Antipsychotic Drugs and Cognitive Function: A Systematic Review and Pairwise Network Meta-Analysis
Lena Feber,Natalie L Peter,Virginia Chiocchia,Johannes Schneider-Thoma,Spyridon Siafis,Irene Bighelli,Wulf-Peter Hansen,Xiao Lin,Daniel Prates-Baldez,Georgia Salanti,Richard S E Keefe,Rolf R Engel,Stefan Leucht
DOI: https://doi.org/10.1001/jamapsychiatry.2024.2890
2024-10-16
JAMA Psychiatry
Abstract:Importance: Cognitive deficits are a substantial part of the symptoms of schizophrenia spectrum disorders (SSDs) and contribute heavily to the burden of disease. Antipsychotic drugs are not cognitive enhancers, but due to their different receptor-binding profiles, they could differ in their effects on cognition. No previous network meta-analysis compared antipsychotics to placebo, which is important to determine whether use of these drugs is associated with cognitive performance in SSDs at all. Objective: To determine the association of treatment with various antipsychotics and cognition in patients with SSDs. Data sources: Cochrane Schizophrenia Trials Register through June 25, 2023. Study selection: Randomized clinical trials examining the effects on cognition of antipsychotic drugs or placebo in participants with SSD. Data extraction and synthesis: A systematic review and random-effects frequentist network meta-analysis was performed following Preferred Reporting Items for Systematic Reviews and Meta-analyses-Network Meta-analysis reporting guideline. Main outcomes and measures: The primary outcome was change in overall cognition score calculated for each study. Secondary outcomes included cognitive domains, quality of life, and functioning. Results: This study included 68 studies involving 9525 participants (mean [SD] age, 35.1 [8.9] years; 5878 male [70%] and 2890 [30%] female; some studies did not provide this information). There were few clear differences between antipsychotics, but first-generation dopamine antagonists haloperidol (standardized mean difference [SMD], 0.04; 95% CI, -0.25 to 0.33) and fluphenazine (SMD, 0.15; 95% CI, -0.39 to 0.69) as well as clozapine (SMD, 0.12; 95% CI, -0.23 to 0.48) ranked low. No individual antipsychotic was associated with a clearly better outcome than placebo, but antipsychotics as a group were, with small effect sizes (mean SMDs: adrenergic/low dopamine, 0.21; serotonergic/dopaminergic, 0.26; muscarinic, 0.28; dopaminergic, 0.40). Conclusion and relevance: Although data are relatively sparse, those reviewed in this study suggest that first-generation dopamine antagonists and clozapine should be avoided when cognitive deficits are a concern. Antipsychotics are not procognitive drugs. The overall small superior outcomes compared to placebo may be explained by less disordered thought patterns associated with fewer positive symptoms rather than cognitive deficits in the proper sense. The findings also suggest that harmonizing measurement of cognitive function in randomized clinical trials would be beneficial.