The association of osteoarthritis risk factors with different patterns of knee pain

Raducha-Grace L, Boudreau R, Hannon M, Newman A, Chu C, Nevitt M, Kwoh CK. The association of osteoarthritis risk factors with different patterns of knee pain. Paper presented at: American College of Rheumatology Annual Scientific Meeting; 2008 Oct 26; San Francisco, CA.

Purpose: Patients with knee osteoarthritis describe varying patterns of knee pain, including localized pain, regional pain and diffuse pain. The factors associated with these different patterns of knee pain may vary. We examined the relationship between risk factors for knee osteoarthritis and patterns of knee pain. Methods: The Knee Pain Map is an interviewer-administered assessment that asks subjects to characterize their knee pain as localized, regional, or diffuse. Localized pain is defined by the subject’s use of one or two fingers to point to a specific location, regional pain by the use of the whole hand to cover a more extensive region, and diffuse pain as unable to localize or regionalize. A total of 1177 participants were studied from the Osteoarthritis Initiative (OAI), a community based cohort of men and women, ages 49-75, with symptomatic knee OA or with risk factors that increase their risk of developing knee OA. 800 of the participants had pain in one or both knees. OAI data release 3.0.0 was used. We used multinomial logistic regression modeling to examine the relationship between risk factors for OA and knee pain patterns, accounting for nesting of knees within an individual. Knees without pain were used as controls. Bivariate models were used for age, BMI, sex, race, family history of knee replacement, history of knee injury, history of knee surgery, and evidence of hand OA on exam. Risk factors with relative risk ratios with p-values < 0.15 for any pain pattern were included in the multivariate model. Results: We compared 1103 knees with pain to 1155 knees without pain. Among painful knees, localized pain was most common (70%) compared to regional (19%) and diffuse (11%) pain. On bivariate analysis using knees without pain as controls, older age, female sex, non-white race, BMI, knee injury, knee surgery, and hand OA were associated with one or more pattern of pain (local, regional or diffuse). In the final multivariate model (see table), older age, female sex, non-white race, injury, surgery and hand OA remained significant. When compared knees without pain, older age was associated with a significantly lower risk of local pain and hand OA was associated with a significantly lower risk of regional pain. Non-white race was significantly associated with local pain and female sex with local and regional pain. Conclusions: We have shown that specific OA risk factors are associated different knee pain patterns. Better understanding of the relationship between OA risk factors and knee pain patterns may help to characterize heterogeneous subsets of knee OA. Relative Risk Ratios in Multivariate Regression Model Local Regional Diffuse Older age (5-year age groups) 0.9, p < 0.01 0.9, p = 0.24 0.9, p = 0.13 Female vs. male 1.4, p = 0.02 2.1, p < 0.01 1.2, p = 0.53 Non-white race vs. white race 1.5, p = 0.02 1.8, p = 0.13 1.0, p = 0.94 + Knee injury 3.3, p < 0.01 4.6, p < 0.01 3.0, p < 0.01 + Knee surgery 2.6, p < 0.01 2.7, p < 0.01 2.0, p = 0.06 + Hand OA 0.9, p = 0.32 0.6, p = 0.04 0.9, p = 0.71