The diagnostic yield of superior mesenteric angiography: correlation with the pattern of gastrointestinal bleeding

Am J Gastroenterol. 1989 Aug;84(8):878-81.

Abstract

In an attempt to evaluate the usefulness of mesenteric angiography in relation to the nature and rate of gastrointestinal bleeding, we reviewed the records of all patients hospitalized at Montefiore Medical Center between 1983 and 1986 who had mesenteric angiography as part of their diagnostic evaluation for gastrointestinal bleeding. Fifty-eight patients were classified according to the pattern of blood loss: 14 patients with chronic occult, 28 patients with recurrent acute, and 16 patients with acute bleeding. The sensitivity and specificity of mesenteric angiography for each group was: chronic occult, 40% sensitivity and 100% specificity; recurrent acute, 30% sensitivity and 100% specificity; acute, 47% sensitivity and 100% specificity. Vascular lesions accounted for most of the diagnosed abnormalities, including four of five lesions in patients with chronic blood loss and four of six lesions in each of the groups with recurrent acute and acute bleeding. A positive angiogram was correlated with a high likelihood of surgery in patients with recurrent acute as well as acute hemorrhage. Five of six in the former group and six of seven in the latter group required an operative procedure. Mesenteric angiography is an integral part of the diagnostic evaluation of patients with gastrointestinal bleeding, although the frequency of positivity varies with the pattern of bleeding.

MeSH terms

  • Acute Disease
  • Adult
  • Aged
  • Aged, 80 and over
  • Blood Transfusion
  • Chronic Disease
  • Diagnostic Errors
  • Gastrointestinal Hemorrhage / diagnostic imaging*
  • Gastrointestinal Hemorrhage / etiology
  • Gastrointestinal Hemorrhage / therapy
  • Humans
  • Mesenteric Arteries / diagnostic imaging*
  • Middle Aged
  • Occult Blood
  • Predictive Value of Tests
  • Radiography
  • Recurrence
  • Retrospective Studies
  • Sensitivity and Specificity