ARIPIPRAZOLE HAS ADVANTAGES compared to other antipsychotics with regard to side-effect profile, including metabolic and cardiac side-effects.1 While these benefits have been reported from Western and US country, the data are still limited in Asia. Given that those metabolic side-effects are expected to vary with ethnic background,2 this lack of data limits the extrapolation of those findings to Asian patients. In this 1-year follow-up study we therefore examined metabolic, hormonal, and cardiac effects of switching to aripiprazole in Japanese patients with schizophrenia.
The purpose of the present study was to evaluate changes in metabolic parameters after switching to aripiprazole in Japanese population. In this 1-year observation study, following a switch to aripiprazole, 32 patients with schizophrenia were observed and assessment was done of bodyweight, total cholesterol, triglyceride, serum prolactin level, and corrected QT (QTc) interval. Significant reductions were observed in these parameters other than QTc interval. Given known detrimental metabolic and hormonal effects of some atypical antipsychotics, a switch to aripiprazole may warrant serious consideration also in Asian patients who suffer those side-effects.
This 1-year open-label study was conducted at two psychiatric hospitals (Oizumi Hospital and Aoyagi Hospital) and one clinic (Asakadai Mental Clinic) in the suburbs of Tokyo, Japan. Details of study procedures have been described elsewhere.3 Briefly, consecutive Japanese inpatients and outpatients aged ≥18 years with schizophrenia (DSM-IV)4 who were treated with a stabilized dose of oral antipsychotics for ≥1 month prior to study entry, were assessed for inclusion in the study. Aripiprazole was initiated at 12 mg/day, maintained at the same dose for the first 2 weeks, and then titrated between 12 and 30 mg/day for the rest of the study period. A total dose of previous antipsychotics was reduced biweekly by 25% and discontinued in 6 weeks while concomitant medications were kept constant throughout the study period.5 Bodyweight, total cholesterol, triglyceride, serum prolactin level, and corrected QT (QTc) interval on a 12-lead electrocardiogram (ECG) were also evaluated at baseline and weeks 14 and 52.
Differences in variables of interest among baseline and 14- and 52-week assessment points were tested, using a repeated measures one-way analysis of variance (anova) or χ2 test, using SPSS version 16.0 (SPSS, Chicago, IL, USA). Two-tailed P < 0.05 was considered statistically significant.
This study was approved by the institutional review board at each site, and all participants provided their written informed consent prior to entering the study.
Forty patients were enrolled; of these, 32 patients completed all scheduled assessments. Eighteen patients (56.3%) were male, and mean ± SD age, duration of illness, and final dose of aripiprazole were 54.6 ± 16.9 years, 27.0 ± 18.4 years, and 18.4 ± 4.6 mg/day, respectively. A total of 14, 11, one, and six patients took olanzapine, risperidone, both of them, and other antipsychotics, respectively. Eight patients (25.0%) experienced a clinical worsening defined by a ≥1-point increase in Clinical Global Impression severity of illness within 1 year. Repeated measures one-way anova found significant difference over time in bodyweight (F(1.4,43.7) = 13.3, P < 0.001), total cholesterol (F(2,62) = 3.6, P < 0.05), triglyceride (F(2,62) = 4.2, P < 0.05), and serum prolactin level (F(1.0,32.4) = 52.5, P < 0.0001), but not in QTc interval (Table 1). Bodyweight and serum prolactin demonstrated a quick drop by 14 weeks followed by a slower decrease whereas a gradual, steady decline over 1 year was found for total cholesterol and triglyceride levels.
|Baseline||14 weeks||52 weeks||Statistics|
|Bodyweight (kg)||63.1 ± 2.6||60.2 ± 2.5||59.0 ± 2.7||P < 0.001, F(1.4,43.7) = 13.3†|
|Total cholesterol (mg/dL)||194.3 ± 6.4||191.1 ± 6.3||181.4 ± 6.0||P = 0.032, F(2,62) = 3.6†|
|Triglyceride (mg/dL)||151.8 ± 17.0||131.1 ± 16.8||106.8 ± 10.1||P = 0.026, F(2,62) = 4.2†|
|Serum prolactin level (ng/mL)||38.9 ± 5.9||7.2 ± 3.7||5.8 ± 3.0||P < 0.001, F(1.0,32.4) = 52.5†|
|QTc interval (s)||0.416 ± 0.007||0.407 ± 0.004||0.414 ± 0.005||n.s.|
The rate of people who presented with an excessive increase in serum prolactin level (>13.7 ng/mL for men or >29.3 ng/mL for women) was significantly decreased from 78.1% at the baseline to 3.1% at week 52 (χ2(1) = 37.3, P < 0.001). The rate of people who demonstrated an excessive increase in total cholesterol (≥220 mg/dL) and triglyceride (≥150 mg/dL) was also decreased from 21.9% to 9.4% and from 34.4% to 18.8%, respectively, but these changes did not reach statistical significance (χ2(1) = 1.90, P > 0.05; χ2(1) = 2.00, P > 0.05, respectively). Changes in triglyceride and prolactin level were greater in women (−81.8 ± 105.0 mg/dL vs −16.4 ± 59.7 mg/dL, t(30) = 2.2, P < 0.05; −44.2 ± 32.3 ng/mL vs −24.5 ± 14.0 ng/mL, t(16.8) = 2.1, P < 0.05, respectively).
This study replicated previously reported benefits of switching to aripiprazole in terms of metabolic side-effects in a Japanese sample group. Although decreases in bodyweight and total cholesterol 1 year after the switch are consistent with previous retrospective chart review (n = 24) that assessed changes for 6 months,6 the present study also showed improvements in serum prolactin levels and triglyceride. Bodyweight and serum prolactin level showed a decrease at 14 weeks and subsequent stabilization, while total cholesterol and triglyceride gradually and steadily declined over 1 year. This is in contrast to the findings from another switching study, using ziprasidone, in which bodyweight reduced continuously over 52 weeks after switching from risperidone (n = 43) or olanzapine (n = 71), while total cholesterol and triglyceride rapidly decreased for the first 6 weeks and then stabilized.7 These discrepancies may be due to different mechanisms underlying metabolic effects between aripiprazole and ziprasidone.
The present findings should be interpreted in light of the small sample size. In addition, other relevant metabolic parameters, including serum insulin level and HbA1c, were not measured, and QTc intervals were not assessed at a scheduled time. Moreover, although no patient reported any change in their lifestyle during the study period, participant lifestyle, which could affect metabolic parameters, was not evaluated. In addition, gender differences in reference ranges of serum prolactin level and QTc interval should also be noted. Still, the results confirmed the benefits of switching to aripiprazole also in a Japanese population. Given the favorable impacts observed in the present study, a switch to aripiprazole may be seriously considered in patients who suffer metabolic and hormonal side-effects.5
The authors thank Drs K. Ishii, S. Katayama, Y. Imasaka, and Y. Goto for their valuable comments.