Efficacy of Low-Dose Mirtazapine in Neuroleptic-Induced Akathisia: A Double-Blind Randomized Placebo-Controlled Pilot Study

METHODS TOP

Twenty-six schizophrenic inpatients who met DSM-IV criteria for acute NIA and scored at least two points (mild akathisia) on the global subscale of the Barnes Akathisia Scale (BAS) 5 participated in the study. All patients were physically healthy and had normal values on routine laboratory tests. The study was approved by the Tirat Carmel Hospital Review Board. Written informed consent was obtained from all participants after full explanation of the protocol.

According to entries on a table of random numbers, the participants were allocated to receive for 5 days (at 8:00 a.m.) either mirtazapine at a dose of 15 mg (n = 13 [5 men and 8 women; age, mean ± SD: 32.5 ± 11.3 years; duration of illness, 5.0 ± 6.8 years]) or placebo (n = 13 [8 men and 5 women; age, 28.5 ± 9.2 years; duration of illness, 3.5 ± 4.2 years]). All patients were treated with conventional antipsychotic agents (haloperidol, 5-15 mg/day; perphenazine, 8-24 mg/day) for at least 2 weeks (range, 2-5) prior to enrollment in the study. Doses of all psychotropic medications remained unchanged for at least 3 days prior to baseline ratings and during the entire study period.

The severity of NIA was rated by one of the authors (S.E.), using the BAS, at baseline and on days 3 and 5. Clinically relevant response was defined as a reduction of at least two points on the BAS global subscale. 6 In addition, the Positive and Negative Symptom Scale (PANSS), 7 Hamilton Rating Scale for Depression (HAM-D), 8 and Simpson-Angus Scale (SAS) 9 for parkinsonism were completed at baseline and on day 5.

Analysis of covariance with repeated measurements (ANCOVA-RM), with the baseline values as covariates, was performed to detect between-group differences for the BAS subscale scores during the study period. These analyses were performed separately for completers (n = 10 in each group) and for all participants (intent-to-treat analysis; n = 13 in each group). Between-group differences in clinical variables and baseline rating scale scores, as well as in changes from baseline in PANSS, HAM-D, and SAS scores, were analyzed with Student's t-test. All tests were two-tailed.

RESULTS TOP

Three patients in each group discontinued treatment for persistent NIA before the second assessment (day 3) because of intolerance. At baseline the mirtazapine group scored higher than the placebo group on the BAS subjective subscale (2.3 ± 0.5 vs. 1.8 ± 0.4; t = 2.47, df = 18, p = 0.024;Table 1). Higher scores, although not statistically significant, were also observed in the BAS objective subscale (1.9 ± 0.6 vs. 1.4 ± 0.5; t = 2.02, df = 18, p = 0.06) and global subscale (3.0 ± 0.7 vs. 2.5 ± 0.5; t = 1.84, df = 18; p = 0.08).

TABLE 1. Clinical characteristics and rating scale scores in mirtazapine and placebo groups with neuroleptic-induced akathisiaANCOVA = RM, Analysis of Variance with Repeated Measures (group × time); BAS, Barnes Akathisia Scale; HAM-D, Hamilton Rating Scale for Depression; PANSS, Positive and Negative Syndrome Scale; SAS, Simpson-Angus Scale for parkinsonism.

The two groups were similar in other rating scale scores and clinical characteristics (Table 1). ANCOVA-RM revealed significant group x time effects in favor of the mirtazapine group among completers and in the intent-to-treat analysis for the BAS global subscale (F [1, 17] = 14.87, p = 0.001, and F [1, 23] = 13.24, p = 0.01, respectively) and objective subscale (F [1, 17] = 8.25, p = 0.011, and F [1, 23] = 7.35, p = 0.012, respectively). Borderline significant superiority was observed for mirtazapine over placebo among completers and in intent-to-treat analyses for the BAS subjective subscale (F [1, 17] = 4.39, p = 0.051, and F [1, 23] = 4.12, p = 0.054, respectively) and distress subscale (F [1, 17] = 4.21, p = 0.056, and F [1, 23] = 3.80, p = 0.064, respectively).

Applying the response criterion (reduction of at least two points on the BAS global subscale) to all participants yields a response rate of 53.8% (7/13) in the mirtazapine group and only 7.7% (1/13) in the placebo group (t = 8.3, df = 1, p = 0.004). A complete disappearance of NIA (BAS global score = 0) occurred in 5 of 13 patients (30.7%) in the mirtazapine group and none (0 of 13) in the placebo group. Within the mirtazapine group no difference was found in treatment response between men and women when response criteria were applied (three of five vs. four of eight, respectively; t = 0.12, df = 1, p = 0.72). There was a modest but significantly greater reduction in the mirtazapine group than in the placebo group on the SAS (-4.8 ± 2.5 and -2.5 ± 2.3, respectively; t = 4.6, df = 18, p = 0.046) and PANSS scores (-9.7 ± 13.6 vs. 0.1 ± 5.9, respectively; t = 4.4, df = 18, p = 0.05) but not HAM-D scores (-4.3 ± 7.9 vs. -0.3 ± 4.3, respectively; t = 2.0, df = 18, p = 0.17;Table 1).

The only side effect was mild sedation, evident in five patients in the mirtazapine group and one in the placebo group.

DISCUSSION TOP

In this study the addition of low-dose mirtazapine (15 mg/day) to ongoing antipsychotic treatment exerted a robust anti-NIA effect in comparison with placebo, even though on the primary outcome measure of BAS subscales the mirtazapine group had significantly higher (subjective) and nearly significantly higher (objective and global) baseline scores. The response rate of 53.8% is higher than previously reported with mianserin (40%) 6 and propranolol (37.5%). 10 However, in order to assess the relative efficacy of mirtazapine versus the other anti-NIA agents, a large-scale head-to-head comparative study is warranted and is currently being conducted by our group.

The analysis of the secondary outcome measures revealed that in contrast to the lack of antiparkinsonian effect of low-dose mianserin, 6 low-dose mirtazapine treatment seems to be associated with a subtle reduction in neuroleptic-induced parkinsonian symptoms. Mirtazapine was also found effective in the treatment of essential and parkinsonian tremor. 10 In the present study, the modest decrease in the global PANSS scores that accompanied the anti-NIA effect of mirtazapine may reflect an association or overlap between the severity scores of psychotic symptoms and NIA.

Mirtazapine was well-tolerated, and the only observed adverse effect was mild sedation. Though weight gain is another side effect associated with long-term mirtazapine treatment, weight monitoring was not conducted in the current study because it is unlikely that it would be clinically relevant during a 5-day trial.

The positive anti-NIA effect of mirtazapine substantiates recently reported beneficial anti-NIA effects of mianserin 6 and cyproheptadine, 2 supporting the possible role of 5-HT2A receptor inhibition in the anti-NIA effect. 1 Mirtazapine also exerts antihistaminic and α₂-antagonistic activity, but it is unlikely that these properties account for its anti-NIA effect, since histaminic and α₂-receptor antagonists seem to be ineffective in the treatment of NIA. 1 Although atypical antipsychotics are associated with a low propensity to induce extrapyramidal side effects, they are not completely devoid of proakathisic effect, a phenomenon that may lead to poor treatment response and nonadherence to treatment. The role of 5-HT2 antagonists in the treatment of akathisia induced by atypical antipsychotic agents merits further investigation.

ACKNOWLEDGMENT TOP

The authors thank Rena Kurs of Sha'ar Menashe Mental Health Center for editorial assistance.

REFERENCES TOP