Reliability of gross visual assessment of specimen adequacy during EUS-guided FNA of pancreatic masses.

摘要

BACKGROUND: In many centers, on-site cytopathologists are not available during EUS-guided FNA (EUS-FNA) examinations. Often, endosonographers request that technologists assess the adequacy of FNA by gross inspection of the slides. To date, there has not been a study that assessed the accuracy of experienced technologists in predicting tissue sampling adequacy by gross inspection before cytologic staining. OBJECTIVES: To assess a grading system used by cytotechnologists and EUS technologists during gross inspection of FNA slides in reliably predicting specimen adequacy compared with the final cytologic diagnoses. DESIGN: Prospective, double-blind, controlled study. SETTING: Academic tertiary-referral center with a high-volume EUS practice. PATIENTS: Fifty-one patients with a suspected solid pancreatic mass who were undergoing planned EUS-FNA. MAIN OUTCOME MEASUREMENTS: The degree of correlation in the assessment of specimen adequacy as exhibited by a weighted kappa statistic between 2 groups of technologists and a board-certified cytopathologist. RESULTS: FNA was performed in 37 cases with 234 individual slide specimens available for analysis. Only fair agreement was observed between cytotechnologists and EUS technologists versus final cytopathologic assessment of adequacy (kappa 0.20 and 0.19, respectively). The routine practice of 6 to 7 FNA passes yielded adequate tissue for assessment in 36 of 37 patients (97%). LIMITATIONS: Interobserver variability, single center, and findings applicable only to solid pancreatic lesions. CONCLUSIONS: Neither trained EUS technologists nor cytotechnologists were able to provide a reliable assessment of pancreatic-mass FNA adequacy by using gross visual inspection of the specimen on a slide. Rapid on-site cytopathology reduced the number of passes, ensured specimen adequacy, provided definitive diagnosis, and should be used in centers where available.