Diagnosis of ventilator-associated pneumonia: agreement between quantitative cultures of endotracheal aspiration and plugged telescoping catheter.

Abstract

To evaluate the diagnostic agreement between quantitative cultures of samples obtained with endotracheal aspiration (ETA) and plugged telescoping catheter (PTC). Prospective study. Medical ICU. Hundred thirty-eight episodes of suspected ventilator-associated pneumonia studied in 100 consecutive patients. For each suspected episode of ventilator-associated pneumonia, ETA and PTC were performed consecutively. The agreement between microbiological results obtained from the two techniques was evaluated (kappa statistic test). Pneumonia was diagnosed on a positive culture result of telescoping catheter with the threshold set at 10(3) cfu/ml or more. The cut-off points evaluated for ETA ranged from 10(2 )to 10(6) cfu/ml. Micro-organisms retrieved by aspiration and telescoping catheter were similar and bacterial counts obtained by the two procedures were well correlated ( r = 0.71 p < 0.001). There was good agreement between positive and negative ETA and PTC specimens (kappa: 0.78) with a diagnostic threshold for ETA of 10(4) cfu/ml. The sensitivity and specificity of ETA for the diagnosis of PTC-confirmed pneumonia were 92% and 85%, respectively. Kappa decreased to 0.48 when the diagnostic threshold was increased to 10(6) cfu/ml. Quantitative cultures of ETA and PTC tallied for both micro-organisms and counts. The simpler ETA appears adequate for determining the presence of pathogenic organisms in significant concentration in the lower respiratory tract.