Study on the Prediction Model of Wound Infection After Spinal Fusion and Internal Fixation Based on Logistic...

Area of Science:

• Spinal surgery
• Infectious disease epidemiology
• Biostatistics

Background:

• Spinal problems increasingly affect both older adults and adolescents.
• Minimizing postoperative wound infection after spinal fusion and internal fixation is a significant clinical challenge.
• Personalized treatment plans are essential for optimizing patient recovery and reducing infection risk.

Purpose of the Study:

• To develop a predictive model for postoperative wound infection risk following spinal fusion and internal fixation.
• To identify key clinical variables associated with the incidence of postoperative infection.
• To inform individualized patient management strategies and enhance surgical outcomes.

Main Methods:

• Logistic regression, SMOTE algorithm, and confusion matrix were employed for predictive modeling.
• Analysis of clinical data from 449 patients undergoing spinal fusion and internal fixation.
• Inclusion of 14 variables, including age, fixation segments, medical history, and intraoperative factors.

Main Results:

• Age, number of internal fusion fixation segments, preoperative hospitalization duration, cerebrospinal fluid leakage, and preoperative ASA score were identified as significant predictors of postoperative infection.
• The study established a classification model to differentiate between postoperative infection and non-infection outcomes.
• Statistical analysis confirmed the importance of these factors in assessing infection probability.

Conclusions:

• The study successfully modeled the probability of infection after spinal fusion and internal fixation.
• Identified factors provide a basis for risk stratification and targeted preventive measures.
• Findings support the development of individualized care plans to minimize postoperative infection rates in spinal surgery patients.