Severe upper gastrointestinal tumor bleeding: endoscopic findings, treatment, and outcome.

BACKGROUND AND STUDY AIMS

The aim of the present study was to review endoscopic findings, treatment, and clinical outcomes in patients with severe upper gastrointestinal bleeding due to tumors.

PATIENTS AND METHODS

A retrospective analysis was made of prospectively gathered data on all patients with severe upper gastrointestinal bleeding who were admitted to two large referral centers during a 45-month period.

RESULTS

Nine hundred thirty-five patients had severe upper gastrointestinal bleeding, of whom 42 (5%) were found to have tumors. Histologically, nearly all of the tumors were of a malignant type. Fifty-two percent of the patients had acute severe upper gastrointestinal bleeding as the initial presentation of their tumor. The most common tumor was gastric adenocarcinoma, and all of these cases were at advanced stages. Endoscopic hemostasis with thermal probes or epinephrine injection, or both, was carried out in seven patients (17%), with successful hemostasis in all of the tumors. Regardless of the treatment given, patients with upper gastrointestinal tumor bleeding, had a 30-day surgery rate of 43%, a 30-day rebleed rate of 33%, a 30-day mortality rate of 10%, and a 1-year mortality rate of 89%.

CONCLUSIONS

Most tumors that cause severe upper gastrointestinal bleeding are of a malignant histologic type and are already at an advanced stage. Endoscopic hemostasis of bleeding upper gastrointestinal tumors is safe and initially effective, and may provide time for elective surgical palliation. Regardless of therapy, upper gastrointestinal tumors with severe bleeding have a poor one-year survival.