ORIGINAL ARTICLE: Clinical Endoscopy Can endosonographers evaluate on-site cytologic adequacy? A comparison with cytotechnologists Alan D. Savoy, MD, Massimo Raimondo, MD, Timothy A. Woodward, MD, Kyung Noh, MD, Surakit Pungpapong, MD, Arthur D. Jones, MD, Julia Crook, PhD, Michael B. Wallace, MD, MPH Jacksonville, Florida, USA Background: On-site determination of cytologic adequacy increases the accuracy of EUS-guided FNA (EUS- FNA); however, on-site cytotechnologists are not available to all endosonographers. We hypothesize that expe- rienced endosonographers can accurately assess whether an on-site FNA specimen is adequate. Objective: To determine the accuracy of on-site cytopathology interpretation of EUS-FNA specimens by com- paring endosonographers with a cytotechnologist. Design: Prospective double-blind controlled trial. Setting: Academic medical center with a high-volume EUS practice. Patients: Consecutive patients undergoing EUS-FNA of lymph nodes or pancreas tumors. Main Outcome Measurements: Accuracy, sensitivity, and specificity of 3 endosonographers and 1 cytotechnol- ogist for interpretation of cytologic specimen adequacy and diagnosis compared with a criterion standard of a board-certified cytopathologist. Results: There were 59 lymph node, 49 pancreas, and 9 liver specimens (117 total). For determination of ad- equacy, none of the endosonographers were statistically equivalent to the cytotechnologist (P Z .004). For de- termination of suspicious/malignant versus benign specimens, all 3 endosonographers were inferior (P ! .001) to the cytotechnologist. Limitations: This study represents a small group of trained endosonographers in a high-volume practice and may not be applicable to other settings. The sample size does not allow an accurate evaluation of different bi- opsy sites (eg, pancreas vs lymph node). Conclusions: Even trained endosonographers have variable and, in some cases, inferior abilities to interpret on-site cytologic adequacy compared with cytotechnologists. (Gastrointest Endosc 2007;65:953-7.) EUS-guided FNA (EUS-FNA) is a minimally invasive, ac- curate, and safe method of diagnosing and staging GI and thoracic malignancy. EUS-FNA is more accurate than EUS imaging alone. 1-13 A key issue in maximizing FNA accuracy is to ensure that an adequate specimen is obtained before termination of the procedure. One study demonstrated that having on-site cytopathology increases the diagnostic yield of EUS-FNA. 14 Interpretation of cytology specimens is challenging and requires adequate cellular material (both benign and malignant) to render a definitive diagno- sis and is generally performed by cytotechnologists (who can interpret adequacy) or physician cytopathologists (who can interpret both adequacy and diagnosis). Despite its benefits, on-site cytopathology may increase the time and the expense of EUS procedures and is not routinely available, especially in ambulatory surgical cen- ters. Historically, EUS has been limited to a few highly ex- perienced centers. However, EUS has rapidly expanded and is now being performed by increasing numbers of gas- troenterology practices around the world. We hypothesized that physicians trained in EUS (includ- ing medical school training in pathology) are capable of in- terpreting cytologic adequacy from EUS-FNA specimens. If that is true, on-site interpretation of cytologic adequacy by the endoscopist could reduce cost and procedure duration, Copyright ª 2007 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2006.11.014 www.giejournal.org Volume 65, No. 7 : 2007 GASTROINTESTINAL ENDOSCOPY 953