Retrospective analysis of risk factors for deep infection in lower limb Gustilo–Anderson type III fractures

BACKGROUND: Open fractures are among the most severe injuries observed in orthopedic patients. Treating open fractures is difficult because such patients with infections may require multiple operations and amputations. Furthermore, only a few studies have focused on antibiotic prophylaxis in open fractures and evaluated how to cover lost soft tissue to increase the success rate of reconstruction. We evaluated the risk factors for deep infection in lower limb Gustilo–Anderson (G–A) type III fractures. MATERIALS AND METHODS: This retrospective study investigated patients who underwent surgical procedures for lower limb G–A type III fractures between January 2007 and January 2017 at our institution. We enrolled 110 patients with 114 lower limb G–A type III fractures (77 G–A type IIIA fractures and 37 G–A type IIIB fractures) who were followed up for at least 2 years. We compared patients presenting infections with those without infections by assessing the following factors: severe contamination, diabetes, smoking, Injury Severity Scale, segmental fracture, location of fracture, G–A classification, damage control surgery, methods of surgery, timing of fixation, combination of antibiotics used, duration of antibiotic prophylaxis, timing of wound closure, and soft-tissue reconstruction failure. RESULTS: Eighteen fractures presented deep infections. Compared with patients without infections, patients developing infections differed significantly in terms of severe contamination (P < 0.01), G–A classification (P < 0.01), duration of antibiotic prophylaxis (P < 0.01), timing of wound closure (P < 0.01), and incidence of soft-tissue reconstruction failure (P < 0.01). Skin grafting was associated with significantly higher failure rates than muscle and free flap reconstructions (P = 0.04). Treatment with antibiotics was significantly longer in patients with drug-resistant bacterial infections than in those without infections (P < 0.01). CONCLUSION: Early flaps rather than skin grafting should be used to cover G–A type IIIB fractures, because skin grafting resulted in the highest failure rate among soft-tissue reconstructions in open fractures. Longer duration of antibiotic use had a significant impact not only on deep infection rates but also on the presence of drug-resistant bacteria. These findings suggest that prolonged use of antibiotics should be avoided in cases of open fractures. LEVEL OF EVIDENCE: Level IV retrospective observational study.