Performing endoscopy during the first 6 hours after a patient with gastrointestinal bleeding is handed off to a specialist does not lower the 30-day mortality rate compared to procedures done 6 to 24 hours after consultation, according to a new study of 516 volunteers reported in Thursday's New England Journal of Medicine. All of the patients were predicted to be at high risk for death or further bleeding. The 30-day all-cause death rates were 8.9% with urgent endoscopy vs 6.6% when the endoscopy was done later, but within 24 hours of consultation (P=0.34). When the research team, led by Dr. James Lau of the Chinese University of Hong Kong in Shatin, looked at the rate of further bleeding, the rates were also not significantly different: 10.9% with urgent endoscopy compared with 7.8% when the procedure was done later. Patients scoped within the first 6 hours were more likely to have endoscopic hemostatic treatment, with rates of 60.1% in the urgent group vs 48.8% when endoscopy was done during the 6- to 24-hour timeframe. Just under 13% of patients who presented with upper GI bleeding fell into the high-risk category, defined as a Glasgow-Blatchford score of 12 or higher. Most had a peptic ulcer. Patients who clearly required urgent intervention due to hypotensive shock or did not stabilize after initial hemodynamic resuscitation were excluded. Among the patients who died during that 30-day window, 48% in the urgent-endoscopy group and 47% in the early-endoscopy group had advanced cancer. Previously, observational data looking at the timing issue has produced "inconsistent, very-low-quality evidence," so the new results are welcome, said Dr. Loren Laine of the Yale School of Medicine in a journal editorial. In practical terms, it took about 8 hours between the time a patient came through the door and there was a gastroenterology consultation with randomization, Dr. Laine noted, thus the time windows were actually larger, with 55% of patients getting scoped more than 24 hours after presentation. Mean time from consultation to endoscopy was 2.5 hours in the urgent group and 16.8 hours in the control group. Current guidelines recommend endoscopy within 24 hours of presentation for upper GI bleeding. In-hospital mortality with upper GI bleeding is typically 10%, according to data from the United Kingdom. The Lau team found no significant difference between the "urgent" and "early" groups in the rate of further bleeding within 30 days, the odds of having surgery, the likelihood of being admitted to intensive care, the duration of hospitalization, the chance of receiving a transfusion, or the number of units of packed red cells received by transfusion. They noted that doctors treating patients in the urgent-endoscopy group saw more ulcers that were actively bleeding during the procedure. "The longer period until endoscopy and longer duration of acid suppression reduced the number of ulcers with active bleeding and major stigmata of bleeding," the team said. "This observation corroborates the findings from an earlier randomized, controlled trial that evaluated the use of a high-dose proton-pump inhibitor before endoscopy. Acid suppression before endoscopy can reduce the need for endoscopic treatment." At this point, "the available evidence suggests that most patients hospitalized with upper gastrointestinal bleeding need not be rushed to immediate endoscopy. Rather, resuscitation and treatment for coexisting active medical conditions should be initiated as appropriate and endoscopy then performed within 24 hours after presentation," said Dr. Laine. "Although the current trial suggests that periods somewhat longer than 24 hours may be acceptable in high-risk patients, I favor endoscopy within 24 hours for patients hospitalized with upper gastrointestinal bleeding, to avoid potentially prolonging hospitalization unnecessarily," Dr. Laine said. —Reuters Staff Source
