Over-the-scope clips more effective than standard treatment for nonvariceal upper GI bleeding

A randomized controlled trial of 190 patients from university teaching hospitals in Hong Kong, mainland China, and Australia finds that initial treatment of nonvariceal upper gastrointestinal (GI) bleeding with over-the-scope clips (OTSCs) reduces the risk of further bleeding vs standard endoscopic haemostatic treatment.

Standard treatment of nonvariceal upper GI bleeding involves haemoclips or contact thermal devices, with or without prior injection of diluted epinephrine. “The failure rate with either method is close to 10 percent, and further bleeding after endoscopic haemostasis is associated with increased mortality,” wrote the researchers. [Ann Intern Med 2019;171:805-822; Gut 2011;60:1327-1335]

“OTSCs are memory-shaped nitinol clips that are currently mostly used in severe or refractory bleeding from nonvariceal causes. In the present study, we hypothesized that OTSCs used as initial treatment could be superior to standard treatment at preventing further bleeding from nonvariceal causes,” they continued. [Ann Intern Med 2023;doi:10.7326/M22-1783]

Adult patients with active bleeding or a nonbleeding visible vessel from a nonvariceal cause on upper gastrointestinal endoscopy were randomized to receive OTSC (n=93) or standard haemostatic treatment (n=97). The primary outcome was 30-day probability of further bleeds. Other outcomes included failure to control bleeding after assigned endoscopic treatment, recurrent bleeding after initial haemostasis, further intervention, blood transfusion, and hospitalization.

The 30-day probability of further bleeding was 3.2 percent in the OTSC group vs 14.6 percent in the standard treatment group (risk difference, 11.4 percentage points; 95 percent confidence interval [CI], 3.3–20.0; risk ratio [RR], 0.22; p=0.006). Failure to control bleeding after the assigned endoscopic treatment at index endoscopy occurred in 1.1 percent vs 6.2 percent of patients in the OTSC vs standard treatment group (risk difference, 5.1 percentage points; 95 percent CI, 0.7–11.8; RR, 0.18). The respective rates of recurrent bleeding after endoscopic haemostasis were 2.2 percent vs 8.8 percent (risk difference, 6.6 percentage points; 95 percent CI, -0.3–14.4; RR, 0.25).

“We did not find significant differences in other secondary outcomes, including red blood cell transfusion and hospitalization,” reported the researchers. “It can be argued that the outcomes after standard treatment and rescue use of OTSCs would be similar to those after primary use of OTSCs, yet the lack of difference in other outcomes may be explained by the insufficient trial size to detect small differences, and although our trial protocol did not allow crossover treatment, rescue use of OTSCs was common.”

There were two deaths within 30 days in the OTSC group and four deaths in the standard treatment group. There were two device-related serious adverse events in the OTSC group. A patient developed pyloric obstruction after OTSC treatment of a small antral ulcer. The captured mucosa formed a pseudopolyp obstructing the pyloric channel. The OTSC had to be cut using a bipolar cutting device. Another patient died of an ulcer perforation.

“We should be cautious in recommending use of OTSCs as initial treatment,” said corresponding author of the study, Professor James Lau of the Department of Surgery, Chinese University of Hong Kong. “OTSCs are costly, and the procedure takes a long time as it involves scope withdrawal, clip mounting and scope reinsertion. Endoscopists require training before using OTSCs. Therefore, the primary use of OTSCs may lie in the treatment of high-risk ulcers predicted to fail standard endoscopic treatment.”