This was a well-designed, prospective, blinded study that addressed the question of whether EUS-guided fine needle aspiration (FNA) cytology of peri-esophageal cancer lymph nodes improves staging accuracy and changes patient management. For N- staging, the study found the accuracy of EUS alone was 81% versus 61% for helical CT scan. They found that the accuracy of EUS/FNA was 87% compared to 51% for CT scan (p<0.001). When comparing EUS-FNA versus EUS alone, there was a significant difference in accuracy of 87% versus 74% (p=0.012), although there was not a statistically significant difference in sensitivity or specificity. When looking at whether EUS-FNA changed management compared to CT or EUS alone, the study found that compared to CT scan, EUS-FNA changed the tumor stage of 33% of their patients to a higher or worse stage, and 5% to a lower stage. This was associated with a greater rate of treatment strategies that were not direct surgeries. There was no significant change in the distribution of tumor stages, or management, when comparing EUS alone versus EUS-FNA.

This study shows that EUS is more accurate than spiral CT scan for staging esophageal cancer. The addition of EUS-FNA of lymph nodes significantly increased the nodal staging accuracy.

The current staging system for esophageal cancer is considered imperfect by some expert surgical oncologists, and revisions have been recommended. It has been proposed that N status be determined by the number of metastatic regional lymph nodes, with N1 (1 or 2 nodes) and N2 (3 or more nodes), rather than by anatomic distribution. It is also proposed that celiac lymph nodes in patients with gastroesophageal junction adenocarcinoma should be considered as regional lymph nodes instead of distant metastatic disease. It is often very difficult, and somewhat arbitrary, to classify these lymph nodes with EUS as celiac (M1a), rather peri-tumor nodes (N1). If celiac nodes are treated as regional lymph nodes, then there may be less importance for obtaining EUS-FNA cytologic diagnosis in the future.

EUS imaging, with or without the addition of FNA, is the most accurate imaging modality for locoregional staging of esophageal cancer and should be considered the "standard of care." As efforts continue to refine the staging system and treatment algorithms for esophageal cancer, EUS-FNA will remain the main method for providing locoregional staging.

This prospective study aimed to assess the safety, yield and clinical utility of EUS with a 19 gauge trucut needle biopsy (TNB) of solid pancreatic lesions. Twenty-three consecutive patients with pancreatic masses discovered on CT scan underwent EUS with TNB. Pancreatic tissue was obtained in 17 of 23 patients. In all 13 patients with pancreatic body lesions, EUS-TNB was successful, but only 4 of 10 patients had successful EUS-FNB when the mass was in the pancreatic head. No complications were observed. EUS-FNA was performed in 12 of 17 patients who had successful EUS-TNB, with an average of three needle passes performed. Of these 12 patients, EUS with FNA established the correct diagnosis in 50%, compared with 83% for EUS-TNB.

Pancreatic tissue core biopsy specimens may have theoretic advantages over cytologic samples. Tissue architecture is better preserved compared to FNA, special staining may be easier to perform, and an on-site cytopathologist is not needed which could lower cost. However, use of this technique via EUS is still limited by technical difficulties with the needle itself. The EUS trucut needle is difficult to use for trans-duodenal biopsy, as demonstrated by only 40% of patients with pancreatic head masses having successful biopsies. Sampling error may also be an issue with the trucut needle as one pass is usually performed, compared to multiple passes with EUS-FNA. The low diagnostic rate of EUS-FNA in this study is probably due to the fact that no cytologist was present during the procedure, and that only a mean of three passes were obtained. As the authors suggest, EUS-TNB may be most useful for obtaining tissue from the pancreatic body and tail. It should be noted that the cost of this needle is roughly twice that of a standard EUS-FNA needle. There may be a role in selected situations for using an EUS trucut needle, but EUS-FNA cytology should still be considered the procedure of choice for pancreatic mass sampling.

Forty-four patients with strictures in the liver hilum diagnosed by CT and/or ERCP that were suspicious for hilar cholangiocarcinoma, and had previous non-diagnostic attempts at tissue diagnosis, were prospectively enrolled. Patients underwent EUS with FNA of the hilar lesions with 22-gauge needles. Adequate material was obtained in 43 of 44 patients. Cytology revealed carcinoma in 26 and other malignancies in 5. Twelve patients had benign cytology. Four of the benign lesions were false negatives. Accuracy, sensitivity and specificity, and positive and negative predictive values were 91%, 89% and 100%, 100% and 67%, respectively. EUS with FNA changed the pre-planned surgical approach in 61%.

EUS-FNA of the proximal bile duct is often not routinely performed. This is due, in part, because it is in a difficult area to visualize. In this study, a plastic biliary stent was used as a guide to help find the hilar lesion. To aid in FNA, the echo-endoscope was placed in the "long position" against the greater curve to provide support for the scope when FNA was performed. While the positive predictive value of FNA was good, the negative predictive value was only 67%. Note that the risk of EUS-FNA causing cholangitis is significantly increased in obstructed biliary systems, and that consideration may first be given towards ERCP and stent placement with biliary drainage, unless the needle can be placed directly into the mass and avoid the bile duct. EUS-FNA should be considered for obtaining tissue diagnosis of proximal biliary and hilar strictures.