Pancreatic cancer represents the fourth most common cause of cancer-related deaths in the United States and in the next 15 years it is expected to become the second most common. The standard of care for borderline resectable, locally advanced, and metastatic pancreatic cancer is chemotherapy, and therefore tissue acquisition for diagnosis is required before treatment is initiated. For the past 25 years, endoscopic ultrasound-guided fine-needle aspiration (FNA), when available, has been the most frequently used technique for acquisition of tissue in these patients. This article summarizes both the author’s opinion and current best practices for performing this procedure.

Dr. John DeWitt is the Professor of Medicine and Director of Endoscopic Ultrasound at Indiana University. He has achieved national and international recognition for research in diagnostic and therapeutic endoscopic ultrasound and pioneered the use of endoscopy for the nonoperative ablation of pancreatic cystic neoplasms. He is internationally renowned and has served in various international endoscopy organizations in leadership roles.

Dr. John DeWitt









Endoscopic ultrasound and biopsy of solid pancreatic lesions

  1. EUS-guided tissue acquisition is performed with a linear echoendoscope. This endoscope permits visualization of the needle as it passes into the target of interest. In the United States, sonographic imaging places orientation of the needle trajectory from a top right to lower left side position. Therefore, it is recommended that the target lesion be placed in the middle or slightly to the left side of the screen. This orientation minimizes the need to use the elevator, the deployment of which can make it harder to perform the biopsy. It is critical to make sure that the sheath is out of the echoendoscope prior to performing the biopsy in order to minimize risk of damage to the endoscope. 

  2. Tissue sampling can be performed with either fine-needle aspiration (FNA) or fine-needle biopsy (FNB) needles. Needles are made from different materials including aluminum, stainless steel, and nitinol. Nitinol needles are more flexible and are preferred for biopsies requiring angulation, such as those from the distal duodenum. All FNA needles from different manufacturers are essentially the same. However, there are three main types of FNB needles. These include: (a) reverse-bevel or, more recently developed, forward-bevel (Pro-Core, Cook Medical, Bloomington, Indiana, USA); (b) fork-tip (Shark Core, Medtronic, Minneapolis, Minnesota, USA); and (c) Franseen (Acquire, Boston Scientific, Marlborough, Massachusetts, USA).

  3. Compared to FNA, EUS-FNB sampling appears to increase the total amount of tissue obtained and decrease the number of passes required to diagnose malignancy in pancreatic masses. It remains unclear whether any needle size (19G vs. 22G vs. 25G) or needle type (FNA vs. FNB) increases overall diagnostic accuracy.

  4. Reverse-bevel FNB needles are probably inferior to the other two FNB needle types, and their use is not recommended. Fork-tip and Franseen needles are probably equivalent.

  5. The use of FNB obviates the need for rapid on-site evaluation (ROSE) in most cases. Therefore, the use of ROSE should be personalized for each endoscopist. When ROSE is not available, 4–6 passes with an FNB needle is recommended rather than use of an FNA needle. Use of thinner 22G or 25G FNB needles may help to increase technical success in difficult locations such as the pancreatic tail or uncinate process.

  6. The greater use of personalized medicine and genomic evaluation should further increase use of FNB compared to FNA. In the future, use of FNA may be limited to tissue acquisition in patients with intervening vessels, to highly vascular tumors such as neuroendocrine tumor, or to intravascular biopsy.

  7. Tissue sampling presents several different technical options. The use of standard 10–20-mL suction appears to increase diagnostic yield compared to absence of suction. However, increasing the negative pressure to 50mL does not appear to improve accuracy and is not recommended.

  8. FNA and FNB needles come with a stylet in the packaging. The use of a stylet does not appear to improve diagnostic accuracy, specimen adequacy, or yield. Therefore, after the initial biopsy has been performed the stylet may be discarded. Slow removal of the stylet, the so-called “slow pull” technique, also does not appear to improve diagnostic accuracy.

  9. Samples obtained by EUS may be processed by several different formats. Specimens may be: (a) air-dried and taken rapidly to the lab; (b) fixated with a spray-based fixative; or (c) transported in a liquid medium. Cell blocks can be created from FNAs and this may often provide a sample similar to that seen with histologic samples. Cell block or histology specimens are preferred for immunostaining or genomic testing.

  10. EUS-FNA and FNB of pancreatic masses is a safe technique. Bleeding is self-limited in almost all cases. Pancreatitis may occur after 1%–2% of biopsies. Antibiotics are not required for biopsy of solid lesions.