Prevalence of extrapyramidal syndromes in psychiatric inpatients and the relationship of clozapine treatment to tardive dyskinesia

Schizophrenia Research 42 (2000) 223–230 www.elsevier.com/locate/schres

Prevalence of extrapyramidal syndromes in psychiatric inpatients and the relationship of clozapine treatment to tardive dyskinesia Jiri Modestin *, Patrik Lukas Stephan, Thomas Erni, Tanja Umari Psychiatric University Hospital, Lenggstrasse 31, CH 8029 Zurich, Switzerland Received 29 July 1998; accepted 30 June 1999

Abstract In 200 inpatients on regular neuroleptics, point prevalence of extrapyramidal syndromes, including Parkinson syndrome, akathisia and tardive dyskinesia ( TD), was studied and found to be 20, 11 and 22%, respectively. A total of 46 patients have currently, and for a longer time, (average about 3 years, median over 1 year) been treated with clozapine, and 127 with typical neuroleptics (NLs). Comparing both groups, higher TD scores were found in the clozapine sample. Investigating the influence of a set of seven clinical variables on the TD score with the help of multiple regression analysis, the influence of the treatment modality disappeared, whereas the age proved to be the only significant variable. Studying the role of past clozapine therapy in patients currently on typical NLs and comparing 10 matched pairs of chronic patients with and without TD in whom a complete life-time cumulative dose of NLs was identified, a relationship between TD and length of current typical NL therapy and life-time typical NL dosage could be demonstrated. On the whole, long-term relatively extensive use of clozapine has not markedly reduced the prevalence of extrapyramidal syndromes in our psychiatric inpatient population. In particular, we failed to demonstrate a beneficial effect of clozapine on prevalence of TD. There are certainly patients who suffer from TD in spite of a long-term intensive clozapine treatment. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Age; Clozapine; Extrapyramidal syndromes; Prevalence; Tardive dyskinesia; Typical neuroleptics

1. Introduction The occurrence of extrapyramidal syndromes, in terms of Parkinson syndrome, akathisia and tardive dyskinesia ( TD), represents a serious complication of neuroleptic (NL) treatment. Therefore, a lower frequency of these extrapyramidal syndromes in patients treated with atypical NL clozapine is considered its major advantage. It is known that clozapine causes no complete Parkinson syn* Corresponding author. Tel.: +41-1-384-2111; fax: +41-1-383-4456

drome, and a low prevalence of akathisia in clozapine-treated patients has been reported (Chengappa et al., 1994). Clozapine should also exert favorable therapeutic effects not only on parkinsonism and chronic akathisia, but also on TD (Spivak et al., 1997). In the Psychiatric University Hospital of Zurich, clozapine has regularly and abundantly been used since the late 1960s. Therefore, we speculated that the occurrence of extrapyramidal syndromes could be lower in our hospital population, including both acute and chronic inpatients. We studied point prevalence of extrapyramidal syndromes in

0920-9964/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0 9 2 0 -9 9 6 4 ( 9 9 ) 0 0 13 3 - 4

224

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

all NL-treated patients hospitalized in August 1995 and correlated the occurrence of these syndromes with the kind of NL therapy the patients received. Special attention was paid to the role of clozapine versus typical NLs with respect to TD prevalence.

2. Methods All inpatients of the hospital who had received continuous NL treatment for at least 3 days were considered for the study. Out of 224 eligible patients, four (2%) were not available as they worked outside the hospital, and 20 (9%) refused to participate. Thus, 200 inpatients, namely 114 (57%) men of average age 45 years (s.d. 16) and 86 (43%) women of average age 53 years (s.d. 21), were included. First, in all these 200 patients, a systematic neurological examination was carried out by one of the authors (P.L.S.) trained for this task for over 20 h. The following instruments were used: 1. Section ‘Motor examination’ of Unified Parkinson’s Disease Rating Scale UPDRS ( Fahn et al., 1987) to assess the extent of Parkinson syndrome. 2. Barnes Akathisia Scale BAS (Barnes, 1989). 3. Abnormal Involuntary Movements Scale AIMS (Guy, 1976) for the rating of TD. Demographic and clinical data, including current treatment, were obtained from clinical charts. Second, the reasons for prescribing some patients clozapine were explored reviewing retrospectively the patients’ clinical charts to make sure that TD had not been the main indication. Having excluded patients treated with the clozapine/typical NLs combination, patients on clozapine were then compared with patients on typical NLs. The differences between both groups with regard to TD (AIMS total score) were analyzed, taking the variables age, sex, diagnosis (in terms of schizophrenia versus other diagnoses), current prescription of antiparkinsonian drugs, duration of current NL therapy (the length of continuous treatment with clozapine or typical NLs, and, in the case of typical NLs, disregarding whether one or more NLs were prescribed simultaneously or consecu-

