale. However, we did find significant differences showing better cognitive functioning (p=0.0001) in patients on atypicals compared to those on typical antipsychotics. We also found a significantly (p=0.00001) better overall functioning in patients taking atypical antipsychotics. In an interesting univariate subanalysis among the 72 patients taking atypical (risperidone vs. olanzapine) we did not find statistically significant differences for neither cognitive nor overall functioning. Nevertheless, there was a nearly sginifican trend for a linear association between those taking olanzapine and higher cognitive test scoring (p=0.068). Conclusion: We did not find differences in efficacy of psychological or behavioural symptoms among patients taking atypical versus typical antipsychotics. However, this study suggests that atypical antipsychotics may contribute to a better cognitive and overall functioning in aged adults who are chronically institutionalised because of a severe mental disorder. This results
P2. Psychotic disorders and antipsychotics
$261
should, however, be interpreted with caution as we cannot rule out potential confounding effects of age,sex, diagnosis or length of admission. However, the sample is very homogeneous for most of this independent variables.
accurate marker of the underlying genetic substrate than the assessed endophenotypes. Grants supported: Hacettepe Univ. Research Foundation, #01 01 101 001.
•
References
Endophenotype assessments in a large schizophrenia pedigree
[1] Leboyer Met al (1998) Psychiatric genetics: Search for phenotypes. Trends Neuroscienci 21: 102-105.
S. Ozer 1 , A.S. Gonul 2, S. Ulusoy 1 , E Heutink 3, S. Claes 4, A.N. Akarsu 5, A. Ulusahin 1 *. IDep. Psychiatry, Hacettepe Univ.
Med. Fac., Ankara, Turkey; 2Dep. Psychiatry, Erciyes Univ. Med. Fac., Kayseri, Turkey; 3Dep. Clin. Genetics Erasmus Univ., Rotterdam, The Netherlands; 4Dep. Psychiatry, Univ. Antwerp, Antwerp, Belgium; 5Gene Map. Lab., Dep. Pediatr., Hacettepe Univ. Med. Fac., Ankara, Turkey Bac kg r o und: Schizophrenia is a clinically and genetically complex disorder. Including endophenotypes may facilitate genetic research. Endophenotype approach focusing on subclinical traits may help identification of genes that increase the susceptibility for the illness. Objective: In a multigenerational large schizophrenia family, we assessed Axis I psychopathology according to RDC criteria, and also neurological soft signs and schizotypal personality traits as possible endophenotypes. M e t ho d: A large multigenerational pedigree demonstrating aggregation of schizophrenia and other psychiatric disorders was assessed. Most of the affected members of the family were treated in the regional university psychiatry department for ten years. The available hospital records of the living and deceased affected members were used to complete the clinical information. Of 90 living members from three generations 54 were interviewed with the SADS-L and 71 blood samples were collected for further molecular genetic study. For endophenotype assessment the Neurological Evaluation Scale (NES) and the schizotypal features related section of the Comprehensive Assessment of Symptoms and History (CASH) were used. Results: We identified a Turkish isolate (Turkmen tribe, province of Bozok) from Central Anatolia through 10 schizophrenic probands. The complete pedigree structure consisted of 358 subjects in total; historical records indicated that the ancestors of the family settled in the region around 1700. Marrying distant relatives was a common pattern in the family. RDC diagnoses of the interviewed members were as follows: schizophrenia and unspecified functional psychosis 10, affective disorders 17, alcohol and drug abuse 5, other diagnoses 16, no diagnosis 19. All schizophrenic members had similar paranoid features. Mean NES scores of schizophrenic and