Numerous case reports of atypical antipsychotics inducing hypomanic/manic symptoms have been published; most concern the use of risperidone and olanzapine (Reference Aubry, Simon and BertschyAubry et al, 2000), but quetiapine (Reference BenazziBenazzi, 2001) and ziprasidone (Reference Lu, Lundgren and EscalonaLu et al, 2002) have also been implicated. A literature search using Medline and PubMed revealed no such reports associated with amisulpride. Although the manufacturer has accumulated a small number of reports of manic symptoms developing during amisulpride treatment, a recent internal review concluded that no causality could be established (Sanofi-Synthelabo, personal communication, 2002). I report a case of amisulpride-induced mania.

A 17-year-old female with a 4-year history of schizophrenia was commenced on amisulpride for persistent negative symptoms. It was cross-titrated with olanzapine, over a 4-week period, to 400 mg. She continued taking citalopram 20 mg, which had been started 6 months previously on the basis that her negative symptoms could be secondary to a masked depression. On commencement of amisulpride her negative symptoms, as rated on the Scale for the Assessment of Negative Symptoms (SANS; Reference AndreasenAndreasen, 1982), rapidly and linearly improved. Her mood, however, continued to rise and by 3 months she had developed a manic episode without psychotic features. She exhibited insomnia, hyperactivity, distractibility, disinhibition and an abnormally and persistently elevated mood that continued despite the immediate cessation of citalopram. There was no evidence of substance misuse or akathisia. These features improved after halving the amisulpride to 200 mg and re-introducing olanzapine 15 mg. They fully remitted within days of stopping the amisulpride. No other concomitant medication was used. The delay in development of overt manic symptoms may reflect having to overcome a baseline SANS score of 68.

The mechanism of action of mood changes induced by atypical antipsychotics is unknown, with speculation centring exclusively on a 5-HT_2a:D₂ economy. Lane et al (Reference Lane, Lin and Chang1998) argue that a higher ratio will increase frontal dopamine release, whereas others point to the combined blockade enhancing the ability of 5-HT_1a to release frontal dopamine (Reference Ichikawa, Ishii and BonaccorsoIchikawa et al, 2001). These theories do not explain the manicogenic effects of amisulpride, which has no serotonin affinity. I propose that the ability of low doses of amisulpride to differentially block presynaptic D₂ and D₃ autoreceptors enhances dopamine transmission in the frontal cortex and can lead to the development of manic symptoms in susceptible subjects. Presumably this mechanism contributes to its antidepressant efficacy, for which it is used in many countries. The theory implies induction of manic features at low doses only.