Association Between Joint Inflammatory Markers and Pain in Knee Osteoarthritis Is Limited by Moderate Evidence

Authors of a systematic review evaluated the literature on the link between markers of inflammation and pain in patients with knee osteoarthritis.

The association between different inflammatory markers and pain in patients with radiographic knee osteoarthritis (OA) is limited by weak to moderate correlation data and conflicting to moderate quality of evidence, according to findings from a systematic review published in Osteoarthritis and Cartilage.

Knee OA is a chronic degenerative joint disease that causes disability and a poor quality of life due to musculoskeletal pain and functional limitations. Although treatment of musculoskeletal pain is key, the etiology of OA-related pain remains unknown.   However, there is evidence that histologic signs of synovial inflammation and changes associated with OA, such as synovial lining hyperplasia, sublining fibrosis, angiogenesis and effusion, are correlated with pain. Inflammatory cytokines, such as interleukin (IL)-6, IL-1b, and tumor necrosis factor (TNF)-α, have major proinflammatory activity in OA and may contribute to activation of innervating nociceptors and increase of C-reactive protein (CRP) levels.

The objective of the study was to systematically review the literature on the link between markers of inflammation and pain in patients with knee OA.

Study authors searched MEDLINE, Web of Science and EMBASE databases. Articles eligible for the current analysis were original interventional, observational, cross-sectional, and cohort studies; published in English, French, German, Dutch or Italian; included at least 80% of the study sample with clinical and radiographic primary knee OA, according to accepted diagnosing criteria (American College of Rheumatology [ACR] clinical and radiological criteria, Osteoarthritis Research Society International Atlas [OARSI], Kellgrene Lawrence [KL], or Ahlback’s criteria); and had data on the association between local signs of inflammation (ie, effusion/synovitis and Baker’s cyst) or signs of systemic inflammation (ie, serum or synovial CRP, interleukins, and TNF-α levels) that were assessed using validated tools (eg, visual analog scale [VAS], Numeric Rating Scale [NRS], Western Ontario and WOMAC, Knee Injury and Osteoarthritis Outcome Score [KOOS]) and experimental measures of pain (sensitivity).

A total of 10,681 articles are were identified from the initial search, with 37 articles recorded for inclusion. Two studies were longitudinal cohort data, 1 study implemented a case control design, and the remaining 34 reported cross-sectional data. A total of 25 cross-sectional studies, 2 cohort studies, and 1 case control study were evaluated as “high quality.” 

Majority of the included studies showing a positive significant association between inflammation and pain; however, the strength of such associations did not exceed the moderate level and had conflicting results. Nine studies investigated a possible association between effusion and pain using ultrasound, with 6 of these studies showing a positive association with pain and 3 not finding a significant association, therefore resulting in a conflict level of evidence. Seven studies that used contrast-enhanced magnetic resonance imaging (CE-MRI) showed a positive association between synovitis and pain, but of moderate quality evidence. Seven studies using non-CE-MRI techniques found a significant correlation with pain; however, 5 of these studies did not find a correlation, and thus, resulted in conflicting evidence.

Study authors also found conflicting quality evidence in the association between Baker’s cyst, a fluid-filled cyst posterior to the knee, and pain. One study showed that the prevalence of Baker’s cyst was higher in symptomatic patients compared with asymptomatic patients (39.2% vs 21.9%; P <.001). However, another study showed that patients had a 5.5-times (95% CI, 1-31.05) increased risk of experiencing pain with the presence of Baker’s cyst. A third study did not find any association at all.

Study authors also noted conflicting evidence in association between IL-6, TNF-α, and CRP, and pain. TNF-α was associated with pain in all, except 2, studies. Four studies indicated positive associations between IL-6 and pain, but 6 studies did not. Serum CRP levels were not correlated with pain, though a low-positive correlation was reported in the only study linking synovial fluid CRP concentrations with pain (r=0.22; 95% CI, 0.033-0.393).

Despite the comprehensive literature search and systematic review, the articles were limited by heterogeneity in radiographic evidence of knee OA, which excluded patients with early knee OA, those who were at increased risk of developing OA, or those who were only diagnosed by clinical criteria.

“Future cohort studies investigating the association between inflammation and pain in patients with knee OA can strengthen the level of evidence of such association and are encouraged to implement appropriate confounders in the analyses,” the authors of the review concluded.


Dainese P, Wyngaert KV, De Mits S, Wittoek R, Van Ginckel A, Calders P. Association between knee inflammation and knee pain in patients with knee osteoarthritis: a systematic review. Osteoarthr Cartil. Published online December 21, 2021. doi:10.1016/j.joca.2021.12.003