Home Feedback Contents

Stress ulcer evidence

Up

Asian Intensive Care: coming of age
International intensive care conference, Hong Kong, December 14th-15th
Register now!
Click here for details

 

Gavin Joynt

Definitions

Overt (macroscopically visible) bleeding

  • Haematemesis, “coffee ground” aspirate

  • Melena, haematocheza

Clinically important bleeding

  • Haemodynamic instability

  • Orthostatic haemodynamic response

  • Decrease in Hb of > 2g/dL

  • Transfusion requirement

Can preventive medical therapies decrease morbidity and mortality from stress ulcers?

Type of medical therapy – compared with Placebo or Control

Relative risk of o

 Relative risk of important bleeding Relative risk of death
Antacids 0.66 (0.37-1.17) 0.35(0.09-1.41) 1.42(0.82-2.47)
Sucrulfate 0.58(0.34-0.99) 1.26(0.12-12.87) 1.06(0.67-1.67)
H2 antagonists 0.58(0.42-0.79)  0.44(0.22-0.88) 1.15(0.86-1.53)

Cook DJ et al. JAMA 1996; 275: 308 [Meta-Analysis]

  • Summary – Evidence suggests that bleeding, but not mortality can be reduced.

What are the best medical preventive therapies?

Type of medical therapy – compared with antacid

Relative risk of o

 Relative risk of important bleeding Relative risk of death
Sucrulfate vs antacid 0.97(0.62-1.51) 1.49(0.42-5.27) 0.73(0.54-0.97)
H2 antagonists vs antacid 0.56(0.37-0.84) 0.86(0.46-1.59) 0.89(0.66-1.21)

Cook DJ et al. JAMA 1996; 275: 308 [Meta-Analysis]

 

Relative risk of o

 Relative risk of important bleeding Relative risk of death
H2 antagonist vs sucralfate 0.44(0.21-0.92) 1.18(0.92-1.51) 1.03(0.84-1.26)

Rate of clinically important bleeding in mechanically ventilated patients receiving ranitidine was 10/596 (1.7%) vs 23/604 (3.8%) in the sucralfate group.

Cook DJ et al. New Engl J Med 1998; 338: 791 [Prospective randomized controlled trial]

  • Summary – evidence suggests that H2 receptor antagonists are most efficient in reducing overt and clinically important bleeding in ICU patients, and do NOT increase the risk of nosocomial pneumonia when compared with sucrulfate. Mortality is not changed by any agent.

Does every ICU patient require preventive therapy?

  • Incidence of stress ulceration is low and decreasing with time and routinely using prophylaxis in all patients is probably not cost effective.

    • Overt bleeding rate

      • 25% in 1980s

      • 5% in 1990s

    • Clinically important bleeding

      • Current rate probably 0.6-3%

  • Direct mortality from stress ulcer bleeding is probably very low (< 0.1%)

Which patients need stress ulcer prophylaxis?

Attempt to identify patients at high and low risk and not to treat those at very low risk.

  • 2252 patients (674 received prophylaxis)

  • Overt bleeding 4.4%(3.6-5.6%)

  • Important bleeding 1.5%(1.0-2.1%)

  • 87% of those with overt bleeding were receiving prophylaxis, they had a high mortality, suggesting that doctors could identify those at high risk and that they were sicker.

  • Independent risk factors

    • Mechanical ventilation >48h (OR 15.6)

    • Coagulopathy (OR 4.2)

    • Hypotension (OR 3.7, P=0.08 – trend only)

    • Rate of important bleeding was very low (0.1%) if you had no factors Vs 3.7% if you had one or more factors 

Note: although a large study, not a comprehensive case mix - did not include significant numbers burn injury or neurosurgical patients.

Cook DJ et al. New Engl J Med 1994; 330:377

  • Mechanically ventilated patients (from 1998 study)

    • Renal failure, enteral nutrition and ranitidine independently associated with bleeding.

    • Enteral nutrition and ranitidine protect against bleeding, and an additive effect is seen.

Cook DJ et al. Crit Care Med 1999; 27:2812

  • Beware of patients with well-known high risk conditions; 

    • Burns “Curling’s ulcer” bleeding rate is about 11%

    • Neurosurgical patients may have a bleeding rate of 30% or more without prophylaxis.

Butterfield WC. Surg Annu 1975; 7:261

Tryba M et al. Drugs 1997; 54:581

  • Summary - Patients who are not mechanically ventilated, are not coagulopathic, are enterally fed and have no major organ failure probably do not require prophylaxis

©Gavin Joynt September 2003

©Charles Gomersall, October, 2009 unless otherwise stated. The author, editor and The Chinese University of Hong Kong take no responsibility for any adverse event resulting from the use of this webpage.
Copyright policy    Contributors