Background: Increasing evidence suggests an association between Modic changes (MC) and subclinical infection and inflammatory reactions. However, the relationship between preoperative MC and surgical site infection (SSI) has not been fully explored. This study aims to investigate the correlation between MC and SSI.
Methods: A retrospective analysis was conducted on patients (n = 646) who underwent single-level lumbar spine surgery for lower back pain in our hospital between 2018 and 2023. According to the Centers for Disease Control and Prevention (CDC) criteria, the patients were divided into an SSI group (n = 40) and a Non-SSI group (n = 606). Univariate analysis was performed to determine the statistical differences in variables between the two groups, and the variables with significant differences were included in a multivariable logistic regression analysis to identify independent risk factors for SSI. Receiver operating characteristic (ROC) curve analysis was performed on the independent risk factors.
Results: The SSI group and the Non-SSI group exhibited significant differences in diabetes prevalence, MC prevalence, Total endplate score (TEPS) and area ratio of MC (P < 0.05). Age, gender, American Society of Anesthesiologists(ASA)score, hypertension, coronary heart disease (CHD), chronic obstructive pulmonary disease (COPD), MC classification, and the location of MC in the endplate showed no significant differences (P > 0.05). Multivariate binary logistic regression analysis was performed on the variables with significant differences, and the results indicated a significant correlation between TEPS (P = 0.009) and the area ratio of MC changes (P = 0.001) with SSI. ROC curve analysis was performed on the TEPS and area ratio of MC changes, and the results showed that the diagnostic value of TEPS (AUC: 0.641; CI: 0.522-0.759) is lower than the area ratio of MC (AUC: 0.722; CI: 0.621-0.824), and the combined diagnosis did not significantly improve the diagnostic value (AUC: 0.747; CI: 0.653-0.842). The area ratio of MC had moderate diagnostic value for SSI (AUC: 0.722; CI: 0.621-0.824), with a cut-off value of 24.62% determined by the Youden index (sensitivity: 69.2%; specificity: 73.1%), and for every 1% increase in the area ratio of MC changes, the risk of SSI in MC patients increased by 10.3% (OR = 1.103; CI: 1.044-1.167).
Conclusion: The area ratio MC and the TEPS are independent risk factors for SSI after lumbar spine surgery. The predictive value of the area ratio of MC is greater than TEPS, and when the two are combined, the predictive value is not significantly improved. When the rate of MC exceeds 24.62%, caution should be exercised regarding the occurrence of SSI.
Keywords: Lumbar spine surgery; Modic changes; Predictive diagnosis; Surgical site infection; Total endplate score.
© 2024. The Author(s).