Comparison of Multi-Drug Resistant Organisms Causing Early Ventilator-Associated Pneumonia in Three Geographically Distinct Trauma Intensive Care Units

<b><i>Introduction:</i></b> It is unclear why differences in patient location change organisms causing ventilator-associated pneumonia (VAP). We investigated VAP organisms in three geographically separate trauma intensive care units (TICUs). <b><i>Patients and Methods:</i></b> A retrospective review of organisms causing VAP (bronchoalveolar lavage [BAL] performed ≤7 d after admission and growing ≥10⁵ cfu/mL) in three geographically separate TICUs was conducted. Patients were treated by similar multidisciplinary teams and protocolized pathways. The primary outcome was the incidence of multi-drug resistant (MDR) VAP. Secondary outcomes were the incidence of inappropriate empiric antimicrobial therapy (IEAT) and the determination of risk factors for MDR VAP. Chi-squared, Kruskal-Wallis, and multi-variable logistic regression analyses were used accordingly. <b><i>Results:</i></b> In total, 271 patients were included: 142 in TICU-1, 63 in TICU-2, and 66 in TICU-3. The incidence of MDR VAP was similar across TICUs at 33.8%, 47.6%, and 39.4%, respectively (p = 0.17). Gram-negative MDRs were more prevalent in TICU-1 (70.8%) versus TICU-2 (60.0%) or TICU-3 (26.9%) (p = 0.001). Gram-positive MDRs were identified more in TICU-3 (73.1%) versus TICU-2 (43.3%) or TICU-1 (35.4%). IEAT did not differ by unit overall but was significantly greater for MDR gram-positive organisms in TICU-3 (70.4%) versus TICU-2 (44.8%) or TICU-1 (37.5%) (p = 0.02) and highest for MDR gram-negative organisms in TICU-1 (64.6%) versus TICU-2 (62.1%) or TICU-3 (55.8%) (p = 0.02). Multi-variable regression analyses revealed antibiotic days before BAL and kidney replacement therapy (KRT) as significant predictors of MDR VAP. <b><i>Conclusions:</i></b> Different TICU locations did not influence the overall incidence of MDR VAP, but differences in MDR organisms were observed. IEAT rates for both gram-positive and gram-negative organisms in different units may necessitate changes in empiric therapy. Antibiotic days prior to the BAL and KRT significantly increased the odds of early MDR VAP.