non-schizophrenic members were 164-9.49 and 9.93+6.00 respectively (Cut-off point of NES for Turkish population is 15). The appearance of glabellar reflex in 5 of the 6 examined schizophrenic members and 9 of the 40 nonschizophrenic subjects was noticable. Schizotypal features were common especially in the schizophrenic members. Social anxiety and suspiciousness were the most prevalent items. Conclusion: In this family with an exceptional prevalence of mental disorders two distinct putative markers and RDC schizophrenia criteria were evaluated. All l0 schizophrenic patients showed a clinically homogeneous phenotype with prominent persecutory delusions. On the other hand we couldn't observe any cosegregation of the assessed endophenotypes with schizophrenia, with the exception of glabbellar reflex. For this particular family, we conclude that the RDC schizophrenia phenotype is a more
•
Influence of long-lasting administration of neuroleptics on glutamatergic transmission in rats
M. Pietraszek 1, J. Wardas 1, K. Golembiowska 2, M. Bijak 3, S. Wolfarth 1, K. Ossowska 1 . Departments of 1Neuro-
Psychopharmacology, 2pharmacology and 3physiology, Institute of Pharmacology, Polish Academy of Sciences, Krakrw, Poland It has been postulated that a dysfunction of glutamatergic transmission, mainly in cortical regions, may contribute to the pathophysiology of schizophrenia (Ishimaru and Tom 1997). In our previous study we showed that chronic administration of neuroleptics increased the number of cortical NMDA receptors labelled with the competitive antagonist [3H]CGP 39653 (Ossowska et al. 1999). The purpose of the present study was to examine the influence of haloperidol, a typical neuroleptic, and clozapine, an atypical one, on AMPA receptors, as well as on the glutamatedependent neuronal activity and the extracellular levels of excitatory amino acids in rat cerebral cortex. Haloperidol (1 mg/kg/day) and clozapine (30 mg/kg/day) were administered to rats in drinking water for 6 weeks or 3 months, and were afterwards withdrawn for 4 days. Extracellular concentrations of basal and veratridine-stimulated glutamate (Glu) and aspartate (Asp) in the fronto-parietal cortex were assessed using an in vivo microdialysis; neuronal discharges that developed in a Mg2+-free medium were recorded extracellularly in cortical slices ex vivo. In autoradiographic studies, AMPA receptors were labelled with [3H]AMPA. Haloperidol elevated the basal, but not the veratridine-evoked, extracellular levels of Glu and Asp and enhanced the activity of cortical neurons. In contrast, clozapine decreased both the basal and the stimulated levels of Glu and Asp in the fronto-parietal cortex, but had no effect on the activity of cortical neurons. None of the neuroleptics influenced the binding of [3H]AMPA to AMPA receptors in the cerebral cortex. The present study shows that the two representatives of typical and atypical neuroleptics, administered chronically, alter the glutamatergic system in cortical regions. They have no effect on AMPA receptors, but differently regulate the extracelluler levels of Glu and Asp and the neuronal activity. The influence of the examined neuroleptics on cortical glutamatergic transmission may be of importance to their clinical profile as antipsychotic drugs.
References [1] Ishimaru, M., Toru, M., 1997. The glutamate hypothesis of schizophrenia. Drugs 7, 47~7. [2] Ossowska, K., Pietraszek, M., Wardas, J., Nowak, G., Wolfarth, S., 1999. Chronic haloperidol and clozapine administration increases the number of cortical NMDA receptors in rats. Naunyn Schmiedeberg's Arch. Pharmacol. 359, 280-287.
$260
References [1] Lopes-Machado EZ, Crippa JAS, Hallak JEC, Guimaraes FS, Zuardi AW. "Electrodermallynonresponsiveschizophrenicsmake more errors in the stroop color word test, indicating selective attention deficit". Schizophrenia Bulletin (in press).
