- Email: [email protected]
Stability of psychopathology dimensions in chronic schizophrenia: Response to clozapine treatment
Stability of Psychopathology Dimensions in Chronic Schizophrenia: Response to Clozapine Treatment Stanley R. Kay? and Jean-Pierre Lindenmayer Current models of schizophrenia postulate that different symptom complexes, including the positive and negative, may relate to fundamental underlying neurobiological distinctions. However, the premise of an underlying stability to the psychopathological profile has not been systematically investigated, particularly in response to pharmacological intervention. The present work aimed to study this issue in 14 chronic schizophrenic inpatients by comparing their symptom clusters before and after a 20-week course of clozapine treatment. The results indicated significant improvement on all eight symptom dimensions, as well as in severity of general psychopathology. Despite the clinical gains, most dimensions remained highly stable, with correlations between prestudy and clozapine week 20 ranging up to .91 (P < .OOOl) for the positive-negative composite score. These findings of stability over time, even in response to potent treatment, support the validity and importance of schizophrenic psychopathology dimensions, which appeared to possess fundamental traitlike characteristics. Copyright 0 1991 by W.B. Saunders Company
I
N THE 1990.3,more than three decades after neuroleptics were introduced for schizophrenia, there is growing recognition that not all facets of the illness are amenable to traditional pharmacotherapy. This has led to a revitalized interest in the symptom profile of schizophrenia and to studies of the syndrome-specific effects of existing and newer drug compounds. Such research has rather consistently demonstrated that negative or deficit symptoms are less responsive than positive or productive symptoms to the classical dopamine-blocking agents.‘-3 Findings of this kind have supported the views that negative features of schizophrenia may be unrelated to dopamine excess, may largely comprise the residual disorder, and, at least in the chronic stage, may portend an ominous course of illness.435Atypical neuroleptics that have been recently developed, such as clozapine and risperidone, reportedly produce a wider spectrum of treatment changes in schizophrenia, including amelioration of both the positive and negative syndromes.6-8 In addition, large-scale factorial studies have supported the validity and independence of the positive and negative features, but have also found that other components, such as excitement and depression, constitute additional independent parameters of schizophrenia.‘.” Any interpretation of the meaning of the distinct psychopathology profiles requires some knowledge about the stability over time and after treatment. The studies of positive and negative syndromes almost invariably have been conducted on already medicated patients, and so there is no certainty that the postmedication profiles are comparable to those that exist in the more natural state. In fact, From the Depattment of Psychiatry, Albert Einstein College of MedicineiMontefiore Medical Center and Bronx Psychiatric Center, Bronx, NY. fDeceased. Address reprint requests to Jean-Pierre Lindenmayer, M.D., Schizophrenia Research Program, Bronx Psychiatric Center, 1500 Waters Place, Bronx, NY 10461. Copyright 0 1991 hy W. B. Saunders Company 0010-440X/91/3201-0001$03.00/0 28
Comprehensive
Psychiatry,
Vol. 32, No. 1 (January/February),
1991: pp 28-35
STABILITY
OF PSYCHOPATHOLOGY
WITH CLOZAPINE
29
given the evidence that neuroleptics target specific aspects of psychopathology,’ there is reason to doubt the comparability of assessment at these two phases. Moreover, our judgments about the validity and significance of a specific psychopathology profile depend on information about its temporal stability and preservation, even in diminished form.” If the neuroleptic-treated patient no longer demonstrates the pretreatment pattern of symptoms, then it can be concluded that the presenting clinical profile was a temporary state-dependent one that might be detectable only in conditions of exacerbation. Conversely, if despite successful amelioration of symptoms the residual profile resembles the initial one, we may be witnessing a stable and characteristic picture that can be diminished, but not fundamentally dismantled. This conclusion is particularly compelling if the clinical profile is unaltered after successful treatment with a compound that addresses a broad range of psychopathology dimensions. In such a case, we would have evidence for a traitlike model of schizophrenic phenomenology, which would imply that the symptomatic configuration is enduring and perhaps intrinsic to the disease process itself. In a previous study, we investigated the stability of positive and negative syndromes in 62 acute and subacute schizophrenic inpatients before and after a 3to 4-month course on haloperidol or chlorpromazine.3 Both symptom dimensions, despite significant improvement, were found to be stable over time, ie, the baseline placebo scores reliably predicted those that remained after treatment. This observation suggested that positive and negative phenomena constitute persistent dimensions of psychopathology that sustain their relative clinical prominence, even with neuroleptic intervention. In contrast to our earlier investigation, the present work aimed to assess the stability of psychopathology dimensions in schizophrenia (1) along a broader range of symptom clusters, (2) with a chronic refractory group, and (3) using, instead of a classical neuroleptic, an atypical neuroleptic (clozapine) that is considered to more effectively intervene in multiple dimensions of the illness while not inducing neurological side effects. METHOD Subjects Fourteen neuroleptic refractory schizophrenic inpatients were recruited from the long-term units of Bronx Psychiatric Center, an urban state hospital in New York. They were selected consecutively on the basis of meeting all of the following criteria: (1) diagnosis of schizophrenia by both DSM-III-RI2 and Research Diagnostic Criteria” as ascertained from Schedule for Affective Disorders and Schizophrenia (SADS) interview14; (2) history of treatment resistance to at least three classes of neuroleptics; (3) presence of active and persistent psychotic symptoms; (4) no history or evidence of central nervous system disease, seizure disorder, or any serious concurrent medical condition; and (5) no diagnosis of substance abuse disorder by DSM-III-R criteria and no record of substance abuse in the past 6 months. The final sample included eight men and six women of mean age 33.6 years (SD 6.63; range, 24 to 45). All had an extensive history of schizophrenic disorder, with a mean of 15.1 years since first psychiatric hospitalization (SD 6.16) and a range of 7 to 26 years.
