Upper GI bleed: bleeding peptic ulcer: elderly patients on omeprazole were less likely to continue bleeding or require surgery.

Clinical bottom line (level 1b)

  1. Patients aged < 60 with a bleeding peptic ulcer who took omeprazole compared with placebo were less likely to continue bleeding (NNT = 7 at 3 days) .
  2. Patients on omeprazole were less likely to require endoscopy, surgery or die (NNT = 9 at 3 days) .
Hasselgren et al: Scandinavian Journal of Gastroenterology 1997; 32: 328-333
Expires October 2002

The study

Double-blinded ?concealed randomised trial without intention-to-treat
Setting: 29 hospitals, Sweden and Norway

333 patients (aged mean 74, 58% male) with melaena or haematemesis within the last 48 hours confirmed on endoscopy due to a bleeding peptic

Excluded if
  • aged < 60
  • taken anticoagulation therapy within last 5 days
  • clinically significant laboratory test abnormalities
  • upper GI malignancy
  • deficient haemostasis (prothrombin <40% or platelet count < 100 x 10 9 /l
  • severe hepatic, renal or cardiac failure


Note:
  • Patients only had endoscopic therapy or surgery if they had received > 4units of blood over 24 hours or > 8 units since admission.

Control Group: (n = ?, 163 analysed): placebo
Experimental Group: (n = ?, 159 analysed): omeprazole 80mg iv bolus over 30 minutes, followed by 8mg/hour for 72 hours
Anti ulcer medication, tranexamic acid and NSAIDs were not permitted.
97% followed for ? weeks
Outcome notes:

  • treatment failure : need for further endoscopic treatment, surgery or death

The evidence

Outcome Time to outcome CEREERRRR
(95% CI)		ARR
(95% CI)		NNT
(95% CI)
any bleeding 3 days 89
(54.6%)		 64
(40.3%)		 26%
(7% to 42%)		 14.4%
(3.55% to 25.1%)		 7
(4 to 28)
treatment failure 3 days 27
(16.6%)		 9
(5.66%)		 66%
(30% to 83%)		 10.9%
(4.16% to 17.7%)		 9
(6 to 24)
surgery 3 days 16
(9.82%)		 4
(2.52%)		 74%
(25% to 91%)		 7.30%
(2.12% to 12.5%)		 14
(8 to 47)
death 3 days 1
(0.61%)		 1
(0.63%)		 -3%
(-1500% to 94%)		 -0.02%
(-1.73% to 1.70%)		 -6500
(NNT = 59 to infinity;
NNH = 58 to infinity)

Comments

  1. Patients bled for a shorter time on omeprazole.
  2. Patients were randomised in blocks of two - this may have jeopardised allocation concealment.

    Citation

    1. Hasselgren G, Lind T, Lundell L, et al: Continuous intravenous infusion of omeprazole in elderly patients with peptic ulcer bleeding: results of a placebo-controlled multicenter study. Scandinavian Journal of Gastroenterology 1997; 32: 328-333
    Contributor: Chris Ball and Musab Hayatli, October 1999
    Reviewer:

    Clinical Question.
    Patient elderly with a bleeding peptic ulcer
    Intervention or Exposure omeprazole
    Comparison placebo
    Outcome continued bleeding, death, surgery, endoscopy