The optimal management of postintubation tracheal stenosis is not well defined.A therapeutic algorithm was designed by thoracic surgeons, ear, nose and throat (ENT) surgeons, anaesthetists and pulmonologists. Rigid bronchoscopy with neodymium±yttrium aluminium garnet (Nd-YAG) laser resection or stent implantation (removable stent) was proposed as first-line treatment, depending on the type of stenosis (web-like versus complex stenosis). In patients with web-like stenoses, sleeve resection was proposed when laser treatment (up to three sessions) failed. In patients with complex stenoses, operability was assessed 6 months after stent implantation. If the patient was judged operable, the stent was removed and the patient underwent surgery if the stenosis recurred.This algorithm was validated prospectively in a series of 32 consecutive patients. Three patients died from severe coexistent illness shortly after the first bronchoscopy. Of the 15 patients with web-like stenosis, laser resection was curative in 10 (66%). Among the 17 patients with complex stenoses, three remained symptom-free after stent removal. Bronchoscopy alone was thus curative in more than one-third of the patients. Six patients underwent surgery, two after failure of laser resection and four after failure of temporary stenting. Surgery was always performed with the patient in good operative condition. Palliative stenting was the definitive treatment in nine cases. Tracheostomy was the definitive solution in two cases.This approach, including an initial conservative treatment, depending on the type of the stenosis, appears to be applicable to almost all patients and allows secondary surgery to be performed with the patient in good condition. Eur Respir J 1999; 13: 888±893.

Continuous control of P(cuff) is associated with significantly decreased microaspiration of gastric contents in critically ill patients.

The diagnostic accuracy of protected-specimen brush (PSB), bronchoalveolar lavage (BAL), and endotracheal aspirates (EA) was prospectively evaluated in a series of 28 mechanically ventilated patients (MV patients) who died within 3 d of the bronchoscopic procedure, using postmortem lung examination as the gold standard for establishing the diagnosis of pneumonia. The entire fixed lungs were carefully dissected along the bronchovascular axes and each segment was cut into 5- to 10-mm thick sections, enabling gross examination of the lung parenchyma. Two tissue blocks were taken from each segment, including grossly abnormal areas whenever present. In several cases, two peripheral (subpleural) lung-tissue blocks were also taken from each lobe prior to systematic dissection of the lungs. Quantitative cultures (QC) and direct cytologic and microbiologic examination (DE) was performed on respiratory samples obtained within 72 h before death. Values of 10(3) cfu/ml of Ringer's solution, 10(4) cfu/ml of retrieved fluid, and 10(6) cfu/ml of respiratory secretions were used as cutoff points for quantitative PSB, BAL, and EA cultures, respectively. The main findings in this study were that: (1) Pneumonia was present in 67% of the patients. (2) Histologic lesions of pneumonia were mainly bilateral and predominated in the dependent lung segments. (3) Coexistence of a variety of noninfectious processes was a common finding in patients with pneumonia. (4) In several cases pneumonia was absent from peripheral lung samples while more central areas of the same segment displayed typical foci of pneumonia. (5) The sensitivity of quantitative cultures was 55%, 57%, and 47% for EA, PSB, and BAL, respectively, and the specificity was 85%, 88%, and 100%, respectively. Reducing the diagnostic threshold of EA to 10(5) cfu/ml of respiratory secretions instead of 10(6) cfu/ml resulted in a sensitivity of 63.1% and a specificity of 75% for EA. The sensitivity of direct examination (DE) was 50%, 47%, and 47%, respectively, and the specificity was 75%, 88%, and 87%. (6) The presence of intracellular organisms (ICO) in BAL had a 36.8% sensitivity and 100% specificity in establishing the diagnosis of pneumonia regardless of their percentage. (7) Although 15 patients (53%) were not on antibiotics or were off antibiotics for more than 48 h before testing, no relationship could be established between the patients' antibiotic status and the result of any diagnostic test. By using a recommended methodology for respiratory sampling techniques together with complete postmortem lung examination as a diagnostic "gold standard," this study provides a realistic insight into the diagnostic values of EA, PSB, and BAL in MV patients with suspected pneumonia.

Diagnostic indications for flexible bronchoscopy in the initial investigation of children with suspected foreign-body (FB) aspiration have not been evaluated prospectively. We prospectively collected history, clinical, and radiologic findings at prebronchoscopic examination of all children referred for suspected FB aspiration between February 1993 and September 1995. Children with asphyxiating FB aspiration, requiring immediate rigid bronchoscopy, were excluded. If there was clear evidence of FB aspiration from the physical and radiographic findings, rigid bronchoscopy was directly performed. If the evidence was not convincing, children underwent diagnostic flexible bronchoscopy under local anesthesia. If an FB was found, rigid bronchoscopy was always performed for extraction. Eighty-three consecutive children (median age: 24 mo) were included. Among 28 who underwent rigid bronchoscopy first, 23 had an FB. Among the 55 children who underwent flexible bronchoscopy first, 17 had an FB. Predictive signs of a bronchial FB were a radiopaque FB, and associated unilaterally decreased breath sounds and obstructive emphysema (positive predictive value = 0.94). We propose the following management algorithm: Rigid bronchoscopy is performed first in case of asphyxia, a radiopaque FB, or association of unilaterally decreased breath sounds and obstructive emphysema. In any other case, flexible bronchoscopy is performed first for diagnostic purposes. If applied retrospectively to the 83 children in our study, this algorithm would have decreased the negative first rigid bronchoscopy rate to 4%. Flexible bronchoscopy is a safe and cost-saving diagnostic procedure in children with suspected FB aspiration.

No abstract