After severe trauma, physicians frequently use multiple antibiotics for prolonged periods of time to prevent sepsis, based on intuition rather than scientific evidence. Over a 1-year period (January-December 1999) we included prospectively 112 critically injured patients who required an operation and/or chest tube insertion and stayed for more than 2 days in the intensive care unit (ICU). Of these patients, 46 received a single prophylactic antibiotic for 24 hours (group SING+SHORT), and 66 received one or more prophylactic antibiotics for more than 24 hours (group MULT+LONG), based on physician discretion. Twenty-seven outcome parameters were collected to compare the effect of the different prophylactic antibiotic regimens. The two groups were similar in regard to overall injury severity, age, gender, mechanism of injury, and physiologic condition on admission. However, more SING+SHORT patients had an abdominal operation (83% versus 62%, P = 0.02), and more MULT+LONG patients had an orthopedic operation (35% versus 15%, P = 0.03). There was no difference in sepsis (41% versus 42%, P = 1.0), organ failures (37% versus 50%, P = 0.18), mortality (7% versus 12%, P = 0.52), ICU stay (14 +/- 2.5 versus 16 +/- 2 days, P = 0.57), hospital stay (26 +/- 3 versus 28 +/- 2 days, P = 0.53), or any other outcome parameter. Independent risk factors for sepsis were blunt mechanism of trauma, Injury Severity Score > or = 25, and more than two units of blood transfused over the first 24 hours, but not the amount of prophylactic antibiotics given. In conclusion, we found that 24-hour prophylaxis with a single broad-antibiotic is as effective as prophylaxis for longer periods of time with multiple spectrum antibiotics for critically injured patients at high risk for sepsis.