We evaluated the performance, clinical role, and diagnostic accuracy of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) in gastrointestinal intramural lesions.

Procedural and pathologic data were reviewed from consecutive patients undergoing EUS-FNA for intramural lesions. Final diagnoses were determined by surgical histopathologic conformation and the diagnosis of malignancy, including clinical follow-up with repeat imaging.

Forty-six patients (mean age, 47 years; 24 males) underwent EUS-FNA. Lesions were located in the stomach (n=31), esophagus (n=5), and duodenum (n=10). The median lesion size was 2 cm (range, 1 to 20.6). Final diagnoses were obtained in 22 patients (48%). EUS-FNA was diagnostic in 40 patients (87%). The diagnostic accuracy of cytology for differentiating between benign and malignant lesions was 82%; diagnostic error occurred in three patients (6%). The cytologic results influenced clinical judgment in 78% cases. The primary reasons for negative or no clinical impact were false-negative results, misdirected patient management, and inconclusive cytology.

EUS-FNA exhibited an 87% diagnostic yield for gastrointestinal intramural lesions; the accuracy of cytology for differentiating malignancy was 82%. The limitations of EUS-FNA were primarily because of nondiagnostic sampling (9%) and probable diagnostic error (6%); these factors may influence the clinical role of EUS-FNA.