Comparative Efficacy of Animal Depression Models and Antidepressant Treatment: A Systematic Review and Meta-Analysis

Abstract

Background: Animal models are critical tools in the study of psychiatric disorders; however, none of the current models fully reflect human stress-related disorders, even though most of the knowledge about the mechanisms of depression comes from animal studies. Animal studies are useful in pharmacological research, whereby we can obtain results that translate into patient treatment by controlling environmental factors, especially in behavioural research. The authors systematically reviewed this issue since medical databases provide access to many primary studies. Methods: A systematic review and meta-analysis were conducted based on 25 primary studies. The studies were identified in databases such as PubMed, Embase, and Web of Science (December 2022) according to the inclusion and exclusion criteria established at the beginning of the research and published in the form of a protocol, following the PRISMA and Cochrane Collaboration methodology for secondary studies and CAMARADES (CAMARADES Berlin, QUEST-BIH Charité) for secondary studies on animals. Forest plot analyses were performed (data presented as Mean Difference, Random Model, Inverse Variance), Risk of Bias assessment (Systematic Review Center for Laboratory animal Experimentation (SYRCLE) evaluation), quality assessment of included studies (Animal research: Reporting of In Vivo Experiments (ARRIVE)), and a range of data from source publications were compiled in tabular form. The study analysed the popularity of both animal depression models (ADM) and rat strains used in pharmacological research to test the efficacy of antidepressant drugs based on the immobility time (IT) factor (Forced Swimming Test). The study examined selective serotonin reuptake inhibitors, namely fluoxetine, sertraline, paroxetine, citalopram, and escitalopram. Additionally, the study addressed issues concerning the “data availability statement” because precise IT data analysis was impossible in the case of 212 papers. Results: Our data confirm that the Chronic Unpredictable Mild Stress (CUMS) model is the most popular and versatile model used in preclinical depression research, while the two most popular rat strains were Wistar and Sprague-Dawley. The quality of included papers based on the ARRIVE assessment showed a ratio value equal to 0.63, meaning that studies were of intermediate overall quality. The Risk of Bias assessment based on the SYRCLE tool revealed a high risk related to the blinding and the random outcome assessment. In the meta-analysis, the results indicate that all analysed drugs demonstrated efficacy in reducing IT, and the most analysed drug was fluoxetine (confirmed based on 17 studies (19 models)). The analysis of the efficacy of ADMs showed that the most effective models were CUMS, Flinders Sensitive Line (genetic model), Social Isolation, Restraint Stress, and Low-dose Lipopolysaccharide (pharmacological model). Only 2.35% (5 out of 212) of corresponding authors responded to our data request. Conclusions: The study highlights the dominance of the CUMS model and the Wistar and Sprague-Dawley rat strains in preclinical depression research, affirming the efficacy of SSRIs, particularly fluoxetine, in reducing IT. The findings underscore the need for better data availability and methodological improvements despite intermediate overall study quality and notable bias risks. Enhanced transparency and rigorous assessment standards are essential for advancing the reliability of animal models in depression research.