Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

Analysis of the Multicenter Osteoarthritis (MOST) Study.

Singh1,  Jasvinder, Neogi2,  Tuhina, Felson3,  David T., Torner7,  James, Baker4,  Kristin, Nevitt5,  Michael C., Tolstykh5,  Irina

Birmigham VA Medical Center and University of Alabama, Minneapolis, MN
Boston Univ Schl of Med, Boston, MA
Boston University School of Medicine, Boston, MA
Boston University School of Medicine
UCSF, San Francisco, CA
University of Alabama
University of Iowa


OA has been labeled an inflammatory disorder but studies have examined only a limited repertoire of inflammatory markers with disease and have reported inconsistent findings. We assessed whether inflammation biomarkers and those representing adipokines are associated with synovitis and knee pain in participants in a cohort study of knee OA.


A subgroup of 100 Caucasian women were included from the NIH-funded Multicenter Osteoarthritis (MOST) Study. MOST participants were age 50 to 79 years at baseline and either with symptomatic knee OA or at high risk of disease. At baseline and 30 month follow-up, subjects were queried about the presence of knee pain on most days, completed a knee-specific Western Ontario McMaster Osteoarthritis Index (WOMAC), and had weight-bearing PA and lateral knee x-rays. New knee pain was present if a subject did not have knee pain on most days at baseline but did at follow-up irrespective of x-ray OA. At 30 months, participants underwent a gadolinium contrast-enhanced MRI using a 1.5 Tesla scanner to detect synovitis, categorized as none/questionable, mild, or a lot/extensive. WORMS readings of MRI's were used to score bone marrow lesions (BML's) and size of effusion. We assessed the multivariable-adjusted association of 1 standard deviation change in each biomarker level at baseline (blood or urine) with synovitis, prevalent pain and incident pain on most days on WOMAC scale at 30 months: TNF-alpha, MCP-1, leptin, adiponectin, PAI-1, CRP, MMP3, TIMP3, ICAM-1, IGF-1, IGF-BP3, oxidized LDL, TGF-beta 1, estradiol (total), DHEA-S, sex hormone binding globulin (SHBG) and cartilage oligomeric matrix protein (COMP).


Mean age (standard deviation) was 59.5 years (7.2), body mass index (BMI) was 29 (4.8) kg/m2, 57% had no radiographic OA at baseline, 20% with unilateral OA and 23% with bilateral OA and 68% were using an pain medications at baseline. Adjusted for site, age, BMI, baseline K/L grade, BML score (>0) and effusion score (>0), the following associations were significant for presence of a lot/extensive synovitis at 30-month visit: baseline adiponectin, OR 2.6 (95% CI:1.3, 4.9) and baseline TGF-beta 1, OR 0.3 (95% CI: 0.2, 0.7).

Significant predictors of outcomes in Multivariable adjusted Odds per 1 Standard deviation increase in each biomarker

 A lot /extensive Synovitis (ref: none/mild)Prevalent knee pain on WOMAC (ref: none)Incident knee pain on WOMAC (ref: none)
 OR (95% CI)p-valueOR (95% CI)p-valueOR (95% CI)p-value
Adiponectin2.6 (1.3, 4.9)0.0051.4 (0.9, 2.4)0.160.9 (0.4, 2.2)0.84
TGF-B10.3 (2.0, 0.7)0.0021.2 (0.7, 2.1)0.431.2 (0.5, 2.9)0.71
TNF-alpha1.2 (0.7, 2.1)0.441.9 (1.1, 3.2)0.0152.4 (0.8, 6.6)0.10
ICAM-11.1 (0.7, 1.8)0.711.1 (0.7, 1.8)0.570.3 (0.1, 0.97)0.044

Similarly, in multivariable-adjusted analyses, the following associations were significant for any synovitis (versus none) at 30-month: baseline ICAM-1, OR 1.9 (1.1, 3.4) and baseline IGF-1, OR 1.9 (1.1, 4.1). Adjusted for all the variables above and use of pain medications, baseline TNF-alpha was significantly associated with prevalent knee pain at 30-months, OR 2.0 (1.2, 3.4) and ICAM-1 was associated with incident knee pain, OR 0.3 (0.1, 0.97).


In this hypothesis-generating study using a longitudinal cohort of Caucasian women, we found several biomarkers predictive of synovitis, prevalent and incident knee pain at 30-month follow-up.

To cite this abstract, please use the following information:
Singh, Jasvinder, Neogi, Tuhina, Felson, David T., Torner, James, Baker, Kristin, Nevitt, Michael C., et al; Analysis of the Multicenter Osteoarthritis (MOST) Study. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :951
DOI: 10.1002/art.28718

Abstract Supplement

Meeting Menu