Situation

We have higher than desired utilization of empiric anti-MRSA and antipseudomonal treatment for patients admitted with respiratory syndromes. Often a specific organism is not isolated, but empiric treatment with vancomycin + either pip/tazo or cefepime is longer than it should be due to concern for possible MRSA or Pseudomonas. Additionally, though less common, patients are being treated for growth of Candida species in respiratory specimens, when pneumonia with this organism is exceedingly rare. 

Background

One institution modified the "usual flora" comment on respiratory specimens to add "No S. aureus or Pseudomonas isolated" and this change resulted in significant decrease in empiric anti-MRSA and antipseudomonal treatment, with concurrent reduction in subsequent acute kidney injury. Microbiology Comment Nudge Improves Pneumonia Prescribing in Open Forum Infectious Diseases 

Another institution limited Candida identification in respiratory specimens and did not increase mortality with patients in the limited identification group. They received less antifungals, had shorter length of stay and lower hospital cost. Improved Outcomes Associated with Limiting Identification of Candida spp. in Respiratory Secretions in Journal of Clinical Microbiology  

Assessment

Inappropriate empiric treatment with any of these antimicrobials can increase length of hospital stay, adverse drug events, contribute to C. difficile infections and antimicrobial resistance. Ongoing antimicrobial stewardship interventions should target these opportunities to reduce overuse for inappropriate empiricism. Other institutions have done this successfully, providing a precedent for us to make decisions.

Recommendation

1. Add the comment “No significant Staphylococcus aureus or Pseudomonas aeruginosa present” to the current “normal flora” statement on growth from respiratory cultures.  
2. If individuals call the microbiology lab wishing for further workup of any possible S. aureus or P. aeruginosa after the revised statement is posted, direct them to call Antimicrobial Stewardship
3. Categorize Candida spp. as "usual flora" and stop reporting it separately, even when there are no bacteria isolated
4. Apply recommendations 1 & 2 above to cultures from all respiratory specimens, including sputum, bronchial washings, and tracheal cultures

Revised November 2019