- Email: [email protected]
Changes in psychopathology and dyskinesia after neuroleptic withdrawal in a double-blind design
Schizophrenia Research, 10 (1993) 261-211 0 1993 Elsevier Science Publishers B.V. All rights
SCHRES
261 reserved
0920-9964/93/$06.00
003 19
Changes in psychopathology and dyskinesia after neuroleptic withdrawal in a double-blind design Lisa Dixona, Gunvant Thakerb, Robert Conleyb, David Rossb, Nicola Cascella’, Carol Tammingab “University of Maryland, Department of Psychiatry, Baltimore, MD 21201, USA, bMaryland Psychiatric Research Center, Baltimore, MD 21228, USA and ‘Pharm Italia Carlo Erba, Milano, Italy (Received
19 October
1993; revision received 2 February
1993; accepted
10 February
1993)
The goal of this study was to assess the time course of change in psychopathology and dyskinesia after neuroleptic withdrawal. Fifteen DSM-III schizophrenic patients were abruptly withdrawn in a doubleblind fashion from stable haloperidol treatment. Weekly ratings of dyskinesia and psychopathology were performed for 4 weeks post-withdrawal. There was an overall increase in dyskinesia ratings over the 4-week period (p < 0.05) beginning in week 2, with dyskinetic movements of the fingers showing the most significant increase (p
Dyskinesia;
Neuroleptic
withdrawal
INTRODUCTION
Neuroleptic drugs have a potent antipsychotic action in schizophrenia and an antidyskinetic effect in neuroleptic-induced dyskinesias. In schizophrenia, the onset of drug action is immediate, but its full effects are gradual (Klein and Davis, 1969), whereas in dyskinesias, the drug effect is immediate and full (Smith et al. 1977). Typically, treatment of schizophrenia with these agents is chronic, due to the persistent nature of psychotic symptoms. However, because treatment is continual, the actual nature of the underlying symptomatology Correspondence to: C. Tamminga, Maryland Psychiatric Research Center, P.O. Box 21247, Baltimore, MD 21228, USA.
is obscured. Withdrawal of neuroleptic medication and the time course of symptom reemergence could be informative about underlying symptoms and perhaps the associated pathophysiology. Neuroleptic withdrawal in animals is followed by a hyperdopaminergic state, associated with the exposed, upregulated dopamine receptors, as seen in behavioral and biochemical measures (Rupniak et al., 1985). In humans, the same phenomena may occur in some or all dopamine-related functions, where neuroleptic withdrawal leads to a rebound effect in the early drug-free period. This assumption has been supported in clinical studies showing prolactin over-reduction immediately postwithdrawal (Kirkpatrick et al., 1989) and changes in sleep parameters (Thaker et al., 1989). Whether this occurs in more complex behavior functions
268
like cognition and movement is a question of this study. Withdrawal of antipsychotic medication has been reported to produce a heterogeneous response in psychopathology. Rapid relapse rates vary from 25 to 60%. Perenyi et al. (1985) abruptly withdrew 14 chronic schizophrenic inpatients from neuroleptic therapy, and 8 relapsed significantly within 5 weeks. Glazer et al. (1989) withdrew 19 schizophrenic patients with tardive dyskinesia (TD) from neuroleptic medication over a 7- 10 day period and found that 5 exhibited ‘psychological relapse’ requiring reinstitution of medication. In contrast, Marder et al. (1979) withdrew 22 psychotic schizophrenic inpatients from neuroleptics and found that 8 patients actually improved during the 30-day drug-free interval. Efforts to predict relapse have led to the finding of an association of increased plasma homovanillic acid (HVA) with relapse. In the study of Glazer et al. (1989), early relapsing patients had lower baseline and a significantly greater increase in plasma HVA after discontinuation than nonrelapsing patients. Davidson et al. (1991) studied 23 schizophrenic inpatients after neuroleptic discontinuation and found that plasma HVA levels were higher in the 9 patients who relapsed compared with those who did not. However, this finding has not been consistently replicated, possibly due to the complex methodologic issues surrounding HVA assessment. Withdrawal dyskinesia and worsening of dyskinetic symptoms in patients with tardive dyskinesia (TD) have also been observed after neuroleptic withdrawal. Ten of the 19 TD patients studied by Glazer et al. (1989) developed withdrawal emergent TD. Patients developing withdrawal emergent TD were younger and had a lower exposure to neuroleptic medication than patients who did not develop withdrawal emergent TD. Seven of the patients (50%) in the study by Perenyi et al. (1985) developed withdrawal dyskinesia. Most of these studies have not employed a double blind design. The purpose of this study was to characterize quantitative and temporal changes in clinical symptoms and dyskinetic movements after blind, abrupt, neuroleptic withdrawal in a structured hospital setting in severely ill, but stable schizophrenic patients. Further, the study aimed to assess the relationship between psychopatho-
logic changes and changes in dyskinetic movements given their probable linkage to brain dopamine systems.
