Background: The aims of this study were to determine the rate of false-positive diagnosis with EUS-guided fine-needle aspiration and to identify factors contributing to this type of error.
Methods: The records of 577 patients undergoing EUS-guided fine-needle aspiration were reviewed and a subset of 188 patients with malignant cytology who underwent surgery was identified. Operative histopathology was compared with EUS-guided fine-needle aspiration cytopathology and false-positive cases were identified. An experienced cytopathologist, who was not involved with the original interpretation of the specimens, reviewed these cases to identify any factor(s) contributing to the errors.
Results: Three cases of false-positive diagnosis were identified (1.6%; 95% CI [0.3%, 4.6%]). By aspiration site, the false-positive rates were as follows: pancreas 2/39 (5.1%), 95% CI [0.6%, 17.3%]; lymph nodes 1/136 (0.7%), 95% CI [0.02%, 4.0%]; and other sites 0/13, 95% CI [0.0%, 24.7%]. In both instances of a false-positive diagnosis for a pancreatic aspiration cytologic specimen, interpretative errors were identified. The false-positive interpretation of cytologic material aspirated from a lymph node occurred in a patient without any evidence for malignancy at surgery. In 111 patients with confirmed esophageal, gastric, or rectal malignancy undergoing EUS-guided fine-needle aspiration of nonperitumoral lymph nodes, there was no false-positive diagnosis, suggesting that specimen contamination by luminal tumor is rare.
Conclusion: The overall rate of false-positive diagnosis for EUS-guided fine-needle aspiration is similar to that reported for other modalities. Most false-positive diagnoses are caused by interpretation errors.