Debate continues over long-term antipsychotic use

Care in choice of agent, caution in efforts to discontinue therapy and the need for more studies were points of agreement when Robin Murray (King’s College, London, UK) and Konstantinos Fountoulakis (Aristotle University of Thessaloniki, Greece) discussed the pros and cons of long-term antipsychotic usage after first episode psychosis.

The good news is that ongoing randomized trials in England, Denmark, Holland and Australia will cast light on whether maintaining antipsychotics is better than dose reduction or discontinuation once first episode psychosis (FEP) has resolved, Professor Murray told the CINP 2021 Virtual debate on long-term treatment.

These studies should give us more information than we have from the Wunderink et al trial,1 which has been much debated. The bad news is that we still have to wait 2-3 years for the answer.

Until then we have the naturalistic studies which are subject to confounding factors and can’t distinguish between association and causation. In particular, it is possible that patients who come off antipsychotics and do well are those with the best prognosis with or without continued treatment.

Even high exposure to antipsychotics is associated with less mortality than no exposure3


Do adverse metabolic effects increase mortality?

Around 16% of the UK population is obese, and that is also true of people who develop FEP. But the proportion rises to 29% in people treated for a year, and in those with a long history of psychosis around 50% are obese.2

So, at the least, it would be helpful to use weight-sparing antipsychotics, Robin Murray advised. But there is controversy about the broader question of whether metabolic syndrome increases long-term mortality.

Scandinavian evidence that even high exposure to antipsychotics is associated with less mortality than no exposure3 is challenged by the argument that those who did not take antipsychotics were older and more likely to be homeless.

The jury is out on cause of brain volume loss

On the other hand, as Professor Fountoulakis said, we have at least some data suggesting antipsychotics reduce mortality,4 and no data suggesting they increase it.


And brain volume?

Longitudinal data from Iowa suggesting antipsychotics are associated with loss of brain volume,5 and similar findings by the ENIGMA Consortium,6 question the dose and duration of therapy. The potentially adverse effects of antipsychotics on brain structure seem supported by animal studies, Professor Murray said.

But the idea that brain loss is due not to treatment but to progressive disease remains plausible, and Professor Fountoulakis  pointed out that the nature of brain loss (animals lose cells, patients do not), its rapidity, and the fact that animals do not experience schizophrenia question the validity of such models.

We can’t predict people who will do well off treatment, and the one-year relapse rate in remitted FEP is around 70% without continued antipsychotics8


Some do well off treatment

Professor Murray noted that the AESOP-10 study found that 19% of people with schizophrenia had no psychotic symptoms at ten years despite taking no antipsychotics.

Professor Fountoulakis responded by saying that we cannot predict in advance who these patients are, and that -- in supporting the welfare of a minority -- we should not put the majority at risk. And real-world data suggest the one-year relapse rate in remitted FEP is around 70% without continued antipsychotic treatment.8

Our correspondent’s highlights from the symposium are meant as a fair representation of the scientific content presented. The views and opinions expressed on this page do not necessarily reflect those of Lundbeck.


1. Wunderink L et al. JAMA Psychiatry 2013;70(9):913-20

2.  Gaughran F et al.  BMC Psychiatry 2017

3. Torniainen  et al.  Schiz  Bull 2015 May;41(3):656-63

4. Taipale H et al. Schizophr Res 2017;197:274-80

5. Ho B-C, Andreasen N et al. Arch Gen Psychiatry.2011 ; 68(2): 128–137

6. Van Erp TGM et al. Biol Psychiatry 2018;84(9):644-654

7. Morgan C et al. Psychological Medicine 2014; 44(13): 2713–2726

8. Di Capite S et al. Early Intervention Psychiatrty  2018 ;12(5):893-899