~ A
pharmacoepidemiological study of diabetes mellitus and antipsychotic treatment in the United States
E Cavazzoni 1, K. Hornbuckle 1, D. Hutchins2, L. Jovanovic3, A. Breier 1, J. Buse 4. 1Lilly Research Laboratories, Indianapolis, U.S.A.; 2Advance PCS, Scottsdale, U.S.A.; 3Sansum Medical Research Institute, Santa Barbara, U.S.A.; 4Diabetes Care Center, Chapel Hill U.S.A. A comparison of the odds of developing diabetes mellitus (DM) during exposure to various antipsychotics. Cases of treatmentemergent DM have been reported for both conventional and atypical antipsychotics. However, large epidemiological studies are needed to determine the existence and extent of any association between development of DM and antipsychotic medications and to resolve whether there is a substantial difference in the risk of developing DM among different antipsychotic therapies. This is a retrospective cohort study based on the prescription claim data of AdvancePCS Inc. in the US covering the period of 12/01/1997 to 8/31/2000. Prescription claims for antipsychotics were used to identify and select adult subjects that began antipsychotic monotherapy between 12/1/1998 and 2/29/2000. The antipsychotic cohorts studied were combined conventional antipsychotics (N = 19,782), haloperidol (N= 8,476), thioridazine (N=3,133), combined atypical antipsychotics (N= 38,969), olanzapine (N=13,863), risperidone (N=20,633), quetiapine (N=4,196), and clozapine (N=277). The AdvancePCS general patient population (N=5,816,473) included all adult AdvancePCS members who had made at least one prescription claim (other than an antipsychotic) between January 1, 2000 and February 29, 2000. Prescription claims for diabetes medications were used to identify subjects with DM. Those with antipsychotic prescription claim(s) 6 months prior to starting anti-diabetic therapy or those with anti-diabetic prescription claims any time prior to starting antipsychotic therapy were excluded. Cox Proportional Hazards regression was used to adjust for differences in age, gender, and duration of antipsychotic exposure between cohorts in the estimation of risk of developing diabetes. Hazard ratios (HRs) of DM in antipsychotic cohorts, were determined relative to the AdvancePCS general patient population and between selected antipsychotic cohorts. The relative risk of developing DM during treatment with conventional and atypical antipsychotics cohorts compared to the AdvancePCS general population were HR=3.5; CI: 3.1-3.9; p<.001 and HR=3.1; CI: 2.9-3.4; p<.001, respectively. The relative risk of developing DM during treatment with individual antipsychotics compared to the AdvancePCS general population were haloperidol HR=3.1; CI: 2.6-3.7; p<.001, thioridazine HR=4.2; CI: 3.2-5.5; p<.001, olanzapine HR=3.0; CI: 2.6-3.5; p<.001, risperidone HR=3.4; CI: 3.1-3.8; p<.001, quetiapine HR=I.7; CI: 1.2-2.4; p=.002, and clozapine HR=3.3; CI: 1.4-8.0; p=.007. The combined conventional and atypical cohorts were compared and there was no significant difference in risk of developing DM (HR=0.97; CI: 0.8-1.1; p=0.6). Compared to the haloperidol cohort, a statistically significant increase in risk of developing DM was observed in the
risperidone cohort HR=I.2; CI: 1.0-1.5; p= 0.04. There was no statistically significant difference between the risk of developing DM in the olanzapine and risperidone cohorts HR=0.9; CI: 0.81.1; p= 0.2. An increased risk of developing diabetes compared to a general reference population was observed in the AdvancePCS prescription-database cohorts during treatment with either conventional or atypical antipsychotics. Though the risk of developing diabetes was significantly greater for patients in the risperidone cohort than in the haloperidol cohort, this analysis did not demonstrate a generally elevated risk between the atypical and conventional antipsychotic cohorts. It remains unclear whether the observed increases are related to factors intrinsic or extrinsic to those psychiatric conditions commonly treated with antipsychotic drugs.