Design and Measures After signing informed consent, the patients were transferred to a closed Research Unit for a 20-week open trial on clozapine (Clozaril, Sandoz, East Hanover, NJ). They were entered onto the research protocol on a staggered basis to guard against behavioral contagion effects. Prior to clozapine treatment, patients were stabilized on typical neuroleptics (generally chlorpromazine or haloperidol)
KAY AND LINDENMAYER
30
in standard dose ranges that were titrated to achieve maximal clinical benefits. After baseline evaluation on the prestudy neuroleptic, the patients were tapered off the medication and subsequently placed on clozapine. The starting dose was 25 to 100 mgid (mean, 68.8), building up gradually to a mean of 296 mg/d by week 4 (range, 50 to 500 mg) and increasing further to means of 482,630, and 673 on weeks 8, 12, and 16, respectively. The clozapine dose in the final study phase (week 20) was stabilized at an average of 680 mg/d, with a range from 125 to 900 mg. Patients were assessed conjointly by two or more independent raters (research fellows in biological psychiatry and psychopharmacology) during the prestudy stabilization phase and again at the end of clozapine week 20. The key clinical efficacy parameter was the Positive and Negative Syndrome Scale (PANSS),” which was selected because it offers a well-operationalized and standardized method for assessing a broad range of symptom dimensions in schizophrenia, including positive and negative syndromes. This 30-item scale applies a formalized patient interview and rating definitions for each item at all levels of symptom severity, with a range from 1 (absent) to 7 (extreme).16 The sum of seven positive symptoms, such as delusions, hallucinations, and conceptual disorganization, constitutes a positive syndrome score; the sum of seven primary negative symptoms, such as blunted affect, passive/apathetic social withdrawal, and difficulty in abstract thinking, constitutes a negative syndrome score. The total positive minus negative score provides a bipolar composite index of the predominance of one syndrome in relation to the other. The remaining 16 items that cannot be definitively classified as positive or negative are summed to yield a general psychopathology score. In addition, since the PANSS incorporates the 18 items from the Brief Psychiatric Rating Scale (BPRS),” it can be similarly scored to assess the five BPRS factors: anergia, thought disturbance, activation, paranoid belligerence, and depression. Data on the validation and standardization of the PANSS have been elsewhere described,‘5,‘8,‘9 and its interrater reliability has been found to range from .81 to .88. For purposes of the present study, all raters were pretrained by standard videotape technique*” and were required to demonstrate an interrater agreement beyond .80 before participating in the research. Although the raters were nonblind to the medications, they were kept blind to the research objectives of the present investigation.
RESULTS Clinical Changes
Comparison of baseline psychopathology measures with those obtained after a 20-week trial on clozapine indicated significant improvements across all parameters (Table 1). General psychopathology changed by 15.5% (P < .OOl), but the most impressive benefits were in the positive spectrum of symptoms (25.5%, P < .OOOl). The positive minus negative composite index indicated far greater progress on the positive score (P < .0005), but this finding is mitigated by the relatively lower negative syndrome score shown by this sample in the baseline, which reflects the screening requirement of an active psychotic presentation. Among the BPRS factors derived from the PANSS, all were significantly improved, with the most prominent changes evident in paranoid belligerence (33.0%, P < .002), thought disturbance (18.2%, P < .OOOl), and activation (15.8%, P < 0.002). Clinical Stability
Despite the symptom reduction in all areas, the baseline prestudy values were in most cases highly correlated with those obtained after 20 weeks of clozapine treatment (Table 1). Thus, patients who scored relatively high or low on particular dimensions during the baseline exhibited a comparable pattern after treatment. Significant stability was shown for both positive and negative syndromes (r = .68 in each case, P < .005), and the positive-negative composite was especially
(7.22) (5.43) (6.75) (7.76) (3.42) (3.99) (2.24) (2.88) (2.68)
(SD) 39.4 21.9 20.1 1.8 9.4 14.4 6.4 7.1 10.1
Mean (SD) (6.09) (4.99) (4.31) (5.29) (1 .QQ) (4.03) (1.60) (2.81) (1.70)
Week 20
Symptoms
15.5 25.5 12.2 t 15.3 18.2 15.8 33.0 11.4
Percent Improvement 4.36 6.17 2.09 4.92 2.33 6.10 3.66 3.49 1.95
t i < < < < < < < <
Paired t Test P’
Treatment
.0005 .OOOl .05 .0002 .02 .OOOl .002 .002 .05
Dimensions After 20 Weeks of Clorapine
cannot be calculated, since the composite index comprises a bipolar scale that lacks a true zero point.
