Criteria for diagnosis of infected internal fixation implants at the time of conversion to total hip arthroplasty (THA) are not clear. The purpose of this study is to identify risk factors for infection in patients undergoing conversion to THA.
We retrospectively reviewed patients at a single institution who underwent conversion to THA from 2009 to 2014. Patients were diagnosed with infection preoperatively using Musculoskeletal Infection Society criteria or postoperatively if they were found to have positive cultures intraoperatively at the time of conversion surgery. Medical comorbidities and preoperative inflammatory markers were compared between infected and noninfected groups. Univariate and multivariate logistic regression analysis were performed to identify independent risk factors for infection. Receiver operating characteristic curves were generated to determine test performance of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). A post hoc power analysis was performed.
Thirty-three patients were included in the study. Six patients (18%) were diagnosed with infection. We found no association between comorbidities and infection in this cohort. The mean ESR and CRP were higher in infected (ESR = 41.6 mm/h, CRP = 2.0 mg/dL) vs noninfected (ESR = 19.3 mm/h, CRP = 1.3 mg/dL) groups (both P < .01). ESR >30 mm/h (odds ratio 28.8, 95% confidence interval 2.6-315.4, P = .001) and CRP >1.0 mg/dL (odds ratio 11.5, 95% confidence interval 1.6-85.2, P = .01) were strongly associated with infection. Receiver operating characteristic curves for ESR (area under the curve [AUC] = 0.89) and CRP (AUC = 0.89) demonstrated good fit.
We report a high incidence of infection in patients who underwent conversion to THA. Preoperative ESR and CRP are effective screening tools though occult infections may still be missed. Patients with borderline or elevated inflammatory markers should raise strong suspicion for infection.