Diagnosis of nosocomial pneumonia in cancer patients undergoing mechanicalventilation - A prospective comparison of the plugged telescoping catheterwith the protected specimen brush

Study objectives: Quantitative culture of protected samples of lower respir atory tract secretions obtained by a fiberoptic protected specimen brush (P SB) is widely accepted for the diagnosis of ventilator-associated pneumonia (VAP), but this diagnostic procedure is time consuming, expensive, and may give rise to iatrogenic complications, especially in cancer patients who o ften present with thrombocytopenia. The plugged telescoping catheter (PTC) could be a satisfactory alternative to the PSB in this setting. The aim of the present study was to evaluate the interest of the PTC to diagnose VAP i n ventilated cancer patients. Design: A prospective observational study. Setting: A 15-bed medical-surgical ICU in a comprehensive cancer center. Patients and interventions: Over a 9-month period, 42 patients suspected of baring bacterial VAP during mechanical ventilation underwent 69 bronchial samplings: a blinded PTC and a fiberoptic PSB were performed successively i n each case. A positive culture for both sampling procedures was defined as the recovery of greater than or equal to 10(3) cfu/mL of at least one pote ntial pathogen. The PSB result was taken as the reference standard. Measurements and results: The overall agreement between the techniques was 87% (60/69), PTC had a sensitivity of 67%, a specificity of 93%, a positive predictive value of 71%, and a negative predictive value of 91%. Conclusions: We conclude that the accuracy of the blinded PTC compares well with that of the PSB for the diagnosis of VAP in cancer patients. The sens itivity of the PTC observed herein, which is slightly lower than that descr ibed in previous studies, may be due to the blinded nature of the method: t he indications for initial or secondary coupling with a directed sampling m ethod in patients with suspicion of localized pneumonia remain to be determ ined.