Difficult Location of Lesions

Whenever an EUS-guided biopsy is attempted, the position of the transducer in relation to the lesion is of major importance. The endoscope should always be advanced to a position that brings the transducer as close to the lesion as possible. However, there are some technical challenges related to the lesion and its surroundings that may influence the biopsy procedure. The most difficult regions to be reached by the echoendoscope are the deep part of the uncinate process, the second duodenum, the gastric fornix, and near the greater curvature. Other lesions, such as adjacent ulcers, scars, or diverticula, impede the best visualization of the lesion, and consequently subsequent EUS-guided puncture. One special situation is when the papilla is situated inside a diverticula. By using water instillation, it is frequently possible to visualize the papillary and suprapapillary regions. Avoiding puncture of vessels and pancreatic ducts should always be attempted ( Fig. 1 ). Lesions located in the hepatoduodenal ligament or its surroundings in particular may pose a challenge because of an interposed vessel, such as the portal vein or the hepatic artery or its branches ( Fig. 2 ). Fortunately, it is often possible to find an adequate window to perform an FNA without the presence of intervening vasculature or ductal structures. Before the procedure, anticoagulants should be stopped for 5 days and antiaggregants (eg, clopidogrel) should be stopped for for 5 to 7 days.

Hypoechoic mass in the pancreatic isthmus with invasion of the splenic vein, portal confluence, and superior mesenteric vein. The FNA needle should avoid puncturing the vessels and the main pancreatic duct.

Tumor in the pancreatic head invading the splenomesenteric confluence, which makes FNA technically difficult using the current position of the echoendoscope.

Characteristics of Lesions

The characteristics of a lesion that is targeted during FNA influences the diagnostic results to a high degree. Diagnostic success is affected in particular if a lesion is hard and fibrotic, such as in solid pancreatic masses. The same is the case in lesions that are highly mobile, such as small submucosal tumors in the stomach. The best location for tumor sampling is thought to be the periphery of the tumor. The central part of the pancreatic tumor should be avoided because of extensive necrosis. The presence of acute pancreatitis, necrosis, or severe chronic pancreatitis with coexisting fibrosis could be misinterpreted as perfusion defects during contrast enhancement or may be mistaken for cancer during elastography. Although these techniques could guide the tissue sampling, their incremental value has not been established yet. The presence of chronic pancreatitis features was associated with a lower accuracy of EUS-FNA for the differential diagnosis of pancreatic masses (73% vs 91%) and might need a higher number of passes for establishing the diagnosis. The presence or absence of stents (either plastic or metallic) usually does not impede EUS-FNA, although some suggest better outcomes when the stent is placed just 1 day before performing EUS-FNA.

Most other authors have shown that there is no difference in diagnostic accuracy when a lesion is more than or less than 3 cm in diameter. However, a single study has shown that the size of the tumor may be a restrictive factor in reaching an adequate diagnosis, especially for tumors less than 1 cm, where the sensitivity was found to be as low as 40%. Concerning the tumors of 1 to 2, 2 to 3, 3 to 4, and greater than 4 cm, the diagnostic sensitivity was 75.9, 86.9, 93.2, and 91.6%, respectively. The presence of intratumoral anechoic foci may require a higher number of passes.

Aspiration from the edge of the lymph node (compared with the center) did not increase the likelihood of a correct diagnosis. Coagulation necrosis within a lymph node on EUS was described as an ill-defined, rounded, and nonshadowing echogenic area. It is different from the linear, nonshadowing, hilar echogenic structure that may be seen in the center of benign lymph nodes and has a 70% accuracy for diagnosing malignancy on FNA.

The results of a retrospective study on 107 patients showed that targeting the cyst wall of cystic lesions of the pancreas produces an increasing diagnostic accuracy of up to 56%.

