Endoscopic ultrasound guided fine needle aspiration of solid pancreatic lesions: Performance and outcomes

BACKGROUND AND AIM: We report our single-centre experience with endoscopic ultrasound guided fine needle aspiration (EUS-FNA) of solid pancreatic lesions with regard to clinical utility, diagnostic accuracy and safety. METHODS: We prospectively reviewed data on 100 consecutive EUS-FNA procedures performed in 93 patients (54 men, mean age 60.6 +/- 12.9 years) for evaluation of solid pancreatic lesions. Final diagnosis was based on a composite standard: histologic evidence at surgery, or non-equivocal malignant cytology on FNA and follow-up. The operating characteristics of EUS-FNA were determined. RESULTS: The location of the lesions was pancreatic head in 73% of cases, the body in 20% and the tail in 7%. Mean lesion size was 35.1 +/- 12.9 mm. The final diagnosis revealed malignancy in 87 cases, including adenocarcinomas (80.5%), neuroendocrine tumours (11.5%), lymphomas (3.4%) and other types (4.6%). The FNA findings were: 82% interpreted as malignant cytology, 1% as suspicious for neoplasia, 1% as atypical, 7% as benign process and 9% as non-diagnostic. No false-positive results were observed. There was a false-negative rate of 5%. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 94.3%, 100%, 100%, 72.2% and 95%, respectively. In 23 (88.5%) of 26 aspirated lymph nodes malignancy was found. Minor complications occurred in two patients. CONCLUSIONS: Our experience confirms that EUS-FNA in patients with suspected solid pancreatic lesions is safe and has a high diagnostic accuracy. This technique should be considered the preferred test when a cytological diagnosis of a pancreatic mass lesion is required.