Study on the Diagnostic Yield and Correlated Factors of EUS-FNA in Pancreatic Masses

Aims: EUS-FNA has become an important tool in diagnosis of pancreatic masses. However, the diagnostic accuracy of the technology is influenced by many factors.

Aim: The aim of this study was to investigate the diagnostic yield and correlated factors of EUS-FNA in pancreatic masses.

Method: This study retrospectively collected and analyzed the clinical data of all patients with pancreatic lesions who underwent EUS-FNA from January 2008 to January 2010. The clinical data enrolled 11 factors including lesion size, lesion characteristics, lesion location, negative pressure suction, number of needle passes, real-time cytological diagnosis, etc.

Results: A total of 101 patients were enrolled and obtained sufficient specimens by EUS-FNA. There were 64 men and 37 women (mean age: 55.6yrs, range: 16–78yrs). The mean size of lesions was 3.53cm (range: 1–8cm). The positive rate of cell smears was 54.5% (55/101)and the positive rate of cell blocks was 34.0% (16/47). The overall diagnostic accuracy, sensitivity, specificity, positive predictive value and negative predictive value of EUS-FNA were as follows: 85.1%, 81.1%, 96.3%, 98.4%, 65.0%, respectively.

Univariable Logistic regression analysis indicated that lesion size, lesion characteristics, negative pressure suction, needle type, operating physician experience were correlated factors of EUS-FNA positive rate. Multivariate variable Logistic regression analysis showed that the lesion size and lesion characteristics were independent correlated factors of EUS-FNA positive rate (OR=2.012, P=0.000; OR=10.218, P=0.002). Lesion size and EUS-FNA positive rate were positively correlated. The positive rate of EUS-FNA in solid lesions was 10.2 times in cystic lesions. Multivariate variable Logistic regression analysis also showed lesion size was the only independent correlated factors of EUS-FNA diagnostic accuracy (OR=1.984, P=0.015), and they were positively correlated. The positive rate between Group a (EUS experience>15 years) and Group b (EUS experience ≤10 years) were significantly different (75.5% vs. 43.8%, P=0.001). Among the total 170 punctures of EUS-FNA, 131 punctures were performed with 5ml negative pressure, 39 with 10ml negative pressure. The results showed that the positive rate of punctures with 5ml negative pressure suction was significantly higher than those with 10ml (41.2% vs. 23.1%, P=0.039). The average number of EUS-FNA passes was not influenced by patients gender, patients age, lesion location, lesion size and needle type, while the average number of passes in cystic lesions significantly more than that in solid lesions (P=0.025).

Conclusions: EUS-FNA can effectively make a pathological diagnosis of pancreatic masses with high diagnostic accuracy and specificity. EUS-FNA positive rate and diagnostic accuracy were both positively correlated with pancreatic lesion size. EUS-FNA positive rate of solid pancreatic lesions was significantly higher than that of cystic lesions.

Keywords: Endoscopic ultrasonography, Fine needle aspiration, Pancreas, Diagnosis