An alternative interpretation of the data
The conclusion of the meta-analysis by Leontiadis et al.  that proton pump inhibitors are of benefit in the acute management of peptic ulcer haemorrhage is clearly open to question.
In terms of mortality, none of the 18 studies included in the analysis of the primary end-point reported any significant reduction in the death from bleeding peptic ulcers with proton pump inhibitors. Indeed, there was a non-significant increase in mortality in patients receiving proton pump inhibitors compared with controls (5.2% v 4.6%). Given this, it is difficult to understand why the authors felt obliged to comment that “… there may have been too few patients in our pooled analysis of mortality data to enable us to detect a difference” which, in the context of the discussion section, must be taken to imply a difference in favour of proton pump inhibitors. With regard to the secondary end-points, only four of 19 studies reported a reduction in re-bleeding and only two of 17 showed a statistically significant reduction in surgery. It is extraordinary, therefore, that the authors stated that “It is, however, a remarkably consistent observation that such treatment reduces the rates of rebleeding and, in general, the need for surgical intervention”. Where, one may ask, is the consistency? And where is the impartiality? Surely authors with such close links to the pharmaceutical industry should make more of an effort to dispel any suggestion of bias in the interpretation of the data.
Of the 21 studies included in the meta-analysis, only seven were placebo-controlled [2-8]; these were the largest trials and, unlike some of the remainder, all were double-blind. The three largest studies were from Europe: none demonstrated any reduction in either mortality or re- bleeding; only one reported a reduction in surgery and in this study the upper 95% confidence interval was close to one. [2-4] The four other placebo-controlled trials were carried out in Asia: again, none reported any reduction in mortality; three showed a statistically significant decrease in re-bleeding while only one reported a reduction in surgery. [5 -8] Are these differences accounted for by variations in response to proton pump inhibitors in different populations or do they reflect other differences in the design, performance or analysis of trials? Furthermore, should these differences be taken into account when considering the external validity of the data?
The meta-analysis by Leontiadis et al. once again draws attention to the limitations of this methodology. The studies analysed were heterogeneous in terms of the patients recruited, the drugs used, the routes of administration and the controls. It is not surprising, therefore, that differences in outcome were observed between different studies, leading to confusion and uncertainty. Yet, when the forest plots are cleared away and the fog of statistics is dispersed, there are some useful conclusions to be gleaned from the data set provided in this study. Firstly, proton pump inhibitors do not reduce mortality from peptic ulcer haemorrhage. Secondly, in European populations, there is no evidence that these drugs reduce the rate of re-bleeding or surgery. And finally, given the absence of any therapeutic benefit in favour of parenteral over oral administration and accepting the failure to reduce mortality, re-bleeding or surgery, there can surely be no justification for the current fashion for intravenous proton pump inhibitors in the management of patients with peptic ulcer haemorrhage.
In one respect, the study by Leontiadis et al. leaves the management of peptic ulcer haemorrhage unchanged. Patients with bleeding ulcers will, of course, continue to receive proton pump inhibitors and, as before, these drugs will be prescribed to promote healing. But the notion that these drugs improve survival or reduce the risk of re-bleeding or surgery – at least in European populations – is without foundation and, hence, the emergency care of patients with bleeding peptic ulcers should no longer include intravenous proton pump inhibitors.
 Leontiadis GI, Sharma VK, Howden CW. Systematic review and meta- analysis of proton pump inhibitor therapy in peptic ulcer bleeding. BMJ 2005;330;568-70.
 Daneshmend TK, Hawkey CJ, Langman MJ, Logan RF, Long RG. Omeprazole versus placebo for acute upper gastrointestinal bleeding: randomised double blind controlled trial. BMJ 1992;304;143-7.
 Hasselgren G, Lind T, Lundell L, Aadland E, Efskind P, et al. Continuous intravenous infusion of omeprazole in elderly patients with peptic ulcer bleeding. Results of a placebo-controlled multicentre study.Scand J Gastroenterol 1997;32;328-33.
 Schaffalitsky de Muckadell OB, Havelund T, Harling H, Boesby S, et al. Effect of omeprazole on the outcome of endoscopically treated bleeding peptic ulcers. Randomised double-blind placebo-controlled multicentre study. Scand J Gastroenterol 1997;32;320-7.
 Khuroo MS, Yattoo GN, Javid G, Khan BA, Shah AA, et al. A comparison of omeprazole and placebo for bleeding peptic ulcer. N Engl J Med 1997;336;1054-58.
 Lau JYW, Sung JJY, Lee KKC, Yung M, Wong SKH, et al. Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med 2000;343;310-6.
 Javid G, Masoodi I, Zargar S, Khan BA, Yatoo GN, et al. Omeprazole as adjuvant therapy to endsocopic combination sclerotherapy for treating bleeding peptic ulcer. Am J Med 2001;111;280-4.
 Kaviani MJ, Hashemi MR, Kazemifar AR, Roozitalab S, et al. Aliment Pharmacol Ther 2003;17;211-6.
Competing interests: None declared
Competing interests: No competing interests