Adequacy of empiric coverage for Gram-negative bacteremias in an adult population

Purpose: This study reviewed Gram-negative bacteremias for appropriateness of antibiotic therapy at three points in time: 1) prior to Gram-stain results, 2) after Gram-stain results, and 3) after susceptibilities are posted.  The study also examined if a single antibiotic was more likely to cover any Gram-negative bacteremia or if the addition of a second agent increased the probability of appropriate coverage. 

Methods: This retrospective cohort examined patients 18 years of age or older with blood cultures positive for Gram-negative bacteremia from April 1^st, 2011 to September 30^th 2011. Charts were reviewed for demographics, co-morbidities, cultures, antibiotic utilization, and in-hospital mortality. Antibiotics were identified as appropriate if the organism was susceptible. 

Results: Antibiotic therapy was appropriate for 53.8% of cultures prior to Gram-stain results, 94.6% of cultures after Gram-stain results, and 99.2% of cultures after susceptibilities were determined.  The monotherapy analysis showed susceptibilities of 88.1% for pipercillin/tazobactam, 94.7% for cefepime and 97.7% for meropenem.  Comparing these agents produced a statistically significant benefit for meropenem over pipercillin/tazobactam (p = 0.0014) and the benefit of cefepime over pipercillin/tazobactam was nearing significance (p=0.062).  There was no significant difference comparing cefepime and meropenem.  Addition of a gentamicin to pipercillin/tazobactam showed a statistically significant increase in appropriate empiric therapy.  A significant increase in coverage was not seen with other combinations. 

Conclusion: Utilization of pipercillin/tazobactam will be further reviewed to determine if a change in recommendations for empiric antibiotics should be recommended.  Gentamicin may be considered as additional therapy if pipercillin/tazobactam is selected for empiric coverage.  Additional studies are needed to analyze the adverse events associated with gentamicin versus the benefit of the double coverage.