46.6 29.4 22.9 6.6 11.1 17.6 7.6 10.6 11.4
General psychopathology Positive syndrome Negative syndrome Positive-negative composite Anergia Thought disturbance Activation Paranoid belligerence Depression
*Directional tests. tPercent improvement
Mean
Symptom Dimension (PANSS)
Prestudy
Table 1. Changes and Stability of Schizophrenic
.59 .68 .68 .Ql .59 .87 .87 .13 .44
r < < < < < < <
.02 ,005 .005 .OOOl .02 .OOOl .OOOl NS NS
P”
Correlation, Prestudy to Week 20
KAY AND LINDENMAYER
32
consistent over time and treatment (r = .91, P < .OOOl).Among the BPRS factor scores from the PANSS, thought disturbance and activation were very stable (r = .87 in each case, P < .OOOl),and anergia was also significantly stable (I = .59, P < .02). Only paranoid belligerence and depression failed to show consistency across the 20-week interval, whereas the severity of general psychopathology was itself rather stable during this period. The associations between prestudy and final study week scores are illustrated in Fig 1 for the positive, negative, composite, and thought disturbance scales. These diagrams indicate that the ratings at clozapine week 20 can be reliably predicted from the prestudy values by regression equations. DISCUSSION The results demonstrate that clozapine is an effective antipsychotic agent even for a chronic refractory cohort of schizophrenic patients. The therapeutic effects were wide ranging, encompassing both positive and negative features, as well as other symptom clusters, such as depression and thought disturbance. Our findings are in accord with the double-blind clozapine studies of Honigfeld et a1,6Kane et al,’ and Lindstrom.” The present research using the PANSS provided confirmation with a scale that offered a broader scope of evaluation (i.e., 30 symptoms), a more strictly operationalized interview and rating system, and a standardized negative symptom scale.19 However, as an open-label study, these results have to be interpreted with caution. Even more intriguing than the drug effects was the observation of stability of symptom complexes. Thus, patients who at baseline were predominantly positive or negative in their clinical presentation remained so after an extensive course of treatment and despite the highly significant reduction of manifest symptoms. The stability of the positive-negative profile, as given by the longitudinal correlation for the composite index, was remarkably high (r = .91), which indicates greater than an 82% shared variance. Positive and negative ratings were equally stable
Clozo~ms Week 20
Cloza,we
Week
20
Clozo~in.
Clompme
Week
20
Week 20
Fig 1. Scatterplot of the correlations between ratings on prestudy and clozapine week 20 for PANSS positive syndrome, negative syndrome, composite index, and thought disturbance.
STABILITY
OF PSYCHOPATHOLOGY
WITH CLOZAPINE
33
(r = .68 each), and th ought disturbance and activation were the most stable factor scores (1.= .87 each). The only facets of psychopathology that appeared to lack stability were depression and paranoid belligerence. The instability of depression might reflect the different kinds of depression that prevail at different stages of treatment. One form of depression found in the active stage of schizophrenia tends to respond favorably to treatment22*u and to predict a successful reconstitution after 2 to 3 years for both acutez4 and chronic patients? By contrast, postpsychotic depression may develop in certain schizophrenics as other symptoms begin to remit,26 producing a heightened state of arousal and leading to a poorer therapy outcome.” The instability of paranoid belligerence may conceivably relate to its hybrid character and its association with depression. This interpretation is based on our recent orthogonal factor analytic study of the 30-item PANSS, which involved 240 schizophrenic inpatients.’ The findings revealed four principal components of psychopathology-negative, positive, excited, and depressive-which, in their interactions, seemed to explain the symptoms of the traditional Kraepelinian subtypes. Thus, for example, paranoid symptoms did not constitute an independent component of psychopathology; instead, these appeared to represent the interplay between distinct but coexisting positive and depressive components.9 The finding that psychopathology dimensions in chronic schizophrenia are stable even after pharmacological intervention suggests that these represent entrenched profiles resembling traitlike characteristics. Accordingly, the same pattern of features that comprised the patient’s clinical profile before treatment are likely to do so afterward. This implies a structural stability and validity of these clinical dimensions as integral components that may serve to subclassify schizophrenia. In this respect, the data support the value of diagnostic system that includes functional-dimensional measurement to characterize the particular dysfunctions within a broad nosological category, such as schizophrenia; the assessment points the way to more meaningful phenomenological patterns which may correspond to distinct biological processes.28 Although the present data were collected openIy, creating the possibility for an expectancy bias, it is noteworthy that a theoretical blind was maintained and that the expectation was more likely that of symptomatic change rather than constancy. Furthermore, our previous double-blind study of the stability of positive and negative syndromes in acute and subacute schizophrenic patients, following 3 to 4 months of treatment with classical neuroleptics (haloperidol or chlorpromazine), yielded the same conclusions.3 The effect of drug treatment in both investigations was to reduce the severity of manifestations, but not to fundamentally alter their predominance within the clinical picture. Such observations suggest that these symptom dimensions may have specific biological underpinnings, which, with our present armament of medications, can be pharmacologically modified but not eradicated. This position is consistent with a growing body of literature that yields evidence for the neurobiological significance of the positive-negative distinction (see reviews by Pogue-Geile and Zubin3” and Kay3l). The current and earlier 3 demonstrations of syndromal stability also suggest that studies conducted under posttreatment
34
KAY AND LINDENMAYER
conditions may be generalizable to the pretreated state. Further research is needed, of course, to determine the survival of these findings in other samples and with other antipsychotic agents. ACKNOWLEDGMENT The authors thank Leonardo Mabugat, M.D., Lisa M. Murrill, M.A., and Sukhwinder Singh, M.D., for their contributions to the data collection and compilation.