tively) and illness duration into consideration. Univariate and multivariate statistical methods were used in this part of the study. Using the same procedure, the influence of the past clozapine treatment on TD was also investigated in the patients currently on typical NLs. Third, in a small group of patients, the influence of the life-time cumulative NL dose on TD was studied. A total of 10 patients were identified who suffered from pronounced TD (AIMS total score >6), fulfilled research criteria for TD (Schooler and Kane, 1982), gave their informed consent for study of their previous charts and in whom a suitable control patient without TD (AIMS total score=0) could be identified, who fulfilled matching criteria and gave his or her consent. The matching criteria included equality of sex, age (±4 years), diagnosis (schizophrenia versus other diagnoses) and illness duration (±4 years). These 20 patients (10 matched pairs) had a total of 140 hospitalizations in 16 different psychiatric hospitals. All their charts, which also included data on their outpatient treatment, were thoroughly scrutinized, and cumulative doses of NLs were calculated, transforming all NL doses given to the patients in equiv. CPZ according to ratios indicated by Clark and Del Guidice (1970) and Dietmaier and Laux (1992) and adding them up.

3. Results Of 200 patients included in the study, 46 (23%) patients were currently on clozapine, 27 (13.5%) were currently treated with both clozapine and typical NLs, and 127 (63.5%) patients received typical NLs. Note that 41 (32%) patients of the latter group received two, and three (2%) patients received three typical NLs simultaneously. The shortest current NL therapy (in two patients) lasted 5 days, whereas the longest therapy was over 39 years in duration at the time of the inquiry. All but one of 46 patients currently on clozapine had previously also been treated with typical NLs; in the group of 127 patients currently on typical NLs, a total of 49 (39%) patients had also been treated with clozapine in the past. The indication for clozapine treatment in 46 patients currently on

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

clozapine was: insufficient response or resistance with respect to typical NLs in 14 (30%) patients; pronounced negative symptoms in two (4%) patients; severe extrapyramidal reactions with typical NLs, namely acute dystonia, Parkinson syndrome and akathisia in two (4%), five (11%) and five (11%) patients, respectively; one patient asked for clozapine herself; and in 11 (24%) patients, none of them suffering from TD, no special reasons for clozapine prescription were given. In six (13%) patients, clozapine was prescribed because of TD (in one of them, the diagnosis of TD was questionable). 3.1. Point prevalence of extrapyramidal syndromes The investigated Parkinson syndrome UPDRS scores ranged from 0 to 29. A total of 32 (16%) patients were entirely symptom-free, whereas 40 (20%) patients fulfilled criteria delineated by Van Harten et al. (1996) for Parkinson syndrome. The UPDRS average total score in these 40 patients reached 9.5 (s.d. 6.1). Patients with Parkinson syndrome were older than symptom-free patients: mean 54 years (s.d. 18) versus 43 years (s.d. 18); (t=2.58, df=70, p=0.006). Also, they had a longer average illness duration of 22 years (s.d. 15) versus 14 years (s.d. 12); (t=2.53, df=70, p= 0.007). Clozapine was given to five (13%) patients with Parkinson syndrome, typical NLs to 31 (78%) patients, clozapine/typical NLs combination to four (10%) patients and antiparkinsonian drugs to eight (20%) patients. The corresponding figures for symptom-free patients are nine (28%), 21 (66%), two (6%) and 12 (38%) — none of the differences was statistically significant. The akathisia BAS global scores ranged from 0 to 3. No akathisia was found in 159 (79.5%) patients, pseudo-akathisia with no inner restlessness in 18 (9%) patients, and true akathisia in 23 (11.5%) patients. The BAS average global score was 1.7 (s.d. 0.7) in the latter group. No significant relationship was found between the occurrence of akathisia and the age and illness duration. Clozapine was given to six (26%) patients with true akathisia, typical NLs to 15 (65%) patients, clozapine/typical NLs combination to two (9%) patients and antiparkinsonian drugs to nine (39%)

