De novo treatment-resistant psychosis following thioridazine withdrawal

To the Editor

Thioridazine was developed in the late 1950s and was in fact the most widely used antipsychotic in the UK prior to the year 2000 (Bisset, 2002). It was also often used in patients with intellectual impairment (Davies et al., 2002) to manage impulsive behaviour. In December 2000, the Committee on Safety of Medicines in the UK issued a controversial directive restricting the use of thioridazine after the emergence of evidence of QTc prolongation associated with its use (Bisset, 2002). In this report we describe a presentation of thioridazine withdrawal treatment-resistant psychosis in an intellectually impaired patient without a history of psychosis.

Mr P is a 28-year-old man who has a history of moderate intellectual impairment. At the age of 12 years he began exhibiting violent, aggressive behaviour at school and was subsequently commenced on thioridazine, maintained at a dose of 175 mg daily (typical dose range 200–800 mg). At the time there had been no evidence of overt psychotic symptoms. Thioridazine was markedly effective at curbing his aggression for 15 years without evidence of psychosis. Up until the age of 27 years, he was able to work in part-time labour and was living at home with his mother. In 2008, when supplies of thioridazine had been exhausted, he was switched over to risperidone as an alternative medication. Within 2 weeks a precipitous decline in his mental state was noted. He once again began demonstrating violent behaviour, but, in addition to this, for the first time he experienced persecutory delusions and distressing auditory hallucinations. From this time he was effectively hospitalised for a 2½-year period, spending most of his time in a closed psychiatric ward. He was trialled on a variety of medications including: quetiapine, amisulpride, olanzapine, zuclopenthixol, flupenthixol, pericyazine and carbamazepine. He was unresponsive to all of these medications. Owing to the marked refractory nature of his illness, clozapine was commenced and titrated to 300 mg daily. Remarkably, he responded completely to clozapine within weeks and has now been in complete remission for the last 18 months. He is now at his baseline level of functioning – equivalent to that which was achieved when he was on thioridazine.

The possibility of thioridazine withdrawal psychosis is verified by our review of the existing literature. In Finland, a nationwide registry study showed that hospitalisation rates for over 2000 schizophrenia patients who were on thioridazine prior to its withdrawal in 2005 doubled compared with the preceding years of 2000–2004 (Purhonen et al., 2012). There is a case description of recurrent thioridazine withdrawal psychosis where on three separate occasions when thioridazine was withdrawn carefully (between 1969 and 1980) the same patient experienced psychotic symptoms within days, requiring hospitalisation on one occasion (Miller, 2009).

A study in an English learning disability team identified 18 patients on thioridazine out of a total population of 155 learning disability patients (Davies et al., 2002). Of these 18 patients, only three of them had a diagnosis of schizophrenia. Following the Committee on Safety of Medicine’s advice in 2000, all 18 patients stopped thioridazine. Seven of these patients (39%) experienced moderate or severe difficulties in the following 3 months. In four patients, despite substitution with alternate antipsychotic medication, significant problems still occurred. Of particular note, three patients (17%) developed de novo psychosis having never experienced definitive psychotic symptoms in the past. This latter group is of particular interest given the presentation of our patient. This is the only case series that describes potential de novo psychosis upon withdrawal of long-term antipsychotic medication in an intellectually impaired population. However, the exact clinical details are not provided in this report but would certainly be of significant interest given the importance of this concerning new finding.

There are only three other cases of short-term de novo psychosis following withdrawal of antipsychotic agents (reserpine and metoclopramide) that have been described (Moncrieff, 2006).

In our patient it could also be possible that long-term thioridazine treatment was masking and treating an underlying psychotic illness that only became noticeable following the withdrawal of thioridazine.

Moncrieff’s review of the literature (2006) on antipsychotic withdrawal suggested the possibility that chronic exposure to antipsychotic medication may alter neuronal receptor profiles, paradoxically creating a susceptibility to psychotic illness (supersensitivity psychosis). A mechanistic explanation remains elusive; however, the upregulation of dopamine receptor density and sensitivity to dopamine in mesolimbic pathways may play some role. Moncrieff (2006) proposed that supersensitivity psychosis would be most readily provoked by drugs with short half-lives and that the withdrawal psychosis could actually be a distinct entity from the underlying illness.

This evidence for supersensitivity psychosis is stronger for clozapine (Moncrieff, 2006) but our case report and review of the literature also suggests that thioridazine is a medication that could induce a withdrawal psychosis. Of particular note is the suggestion that our patient may have developed a de novo treatment-resistant psychosis following years of thioridazine treatment. This presents the possibility that certain vulnerable populations such as intellectually impaired individuals may be at higher risk of enduring brain adaptations to long-term dopamine blockade, which could lead to an iatrogenic and long-lasting psychotic syndrome. It may well be that this is an under-recognised clinical entity, particularly in the intellectually impaired population. Clinicians should therefore exercise caution in such populations before prescribing long-term antipsychotic medication.

Intriguingly, our case report points to the possibility of iatrogenic and enduring psychosis triggered by long-term administration of antipsychotic medication.