METHODS
Fifteen RDC-diagnosed DSM-III schizophrenic patients admitted to the Maryland Psychiatric Research Center inpatient unit from 1985 to 1989 were enrolled in the study after giving informed consent. Patients were seriously ill with persistent symptoms. However, they were clinically stable on the ward for at least 2 months at the time of the study. As soon as possible after admission to the ward, all patients were placed on a dose of haloperidol equivalent to the dose of neuroleptic they had received prior to admission using the conversion of Baldessarini equivalences (1985). Haloperidol doses ranged from 5 mg to 40 mg. Antiparkinsonian agents were used sparingly on an as needed basis. All patients were free of parenteral neuroleptic medication for at least six months prior to the study. The withdrawal study was planned to last for 8 weeks. In a modified double-blind, placebo controlled design, all patients were withdrawn from haloperidol abruptly during the first four weeks of the study. At the time of withdrawal, patients received placebo capsules which were identical in appearance to active haloperidol. Both patients and staff were aware that patients would begin to receive placebo sometime within the first four weeks of the study, but neither group was aware of exactly when this would occur. Patients had drug-free ratings for at least 4 weeks of the study. Psychopathology was assessed using the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962) which was performed weekly by two blind raters who then developed a consensus rating. Reliability was adequate for the thought disorder (0.86), anxiety (0.79) and hostility (0.78) subscales. It was marginal for the withdrawal subscale (0.55) and poor for activation (0.24). Dyskinesia ratings were assessed with the MPRC Involuntary Movement Scale. The MPRC scale is an 8-point scale, rating movements from absent to severe based on severity and persistence of movements. All dyskinesia evaluations were videotaped
269
and ratings of videotapes were performed by blind raters. All raters were trained in the use of the MPRC Involuntary Movement Scale, and the overall reliability at the center of this scale is 0.81. The same rater performed all dyskinesia ratings for a given patient. Data analysis was done using a repeated measures analysis of variance on BPRS total scores and dimensions including thought disorder, anxiety, activation, hostility and withdrawal as well as on global TD scores and and individual body area scores. Four patients had incomplete ratings of dyskinesia due to deterioration of videotapes and were thus excluded from repeated measures analysis. Polynomial contrasts were used to assess trends over time. Patients were given a diagnosis of TD based on the Schooler and Kane criteria (1982); TD subjects all had a score of 2 or more on the global dyskinesia scale at baseline indicative of at least mild TD.
RESULTS
Patients were relatively young, predominantly Caucasian males, with an age of onset of illness of 18.7 (4.1) years (Table 1). All had a diagnosis of schizophrenia, and 6 had a diagnosis of TD. A significant linear trend was found for global dyskinesia ratings from baseline to week 4 postneuroleptic withdrawal (F= 5.35, df = (1, lo), ~~0.05). Analysis of all patients showed an increase in weekly global dyskinesia rating over this time period from 1.36 to 2.36 (Table 2). Dyskinetic movements of the fingers showed the most significant increase. There was no interaction
TABLE
1
Demographic and Illness Schizophrenic Patients
Age Years of neuroleptic treatment Years of illness Age of onset of illness Sex
Characteristics
of
DSM-III- R
Mean _+SD
Range
28k5.2 8.4k4.4 9.3 t_ 4.2 18.7+4.1 9 male
IS-40 4-20 4-20 9-29 6 female
between the presence of baseline TD and change in TD over time (Table 2). In contrast to dyskinesia, repeated measures analysis of variance revealed no changes in psychopathology from baseline to week 4 post-withdrawal on any BPRS dimension or item (Table 3). However, assessment of change in thought disorder in individual patients suggested a bimodal distribution. Six patients (40%) had increases of 3-7 points indicative of relapse (mean increase= 5.0 f 1.55)). Eight patients (53%) were essentially unchanged with a change in thought disorder score of less than 2 (mean change= - 0.511.20)). One patient (7%) improved with a reduction in thought disorder of 5 points. Only 3 patients (20%) demonstrated increases of 3 or more points on the withdrawal subscale, and each of those three patients also deteriorated in thought disorder. None of the demographic or clinical variables including age, sex, neuroleptic dose, years of illness, age of onset of illness or baseline TD were predictive of relapse. However, the homogeneity of the sample limits the power to find any such predictors. Of the six patients who relapsed, 4 had baseline TD and 2 did not. Further, there was no interaction between change in TD over time and psychotic relapse (Table 2).