225

patients. The corresponding figures for akathisiafree patients are 38 (24%), 100 (63%), 21 (13%) and 42 (26%) — none of the differences was statistically significant. Regarding TD, AIMS scores ranged from 0– 22; 125 (62.5%) patients scored 0, whereas in 28 (14%) patients, pronounced TD (AIMS total score >6) was identified. Of these 28 patients, 13 were on clozapine. Two of them had presented TD before clozapine was started, and in 11 patients, TD must have appeared in the course of clozapine treatment. As mentioned, in six patients, clozapine had been prescribed because of TD. In four of them, TD was not demonstrable any more. TD research criteria were fulfilled in 45 (22.5%) patients, whose average AIMS total score was 8.6 (s.d. 3.9). These patients were older than patients without TD: mean 54 years (s.d. 17) versus 47 years (s.d. 19); (t=2.18, df=168, p=0.015). Also, they had a longer average illness duration of 23 years (s.d. 15) versus 15 years (s.d. 15); (t= 3.11, df=168, p=0.001). Clozapine was given to 16 (36%) patients with TD, typical NLs to 25 (55%) patients, clozapine/typical NLs combination to four (9%) patients and antiparkinsonian drugs to 13 (29%) patients. The corresponding figures for patients without TD are 25 (20%), 85 (68%), 15 (12%) and 33 (26%) — none of the differences was statistically significant. Taking all three extrapyramidal syndromes together, at least one was present in 84 (42%) of 200 patients who were assessed. Eleven (24%) out of 45 patients with TD also suffered from Parkinson syndrome and the same number from akathisia. 3.2. Current clozapine versus typical NLs regarding TD Table 1 compares 46 patients currently on clozapine with 127 patients currently treated with typical NLs with regard to the basic clinical characteristics and the main extrapyramidal side-effects. Patients on clozapine were more frequently diagnosed as schizophrenia, received their current NL drug for a longer time and were ill for a longer time. They scored significantly higher on AIMS, whereas no differences were found with regard to Parkinson syndrome and akathisia including objec-

226

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

Table 1 Comparison of psychiatric inpatients on clozapine versus typical NLs with regard to basic clinical characteristics and main extrapyramidal side-effects (percentages in parentheses)

Age (years): mean±s.d. Sex: males Diagnosis: schizophrenic disorder Duration of current NL therapy (years): mean±s.d. Current NL daily dose (mg/CPZ mg equiv.): mean±s.d. Illness duration (years): mean±s.d. AIMS total score: mean±s.d. UPDRS total score: mean±s.d. BAS global score: mean±s.d.

Clozapine

Typical NLs

Significancea

n=46

n=127

x2

p

53.0±19.3 25(54) 31(67) 3.1±4.2 291±171 21.2±16.0 3.8±5.3 3.3±4.7 0.2±0.6

49.1±19.1 71 (56) 57(45) 2.6±5.8 586±938 14.9±13.6 2.0±3.5 4.2±4.3 0.3±0.6

0.24 0.00 5.97 6.52

>0.1 >0.1 0.015 0.011

5.12 5.13 1.99 0.65

0.024 0.024 >0.1 >0.1

a Chi-squared test or Kruskal–Wallis test with 1 df.

