BACKGROUND: It is unclear whether appropriate empiric antimicrobial therapy improves outcomes in patients with bacteremia due to Escherichia coli or Klebsiella. The objective of this study is to assess the impact of appropriate empiric antimicrobial therapy on in-hospital mortality and post-infection length of stay in patients with Escherichia coli or Klebsiella bacteremia while adjusting for important confounding variables. METHODS: We performed a retrospective cohort study of adult patients with a positive blood culture for E. coli or Klebsiella between January 1, 2001 and June 8, 2005 and compared in-hospital mortality and post-infection length of stay between subjects who received appropriate and inappropriate empiric antimicrobial therapy. Empiric therapy was defined as the receipt of an antimicrobial agent between 8 hours before and 24 hours after the index blood culture was drawn and was considered appropriate if it included antimicrobials to which the specific isolate displayed in vitro susceptibility. Data were collected electronically and through chart review. Survival analysis was used to statistically assess the association between empiric antimicrobial therapy and outcome (mortality or length of stay). Multivariable Cox proportional hazards models were used to calculate hazard ratios (HR) and 95% confidence intervals (CI). RESULTS: Among 416 episodes of bacteremia, 305 (73.3%) patients received appropriate empiric antimicrobial therapy. Seventy-one (17%) patients died before discharge from the hospital. The receipt of appropriate antimicrobial agents was more common in hospital survivors than in those who died (p = 0.04). After controlling for confounding variables, there was no association between the receipt of appropriate empiric antimicrobial therapy and in-hospital mortality (HR, 1.03; 95% CI, 0.60 to 1.78). The median post-infection length of stay was 7 days. The receipt of appropriate antimicrobial agents was not associated with shortened post-infection length of stay, even after controlling for confounding (HR, 1.11; 95% CI 0.86 to 1.44). CONCLUSION: Appropriate empiric antimicrobial therapy for E. coli and Klebsiella bacteremia is not associated with lower in-hospital mortality or shortened post-infection length of stay. This suggests that the choice of empiric antimicrobial agents may not improve outcomes and also provides data to support a randomized trial to test the hypothesis that use (and overuse) of broad-spectrum antibiotics prior to the availability of culture results is not warranted.