Upper GI bleed: acute hepatic failure and chronic renal failure increased the risk of bleeding in intensive care patients

Clinical bottom line (level 2b)

  1. Around one tenth of elderly patients in intensive care had an upper gastrointestinal bleeding during their stay.
  2. Patients in intensive care who had the following factors were at increased risk of having an upper GI bleed
    • acute hepatic failure
    • chronic renal failure
    • nasogastric tube in situ for 6 days or more
    • history of alcohol abuse
    • H. pylori infection
Ellison et al: Critical Care Medicine 1996; 24: 1974-1981
Expires October 2002

The study

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

Setting: 6 Veterans Affairs hospitals, USA

874 patients (aged mean 63, 98% male) admitted to intensive care units

Excluded if
  • aged < 18
  • duodenal or gastric ulcer disease
  • pre-existing GI bleeding
  • gastric cancer
  • esophageal varices
  • expected survival or ICU stay < 2 days
  • ICU stay for an uncomplicated cardiac or peripheral vascular surgical procedure
  • history of serum sickness or reaction to gamma globulin
  • use of plasmaphoresis therapy
  • hypervitaminosis A
  • pregnancy



    • Factors studied:
  • platelet count, prothrombin time, NSAID use, heparin, warfarin, corticosteroids, prophylactic therapy, respiratory failure, septic shock
  • acute hepatic failure
  • chronic renal failure
  • nasogastric tube in-situ for 6 days
  • history of alcohol abuse
  • Helicobacter pylori IgA level > 1
  • history of malignancy


    • 74% of patients received prophylactic treatment for upper GI bleeding (antacids, histamine antagonists or sucralfate).

    Multivariate regression analysis performed on risk factors.

    ?100% followed for stay in ICU
    Outcomes studied:
  • acute upper GI bleeding hematemesis, grossly bloody nasogastric tube drainage or upper GI bleed requiring transfusion

    • Study retrospective analysis of immunoglobulin RCT.

    The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)    		 NNF
    (95% CI)
    acute upper GI bleeding stay in ICU 76/874 8.7%
    (6.8% to 11%)    		 12
    (9 to 15)

    prognostic factor for
    acute upper GI bleeding     		time to outcome control rate (%) adjusted OR
    (95% CI)    		 NNF+
    (95% CI)
    acute hepatic failure ? 76/874
    (8.7%)     		6.67
    ( to )     		3
    ( to )
    chronic renal failure ? 76/874
    (8.7%)     		3.03
    ( to )     		7
    ( to )
    nasogastric tube in-situ for 6 days ? 76/874
    (8.7%)     		2.59
    ( to )     		9
    ( to )
    history of alcohol abuse ? 76/874
    (8.7%)     		2.23
    ( to )     		11
    ( to )
    Helicobacter pylori IgA level > 1 ? 76/874
    (8.7%)     		1.92
    ( to )     		15
    ( to )
    history of malignancy ? 76/874
    (8.7%)     		0.30
    ( to )     		-17
    ( to )

    • No confidence intervals provided with odds ratios.

    Comments

    1. The results of this study are limited due to the selected patient population. The patients represent a small percent of all ICU patients since only those who consented to the intervention trial were included.
    2. There are better studies for general ITU patients (e.g. Cook et al 1994)

    Citation

    1. Ellison RT, Perez-Perez G, Welsh CH, et al: Risk factors for upper gastrointestinal bleeding in intensive care patients: role of Helicobacter pylori. Critical Care Medicine 1996; 24: 1974-1981
    Contributor: Chris Ball and Musab Hayatli, October 1999
    Reviewer: Janice L Zimmerman

    Clinical Question.
    Patient
    Intervention or Exposure
    Outcome