Is there a causal link between polyunsaturated fatty acids and osteoarthritis?

By Vijay Kumar MalesuReviewed by Lily Ramsey, LLMNov 17 2023

In a recent study published in Nutrients, a group of researchers explored the causal effects of polyunsaturated fatty acids (PUFAs) on knee and hip osteoarthritis (OA)  risk using Mendelian randomization (MR) and genome-wide association studies (GWASs) data.

Study: The Role of Polyunsaturated Fatty Acids in Osteoarthritis: Insights from a Mendelian Randomization Study. Image Credit: KucherAV/Shutterstock.com

Background 

OA affects over 500 million people globally and is a prevalent chronic joint disease, especially in knee and hip joints. This estimate, representing about 7% of the world's population, is projected to rise significantly by 2050.

The growing prevalence and the limitations of current treatments, primarily end-stage surgeries, emphasize the need for alternative approaches. Nutritional interventions, PUFAs like omega-3 and omega-6, have been proven beneficial due to their anti-inflammatory properties.

While omega-3s show promise in OA management, the role of omega-6s remains less clear. Further research is necessary to validate the potential benefits of PUFAs in osteoarthritis treatment and prevention, given the high global prevalence of OA, the limited effectiveness of current medications, and the inconsistent findings from previous studies on omega-3 and omega-6 fatty acids.

About the study 

In the present study, researchers utilized MR, leveraging GWAS data, to investigate the relationship between PUFAs, particularly omega-3 and omega-6, and OA, focusing on knee and hip OA.

They adhered to MR's core principles: genetic variants' significant association with PUFAs, no direct variant impact on OA except via PUFAs, and no confounder correlation. The team rigorously selected instrumental variables (IVs) strongly linked to PUFAs and independent of confounders.

They also conducted single-nucleotide polymorphism (SNP) pruning and linkage disequilibrium (LD) checks, along with evaluating weak instrument bias using the F-statistic.

The researchers sourced SNP summary data for PUFAs from the Nightingale Health UK Biobank Initiative, providing omega-6 and omega-3 fatty acids information. Genetic instrumental variables for these fatty acids were derived from the United Kingdom (UK) Biobank study involving over 100,000 European individuals.

The study utilized a high-performance Nuclear Magnetic Resonance (NMR) metabolomics platform for plasma sample analysis. For OA, data were obtained from the Arthritis Research UK Osteoarthritis Genetics (arcOGEN) project and the UK Household Longitudinal Study (UKHLS), involving hundreds of thousands of individuals.

MR analysis was performed using the R package TwoSample MR, employing the random-effects inverse-variance weighted (IVW) method as the primary technique, complemented by MR-Egger and weighted median methods for estimation.

These methods accounted for the potential invalidity of genetic variants and horizontal pleiotropy. Results were presented as odds ratios with confidence intervals, and Bonferroni correction was applied for multiple tests.

The researchers also conducted sensitivity analyses, including funnel plots, Cochran’s Q test, MR-Egger intercept, and leave-one-out analyses. These tests assessed heterogeneity and pleiotropy, validating the study of the results. Significant associations were carefully interpreted, considering the potential for suggestive associations and directional pleiotropy.

Study results 

In the study, researchers conducted MR-Egger regression and IVW analyses to assess heterogeneity in the causal effects of omega-3 fatty acids on knee OA (KOA) and Hip OA (HOA).

They observed heterogeneity in the MR analyses for both KOA and HOA, which led to using random-effects IVW estimation. The IVW results consistently supported a causal relationship between omega-3 fatty acids and KOA risk. 

Furthermore, MR-Egger intercept tests showed no horizontal pleiotropy in the analyses, as indicated by p-values exceeding 0.05. Scatter plots illustrated the estimated impacts of SNPs on omega-3 fatty acids and OA, and leave-one-out analysis confirmed that no outlier instrumental variables significantly influenced the overall results.

The funnel plot indicated no apparent horizontal pleiotropy, showing symmetrical variation in effect size around the point estimate.

The study also explored the causal effects of omega-6 fatty acids on KOA and HOA. Researchers selected 61 and 59 LD-IVs from GWASs for KOA and HOA, respectively. The IVW method suggested a potential association between genetically predicted omega-6 fatty acids and a reduced risk of KOA, with similar significant results obtained from the weighted median analysis.

The MR-Egger analysis showed consistent but non-significant results, and these findings indicated a possible causal relation of omega-6 fatty acids with a decreased risk of KOA. For HOA, the IVW method showed that omega-6 fatty acids were associated with reduced risk, a finding supported by significant results in the MR-Egger and weighted median methods.

Given the heterogeneity observed in the MR analyses of omega-6 for both KOA and HOA, the study employed the random-effects IVW method. This approach further supported the causal relationship between omega-6 and OA.

The analyses showed no horizontal pleiotropy, as confirmed by intercepts with p-values greater than 0.05. Scatter plots displaying individual causal estimates were presented, and the results were confirmed through leave-one-out tests. The funnel plots for the analyses of omega-6 and OA demonstrated the consistency and reliability of the findings.

Conclusion

Overall, the study provided evidence of a potential causal relationship between omega-3 and omega-6 fatty acids and the risk of developing KOA and HOA, suggesting that these nutrients may play a role in the management and prevention of OA.