Is nonoperative treatment of pediatric type I open fractures safe and effective?

Prompt surgical debridement is considered a mainstay for care of open fractures. Recent studies, however, have challenged this concept, proving that time to washout is not an independent risk factor for wound infections in open fractures [6, 7]. Multiple studies have shown that the timing, choice, and duration of antibiotics are important factors in the treatment of open fractures [7, 8, 9]. The benefits of any open surgical debridement of the bone under these minor skin violations of type I fractures, by contrast, have not been shown. The wounds of many type I open fractures occurring in the extremities are often small poke holes; thus, further exposure of these types of fractures in the operating room would likely lead to more periosteal stripping and devascularization. Although treating these type I open fractures nonoperatively cannot be proven superior given the current study design, it is the senior author’s opinion that, compared with operative treatment, nonoperative treatment decreases the potential morbidity associated with general anesthesia without worsening the infection rate.

Type I open fractures in the pediatric population differ from those in adults; in children, the fractures are bridged by a thick, vascular periosteum that facilitates fracture stability and healing. Pediatric fractures have not been as extensively studied as adult fractures, but several publications [2, 3, 10, 11, 12] have provided insight into the generally good prognosis in children. Luhmann et al. [10] reported on 65 pediatric patients with open forearm fractures treated operatively. Of the 65 fractures, 52 were type I, and 47 (90 %) of those had excellent to good results. The authors reported only one infection; it occurred in a 12-year-old with a type II open fracture. They did not find any statistically significant association between infection and fracture type or infection and time to surgical debridement. Yang and Eisler [12] studied 91 patients, 13 of whom were children, with type I fractures treated without operative irrigation and debridement. Those authors reported a 0 % infection rate with their nonoperative treatment of type I open fractures; however, 32 patients were later taken to the operating room for definitive treatment of their fractures; they did not specify how many of these patients were children. Iobst et al. [3] performed a retrospective review of 40 pediatric patients with type I open fractures in a variety of anatomic locations who were treated nonoperatively. None of the open wounds were closed primarily, and all were washed out with a povidone-iodine-saline solution at the bedside. In contrast to our study, all of their patients were admitted for intravenous antibiotics, whereas all of our patients were discharged home from the ED. The authors reported a 2.5 % infection rate. Doak and Ferrick [2] questioned whether patients with type I open fractures that were treated nonoperatively required admission for antibiotics. Their retrospective study consisted of 25 pediatric patients with type I open fractures who were discharged from the ED immediately or after 24 h of observation. They reported one case of a wound infection, and there was no delayed union or nonunion in any of their patients [2]. Patients spent an average of 7.7 ± 3.8 h in the ED during their workup and nonoperative management, a considerably shorter time than the 24 h spent by patients receiving intravenous antibiotics in other institutions. It is the senior author’s opinion that admitting the patient overnight simply to receive 24 h of antibiotics likely provides no benefit compared with receiving IV antibiotics in the ED and being discharged with oral antibiotics. In an evidence-based review, Pace et al. [11] compiled the data from the studies of Doak and Ferrick [2] and Iobst et al. [3] to make a level III recommendation for the nonoperative treatment of pediatric type I open fractures. However, they conceded that an eventual prospective level I or II study with sufficient power is still needed to make a solid recommendation for nonoperative treatment.

The decision to treat or not to treat a type I open fracture nonoperatively should still be based upon the clinical judgment of the pediatric orthopedic surgeon. Although our study shows that in most cases nonoperative treatment of type I open fractures of the forearm or tibia should be attempted, these results may not be generalizable to all patients. Operative treatment may be deemed necessary if there is worry about a concomitant compartment syndrome after high-energy trauma; if there is gross contamination after a farm accident involving dirt, pesticides, or animal feces; or if the underlying fracture pattern requires internal fixation.

Based on further literature review, another area where a level I or II study is needed is the topic of choice and duration of antibiotics for pediatric open fractures. In our study, 90 % of our patients received a dose of intravenous antibiotics in the ED, but only 70 % of the patients were discharged home with oral antibiotics, most commonly cephalexin. Although most of those patients were prescribed a 7-day course of antibiotics, there was a considerable variability, with duration ranging from 3–14 days. Lavelle et al. [13] conducted a web-based survey of academic orthopedic residency programs with regard to the treatment practices for pediatric open fractures. They found that 68 (97 %) of the 70 programs treated pediatric type I open fractures with a cephalosporin alone and that 87 % treated them with intravenous antibiotics for ≤48 h. Wound closure was also evaluated in this survey, and 90 % of programs closed the wounds in pediatric patients with type I open fractures. This finding is in contrast to our study where no patients had their wounds closed and there were no reported infections.

The major weakness of our study was the limited number of patients. The rate of infection after operative treatment of type I open fractures has been reported in the literature to be about 1.9 % [3]. To detect a 1 % increase in the rate of infection with nonoperative treatment, with a power of 0.8 and one-sided alpha of 0.05, the ideal study would need to enroll 3,210 patients in each arm. To detect a 2 % increase in the rate of infection with nonoperative treatment, the ideal study would need to enroll 997 patients in each arm. Thus, the number of patients required to enroll in randomized trials for level I evidence is quite large and will require concerted multi-institutional effort. If we pool the data from the Iobst et al. [3] and Doak and Ferrick [2] studies and combine it with our results, we find that of the 105 pediatric patients with type I open fractures treated nonoperatively among the three studies, the infection rate was 1.9 %, which is identical to that reported in the literature. In addition, there is always the possibility that patients may have experienced infections that were not documented and that the parents did not recall when they were contacted. Given the overall satisfaction of the patients at final follow-up, the presence of any latent infections was likely quite small.

In summary, nonoperative treatment of pediatric type I open fractures with subsequent discharge home from the ED appears to be safe; however, additional prospective, randomized clinical trials are needed to make a definitive level I recommendation regarding nonoperative management.