Objective: To analyze the factors affecting the prognosis of patients with knee osteoarthritis, and to construct a nomogram prediction model in conjunction with multi-dimensional clinical indicators.
Methods: The clinical data of 234 patients with knee osteoarthritis who were treated in our hospital from January 2015 to June 2021 were retrospectively analyzed, including 126 males and 108 females；age more than 60 years old for 135 cases, age less than 60 years old for 99 cases. Lysholm knee function score was used to evaluate the prognosis of the patients, and the patients were divided into good prognosis group for 155 patients and poor prognosis group for 79 patients according to the prognosis. The clinical data of the subjects in the experimental cohort were analyzed by single factor and multiple factors. The patients were divided into experimental cohort and verification cohort, the results of the multiple factor analysis were visualized to obtain a nomogram prediction model, the receiver operating characteristic curve（ROC）, calibration curve and decision curve were used to evaluate the model's discrimination, accuracy and clinical benefit rate.
Results: The results of multivariate analysis showed that smoking, pre-treatment K-L grades of Ⅲ to Ⅳ, and high levels of interleukin 6 （IL-6） and matrix metallo proteinase-3 （MMP-3） were risk factors for the prognosis of patients with knee osteoarthritis. ROC test results showed that the area under the curve of the nomogram model in the experimental cohort and validation cohort was 0.806[95%CI（0.742, 0.866）] and 0.786[（95%CI（0.678, 0.893）], respectively. The results of the calibration curve showed that the Brier values of the experimental cohort and verification cohort were 0.151 points and 0.134 points, respectively. When the threshold probability value in the decision curve was set to 31%, the clinical benefit rates of the experimental cohort and validation cohort were 51% and 56%, respectively.
Conclusion: The prognostic model of patients with knee osteoarthritis constructed based on multi-dimensional clinical data has both theoretical and practical significance, and can provide a reference for taking targeted measures to improve the prognosis of patients.
Keywords: Knee osteoarthritis; Multivariate analysis; Nomogram model; Prognosis.