tive restlessness, subjective awareness of restlessness and distress related to restlessness. To find out whether the AIMS difference was actually due to different treatments, the data of the patients of both groups were further analyzed using non-parametric multivariate methods (Iman, 1982). In the multiple regression analysis, a significant relationship was demonstrated (F=4.20, df=7, p=0.0003) between AIMS total score as a dependent variable and the set of seven independent variables including current NL therapy and its duration, age, sex, diagnosis, current antiparkinsonian therapy, and illness duration. Univariate regression analysis showed that both groups differed significantly from each other as far as the AIMS total score was concerned (F=5.25, df=1,

p=0.023). However, a significant influence of the variables age (F=7.19, df=1, p=0.008), duration of current NL therapy (F=13.09, df=1, p= 0.0004) and illness duration (F=13.12, df=1, p= 0.0004) was also demonstrated. Analyzing the influence of individual variables on AIMS total score, partialling out all other variables in the multiple regression model, the only variable that remained significant was that of age ( Table 2). The difference between both groups (=difference between both treatments) was no longer significant. 3.3. Past clozapine therapy and TD The group of patients on current typical NLs was split into a subgroup of 49 patients who had

Table 2 Results of the non-parametric multiple regression analysis (df=1 for all variables) in two patient groups with regard to the AIMS total score as a dependent variable Variable

Group Age Sex Diagnosis Antiparkinsonian drugs Duration of current NL therapy Illness duration

All patients: 46 on clozapine, 127 on typical NLs

Patients on typical NLs: 49 on clozapine in the past, 78 never on clozapine

F-value

F-value

3.12 6.68 0.32 1.30 2.76 2.82 0.87

p 0.079 0.011 0.575 0.257 0.098 0.095 0.351

3.29 5.60 0.01 0.02 3.81 4.24 0.01

p 0.072 0.020 0.935 0.890 0.053 0.042 0.901

227

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230 Table 3 Comparison of 10 matched pairs of patients with and without TD, using a paired t-test with 9 df

Male sex Age (years): mean±s.d. Illness duration (years): ±s.d. Diagnosis of schizophrenia Clozapine cumulative dose (g): mean±s.d. Typical NLs cumulative dose (g equiv. CPZ ): mean±s.d. Antiparkinsonian drugs cumulative dose (g equiv. biperiden): mean±s.d.

Patients with TD

Patients without TD

Significance

n =10 1

n =10 2

t

p

7 43.6±13.0 18.5±14.7 7 918a±1134 1243±939 2.3±2.8

7 43.1±12.3 18.0±14.9 7 363b±598 590±697 2.7±3.0

1.64 2.06 0.33

0.068 0.035 >0.1

a Including one patient who never received clozapine. b Including three patients who never received clozapine.

received clozapine therapy in the past and a subgroup of 78 patients without past clozapine therapy. Again, a significant relationship was found (F=3.47, df=7, p=0.002) between total AIMS score as a dependent variable and the set of seven independent variables enumerated above in the multiple regression analysis. Analyzing both subgroups with the help of the same univariate regression model, a significant relationship emerged between both subgroups and the AIMS total score (F=4.79, df=1, p=0.030); furthermore, a significant influence of the age variable (F=4.90, df=1, p=0.029), duration of current NL therapy (F= 11.75, df=1, p=0.0008) and illness duration (F= 7.41, df=1, p=0.007) on the AIMS total score was also found. Taking the influence of all other variables on the relationship between one variable and AIMS total score into consideration, the subgroup influence and the influence of the variables diagnosis on AIMS total score disappeared, whereas the influence of age and duration of current NL therapy could be confirmed ( Table 2 on the right). Thus, very similar results were obtained in comparing the patients on current clozapine versus current typical NLs as well as the patients on current typical NLs with versus without past clozapine therapy. 3.4. Life-time cumulative NL dose and TD In Table 3, all 10 matched pairs of patients with and without TD are compared with regard to matching criteria and life-time cumulative dose of

NL and antiparkinsonian drugs. As can be seen, seven pairs were male, and seven pairs received the diagnosis of schizophrenia. In comparisons performed with the help of the paired t-test, patients with TD received a higher cumulative dose of typical NLs; however, evaluating the data with the Wilcoxon test for matched pairs, the difference was not statistically significant ( p= 0.08).

