Biochemical Markers for the Early Identification of Osteoarthritis: Systematic Review and Meta-Analysis

There is a desperate need for the reliable detection of osteoarthritis (OA) at the early stage when patients are likely to benefit most from disease interventions. A variety of biochemical markers have been proposed, but their reliability varies among studies.

In this review, we aimed to answer the following questions: (1) are there biochemical markers that are differentially expressed in early OA versus healthy subjects, and (2) if so, what is the diagnostic value of these biomarkers for early OA?

Embase, PubMed, and Web of Science were searched to obtain all relevant studies up to March 2018, and studies comparing the biochemical markers between early OA and healthy controls were selected. The Downs and Black checklist was used to assess the risk of bias. Biomarkers that were investigated in five or more different populations were pooled for meta-analysis. A meta-regression analysis was performed to explore possible explanations for the heterogeneity of studies.

In total, 26 articles met the criteria for the qualitative synthesis and 17 articles for the final quantitative synthesis. N-terminal crosslinked telopeptide of type I collagen (NTX-I) was the only biomarker found to be differently expressed in patients with early OA versus controls, without significant heterogeneity among studies (I² = 0%, [Formula: see text] = 1.695, p = 0.792). The meta-regression analysis identified that sample size and affected joint possibly explained the heterogeneity among studies.

Although a wide range of biomarkers has been previously investigated in early OA, the diagnostic value of these biomarkers could not be determined because due to a low number of studies regarding any given biomarker. Large prospective and adequately powered studies are therefore required to validate these (and other) biomarkers for identifying early OA.