Interest is growing in the use of registries to address such issues as outcomes research, health care quality assessment, best-practices determination, resource allocation, and reimbursement. Endobronchial ultrasound guided transbronchial needle aspiration (EBUS-TBNA) is a relatively new technology that is rapidly becoming widely accepted in clinical practice. However, there are little outcomes data available regarding risk adjusted diagnostic -yield for EBUS-TBNA outside of the research setting.
Data was prospectively collected from 8 centers participting in the ACCP Quality Improvement Registry, Evaluation, and Education (AQuIRE) diagnostic bronchoscopy registry on consecutive patients undergoing EBUS-TBNA. This included information on demographics, indications, procedure details, anesthesia, comorbidities, general health status, lymph node size, diagnostic yield, final diagnosis, complications, morbidity and mortality, escalations in level of care, and procedural resource utilization.
A total of 405 patients were entered into the database over a 5 month period. Of these, 317 had completed follow-up and 205 had EBUS-TBNA performed. Among the EBUS-TBNA cases, the average number of lymph nodes biopsied was 1.92 + 0.79. EBUS-TBNA obtained a specific diagnosis in 111/205 (54%) cases and adequate lymphoid tissue without a specific diagnosis was obtained in another 80/205 (39%). Adequate EBUS-TBNA samples, defined as either a specific diagnosis or adequate lymphocytes on one or more stations on a per patient basis, was 93%. The probability of obtaining a specific diagnosis was higher in patients with larger lymph nodes by CT scan (Largest lymph node 1-10 mm 41%, 11-20 mm 48%, 21-30 mm 65%, >30 mm 71%; p=0.025), when deep or general anesthesia was used (76/127 (60%) vs. 35/78 (45%); p=0.044), and when fellows were involved (98/168 (58%) vs. 13/36 (36%); p=0.017). In multivariable models the probablity of obtaining a specific diagnosis was only associated with deep or general anesthesia (OR 2.31, 95% CI 1.25 –4.26; p=0.008) and all lymph nodes being less than 11 mm in size (OR 0.38, 95% CI 0.18-0.81; p=0.013). The probability of an adequate EBUS-TBNA sample was not affected by size of the largest lymph node (largest lymph node 1-10 mm 97%, 11-20 mm 92%, 21-30 mm 95%, >30 mm 86%; p=0.27) or by involvement of fellows (158/168 (94%) vs. 33/36 (92%); p=0.71). Adequacy of EBUS-TBNA was higher when deep or general anesthesia was used (122/127 (96%) vs. 69/78 (88%); p=0.047). On-site cytology was used in 175/205 (85%) cases. We failed to demonstrate an association between the use of on-site cytology and the probability of obtaining a specific diagnosis (on-site cytology 95/175 (54%) vs. none 16/30 (53%), p=1.0) or the probability of an adequate EBUS-TBNA (on-site cytology 165/175 (94%) vs. none 26/30 (87%); p=0.13). In multivariable models the adequacy of EBUS-TBNA was only related to the use of deep or general anesthesia (OR 3.18, 95% CI 1.03-9.88; p=045). Among institutions contributing 10 or more cases there were significant differences in the use of deep or general anesthesia (p<0.001). Complications occurred in 3/205 (1.5%) cases, and 2 of these required an escalation of care. There was no associated mortality.CONCLUSIONS: Diagnostic yield for a specific diagnosis is higher in patients with enlarged lymph nodes by CT and when deep or general anesthesia is used. Adequacy of EBUS-TBNA is higher when deep or general anesthesia is used. Complication rates are low with EBUS-TBNA. There is considerable practice variation in the level of anesthesia used for EBUS-TBNA and its impact on outcome must be weighed against the added resource utilization, best local practices and potential risk involved.
The use of internet based registry data is feasible and can facilitate quality improvement, benchmarking, outcomes research, establishment of best practices and can be used to measure risk-adjusted diagnostic yield. It can also serve as an infrastructure for comparative effectiveness research.DISCLOSURES: Industry Grants: Kevin Kovitz- Spiration, Broncus (both with practice and/or institution but not directly with Dr. Kovitz, he is the principle investigator). Consultant fee, speaker bureau, advisory committee, etc: Kevin Kovitz - Consultant for Broncus and Pulmonx