CONCLUSIONS
The results of this study should be interpreted with caution because of the small sample size and rather homogeneous patient population. This study showed overall worsening in dyskinetic movements within four weeks of abrupt neuroleptic withdrawal, especially in the distal extremities. This worsening occurred regardless of the presence of baseline TD and change in psychotic pathology. There are several discontinuation studies in which a subset of patients demonstrated worsening of dyskinesia, withdrawal emergent or covert TD. This perhaps confirms the notion that patients with and without overt TD are vulnerable to the effects of neuroleptic-induced supersensitivity and subject at least to withdrawal-emergent dyskinesias. The fact that there did not appear to be a relationship between relapse of psychosis and of
270 TABLE
2
Mean weekly dyskinesia
scores Maryland
Psychiafric
Baseline
Global (n = 1l)b Tongue Perioral Finger’ Patient subgroups global scores TD patients (n = 5) Non TD patients (n = 6) Relapsing pts (n = 5) Non-relapsing pts (n = 6)
Research
Center involuntary
Movement
Scale” (mean f SD)
Post-discontinuation Week I
Week 2
Week 3
Week 2
1.12 1.20 1.02 1.04
1.82+ 1.08 1.30+ 1.49 1.09* 1.38 1.55k1.04
1.91+ 1.04 1.20+ 1.55 0.45 & 0.93 2.09* 1.38
2.09* 1.38 l.lOk 1.60 0.55+0.82 2.73 + 1.68
2.36* 1.29 1.50* 1.51 0.64kO.92 2.4Sk 1.81
2.4OkO.55 0.50+0.55 1.80&0.84 l.OO+ 1.26
2.60k0.89 1.17kO.75 2.20+ 1.10 1.50+1.05
2.2OkO.84 1.67+ 1.21 1.00* 1.00 1.39* 1.17
2.80_+ 1.50+ 2.40+ 1.83*
3.00* 1.82* 1.80+ 2.00+
1.36& 0.90+ 0.64_+ 1.09+
1.30 1.22 1.14 1.60
1.00 1.22 1.10 1.41
“O-l = None-questionable, 2-3 = mild, 4-5 = moderate, 6-7 = severe. bRepeated measures analysis of variance, F= 3.00, df-(4.40), p< 0.05. ‘Repeated measures analysis of variance, F= 8.54, df-(4,40), p < 0.0001. TABLE
3
Mean weekly Brief Psychiatric
Rating Scale scores (mean + SD) Baseline
Total (n = 15) Thought Disorder Withdrawal Anxiety Activation Hostility Scores represent
31.93+8.88 6.73k4.08 5.33 k 2.74 7.93k3.17 4.67 k 1.23 4.60* 1.45 means
of total scores of indicated
Post-discontinuation Week I
Week 2
Week 3
Week 4
33.8Ok7.89 6.53 + 3.00 5.6Ok2.26 8.87 + 3.36 4.60&1.64 5.47i: 1.89
34.43 & 7.90 6.79& 1.97 5.71 k2.40 8.21 k4.06 5.29 k 1.86 4.93kl.86
33.93 k 7.72 7.29k3.69 4.93 & 2.27 8.5Ok2.66 5.07+ 1.29 5.21 k2.64
36.33 f 11.02 8.00+3.40 6.8Oi 3.91 7.07 + 2.66 5.4Ok2.74 6.13k3.80
BPRS dimension
dyskinesia could suggest a possible differential dopamine upregulation in the mesocortical/mesolimbic systems associated with psychosis and that in the nigrostriatal system associated with dyskinesia. The small sample size could also have prevented us from detecting such an interaction. Alternatively, these data might imply that dyskinesias have a fairly direct relationship to dopaminemediated neuronal transmission whereas the influence of the dopaminergic system on psychosis might be indirect, complex, or buffered by another transmitter system(s). Interpretation of these data have implications for the therapeutic mechanisms of antipsychotic drugs. Another important negative finding of this study was that there were no overall significant changes in psychopathology within the four weeks of the study when the group was assessed as a whole. This finding may reflect on the relative safety of
