Presence of erosions is not a risk factor for the development of knee OA in a hand OA population: Results from the Framingham OA study

Abstract

Objective

This study aimed to investigate whether there is a difference in risk of knee OA between individuals with erosive hand osteoarthritis (OA) compared to non-erosive hand OA, and to identify other risk factors for knee OA presence and development in hand OA patients.

Methods

Subjects were selected from the Framingham OA study's Offspring and Community cohorts. Bilateral knee and hand radiographs were scored. Generalized linear models compared knee OA prevalence and incidence rates among hand OA-free, non-erosive and erosive hand OA. Multivariable logistic regression identified risk factors for symptomatic knee OA (presence of hand OA, anatomical location, erosions, number of affected finger joints, radiographic grade changes), adjusted for age, sex and BMI.

Results

In total, 2367 participants were studied. Hand classification (hand OA-free, non-erosive and erosive hand OA) significantly impacted the prevalence of knee OA (P ​< ​0.0001). Prevalence rates at baseline were 6.3 ​%, 17.9 ​% and 26.8 ​% for radiographic knee OA and 2.9 ​%, 9.7 ​% and 12.7 ​% for symptomatic knee OA for hand OA-free, non-erosive and erosive hand OA respectively. Post hoc analysis indicated differences were primarily between hand OA-free and hand OA groups (p ​< ​0.001), with no significant differences between erosive and non-erosive hand OA (p ​= ​0.7 and p ​= ​0.8). Overall, hand OA was identified as a risk factor for knee OA development. Presence of erosions did not increase risk, but the number of affected finger joints and changes in Kellgren-Lawrence grade did.

Conclusion

Hand OA is associated with knee OA presence and incidence with no difference in the risk of knee OA between those with non-erosive vs erosive hand OA.