In the November 18, 2015 edition of JBJS, Lawing et al. present a well-documented cohort study comparing the outcomes of open-fracture management with local administration of aminoglycoside antibiotics plus systemic antibiotics, versus systemic antibiotics alone. The impact of this intervention on the ultimate rate of deep infection is eye-catching. The deep-infection rate in the local-antibiotic group was 6%, compared to 14.2% in the control group (p = 0.011). Moreover, locally administered aminoglycosides did not have a negative impact on nonunion rates, as one might expect due to the osteocyte toxicity reportedly associated with some aminoglycosides.
There are, however, issues of administrator bias with this study, because the use of local antibiotics was based on attending-surgeon preference. In addition, surgeons make other individual judgments about open-fracture management, such as debridement technique, that were not controlled for in this study. We also went through a period of using local antibiotic drips and catheter pumps in the 1990s that did not seem to yield reproducible results.
Lawing et al. conclude with the hope that their study “will provide support for future prospective, blinded, and randomized trials” focused on this intervention. I believe the data here are compelling enough for one of our trauma clinical-trials networks to plan and conduct an adequately powered trial complete with prospective criteria and blinded outcome adjudication. One reason we publish cohort studies in The Journal is to stimulate just that sort of response in the orthopaedic-research community. It is my hope that within a few years, JBJS editors will be reviewing an RCT manuscript that completes the investigative cycle on this important clinical question.
Marc Swiontkowski, MD