Pharmacy-driven initiative improves care for severe community-acquired pneumonia
Bianchini M, Mercuro N, Kenney R, Peters M, Swiderek J, Samuel L, and Davis S. Pharmacy-driven initiative improves care for severe community-acquired pneumonia. Crit Care Med 2019; 47(1).
Crit Care Med
Learning Objectives: Community-acquired pneumonia (CAP) is associated with low diagnostic yield, which can lead to suboptimal antimicrobial therapy. The purpose of this study was to improve management and outcomes of severe CAP through implementation of a pharmacist-driven bundle for ordering evidence-based microbiologic diagnostic tests in a medical intensive care unit (MICU). Methods: Single-center, IRB-approved, quasi-experiment for CAP in a 165-bed MICU before and after implementation of a diagnostic stewardship bundle. Intervention Nov 2017 - Mar 2018: collaborative practice agreement for MICU pharmacists to order criteria-driven tests, including: influenza PCR, respiratory viral panel, Legionella urine antigen test, and procalcitonin for CAP. Pre-intervention: Nov 2016 - Mar 2017. Inclusion: Adult, MICU, empiric antibiotics for CAP. Primary outcome: duration of antimicrobial therapy. Secondary outcomes: diagnostic tests, de-escalation, length of stay, mortality, and safety. Results: 91 patients were included per group. 42 (46%) pre- and 46 (51%) post-intervention had a risk factor for MRSA or pseudomonas. Median duration in pre- and post-intervention: 7 days, IQR 6-10 vs 7 days, IQR 6-8. Pathogen identification: 34% vs 51%, p <0.035. Any antimicrobial de-escalation: 26% vs 58%, p<0.001. All antimicrobials discontinued: 0% vs 17%; anti-pseudomonal de-escalation: 18% vs 29%; anti-MRSA de-escalation: 18% vs 32%; atypical de-escalation: 18% vs 21%. Median time to de-escalation was 3 days in each group. No differences detected between groups for length of ICU stay (4 vs 6 days), inpatient allcause mortality (14% vs 8%), retreatment (15% vs 12%), 30-day readmission (18% vs 14%), or C. difficile (1% vs 0%). After controlling for other variables, the intervention group was the only variable that was independently associated with antimicrobial deescalation (adjusted OR 4.03, 95% CI 1.96-7.35). Conclusions: Implementation of a pharmacy-driven pneumonia diagnostic stewardship bundle improved the use of evidencebased diagnostics which led to an increase in pathogen identification for patients admitted to the MICU with CAP. This intervention was associated with more antimicrobial de-escalation without negative impact on safety outcomes.