[384] Diagnostic Utility of Endobronchial Ultrasound-Guided Fine-Needle Aspiration (EBUS-FNA): A Review of 593 Cases

Arivarasan Karunamurthy, Guoping Cai, Sanja Dacic, Walid E Khalbuss, Liron Pantanowitz, Sara E Monaco. University of Pittsburgh Medical Center, Pittsburgh, PA; Yale University School of Medicine, New Haven, CT

Background: Establishing a definitive and accurate diagnosis is important for the diagnosis of mediastinal lesions by endobronchial ultrasound-guided fine needle aspiration (EBUS-FNA) and can decrease the need for additional diagnostic work up. Our aim was to evaluate the diagnostic utility of EBUS-FNAs from our institution.
Design: A retrospective review of all EBUS-FNA biopsies procured at our institution over a three year period (2007 to 2010) was performed to look at the adequacy, cytological diagnosis, and the available histological follow-up. Cases were given an adequacy statement (satisfactory/SAT, less than optimal/LTO, or unsatisfactory/UNSAT) and a descriptive diagnosis at the time of final cytological diagnosis that was based on cytomorphology and in some cases ancillary study results. Indeterminate cases were those cases with a final diagnosis of atypical or suspicious.
Results: A total of 593 EBUS FNA specimens were obtained from 357 patients with a mean age of 61years (ranging from 23-90 years), including 420 SAT cases (71%), 107 LTO cases (18%) and 66 UNSAT cases (11%). Histological follow-up was available in 203 cases (34.2%).

Table 1: Cytological-Histological Correlation in 203 of 593 EBUS-FNA Biopsies

FNA adequacy & diagnosis	Histology: Benign	Histology: Granulomas	Histology: Neoplasm	Histology: Malignant	TOTAL
UNSAT	17	6	0	7	30
LTO	32	12	0	11	55
Negative	28	9	0	6	43
Granulomas	0	3	0	0	3
Atypical	4	0	0	4	8
Suspicious	0	0	0	1	1
SAT	58	34	1	25	118
Negative	45	11	0	4	60
Granulomas	0	21	0	0	21
Atypical	12	2	0	1	14
Suspicious	1	0	1	4	6
Positive	0	0	0	16	16
TOTAL	107	52	1	43	203

Malignancy was identified in 7 UNSAT (23%), 6 LTO negative (11%), and 4 SAT negative (3%) cases with a negative cytological diagnosis. The missed malignant tumors included 11 lymphomas (65%), 5 non-small cell carcinomas (29%), and 1 renal cell carcinoma (6%). There were no false positive diagnoses identified. The overall sensitivity and specificity were 80% and 100%, respectively. Of the 29 (14%) indeterminate cases, 10 (35%) were found to be malignant.
Conclusions: EBUS-FNA has a high specificity (100%) and sensitivity (80%) with few false negative diagnoses and no false positive diagnoses. Lymphomas were the most common cause for false negative results, followed by carcinomas. Ensuring adequate sampling is crucial in minimizing false negative diagnoses given that the percent of false negative diagnoses decreased from 23% in UNSAT cases to 3% in SAT cases.
Category: Cytopathology

Monday, March 19, 2012 9:30 AM

Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 67, Monday Morning