•
Cognitive and overall functioning in insUtutionalised patients taking typical versus atypical antipsychotics
J. Cervilla 1, R. Ana2, G. Hoyos 3, L. Minguez3, E. Gonzalez de Pablos 3. 1Sant Joan de D~u - Serveis de Salut Mental, Servei Salut Mental Garraf Vilanova i la Geltr{t, Barcelona, Spain; 2Hospital Santa Clotilde, Santander, Spain; 3Complejo Hospitalario San Luis, Palencia, Spain Atypical antipsychotics have been postulated as having general functioning and cognitive advantages over typical antipsychotics. Objective: The objective of this study was to investigate such notion using a sample of 149 (n=149) patients. Sample: 141 participants out of the 149 were women,the overall mean age 64.5 years) and two thirds had a diagnosis of schizofrenia. They were all long-term in-patients living in a psychitric institution (Complejo Hospitalario San Luis, in Palencia, Spain). Methods: A trained clinical psychologist and a geriatritian administrated a psychopathological scale (BPRS), a cognitive screening scale (the Spanish version of the Minimental State Examination validated by Lobo et al.), a functionality scale (Barthel index) and a behavioural disorders scale (The Spanish version of Cohen-Mansfield's Agitation inventory for the Elderly).We compared using non-parametric univariate statistics differences of scores in all such scales among those subjects taking an atipical antipsychotic (72 patiensts who took riperidone or olanzapine) and typical ones (haloperidol or chlorpromazine). Results: We did not find significant differences amongst the two groups for neither the psychopathological nor the behavioural/agitation scale. However, we did find significant differences showing better cognitive functioning (p=0.0001) in patients on atypicals compared to those on typical antipsychotics. We also found a significantly (p=0.00001) better overall functioning in patients taking atypical antipsychotics. In an interesting univariate subanalysis among the 72 patients taking atypical (risperidone vs. olanzapine) we did not find statistically significant differences for neither cognitive nor overall functioning. Nevertheless, there was a nearly sginifican trend for a linear association between those taking olanzapine and higher cognitive test scoring (p=0.068). Conclusion: We did not find differences in efficacy of psychological or behavioural symptoms among patients taking atypical versus typical antipsychotics. However, this study suggests that atypical antipsychotics may contribute to a better cognitive and overall functioning in aged adults who are chronically institutionalised because of a severe mental disorder. This results
P2. Psychotic disorders and antipsychotics
$261
should, however, be interpreted with caution as we cannot rule out potential confounding effects of age,sex, diagnosis or length of admission. However, the sample is very homogeneous for most of this independent variables.
accurate marker of the underlying genetic substrate than the assessed endophenotypes. Grants supported: Hacettepe Univ. Research Foundation, #01 01 101 001.
•
References
Endophenotype assessments in a large schizophrenia pedigree
[1] Leboyer Met al (1998) Psychiatric genetics: Search for phenotypes. Trends Neuroscienci 21: 102-105.
S. Ozer 1 , A.S. Gonul 2, S. Ulusoy 1 , E Heutink 3, S. Claes 4, A.N. Akarsu 5, A. Ulusahin 1 *. IDep. Psychiatry, Hacettepe Univ.
Med. Fac., Ankara, Turkey; 2Dep. Psychiatry, Erciyes Univ. Med. Fac., Kayseri, Turkey; 3Dep. Clin. Genetics Erasmus Univ., Rotterdam, The Netherlands; 4Dep. Psychiatry, Univ. Antwerp, Antwerp, Belgium; 5Gene Map. Lab., Dep. Pediatr., Hacettepe Univ. Med. Fac., Ankara, Turkey Bac kg r o und: Schizophrenia is a clinically and genetically complex disorder. Including endophenotypes may facilitate genetic research. Endophenotype approach focusing on subclinical traits may help identification of genes that increase the susceptibility for the illness. Objective: In a multigenerational large schizophrenia family, we assessed Axis I psychopathology according to RDC criteria, and also neurological soft signs and schizotypal personality traits as possible endophenotypes. M e t ho d: A large multigenerational pedigree demonstrating aggregation of schizophrenia and other psychiatric disorders was assessed. Most of the affected members of the family were treated in the regional university psychiatry department for ten years. The available hospital records of the living and deceased affected members were used to complete the clinical information. Of 90 living members from three generations 