withdrawing patients from neuroleptics, either for evaluative purposes or for a pre-clozapine washout period. However, a substantial minority of quickly relapsing schizophrenic patients do exist, and their relapse was not subtle. This finding is very consistent with the reports of others (Perenyi et al., 1985, Zander et al., 1981). It suggests the possibility that there exists a subgroup of patients who are exquisitely sensitive to the neuroleptic medications with rapid relapse after withdrawal, and a subgroup of patients who are not, either demonstrating delayed relapse or no relapse. The risks and dangers of relapse in individuals must thus be carefully considered on a case by case basis. It would be useful clinically to know who is a risk for rapid relapse. These patients should be strongly considered for maintenance depot medication. Further work needs to be done to clarify which patients are at risk for relapse. More theoretically, the data are
271
consistent with the idea that schizophrenia in many cases is a relapsing, not a continuously symptomatic illness.
ACKNOWLEDGEMENTS
Supported by NIMH Grants and 5ROl MH42234-02.
4R37MH
37073-09
REFERENCES
Baldessarini RJ: Chemotherapy in Psychiatry Principles and Practice. Cambridge, MA, Harvard UP, 1985. Davidson M, Kahn RS, Powchik P, Warne P, Losonczy MF, Kaminsky R, Apter S, Jaff S, Davis KL (1991) Changes in plasma homovanillic acid concentrations in schizophrenic patients following neuroleptic discontinuation, Arch. Gen. Psychiatry 48, 73-76. Glazer WM, Bowers MB, Charney DS, Heninger GR (1989) The effect of neuroleptic discontinuation on psychopathology, involuntary movements, and biochemical measures in patients with persistent tardive dyskinesia. Biol. Psychiatry 26, 2244233.
Kirkpatrick B, Buchanan RW, Maeda K, Carpenter WT, Jauch D, Tamminga CA (1989) Effect of neuroleptic withdrawal on plasma prolactin: A possible marker of receptor adaptation. Biol. Psychiatry 26, 131-138. Klein DF, Davis JM (1969) Diagnosis and Drug Treatment of Psychiatric Disorders. Williams and Wilkins, Baltimore, MD, 1969. Marder SR, van Kammen DP, Docherty JP, Rayner J, Bunney WE (1979) Predicting drug-free improvement in schizophrenic psychosis. Arch. Gen. Psychiatry 36, 1080-1085. Overall JE, Gorham DR (1962) The Brief Psychiatric Rating Scale. Psychol. Rep. 10, 799-812. Perenyi A, Frecska E, Bagdy G, Revai K (1985) Changes in mental condition, hyperkinesias and biochemical parameters after withdrawal of chronic neuroleptic treatment. Acta Psychiatr Stand 72, 430-435. Rupniak MN, Jenner P, Marsden CD (1985) The effect of chronic neuroleptic administration on cerebral dopamine receptor function. Life Sci. 32, 2289-2311. Schooler NR, Kane JM (1982) Research diagnoses for tardive dyskinesia. Arch. Gen. Psychiatriy 39, 486-487. Smith RC, Tamminga CA, Haraszti J, Pandey G, Damis JM (1977) Effects of dopamine agonists on tardive dyskinesia. Am. J. Psychiatry 134, 763-768. Thaker GK, Wagman AM, Kirkpatrick B, Tamminga CA (1989) Alterations in sleep polygraphy after neuroleptic withdrawal: A putative supersensitive dopaminergic mechanism. Biol. Psychiatry 25, 75-86. Zander KJ, Fischer B, Zimmer R, Ackenheil M (1981) Longterm neuroleptic treatment of chronic schizophrenic patients: clinical and biochemical effects of withdrawal. Psychopharmacology 73,43347.
