Abstract
No gold standard method exists for the diagnosis of ventilator-associated pneumonia despite the availability of multiple techniques.
A prospective, crossover study was performed on mechanically ventilated patients meeting with suspected pneumonia. Eighteen paired samples were obtained on 15 patients, comparing the results of quantitative tracheal lavage (QTL) to bronchoscopic protected brush specimen (PSB) by quantitative culture and gram stain examination.
The sensitivity, specificity, positive and negative predictive values, and accuracy are affected by the growth density threshold selected, and whether the same organisms are expected by both methods. There is a significant relationship between QTL and PSB (P = 0.0048; R = 0.632), gram stain and PSB (P <0.001; R = 0.791), and gram stain and QTL (P = 0.0125; R = 0.575), by Spearman rank order correlation.
QTL may have a role for diagnosing and directing treatment of ventilator-associated pneumonia, allowing reservation of bronchoscopic PSB for secondary, high risk and refractory cases.