Ventilator-associated pneumonia (VAP) remains a challenge in critical care. Transfusion (TX) of red cells (pRBCs) has been associated with nosocomial infection generally. Little data exist on the relationship between TX and VAP.
We conducted a secondary analysis of a large, multicenter prospective, observational study of TX practice in critical care (CRIT Trial). As part of this study, pRBC TX rates were prospectively tracked, as was the development of pneumonia. Pneumonia was prospectively defined based on commonly employed clinical criteria. We defined VAP as pneumonia in patients who had received ≥48 hrs of mechanical ventilation (MV) but did not have pneumonia on admission to the ICU. Additional data collected included: demographics, ICU type, severity of illness, comorbidities, use of antibiotics, and process of care. To explore risk factors for VAP we compared patients with VAP to those not diagnosed with VAP.
Of 4,982 patients enrolled, 1,518 received ≥48 hrs of MV and lacked pneumonia on ICU admission. VAP developed in 311 subjects (20.5%). In patients with VAP the mean duration of MV prior to the diagnosis of VAP was 5.1±4.2 days. Approximately 60% of patients with VAP received pRBCs during the observation period prior to developing VAP vs. 51.4% of controls (p<0.0001). Factors independently associated with VAP are shown below.
VariableAdjusted Odds Ratio (95% CI)PMale gender1.52 (1.14–2.04)0.004Trauma admission1.97 (1.36–2.86)<0.001Heavy sedation1.51 (1.14–2.01)0.004Parentral nutrition3.31 (2.30–3.77)<0.001Enteral nutrition within 48 hrs of MV2.77 (2.03–3.77)<0.0011–2 u pRBC TX1.81 (1.23–2.66)0.002>2 u pRBC TX1.89 (1.28–2.81)0.002
There was no relationship between admission hemoglobin and VAP.CONCLUSIONS: TX is common in MV patients and increases the risk for VAP.CLINICAL IMPLICTIONS: Physicians need to reevaluate their approach to TX in MV patients. TX practice is one of the few process of care variables amenable to intervention. Alternatives to TX are needed.
A.F. Shorr, OBI, Grant monies.