Abstract
BACKGROUND: Emerging knowledge has highlighted the role of periostin (POSTN) in osteoarthritis (OA) process; however, whether POSTN is suitable as a biomarker of OA remains unclear. This study aimed to investigate the potential value of POSTN as a biomarker of OA. METHODS: Ten 6-month-old female Sprague-Dawley (SD) rats were used in this study. Five rats underwent ovariectomy (OVX) operation and the others were carried out sham operation. Thirty-two patients with OA and eighteen patients who had meniscus injuries or ligament injuries but with intact articular cartilages were recruited in this study from January to July 2019 at the Peking University International Hospital. We first detected the expression of POSTN in the cartilage of OVX induced OA rats and different compartments of the knee joint in patients with OA using immunohistochemistry. Besides, serum POSTN levels in patients with or without OA were examined using enzyme-linked immunosorbent assay (ELISA). The associations among serum POSTN levels, clinical symptoms, and radiological severity were assessed according to the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores and, Kellgren-Lawrence (KL) grading, respectively. Finally, multivariable cumulative link models were established to evaluate the validity of serum POSTN level as a predictor of knee OA. RESULTS: The significantly higher POSTN expression was found in OVX-OA rats than Sham rats, while, the expression of POSTN was significantly higher in the torn cartilage of patients with OA. However, the serum POSTN level did not differ significantly between patients with and without OA. Additionally, we found no remarkable associations between serum POSTN level and WOMAC scores and KL grading. Subsequent analysis revealed that serum POSTN was not a significant predictor of OA. CONCLUSION: Thus, although POSTN may be involved OA process and local POSTN is valuable in disease diagnosis and distinguishing of the severity of disease, its serum level is not sufficient to serve as a candidate biomarker of OA given the current analysis technology.