OBJECTIVE: A standard diagnostic evaluation including upper and/or lower en
doscopy, tagged red blood cell scintigraphy, and visceral angiography ident
ifies the source of GI bleeding in the majority of patients who present wit
h acute GI hemorrhage. However, in a small group of patients the source of
bleeding remains obscure; this form of GI hemorrhage is uncommon but repres
ents a considerable diagnostic challenge. Some investigators have advocated
provocation of bleeding with vasodilators, anticoagulants, and/or thrombol
ytics in association with tagged red blood cell scans or angiography. Unfor
tunately, the available literature on this topic is limited. Therefore, the
purpose of this study is to report our experience with provocative GI blee
ding studies.
METHODS: The radiology databases at Duke University Medical Center and the
Durham Veterans Administration Medical Center were reviewed from 1994 to 19
99. Any patient who received a vasodilator, anticoagulant, or thrombolytic
to induce bleeding during a tagged red blood cell scan or visceral angiogra
m was included.
RESULTS: Seven provocative bleeding studies were performed on seven patient
s. All patients underwent a visceral angiogram with intra-arterial administ
ration of tolazoline (a vasodilator), heparin (an anticoagulant), and/or ur
okinase (a thrombolytic). Of the seven provocative studies, only two induce
d angiographically identifiable bleeding. Both of these patients underwent
surgical therapy. There were no complications attributed to the provocative
bleeding studies.
CONCLUSIONS: These results suggest that provocative GI bleeding studies can
be performed safely. However, because an active bleeding source was identi
fied in only a small proportion of patients, we believe that further study
is required to optimize patient selection and to clarify the cost-effective
ness of this approach in patients with GI hemorrhage of obscure origin. (Am
J Gastroenterol 2000,95: 2807-2812. (C) 2000 by Am. Coll. of Gastroenterol
ogy).