Constructing a Diagnosis Model and Visualizing the Risk Relationship between Biomarkers and Overuse Injuries in Well-trained Wrestlers

This study aimed to investigate the association between biomarkers and overuse injuries in well-trained wrestlers. Seventy-six well-trained wrestlers on a national team completed two blood sample collections, two clinical overuse injuries diagnoses, and a questionnaire survey at a 2- week interval. Multivariate logistic regression analysis and receiver operating characteristic curve were used to screen for related factors and construct the prediction probability model of overuse injuries. Using a restricted cubic spline further clarifies the relationship between biomarker levels and the risk of overuse injuries. Creatine kinase (CK), cortisol, rheumatoid factor, testosterone in men, and C-reactive protein (CRP) levels in the overuse injuries group were significantly different compared to those in the non-overuse injuries group. The diagnostic efficiency of the prediction probability model was more valuable than any single variable (area under the curve = 0.96, Specificity = 0.91, Sensitivity = 0.89, high accuracy). A J-shaped relationship was noted between biomarkers (cortisol, CRP, and CK) and the risk of overuse injuries (cutoff point: 17.95 mu g.dL(-1), 4.72 mg.L-1, and 344 U.L-1; p for nonlinearity: < 0.001, 0.025, and 0.043, respectively). In conclusion, a predictive model based on biomarkers (cortisol, CRP, and CK) predicted the overuse injuries risk of well-trained wrestlers. High levels of these three biomarkers were associated with a higher risk of overuse injuries, and a J-shaped relationship was observed between them.