Background/Purpose: Inflammation plays a role in OA, and catabolic cytokines including IL-1β potentiate OA joint space narrowing. Elevated serum urate (sU) levels promote crystal-induced stimulation of IL-1β production by the NLRP3 inflammasome, potentially contributing to OA incidence and/or progression. In particular, patients with medial compartment knee OA and elevated sU levels may be at higher risk for progression of joint space narrowing than patients without elevated sU levels. Additionally, intraarticular urate concentrations have been associated with radiographic knee OA (RKOA) severity. However, there is limited research on associations of asymptomatic hyperuricemia (AH) and knee OA outcomes. Therefore, we sought to examine the association of AH with RKOA and symptomatic RKOA (sRKOA) in a large nationally representative sample (NHANES III). We also examined whether BMI modifies the association between AH and RKOA.
Methods: NHANES III was a cross-sectional health examination survey conducted between 1988 and 1994. It used a multistage, stratified probability cluster design to select a representative sample of non institutionalized civilian in US, and included data on sUA, gout, clinical and radiographic knee OA. We analyzed data (n=2213) for adults over age 60, excluding individuals with self-reported gout. Hyperuricemia was defined as serum urate > 6.8, mg/dL. One non-weight bearing AP knee X-ray was performed with RKOA defined as Kellgren Lawrence grade ≥ 2, and sRKOA as RKOA plus pain in the affected joint on most days for the prior 6 weeks. Wald chi-square tests were used to examine differences in proportions between different study characteristics. Multivariate log binomial models were used to examine the association between AH and knee OA outcomes and estimate prevalence ratios (PRs) and 95% confidence intervals (CIs).
Results: Among US adults age 60 years and older, prevalence of AH was 17.9% (CI 15.3-20.5). AH prevalence was significantly greater among men vs women (24.5% vs. 13.3%, p=<0.01), and obese (BMI>30) vs non-obese patients (27.4% vs. 14.8%, p=<0.01). Prevalence of RKOA was 37.7% overall (CI 35.0-40.3), and was significantly greater in women vs men (42.1% vs. 31.3%, p=0.01). The prevalence of RKOA was highest among subjects with greater age, obesity, non-Hispanic Black race, and less education. RKOA prevalence among patients with AH was 44.0% vs 36.3% for those with normuricemia (p = 0.056). Importantly, sRKOA was significantly higher in the AH group (17.4 vs 10.9%, p=0.04). After adjusting for age, sex, race, and education, patients with AH were more likely to have both RKOA (PR = 1.26, 95% CI: 1.06, 1.36) and sRKOA (PR = 1.69, 95% CI: 1.19, 2.42). These associations remained for non-obese adults, but were severely attenuated among obese adults, suggesting that obesity status may modify the association between AH and knee OA.
Conclusion: We identified a greater prevalence of RKOA among patients with AH, along with a greater prevalence of sRKOA, suggesting that urate may participate in OA pathogenesis. This association appeared to be modified by obesity status, with non-obese but not obese adults experiencing a greater prevalence of knee OA among participants with AH. Longitudinal studies are needed to verify these findings.