54 were interviewed with the SADS-L and 71 blood samples were collected for further molecular genetic study. For endophenotype assessment the Neurological Evaluation Scale (NES) and the schizotypal features related section of the Comprehensive Assessment of Symptoms and History (CASH) were used. Results: We identified a Turkish isolate (Turkmen tribe, province of Bozok) from Central Anatolia through 10 schizophrenic probands. The complete pedigree structure consisted of 358 subjects in total; historical records indicated that the ancestors of the family settled in the region around 1700. Marrying distant relatives was a common pattern in the family. RDC diagnoses of the interviewed members were as follows: schizophrenia and unspecified functional psychosis 10, affective disorders 17, alcohol and drug abuse 5, other diagnoses 16, no diagnosis 19. All schizophrenic members had similar paranoid features. Mean NES scores of schizophrenic and non-schizophrenic members were 164-9.49 and 9.93+6.00 respectively (Cut-off point of NES for Turkish population is 15). The appearance of glabellar reflex in 5 of the 6 examined schizophrenic members and 9 of the 40 nonschizophrenic subjects was noticable. Schizotypal features were common especially in the schizophrenic members. Social anxiety and suspiciousness were the most prevalent items. Conclusion: In this family with an exceptional prevalence of mental disorders two distinct putative markers and RDC schizophrenia criteria were evaluated. All l0 schizophrenic patients showed a clinically homogeneous phenotype with prominent persecutory delusions. On the other hand we couldn't observe any cosegregation of the assessed endophenotypes with schizophrenia, with the exception of glabbellar reflex. For this particular family, we conclude that the RDC schizophrenia phenotype is a more
•
Influence of long-lasting administration of neuroleptics on glutamatergic transmission in rats
M. Pietraszek 1, J. Wardas 1, K. Golembiowska 2, M. Bijak 3, S. Wolfarth 1, K. Ossowska 1 . Departments of 1Neuro-
Psychopharmacology, 2pharmacology and 3physiology, Institute of Pharmacology, Polish Academy of Sciences, Krakrw, Poland It has been postulated that a dysfunction of glutamatergic transmission, mainly in cortical regions, may contribute to the pathophysiology of schizophrenia (Ishimaru and Tom 1997). In our previous study we showed that chronic administration of neuroleptics increased the number of cortical NMDA receptors labelled with the competitive antagonist [3H]CGP 39653 (Ossowska et al. 1999). The purpose of the present study was to examine the influence of haloperidol, a typical neuroleptic, and clozapine, an atypical one, on AMPA receptors, as well as on the glutamatedependent neuronal activity and the extracellular levels of excitatory amino acids in rat cerebral cortex. Haloperidol (1 mg/kg/day) and clozapine (30 mg/kg/day) were administered to rats in drinking water for 6 weeks or 3 months, and were afterwards withdrawn for 4 days. Extracellular concentrations of basal and veratridine-stimulated glutamate (Glu) and aspartate (Asp) in the fronto-parietal cortex were assessed using an in vivo microdialysis; neuronal discharges that developed in a Mg2+-free medium were recorded extracellularly in cortical slices ex vivo. In autoradiographic studies, AMPA receptors were labelled with [3H]AMPA. Haloperidol elevated the basal, but not the veratridine-evoked, extracellular levels of Glu and Asp and enhanced the activity of cortical neurons. In contrast, clozapine decreased both the basal and the stimulated levels of Glu and Asp in the fronto-parietal cortex, but had no effect on the activity of cortical neurons. None of the neuroleptics influenced the binding of [3H]AMPA to AMPA receptors in the cerebral cortex. The present study shows that the two representatives of typical and atypical neuroleptics, administered chronically, alter the glutamatergic system in cortical regions. They have no effect on AMPA receptors, but differently regulate the extracelluler levels of Glu and Asp and the neuronal activity. The influence of the examined neuroleptics on cortical glutamatergic transmission may be of importance to their clinical profile as antipsychotic drugs.
References [1] Ishimaru, M., Toru, M., 1997. The glutamate hypothesis of schizophrenia. Drugs 7, 47~7. [2] Ossowska, K., Pietraszek, M., Wardas, J., Nowak, G., Wolfarth, S., 1999. Chronic haloperidol and clozapine administration increases the number of cortical NMDA receptors in rats. Naunyn Schmiedeberg's Arch. Pharmacol. 359, 280-287.