External Validation of a Predictive Score for Fracture-Related Infections in Orthopedic Trauma Surgery.

Abstract

Background: The purpose of this study was to externally validate a predictive score for fracture-related infections, establishing generalizability for absolute and relative risk of infection in the setting of orthopedic fracture surgery.

Materials and methods: This was a retrospective, case-control study performed at a level I academic trauma center that included 147 patients with fracture-related infection in the study group and 300 control patients. We analyzed the same 8 independent predictors of fracture-related infection cited by a previous study. We then used the area under the receiver operating characteristics curve (AUC) to compare the derivation and validation cohorts. The validation and derivation cohorts were then compared by grouping patients into 4 strata of Wise score groups. This allowed for comparison of AUC and risk of fracture-related infection in our institution with those in the previously studied institution.

Results: The resulting data yielded an AUC (0.74) nearly identical to that of the previously studied institution. It was also found that the relative risk of infection correlated with the Wise score in the same way the initial model did with the absolute risks being similar.

Conclusion: The previous predictive model was externally validated and shown to be generalizable to a different patient population. The relative risk of a fracture-related infection can be determined using this scoring model preoperatively with the goal of aiding in patient counseling and surgical decision-making, giving a quantitative value to patient risk factors. [Orthopedics. 202x;4x(x):xx-xx.].