SCHRES
261 reserved
0920-9964/93/$06.00
003 19
Changes in psychopathology and dyskinesia after neuroleptic withdrawal in a double-blind design Lisa Dixona, Gunvant Thakerb, Robert Conleyb, David Rossb, Nicola Cascella’, Carol Tammingab “University of Maryland, Department of Psychiatry, Baltimore, MD 21201, USA, bMaryland Psychiatric Research Center, Baltimore, MD 21228, USA and ‘Pharm Italia Carlo Erba, Milano, Italy (Received
19 October
1993; revision received 2 February
1993; accepted
10 February
1993)
The goal of this study was to assess the time course of change in psychopathology and dyskinesia after neuroleptic withdrawal. Fifteen DSM-III schizophrenic patients were abruptly withdrawn in a doubleblind fashion from stable haloperidol treatment. Weekly ratings of dyskinesia and psychopathology were performed for 4 weeks post-withdrawal. There was an overall increase in dyskinesia ratings over the 4-week period (p < 0.05) beginning in week 2, with dyskinetic movements of the fingers showing the most significant increase (p
Dyskinesia;
Neuroleptic
withdrawal
INTRODUCTION
Neuroleptic drugs have a potent antipsychotic action in schizophrenia and an antidyskinetic effect in neuroleptic-induced dyskinesias. In schizophrenia, the onset of drug action is immediate, but its full effects are gradual (Klein and Davis, 1969), whereas in dyskinesias, the drug effect is immediate and full (Smith et al. 1977). Typically, treatment of schizophrenia with these agents is chronic, due to the persistent nature of psychotic symptoms. However, because treatment is continual, the actual nature of the underlying symptomatology Correspondence to: C. Tamminga, Maryland Psychiatric Research Center, P.O. Box 21247, Baltimore, MD 21228, USA.
is obscured. Withdrawal of neuroleptic medication and the time course of symptom reemergence could be informative about underlying symptoms and perhaps the associated pathophysiology. Neuroleptic withdrawal in animals is followed by a hyperdopaminergic state, associated with the exposed, upregulated dopamine receptors, as seen in behavioral and biochemical measures (Rupniak et al., 1985). In humans, the same phenomena may occur in some or all dopamine-related functions, where neuroleptic withdrawal leads to a rebound effect in the early drug-free period. This assumption has been supported in clinical studies showing prolactin over-reduction immediately postwithdrawal (Kirkpatrick et al., 1989) and changes in sleep parameters (Thaker et al., 1989). Whether this occurs in more complex behavior functions
268
like cognition and movement is a question of this study. Withdrawal of antipsychotic medication has been reported to produce a heterogeneous response in psychopathology. Rapid relapse rates vary from 25 to 60%. Perenyi et al. (1985) abruptly withdrew 14 chronic schizophrenic inpatients from neuroleptic therapy, and 8 relapsed significantly within 5 weeks. Glazer et al. (1989) withdrew 19 schizophrenic patients with tardive dyskinesia (TD) from neuroleptic medication over a 7- 10 day period and found that 5 exhibited ‘psychological relapse’ requiring reinstitution of medication. In contrast, Marder et al. (1979) withdrew 22 psychotic schizophrenic inpatients from neuroleptics and found that 8 patients actually improved during the 30-day drug-free interval. Efforts to predict relapse have led to the finding of an association of increased plasma homovanillic acid (HVA) with relapse. In the study of Glazer et al. (1989), early relapsing patients had lower baseline and a significantly greater increase in plasma HVA after discontinuation than nonrelapsing patients. Davidson et al. (1991) studied 23 schizophrenic inpatients after neuroleptic discontinuation and found that plasma HVA levels were higher in the 9 patients who relapsed compared with those who did not. However, this finding has not been consistently replicated, possibly due to the complex methodologic issues surrounding HVA assessment. Withdrawal dyskinesia and worsening of dyskinetic symptoms in patients with tardive dyskinesia (TD) have also been observed after neuroleptic withdrawal. Ten of the 19 TD patients studied by Glazer et al. (1989) developed withdrawal emergent TD. Patients developing withdrawal emergent TD were younger and had a lower exposure to neuroleptic medication than patients who did not develop withdrawal emergent TD. Seven of the patients (50%) in the study by Perenyi et al. (1985) developed withdrawal dyskinesia. Most of these studies have not employed a double blind design. The purpose of this study was to characterize quantitative and temporal changes in clinical symptoms and dyskinetic movements after blind, abrupt, neuroleptic withdrawal in a structured hospital setting in severely ill, but stable schizophrenic patients. Further, the study aimed to assess the relationship between psychopatho-
logic changes and changes in dyskinetic movements given their probable linkage to brain dopamine systems.
