Background/Purpose : It is debated whether erosive hand osteoarthritis (OA) is a separate inflammatory subset of hand OA, or just a severe form of the same disease. Cross-sectional studies on ESR and CRP are limited in numbers, performed on small cohorts, and have show conflicting results, with higher as well as lower serum levels in erosive OA compared to non-erosive OA.[1-4] With a large OA cohort, we have the opportunity to explore this further, and hereby compare the serum levels of inflammatory biomarkers in erosive and non-erosive hand OA.

Methods: We included subjects with radiographic hand OA (i.e. Kellgren-Lawrence grade ≥ 2 in one or more finger joint(s)) from a community-based OA cohort, and measured their plasma levels of erythrocyte sedimentation rate (ESR) and C-reactive protein (using a high-sensitivity assay; hsCRP). Erosive status was defined as having one or more interphalangeal joint(s) with radiographic central erosions, and rheumatic inflammatory comorbidities were excluded. In crude linear regression analyses, we explored whether erosive hand status was associated with higher levels of ESR and hsCRP (log-transformed) as compared to non-erosive hand OA. In adjusted analyses, age, sex and body mass index (BMI) were added to the model. To explore the association between OA-related joint inflammation and systemic inflammatory markers, we performed additional analyses, in which the sum scores of ultrasound-detected synovitis in the hands (range 0-90) and the knees (range 0-6) were included in the model. Significance level was set at p<0.05.

Results: Non-erosive and erosive hand OA were present in 373 (76%) and 119 (24%) subjects, respectively (Table).

In crude analyses, there were significant associations between erosive hand OA status and higher log-transformed ESR (ß = 0.23, 95% CI 0.05,0.41) as well as hsCRP (ß = 0.22, 95% CI –0.1,0.45). After adjusting for age, sex and BMI, the associations were not longer significant for neither ESR (ß =, 95% CI) nor hsCRP (ß = 0.11, 95% CI –0.12,0.35). Ultrasound-detected synovitis sum scores for the hands and knees were not related to higher systemic inflammatory markers with ß = 0.02 (95% CI –0.01,0.04) and ß = –0.02 (95% CI –0.10,0.05) for ESR and ß = 0.02 (95% CI –0.01,0.04) and ß = –0.01 (95% CI –0.11,0.09) for hsCRP, respectively.

Conclusion: Persons with erosive hand OA have higher ESR and hsCRP as compared to non-erosive hand OA. However, the higher levels of systemic inflammatory markers are explained by age, sex and BMI, rather than local joint inflammation. Both erosive and non-erosive hand OA is associated with local joint inflammation with no effect on systemic inflammatory markers.

Table:

References: 1) Vannini A, et al. Acta reumatologica portuguesa; 2013. 2) Punzi L, et al. Ann Rheum Dis; 2005. 3) Olejarova M, et al. Joint, bone, spine: revue du rhumatisme; 2000. 4) Bijsterbosch J, et al. Ann Rheum Dis; 2011.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-effect-of-erosive-status-on-systemic-inflammatory-biomarkers-in-hand-osteoarthritis/