Abstract
Aim
Local antibiotics, delivered to the site of infection, achieve high tissue concentrations and are used as an adjunct to systemic therapy. Local gentamicin provides levels well above the minimum inhibitory concentration and may be sufficient on its own, however, the efficacy of single or combination local antibiotics has not been studied. This retrospective study evaluated the effect of combination aminoglycoside and vancomycin local antibiotic treatment compared to aminoglycoside alone in the surgical management of bone infection.
Method
We studied patients with microbiologically confirmed osteomyelitis and fracture-related infection, who had implantation of antibiotic carriers as part of their surgical management. Data including patient demographics, type of surgery, microbiological characteristics, BACH score, duration of antibiotic treatment and clinical outcomes were collected. Failure of therapy was a composite of recurrence of infection, continued or new antimicrobial therapy, or reoperation with suspected or confirmed infection at one year after index surgery.
Results
There were 266 patients who met the inclusion criteria. Nine patients died before the outcome endpoint at 12 months and five patients were lost to follow up so were excluded. 252 patients were included in the final analysis and were well matched with regard to demographics, BACH score and microbiology. 113 patients had treatment with aminoglycoside alone and 139 patients had combination aminoglycoside and vancomycin.
There was no difference in the failure rate between groups; 10/113 (8.8%) in the aminoglycoside alone and 12/139 (8.6%) in the combination group, p = 0.934. There was no difference for reoperation, ongoing suppressive antibiotic use, or clinical suspicion of infection. Multivariate analysis showed that there was no added benefit of combination therapy (OR 1.54: 95%CI 0.59-4.04, p=0.38). BACH score and low BMI were associated with increased risk of failure (BACH OR 3.49: 95%CI 1.13-10.76, p=0.03; Low BMI OR 0.91: 95%CI 0.84-0.99, p-0.037).
The form of the carrier material (pellets or injectable paste) had no effect on failure rate (p=0.434). Aminoglycoside resistance (confirmed and presumed) occurred in 39/113 (34.5%) of the aminoglycoside only group and 36/139 (25.9%) of the combination group (p=0.137). The presence of aminoglycoside resistance had no effect on failure rate (OR 0.39: 95%CI 0.05-3.01, p=0.37).
Conclusions
Clinical outcome was not improved by the addition of vancomycin to aminoglycoside alone as local therapy for the management of osteomyelitis and FRI. Laboratory measured resistance, using currently accepted breakpoints, may not be relevant in local therapy.
