Upper GI bleed: Endoscopic haemostasis of bleeding peptic ulcers reduces bleeds, emergency surgery and death.

Clinical bottom line (level 1a)

  1. Patients with bleeding peptic ulcers who receive endoscopic haemostasis compared with anti ulcer therapy are less likely to continue or restart bleeding (NNT = 4 at unknown) , require emergency surgery (NNT = 6 at unknown) or die in hospital (NNT = 30 at unknown)
Sacks et al: Journal of the American Medical Association 1990; 264 (4): 494-499
Expires October 2002

The study

Systematic review of all randomised controlled trials of
  • Patients: bleeding peptic ulcers
  • Intervention: endoscopic haemostasis compared with common anti ulcer treatment and emergency surgery
  • Outcome: continued or recurrent bleeding, emergency surgery, in-hospital mortality


    • Articles found in English using MEDline, ? (search terms: ? ) and weekly searches of Current Contents ofClinical Medicine, selecting review articles, and checking bibliographic references in the articles.

    Selection criteria: detailed in text
    Appraisal criteria: blinded review by 2 researchers using a points system based on study design, blinding and statistical analysis.
    Articles excluded if:
    • none of the three endpoints were reported
    • control group received any treatment other than common anti ulcer therapy and emergency surgery


    25 randomised controlled trials found: 6 used mono polar electrocoagulation, 4 bipolar electrocoagulation, 8 used laser, 6 injections
    • Results were reported as pooled rate differences (equivalent to actual risk reduction)
    Study heterogeneity not reported

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    continued or recurrent bleeding unknown 680
    (38.9%)
    (%)    		 69%
    (% to %)    		 27%
    (12% to 43%)    		 4
    (2 to 8)
    emergency surgery unknown 461
    (25.7%)
    (%)    		 62%
    (% to %)    		 16%
    (11% to 21%)    		 6
    (5 to 9)
    in-hospital mortality unknown 185
    (9.9%)
    (%)    		 30%
    (% to %)    		 3%
    (1% to 5%)    		 30
    (20 to 100)

    Comments

    1. There were too few studies to indicate which treatment method was most efficacious, though none has a very clear advantage.
    2. By limiting the search to only English language papers, other important studies may have been missed.

    Citation

    1. Sacks HS, Chalmers TC, Blum AL, et al: Endoscopic haemostasis: an effective therapy for bleeding peptic ulcers. Journal of the American Medical Association 1990; 264 (4): 494-499
    Contributor: Chris Ball and Musab Hayatli, October 1999
    Reviewer: Kenneth Locke

    Clinical Question.
    Patient bleeding peptic ulcers
    Intervention or Exposure haemostasis by endoscopy
    Outcome death, recurrent bleeding, emergency surgery