Empiric therapy of sepsis in the surgical intensive care unit with broad-spectrum antibiotics for 72 hours does not lead to the emergence of resistant bacteria.

Background: It is our practice to treat suspected sepsis with imipenem/cilastatin and gentamicin (IMP/GENT) for 72 hours while awaiting culture results. We wanted to determine if this practice engenders antimicrobial resistance.

Methods: Review of prospectively collected data regarding use of IMP/GENT and microbial sensitivity to imipenem/cilastatin during the first and last 7 months of a 19-month study period (October 1, 1995, to April 30, 1997).

Results: The susceptibility of appropriate organisms to imipenem/cilastatin was 76% in the early period and 80% in the late period (p = 0.42). Pseudomonas aeruginosa was more susceptible in the late period (88 vs. 62%; p = 0.007). Resistance to gentamicin (30% early vs. 21% late; p = 0.02) and representative cephalosporins (cefoxitin, 52% early vs. 61% late; p = 0.35; ceftazidime, 26% early vs. 23% late; p = 0.76) did not develop during the study period. The incidence of fungemia was the same in both periods (4 of 467 admissions vs. 3 of 599 admissions; p = 0.48).

Conclusion: This protocol did not lead to the emergence of resistant bacteria.