REFERENCES 1. Meltzer HY, Sommers AA, Luchins DJ: The effects of neuroleptics and other psychotropic drugs on negative symptoms in schizophrenia. J Clin Psychopharmacol6:329-338, 1986 2. Breier A, Wolkowitz OM, Doran AR, et al: Neuroleptic responsivity of negative and positive symptoms in schizophrenia. Am J Psychiatry 144:1549-1555,1987 3. Kay SR, Singh MM: The positive-negative distinction in drug-free schizophrenic patients: Stability, response to neuroleptics, and prognostic significance. Arch Gen Psychiatry 46:711-718,1989 4. Crow TJ: Positive and negative schizophrenic symptoms and the role of dopamine. Br J Psychiatry 137:383-386,198O 5. Andreasen NC: Positive versus negative schizophrenia: A critical evaluation. Schizophr Bull 380-389,1985 6. Honigfeld G, Patin J, Singer J: Clozapine: Antipsychotic activity in treatment-resistant schizophrenia. Adv Ther 1:77-91,1984 7. Kane JM, Honigfeld G, Singer J, et al: Clozapine in treatment-resistant schizophrenia. Arch Gen Psychiatry 45:789-796,1988 8. De Cuyper HJA: Risperidone in the treatment of chronic psychotic patients: An overview of the double-blind comparative studies, in Ayd FJ Jr (ed): Thirty Years Janssen Research in Psychiatry. Baltimore, MD, Ayd Medical Communications, 1989, p 115 9. Kay SR, Sevy S: Pyramidical model of schizophrenia. Schizophr Bull voll6,1990 10. Lepine JP, Piron J, Chapotot E: Factor analysis of the PANSS in schizophrenic patients, in Stefanis CN, Soldatos CR, Rabavilas AD (eds): Psychiatry Today: Accomplishments and Promises. Amsterdam, The Netherlands, Excerpta Medica, 1989, p 828 11. Sommers AA: “Negative symptoms”: Conceptual and methodological problems. Schizophr Bull 11:364-379,1985 12. American Psychiatric Association: DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders (ed 3, revised). Washington, DC, American Psychiatric Association, 1987 13. Spitzer RL, Endicott JE, Robins E: Research Diagnostic Criteria (RDC) for a Selected Group of Functional Disorders (ed 3). New York, NY, Biometrics Research, New York State Psychiatric Institute, 1978 14. Endicott J, Spitzer RL: A diagnostic interview: The Schedule for Affective Disorders and Schizophrenia. Arch Gen Psychiatry 35:837-844,1978 15. Kay SR, Fiszbein A, Opler LA: The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull 13:261-276,1987 16. Kay SR, Opler LA, Fiszbein A: Positive and Negative Syndrome Scale (PANSS) Manual. Toronto, Canada, Multi-Health Systems (in press) 17. Overall JE, Gorham DR: Brief Psychiatric Rating Scale. Psycho1 Rep 10~799812, 1962 18. Kay SR, Opler LA, Lindenmayer JP: Reliability and validity of the Positive and Negative Syndrome Scale for schizophrenics. Psychiatry Res 23:99-110,1988 19. Kay SR, Opler LA, Lindenmayer JP: The Positive and Negative Syndrome Scale (PANSS): Rationale and standardization. Br J Psychiatry 155:59-65,1989 (suppl7) 20. Kay SR, Opler LA, Fiszbein A: PANSS Training Videotapes (vol 1 and 2). Piscataway, NJ, Janssen Research Foundation, 1989 21. Lindstrom LH: The effect of long-term treatment with clozapine in schizophrenia. Acta Psychiatr Stand 77:524-529,1988 22. Vaillant G: Prospective prediction of schizophrenic remission. Arch Gen Psychiatry 11:509-518, 1964 23. Astrup C, Noreik K: Functional Psychosis: Diagnostic and Prognostic Models. Springfield, IL, Thomas. 1966
STABILITY OF PSYCHOPATHOLOGY
WITH CLOZAPINE
35
24. Lindenmayer JP, Kay SR, Friedman C: Negative and positive schizophrenic syndromes after the acute phase: A prospective follow-up. Compr Psychiatry 27:276-286,1986 25. Kay SR, Murrill LM: Predicting outcome of schizophrenia: Significance of symptom profiles and outcome dimensions. Compr Psychiatry 31:91-102, 1990 26. McGlashan T, Carpenter WT Jr: An investigation of the postpsychotic depressive syndrome. Am J Psychiatry 133:14-19,1976 27. Singh MM, Kay SR: Dysphoric response to neuroleptic treatment in schizophrenia: Its relationship to autonomic arousal and prognosis. Biol Psychiatry 14:277-294,1979 28. van Praag HM, Leijnse B: Neubewertung des Syndroms: Skizze einer functionallen Pathologie. Psychiatr Neurol Neurochir 68:50-66,1965 29. van Praag HM, Korf J, Lakke JPWF, et al: Dopamine metabolism in depression, psychoses, and Parkinson’s disease: The problem of the specificity of biological variables in behavioral disorders. Psycho1 Med 5:138-146, 1965 30. Pogue-Geile MF, Zubin J: Negative symptomatology in schizophrenia: A conceptual and empirical review. Int J Ment Health 16:3-45, 1988 31. Kay SR: Positive and Negative Syndromes in Schizophrenia: Assessment and Research. New York, NY, Brunner/Mazel (in press)
I
