Endosonography-guided fine needle aspiration biopsy of solid pancreatic masses - accuracy and impact on the treatment of 358 patients

Abstract

The aim of our study was to evaluate the accuracy, safety and benefit of EUS-FNA in diagnosing and treating patients with unclear solid pancreatic mass. Diagnostic accuracy of EUS-FNA was evaluated through the comparison of EUS-FNA cytology with either histology of rescted specimen or clinical and radiological follow-up. The benefit was defined as impact of EUS-FNA cytology on subsequent treatment. In a cohort of 358 patients with a solid pancreatic mass, the sensitivity and specificity of EUS-FNA in diagnosing pancreatic cancer was 93% (95% Cl: 89.2-95.6) and 90.1% (95% Cl: 80.1-95.6), respectively. Complications ocured in 2.2% of patients. False-negative cytology was less frequent in EUS-FNA with rapid on-site cytopathology evaluation (ROSE) compared with EUS-FNA without ROSE (3.9% vs 14.3%; p = 0.006). The positive and negative impact of EUS-FNA on subsequent treatment was observed in 63.9% and 0.5% of patients, respectively. In 3.4% of patients with pancreatic cancer, EUS-FNA diagnosed either subtle malignant ascites or liver metastases (mean ± SD: 8,9 ± 3,5 mm), that were not detected on previous contrast-enhanced CT scans. We have confirmed that EUS-FNA is accuarate and safe method in diagnosing pancreas carcinoma and has positive impact on subsequent treatment in 2/3 of patients with unclear solid pancreatic mass. ROSE significantly increases the accuracy of the method.