BACKGROUND: Ventilator-associated pneumonia (VAP) is one of the most common nosocomial infections; however, its diagnosis remains difficult to establish in the critical care setting. We investigated the potential role of neutrophil CD64 (nCD64) expression as an early marker for the diagnosis of VAP. METHODS: Forty-nine consecutive patients with clinically suspected VAP were prospectively included in a single-center study. The levels of nCD64, C-reactive protein (CRP), and serum procalcitonin (PCT) were analyzed for diagnostic evaluation at the time of intubation (baseline), at day 0 (time of diagnosis), and at day 3. The receiver operating characteristic curves were analyzed to identify the ideal cutoff values. RESULTS: VAP was confirmed in 36 of 49 cases. In patients with and without VAP, the median levels (interquartile range, IQR) of nCD64 did not differ either at baseline [2.4 (IQR, 1.8-3.1) and 2.6 (IQR, 2.3-3.2), respectively; p=0.3] or at day 0 [2 (IQR, 2.5-3.0) and 2.6 (IQR, 2.4-2.9), respectively; p=0.8]. CRP showed the largest area under the curve (AUC) at day 3. The optimum cutoff value for CRP according to the maximum Youden index was 133 mg/dL. This cutoff value had 69% sensitivity and 76% specificity for predicting VAP; the AUC was 0.73 (95% CI, 0.59-0.85). The nCD64 and PCT values could not discriminate between the VAP and non-VAP groups either at day 0 or day 3. CONCLUSIONS: The results of this pilot study suggest that neutrophil CD64 measurement has a poor role in facilitating the diagnosis of VAP and thus may not be practically recommended to guide the administration of antibiotics when VAP is suspected.
BACKGROUND: Ventilator-associated pneumonia (VAP) is one of the most common nosocomial infections; however, its diagnosis remains difficult to establish in the critical care setting. We investigated the potential role of neutrophil CD64 (nCD64) expression as an early marker for the diagnosis of VAP. METHODS: Forty-nine consecutive patients with clinically suspected VAP were prospectively included in a single-center study. The levels of nCD64, C-reactive protein (CRP), and serum procalcitonin (PCT) were analyzed for diagnostic evaluation at the time of intubation (baseline), at day 0 (time of diagnosis), and at day 3. The receiver operating characteristic curves were analyzed to identify the ideal cutoff values. RESULTS: VAP was confirmed in 36 of 49 cases. In patients with and without VAP, the median levels (interquartile range, IQR) of nCD64 did not differ either at baseline [2.4 (IQR, 1.8-3.1) and 2.6 (IQR, 2.3-3.2), respectively; p=0.3] or at day 0 [2 (IQR, 2.5-3.0) and 2.6 (IQR, 2.4-2.9), respectively; p=0.8]. CRP showed the largest area under the curve (AUC) at day 3. The optimum cutoff value for CRP according to the maximum Youden index was 133 mg/dL. This cutoff value had 69% sensitivity and 76% specificity for predicting VAP; the AUC was 0.73 (95% CI, 0.59-0.85). The nCD64 and PCT values could not discriminate between the VAP and non-VAP groups either at day 0 or day 3. CONCLUSIONS: The results of this pilot study suggest that neutrophil CD64 measurement has a poor role in facilitating the diagnosis of VAP and thus may not be practically recommended to guide the administration of antibiotics when VAP is suspected.