Supplement Abstracts Submitted for the 73rd Annual Scientific Meeting of the American College of Gastroenterology: STOMACH
Purpose: To assess interval size change and management recommendations of gastric GISTs from index to surveillance EUS
Methods: A retrospective chart review of all EUS for presumed gastric GISTs was performed at Duke University Medical Center. Patients with at least one surveillance EUS performed during January 1, 2000 to November 1, 2007 were included. The following EUS findings were recorded: length of follow up, interval size change, use/yield of FNA, and number of exams. For each tumor, the width (x) and length (y) were used to calculate a cross sectional area [πxy (mm2)].
Results: Forty-five patients underwent at least one surveillance EUS for a presumed gastric GIST. The average time interval between index EUS and first surveillance was 16.53 months (±12.97); range 1–59 months. Patients underwent an average of 2.44 (±0.69) procedures within the 7 year interval. FNA was performed in 31.1% patients with a diagnostic yield of 50%: GIST (57.1%), leiomyoma (28.6%) and foveolar hyperplasia (14.3%). Surgical resection was recommended in 4 patients: increase in size on surveillance imaging (2), bleeding (1), referring surgeon's opinion (1). One patient elected for surgical excision. To date, two of the 5 patients have undergone resection. The presence of a GIST was confirmed by FNA in only on 1 of the surgically resected tumors. The tumor size on the surgical pathology specimens demonstrated similar measurements in comparison to the prior EUS measurements. The range in cross sectional area of the presumed GIST on index EUS was 220 mm2–1105 mm2. Twenty-five GISTS (55.6%) increased in size [average increase in size 59.8 mm2 (30.9%)] after an average of 20.8 months between index and surveillance EUS. Stability in size was detected in 15.6% of patients and surveillance was discontinued.
Conclusion: In this small retrospective study, EUS was frequently utilized in the surveillance of presumed gastric GISTs. When performed, FNA had a poor diagnostic yield. Gastric GISTs are more likely to increase in size than remain stable through a surveillance interval. Continued surveillance of GISTs is recommended, however the optimal intervals remain unclear. Prospective studies are required and should be the focus of future investigation.