Upper GI bleed: peptic ulcer:previous peptic ulcers and NSAID use increased the risk.

Clinical bottom line (level 2b)

  1. patients with the following prognostic factors were at increased risk of bleeding peptic ulcers
    • previous peptic ulcers
    • NSAID use
    • older than 75
    • alcohol-related disorders
    • liver cirrhosis, oesophageal varices or portal vein thrombosis
Hallas et al: ScandinavianJournal of Gastroenterology 1995; 30: 438-444
Expires October 2002

The study

Retrospective cohort study with objective outcomes, adjusted for confounding factors, not validated in an independent set of patients.

Setting: county,Denmark

138700 patients living in hospital catchment area

Factors studied:
  • age, sex, previous peptic ulcer disease, alcohol-related diagnoses,liver cirrhosis, oesophageal varices, malignancy, smoking, use of anti-ulcer medication, steroids, potassium supplements, oral anticoagulants,NSAIDs
  • history of a peptic ulcer
  • use ofNSAID
  • aged >75
  • alcohol-related diagnosis
  • liver cirrhosis, oesophageal varices or portal vein thrombosis
  • age 55 to74
  • male sex
  • use of aspirin




    • Multivariate Poisson regression analysis performed on prognostic factors.

    ?100% followed for 19 months
    Outcomes studied:
  • upper GI bleed defined as melena or anaemia or blood transfusion required AND peptic ulcers or erosions on endoscopy thought to be the source of bleeding

    • Patients taking NSAIDswere identified using a prescription database. For each NSAID user, 4non-users matched for age and sex were randomly extracted.

    The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)    		 NNF
    (95% CI)
    upper GI bleed 19 months 183/138700 0.13%
    (0.11% to 0.15%)    		 758
    (662 to 886)

    prognostic factor for
    upper GI bleed     		time to outcome control rate (%) adjusted OR
    (95% CI)    		 NNF+
    (95% CI)
    history of a peptic ulcer 19 months 183/138700
    (0.13%)     		6.5
    (4.4 to 9.7)     		140
    (88 to 220)
    use ofNSAID 19 months 183/138700
    (0.13%)     		5.8
    (4.0 to 8.5)     		160
    (100 to 250)
    aged >75 19 months 183/138700
    (0.13%)     		5.1
    (3.3 to 7.8)     		190
    (110 to 330)
    alcohol-related diagnosis 19 months 183/138700
    (0.13%)     		4.1
    (2.3 to 7.5)     		250
    (120 to 590)
    liver cirrhosis, oesophageal varices or portal vein thrombosis 19 months 183/138700
    (0.13%)     		4.1
    (2.2 to 7.6)     		250
    (120 to 630)
    age 55 to74 19 months 183/138700
    (0.13%)     		2.4
    (1.6 to 3.6)     		540
    (290 to 1300)
    male sex 19 months 183/138700
    (0.13%)     		2.1
    (1.5 to 2.8)     		690
    (420 to 1500)
    use of aspirin 19 months 183/138700
    (0.13%)     		1.9
    (1.2 to 2.9)     		840
    (400 to 3800)

    Comments

    1. Broad definition of UGIbleeding used my explain differences between this study and McMahon 1997

    Citation

    1. Hallas J, Lauritsen J, DalsgardVilladsen H, et al: Non-steroidal anti-inflammatory drugs and upper gastrointestinal bleeding, identifying high-risk groups by excess risk estimates. ScandinavianJournal of Gastroenterology 1995; 30: 438-444
    Contributor: Chris Ball and Musab Hayatli, October 1999
    Reviewer: Daniel Sontheimer

    Clinical Question.
    Patient
    Intervention or Exposure NSAID use, previous peptic ulcer
    Outcome upper GIbleed