Diagnosis of gastrointestinal tract lesions by endoscopic ultrasound-guided fine-needle aspiration biopsy

BACKGROUND: Endoscopic ultrasound-guided fine-needle aspiration biopsy (EUS-FNA) allows detailed imaging of both intramural and extramural structures of the gastrointestinal (GI) tract and also allows tissue samples to be obtained from masses and lesions in the GI tract. The objective of the current study was to determine the diagnostic utility of EUS-FNA in evaluating intramural and extramural GI tract lesions. METHODS: The authors evaluated all EUS-FNA specimens of GI tract lesions obtained over a 30-month period (from August 2000 to February 2003). Samples of pancreatic and intrabdominal/mediastinal lymph nodes were excluded from the study. A single endosonographer performed all procedures. An attending cytopathologist also was present on site to assess specimen adequacy. Cytologic diagnoses were analyzed for correlations with final diagnoses, which were based on histologic examination of biopsied/resected pathology materials and/or clinical follow-up findings. RESULTS: Sixty-two EUS-FNA specimens of intramural and extramural GI tract lesions were obtained from a total of 60 patients. The mean patient age was 58.8 years (standard deviation, 15.3 years). Thirty-six patients (60%) were male, and 24 (40%) were female. Twenty-eight patients had surgical pathologic evaluation of the corresponding lesions. The remaining 32 patients were followed clinically for a mean duration of 9.5 months (standard deviation, 7.7 months). The anatomic sites of the lesions were as follows: esophagus in 23 patients (37%), stomach in 13 patients (21%), duodenum in 15 patients (24%), and rectum/sigmoid in 11 patients (18%). It is noteworthy that 29 patients (43%) previously had experienced unsuccessful attempts at tissue diagnosis by endoscopic forceps biopsy. Of the 62 EUS-FNA specimens, 43, 4, and 15 were reported as being positive for a neoplasm, suspicious, and benign, respectively. Neoplastic lesions included carcinoma (n = 24), gastrointestinal stromal tumor (GIST; n = 18), neuroendocrine neoplasm (n = 2), and lymphoma (n = 1). There were two cases of endometriosis, three foregut duplication cysts, and one case of diverticulosis. There were two lesions that yielded false-negative findings (one gastric lymphoma and one GIST) secondary to sampling or interpretive error. There also were three cases that yielded false-positive findings (one case of endometriosis, one case of duodenal diverticula with smooth muscle hyperplasia, and one case of normal pancreas, which presented as a periduodenal mass). The sensitivity, specificity, and diagnostic accuracy of EUS-FNA in diagnosing GI tract neoplastic lesions were 89%, 88%, and 89%, respectively. CONCLUSIONS: EUS-FNA provides accurate tissue diagnosis in a wide variety of extraintestinal mass lesions and intramural GI tumors, particularly in patients for whom previous endoscopic forceps biopsy was unsuccessful in establishing a diagnosis.