Current issues of ACP Journal Club are published in Annals of Internal Medicine


Endoscopic re-treatment of bleeding peptic ulcers had fewer complications than surgery

ACP J Club. 1999 Nov-Dec;131:64. doi:10.7326/ACPJC-1999-131-3-064

Source Citation

Lau JY, Sung JJ, Lam Y-H, et al. Endoscopic retreatment compared with surgery in patients with recurrent bleeding after initial endoscopic control of bleeding ulcers. N Engl J Med. 1999 Mar 11;340:751-6. [PubMed ID: 99156341]



In patients with recurrent bleeding after initial endoscopic treatment of bleeding peptic ulcers, what is the efficacy and safety of endoscopic re-treatment compared with surgery?


Randomized, unblinded, controlled trial with follow-up at 6 weeks.


A specialized center in Hong Kong with expertise in endoscopic treatment of bleeding peptic ulcers.


94 patients who were admitted with recurrent bleeding after initial endoscopic therapy for bleeding peptic ulcer. Recurrent bleeding was defined as vomiting of fresh blood, development of hypotension and melena, or need for > 4 units of blood within 72 hours after endoscopic therapy. Patients with terminal cancer were excluded. 92 patients (mean age 65 y, 76% men) were included in the analysis.


48 patients had re-treatment with endoscopic injection of epinephrine and thermocoagulation, and 44 patients had surgery with the surgeon's choice of procedure. Patients in the endoscopy group and patients in the surgery group who had simple ulcer plication or excision alone received intravenous omeprazole, 40 mg every 12 hours, until they resumed an oral diet.

Main outcome measures

Mortality at 30 days, duration of hospitalization after treatment, intensive care unit (ICU) stay, need for blood transfusion, and treatment-related complications.

Main results

Analysis was by intention to treat. The endoscopic re-treatment and surgery groups did not differ for mortality at 30 days (10% vs 18%, P = 0.37), duration of hospitalization (median 10 vs 11 d, P = 0.59), need for or length of ICU stay (5 vs 10 patients, 59 d for both, P = 0.16), or units of blood transfused (median 8 vs 7 units, P = 0.27). Patients who had surgery were more likely to have complications (Table).


For patients with recurrent bleeding after initial endoscopy for bleeding peptic ulcer, endoscopic re-treatment did not differ from surgery for 30-day mortality, duration of hospitalization, intensive care unit stay, or units of blood transfused. Patients who received endoscopic treatment were less likely to have complications.

Source of funding: No external funding.

For correspondence: Dr. S.C. Chung, Department of Surgery, Prince of Wales Hospital, Sharin, Hong Kong, China. FAX 852-26350075.

Table. Endoscopic re-treatment vs surgery for recurrent bleeding after initial endoscopy for bleeding peptic ulcer*

Outcome by 6 wk Endoscopy Surgery RRR (95% CI) NNT (CI)
Complications 15% 36% 60% (15 to 82) 5 (3 to 25)

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.


Lau and colleagues report that endoscopic re-treatment was as effective as surgery in the management of recurrent ulcer bleeding after initial endoscopy. The groups studied were equal in the spectrum of bleeding severity observed at randomization. However, patients in whom endoscopic re-treatment ultimately failed were more likely to have had hypotension and ulcers > 2 cm at the time of randomization. These patients are most likely to have large-diameter arteries as the source of ulcer bleeding (1).

In vitro experiments done by Hepworth and colleagues (2) showed that coaptive thermocoagulation was ineffective when bleeding occurred from an artery > 2 mm in diameter. Rutgeerts and colleagues (3) showed, in vitro, that injection therapy was ineffective when bleeding occurred from arteries larger than 0.5 mm. In vivo, it is difficult to seal bleeding ulcer vessels > 1 mm in diameter with endoscopic coaptive thermocoagulation techniques. It is uncertain whether in vivo endoscopic injection therapy achieves a hemostatic effect with respect to vessel size similar to that observed in vitro.

Injection therapy is simple to perform, readily available, and popular. Although several studies have shown that endoscopic injection therapy alone can achieve effective initial hemostasis, diameters of bleeding vessels were not reported. Endoscopic coaptive thermocoagulation is more commonly delivered by bipolar probe than by heat probe. With correct technique, the 2 devices deliver equivalent hemostatic results. To achieve the results reported by Lau and colleagues, endoscopists must avoid the temptation to rely on injection therapy alone as a retreatment strategy.

James S. Barthel, MD
H. Lee Moffitt Cancer CenterTampa, Florida, USA


1. Swain CP, Storey DW, Bown SG, et al. Nature of the bleeding vessel in recurrently bleeding gastric ulcers. Gastroenterology. 1986;90:596-608.

2. Hepworth CC, Kadirkamanathan SS, Gong F, Swain CP. A randomised controlled comparison of injection, thermal, and mechanical endoscopic methods of haemostasis on mesenteric vessels. Gut. 1998;42:462-9.

3. Rutgeerts P, Geboes K, Vantrappen G. Experimental studies of injection therapy for severe nonvariceal bleeding in dogs. Gastroenterology. 1989;97:610-21.