Urgent endoscopy not linked to improved outcomes in acute upper GI bleeding

Endoscopy performed ≤6 hours after gastroenterological consultation is not associated with a lower 30-day mortality rate than that performed 6–24 hours after consultation among patients in stable condition who were hospitalized with acute upper gastrointestinal (GI) bleeding and who were at high risk of further bleeding or death, a randomized trial by the Chinese University of Hong Kong (CUHK) has shown.

The 30-day mortality rate did not differ significantly between patients who underwent endoscopy at ≤6 hours (urgent-endoscopy group) and 6–24 hours (early-endoscopy group) (8.9 percent vs 6.6 percent; hazard ratio, 1.35; 95 percent confidence interval [Cl], 0.72 to 2.54; p=0.34). [N Engl J Med 2020;382:1299-1308]

The rate of further bleeding (defined as persistent or recurrent bleeding) within 30 days was also similar between the urgent-endoscopy and early-endoscopy groups (10.9 percent vs 7.8 percent; relative risk [RR], 1.46; 95 percent CI, 0.83 to 2.58)

Endoscopic haemostatic therapy was administered during initial endoscopy in 60.1 percent and 48.4 percent of patients in the urgent-endoscopy group and early-endoscopy group, respectively (RR, 1.24; 95 percent CI, 1.06 to 1.46).

The median duration of hospitalization was 5 days for both urgent-endoscopy and early-endoscopy groups, with a comparable number of patients admitted to intensive care unit in both groups (4 and 3, respectively).

The standard of endoscopy procedures performed after hours (6:00 pm to 5:59 am) was noninferior to those performed during office hours (6:00 am to 5:59 pm), as shown by results of a post-hoc analysis (further bleeding, 10.7 percent for endoscopy performed during office hours vs 10.4 percent for endoscopy performed after hours) (death, 7.5 percent vs 10.4 percent).

In this randomized trial, 516 high-risk patients with overt signs of acute upper GI bleeding (haematemesis, melena, or both) and a Glasgow-Blatchford score of ≥12 were randomized (1:1) to undergo urgent endoscopy or early endoscopy after gastroenterological consultation. Inpatients who developed GI bleeding during hospitalization for other medical illnesses were also included in the study.

An intravenous (IV) high-dose infusion of a proton-pump inhibitor (PPI) (80 mg bolus followed by 8 mg per hour) was administered to patients in both groups on admission and at the first sign of bleeding during hospitalization. Patients with suspected variceal bleeding were also prescribed vasoactive drugs and IV antibiotics.

At baseline, peptic ulcer was the source of bleeding in >61 percent of patients in the urgent-endoscopy and early-endoscopy groups, while oesophageal or gastric varices accounted for 9.7 percent and 7.4 percent of bleeding, respectively.

“Urgent endoscopy with haemostatic treatment has been recommended for patients with acute upper GI bleeding. Nevertheless, results of our study showed that urgent endoscopy did not significantly reduce mortality rate, the incidence of further bleeding, and the duration of hospitalization. These findings are in contrast with our hypothesis that urgent endoscopy would be associated with improved treatment outcomes,” said first author Professor James Lau, Chairmen of the Department of Surgery, CUHK.

“Those who underwent endoscopy at 6–24 hours after gastroenterological consultation received a longer duration of PPI, which reduced the number of ulcers with active bleeding. Thus, acid suppression before endoscopy can reduce the need for endoscopic treatment,” he suggested.