METHODS
Fifteen RDC-diagnosed DSM-III schizophrenic patients admitted to the Maryland Psychiatric Research Center inpatient unit from 1985 to 1989 were enrolled in the study after giving informed consent. Patients were seriously ill with persistent symptoms. However, they were clinically stable on the ward for at least 2 months at the time of the study. As soon as possible after admission to the ward, all patients were placed on a dose of haloperidol equivalent to the dose of neuroleptic they had received prior to admission using the conversion of Baldessarini equivalences (1985). Haloperidol doses ranged from 5 mg to 40 mg. Antiparkinsonian agents were used sparingly on an as needed basis. All patients were free of parenteral neuroleptic medication for at least six months prior to the study. The withdrawal study was planned to last for 8 weeks. In a modified double-blind, placebo controlled design, all patients were withdrawn from haloperidol abruptly during the first four weeks of the study. At the time of withdrawal, patients received placebo capsules which were identical in appearance to active haloperidol. Both patients and staff were aware that patients would begin to receive placebo sometime within the first four weeks of the study, but neither group was aware of exactly when this would occur. Patients had drug-free ratings for at least 4 weeks of the study. Psychopathology was assessed using the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962) which was performed weekly by two blind raters who then developed a consensus rating. Reliability was adequate for the thought disorder (0.86), anxiety (0.79) and hostility (0.78) subscales. It was marginal for the withdrawal subscale (0.55) and poor for activation (0.24). Dyskinesia ratings were assessed with the MPRC Involuntary Movement Scale. The MPRC scale is an 8-point scale, rating movements from absent to severe based on severity and persistence of movements. All dyskinesia evaluations were videotaped
269
and ratings of videotapes were performed by blind raters. All raters were trained in the use of the MPRC Involuntary Movement Scale, and the overall reliability at the center of this scale is 0.81. The same rater performed all dyskinesia ratings for a given patient. Data analysis was done using a repeated measures analysis of variance on BPRS total scores and dimensions including thought disorder, anxiety, activation, hostility and withdrawal as well as on global TD scores and and individual body area scores. Four patients had incomplete ratings of dyskinesia due to deterioration of videotapes and were thus excluded from repeated measures analysis. Polynomial contrasts were used to assess trends over time. Patients were given a diagnosis of TD based on the Schooler and Kane criteria (1982); TD subjects all had a score of 2 or more on the global dyskinesia scale at baseline indicative of at least mild TD.
RESULTS
Patients were relatively young, predominantly Caucasian males, with an age of onset of illness of 18.7 (4.1) years (Table 1). All had a diagnosis of schizophrenia, and 6 had a diagnosis of TD. A significant linear trend was found for global dyskinesia ratings from baseline to week 4 postneuroleptic withdrawal (F= 5.35, df = (1, lo), ~~0.05). Analysis of all patients showed an increase in weekly global dyskinesia rating over this time period from 1.36 to 2.36 (Table 2). Dyskinetic movements of the fingers showed the most significant increase. There was no interaction
TABLE
1
Demographic and Illness Schizophrenic Patients
Age Years of neuroleptic treatment Years of illness Age of onset of illness Sex
Characteristics
of
DSM-III- R
Mean _+SD
Range
28k5.2 8.4k4.4 9.3 t_ 4.2 18.7+4.1 9 male
IS-40 4-20 4-20 9-29 6 female
between the presence of baseline TD and change in TD over time (Table 2). In contrast to dyskinesia, repeated measures analysis of variance revealed no changes in psychopathology from baseline to week 4 post-withdrawal on any BPRS dimension or item (Table 3). However, assessment of change in thought disorder in individual patients suggested a bimodal distribution. Six patients (40%) had increases of 3-7 points indicative of relapse (mean increase= 5.0 f 1.55)). Eight patients (53%) were essentially unchanged with a change in thought disorder score of less than 2 (mean change= - 0.511.20)). One patient (7%) improved with a reduction in thought disorder of 5 points. Only 3 patients (20%) demonstrated increases of 3 or more points on the withdrawal subscale, and each of those three patients also deteriorated in thought disorder. None of the demographic or clinical variables including age, sex, neuroleptic dose, years of illness, age of onset of illness or baseline TD were predictive of relapse. However, the homogeneity of the sample limits the power to find any such predictors. Of the six patients who relapsed, 4 had baseline TD and 2 did not. Further, there was no interaction between change in TD over time and psychotic relapse (Table 2).
