VARICEAL BLEEDING

Hayes, PC., et al (1990) “Meta-analysis of value of propanolol in prevention of variceal haemorrhage” Lancet 336:153-156

• meta-analysis of controlled trials
• looking at prevention of primary and secondary variceal bleeding
• propanolol works for primary prevention
• endoscopic band ligation for acute bleeding

ULCER PREVENTION AND TREATMENT

Lay, JY., et al (2000) “Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers” N Eng J Med 343:310-316

• double blind RCT
• n = 240
• actively bleeding ulcers or non-bleeding visible vessels after treatment with adrenaline + thermocoagulation
• omeprazole 80mg bolus -> 8mg/hr for 72 hours VS placebo
• all patients received 20mg PO omeprazole for 8/52
• significant reduction in rate of bleeding within 30 days (most of which happened in first 3 days) in treatment group
• no difference in mortality
• no difference in number of patient requiring surgery

Messori, A., et al (2000) “Bleeding and pneumonia in intensive care patients given ranitidine and sucralfate for prevention of stress ulcer: meta-analysis of randomised controlled trials” BMJ 321:1103-1106

• incidence of pneumonia and bleeding -> thought to be potential adverse effect of gastric acid suppression c/o bacterial overgrowth
• small numbers in studies
• no benefit from ranitidine vs placebo
• no benefit from sucralfate vs placebo
• increase risk of nosocomial pneumonia in ranitidine vs sucralfate
  -> muscosal ischaemia more important than acidity in stress ulcer formation
  -> despite this evidence use of PPI is widespread in critical care

Parsonnet, J. et al (2005) “Clinician-discoverers – Marshal, Warren and H. pylori” N Engl J Med, 353:2421-2423

• acknowledgement of these guys discovery
• H.pylori causes: chronic superficial gastritis, chronic active gastritis, peptic ulcer disease and gastric adenocarcinoma

Cook, D et al (1998) “A comparison of sucralfate and ranitidine for the prevention of upper gastrointestinal bleeding in patients requiring mechanical ventilation” Canadian Critical Care Trials Group, NEJM, 338:791-7

• MRCT
• placebo vs ranitidine vs sucralfate
  -> GIH: placebo ( ), ranitidine (1.7%), sucralfate (3.8%) – P < 0.05
  -> pneumonia incidence: ranitidine (19.2%), sucralfate (16.2%) – P > 0.05
  -> no change in LOS or mortality

Mariki, P. E. et al (2010) “Stress ulcer prophylaxis in the new millennium: A systematic review and meta-analysis” Critical Care Medicine: Volume 38 – Issue 11 – pp 2222-2228

Background

• – stress ulceration uncommon (1%)
  – prophylaxis may be unwarranted if feeding can be established early
  – prophylaxis may increase risk of hospital-acquired pneumonia and Clostrodium difficile infection
• – meta-analysis of RCT’s
  – histamine-2 receptor blockers vs placebo
  – 17 studies (1836 patients)
• – primary end point: clinically significant GIH
  – secondary end points: incidence of HAP and hospital mortality.
  – sub group analysis performed by grouping studies by enteral nutrition or no enteral nutrition
• -> significant decrease in risk of gastrointestinal bleeding (OR 0.47, p < 0.02) -> BUT only noted in a subgroup of patients who did not receive enteral nutrition (OR 1.26, CI 0.43-3.7)
  -> if patients fed, prophylaxis did not alter the risk of GI bleeding
  -> no increase in risk of HAP overall
  -> BUT, those who were fed had an increased risk of HAP (p 0.02, or 2.81)
  -> stress ulcer prophylaxis did not change mortality
  -> the subgroup who received stress ulcer prophylaxis AND enteral feeding had a elevated HAP rate (? both increasing gastric pH and thus allowing gastric multiplication of bacteria and subsequent aspiration)

Internal Validity

• clinically significant GIH was defined by each individual study
• if there wasn’t a definition of clinically significant bleeding then endoscopic bleeding was used.
• used H2 antagonists: cimetidine, ranitidine, famotidine
• varying doses
• some studies ran infusions
• only 3 studies looked at enteral nutrition

External Applicability

• we use omeprazole
• those who were fed may have been sicker than those that weren’t which may explain increase in HAP and death (confounding)

AN APPROACH

• feed early unless contraindicated
• don’t use prophylaxis if feeding established
• if unable to establish enteral feed, risk assess for GIH: if high risk -> use prophylaxis
• if develops stress ulcers treat
• vigilance for the development of hospital acquired pneumonia
• may need to take ulcer prophylaxis of FASTHUG sticker!