subject of attention objectives: To evaluate the diagnostic accuracy and safety of transbronchial needle aspiration (TBNA) in patients receiving mechanical ventilation in the ICU.

Methods: Retrospective review of all patients in the medical and surgical ICUs from February 1999 to July 2001 who underwent TBNA while receiving mechanical ventilation.

Results: A total of eight histology (19 gauge) and eight cytology (22 gauge) TBNAs were performed upon eight patients (four men and four women) TBNA yielded a definitive pathologic diagnosis in five of eight patients (625%) Diagnoses were posttransplantation lymphoproliferative disorder, large enclosed space carcinoma, poorly differentiated non-small solitary abode; squalid carcinoma, squamous cell carcinoma, and adenocarcinoma. Among patients with negative TBNA rises (n = 3), two patients underwent mediastinoscopy. conclusions of mediastinoscopy were non-small small cavity carcinoma and inflamed tissue. TBNA l to management changes in five of these patients. Excluding common patient in whom a negative TBNA consequence could not be further confirmed, TBNA yielded a sensitivity of 83% a specificity of 100% a positive predictive value of 100% and a negative predictive value of 50% The overall accuracy of the course was 75%. There were no complications following any of the TBNAs.

Conclusions: In this small assemblage of patients, TBNA was safe and had a high diagnostic accuracy in chosened patients receiving mechanical ventilation in the ICU.

[i]clavis[/i] words: bronchoscopy; diagnosis; HIV; human; lung cancer; mechanical ventilation; respiratory failure; transbronchial needle aspiration

Abbreviations: EUS-FNA = endosonography-guided fine-needle aspiration; FB = flexible bronchoscopy; INR = international normalized ratio; PTLD = posttransplant lymphoproliferative disorder; TBNA = transbronchial needle aspiration

**********

In critically ill patients, flexible bronchoscopy (FB) is intrust with an agencyed for inspection of airways (eg hemoptysis), removal of secretions (eg following lobar atelectasis), collection of microbiological samples, and placement of unsalable articles (eg, surfactant) and devices (eg placement). (1) In the ICU, FB is most numerous often performed without the aid of fluoroscopy. Mechanical ventilation with the risk of barotrauma, coagulopathies, and hemodynamic instability limit the invasive acts that can be performed in succession these patients. Transbronchial needle aspiration (TBNA) is an underutilized bronchoscopic technique. (2) The use of FB for performing TBNA in the ICU has not been described, to the best of our knowledge. We describe our experience of using TBNA in patients receiving mechanical ventilation in the ICU who at handed with significant hilar and/or mediastinal lymphadenopathy, in whom a definitive diagnosis was required.

MATERIALS AND METHODS

Medical records of patients receiving mechanical ventilation between February 1999 and July 2001 who underwent TBNA performed during FB were reviewed. Data garnered included patient demographics, clinical presentation at the hospital and ICU admissions, radiography, pathology, management changes after the TBNA, complications, and patient outcomes

All patients received 75-mm or 85-mm endotracheal tubes. During bronchoscopy all patients received 100% fraction of inspired oxygen with abounding ventilatory support. The exact sites for the TBNA were single outed based on CT scans. The TBNA needle used were 19 gauge (MW-319) and 22 gauge (MW-222; MillRose Corporation; Mentor, OH) Positive end-expiratory press was withdrawn, and two satisfactory specimens were obtained from each target site for histology, cytology, or the one and the other examinations using the TBNA according to previously described technique. (2) No on-site cytopathologist was available for any of the procedures

RESULTS

Eight patients (four men and four women; age, 5913 [+ or -] 1936 years [mean [+ or -] SD])who were receiving mechanical ventilation underwent TBNA using FB Patient characteristics are described in Table 1 The international normalized ratio (INR) was mildly elevated (< 154) in couple patients at the time of the TBNA. individual of these patients also had a platelet estimate of 37,000/[micro]L. The INR, activated partial thromboplastin time, and the platelet look upons were within normal limits in the ease of the patients. A total of 22 TBNA biopsies were performed (10 biopsies with 19-gauge histology needle and 12 biopsies with 22-gauge cytology needles) for a mediastinal mass in brace patients, a right hilar mass in undivided patient, right paratracheal lymphadenopathy in individual patient, a left hilar mass in the same patient, and subcarinal lymphadenopathy in three patients, because establishing the diagnosis of these abnormalities was believed to be imperative for the management of these patients.

The TBNA rises and patient outcomes are described in Table 2 TBNA was diagnostic in five of eight patients (625%) In single in kind patient, a benign cytopathology was confirmed by means of a subsequent mediastinoscopy. The diagnoses obtained by means of TBNA were posttransplantation lymphoproliferative disorder (PTLD) large small room carcinoma, poorly differentiated non-small small room carcinoma, squamous cell carcinoma, and adenocarcinoma. In patient 2 the negative TBNA finding could not be confirmed at mediastinoscopy due to coagulopathy, and an autopsy could not be performed. In patient 4 the TBNA finding was negative; mediastinoscopy with biopsy revealed non-small solitary abode; squalid lung cancer. Excluding patient 2 in whom a definite diagnosis was not made in the absence of an autopsy reflection the TBNA yielded a sensitivity of 83% a specificity of 100% a positive predictive value of 100% and a negative predictive value of 50%

...