CONCLUSIONS
The results of this study should be interpreted with caution because of the small sample size and rather homogeneous patient population. This study showed overall worsening in dyskinetic movements within four weeks of abrupt neuroleptic withdrawal, especially in the distal extremities. This worsening occurred regardless of the presence of baseline TD and change in psychotic pathology. There are several discontinuation studies in which a subset of patients demonstrated worsening of dyskinesia, withdrawal emergent or covert TD. This perhaps confirms the notion that patients with and without overt TD are vulnerable to the effects of neuroleptic-induced supersensitivity and subject at least to withdrawal-emergent dyskinesias. The fact that there did not appear to be a relationship between relapse of psychosis and of
270 TABLE
2
Mean weekly dyskinesia
scores Maryland
Psychiafric
Baseline
Global (n = 1l)b Tongue Perioral Finger’ Patient subgroups global scores TD patients (n = 5) Non TD patients (n = 6) Relapsing pts (n = 5) Non-relapsing pts (n = 6)
Research
Center involuntary
Movement
Scale” (mean f SD)
Post-discontinuation Week I
Week 2
Week 3
Week 2
1.12 1.20 1.02 1.04
1.82+ 1.08 1.30+ 1.49 1.09* 1.38 1.55k1.04
1.91+ 1.04 1.20+ 1.55 0.45 & 0.93 2.09* 1.38
2.09* 1.38 l.lOk 1.60 0.55+0.82 2.73 + 1.68
2.36* 1.29 1.50* 1.51 0.64kO.92 2.4Sk 1.81
2.4OkO.55 0.50+0.55 1.80&0.84 l.OO+ 1.26
2.60k0.89 1.17kO.75 2.20+ 1.10 1.50+1.05
2.2OkO.84 1.67+ 1.21 1.00* 1.00 1.39* 1.17
2.80_+ 1.50+ 2.40+ 1.83*
3.00* 1.82* 1.80+ 2.00+
1.36& 0.90+ 0.64_+ 1.09+
1.30 1.22 1.14 1.60
1.00 1.22 1.10 1.41
“O-l = None-questionable, 2-3 = mild, 4-5 = moderate, 6-7 = severe. bRepeated measures analysis of variance, F= 3.00, df-(4.40), p< 0.05. ‘Repeated measures analysis of variance, F= 8.54, df-(4,40), p < 0.0001. TABLE
3
Mean weekly Brief Psychiatric
Rating Scale scores (mean + SD) Baseline
Total (n = 15) Thought Disorder Withdrawal Anxiety Activation Hostility Scores represent
31.93+8.88 6.73k4.08 5.33 k 2.74 7.93k3.17 4.67 k 1.23 4.60* 1.45 means
of total scores of indicated
Post-discontinuation Week I
Week 2
Week 3
Week 4
33.8Ok7.89 6.53 + 3.00 5.6Ok2.26 8.87 + 3.36 4.60&1.64 5.47i: 1.89
34.43 & 7.90 6.79& 1.97 5.71 k2.40 8.21 k4.06 5.29 k 1.86 4.93kl.86
33.93 k 7.72 7.29k3.69 4.93 & 2.27 8.5Ok2.66 5.07+ 1.29 5.21 k2.64
36.33 f 11.02 8.00+3.40 6.8Oi 3.91 7.07 + 2.66 5.4Ok2.74 6.13k3.80
BPRS dimension
dyskinesia could suggest a possible differential dopamine upregulation in the mesocortical/mesolimbic systems associated with psychosis and that in the nigrostriatal system associated with dyskinesia. The small sample size could also have prevented us from detecting such an interaction. Alternatively, these data might imply that dyskinesias have a fairly direct relationship to dopaminemediated neuronal transmission whereas the influence of the dopaminergic system on psychosis might be indirect, complex, or buffered by another transmitter system(s). Interpretation of these data have implications for the therapeutic mechanisms of antipsychotic drugs. Another important negative finding of this study was that there were no overall significant changes in psychopathology within the four weeks of the study when the group was assessed as a whole. This finding may reflect on the relative safety of
