Antipsychotic drugs have been the mainstay of treatment for people diagnosed with schizophrenia or psychosis since they were introduced in the 1950s. They are recommended for acute psychosis, and routinely prescribed on a long-term basis for the amelioration of ongoing symptoms and the prevention of relapse. The National Institute for Health and Care Excellence (NICE) clinical guideline assumes that most people with schizophrenia will need long-term antipsychotic treatment (NICE 2014). New evidence suggests that psychiatric practice may need to change, however.
Evidence for long-term treatment
The evidence base for the long-term use of antipsychotic drugs consists of randomised controlled trials in which people already established on treatment are randomised to continue with maintenance treatment or to have their antipsychotics discontinued. These trials find higher rates of relapse among people randomised to discontinue antipsychotics (Reference Leucht, Tardy and KomossaLeucht 2012).
Problems with this evidence base have been recognised for some time. First, follow-up has mostly been short, with few trials lasting longer than a year. Second, relapse has been defined in a wide variety of ways, and many definitions have not specified the presence of psychotic symptoms. Third, discontinuation has usually been abrupt, with the consequence that antipsychotic withdrawal symptoms may have been misdiagnosed as relapse in placebo-treated patients, especially in studies in which relapse was defined as a small increase in general symptoms. Fourth, few studies have evaluated outcomes other than relapse (Reference Leucht, Tardy and KomossaLeucht 2012).
Meta-analyses suggest that studies with longer follow-up tend to find smaller differences in relapse rates than those lasting for shorter periods (Reference Leucht, Tardy and KomossaLeucht 2012). Consequently, the lack of longer-term follow-up of the outcome of antipsychotic discontinuation presents a major obstacle to the evaluation of antipsychotic therapy.
The Wunderink et al study
In 2013, data from a 7-year follow-up of a randomised antipsychotic discontinuation trial were published (Reference Wunderink, Nieboer and WiersmaWunderink 2013). The original study (Reference Wunderink, Nienhuis and SytemaWunderink 2007), conducted in The Netherlands, involved 131 people who had recovered from a first episode of psychosis and had been in remission for 6 months. Participants were randomised to antipsychotic maintenance treatment or to a gradual and flexible antipsychotic reduction strategy. At the 18-month follow-up (128 participants), only 54% of the discontinuation group had completely discontinued antipsychotics, and only 22% discontinued and remained off. Thirty-two per cent stopped and then restarted their antipsychotic medication and the rest did not manage to discontinue at all. At 18 months, relapses were twice as common in the discontinuation group than in the maintenance group (43% v. 21%), although time spent in hospital was not increased. There were no differences in measures of social functioning between the groups, but there was a non-significant trend towards more people being in work in the discontinuation group (Reference Wunderink, Nienhuis and SytemaWunderink 2007). A neurocognitive substudy also showed better performance in the discontinuation group (Reference Faber, Smid and van GoolFaber 2011). Despite this, Wunderink et al concluded that antipsychotic discontinuation is a strategy that should be reserved for people who are unwilling to adhere to maintenance treatment.
At 7-year follow-up, however, the situation had changed. Rates of relapse had evened out, with 62% of the discontinuation group experiencing a relapse v. 69% of the maintenance group. Although symptomatic recovery was the same in the two groups, people randomised to the antipsychotic discontinuation strategy were twice as likely to show ‘functional remission’ than those randomised to maintenance (46% v. 20%). At this point, 21% of the discontinuation group were off antipsychotics completely, and a further 21% were taking very low doses (less than 1 mg haloperidol equivalent per day) (Reference Wunderink, Nieboer and WiersmaWunderink 2013). A post hoc analysis according to actual treatment received showed that 56% of patients who discontinued antipsychotics or reduced to a very low dose showed functional remission compared with 22% of those who continued antipsychotics (Reference Wunderink, Nieboer and WiersmaWunderink 2013). Although this study recruited only people with first-episode psychosis, a recent meta-analysis found no difference in the outcome of antipsychotic discontinuation between studies involving patients with a first episode of psychosis and those with recurrent illness (Reference Leucht, Tardy and KomossaLeucht 2012).
Antipsychotics and the brain
Other important developments include emerging evidence about the effects of antipsychotics on brain matter. In 2005, a study with macaque monkeys showed that these drugs reduce brain matter by around 10% over 18 months of treatment (Reference Dorph-Petersen, Pierri and PerelDorph-Petersen 2005). Evidence in patient populations is complex, but some studies indicate decreases in brain matter associated with antipsychotic treatment levels (Reference Moncrieff and LeoMoncrieff 2010), including results of the largest and longest cohort study of first-episode schizophrenia published to date (Reference Andreasen, Nopoulos and MagnottaAndreasen 2011).
Whether these structural effects are reflected in intellectual functioning is uncertain. Some studies suggest not, but others have found an association between reduced brain matter and cognitive capacity (Reference Moncrieff and LeoMoncrieff 2010).
Conclusions
Antipsychotics can effectively reduce acute psychotic symptoms in many people, but evidence on the benefits of long-term treatment is more equivocal, and recent evidence underlines their potentially negative effect on brain volume, alongside other serious physical complications.
NICE (2014) recommends more research into long-term outcomes of antipsychotic discontinuation. In the meantime, recent evidence suggests that a trial of supported antipsychotic discontinuation should be considered as one option in the routine treatment of people with psychotic conditions. This strategy may help some people to improve their functioning and quality of life, and reduce the risks to their physical health.
eLetters
No eLetters have been published for this article.