Predictive modeling for identifying infection risk following spinal surgery: Optimizing patient management

Area of Science:

• Orthopedic Surgery
• Infectious Disease Epidemiology
• Clinical Predictive Modeling

Background:

• Post-spinal surgery infection is a significant complication impacting patient prognosis.
• Predictive modeling can identify high-risk patients for optimized management.
• Scoliosis surgery carries a notable risk of all-cause postoperative infection.

Purpose of the Study:

• To investigate the occurrence of all-cause infection in patients undergoing scoliosis surgery.
• To develop and validate a predictive model for in-hospital infection following scoliosis surgery.

Main Methods:

• Retrospective analysis of 370 scoliosis surgery patients (January 2016 - October 2022).
• Comparison of clinicodemographic, surgical, and laboratory variables between infected and non-infected groups.
• Logistic regression used to build and internally validate an infection prediction model.

Main Results:

• All-cause in-hospital infection occurred in 17.8% (66/370) of patients.
• Key predictors identified: Sex, ASA classification, BMI, diabetes, hypertension, preoperative white blood cells, preoperative CRP, and surgery duration.
• The predictive model achieved an Area Under the Curve (AUC) of 0.776 on internal validation.

Conclusions:

• Dynamic nomograms incorporating identified variables may predict all-cause infection risk after scoliosis surgery.
• The developed model offers potential for real-time risk assessment and clinical decision support.
• This predictive tool could enhance pre- and postoperative management strategies for scoliosis patients.