Effect of Clozapine on Caudate Nucleus Volume in Relation to Symptoms of Schizophrenia

According to magnetic resonance imaging (MRI), an increase in the caudate volume of patients with schizophrenia was found after treatment with typical antipsychotics (1, 2), but the clinical relevance of this increase remains unclear. Chakos et al. (2) found no correlation between caudate volume increase and symptom improvement during treatment with typical antipsychotics. However, larger caudate volumes after typical antipsychotic treatment were associated with poorer neuropsychological test performance (1), the deficit syndrome (3), and greater severity of symptoms (4), suggesting that medication-induced hypertrophy of the caudate nucleus may moderate the clinical effects of antipsychotic treatment (4). When patients switched from treatment with typical antipsychotics to treatment with the atypical antipsychotic clozapine, a decrease in caudate volume was found (5, 6).

This study investigated the effects of switching from treatment with typical antipsychotics to clozapine on caudate volume and clinical symptoms after 52 weeks of clozapine treatment in a group of patients with schizophrenia who were not responding to treatment with typical antipsychotics. The current study is an extension of our earlier study (6), which examined patients after treatment with clozapine for 24 weeks. We found a significant decrease in caudate volume in that study, but this decrease was not significantly correlated with symptom improvement. Since the optimal clinical efficacy of clozapine may not be reached until after 6 months of treatment (7), we hypothesized that caudate volume change would be related to change in symptoms after a prolonged period of clozapine treatment (52 weeks).

Method

Twenty-eight patients with schizophrenia (mean age=35 years, SD=10) participated in the study. All patients were treated at the University Medical Center in Utrecht. Patients were evaluated with the Comprehensive Assessment of Symptoms and History interview (8), and all met DSM-IV criteria for schizophrenia. Patients had been treated with typical antipsychotics for a minimum of 4 weeks (mean duration of treatment was 103 months, SD=101, and mean lifetime dose of haloperidol equivalent was 10.4 mg/day, SD=7.1). All had failed to show adequate responses according to criteria described elsewhere (6). Six patients did not complete the study because they refused further participation (N=1) or needed other psychotropic medication (N=5). After a complete description of the study, written informed consent was obtained from all subjects.

Patients were withdrawn from typical antipsychotics and then treated with clozapine for 52 weeks (mean dose of clozapine=346 mg/day, SD=61). MRI was used to assess neuroanatomical variables before the typical antipsychotics were withdrawn, after 24 weeks of clozapine treatment, and after 52 weeks of clozapine treatment. Psychopathology was rated just before clozapine treatment and once a month thereafter by using the Positive and Negative Syndrome Scale (9). A 20% or greater reduction in total Positive and Negative Syndrome Scale score from baseline to week 52 was defined a priori as treatment response. Additional information about assessment and MRI procedures is available in our previous report (6).

Analysis of caudate volume was performed by using repeated measures analysis of variance with the factors time (during typical antipsychotic treatment, after 24 weeks of clozapine treatment, and after 52 weeks of clozapine treatment), side (right and left caudate nucleus), and group (patients who responded to clozapine [N=16] and patients who did not respond to clozapine [N=6]).

Paired t tests were performed to compare change in caudate volume or Positive and Negative Syndrome Scale score at week 24 of clozapine treatment compared with baseline and at week 52 compared with week 24 of clozapine treatment.

Regression analysis was used to examine the relationship between caudate volume change following 24 weeks or 52 weeks of clozapine treatment and degree of psychiatric symptom improvement after 52 weeks of clozapine treatment.

Results

After 52 weeks of clozapine treatment, patients who responded to clozapine showed a significant difference from those who did not respond to clozapine in caudate volume change between the left and right caudate nucleus (F=3.4, df=2, 19, p<0.05). Clozapine treatment significantly reduced left (F=3.9, df=2, 20, p<0.05) but not right (F=2.4, df=2, 20, p=0.11) caudate volume over time. The change in left caudate volume was significant in responders (F=7.8, df=2, 14, p<0.005) but not in nonresponders (F=0.65, df=2, 4, p=0.55). The change in left caudate volume was significant after 24 weeks of clozapine treatment compared with baseline (t=3.1, df=1, 26, p<0.005) but not at week 52 compared with week 24 (t=–0.2, df=1, 21, p=0.85).