N THE 1990.3,more than three decades after neuroleptics were introduced for schizophrenia, there is growing recognition that not all facets of the illness are amenable to traditional pharmacotherapy. This has led to a revitalized interest in the symptom profile of schizophrenia and to studies of the syndrome-specific effects of existing and newer drug compounds. Such research has rather consistently demonstrated that negative or deficit symptoms are less responsive than positive or productive symptoms to the classical dopamine-blocking agents.‘-3 Findings of this kind have supported the views that negative features of schizophrenia may be unrelated to dopamine excess, may largely comprise the residual disorder, and, at least in the chronic stage, may portend an ominous course of illness.435Atypical neuroleptics that have been recently developed, such as clozapine and risperidone, reportedly produce a wider spectrum of treatment changes in schizophrenia, including amelioration of both the positive and negative syndromes.6-8 In addition, large-scale factorial studies have supported the validity and independence of the positive and negative features, but have also found that other components, such as excitement and depression, constitute additional independent parameters of schizophrenia.‘.” Any interpretation of the meaning of the distinct psychopathology profiles requires some knowledge about the stability over time and after treatment. The studies of positive and negative syndromes almost invariably have been conducted on already medicated patients, and so there is no certainty that the postmedication profiles are comparable to those that exist in the more natural state. In fact, From the Depattment of Psychiatry, Albert Einstein College of MedicineiMontefiore Medical Center and Bronx Psychiatric Center, Bronx, NY. fDeceased. Address reprint requests to Jean-Pierre Lindenmayer, M.D., Schizophrenia Research Program, Bronx Psychiatric Center, 1500 Waters Place, Bronx, NY 10461. Copyright 0 1991 hy W. B. Saunders Company 0010-440X/91/3201-0001$03.00/0 28
Comprehensive
Psychiatry,
Vol. 32, No. 1 (January/February),
1991: pp 28-35
STABILITY
OF PSYCHOPATHOLOGY
WITH CLOZAPINE
29
given the evidence that neuroleptics target specific aspects of psychopathology,’ there is reason to doubt the comparability of assessment at these two phases. Moreover, our judgments about the validity and significance of a specific psychopathology profile depend on information about its temporal stability and preservation, even in diminished form.” If the neuroleptic-treated patient no longer demonstrates the pretreatment pattern of symptoms, then it can be concluded that the presenting clinical profile was a temporary state-dependent one that might be detectable only in conditions of exacerbation. Conversely, if despite successful amelioration of symptoms the residual profile resembles the initial one, we may be witnessing a stable and characteristic picture that can be diminished, but not fundamentally dismantled. This conclusion is particularly compelling if the clinical profile is unaltered after successful treatment with a compound that addresses a broad range of psychopathology dimensions. In such a case, we would have evidence for a traitlike model of schizophrenic phenomenology, which would imply that the symptomatic configuration is enduring and perhaps intrinsic to the disease process itself. In a previous study, we investigated the stability of positive and negative syndromes in 62 acute and subacute schizophrenic inpatients before and after a 3to 4-month course on haloperidol or chlorpromazine.3 Both symptom dimensions, despite significant improvement, were found to be stable over time, ie, the baseline placebo scores reliably predicted those that remained after treatment. This observation suggested that positive and negative phenomena constitute persistent dimensions of psychopathology that sustain their relative clinical prominence, even with neuroleptic intervention. In contrast to our earlier investigation, the present work aimed to assess the stability of psychopathology dimensions in schizophrenia (1) along a broader range of symptom clusters, (2) with a chronic refractory group, and (3) using, instead of a classical neuroleptic, an atypical neuroleptic (clozapine) that is considered to more effectively intervene in multiple dimensions of the illness while not inducing neurological side effects. METHOD Subjects Fourteen neuroleptic refractory schizophrenic inpatients were recruited from the long-term units of Bronx Psychiatric Center, an urban state hospital in New York. They were selected consecutively on the basis of meeting all of the following criteria: (1) diagnosis of schizophrenia by both DSM-III-RI2 and Research Diagnostic Criteria” as ascertained from Schedule for Affective Disorders and Schizophrenia (SADS) interview14; (2) history of treatment resistance to at least three classes of neuroleptics; (3) presence of active and persistent psychotic symptoms; (4) no history or evidence of central nervous system disease, seizure disorder, or any serious concurrent medical condition; and (5) no diagnosis of substance abuse disorder by DSM-III-R criteria and no record of substance abuse in the past 6 months. The final sample included eight men and six women of mean age 33.6 years (SD 6.63; range, 24 to 45). All had an extensive history of schizophrenic disorder, with a mean of 15.1 years since first psychiatric hospitalization (SD 6.16) and a range of 7 to 26 years.
