A Mathematical Prediction Model for Postoperative Infection Based on Logistic Multiple Regression Analysis in the Assessment of Surgical Outcome and Prediction of Infection in Elderly Spinal Fractures.

Investigating postoperative infection risk factors in elderly spinal fracture patients is crucial for optimizing surgical outcomes, improving patient safety, and guiding clinical decision-making in the management of these complex cases. To investigate the risk factors for postoperative infection in elderly patients with spinal fractures, with the goal of giving clinical care guidelines.From January 2019 to January 2022, 120 elderly patients admitted to our hospital for elective spinal fracture surgery were featured as the study subjects, and the patients were divided into infected and non-infected categories according to whether they had postoperative infection or not. A mathematical prediction model was built after using logistic multiple regression to investigate the parameters influencing postoperative infection of a spinal fracture.There were 20 patients in the infected category and 100 patients in the non-infected category. Univariate analysis showed that the proportion of patients in the infected category with male, age ≥65 years, multiple fractures, use of hormones or combined diabetes was notably higher than that in the non-infected category (P < .05). Male gender, age ≥65 years, multiple fractures, and certain medical conditions are independent risk factors for postoperative infection.Logistic regression analysis revealed that male, age ≥65 years, multiple fractures, use of hormones, or combined diabetes was an independent risk factor for postoperative infection in elderly patients with spinal fracture. Our study provides valuable insights that can guide clinical care and decision-making for elderly patients with spinal fractures. By applying these findings in practice, clinicians can refine their treatment strategies, improve patient outcomes, and enhance the overall quality of care provided to this vulnerable population.