Impaired Passage or Altered Anatomy

The presence of an esophageal stricture can preclude echoendoscope passage. Then, either dilation or the use of miniprobe/slim scopes may be proposed. Miniprobes or nonoptic radial endoscopes enable only diagnostic evaluation, thus to perform FNA, a slim, linear echoendoscope or an endobronchial echoendoscope could be the answer. The endobronchial ultrasound videoscope with a thinner diameter than the conventional echoendoscope (6.2 mm instead of 12.8 mm) has been used successfully in several cases to guide FNA. The main limitation is the impossibility to reach the head of the pancreas because the working length of this endoscope is too short. The lack of insufflation can be resolved by manual air insufflation that enables advancement of the scope and decreases the risk of esophageal perforation.

The presence of food or fluid inside the esophagus or stomach, due to either an esophageal or a pyloric or duodenal stenosis, may impede the examination. In this case, the procedure should be postponed and the patient kept on a liquid diet for 2 to 3 days.

Previous surgery, a dominant pancreatic mass, or a duodenal stenosis may prevent the echoendoscope from reaching the second part of the duodenum. In most situations, the FNA can be performed from the duodenal bulb, with the help of an assistant, to maintain the scope position, or from the antrum, with much care to avoid vessels when targeting pancreatic head lesions. The main limitations are the view of the head of the pancreas after Roux-en-Y surgery. Nevertheless, in one study, most patients were able to get a complete pancreaticobiliary examination, including FNA, after Billroth II and Whipple surgeries, provided that the afferent limb could be intubated. Deflecting the tip of the echoendoscope at the level of anastomosis increases the visual field for sonographic examination.

Inadequate EUS Imaging

A successful biopsy specimen is dependent foremost on adequate EUS imaging so that the position of the needle can be monitored at all times. There are several factors that may impair optimal ultrasonic visualization, such as problems with balloon inflation, air in front of the transducer either due to air in the balloon or in the lumen, impaired acoustic coupling, or endoscope torquing.

It is rarely necessary to fill the balloon with water before or during EUS-guided biopsy. However, in a few indications this may be needed, either to anchor the transducer in the duodenum or to optimize needle visualization during biopsy. The presence of air in the balloon represents a technical problem that may impair the examination. In this situation, aspiration of the balloon content and subsequent refilling may solve the problem but often complete removal of the echoendoscope is necessary to remove the air bubbles completely from within the balloon. The presence of the needle inside the working channel may impede supplementary air suction, especially for nontherapeutic linear echoendoscopes with a thin working channel. If there is an accumulation of air in front of the targeted lesion, the procedure cannot be continued, because of the lack of visual control of the needle at the puncture site.

There are several situations in which the endoscopist has to torque or bend the echoendoscope, thus making the FNA difficult or impossible. The first situation is encountered in the second portion of the duodenum, where the tip of the endoscope is angulated and the needle exit for targeting lesions in the head of the pancreas or distal common bile duct is sometimes challenging. If the needle is forced to exit, the transducer might be damaged by accidental puncture. This damage could be avoided by having the sheath of the needle in the endoscopic view or by seeing the needle tip in the ultrasound view. For transduodenal targeting of the lesions, shortening of the echoendoscope is the best course of action. FNA can be performed safely with the help of an assistant to maintain the position of the echoendoscope. However, if a periduodenal compressive lesion limits straightening of the echoendoscope, then the position of the scope is maintained, the deflection of the tip of the scope is loosened, and the needle is advanced until it is in endoscopic view. The tip of the echoendoscope is then deflected again so as to bring the lesion within the trajectory of the needle. Air suction also helps to bring the gut wall closer to the probe.

The second situation in which torquing of the echoendoscope may impede FNA is the location of a lesion in the gastric fundus. If attempts to reach the lesion from the level of cardia fail, the needle must be advanced out of the working channel with the echoendoscope in a straight position before the scope is looped to reach the lesion.

A third situation in which torquing of the echoendoscope may influence FNA is in the stomach, due to the large gastric volume and the mobile gastric wall. When a lesion inside the gastric wall is targeted and biopsy is difficult or unsuccessful, one possibility is to attempt a 2-step puncture: the first step is to puncture the wall; the second step is then to target the lesion itself. Another possibility is to advance the scope further into the stomach to form a long loop, against the opposite gastric wall, so that the tip of the scope acts as a hinge. Rapid (sharp) movements of the needle during transgastric FNA is preferable to avoid “bouncing” of the needle due to gastric mobility.

