Provocative Mesenteric Angiography for Obscure Gastrointestinal Hemorrhage: An Update on Outcomes, Safety, and Predictors of Success.

PURPOSE: To confirm the safety and modest effectiveness of provocative mesenteric angiography (PMA) in patients with occult gastrointestinal hemorrhage and identify predictors of success.

MATERIALS AND

METHODS

Retrospective review was performed on 119 provocative mesenteric angiograms on 109 patients (males, 73; mean age, 67.6 years; range, 7-94 years) over a 22-year period. PMA was defined as transcatheter administration of tissue plasminogen activator (tPA) into mesenteric arteries. Clinical, radiologic, and laboratory notes were reviewed. Adverse event rates within 30 days of the procedure were assessed.

RESULTS

Of 119 provocative mesenteric angiograms, active extravasation was induced in 33 (27.7%) of procedures. On univariable analysis, older age, admission for acute gastrointestinal bleeding (GIB), hematochezia, prior positive radiologic study for GIB, transfusion of 2 or more units of blood, estimated nitroglycerin concentration ratio, estimated tPA concentration ratio, and increasing vessel selectivity for tPA administration were significant predictors of positive PMA (P < .05 for each). However, on multivariable analysis, only hematochezia (P = .036) and prior positive radiologic study (P = .016) were significant predictors of positive PMA. No patients with melena and negative radiologic studies had a positive PMA. No major bleeding adverse events occurred.

CONCLUSIONS

PMA yielded a modest 27.3% positivity rate without clinically significant hemorrhagic adverse events. Prior positive radiologic studies for GIB and character of stool (melena vs hematochezia) significantly predicted provocation of bleeding. PMA is not recommended for patients with melena and negative radiologic imaging studies.