Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.
Association of Medicaons and Rheumatoid Arthritis Susceptibility Polymorphism with Lipid Profiles in Patients with Rheumatoid Arthritis.
Davis1, Lisa A., Pointer2, Lauren M., Wolff3, Roger K., Reimold4, Andreas M., Kerr5, Gail S., Mikuls6, Ted R., Cannon7, Grant W.
Univ of Colorado School of Med, Aurora, CO
Denver Veterans Affairs Medical Center, Denver, CO
University of Utah, Salt Lake City, UT
Dallas VA and University of Texas Southwestern, Dallas, TX
Washington DC VA and Georgetown University, Washington, DC
Omaha VA and University of Nebraska, Omaha, NE
Salt Lake City VA and University of Utah, Salt Lake City, UT
Denver VA and University of Colorado, Aurora, CO
Background/Purpose:
Rheumatoid arthritis (RA) patients have an increased risk of cardiovascular disease (CVD), and though the mechanism is unknown, there are independent associations with RA-related factors (del Rincon I Arth Rheum 2005; Wolfe F, Arth Rheum 2008; Goodson Arth Rheum 2002). We examined whether single nucleotide polymorphisms (SNPs) linked with RA susceptibility and treatment are associated with low density lipoprotein (LDL) and triglyceride (TG) levels in subjects with RA.
Methods:
Patients (n=1362) enrolled in the prospective Veterans Affairs RA (VARA) registry were genotyped for multiple SNPs with DNA samples derived from whole blood. Genes and associated SNPs included: REL (rs10203477, rs842647, rs13031237, rs9309331); DDAH2 (rs15574), interleukin-10 (rs3024493, rs1800872, rs1800896); methylenetetrahydrofolate reductase (rs1801131, rs1801133); tumor necrosis factor/ lymphotoxin A (rs3093662, rs1800629, rs3093668); and tumor necrosis factor receptor associated factor 1 (rs1014529, rs1014530). Covariates included: patient characteristics (age, ethnicity, gender, body mass index [BMI], diabetes, smoking status, education); RA severity markers (anti-CCP and rheumatoid factor status, C reactive protein level); lipid-lowering agent use (statin, fibrate, bile acid sequestrant); and lipid profile data. Multivariate linear regression was performed to determine factors associated with TG and LDLs. A p-value <0.01 was deemed significant in the final model.
Results:
Factors associated with lower LDL level were age, diabetes, statin use, and the REL-related SNP (rs9309331), where the presence of CC conferred a 13 point lower LDL value compared to GG (see Table). The distribution of TG values required a lognormal transformation of the data, and results are interpretable as percentages relative to subjects without the described trait. For example, subjects using fibrates still demonstrated TG values 87% higher than subjects not taking fibrates, despite use of these agents. Factors associated with higher TG levels were Caucasian race, BMI, diabetes, smoking status, and fibrate use. There was an interaction term between diabetes and fibrate use that was associated with a decreased TG level.
Table. Predictors of LDL and TG levels in patients with rheumatoid arthritis
| Variable | Parameter estimate | SE | Transformed percentage | p-value | |
|---|---|---|---|---|---|
| Low density lipoprotein (LDL) model | |||||
| Age | -0.32 | 0.09 | — | <0.001 | |
| Presence of diabetes | -7.25 | 2.42 | — | 0.003 | |
| rs9309331 | CC | -12.92 | 4.89 | — | 0.008 |
| GC | -4.33 | 2.16 | — | 0.045 | |
| GG | referent | — | — | ||
| Statin Use | -15.65 | 3.20 | — | <.0001 | |
| Triglyceride (TG) model | |||||
| Caucasian | 0.24 | 0.04 | 27.29% | <.0001 | |
| Body mass index | 0.02 | 0.00 | 1.84% | <.0001 | |
| Presence of diabetes | 0.17 | 0.04 | 18.04% | <.0001 | |
| Smoking status | current | 0.15 | 0.05 | 16.59% | 0.001 |
| former | 0.08 | 0.04 | 8.87% | 0.037 | |
| never | referent | ||||
| Fibrate use | 0.62 | 0.15 | 86.66% | <.0001 | |
| Diabetes* fibrate interaction term | -0.70 | 0.25 | -49.50% | 0.005 |
Conclusion:
RA subjects on statins demonstrate lower LDL compared to RA subjects not receiving statins, but this was not found for fibrate use and TGs (unless the subject was also diabetic). A possible connection between rs9309331 and LDL levels should be further investigated.
To cite this abstract, please use the following information:
Davis, Lisa A., Pointer, Lauren M., Wolff, Roger K., Reimold, Andreas M., Kerr, Gail S., Mikuls, Ted R., et al; Association of Medicaons and Rheumatoid Arthritis Susceptibility Polymorphism with Lipid Profiles in Patients with Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :1181
DOI:
