Several manufacturers, including Wilson-Cook, Boston Scientific, Olympus, and Medi-Globe, produce commercially available specialized needle assemblies for use with EUS . Our ­gastroenterology team typically uses 19-, 22-, and 25-gauge Wilson-Cook needles with an excursion of up to 8 cm beyond the distal end of the 125-cm-long echoendoscope . A 25-gauge needle is routinely used for the sampling of most lesions and produces sufficient quantities of minimally bloody specimens. When needed, a 19-gauge needle is used for obtaining tissue fragments from suspected stromal tumors and for performing immunohistochemical (IHC) stains in cell-block slides. About 22- and 19-gauge needles may be used for draining of fluid collections and sampling of mediastinal lymph nodes with sclerosis. Ancillary studies for lymphoma work-up are performed successfully in aspirates, with use of a 25-gauge needle. The use of a 19-gauge Trucut needle may result in more frequent complications (i.e., infection and bleeding) and may not increase the accuracy , particularly when mediastinal lesions are evaluated; however, the approach does appear safe and accurate for the sampling of intestinal wall and abdominal masses.

EUS identification, staging, and FNA , particularly of pancreatic masses, are technically challenging and require special training. Accuracy is operator dependent and correlates with experience. EUS competence is difficult to achieve even for advanced endoscopy trainees. The current recommendation of the American Society for Gastrointestinal Endoscopy (ASGE) for EUS competence is 150 EUS procedures; however, this number may not be sufficient and most trainees may require 250 or more. Factors that influence this variable rate include cognitive and technical skills of the operator and station/organ examined, i.e., esophagogastric versus pancreatobiliary imaging; other factors are availability of cases in the trainer center and continuous access to perform EUS exams. Ideally, EUS-FNA competence starts once EUS competence is reached. EUS-FNA sampling accuracy has been shown to increase from 33 to 91 % following a 2-month period of formal mentored training. The ASGE recommends a minimum of supervised 50 EUS-FNA procedures, 25–30 of which should be aimed to pancreas masses to reach the level of competence. The ASGE recommendation of 25–30 supervised EUS-FNA procedures for the diagnosis of pancreatic adenocarcinoma is supported by a study showing that there was a significant increase in the sensitivity of EUS-FNA diagnosis of pancreatic cancer, which plateaued at 80–90 % after 30 procedures. EUS-FNA cytology interpretation errors during the initial learning phase are primarily due to inadequate specimens.

The needle selection should be based on the physical characteristics and anatomic location of the target lesion.

One of the basic principles of aspiration of solid lesions is that the thicker the needle is and the more negative pressure applied, the more bloody and diluted the sample can be. Therefore for most cases, the tendency is to use one of the smallest caliber needles available: a 25-gauge needle.

A 25-gauge needle is usually initially utilized in most solid lesions. It is better to have the stylet in place completely occluding the needle lumen until the target lesion is reached. Having the stylet occluding the needle lumen before reaching the target tumor avoids carryover of nonlesion material/contaminants. These contaminants include gastrointestinal mucosal epithelial fragments, fluids from the lumen, or any tissue fragments form organs trespassed with the needle before reaching the tumor. These nondiagnostic elements cause marked dispersion of the diagnostic material, which gets diluted, swollen, and blurred by fluid or widely scattered between numerous non-lesional tissue fragments making it more difficult to interpret.

Once the stylet is completely removed, it is more useful to do the aspiration without suction (the cellular material gets in the needle lumen by the action of capillary forces due to the fast movement of the needle within the lesion). The use of a 25-gauge needle allows the operator to have a “feel” of the consistency of the lesion. On occasions when there is no too much physical distance to move the needle within the lesion or when after the first pass a very hard lesion is felt, application of negative pressure with a syringe is recommended.

When the needle is in the lesion, it is recommended to move it back and forth as fast as possible within the target lesion. Keeping the needle still or longer than 5–10 seconds (depending on Doppler examination) within the lesion often causes the specimen to clot or be too bloody resulting in “sausages” of blood and fibrin obscuring the diagnostic elements, which are difficult to smear onto a glass slide and interpret on cytological rapid ­on-site interpretation. When clotting occurs, attempts to make smears should be avoided and the material should be placed in fixative and submitted to the pathology laboratory to be processed as a cell block.

The consistency of the lesion felt by the operator who is moving the needle is also important in deciding the next step. For example, in gastric wall lesions, a soft consistency generally narrows the differential diagnosis to lipomas, neuroendocrine tumors , or heterotopic pancreatic tissue. Gastrointestinal stromal tumors (GISTs), Schwannomas, and leiomyomas are generally firm rather than soft.

Solid masses for practical purposes could be approached in three ways based on their consistency: hard, gritty or rubbery, and soft.

Hard masses: Most of these lesions are associated with a very fibrotic or desmoplastic stroma. In these cases, the lesion “grabs” the needle and it is difficult to move the needle within the lesion; not uncommonly the needle gets bent because of the attempts to move the needle fast. The best approach is to use a 22- or a 19-gauge needle and perform the FNA applying suction. The smears are generally not very cellular but are often diagnostic. Several passes might be required. It is advised to try as much as possible to sample different areas of the mass when diagnostic material is not obtained with the first attempt. The use of a 19-gauge needle is recommended when GIST is suspected to obtain material for a cell-block preparation and perform immunostains and molecular tests in the paraffin-embedded material.

Gritty or rubbery masses: These lesions could be sampled using a 25-gauge needle, with or without suction. The movement of the needle within the lesion has to be as fast as the size and location of the tumor allow it. Generally, 1–3 passes provide enough diagnostic material. It is important to keep in mind that when additional studies might be required (e.g., metastatic carcinoma of unknown origin, metastatic lung adenocarcinomas where epidermal growth factor receptor (EGFR) testing might be requested, metastatic colorectal cancer where KRAS gene mutations are often needed, etc.), additional passes up to the discretion of the cytopathologist and interventional gastroenterologist are often required.