Intended for healthcare professionals

Rapid response to:


Systematic review and meta-analysis of proton pump inhibitor therapy in peptic ulcer bleeding

BMJ 2005; 330 doi: (Published 10 March 2005) Cite this as: BMJ 2005;330:568

Rapid Response:

Evidence that omperazole causes organ dysfunctions?

This is a important study because it illustrates the pitfalls in failing to include an objective metabolic measure of shock, the most relevant variable, in the analysis of the data.

The authors concluded that, "Proton pump inhibitor treatment had no significant effect on mortality (odds ratio 1.11, 95% confidence interval 0.79 to 1.57; number needed to treat (NNT) incalculable) but reduced rebleeding (0.46, 0.33 to 0.64; NNT 12) and surgery (0.59, 0.46 to 0.76; NNT 20)" (1). They added that, "Results were similar when the meta- analysis was restricted to the 10 trials with the highest methodological quality:". Yet as James Penston oberved in his rapid rsponse, "there was a non-significant increase in mortality in patients receiving proton pump inhibitors compared with controls (5.2% v 4.6%)". Might we be looking at two opposing effects, a beneficial effect on bleeding and need for surgery to stop it and an adverse effect upon the ability to withstand the bleeding and/or surgery?

A case has been made for proton pump inhibitors having a mitochondrial effect in all cells, one that impairs their cytpoprotctive response to acute reductive stress such as that likely to be imposed by haemorrhage (2). The fear is that in so doing it may increase the risk of developing organ dysfunctions and death from them. Even in the absence of acute reductive stress proton pump inhibitors have caused organ injury, microscopic colitis and interstitial nephritis. If proton pump inhibitors can cause organ dysfunctions in patients exposed to acute reductive stress the anaesthetic practice of administering proton pump inhibitors before elective surgery to reduce the risk of aspiration pneumonia (3) could be having a similar effect.

The problem is that organ dysfunctions and deaths can occur so long after the intitiating event (4) that they are often considered unrelated events and can be omitted from analyses of outcome even after surgery. A myocardial infarct if often interpeted as an unavoidable act of God. Deaths occurring mor than 30 days later are also often considered unrelated events. They may not be. Consider fractured hips.

There is a significant difference in mortality between patients having surgery within 24 hours of admission (20%) and those having surgery beyond 24 hours of admission (50%). Even when only the healthy subgroup of ASA I and II patients were considered, the relative risk of death was 4.5 times greater if surgery occurred after 24 hours from admission. More importantly the attrition continues threafter, the overall mortality at 1 year being 14%, at 2 years 26%, and at 3 years 33% (5). The same trend almost certainly occurs in the case of bleeding ulcers. A much longer view is required in measuring outcomes from therapeutic interventions for acute events.

Proton pump inhibitors are most likely to cause organ dysfunctions in those who have developed severe acute reductive stress. The very likely inclusion of a majority of patients who did not, in the studies included in this meta-analysis, might have concealed a real effect. The increase in mortality from 4.6% to 5.2% might be real and better defined by stratifying patients by the severity of their shock and/or number of units of blood they received. In the absence of measurements of gastric intramucosal pH shock is unlikely to be accurately defined especially in its degree and duration (6,7).

The real concern might be for the long term administration of proton pump inhibitors. Might it increase the risk of organ dyfunctions and death in all acute illnesses, including all accidents, acute cardiovascular events, and systemic infections? Might the hypothetical risk of proton pump inhibitors even be cummulative? What of the potential for the adverse effect being additive with those of other drugs and also causing chronic organ dyfunctions (8)?

These pressing questions are never going to be answered by The Adverse Event Reporting System (AERS) designed to support the FDA's post- marketing safety surveillance program for all approved drug and therapeutic biologic products (9).

1. Grigoris I Leontiadis, Virender K Sharma, and Colin W Howden Systematic review and meta-analysis of proton pump inhibitor therapy in peptic ulcer bleeding BMJ 2005; 330: 568

2. Microscopic colitis: the tip of an omeprazole iceberg? Richard G Fiddian-Green eCMAJ, 25 Nov 2004 eLetter re: Sandra Dial Clostridium difficile colitis: A marker for ischemic colitis? CMAJ 2004; 171: 1326-1327

3. Memis D, Turan A, Karamanlioglu B, Saral P, Ture M, Pamukcu Z. The effect of intravenous pantoprazole and ranitidine for improving preoperative gastric fluid properties in adults undergoing elective surgery. Anesth Analg. 2003 Nov;97(5):1360-3.

4. Associations between intramucosal acidosis in the gut and organ failure. Crit Care Med. 1993 Feb;21(2 Suppl):S103-7.

5. Hamlet WP, Lieberman JR, Freedman EL, Dorey FJ, Fletcher A, Johnson EE. Influence of health status and the timing of surgery on mortality in hip fracture patients. Am J Orthop. 1997 Sep;26(9):621-7.

6. Hamilton-Davies C, Mythen MG, Salmon JB, Jacobson D, Shukla A, Webb AR. Comparison of commonly used clinical indicators of hypovolaemia with gastrointestinal tonometry. Intensive Care Med. 1997 Mar;23(3):276-81.

7. Fiddian-Green RG, Haglund U, Gutierrez G, Shoemaker WC. Goals for the resuscitation of shock. Crit Care Med. 1993 Feb;21(2 Suppl):S25-31.

8. Iatrogenic diseases with a common cause? Richard G Fiddian-Green (25 October 2002) eLetter re: Edward H Wagner and Trish Groves Care for chronic diseases BMJ 2002; 325: 913-914

9. Adverse Event Reporting System.

Competing interests: Patents issued in my name

Competing interests: No competing interests

15 March 2005
Richard G Fiddian-Green
Former chairman genral surgery, U Massachusetts