2100: Paradoxical Antagonistic Association of Visceral and Subcutaneous Adipose Tissue with Circulating C-Reactive Protein in Rheumatoid Arthritis

Jon Giles and Joan Bathon, Columbia University, New York, NY

Background/Purpose: C-reactive protein (CRP) is used as an indicator of disease activity in people with RA.However, adipose tissue also expresses inflammatory cytokines that induce hepatic CRP production.Both visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) express inflammatory cytokines, with effects on CRP that are additive.However, whether this is also true for RA has not been explored.

Methods: RA patients were from two independent cohorts exploring subclinical cardiovascular disease.Non-RA controls were from a large multi-center cohort study, also studying subclinical cardiovascular disease.Known cardiovascular disease was an exclusion criterion for all 3 cohorts.In all 3 cohorts, participants underwent abdominal CT scanning with quantification of the areas of VAT and SAT at the L4/L5 disc interspace.VAT and SAT were interpreted by the same reader for both RA cohort 1 and the non-RA cohort.Readers were unaware of patient characteristics.For each cohort, stored serum samples were assayed for CRP via a central lab in single batches.Generalized linear models were constructed to explore the interaction of VAT and SAT with CRP.

Results: There were 131 RA patients in RA cohort 1 and 121 RA patients in RA cohort 2 (total n=252).The mean age of the combined RA cohorts was 57 years.The RA patients were mostly women (71%) of non-Hispanic White race (63%) with a median RA duration of 8.1years, 80% seropositive for RF or anti-CCP and a median DAS28 of 3.7.Non-biologics, biologics, and prednisone were used in 81%, 39, and 35%, respectively.In the non-RA cohort, higher SAT was associated with significantly higher CRP levels within each tertile of VAT, with progressively higher CRP with both higher SAT and VAT (Fig).In contrast, in both RA cohorts, there was a strong and significant association of higher SAT with CRP among those in the lowest tertile of VAT.However, the association of SAT with CRP among those in the highest tertile of VAT was weaker and not statistically significant in both RA cohorts (Fig).For those in the highest tertile of SAT, CRP was significantly lower among those in the highest vs lowest tertile of VAT, despite those in the highest tertile of VAT having a higher BMI (35.6 vs 32.6 kg/m2 for the combined RA cohorts) and similar DAS28, swollen, and tender joints counts.Use of biologic and non-biologic DMARDs and prednisone was not different between the groups. Adjusting for demographics and BMI did not reduce the significance of the antagonistic interaction.

Conclusion: We observed a similar paradoxical association of VAT and SAT on circulating CRP level in two independent cohorts of RA patients.These findings have implications on how elevated CRP levels are interpreted in RA patients and warrant additional study into the mechanism of how higher VAT may suppresses SAT-associated inflammation in RA.




J. Giles: AbbVie, 2, Eli Lilly, 2, Gilead, 2, Novartis, 2, Pfizer, 2; J. Bathon: None.