Objective To investigate the diagnostic value of endoscopic ultrasound-guided fine needle aspiration ( EUS-FNA) on malignant lesions in gastrointestinal adjacent tissue, and further to analyze the risk factors influencing positive rate of EUS-FNA. Methods The clinical data of 171 patients undergoing EUS-FNA from January 2009 to May 2016 were collected. The lesion location, size and characteristics, the number of needle passes, puncture suction negative pressure, size of puncture needle, and years of operator experience in EUS were retrospectively analyzed. Results The overall sensitivity, specificity, and accuracy of EUS-FNA in the diagnosis of malignant lesions were 78. 3％ ( 83/106) , 100. 0％ ( 65/65) , and 86. 5％( 148/171) , respectively. The univariable logistic regression analysis demonstrated that the risk factors of EUS-FNA were lesion location, lesion characteristics, and lesion size. In multivariate analysis, larger lesion size ( OR=1. 029, 95％CI: 1. 011-1. 047, P=0. 001) and lesion characteristics of solid ( OR=5. 098, 95％CI:1. 324-19. 633, P=0. 018) were independent factors affecting the positive rate of EUS-FNA. Among 171 cases performed by EUS-FNA, the incidence of postoperative complications was 1. 75％ ( 3/171 ) included 2 cases of fever and 1 case of acute pancreatitis, which were improved after conservative treatment. Conclusion EUS-FNA is a safe and effective method of cytological and histological diagnosis with high accuracy and sensitivity, importantly in distinguish malignancy from benign lesion in gastrointestinal adjacenttissue. Positive rate of diagnosis on malignant lesions by EUS-FNA is positively correlated with lesion size, and EUS-FNA positive rate of solid malignant lesions is significantly higher than that of cystic lesions.

OBJECTIVETo identify the predictive factors for differentiating pancreatic ductal adenocarcinoma (PDAC) from other neoplastic solid pancreatic lesions and assess the accuracy of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) for diagnosis of PDAC.

METHODSWe retrospectively analyzed the clinical data of patients referred for EUS-FNA evaluation of pancreatic lesions in the Digestive Endoscopic Center of Nanfang Hospital between January, 2009 and May, 2016. The cases with unknown diagnosis, missing data, repeated punctures, cystic lesions and benign lesions were excluded from the analysis. The positivity rates of EUS-FNA were compared between patients with PDAC and those with non-PDAC lesions, and the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of EUS-FNA were assessed in the diagnosis of PDAC. Univariate and multivariate logistic regression analyses were used to identify the factors for differentiating PDAC from non-PDAC lesions based on the demographic characteristics, clinical presentations, laboratory data, and endoscopic ultrasonography imaging features of the patients.

RESULTSAmong the 75 patients with solid neoplastic pancreatic lesions, 54 (72.0%) were found to have PDAC and 21 (28.0%) had non-PDAC lesions. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of EUS-FNA for the diagnosis of PDAC were 77.8%, 100.0%, 100.0%, 63.6% and 84.0%, respectively. No significant difference was found in the positivity rate of EUS-FNA between patients with PDAC and those with non-PDAC lesions (77.8% 76.2%, > 0.05). Multivariate regression analysis identified abdominal pain (=5.163, 95%: 1.093-24.389, =0.038), lesion size (=0.926, 95%: 0.877-0.978, =0.006), characteristics of the solid lesions (=7.105, 95%: 1.440-35.043, =0.016), and evidence of metastases (=6.165, 95%: 1.332-28.533, =0.020) as the independent factors for predicting PDAC.

CONCLUSIONSThe pretest characteristics including abdominal pain, evidence of metastases, and lesion size and lesion characteristics defined by endoscopic ultrasonography findings can reliably predict a diagnosis of PDAC. EUS-FNA has a high sensitivity and a high specificity for the diagnosis of PDAC.