Improving the Accuracy of Ultrasound-Endoscopy Fine- Needle Aspiration (EUS-FNA): Factors Predicting a Correct Diagnosis and Usefulness of an Attendant Pathologist

Abstract

Background: EUS-FNA is a well established technique for staging gastrointestinal and lung cancer and for diagnosing pancreatic masses. Aim: To evaluate the predictive factors of a correct diagnosis by EUS-FNA and to establish the usefulness of an attendant pathologist. Patients and methods: Between January 2001 and February 2003, we evaluated all consecutive patients referred for EUS-FNA. The procedure was carried out with a linear echoendoscop and a 22G needle. An attendant pathologist examinated the samples. Epidemiological data, EUS characteristics,and citological results were registered. EUS-FNA diagnosis was compared with the gold standart (pathology of resected specimen or clinical follow-up). In order to establish the importance of an attendant cytopathologist, the actual results were compared with those that would have been obtained if a particular number of passes had been performed without on-site evaluation. Once a price per pass was established, a cost-minimization study of the availability of an attendant pathologist was performed. Results: 262 lesions were sampled: pancreatic masses (n= 115), lymph nodes (n= 96), cysts (n= 40) and intramural lesions (n= 11). A total of 551 samples was obtained, thus representing 2.1 +/− 1.1 passes per patient on average (range, 1-6). EUS-FNA ascertained the correct diagnosis in 235 (overall accuracy, 89%). Location within the gastrointestinal wall was the only independent factor predicting failure in the diagnosis (multivariate regression analyses). Effectiveness of EUS-FNA in the whole series increased with every pass from 36% to 89% plateauing in the 4th pass. This curve was similar for pancreatic masses, lymph nodes and cyst but the plateau appeared in the 3rd pass in these last two types of lesions. The effectiveness of EUS-FNA for the intramural lesions was always lower, increasing from 0.9% to 45% and reaching a plateau in the 4th pass. The assistance of an attendant cytopathologist was costless with respect to the corresponding strategy without on-site examination for the whole series and for lymph nodes, masses and cysts. Conclusion:1) Location within the gastrointestinal wall was the only independent factor predicting failure in the diagnosis. 2) The availability of an attendant pathologist seems to increase the diagnostic yield of EUS-FNA, minimizing the number of passes and resulting in a costless strategy.