Ultrasound-guided fine-needle aspiration represents the gold-standard for the pathological diagnosis of solid pancreatic lesions. New needles design allowed to obtain samples suitable for histological evaluation (endoscopic ultrasound-guided fine needle biopsy). the aim was to compare two different techniques during ultrasound-guided fine needle biopsy, for diagnosis of suspect pancreatic solid lesions.

Materials and methods This is a prospective, randomized, multicenter study comparing the slow-pull technique with the standard suction technique during Ultrasound-guided fine-needle biopsy in patients with pancreatic solid lesions. The study involved three referral centers for pancreatico-biliary diseases: the Gastroenterology and Endoscopy Unit. ARNAS Civico-Di Cristina-Benfratelli Hospital, Palermo, Italy; the Gastroenterology and Endoscopy Unit. Marche Polytechnic University, Ospedale A. Murri, Fermo, Italy; the Gastroenterology Unit. AUSL Bologna, Bellaria-Maggiore, Bologna, Italy. The protocol was approved by the institutional medical ethical committee at each participating institution and written informed consent was obtained from all patients for participation in the study. All consecutive patients with a newly diagnosed pancreatic solid lesions suitable for endoscopic ultrasound-guided tissue acquisition were considered for enrollment. Written informed consent was obtained from all patients before the procedure. All the endoscopic ultrasound procedures were carried out by a skilled endosonographer with the patients under conscious sedation or deep sedation according to the anesthesiological guidelines approved in each participating center. All Ultrasound-guided fine-needle biopsy were done using a 20 Gauge needle (EchoTip ProCore 20G with ReCoil Stylet™, Cook Medical, Bloomington, IN, USA). The ReCoil Stylet´s has an automatic recoiling capability designed to help users to manage more easily the stylet, thus minimising the risk of contamination. Before puncturing, color doppler examination was performed to exclude the presence of interposing vessels. The needle tip was sharpened by withdrawing the stylet of approximately 2 mm, and then it was advanced into the lesion under real-time endoscopic ultrasound guidance. In patients randomized to the stylet slow-pull techniques, 15 to-and-fro movements within the lesion were performed while gently pulling the needle stylet slowly and continuously upwards. In patients randomized to the standard suction technique, 15 to-and-fro movements within the lesion were performed using a 10-mL suction syringe. In both sampling procedures the fanning technique was used to maximized sample acquisition. Tissue samples were expelled onto slides by reinserting the stylet and the visible core was physically lifted off the slides placing it into a formalin vial for histological evaluation. For the macroscopic adequacy, the endosonographer evaluated for the presence of a visible core defined as a worm-like material whitish/yellowish or red, not including fluid-like specimens in the formalin vial, which was judged to be adequate to perform a pathological analysis. The pathologists were "blinded" on which group the vials received were belonging to.

Randomization was performed using random sequences generated by a computer and then closed in consecutive numbered envelopes.

The degree of blood contamination of the specimen was based on the pathologist evaluation of the formalin fixed and paraffin embedded tissue samples and defined as significant when a large amount of blood cells made pathological diagnosis difficult to be made. On the other hand, blood contamination of the specimen was defined as not significant when no or only few blood cells were present without any influence on the pathological diagnosis.

Sensitivity was defined as the true positive rate whereby the test was the final cytologic diagnosis, Specificity as the true negative rate (proportion of actual negatives that are correctly identified as such), positive predictive value was defined as the number of true positive divided the number of positive calls (true positive and false positive), negative predictive value was defined as the number of true negative divided the number of negative calls (true negative and false negative). Diagnostic accuracy was defined as the ratio between the sum of true positive and true negative values divided by the total number of masses.

Technical success was defined as the capability of sampling the target lesion associated with the presence of a visible core, according to endosonographer's judgment, potentially useful for the final pathological analysis.

Adverse events were defined based on the criteria expressed by Cotton et al. [Cotton PB. Gastrointest Endosc 2010;71:446-54] Samples positive for malignancy were considered diagnostic for malignancy, while in patients with negative Ultrasound-guided fine-needle biopsy, surgical specimen evaluation, results of other diagnostic investigations and/or a clinical follow-up of at least 6 months were used to establish the definitive diagnosis.

Statistical Analysis The sample size was calculated based on the results of a previous study [Nakai Y H. Dig Dis Sci. 2014;59:1578-85] that showed that blood contamination of the specimen was lower with the stylet slow-pull technique as compared with the standard suction technique (25% vs 70%), with a consequent increase in the diagnostic accuracy. Based on these data we calculated for a difference of 25%, at a power of 80% and an alpha of 0.05 (two-sided test), a final sample size of 110 patients (55 patients in each group). Continuous variables will be reported as mean +/- standard deviation (SD) or as median with interquartile range. Categorical variables will be presented as numbers and percentage and will be compared either using the chi-square test (with Yates' correction when appropriate) or Fisher exact test.

Statistical was conducted using the Fisher's exact test for categorical variables and the Mann-Whitney U-test for continuous variables. Sensitivity, specificity, positive predictive value, negative predictive value, and overall diagnostic accuracy for each arm were investigated by comparing the results with the definitive diagnosis. Statistical tests were considered significant at a corresponding p value of <0.05. Data handling and analyses were done with SPSS 14.

Inclusion Criteria:

• Age >18years old
• Solid pancreatic lesion on imaging MRI and/or CT-scan referred for tissue acquisition
• Lesion can be visualized with EUS and needle puncturing can be technically feasible
• Able to sign informed consent

Exclusion Criteria:

• < 18 years-old
• Cystic pancreatic lesions
• Extra-pancreatic lesions or inaccessible/non-visualized lesions
• Previous gastrectomy
• International normalized ratio > 1.5
• Impossibility to suspend anticoagulant therapy
• Platelet count < 50.000 cells/cubic millimeter
• Severe or unstable clinical conditions
• Pregnancy
• Inability to give informed consent
• Refusal to participate to the study

Randomization was performed using random sequences generated by a computer and then closed in consecutive numbered envelopes.