BACKGROUND
Endoscopic ultrasound (EUS)-guided fine-needle aspiration biopsy (EUS-FNA) is increasingly used for the diagnosis of malignant and benign disease in the region of the upper GI tract. We prospectively investigated the clinical accuracy and safety of this method in unselected patients under routine conditions.
PATIENTS AND METHODS
101 consecutive patients (median 61.5 years; 56 female) were enrolled in the study, in whom a total of 106 tissue biopsies were obtained by using EUS-FNA. Major indications for EUS-FNA were suspicious lesions located in the mediastinum, esophagus, stomach, pancreas, liver, biliary system, adrenals or retroperitoneum. A longitudinal echoendoscope (HITACHI FG-34UX) equipped with a standard 22G -aspiration needle was used. The aspirated specimens were analyzed further by using standard cytology and/or histology. Lymph-node biopsies were additionally subjected to flow-cytometry (FACS-light-chain restriction). Surgery was used for reference (where available). In the remaining cases the final diagnosis obtained by the clinical course and all available imaging and histologic informations (ultrasound, CT, MRT) was used for reference.
RESULTS
EUS-FNA caused no serious complications. In 6/106 specimen (5.6 %) no sufficient cell material could be aspirated. In the remaining 100 specimens EUS-FNA reached an overall sensitivity of 78 % and a specificity of 100 %, while the accuracy was 89 % and the positive and negative predictive values were 100 % and 81 %, respectively. The greatest diagnostic accuracy was achieved in mediastinal and retroperitoneal lesions, while the accuracy of EUS-FNA in pancreatic lesions and perigastric lymph nodes was distinctly smaller (<80 %). Addition of FACS studies in patients with suspected malignant lymphoma increased the diagnostic accuracy in the small number of patients included in the study.
CONCLUSION
EUS-FNA improves the tissue-based diagnosis of suspicious lesions in locations that are difficult to access (e. g., posterior mediastinum). EUS-FNA is safe, while its diagnostic accuracy is relatively high. Our preliminary data suggest that flow-cytometry may improve the fine-needle based diagnosis of non-Hodgkin s lymphoma, which should be further investigated.