For patients with knee osteoarthritis (OA), urine C-telopeptide of cross-linked collagen type II (uCTX-II) levels are associated with radiographic severity of OA and pain intensity, according to results published in Arthritis Research & Therapy.

The study included data from 2 randomized clinical trials of participants with radiographic OA and presence of pain (n=2206). The researchers performed a post hoc analysis, as well as an analysis of a subgroup with available urine samples and evaluable radiographs for both knees (n=1241). To assess the contribution of joints at different stages, including some patients with rheumatoid arthritis, they used urine CTX-I, urine CTX-II, and serum osteocalcin analysis to determine associations with combined Kellgren-Lawrence (KL) scores, sex, and pain for both knees.

Related Articles

Of all participants, 3.6% had unilateral knee OA; 68.2% had bilateral, early knee OA; 7.5% had bilateral, late knee OA; and 19.7% had bilateral knee OA of mixed stages. For participants with bilateral knee combinations of KL grades 3/4, 3/3, and 2/4, mean uCTX-II concentration was significantly higher compared with those with KL grades 0/2.

The results indicated that pain, body mass index, age, sex, and KL grade were significantly associated with uCTX-II. All these associations remained significant after adjusting for covariates. However, only weight-bearing pain remained significantly associated with CTX-II, whereas non-weight-bearing pain did not.

With higher radiographic severity of each knee, the researchers found that the level of uCTX-II incrementally increased. Levels of bone markers CTX-I and osteocalcin were significantly associated with body mass index and sex. However, neither CTX-O nor osteocalcin was associated with radiographic severity. Compared with male participants with identical KL scores, female participants in some KL score groups had higher biomarker levels.

The study had several limitations, including its cross-sectional design. The researchers noted that their results are limited by the lack of radiographic assessment of nonknee joints.

“The data suggest that biomarker differences between genders should be taken into account when evaluating these markers in the context of structural features of OA,” the researchers wrote.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Bihlet AR, Byrjalsen I, Bay-Jensen A, et al. Associations between biomarkers of bone and cartilage turnover, gender, pain categories and radiographic severity in knee osteoarthritis [published online September 3, 2019]. Arthritis Res Ther. doi:10.1186/s13075-019-1987-7