14 June 2003

Dear Editor,


We report a very serious and unexpected side effect with the off-label use (i.e. the use of medications outside the terms of Australian approved product information) of a selective serotonin reuptake inhibitor (SSRI), fluoxetine, for the treatment of depression in an adolescent male.

A 16-year-old boy presented with depression and suicidal ideation. He was assessed and was found to have a major depressive disorder, he was initially managed with individual psychotherapy and parental support. After 8 weeks there was no improvement and he continued to deteriorate in his school and social functioning and so was commenced on fluoxetine 20 mg mane. There were no psychotic symptoms. After 3 days on fluoxetine, he presented acutely with auditory hallucinations telling him to kill his mother, father, sister and himself. He recognized these voices as ego-dystonic and was extremely distressed by them. His medication was ceased and he required a short course of clonazepam to sedate him for his marked agitation. Three days later his hallucinations ceased. There have been no recurrences. After 6 months of individual psychotherapy, he has made a good recovery from his depressive episode.

The use of antidepressants in children has increased1. Much of this use in children and adolescents is off label. With this increasing off-label usage, clinicians need to be vigilant for unusual adverse events; this is because it is unlikely that rare side effects, as reported in this letter, would be identified in a clinical trial. With regards to SSRI, there have been reports of mania and agitation, but an extensive literature search has failed to demonstrate previous reports of isolated auditory hallucinations in adolescents. There have, however, been reports of hallucinations with use of fluoxetine in adults. This present case demonstrated a good temporal relationship between the use of the drug and the onset of symptoms; with cessation of the drug there was cessation of his symptoms. Although improvement on de-challenge is an important indicator of causality, it should be noted that without rechallenge it cannot be excluded that the hallucinations occurred independently of the start of fluoxetine. Rechallenge was not possible in this case as the patient was very distressed by his symptoms and was very reluctant to restart fluoxetine or another related drug.

The data on which this increased use of antidepressants in children have been based on have largely been extrapolated from work in adults. There are few randomized controlled trials exploring the efficacy of SSRI in children and adolescents with depression. In only two randomised-controlled trials using fluoxetine for paediatric depression was a response above that of placebo seen2,3. Given the limited efficacy data and the potential serious side effects, we would currently advise caution in the use of SSRI for off-label use in children and adolescents.