Design and Measures After signing informed consent, the patients were transferred to a closed Research Unit for a 20-week open trial on clozapine (Clozaril, Sandoz, East Hanover, NJ). They were entered onto the research protocol on a staggered basis to guard against behavioral contagion effects. Prior to clozapine treatment, patients were stabilized on typical neuroleptics (generally chlorpromazine or haloperidol)
KAY AND LINDENMAYER
30
in standard dose ranges that were titrated to achieve maximal clinical benefits. After baseline evaluation on the prestudy neuroleptic, the patients were tapered off the medication and subsequently placed on clozapine. The starting dose was 25 to 100 mgid (mean, 68.8), building up gradually to a mean of 296 mg/d by week 4 (range, 50 to 500 mg) and increasing further to means of 482,630, and 673 on weeks 8, 12, and 16, respectively. The clozapine dose in the final study phase (week 20) was stabilized at an average of 680 mg/d, with a range from 125 to 900 mg. Patients were assessed conjointly by two or more independent raters (research fellows in biological psychiatry and psychopharmacology) during the prestudy stabilization phase and again at the end of clozapine week 20. The key clinical efficacy parameter was the Positive and Negative Syndrome Scale (PANSS),” which was selected because it offers a well-operationalized and standardized method for assessing a broad range of symptom dimensions in schizophrenia, including positive and negative syndromes. This 30-item scale applies a formalized patient interview and rating definitions for each item at all levels of symptom severity, with a range from 1 (absent) to 7 (extreme).16 The sum of seven positive symptoms, such as delusions, hallucinations, and conceptual disorganization, constitutes a positive syndrome score; the sum of seven primary negative symptoms, such as blunted affect, passive/apathetic social withdrawal, and difficulty in abstract thinking, constitutes a negative syndrome score. The total positive minus negative score provides a bipolar composite index of the predominance of one syndrome in relation to the other. The remaining 16 items that cannot be definitively classified as positive or negative are summed to yield a general psychopathology score. In addition, since the PANSS incorporates the 18 items from the Brief Psychiatric Rating Scale (BPRS),” it can be similarly scored to assess the five BPRS factors: anergia, thought disturbance, activation, paranoid belligerence, and depression. Data on the validation and standardization of the PANSS have been elsewhere described,‘5,‘8,‘9 and its interrater reliability has been found to range from .81 to .88. For purposes of the present study, all raters were pretrained by standard videotape technique*” and were required to demonstrate an interrater agreement beyond .80 before participating in the research. Although the raters were nonblind to the medications, they were kept blind to the research objectives of the present investigation.
RESULTS Clinical Changes
Comparison of baseline psychopathology measures with those obtained after a 20-week trial on clozapine indicated significant improvements across all parameters (Table 1). General psychopathology changed by 15.5% (P < .OOl), but the most impressive benefits were in the positive spectrum of symptoms (25.5%, P < .OOOl). The positive minus negative composite index indicated far greater progress on the positive score (P < .0005), but this finding is mitigated by the relatively lower negative syndrome score shown by this sample in the baseline, which reflects the screening requirement of an active psychotic presentation. Among the BPRS factors derived from the PANSS, all were significantly improved, with the most prominent changes evident in paranoid belligerence (33.0%, P < .002), thought disturbance (18.2%, P < .OOOl), and activation (15.8%, P < 0.002). Clinical Stability
Despite the symptom reduction in all areas, the baseline prestudy values were in most cases highly correlated with those obtained after 20 weeks of clozapine treatment (Table 1). Thus, patients who scored relatively high or low on particular dimensions during the baseline exhibited a comparable pattern after treatment. Significant stability was shown for both positive and negative syndromes (r = .68 in each case, P < .005), and the positive-negative composite was especially
(7.22) (5.43) (6.75) (7.76) (3.42) (3.99) (2.24) (2.88) (2.68)
(SD) 39.4 21.9 20.1 1.8 9.4 14.4 6.4 7.1 10.1
Mean (SD) (6.09) (4.99) (4.31) (5.29) (1 .QQ) (4.03) (1.60) (2.81) (1.70)
Week 20
Symptoms
15.5 25.5 12.2 t 15.3 18.2 15.8 33.0 11.4
Percent Improvement 4.36 6.17 2.09 4.92 2.33 6.10 3.66 3.49 1.95
t i < < < < < < < <
Paired t Test P’
Treatment
.0005 .OOOl .05 .0002 .02 .OOOl .002 .002 .05
Dimensions After 20 Weeks of Clorapine
cannot be calculated, since the composite index comprises a bipolar scale that lacks a true zero point.
