Abstract: Objective To explore the risk factors affecting early unplanned readmissions in the elderly patients undergoing total knee arthroplasty(TKA), and to construct a clinical prediction model in the form of nomogram. Methods A retrospective study was conducted to collect 919 elderly patients who underwent TKA for primary knee osteoarthritis at Northern Jiangsu People's Hospital from January 2020 to April 2022. All patients were divided into readmission group and control group according to whether they were readmitted within 30 days after discharge. Multivariate Logistic regression analysis was performed to screen the risk factors of readmission. And the risk factors were used to construct a clinical prediction model in the form of nomogram, and receiver operator characteristic(ROC) curve was plotted to evaluate the performance of the nomogram. Results The incidence rate of unplanned readmission within 30 days after discharge was 5.11%(47/919) in the elderly patients with TKA. Binary Logistic regression analysis showed age, body mass index(BMI), literacy, payment type, age-adjusted Charlson comorbidity index(AACI), operation time, intraoperative bleeding, and discharge destination were the independent risk factors for unplanned readmission within 30 days after discharge in the elderly patients with TKA, and a clinical prediction model was established in the form of a nomogram. The area under the ROC curve of the nomogram was 0.834(95%CI: 0.776-0.892), with a specificity of 78.7%, and a sensitivity of 78.0%. Conclusions The nomogram established in this study has good predictive efficacy for the readmission risk in the elderly patients with TKA.

Key words: osteoarthritis, total knee arthroplasty, aged, unplanned readmission, risk factors, nomogram, prediction model