Retrospective Analysis of Infection Factors in Secondary Internal Fixation after External Fixation for Open Fracture of a Long Bone: A Cohort of 117 Patients in a Two-Center Clinical Study

Abstract

Purpose. To investigate infection risk factors after secondary internal fixation (IF) of open fracture of a long bone with removed fixation frame and explore the safe range of feasible operation for abnormal inflammatory indicators. Methods. Clinical data of 117 cases of open fracture of a long bone that underwent temporary external fixation (EF) in one stage and IF in two stages were retrospectively analyzed. Collected data included age, sex, Gustilo type, multiple injuries, debridement time, duration of EF, needle infection, interval of conversion to IF after external fixator, preoperative white blood cell (WBC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), albumin (ALB), blood glucose, and prognosis. We selected these factors for univariate analysis of postoperative surgical site infection (SSI) and multivariate logistic regression analysis of statistically significant risk factors and created receiver operating characteristic (ROC) curves to compare the diagnostic efficiency of each index and determine the optimal screening point. Results. We followed up 117 patients, with 130 limbs affected. Univariate analysis showed that ESR, CRP, ALB, WBC, EF time, and Gustilo fracture type were significantly associated with SSI. Multivariate logistic regression analysis showed that CRP, duration of EF, and Gustilo fracture type were independently associated with postoperative infection. Area under ROC curves for WBC, ESR, and CRP were 69.7%, 73.2%, and 81.2%. Conclusions. We demonstrated the role of Gustilo classification of open fractures in predicting postoperative infection, especially for open fractures above type III. If the inflammatory indexes return to normal or show a downward trend, and the second-stage IF operation is performed within the cutoff values, postoperative recurrent infection was reduced.