46.6 29.4 22.9 6.6 11.1 17.6 7.6 10.6 11.4
General psychopathology Positive syndrome Negative syndrome Positive-negative composite Anergia Thought disturbance Activation Paranoid belligerence Depression
*Directional tests. tPercent improvement
Mean
Symptom Dimension (PANSS)
Prestudy
Table 1. Changes and Stability of Schizophrenic
.59 .68 .68 .Ql .59 .87 .87 .13 .44
r < < < < < < <
.02 ,005 .005 .OOOl .02 .OOOl .OOOl NS NS
P”
Correlation, Prestudy to Week 20
KAY AND LINDENMAYER
32
consistent over time and treatment (r = .91, P < .OOOl).Among the BPRS factor scores from the PANSS, thought disturbance and activation were very stable (r = .87 in each case, P < .OOOl),and anergia was also significantly stable (I = .59, P < .02). Only paranoid belligerence and depression failed to show consistency across the 20-week interval, whereas the severity of general psychopathology was itself rather stable during this period. The associations between prestudy and final study week scores are illustrated in Fig 1 for the positive, negative, composite, and thought disturbance scales. These diagrams indicate that the ratings at clozapine week 20 can be reliably predicted from the prestudy values by regression equations. DISCUSSION The results demonstrate that clozapine is an effective antipsychotic agent even for a chronic refractory cohort of schizophrenic patients. The therapeutic effects were wide ranging, encompassing both positive and negative features, as well as other symptom clusters, such as depression and thought disturbance. Our findings are in accord with the double-blind clozapine studies of Honigfeld et a1,6Kane et al,’ and Lindstrom.” The present research using the PANSS provided confirmation with a scale that offered a broader scope of evaluation (i.e., 30 symptoms), a more strictly operationalized interview and rating system, and a standardized negative symptom scale.19 However, as an open-label study, these results have to be interpreted with caution. Even more intriguing than the drug effects was the observation of stability of symptom complexes. Thus, patients who at baseline were predominantly positive or negative in their clinical presentation remained so after an extensive course of treatment and despite the highly significant reduction of manifest symptoms. The stability of the positive-negative profile, as given by the longitudinal correlation for the composite index, was remarkably high (r = .91), which indicates greater than an 82% shared variance. Positive and negative ratings were equally stable
Clozo~ms Week 20
Cloza,we
Week
20
Clozo~in.
Clompme
Week
20
Week 20
Fig 1. Scatterplot of the correlations between ratings on prestudy and clozapine week 20 for PANSS positive syndrome, negative syndrome, composite index, and thought disturbance.
STABILITY
OF PSYCHOPATHOLOGY
WITH CLOZAPINE
33
(r = .68 each), and th ought disturbance and activation were the most stable factor scores (1.= .87 each). The only facets of psychopathology that appeared to lack stability were depression and paranoid belligerence. The instability of depression might reflect the different kinds of depression that prevail at different stages of treatment. One form of depression found in the active stage of schizophrenia tends to respond favorably to treatment22*u and to predict a successful reconstitution after 2 to 3 years for both acutez4 and chronic patients? By contrast, postpsychotic depression may develop in certain schizophrenics as other symptoms begin to remit,26 producing a heightened state of arousal and leading to a poorer therapy outcome.” The instability of paranoid belligerence may conceivably relate to its hybrid character and its association with depression. This interpretation is based on our recent orthogonal factor analytic study of the 30-item PANSS, which involved 240 schizophrenic inpatients.’ The findings revealed four principal components of psychopathology-negative, positive, excited, and depressive-which, in their interactions, seemed to explain the symptoms of the traditional Kraepelinian subtypes. Thus, for example, paranoid symptoms did not constitute an independent component of psychopathology; instead, these appeared to represent the interplay between distinct but coexisting positive and depressive components.9 The finding that psychopathology dimensions in chronic schizophrenia are stable even after pharmacological intervention suggests that these represent entrenched profiles resembling traitlike characteristics. Accordingly, the same pattern of features that comprised the patient’s clinical profile before treatment are likely to do so afterward. This implies a structural stability and validity of these clinical dimensions as integral components that may serve to subclassify schizophrenia. In this respect, the data support the value of diagnostic system that includes functional-dimensional measurement to characterize the particular dysfunctions within a broad nosological category, such as schizophrenia; the assessment points the way to more meaningful phenomenological patterns which may correspond to distinct biological processes.28 Although the present data were collected openIy, creating the possibility for an expectancy bias, it is noteworthy that a theoretical blind was maintained and that the expectation was more likely that of symptomatic change rather than constancy. Furthermore, our previous double-blind study of the stability of positive and negative syndromes in acute and subacute schizophrenic patients, following 3 to 4 months of treatment with classical neuroleptics (haloperidol or chlorpromazine), yielded the same conclusions.3 The effect of drug treatment in both investigations was to reduce the severity of manifestations, but not to fundamentally alter their predominance within the clinical picture. Such observations suggest that these symptom dimensions may have specific biological underpinnings, which, with our present armament of medications, can be pharmacologically modified but not eradicated. This position is consistent with a growing body of literature that yields evidence for the neurobiological significance of the positive-negative distinction (see reviews by Pogue-Geile and Zubin3” and Kay3l). The current and earlier 3 demonstrations of syndromal stability also suggest that studies conducted under posttreatment
34
KAY AND LINDENMAYER
conditions may be generalizable to the pretreated state. Further research is needed, of course, to determine the survival of these findings in other samples and with other antipsychotic agents. ACKNOWLEDGMENT The authors thank Leonardo Mabugat, M.D., Lisa M. Murrill, M.A., and Sukhwinder Singh, M.D., for their contributions to the data collection and compilation.
