Duration of postoperative antibiotic therapy following revision for infected knee and hip arthroplasties.

Abstract

In the period from 1990 to 1998 28 patients (14 males and 14 females) were treated for infected hip (19 cases) and knee (9 cases) arthroplasties. The median age at index operation was 72 y (range 34-82 y) and at revision surgery 74 y (range 36-83 y). The primary diagnosis was osteoarthritis (16 cases), rheumatoid arthritis (5 cases), failed femoral neck fracture (6 cases) and arthritis secondary to congenital hip dislocation (1 case). Nine cases were infected early, i.e. within 3 months postoperatively, and 19 had late infections. The median interval from index operation to diagnosis of the infected arthroplasty was 18 months (range 1-156 months). Postoperatively, all patients received antibiotics based on susceptibility studies of the causative organism according to culture specimens. The erythrocyte sedimentation rate and CRP level were controlled twice a week during hospitalization and antibiotics were continued until normalization of CRP. The indication for the shift from intravenous to oral antibiotics was a rapid fall in CRP. Antibiotic therapy was terminated when CRP was close to or below 10 mg/l and there was evidence of clinical recovery. Based on the screening of ESR and CRP in this study we believe that CRP is a valuable parameter in deciding when to stop antibiotic therapy. However, it is not clear whether a shorter treatment period is effective or not.