The effect of first- and second-generation antipsychotics on brain morphology in schizophrenia: A systematic review of longitudinal magnetic resonance studies with a randomized allocation to treatment arms

Abstract

Schizophrenia manifests as loss of brain volume in specific areas in a progressive nature and an important question concerns whether long-term treatment with medications contributes to this. The aim of the current PRISMA systematic review was to search for prospective studies involving randomization to treatment. PROSPERO ID: CRD42020197874. The MEDLINE/PUBMED was searched and it returned 2638 articles; 3 were fulfilling the inclusion criteria. A fourth was published later; they included 359 subjects, of whom 86 were healthy controls, while the rest were first-episode patients, with 91 under olanzapine, 93 under haloperidol, 48 under risperidone, 5 under paliperidone, 6 under ziprasidone, and 30 under placebo. Probably one-third of patients were suffering from a psychotic disorder other than schizophrenia. The consideration of their results suggested that there is no significant difference between these medications concerning their effects on brain structure and also in comparison to healthy subjects. There does not seem to be any strong support to the opinion that medications that treat psychosis cause loss of brain volume in patients with schizophrenia. On the contrary, the data might imply the possible presence of a protective effect for D2, 5-HT2, and NE alpha-2 antagonists (previously called SGAs). However, the literature is limited and focused research in large study samples is essential to clarify the issue, since important numerical differences do exist. The possibility of the results and their heterogeneity to be artifacts secondary to a modification of magnetic resonance imaging (MRI) signal by antipsychotics should not be easily rejected until relevant data are available.

Do antipsychotic drugs shrink the brain? Probably not

“…The literature to date Fountoulakis and Stahl (2022), in this issue, have performed a pre-registered systematic review, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidance, of RCTs with longitudinal sMRI to see if antipsychotic drugs reduce brain volumes. Introducing the review and discussing the results, they also give a robust and rigorous review of the neuroimaging literature in schizophrenia more generally.…”

“…As such, they may best be regarded as having amphetamine or other drug-induced psychoses, which may notably differ from schizophrenia neurobiologically. While one cannot dismiss animal studies that point to 6-15% volume reductions on haloperidol compared to placebo, as Fountoulakis and Stahl (2022) point out, the animals do not have schizophrenia and the findings could even be artefactual due to, for example, sMRI signal disruption by antipsychotic drugs.…”

“…What then of the observational and animal studies? As Fountoulakis and Stahl (2022) point out, even if one takes this literature at face value, the maximum meta-correlation across observational studies is around 0.36. This would translate to a maximum effect of antipsychotic drug exposure explaining up to 10% of the variance in longitudinal volume loss in schizophrenia.…”