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.

Racial/Ethnic and Sex Differences in Somatosensory Abnormalities in Subjects with or at High Risk of Knee Osteoarthritis: Preliminary Analysis of the Multicenter Osteoarthritis (MOST) Study.

Singh2,  Jasvinder, Jingbo5,  Niu, Freylaw9,  Laura, Scholz6,  Joachim, Arendt-Nielsen1,  Lars, Lewis8,  Cora E., Woolf6,  Clifford

Aalborg University, Aalborg, Denmark
Birmingham VA Medical Center and University of Alabama, Birmingham, Minneapolis, MN
Boston Univ Schl of Med, Boston, MA
Boston University School of Medicine, Boston, MA
Boston University School of Medicine
Childrens Hospital and Harvard School of Medicine
UCSF, San Francisco, CA
University of Alabama, Birmingham
University of Iowa


There may be racial/ethnic and sex differences in the pain experience of knee OA, which may be reflected in differences in somatosensory evaluations. Our objective was to assess differences in somatosensory findings between Caucasians and African-Americans (AA), and between men and women.


MOST is a cohort study of persons aged 50–79 years of age, with or at high risk of knee OA. At the 60-month clinic visit, participants underwent knee radiography, answered pain questionnaires, and had the following somatosensory assessments: static allodynia using a 26g monofilament, and hyperalgesia using a pin. Temporal summation (TS), an augmented pain response to repetitive mechanical stimuli thought to reflect central sensitization, was also assessed. These assessments were carried out separately over each patella and wrist. An abnormal response to 26g or pinprick was defined as the participant experiencing pain or a non-response on 3 out of 4 trials, respectively. TS was defined as being present when, after touching the skin with a 60g monofilament repeatedly at a frequency of 1Hz for 30 seconds, the subject reported increased pain or new pain at the site being tested. Univariate and multivariate-adjusted prevalences (age, BMI, clinic and race and sex, as applicable) of the somatosensory tests were obtained for each site and summated as being present overall at any site, stratified by race/ethnicity and gender.


Of the 770 with somatosensory data, there were 652 Caucasians (438 women, 214 men) and 118 AAs (73 women, 45 men). Mean age, BMI and K-L grade was similar for all groups (Table 1).

Table 1. Clinical and demographic characteristics of study cohort

  University of Alabama (n = 371) University of Iowa (n = 399)
All subjectsWhite men (N = 125)White women (N = 128)Black men (N = 45)Black women (N = 73)White men (N = 89)White women (N = 310)
Age, mean (SD), year65.9 (8.3)68.1 (7.2)62.8 (7.0)65.2 (8.0)71.2 (7.9)68.1 (7.9)
BMI, mean (SD), kg/m230.8 (5.3)28.5 (5.4)31.3 (6.0)32.6 (5.8)30.3 (5.8)30.2 (6.3)
K-L grade N (%)      
    0–168 (54%)57 (45%)21 (47%)26 (36%)43 (48%)159 (52%)
    214 (11%)28 (22%)6 (13%)15 (21%)16 (18%)68 (22%)
    3–443 (34%)43 (34%)18 (40%)32 (44%)30 (34%)82 (27%)

Adjusted prevalence of static allodynia with 26g was higher in African-American women and almost achieved statistical significance (p=0.06) (Table 2). Adjusted prevalences of the pin-prick and temporal summation somatosensory evaluations were not significantly different among the sex and racial groups.

Table 2. Age, BMI- and clinic-adjusted prevalence rates (95% CI) of pain and Temporal summation (any site, person-based)

 White menWhite womenBlack menBlack women 
 Prev (%)95% CIPrev (%)95% CIPrev (%)95% CIPrev (%)95% CIp-value
26 ga3.31.65.23, 74.9-3, 1214.46.230.06
Pinprick6.63, 107.55, 109.6-1, 2091, 170.91
Temporal summation62.756, 6957.753, 6266.554, 7964.654, 750.45
a Model was adjusted for sex, race, BMI and clinic site; additional adjustment for age not done since model became unstable and age was not significant for this test.
P-value is from an Analysis of variance

Adjustment for presence/absence of osteoarthritis did not impact the results for the pinprick and temporal summation tests; however this model did not converge for static allodynia for 26g.


African-Americans and women had greater prevalence of somatosensory abnormalities compared with Caucasians and men, but most of these differences were not statistically significant. This preliminary study is the first to demonstrate racial/ethnic difference in response to the 26g von Frey hair stimulus in patients with knee OA. Further exploration regarding factors that may contribute to such differences is warranted.

To cite this abstract, please use the following information:
Singh, Jasvinder, Jingbo, Niu, Freylaw, Laura, Scholz, Joachim, Arendt-Nielsen, Lars, Lewis, Cora E., et al; Racial/Ethnic and Sex Differences in Somatosensory Abnormalities in Subjects with or at High Risk of Knee Osteoarthritis: Preliminary Analysis of the Multicenter Osteoarthritis (MOST) Study. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1557
DOI: 10.1002/art.29323

Abstract Supplement

Meeting Menu