EUS-guided FNA has been proved to be a safe and useful method for tissue sampling of gastrointestinal track lesions and other organ lesions including mediastinal and intra-abdominal lymph nodes, pancreas and hepatobiliary tree. The usefulness of EUS-FNA depends on several factors. For example, experience of the endosonographers, adequate sampling, sample preparing, accurate interpretation by the cytopathologist and on-site cytopathology interpretation. However, in many hospitals, no cytopathologist can be present during EUS-FNA. Therefore, determining appropriate methods to obtain and prepare EUS-guided FNA are important to make correct a diagnosis without on-site cytopathologist.

Suction with a self-retracting 10-mL syringe will likely bring in more cellularity but also more blood. Some endosonographers use no suction, others use constant suction. Usually specimen is expelled from a needle with pushing the stylet into the needle. But use of the stylet during EUS-FNA is difficult and time consuming process. Injecting air was not recommended, because of spraying out uncontrollably, increasing risk of air artifact and specimen clotting. However, there is no further study which one is the appropriate, suction or no suction and pushing the stylet or injecting air until now.

The hypothesis and aim of the prospective randomized controlled trials are as follows:

First hypothesis: There was no difference in the adequacy, cellularity, bloodiness, contamination, air artifact in specimen obtained by each methods, suction or no suction and pushing the stylet or injecting air.

Aim #1 : To compare the adequacy, cellularity, bloodiness, contamination, air artifact in specimen obtained by each methods, suction or no suction and pushing the stylet or injecting air.

Second hypothesis: There was no difference in sensitivity, specificity, diagnostic accuracy, positive predictive value, negative predictive value and statistical agreement in specimen obtained by each methods, suction or no suction and pushing the stylet or injecting air.

Aim #2 : To compare the sensitivity, specificity, diagnostic accuracy, positive predictive value, negative predictive value and statistical agreement in specimen obtained by each methods, suction or no suction and pushing the stylet or injecting air.

On-site cytopathology interpretation during EUS-FNA has a significant clinical impact by increasing the diagnostic yield of the FNA. However, many hospitals do not have provision for on- site diagnosis of EUS FNA specimen. Therefore, optimal specimen obtainment and preparation of EUS FNA are is important to accurate diagnosis. EUS FNA is performed with or without self-retracting 10-mL syringe according to endosonographer's preference. In general, if there is too little cellularity on immediate cytologic evaluation, then use more suction, and, conversely, if the samples are too bloody, then use less suction. However, it is hard to determine the quality of specimen without on-site cytopathology interpretation. The actual way material is expressed from the needle may influence yield. Injecting air to express the material is problematic, because the material may spray out uncontrollably, risk of air artifact and specimen clotting may increase . A more controlled method is to pushing the stylet and slowly advance it to the needle tip. But use of the stylet during EUS-FNA is not only difficult and time consuming process but also increasing the risk of accidental needle stick injury. However, there is no firm evidence which one is the appropriate, suction or no suction and pushing the stylet or injecting air. If there was no significant difference between each methods, appropriate methods of obtaining and preparing aspirates undergoing EUS-FNA is no suction and injecting air and EUS FNA may perform more easier and shorter.

In this prospective randomized controlled trial, patients with peripancreatic mass for EUS-FNA will be included. One patient will be underwent at least four puncture during EUS-FNA. The sequence of EUS-FNA with first four methods will be selected by using a randomization scheme obtained from a sealed envelope : (1) negative pressure suction with 10 mL syringe and pushing the stylet; (2) negative pressure suction with 10 mL syringe and injecting air; (3) without negative pressure suction and pushing the stylet; (4) without negative pressure suction and injecting air. EUS FNA will be performed by two experienced endosonographers. All specimens will be read by same experienced pathologist blinded to methods.

The investigators will compare between the four groups in terms of adequacy, cellularity, bloodiness, contamination, air artifact. The sensitivity, specificity, diagnostic accuracy, positive predictive value, negative predictive value and statistical agreement for the different methods will be calculated.

Inclusion Criteria:

1. Age greater than 18 years
2. Presence of peri pancreatic mass, mediastinal or intra-abdominal lymphadenopathy confirmed by investigational modality - CT scan, magnetic resonance imaging, EUS.
3. Capable of providing informed consent

Exclusion Criteria:

1. Severe coagulopathy (INR > 1.5) or thrombocytopenia (platelet count < 50,000)
2. History of acute pancreatitis in the preceding 4 weeks
3. Pregnancy
4. Inability to provide informed consent