The duration of antibiotic administration in penetrating abdominal trauma

The epidemiology of penetrating abdominal trauma is changing to reflect an increasing incidence of multiple injuries. Not only do multiple injuries increase the risk of infection, a very high risk of serious infection is conferred by immunosuppression from hemorrhage and transfusion and the high likelihood of intestinal injury, especially to the colon. Optimal timing and choice of presumptive antibiotic therapy has been established for penetrating trauma, but duration has not been studied extensively in such seriously injured patients. The purpose of this study was to test the hypothesis that 24 hours of antibiotic therapy remains sufficient to reduce the incidence of infection in penetrating abdominal trauma.

Three hundred fourteen consecutive patients with penetrating abdominal trauma were prospectively randomized into two groups: Group I received 24 hours of intravenous cefoxitin (1 g q6h) and group II received 5 days of intravenous cefoxitin. The development of a deep surgical site (intra-abdominal) infection as well as any type of nosocomial infection, as defined by the Centers for Disease Control and Prevention, (ie, surgical site infections, catheter-related infections, urinary tract, pneumonia), was recorded. Hospital length of stay was a secondary endpoint. Statistical analysis included chi-square tests for coordinate variables and two-tailed unpaired t tests for continuous variables. The independence of risk factors for the development of infection was assessed by multivariate analysis of variance. Significance was determined when P <0.05.

Three hundred patients were evaluable. There was no postoperative mortality, and no differences in overall length of hospitalization between groups. The duration of antibiotic treatment had no influence on the development of any infection (P = 0.136) or an intraabdominal infection (P = 0.336). Only colon injury was an independent predictor of the development of an intraabdominal infection (P = 0.0031). However, the overall infection incidence was affected by preoperative shock (P = 0.003), colon (P = 0.0004), central nervous system (CNS) injuries (P = 0.031), and the number of injured organs (P = 0.026). Several factors, including intraoperative shock (P = 0.021) and injuries to the colon (P = 0.0008), CNS (P = 0.0001), and chest (P = 0.0006), were independent contributors to prolongation of the hospital stay.

Twenty-four hours of presumptive intravenous cefoxitin versus 5 days of therapy made no difference in the prevention of postoperative infection or length of hospitalization. Infection was associated with shock on admission to the emergency department, the number of intra-abdominal organs injured, colon injury specifically, and injury to the central nervous system. Intra-abdominal infection was predicted only by colon injury. Prolonged hospitalization was associated with intraoperative shock and injuries to the chest, colon, or central nervous system.