Abstract
Background
Previous observational studies have shown controversial results about the relationship between lipid levels and low back pain (LBP). Herein, we aimed to explore the potential causal relationship between lipid levels and LBP by using the mendelian randomization (MR) analysis.
Methods
In this two-sample MR study, data were extracted from publicly available MRC Integrative Epidemiology Unit database. Three single-nucleotide polymorphisms (SNPs) of lipid levels [high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), and triglycerides (TG)] and two SNPs of LBP risk (LBP and back pain) were retrieved and used as genetic instrumental variables. Inverse-variance weighted (IVW), weighted median, MR-Egger, robust adjusted profile score (MR-RAPS), and MR-PRESSO were used to examine the potential causal association between lipid levels and LBP.
Results
IVW (fixed effect) estimation indicated that increased HDL-C level was negatively related to the odds of LBP for European populations. [odds ratio (OR) = 0.923, 95% confidence interval (CI): 0.857–0.993, P = 0.0323]. Similar results were also found in IVW (random effect) (OR = 0.923, 95% CI: 0.866–0.983, P = 0.0134), MR-Egger (OR = 0.858, 95%CI: 0.757–0.973, P = 0.0177), MR-RAPS (OR = 0.932, 95%CI: 0.871–0.997, P = 0.0419), and MR-PRESSO (OR = 0.933, 95%CI: 0.880–0.989, P = 0.0198) analyses. Whereas no causal link was observed between LDL-C/TG and LBP risk (P>0.05).
Conclusion
This two-sample MR study demonstrated a causal relationship between lipid levels and LBP risk. Further investigations are necessary to elucidate the causal association and specific underlying mechanisms by which lipid levels contribute to the development of LBP.