Blind and bronchoscopic sampling methods in suspected ventilator-associated pneumonia. A multicentre prospective study.

To compare four sampling methods: blind tracheal aspirate (blind TA), blind protected telescoping catheter (blind PTC), bronchoscopic PTC and bronchoscopic bronchoalveolar lavage (bronchoscopic BAL), for diagnosis of ventilator-associated pneumonia (VAP). Prospective multicentre study. Five intensive care units in France. Sixty-three patients with mechanical ventilation for more than 48 h, no recent antibiotic change (<72 h) and suspected nosocomial pneumonia. All patients underwent the four sampling methods. Direct examination and quantitative cultures of the four specimens were performed. Visible secretions expelled from the catheter were present 40 times (63%) for blind PTC and 45 times (71%) for bronchoscopic PTC. After exclusion of 11 uncertain cases, 34 VAP were diagnosed. Direct examination of PTC (either blind or bronchoscopic) did not differ from direct examination of bronchoscopic BAL in predicting VAP diagnosis and in guiding initial antibiotic treatment correctly. Compared to that of bronchoscopic BAL (0.98), the area under receiver operating characteristics (ROC) curve was smaller for blind TA (0.78, p=0.002), blind PTC (0.83, p=0.009) and bronchoscopic PTC (0.85, p=0.01). When samples with visible secretions expelled from the catheter were considered, blind and bronchoscopic PTC had areas under ROC curve close to that of bronchoscopic BAL (0.90, p=0.22 and 0.91, p=0.27, respectively). Blind PTC appears to be a good alternative to bronchoscopic sampling for VAP diagnosis, provided that the sample contains visible secretions expelled from the catheter.