Analyzing the impact of heavy metal exposure on osteoarthritis and rheumatoid arthritis: an approach based on interpretable machine learning Original paper
Researched by:
-
Karen Pendergrass
ID
Karen Pendergrass
Karen Pendergrass is a microbiome researcher specializing in microbiome-targeted interventions (MBTIs). She systematically analyzes scientific literature to identify microbial patterns, develop hypotheses, and validate interventions. As the founder of the Microbiome Signatures Database, she bridges microbiome research with clinical practice. In 2012, based on her own investigative research, she became the first documented case of FMT for Celiac Disease—four years before the first published case study.
-
Rheumatoid Arthritis
Rheumatoid Arthritis
OverviewRheumatoid arthritis (RA) is a systemic autoimmune disease marked by chronic joint inflammation, synovitis, and bone erosion, driven by Treg/Th17 imbalance, excessive IL-17, TNF-α, and IL-1 production, and macrophage activation. Emerging evidence links microbial dysbiosis and heavy metal exposure to RA, [1][2] with gut microbiota influencing autoimmune activation via Toll-like receptor (TLR) signaling, inflammasome activation, […]
-
Karen Pendergrass
Researched by:
-
Karen Pendergrass
ID
Karen Pendergrass
Microbiome Signatures identifies and validates condition-specific microbiome shifts and interventions to accelerate clinical translation. Our multidisciplinary team supports clinicians, researchers, and innovators in turning microbiome science into actionable medicine.
Karen Pendergrass
What Was Studied?
This study examined the impact of heavy metal exposure on the prevalence and differentiation of osteoarthritis (OA) and rheumatoid arthritis (RA) using interpretable machine learning models. Researchers analyzed data from the National Health and Nutrition Examination Survey (NHANES) (2003–2020) to assess how various heavy metals contribute to arthritis risk.
Who Was Studied?
The study population consisted of 14,319 participants from NHANES who met specific inclusion criteria, including age ≥ 20 years and confirmed arthritis status via blood and urine heavy metal testing.
Key Findings
Using machine learning techniques such as LASSO regression and SHapley Additive exPlanations (SHAP), the study identified tungsten, cobalt, cadmium, antimony, arsenic, and blood cadmium as significant risk factors for arthritis, while molybdenum, thallium, lead, and mercury appeared to have a protective or neutral association. Cadmium exposure showed a strong correlation with rheumatoid arthritis (RA), likely due to its role in oxidative stress and inflammation, while arsenic exposure was linked to both osteoarthritis (OA) and RA, with previous studies indicating its contribution to cartilage degradation. Tungsten and antimony emerged as newly recognized risk factors, though their mechanisms remain unclear. In contrast, molybdenum exhibited a potential protective effect, possibly by counteracting inflammation. The study’s machine learning models demonstrated high predictive accuracy, with XGBoost achieving 81% accuracy in identifying arthritis and LightGBM distinguishing between OA and RA with 76% accuracy.
Greatest Implications
This study reinforces the environmental component of arthritis development, suggesting that heavy metal exposure contributes to arthritis risk and progression. Machine learning models, particularly SHAP-based interpretations, provide valuable predictive tools for early detection. The findings highlight tungsten, cobalt, cadmium, and arsenic as potential modifiable risk factors, paving the way for targeted interventions to reduce arthritis prevalence.