Lipid profile and suicide attempt in bipolar disorder: A meta-analysis of published and unpublished data

Abstract

Evidence suggests an association between low lipid levels and suicidality in subjects with severe mental disorders. This is the first systematic review and meta-analysis aimed at exploring differences in lipid profile between suicide attempters and non-attempters with bipolar disorder. We included observational studies providing comparative cross-sectional data on total cholesterol, LDL-cholesterol and triglycerides levels. We searched main Electronic Databases, identifying 11 studies that met our inclusion criteria, including also unpublished data. Meta-analyses based on random-effects models were carried out, generating pooled standardized mean differences (SMDs). Heterogeneity among studies was estimated using the I2 index. The meta-analyses included data on lipid profile from 11 studies based on 288 subjects with and 754 without suicide attempt, respectively. No differences in total cholesterol (SMD: −0.10; 95%CI: −0.30 to 0.10; p=0.34), LDL-cholesterol (SMD: −0.26; 95%CI: −0.65 to 0.13; p=0.19), and triglycerides (SMD: −0.06; 95%CI: −0.31 to 0.19; p=0.63) were detected. Heterogeneity across studies was low-moderate and no risk of publication bias was found. Subgroup analyses showed no differences on effect size across different study characteristics, including different time-frames of suicide attempt, except for small sample size. Therefore, the evidence for an association between serum lipid profile and suicidality in bipolar disorder cannot be claimed. More research is needed to better understand the mechanisms underlying suicidal behaviours in bipolar patients, exploring further peripheral biomarkers as this may help clinicians screen and prevent suicidality.