4. Discussion Altogether, 42% of our patients suffered from one of the extrapyramidal syndromes that we studied: 20% from Parkinson syndrome, 11% from akathisia and 22% from TD. The prevalence of NL-induced Parkinson syndrome was estimated to be 26% ( Hansen et al., 1992) and of NL-induced acute and chronic akathisia 20–30% and 30–40%, respectively (Sachdev, 1995a,b). An older survey indicated a TD-prevalence in NL-treated samples of 20% (s.d. 14), whereas the corresponding figure in untreated samples was 5% (s.d. 9) ( Kane and Smith, 1982). A comparable overall mean prevalence of TD among chronic NL-treated patients of 24% was indicated more recently (Jeste and Caligiuri, 1993). Compared with these figures, the occurrence of extrapyramidal syndromes in our patients appears to be about the same, perhaps with the exception of a slightly lower rate of akathisia. Nevertheless, Van Harten et al. (1996), who used the identical study instruments, also

228

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

indicated the prevalence of 9% for akathisia (36% for Parkinson syndrome and 40% for TD). Comparing patients currently on clozapine with the patients currently on typical NLs (Table 1), more TD (significantly higher AIMS total score) was found in the clozapine group. Generally, such a finding could be due to a selective prescription of clozapine in patients suffering from TD; clozapine has been reported to exert a favorable effect on TD in a proportion of patients (Lieberman et al., 1991; Moore et al., 1992; Gerlach and Peacock, 1994), and it has even been claimed to be the drug of choice for patients with TD (LarachWalters et al., 1997). In several more recent studies, in small groups of TD patients treated with clozapine over an average period of 4–10 months, the AIMS scores decreased by 41, 74 and 85%, respectively (Spivak et al., 1997; Bassitt and Louza Neto, 1998; Dalack et al., 1998). However, individual cases of clozapine-aggravated or clozapine-induced TD have been reported (Doepp and Buddeberg, 1975; de Leon et al., 1991; Kane et al., 1993; Dave´, 1994). Anticholinergic drugs may worsen TD (Greil et al., 1984), and dyskinesias upon abrupt clozapine withdrawal have been described (Ahmed et al., 1998). In any case, as the extensive review of the clinical charts revealed, no selective clozapine prescription in TD took place in our patients; 13% of those prescribed clozapine were clinically diagnosed as suffering from TD, which is practically identical with 14% of pronounced TD found in the entire sample studied systematically with regard to the prevalence of extrapyramidal syndromes. Our patients on clozapine received the diagnosis of a schizophrenic disorder more frequently and had a considerably longer illness duration. Therefore, they represent a more impaired population and have been exposed to typical NLs for a longer time. Illness severity and cumulative duration of exposure to typical NLs are considered risk factors for TD (Casey, 1998). Their current neuroleptic therapy was also of a slightly longer duration. Note, however, that the majority of patients in both groups were under current NL therapy for a considerable period of time: the average was almost 3 years, and the median still 13 months for the whole sample comprising both groups.

Thus, there were more differences between our patients on clozapine and on typical NLs, and the variables that we examined (such as diagnosis and illness duration) could have been intercorrelated. Therefore, both groups were compared using multivariate statistical methods to find out whether the AIMS difference could really be explained by the different NL treatments. Considering the influence of the whole set of the variables indicated in Table 2, the difference between both groups was indeed no more statistically significant with regard to the AIMS total score. In contrast, a statistically significant relationship between AIMS total score and age was found. The latter finding could be replicated in the analysis of the possible role of past clozapine treatment comparing patients currently on typical NLs who had or had not been treated with clozapine in the past. Again, considering the whole set of the variables, the group difference was no more statistically significant, whereas the influence of age was confirmed. Age has indeed been identified as a major risk factor for tardive diskinesia in other studies (Jeste and Caligiuri 1993; Van Os et al., 1997). In the comparison of both subgroups of patients on typical NLs (patients with and without past clozapine therapy), the duration of current NL therapy also proved to be important regarding the AIMS total score when considering the whole set of independent variables. Correspondingly, in the sample of 127 patients on current typical NLs, a significant positive correlation between AIMS total score and duration of current NL therapy was found (r=0.29, p=0.0009). Investigating 10 matched pairs of chronic patients with and without TD, those with TD received a higher average lifetime cumulative dose of typical NLs. Whether there is a direct relationship between the cumulative dose and the length of typical NL exposure and TD — as these results would indicate — remains uncertain; controversial results were reported concerning NL life-time dosage and length of treatment as TD risk factors ( Kane and Smith, 1982; Jeste and Caligiuri, 1993). Even though tardive dyskinesia was found more frequently in the patients currently (and for a longer time) treated with clozapine than with typical NLs, in the multivariate analysis of the data,