withdrawing patients from neuroleptics, either for evaluative purposes or for a pre-clozapine washout period. However, a substantial minority of quickly relapsing schizophrenic patients do exist, and their relapse was not subtle. This finding is very consistent with the reports of others (Perenyi et al., 1985, Zander et al., 1981). It suggests the possibility that there exists a subgroup of patients who are exquisitely sensitive to the neuroleptic medications with rapid relapse after withdrawal, and a subgroup of patients who are not, either demonstrating delayed relapse or no relapse. The risks and dangers of relapse in individuals must thus be carefully considered on a case by case basis. It would be useful clinically to know who is a risk for rapid relapse. These patients should be strongly considered for maintenance depot medication. Further work needs to be done to clarify which patients are at risk for relapse. More theoretically, the data are
271
consistent with the idea that schizophrenia in many cases is a relapsing, not a continuously symptomatic illness.
ACKNOWLEDGEMENTS
Supported by NIMH Grants and 5ROl MH42234-02.
4R37MH
37073-09
REFERENCES
Baldessarini RJ: Chemotherapy in Psychiatry Principles and Practice. Cambridge, MA, Harvard UP, 1985. Davidson M, Kahn RS, Powchik P, Warne P, Losonczy MF, Kaminsky R, Apter S, Jaff S, Davis KL (1991) Changes in plasma homovanillic acid concentrations in schizophrenic patients following neuroleptic discontinuation, Arch. Gen. Psychiatry 48, 73-76. Glazer WM, Bowers MB, Charney DS, Heninger GR (1989) The effect of neuroleptic discontinuation on psychopathology, involuntary movements, and biochemical measures in patients with persistent tardive dyskinesia. Biol. Psychiatry 26, 2244233.
Kirkpatrick B, Buchanan RW, Maeda K, Carpenter WT, Jauch D, Tamminga CA (1989) Effect of neuroleptic withdrawal on plasma prolactin: A possible marker of receptor adaptation. Biol. Psychiatry 26, 131-138. Klein DF, Davis JM (1969) Diagnosis and Drug Treatment of Psychiatric Disorders. Williams and Wilkins, Baltimore, MD, 1969. Marder SR, van Kammen DP, Docherty JP, Rayner J, Bunney WE (1979) Predicting drug-free improvement in schizophrenic psychosis. Arch. Gen. Psychiatry 36, 1080-1085. Overall JE, Gorham DR (1962) The Brief Psychiatric Rating Scale. Psychol. Rep. 10, 799-812. Perenyi A, Frecska E, Bagdy G, Revai K (1985) Changes in mental condition, hyperkinesias and biochemical parameters after withdrawal of chronic neuroleptic treatment. Acta Psychiatr Stand 72, 430-435. Rupniak MN, Jenner P, Marsden CD (1985) The effect of chronic neuroleptic administration on cerebral dopamine receptor function. Life Sci. 32, 2289-2311. Schooler NR, Kane JM (1982) Research diagnoses for tardive dyskinesia. Arch. Gen. Psychiatriy 39, 486-487. Smith RC, Tamminga CA, Haraszti J, Pandey G, Damis JM (1977) Effects of dopamine agonists on tardive dyskinesia. Am. J. Psychiatry 134, 763-768. Thaker GK, Wagman AM, Kirkpatrick B, Tamminga CA (1989) Alterations in sleep polygraphy after neuroleptic withdrawal: A putative supersensitive dopaminergic mechanism. Biol. Psychiatry 25, 75-86. Zander KJ, Fischer B, Zimmer R, Ackenheil M (1981) Longterm neuroleptic treatment of chronic schizophrenic patients: clinical and biochemical effects of withdrawal. Psychopharmacology 73,43347.