Stigmata of recent haemorrhage in diagnosis and prognosis of upper gastrointestinal bleeding

Publisher Website
Google Scholar
Abstract
In 277 consecutive episodes of suspected upper gastrointestinal bleeding, lesions bearing stigmata of recent haemorrhage (stigmata) were found by endoscopy in 110 (47%) out of 233 patients who were judged to have bled; 78 (33%) had lesions without stigmata, and in 45 (19%) no lesion was seen. Results in 176 entirely unselected admissions for upper gastrointestinal bleeding were similar. Forty-eight chronic duodenal and 41 chronic gastric ulcers were identified by endoscopy. Stigmata were found in 27 (56%) and 33 (80%) of these cases respectively. Sixteen patients had multiple lesions, and in 12 (75%) the presence of stigmata permitted diagnosis of the source of the haemorrhage. Stigmata were more likely to be seen in cases of duodenal ulcer, Mallory-Weiss lesions, and oesophageal varices when endoscopy was performed within 12 hours of bleeding, but were as common in cases of gastric ulcer after longer intervals. In the absence of stigmata one out of 21 patients with duodenal ulcer had further haemorrhage and one other needed emergency surgery; no patient with gastric ulcer had further haemorrhage or needed emergency surgery. In contrast, when stigmata were present 15 of the 27 patients with duodenal ulcer (56%) had further haemorrhage and 17 (63%) needed emergency surgery; of the 33 patients with gastric ulcer, 10 (30%) had further haemorrhage and 15 (45%) required emergency surgery. Superficial mucosal lesions may have been the source of haemorrhage when an ulcer unmarked by stigmata was seen at endoscopy. Stigmata were superior to any other single factor or combination of factors in predicting rebleeding and the need for emergency surgery.