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

neither the influence of the current treatment in terms of clozapine versus typical NLs nor the influence of the past clozapine therapy versus no past clozapine therapy could be demonstrated. In this analysis, no beneficial effect of a larger-scale clozapine prescription on TD prevalence could be shown; age turned out to be the variable significantly correlated with the occurrence of TD. This finding does not preclude the possibility that there might be TD patients who profit from clozapine treatment; however, it is certain ( Table 3) that there are TD patients treated with clozapine long enough to reach a very high cumulative clozapine dose, who, in spite of this treatment, are presenting pronounced TD. Thus, it is possible that TD amelioration in the proportion of patients suffering from this condition and switched from typical NLs to clozapine is not due to the clozapine therapy as such but is due to a spontaneous remission, clozapine allowing the passage of time without an offending drug to result in a decrease in TD (Raja, 1996). Unfortunately, our patients were not assessed systematically at the beginning of their clozapine therapy with regard to the occurrence of TD. Allowing for this reservation, our data (along with the case reports quoted above) even suggest that in some patients, TD manifests in the course of clozapine treatment for the first time. It remains unclear whether this should be due to increasing age, clozapine, perhaps in interaction with some latent damage caused by previous typical NLs, or other factors.

References Ahmed, S., Chengappa, K.N., Naidu, V.R., Baker, R.W., Parepally, H., Schooler, N.R., 1998. Clozapine withdrawal-emergent dystonias and dyskinesias: a case series. J. Clin. Psychiatry 59, 472–477. Barnes, T.R.E., 1989. A rating scale for drug-induced akathisia. Br. J. Psychiatry 154, 672–676. Bassitt, D.P., Louza Neto, M.R., 1998. Clozapine efficacy in tardive dyskinesia in schizophrenic patients. Eur. Arch. Psychiatry Clin. Neurosci. 248, 209–211. Casey, D.E., 1998. Effects of clozapine therapy in schizophrenic individuals at risk for tardive dyskinesia. J. Clin. Psychiatry 59, Suppl. 3, 31–37. Chengappa, K.N., Shelton, M.D., Baker, R.W., Schooler, N.R., Baird, J., Delaney, J., 1994. The prevalence of akathisia in