REFERENCES 1. Meltzer HY, Sommers AA, Luchins DJ: The effects of neuroleptics and other psychotropic drugs on negative symptoms in schizophrenia. J Clin Psychopharmacol6:329-338, 1986 2. Breier A, Wolkowitz OM, Doran AR, et al: Neuroleptic responsivity of negative and positive symptoms in schizophrenia. Am J Psychiatry 144:1549-1555,1987 3. Kay SR, Singh MM: The positive-negative distinction in drug-free schizophrenic patients: Stability, response to neuroleptics, and prognostic significance. Arch Gen Psychiatry 46:711-718,1989 4. Crow TJ: Positive and negative schizophrenic symptoms and the role of dopamine. Br J Psychiatry 137:383-386,198O 5. Andreasen NC: Positive versus negative schizophrenia: A critical evaluation. Schizophr Bull 380-389,1985 6. Honigfeld G, Patin J, Singer J: Clozapine: Antipsychotic activity in treatment-resistant schizophrenia. Adv Ther 1:77-91,1984 7. Kane JM, Honigfeld G, Singer J, et al: Clozapine in treatment-resistant schizophrenia. Arch Gen Psychiatry 45:789-796,1988 8. De Cuyper HJA: Risperidone in the treatment of chronic psychotic patients: An overview of the double-blind comparative studies, in Ayd FJ Jr (ed): Thirty Years Janssen Research in Psychiatry. Baltimore, MD, Ayd Medical Communications, 1989, p 115 9. Kay SR, Sevy S: Pyramidical model of schizophrenia. Schizophr Bull voll6,1990 10. Lepine JP, Piron J, Chapotot E: Factor analysis of the PANSS in schizophrenic patients, in Stefanis CN, Soldatos CR, Rabavilas AD (eds): Psychiatry Today: Accomplishments and Promises. Amsterdam, The Netherlands, Excerpta Medica, 1989, p 828 11. Sommers AA: “Negative symptoms”: Conceptual and methodological problems. Schizophr Bull 11:364-379,1985 12. American Psychiatric Association: DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders (ed 3, revised). Washington, DC, American Psychiatric Association, 1987 13. Spitzer RL, Endicott JE, Robins E: Research Diagnostic Criteria (RDC) for a Selected Group of Functional Disorders (ed 3). New York, NY, Biometrics Research, New York State Psychiatric Institute, 1978 14. Endicott J, Spitzer RL: A diagnostic interview: The Schedule for Affective Disorders and Schizophrenia. Arch Gen Psychiatry 35:837-844,1978 15. Kay SR, Fiszbein A, Opler LA: The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull 13:261-276,1987 16. Kay SR, Opler LA, Fiszbein A: Positive and Negative Syndrome Scale (PANSS) Manual. Toronto, Canada, Multi-Health Systems (in press) 17. Overall JE, Gorham DR: Brief Psychiatric Rating Scale. Psycho1 Rep 10~799812, 1962 18. Kay SR, Opler LA, Lindenmayer JP: Reliability and validity of the Positive and Negative Syndrome Scale for schizophrenics. Psychiatry Res 23:99-110,1988 19. Kay SR, Opler LA, Lindenmayer JP: The Positive and Negative Syndrome Scale (PANSS): Rationale and standardization. Br J Psychiatry 155:59-65,1989 (suppl7) 20. Kay SR, Opler LA, Fiszbein A: PANSS Training Videotapes (vol 1 and 2). Piscataway, NJ, Janssen Research Foundation, 1989 21. Lindstrom LH: The effect of long-term treatment with clozapine in schizophrenia. Acta Psychiatr Stand 77:524-529,1988 22. Vaillant G: Prospective prediction of schizophrenic remission. Arch Gen Psychiatry 11:509-518, 1964 23. Astrup C, Noreik K: Functional Psychosis: Diagnostic and Prognostic Models. Springfield, IL, Thomas. 1966
STABILITY OF PSYCHOPATHOLOGY
WITH CLOZAPINE
35
24. Lindenmayer JP, Kay SR, Friedman C: Negative and positive schizophrenic syndromes after the acute phase: A prospective follow-up. Compr Psychiatry 27:276-286,1986 25. Kay SR, Murrill LM: Predicting outcome of schizophrenia: Significance of symptom profiles and outcome dimensions. Compr Psychiatry 31:91-102, 1990 26. McGlashan T, Carpenter WT Jr: An investigation of the postpsychotic depressive syndrome. Am J Psychiatry 133:14-19,1976 27. Singh MM, Kay SR: Dysphoric response to neuroleptic treatment in schizophrenia: Its relationship to autonomic arousal and prognosis. Biol Psychiatry 14:277-294,1979 28. van Praag HM, Leijnse B: Neubewertung des Syndroms: Skizze einer functionallen Pathologie. Psychiatr Neurol Neurochir 68:50-66,1965 29. van Praag HM, Korf J, Lakke JPWF, et al: Dopamine metabolism in depression, psychoses, and Parkinson’s disease: The problem of the specificity of biological variables in behavioral disorders. Psycho1 Med 5:138-146, 1965 30. Pogue-Geile MF, Zubin J: Negative symptomatology in schizophrenia: A conceptual and empirical review. Int J Ment Health 16:3-45, 1988 31. Kay SR: Positive and Negative Syndromes in Schizophrenia: Assessment and Research. New York, NY, Brunner/Mazel (in press)