The aim of this study was to assess the accuracy of protected telescoping catheter performed using: 1) protected distal aspiration, or 2) protected mini-bronchoalveolar lavage for the diagnosis of ventilator-associated bacterial pneumonia. Twenty-seven patients who died after receiving mechanical ventilation for at least 72 h were included in a comparative prospective post-mortem study. The two microbiological sampling procedures were performed immediately after death. Surgical pneumonectomies and biopsies were performed within 30 min of death at the bedside for histological examination. The results of the two techniques were compared with histological post-mortem lung examination or biopsies. Histological examination of the parenchyma showed signs of pneumonia in 14 cases. Lung tissue culture was positive in nine of these 14 cases. When ventilator-associated bacterial pneumonia was defined by the association of histological signs and positive lung tissue culture the sensitivity was 78% for both sampling techniques, specificity was 86% for mini-bronchoalveolar lavage and 100% for protected distal aspiration (at a threshold of 1 x 10(3) cfu x mL(-1)). Both techniques protected the distal aspiration and mini-bronchoalveolar lavage, and provided good specificity with an acceptable sensitivity for the diagnosis of ventilator-associated bacterial pneumonia.