Descriptions of debridement for treatment of open fractures are at least as old as The Iliad1, but the question of the timing of debridement is still unresolved after more than a century of study. The origins of the “six-hour rule” are in animal studies, and even today much of what we know (and think we know) about the timing of debridement and antibiotics and about the relation of such timing to the risk of infection is the result of investigations involving rabbits and rats. The most recent animal studies of the timing of antibiotics, timing of lavage, and timing of debridement still suggest that these are each important in reducing infection when considered separately2, but even these three relatively easy-to-control variables have not been studied in combination, nor have the added variables that would mimic the reality of our modern Emergency Medical Services system, emergency departments, trauma centers, and personnel trained in fracture assessment and stabilization, early use of antibiotics, wound lavage, dressing care, and triage.
In 2010, a review of 315 patients by the Lower Extremity Assessment Project (LEAP) Study Group found that the time from injury to operative debridement was not a significant predictor of infection risk3. The authors of the current study conducted a systematic review and meta-analysis of human studies of open long-bone fractures to determine whether a much larger cohort could establish an association between time to initial debridement and development of infectious complications. They found sixteen studies with appropriate data on debridement timing and infection rates to include in a systematic review. The challenges that they encountered in this process should remind us of the difficulty of studying the question at hand. The cutoff used to distinguish “early” and “late” debridement ranged from five to twelve hours. Only fourteen studies provided early and late debridement times and infection rates. No studies were randomized. Three studies failed to clearly describe the antibiotic intervention. Some studies involved a cohort of upper and lower-extremity fractures, some involved only femora, and one characterized fractures as either “tibial” or “non-tibial.” It is of considerable credit to the authors’ skills, as displayed in their excellent results section, that they were able to sort this out in a meaningful way. Ultimately, although there was a wide divergence in overall infection rate among studies—from 4% to 63%—the authors found no significant overall difference in infection rate between early and late debridement. This was true regardless of whether the cutoff time was set at five, six, eight, or twelve hours. The results were also statistically nonsignificant when considering solely deep infections or more severe open fractures.
The authors do a good job of reminding us of the limitations of this study. In particular, a large proportion of the data came from retrospective studies. However, the differing interpretations and definitions of key measures and factors (not only “early” and “late” debridement but the timing of antibiotic administration, definition of infection, and irrigation methods) also created consideration hurdles for the investigators. The strengths of the study are the large number of fractures that could be evaluated on the basis of time to operative debridement and the fact that the numbers were large enough to permit subgroup analyses to be conducted. As the authors recognize, consideration of all of the data ultimately tells us that, although we don’t know definitively that the six-hour rule is incorrect, we see no data supporting it either, even among patients with the most severe injuries.
The real values of the study, in this commentator’s opinion, are threefold. First, the authors offer a reasonably effective challenge to our conventional wisdom without recommending that we abandon the wisdom of experience. Second, they challenge the conventional wisdom that “operating early” frequently has to mean “operating in the middle of the night,” when we might not have the best surgical team or the patient might not be optimized for surgery; this gives me some comfort, given the changes in resident duty hours, crew resource management, checklists, and electronic medical records. Finally, in describing the shortfalls in data collection and the differing approaches to recording data, they demonstrate many of the challenges that any investigator will face in producing a coherent and valid result. Given the emergency nature and severity of many of these fractures, and given how much fracture care has changed in the last 150 years, it is easy to understand why the large, prospective, controlled study that would perhaps provide more definitive information has yet to be conducted. In fact, the authors do such a good job that it makes me wonder whether the definitive Level-I study can ever be done because of all of the variables that we presently know of and how rapidly we continue to find and improve other variables that affect our patients’ outcomes (such as local and systemic hypoxia and hypothermia due to shock and multiple injuries; the comorbidities of diabetes, renal failure, drug use, and smoking; other medications; and even genetic factors). It has become trite to end systematic reviews of the literature with the phrase “More studies are needed,” but the authors effectively demonstrate that this is the case while not making the recommendation that we do away with the “parachute” of “debridement as early as possible,” because a study of deliberately delayed debridement can’t be performed. They should be congratulated for setting a clearer path for the work that is yet to be done.