Choice of Needle

Currently, there are several types of needles for FNA and biopsy (histologic analysis): 22 G, 19 G, and 25 G. The choice of needle is based on various considerations such as the potential for complications, adequacy of specimen, and cost.

The most commonly used needle is a 22 G, which is flexible and enables cytologic assessment without a significant risk for complications, although a 2% risk of acute pancreatitis was reported in a large retrospective series. Nevertheless, hard lesions are sometimes better penetrated with the 25-G needle, which is thinner and easier to use in the distal duodenum (uncinate process) or to access lesions farther away from the transducer. For uncinate process lesions, the use of the 22-G needle has been reported to be unsuccessful in up to 33% of cases. Experimental laboratory simulations that compare resistance to needle advancement have shown that only the 22-G and 25-G needles are suitable for insertion into the target regions if tight angulation is present.

The diagnostic accuracy for pancreatic masses when using the 22-G needles is up to 95% and as low as 68% for subepithelial lesions. A meta-analysis that compared the 22-G and 25-G needles for FNA of pancreatic and peripancreatic masses showed nonsignificant differences in sensitivity (78% vs 91%), specificity (100%), and complications ( Table 1 ). Repeating the EUS-FNA in the case of initial negative cytology increases the yield of diagnosis.

The overall rates of diagnostic adequacy for sampling pancreatic masses for cytology using the 22-G needle are variable compared with histology (82%–93% vs 84%–87%). The overall diagnostic accuracy for histology on each pass is only 60% for the 25-G needle and 75% for the 22-G neeedle. Histology showed a trend for superiority over cytology only in characterizing nonadenocarcinoma tumor types or inflammatory masses with a diagnosis being established in 90% to 100% of cases. For submucosal lesions, the adequacy of FNA for histology is reported to be 43% to 100%. The European Society of Gastrointestinal Endoscopy (ESGE) recommends routine procurement of tissue for histology when sampling subepithelial lesions.

The 19-G aspiration needles are more rigid and transduodenal biopsies are difficult and sometimes impossible to use ; thus, their use is not routinely recommended. The diagnostic accuracy for body/tail pancreatic lesions was shown to be superior when using 19-G needles compared with 22-G needles, offering more cytologic material in some circumstances. A large step forward when using the 19-G needle is the ability to procure histology samples that offer more information about the diagnosis of rare tumors, metastases, lymphomas, and tumor differentiation for neuroendocrine tumors or subepithelial lesions. However, there seems to be no advantage for the 19-G needles over 22-G needles for sampling subepithelial lesions. In a large multicenter European study, it was feasible to use the 19-G FNA needle in only 65% of cases and 41% of these were not relevant for diagnosis. Immunohistochemistry was feasible in most cases with successful tissue procurement (91%). The overall diagnostic yield was only 52%, with mild bleeding occurring in 22% of patients and even a case of death due to abscess formation.

Trucut biopsies (TCB) using 19-gauge EUS-TCB needles (Quick-Core; Cook Medical, Limerick, Ireland) are recommended when EUS-FNA turns out to be nondiagnostic, due to insufficient biopsy material (tumor cells present but of insufficient quality for definitive diagnosis or bloody specimen) as well as in cases where tissue architecture is required for diagnosis but not present (subepithelial masses, autoimmune pancreatitis, lymphoma). It is recommended to use with caution in the esophagus, rectum, and most of the stomach. Nevertheless, the device functions with difficulties in the antrum, fundus, and duodenal bulb, where echoendoscope angulation produces sluggish advancement of the cutting sheath over the specimen tray. Their lack of flexibility as well as the mechanical friction to the firing mechanism produced by the torqued echoendoscope when transduodenal biopsies are required limits their clinical usefulness. When performing transduodenal biopsies or at acute angulations, there is a need for straightening the tip of the echoendoscope, by releasing the elevator and the tip deflection; otherwise, the working channel of the echoendoscope may be damaged when the needle is fired. Close attention should be paid to avoid traversal of the lesion and accidental puncture of adjacent structures because the needle tip advances to 20 mm on being fired.

The flexible 19-G needle (Flex 19; Boston Scientific, Natick, MA, USA), made of nitinol, could be released easily, irrespective of whether the echoendoscope tip is in a torqued or a straight position, and it did not lose its configuration, despite multiple passes. In a study of 38 patients, core tissue procurement was optimal in all but 2 patients. Although on-site pathology was normal in 92% of patients, histology was required to establish a final diagnosis in 3 patients. Both the 19-G flexible needle and the latest 19-G histologic needles (ProCore; Wilson-Cook, Limerick, Ireland) have largely replaced the 19-G trucut needle for performing core biopsies. In another study, the high-definition 22-G histologic needle (ProCore; Wilson-Cook Limerick, Ireland) yielded adequate diagnostic samples with fewer passes. A recent randomized study compared conventional 22-G needles and histologic 22-G needles (ProCore; Wilson-Cook, Limerick, Ireland) for evaluation of pancreatic masses with no differences in tumor location or size between both cohorts. The diagnostic accuracy was 100% in the case of the FNA needle and 89% in the case of ProCore needle, with the safety factor being similar for both needles. Likewise, a 25-G histologic needle used in 50 consecutive patients showed a cumulative sensitivity of 83%, 91%, and 96% when cytologic analysis was performed on passes 1, 2, and 3, respectively.

A pilot study of using echo-brush for sampling cystic pancreatic lesions showed superiority for diagnosis in 7 of 10 cases, compared with EUS-FNA, while some cases of intracystic bleeding were observed. Despite initial optimism for its use, the adequacy of cytology for solid lesions was 53% and for cystic lesions of the pancreas it was only 50%. Lower results were obtained for subepithelial lesions and lymph nodes.

Recently, an algorithm has been proposed to choose needles: 25-G needles for transduodenal FNA, 22-G or 25-G needles for other FNA, 19-G flexible needles for transduodenal interventions, and standard 19-G needles for interventions via other routes.

Choice of Biopsy Method

The use of a stylet is considered a necessity to reduce the contamination with cells from superficial layers of the gut wall and to offer higher stability to the needle, although it enhances the rigidity of the needle apparatus when performing FNAs. Working without a stylet is possible; however, for inexperienced endosonographers the authors strongly advise using the stylet for protection of the echoendoscope channel during introduction of the needle. Several trials revealed that the diagnostic yield of malignancy, amount of blood, cellularity, contamination, and adequacy of the specimen did not differ significantly whether the stylet was used or not. No conclusion has been drawn yet, but ESGE recommendations are that the decision on use of stylet be left to the discretion of the endosonographer performing the procedure.

It is controversial whether the use of suction during EUS-FNA improves the diagnostic accuracy. However, due to the capillary effect, especially in vascular lesions, the risk of bloody samples is higher. In the authors’ study of 52 patients with solid masses in whom 3 passes were randomized to suction or without suction, the use of suction was associated with higher diagnostic sensitivity and negative predictive values but without increasing specimen bloodiness. High-pressure suction applied to a commercially available 22-G FNA needle yielded tissue samples for histologic examination in 96% of the cases and also enabled immunostaining. The recommendation of ESGE is to use suction for solid masses and cystic lesions and to avoid it for lymph nodes sampling.

The fanning technique, which involves sampling multiple areas within a lesion with each pass, was compared in a randomized trial with the standard to-and-fro technique. Although there was no significant difference in diagnostic accuracy between both techniques (76% vs 96%), the fanning technique facilitated a first-pass diagnosis in 85% of patients compared with less than 60% with the standard technique. Using the standard technique, a survey between endosonographers reported that the accuracy for pancreatic FNA was higher than 80%. The predictive factors for higher diagnostic accuracy were more than 100 EUS-FNA case loads/year with more than 7 FNA passes being performed and the availability of rapid on-site cytopathologic evaluation or routine procurement of microcore.

