Causal relationships between plasma lipidome and diabetic neuropathy: a Mendelian randomization study

Aims: This study aimed to assess whether genetically predicted levels of specific plasma lipid species, particularly structural lipids such as phosphatidylcholines, are associated with the risk of diabetic neuropathy (DPN), using a Mendelian randomization (MR) approach.

Methods: The authors used a two-sample MR approach to assess whether genetically predicted levels of 179 plasma lipid species — such as phosphatidylcholines, sphingolipids, and triglycerides — are associated with the risk of DPN. Genetic instruments were derived from genome-wide association studies (GWAS) involving approximately 20,000 individuals.Outcome data were sourced from the FinnGen biobank, which includes over 140,000 genotyped individuals with linked hospital records. DPN cases were identified using diagnostic codes, allowing association testing with the same SNPs used in the lipid dataset. MR analyses used inverse-variance weighting, with additional methods applied to assess robustness. Adjustments were made for HbA1c and BMI.

Results: Two phosphatidylcholine species — PC(16:0_20:2) and PC(16:1_18:1) — were significantly associated with a lower risk of DPN, with odds ratios ranging from 0.83 to 0.88 per standard deviation increase in genetically predicted lipid levels. These associations remained robust after adjusting for confounders and across multiple MR models, with no evidence of horizontal pleiotropy. Both species are structural lipids involved in membrane composition and lipoprotein metabolism, potentially relevant to nerve function.

Conclusions: Higher levels of certain phosphatidylcholine species may be linked to a lower risk of DPN, independently of glycaemic control and body weight. These findings support the hypothesis that membrane lipid composition may influence small fibre health and offer new directions for metabolic profiling in neuropathy risk assessment.

Comments: This study provides new evidence supporting a metabolic contribution to DPN, with a focus on lipid subclasses not routinely captured by standard lipid panels. The identification of phosphatidylcholine species associated with lower DPN risk complements existing data on the role of triglycerides and cholesterol in DPN. These phosphatidylcholines are functionally linked to lipoprotein metabolism and membrane integrity, suggesting potential relevance to both lipid transport and nerve structure.

The use of Mendelian randomization strengthens the validity of the findings by reducing confounding and reverse causation, and the consistency across multiple analytical methods adds confidence. However, the outcome was defined using diagnostic codes rather than clinical phenotyping, and the lack of nerve-specific assessments such as corneal confocal microscopy or nerve conduction studies limits mechanistic interpretation. The study population was also limited to individuals of European ancestry. Despite these limitations, the findings highlight the potential of lipidomic profiling to identify novel markers of DPN risk and support the broader exploration of lipid metabolism as a therapeutic target in this context.

Maryam Ferdousi

Reference: Wang Z, Liu Z, Yang Q, Qiao H, Yin Y, Zhao Z, Shao X. Causal relationships between plasma lipidome and diabetic neuropathy: a Mendelian randomization study. Front Endocrinol (Lausanne). 2025 Jan 15;15:1398691. doi: 10.3389/fendo.2024.1398691. PMID: 39882262; PMCID: PMC11774734.

🔗https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1398691/full