3894 Prospective study of the ability of endoscopic ultrasound to localize duodenal gastrinomas

• Published in: Gastrointestinal endoscopy 2000-04, Vol.51 (4), p.AB139-AB139
• Main Authors: Tio, Thian L., Benjamin, Stanley B., Gibril, Fathia, Abou-Saif, Alaa, Roy, Praveen K., Shojamanesh, Homayoun, Alexander, Richard A., Jensen, Robert T.
• Format: Article
• Language: English
• ISSN: 0016-5107; 1097-6779
• DOI: 10.1016/S0016-5107(00)14304-4

The role of EUS in localizing duodenal gastrinomas remains controversial because of lack of prospective assessments. To address this question we prospectively studied 40 consecutive patients with Zollinger-Ellison syndrome (ZES) admitted to the NIH prior to surgery. All patients had conventional imaging studies (CT, MRI, ultrasound, angiography), somatostatin receptor scintigraphy (SRS), EUS (Olympus UM 20), catheter US [UM3R] with endoscopy [side-viewing endoscope-(TJF)] (CUS.Endo), followed by a standardized laparotomy with a Kocher maneuver, intra-operative ultrasound and duodenotomy. Seven patients had MEN-1 and ZES and 33 had sporadic ZES. Prior to surgery, at least one nonEUS imaging study localized a possible duodenal gastrinoma in 19 patients (48%); conventional imaging only in 28%; SRS in 42%; EUS alone, 35%; CUS.Endo, 45%; and EUS or CUS.Endo, 52%. At exploration duodenal gastrinomas were found in 18 patients (45%). EUS, CUS.Endo, both (EUS and CUS.Endo), and any other imaging (other imaging) had sensitivities of 47%, 64%, 71% and 61% and specificities of 65%, 58%, 52% and 64%, respectively. In the 8 patients with all other imaging negative with a duodenal gastrinoma, EUS was positive in 50% (4/8), CUS.Endo in 38%, or either in 62%. In the 22 patients without a duodenal gastrinoma a false positive localization occurred with EUS in 5/22 (23%); CUS.Endo in 32%; both, 36%; and with other imaging in 36%. The sensitivity of SRS alone for localizing a duodenal tumor was 56%; SRS + EUS, 83%; SRS + CUS.Endo, 83%; and SRS + both, 88%. These results demonstrate that EUS alone will localize 47% of duodenal gastrinomas, which is slightly less than the 61% seen with a combination of other imaging studies. However, if combined with careful endoscopy using a side-viewing endoscope and catheter ultrasound, it will localize 71%. Both procedures are complementary to SRS, correctly localizing a duodenal gastrinoma in 88% of patients; however, the combination of both endoscopic procedures also increases the false positive rate and decreases the specificity.