Initial studies indicated that the number of passes for pancreatic lesions should be 5 to 6 and for the liver or lymph nodes should be 2 to 3. Whatever the type of lesion, 3 or more passes increase the diagnostic accuracy. A later study observed that for pancreatic lesions no more than 7 passes (22-G needle) are necessary to obtain an accuracy of greater than 80%. Another study observed that when performing 4 passes using the 25-G needle on solid pancreatic mass lesions, the diagnostic accuracy was similar whether an onsite cytopathologist was present or absent. In a retrospective study, a mean of 2 passes with combined histology and cytology provided adequate tissue for evaluation of pancreatic masses. When trucut biopsy is performed, more than 2 passes are usually necessary for improved diagnosis, whereas for subepithelial lesions, 5 passes were associated with a diagnostic accuracy of 49% to 61%.

Although the yield of cytology for cystic lesions of the pancreas is reported to be as low as 20% to 30%, when the cysts have a solid component, performing a higher number of passes improves the diagnostic accuracy (78% vs 44%). Inadequate fluid content sampling might occur in about 50% of the cases for several reasons: the cyst content might be very viscous contributing to false negative results and the limited number of passes performed (due to potential risk of infection) could lead to minimal fluid extraction and thereby yield inadequate or nondiagnostic samples. The use of 19-G needles to facilitate large volume cyst aspiration or targeting the cyst wall at FNA may yield better results.

The ESGE recommends performing 3 needle passes when sampling lymph nodes and liver lesions, 5 passes when sampling solid pancreatic masses, and, a single pass when sampling pancreatic cysts due to the potential for inducing an infection.

It has been demonstrated that gross visual inspection of the specimen on a slide by a trained EUS technologist or cytotechnologist was not able to predict the final cytopathologic assessment for adequacy. The cytopathologic on-site rapid assessment of slides is reported to be better than monolayer prepared slides. However, it is not clear if the presence of a cytopathologist ultimately improves the diagnostic accuracy. Nevertheless, for suspicious lymph nodes or pancreatic masses, the diagnostic accuracy is as high as 96% with a cytopathologist versus 84% without a cytopathologist. The presence of the cytopathologist also determined the need for fewer passes, lower number of inadequate samples, and higher diagnostic sensitivity. Other studies showed that even the presence of a cytotechnologist can improve the interpretation for diagnostic adequacy at tissue sampling. In the absence of a cytopathologist, the cell block evaluation of different lesions yield better diagnostic results than smear cytology (92% vs 60%). For lesions accessible through the esophagus or stomach, the accuracy of EUS-TCB was equal to FNA when no on-site cytopathologist was present. Specimen adequacy can be also assessed directly by the endosonographer.

Technical Challenges During Biopsy

There exists several technical challenges during biopsy, some of which are related to designs of the needle or the echoendoscope. The rate of needle dysfunction was reported to be as high as 14% for the 22-G needle, up to 25% for the 25-G needle, and even higher for the 19-G/Trucut needle (5%–50%). Even the use of 22/25 G needles was associated with technical difficulties during transduodenal punctures (24% vs 4%).

Friction between the needle and the sheath is often seen during FNA via the second part of the duodenum. If this becomes problematic, it may be necessary to withdraw the echoendoscope into the stomach, adjust the length of the needle sheath as desired, and then go back through the pylorus into the duodenum and perform the FNA.

When a needle is advanced in a therapeutic channel echoendoscope, the size of the needle with its sheath covering may be unstable inside the oversized channel, resulting in insufficient needle monitoring because the unstable needle bends out of the image plane of the transducer. The solution to such a problem is to choose a needle with an appropriate sheath stabilizer (Medi-Globe GmbH, Grassau, Germany) that fits better to the size of the biopsy channel or to exchange the scope for one with a smaller channel ( Figs. 3 and 4 ).

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

New Methods to Obtain Better Tissue Samples with EUS-guided FNA Preface: EUS-Guided Tissue Acquisition The Changing Paradigm in EUS-Guided Tissue Acquisition Endoscopic Ultrasound-Guided Fine-Needle Aspiration Needles Techniques for EUS-guided FNA Cytology Techniques for Endoscopic Ultrasound-Guided Fine-Needle Biopsy

Stay updated, free articles. Join our Telegram channel

Join