Clozapine treatment significantly improved symptoms over time (all symptoms: F=25.2, df=2, 24, p<0.001; positive symptoms: F=22.7, df=2, 24, p<0.001; negative symptoms: F=8.7, df=2, 24, p<0.005; general symptoms: F=15.7, df=2, 24, p<0.001). Post hoc analysis showed a significant improvement in Positive and Negative Syndrome Scale score after 24 weeks of clozapine treatment compared with baseline (t=4.4, df=1, 26, p<0.001) and at week 52 compared with week 24 (t=2.9, df=1, 22, p<0.01).

Across all subjects, change in left caudate volume was significantly related to symptom improvement after 52 weeks of clozapine treatment (F=8.0, df=1, 20, p<0.05) (Figure 1). A significant relationship between left caudate volume change and a reduction in positive symptoms (F=7.1, df=1, 20, p<0.05) and general symptoms (F=8.4, df=1, 20, p<0.01) but not in negative symptoms (F=2.0, df=1, 20, p=0.17) was found.

No relationship was found between caudate volume change after 24 weeks and symptom improvement after 52 weeks of clozapine treatment (F=2.13, df=1, 23, p=0.16).

Discussion

This study revealed that 52 weeks of clozapine treatment significantly reduced left caudate volume in patients with schizophrenia who had been treated previously with typical antipsychotics. This reduction was limited to responders to clozapine treatment. Moreover, the degree of left caudate volume decrease was significantly correlated with the extent of improvement in positive and general symptoms but not in negative symptoms across all subjects. The decrease in left caudate volume occurred at 24 weeks, but symptom improvement continued up to 52 weeks of treatment.

Our data are consistent with those of previous studies reporting a decrease in caudate volume after clozapine treatment (5) and symptom improvement beyond 24 weeks of clozapine treatment (7). The fact that we found no significant relationship between caudate volume reduction after 24 weeks and symptom improvement after 52 weeks suggests that, beyond 24 weeks of clozapine treatment, symptom improvement may reflect a further (although not significant) adaptation of the brain.

Treatment with typical antipsychotics leads to an increase in caudate volume, particularly on the left side (2, 3). This volume increase may be an ineffective adaptation, as was suggested by Gur et al. (4), since large caudate volumes were associated with greater severity of symptoms (1, 3, 4). In contrast, clozapine lowers caudate volume in patients previously treated with typical antipsychotics (5). It may be postulated that this is a restoring effect, “correcting” the unwanted increase after treatment with typical antipsychotics, since caudate volume decrease is associated with symptom improvement.

It should be noted that the present study included only patients who had not responded to typical antipsychotics. Therefore, results may not be generalizable to all patients with schizophrenia.

To conclude, this study found that switching from treatment with typical antipsychotics to clozapine resulted in a significant decrease in the left caudate volume in patients with refractory schizophrenia who responded to clozapine. This reduction was related to improvement in positive and general symptoms when clozapine treatment was extended over 52 weeks. These results suggest that the caudate nucleus may play a role in the positive and general symptoms of schizophrenia.

Received Aug. 7, 2000; revision received Nov. 17, 2000; accepted Nov. 27, 2000. From the University Medical Center, Department of Psychiatry, Utrecht. Address reprint requests to Dr. Scheepers, University Medical Center, Department of Psychiatry (A00.241), P.O. Box 85500, 3508 GA Utrecht, Netherlands; [email protected] (e-mail). Supported in part by Novartis Pharma B.V. medical research department. The authors thank Herman J. Wynne, Ph.D., for statistical advice.