Abstract 4139045: Suicide and Self-Harm with GLP-1 Receptor Agonists: A Meta-Analysis of Randomized Controlled Trials

• Published in: Circulation (New York, N.Y.) Vol. 150; no. Suppl_1; p. A4139045
• Main Authors: Hosseini, Kaveh, Ebrahimi, Pouya, Batlle, Juan, Maqsood, Muhammad Haisum, Ayati, Aryan, Long, Clarine, McGowan, Natalie, Tarabanis, Constantine, Laynor, Gregory, Heffron, Sean
• Format: Journal Article
• Language: English
• Published: Hagerstown, MD Lippincott Williams & Wilkins 12.11.2024
• Subjects: Diabetes (Type II); Weight loss
• ISSN: 0009-7322; 1524-4539
• DOI: 10.1161/circ.150.suppl_1.4139045

Background: GLP-1 receptor agonists (RA) improve diabetes, reduce cardiovascular events, and cause substantial weight loss. However, concerns have been raised about increased adverse psychiatric effects - namely suicide. Given the explosive increase in use of these agents, detection of even rare side effects is of great clinical importance. We aimed to perform the first meta-analysis of adverse psychiatric outcomes (suicidal ideation/attempt/completion or self-harm) in randomized, placebo-controlled, clinical trials (RCTs) of GLP-1 RA. Methods: A comprehensive literature search through 8/29/23 was conducted to identify RCTs involving adults with diabetes and/or overweight/obesity treated with GLP-1 RA or placebo for ≥6 months. 7,229 non-duplicate articles were screened for inclusion and 144 had data extracted. After querying all available data sources, including direct contact with principal investigators or corresponding authors, 21 of the 144 studies were found to have recorded incidence of adverse psychiatric outcomes. A random-effects meta-analysis was performed to estimate risk ratios and 95% confidence intervals (CI) for the primary outcome. Results: In total, 36,168 subjects received GLP-1 RA and 30,445 placebo. The event rate was very low (63 - GLP-1 RA; 45 - placebo). The pooled odds ratio for incidence of the primary outcome was 0.88 [95% CI, 0.60-1.29] Figure. Notably, 6 studies, including a total of 11,828 subjects, had a history of suicide attempt/depression as exclusion criteria for enrollment. Removal of these studies did not change the overall findings (OR 0.77 [95% CI, 0.38, 1.57]). Conclusion: Our comprehensive meta-analysis of placebo-controlled RCTs does not support an association of GLP-1 RA use with increased adverse psychiatric events among adults with diabetes or overweight/obesity. However, these outcomes were recorded in <15% of all trials, and >1/5 excluded subjects predisposed to these outcomes. More research is necessary, and continued surveillance remains warranted, to establish the safety of GLP-1 RA, particularly in patients at risk for self-harm/suicide.