PURPOSE: 1] To determine incidence of complications related to EBUS-TBNA. 2] To measure outcomes of these complications.
METHODS: Data regarding demographics, procedural details, indication, lesion specifics and complications were extracted from the ACCP Quality Improvement Registry Education and Evaluation (AQuIRE) database and reviewed. The database is an ongoing multicenter bronchoscopy registry where comprehensive information on consecutive eligible patients are prospectively captured along with details of the procedures performed and clinical outcomes.
RESULTS: 1323 EBUS-TBNA bronchoscopies were performed in 635 females and 688 males with a mean of age 63± 13 years. A trainee was involved in 1084(82%) of these bronchoscopies. In addition to EBUS-TBNA, BAL (20%), transbronchial biopsies (14%), bronchial brushings (14%) and endobronchial biopsies (12%) were also performed. TBNA was performed at 1 site in 428 (32%) patients, 2 sites in 403 (30.5%), 3 sites in 307(23%), 4 sites in 114(9%), 5 sites in 55 (4%) and 6 sites in 16(0.5%) The dedicated 22-gauge needle was utilized in 1034(78%), 21-gauge needle in 264(20%), a conventional 20-gauge needle in 5 (0.4%) and the Wang 19-gauge needle in 20 (1.6%) bronchoscopies.. Procedure related complications occurred in 19 patients (1.5%) that included bleeding (n=3), pneumothorax (n=7), sustained hypoxia (n=4), unexplained hypotension (n=1), respiratory failure (n=3) and significant airway injury leading to death (n=1). Among patients with EBUS-TBNA related complications, 11 required escalation in the level of care with 5 requiring Intensive Care Unit admission, 5 requiring hospital admission and 1 requiring observation in the Emergency Center. Of the 7 patients with pneumothorax, 4 required chest tube drainage.
CONCLUSIONS: Endobronchial ultrasound is associated with a relatively low complication rate. This is however the first report of associated mortality. When complications occur, majority of patients require escalation of care.
CLINICAL IMPLICATIONS: Prospective registries capturing outcomes and complications of all invasive procedures can provide vitally useful information regarding the real world performance of invasive diagnostic tests. Data from such registries can be used for quality improvement and benchmarking.
DISCLOSURE: The following authors have nothing to disclose: Archan Shah, David Ost, Carlos Jimenez, Rodolfo Morice, Lonny Yarmus, Joshua Filner, George Eapen
No Product/Research Disclosure Information