Changing epidemiology of gastrointestinal angiodysplasia with increasing recognition of clinically milder cases: angiodysplasia tend to produce mild chronic gastrointestinal bleeding in a study of 47 consecutive patients admitted from 1980-1989.

It is important to recognize patients with gastrointestinal bleeding who are at high risk of having angiodysplasia, because these patients should be evaluated by endoscopy rather than barium studies. Sixty-two clinical and epidemiologic parameters were compared between 47 consecutive patients bleeding from angiodysplasia and 47 consecutive controls bleeding from other lesions admitted to two university teaching hospitals from 1980 through 1989. This study demonstrated statistically significant differences between these two groups. The patients with angiodysplasia generally presented with symptoms and clinical findings compatible with hemodynamically well-compensated, chronic bleeding: they were more likely than other gastrointestinal bleeders to experience weakness or fatigue, less likely to experience dizziness or syncope, and less likely to be orthostatic or hypotensive. They had more prior admissions for gastrointestinal bleeding, particularly for gastrointestinal bleeding of undetermined etiology. They were more likely than other gastrointestinal bleeders to be smokers. Patients with angiodysplasia had a milder hospital course: they had fewer transfusions of packed erythrocytes, shorter hospitalizations, and a lower mortality. The in-hospital mortality of patients bleeding from angiodysplasia was 2.1%. Despite the futility of diagnosing angiodysplasia by barium studies, patients ultimately diagnosed as having angiodysplasia were more often initially evaluated by barium studies than the other gastrointestinal bleeders. The currently identified risk factors for bleeding from angiodysplasia should help to select which gastrointestinal bleeders should be evaluated initially by endoscopy.