229

patients receiving stable doses of clozapine. J. Clin. Psychiatry 55, 142–145. Clark, W.G., Del Guidice, J., 1970. Principles of Psychopharmacology. Academic Press, New York. Dalack, G.W., Becks, L., Meador-Woodruff, J.H., 1998. Tardive dyskinesia, clozapine, and treatment response. Prog. Neuropsychopharmacol. Biol. Psychiatry 22, 567–573. Dave´, M., 1994. Clozapine-related tardive dyskinesia. Biol. Psychiatry 35, 886–887. de Leon, J., Moral, L., Camunas, C., 1991. Clozapine and jaw dyskinesia: a case report. J. Clin. Psychiatry 52, 494–495. ¨ bersichtstabellen. In: Riederer, Dietmaier, O., Laux, G., 1992. U P., Laux, G., Po¨ldinger, W. ( Eds.), Neuro-Psychopharmaka, Bd. 4. Springer, Wien, pp. 197–213. Doepp, S., Buddeberg, C., 1975. Extrapyramidale Symptome unter Clozapin. Nervenarzt 46, 589–590. Fahn, S., Elton, R.L., members of the UPDRS Development Committee, 1987. Unified Parkinson’s Disease Rating Scale. In: Fahn, S., Marsden, C.D., Calne, D.B., Goldstein, M. ( Eds.), Recent Developments in Parkinson’s Disease Vol. 2. Macmillan Health Care Information, Florham Park, NJ, pp. 153–164. Gerlach, J., Peacock, L., 1994. Motor and mental side effects of clozapine. J. Clin. Psychiatry 55, 9, Suppl. B, 107–109. Greil, W., Haag, H., Rossnagl, G., Ru¨ther, E., 1984. Effect of anticholinergics on tardive dyskinesia. Br. J. Psychiatry 145, 304–310. Guy, W., 1976. ECDEU Assessment Manual for Psychopharmacology Revised. United States Department of Health, Education and Welfare, Washington, DC, pp. 534–537. Hansen, T.E., Brown, W.L., Weigel, R.M., Casey, D.E., 1992. Underrecognition of tardive dyskinesia and drug-induced parkinsonism by psychiatric residents. Gen. Hosp. Psychiatry 14, 340–344. Iman, R.L., 1982. In; Some Aspects of the Rank Transformation in Analysis of Variance Problems. Seventh SUGI Conf. Proc, 676–680. Jeste, D.V., Caligiuri, M.P., 1993. Tardive dyskinesia. Schizophr. Bull. 19, 303–315. Kane, J.M., Smith, J.M., 1982. Tardive dyskinesia: prevalence and risk factors 1959 to 1979. Arch. Gen. Psychiatry 39, 473–481. Kane, J.M., Woerner, M.G., Pollack, S., Safferman, A.Z., Lieberman, J.A., 1993. Does clozapine cause tardive dyskinesia? J. Clin. Psychiatry 54, 327–330. Larach-Walters, V., Tognolini, R.Z., Rueda, H.M., Rueda, R.M., Torres, R.G., 1997. New strategies for old problems: tardive dyskinesia ( TD). Review and report on severe TD cases treated with clozapine, with 12, 8 and 5 years of video follow-up. Schizophr. Res. 28, 231–246. Liebermann, J.A., Saltz, B.L., Johns, C.A., Pollack, S., Borenstein, M., Kane, J., 1991. The effects of clozapine on tardive dyskinesia. Br. J. Psychiatry 158, 503–510. Moore, N.C., LeWitt, P.A., Gershon, S., 1992. Reduction of dyskinesias and psychiatric disability with clozapine over different time courses. J. Psychopharmacol. 6, 514–518. Raja, M., 1996. The treatment of tardive dyskinesia. Schweiz. Arch. Neurol. Psychiat. 146, 13–18.

230

J. Modestin et al. / Schizophrenia Research 42 (2000) 223–230

Sachdev, P., 1995a. The epidemiology of drug-induced akathisia. Part I. Acute akathisia. Schizophr. Bull. 21, 431–449. Sachdev, P., 1995b. The epidemiology of drug-induced akathisia. Part II. Chronic, tardive, and withdrawal akathisias. Schizophr. Bull. 21, 451–461. Schooler, N.R., Kane, J.M., 1982. Research diagnosis for tardive dyskinesia. Arch. Gen. Psychiatry 39, 486–487. Spivak, B., Mester, R., Abesgaus, J., Wittenberg, N., Adlersberg, S., Gonen, N., Weizman, A., 1997. Clozapine treat-

ment for neuroleptic-induced tardive dyskinesia, parkinsonism, and chronic akathisia in schizophrenic patients. J. Clin. Psychiatry 58, 318–322. Van Harten, P.N., Matroos, G.E., Hoek, H.W., Kahn, R.S., 1996. The prevalence of tardive dystonia, tardive dyskinesia, parkinsonism and akathisia. The Curac¸ao Extrapyramidal Syndromes Study: I. Schizophr. Res. 19, 195–203. Van Os, J., Fahy, T., Jones, P., Harvey, I., Toone, B., Murrey, B., 1997. Tardive dyskinesia: who is at risk? Acta Psychiatr. Scand. 96, 206–216.