Nonvariceal upper gastrointestinal bleeding.

Nonvariceal upper gastrointestinal bleeding (NVUGIB) remains an important cause of morbidity and mortality. The etiological role and the relative risk of nonsteroidal anti-inflammatory drugs (NSAIDs) in association with other clinical factors are the focus of several papers published in the last year. Data from studies that evaluate the interaction of NSAIDs with Helicobacter pylori suggest that there is a clear role for eradication in aspirin-related bleeding. Although the benefit of H. pylori eradication in nonaspirin NSAID-related bleeding is being debated, they should both be eliminated given their relative contribution to the causation of peptic ulcer hemorrhage. The search for reliable and accurate tools to predict outcomes after NVUGIB continues. Recent important contributions in this area of research are the use of a clinical prediction guide to identify patients who are likely to require endoscopic intervention, and the use of Doppler ultrasound examination of the ulcer base to predict rebleeding. The role of antisecretory therapy in the setting of acute NVUGIB has been revisited in two meta-analyses and one randomized, placebo-controlled trial of omeprazole as an adjunct to endoscopic intervention. Until more definitive data are available, it appears that the effect of acid-decreasing medications is more important when endoscopic therapy is not provided than when effective endoscopic intervention is instituted. With regard to therapeutic endoscopy for upper gastrointestinal bleeding, the few randomized trials that have been published in full form in the last year focus on the use of mechanical methods of hemostasis for peptic ulcer bleeding and other etiologies such as Mallory-Weiss tears and Dieulafoy lesions.

Current status of endoscopic therapy for ulcer bleeding.

This article provides an overview of the therapeutic endoscopic modalities available for the treatment of peptic ulcer bleeding. The benefits of endoscopic haemostasis have been fully demonstrated by three meta-analyses, which included most of the controlled trials published until 1992. In this review, an emphasis is placed on randomized, prospective comparative trials published during the past 20 years. Using an evidence-based medicine approach, the results of meta-analyses are translated into efficacy measures known as relative and absolute risk reductions, and number needed to treat. Single-modality treatments with injection agents such as epinephrine, sclerosants and thrombogenic substances, or with thermal therapies, are efficacious and comparable. Combination therapy involving injection and thermal techniques may offer an advantage over single-method therapy. The differences in the results between clinical trials and routine clinical practice, and among the various randomized studies, are probably related to operators' experience and variations in technique rather than to inconsistency of endoscopic haemostasis.

Ulcers and nonvariceal bleeding.

This review of nonvariceal upper gastrointestinal bleeding concentrates mainly on peptic ulcer bleeding, and is based on papers published during the last two years. While the focus is on outcome prediction models and therapeutic interventions, brief reference is also made to certain epidemiological aspects, pathological and animal studies, and risk factors for bleeding peptic ulcer.