Abstract
Purpose
Recent 2005 American Thoracic Society and Infectious Diseases Society of America ventilator-associated pneumonia (VAP) guidelines suggest combination antimicrobial therapy for Pseudomonas aeruginosa VAP based on limited non-VAP evidence. We reviewed antimicrobial use in patients with P. aeruginosa VAP within our tertiary teaching institution to evaluate patient factors, antimicrobial selection, and outcomes.
Methods
Intubated patients with a P. aeruginosa sputum culture, infiltrate on chest x-ray, and increased white blood cell count were retrospectively enrolled. Collected data identified illness acuity using the Simplified Acute Physiology Score II (SAPS), culture sensitivities, antimicrobial use, and patient survival at hospital discharge.
Results
Fifty-nine patients were analyzed between January 2003 and November 2004. Average age was 57 ± 15.7, body mass index (BMI) of 28.5 ± 6.1, and SAPS of 45.1 ± 13.7. Survival was significantly higher among patients on two culture-sensitive antimicrobials (23/28, 82%) vs less than two agents (12/31, 39%; P = 0.0018). Almost all patients who received the beta-lactam and aminoglycoside (BL/Ag) combination survived (15/16; P = 0.0015). While fewer severely ill patients lived (SAPS greater than 54; 4/16; P = 0.0029), none were placed on the BL/Ag combination, and few on appropriate antimicrobials (6/37; P = 0.0324). Lower survival was seen among obese patients (BMI higher than 30; 9/25; P = 0.0043); although, 5/6 obese patients on BL/Ag survived (P = 0.0237). Patients more than 64-years-old were more often on two appropriate antimicrobials (P = 0.0143) and had a higher survival (P = 0.0241). Unfortunately, we observed a lack of culture follow-up, as only 60% (30/50) of inappropriate antimicrobials were switched to sensitive agents.
Conclusions
Our results emphasize the need to use combination BL/Ag antimicrobials for P aeruginosa VAP as the current guidelines suggest, and to ensure that antimicrobials are optimized based on culture results.
Get full access to this article
View all access options for this article.
