Portal de Periódicos da CAPES

Should every patient with suspected upper GI bleeding receive a proton pump inhibitor while awaiting endoscopy?

2008; Elsevier BV; Volume: 67; Issue: 7 Linguagem: Inglês

10.1016/j.gie.2008.02.040

1097-6779

Alan Barkun,

Esophageal and GI Pathology

It is important to contrast the pre-endoscopy from postendoscopic hemostasis uses of high-dose intravenous PPI use in patients with bleeding ulcers. Indeed, the latter is more effective and less costly in most settings.In this issue of Gastrointestinal Endoscopy, Tsoi et al1Tsoi K.K.F. Lau J.Y.W. Sung J.J.Y. Cost-effectiveness analysis of high-dose omeprazole infusion before endoscopy for patients with upper-GI bleeding.Gastrointest Endosc. 2008; 67: 1056-1063Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar provide us with an important economic analysis that complements the authoritative placebo-controlled randomized trial published last year by Lau et al2Lau J.Y. Leung W.K. Wu J.C. et al.Omeprazole before endoscopy in patients with gastrointestinal bleeding.N Engl J Med. 2007; 356: 1631-1640Crossref PubMed Scopus (254) Google Scholar on the use of proton pump inhibitors (PPIs) while awaiting an early endoscopy. Such cost-effectiveness considerations are especially important because the trial found no statistically significant between-group differences in rebleeding, surgery, or mortality rates. It is important to contrast the pre-endoscopy from postendoscopic hemostasis uses of high-dose intravenous PPI use in patients with bleeding ulcers. Indeed, the latter is more effective and less costly in most settings. Overall, 631 patients were recruited, and 377 were eventually found to be bleeding from ulcers. Among these, owing to a difference in the prevalence of high-risk lesions found in each arm, 60 (19.1%) patients in the PPI and 90 (28.4%) patients in the placebo group required endoscopic hemostasis at the time of gastroscopy (performed on average 14-15 hours after the onset of PPI administration). Corresponding rates for patients with nonulcer causes of bleeding were 5.7% and 6.3%, respectively. Tsoi et al1Tsoi K.K.F. Lau J.Y.W. Sung J.J.Y. Cost-effectiveness analysis of high-dose omeprazole infusion before endoscopy for patients with upper-GI bleeding.Gastrointest Endosc. 2008; 67: 1056-1063Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar tabulated direct costs by using unit prices from the hospital pharmacy of the Prince of Wales Hospital and the Hong Kong Government Gazette, assuming a 30-day time horizon. Overall average per patient costs were U.S. $2813 for the intravenous PPI (80 mg bolus followed by an 8 mg per hour infusion till the endoscopy), and U.S. $2948 in the placebo group. PPI administration also reduced endoscopic therapy by 7.4% and was thus both less costly and more effective than placebo, making it a dominant strategy in economic terms. The conclusions remained robust across a wide range of assumptions for the sensitive variables that included the proportion of patients undergoing endoscopic hemostasis, and the costs of PPIs, endoscopic procedures, and hospital stay. The clinical question addressed is one of the most common and disputed3Barkun A. Bardou M. Marshall J.K. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding.Ann Intern Med. 2003; 139: 843-857Crossref PubMed Scopus (498) Google Scholar quandaries facing a clinician managing patients with suspected upper GI bleeding, and Tsoi et al1Tsoi K.K.F. Lau J.Y.W. Sung J.J.Y. Cost-effectiveness analysis of high-dose omeprazole infusion before endoscopy for patients with upper-GI bleeding.Gastrointest Endosc. 2008; 67: 1056-1063Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar provide us with important information that adds to the existing literature. When approaching an economic analysis, the clinician must consider the validity of the model structure and the generalizability of assumptions of both effectiveness and costs. Previous decision models have suggested the dominance of the PPI strategy but were developed before the availability of contemporary trial data.4Enns R.A. Gagnon Y.M. Rioux K.P. Cost-effectiveness in Canada of intravenous proton pump inhibitors for all patients presenting with acute upper gastrointestinal bleeding.Aliment Pharmacol Ther. 2003; 17: 225-233Crossref PubMed Scopus (55) Google Scholar, 5Gagnon Y.M. Levy A.R. Eloubeidi M.A. et al.Cost implications of administering intravenous proton pump inhibitors to all patients presenting to the emergency department with peptic ulcer bleeding.Value Health. 2003; 6: 457-465Abstract Full Text PDF PubMed Scopus (33) Google Scholar The strength of the analysis by Tsoi et al1Tsoi K.K.F. Lau J.Y.W. Sung J.J.Y. Cost-effectiveness analysis of high-dose omeprazole infusion before endoscopy for patients with upper-GI bleeding.Gastrointest Endosc. 2008; 67: 1056-1063Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar resides principally on the model structure and the resource utilization assumptions, both drawn from actual trial data. Although there exist other published studies6Wallner G. Ciechanski A. Wesolowski M. et al.Treatment of acute upper gastrointestinal bleeding with intravenous omeprazole or ranitidine.Eur J Clin Res. 1996; 8: 235-243Google Scholar, 7Daneshmend T.K. Hawkey C.J. Langman M.J. et al.Omeprazole versus placebo for acute upper gastrointestinal bleeding: randomised double blind controlled trial.BMJ. 1992; 304: 143-147Crossref PubMed Scopus (216) Google Scholar, 8Hawkey G.M. Cole A.T. McIntyre A.S. et al.Drug treatments in upper gastrointestinal bleeding: value of endoscopic findings as surrogate end points.Gut. 2001; 49: 372-379Crossref PubMed Scopus (51) Google Scholar and even a Cochrane meta-analysis9Dorward S. Sreedharan A. Leontiadis G. et al.Proton pump inhibitor treatment initiated prior to endoscopic diagnosis in upper gastrointestinal bleeding.Cochrane Database Syst Rev. 2006; (CD005415)PubMed Google Scholar on the use of PPI before endoscopy, the Lau trial (and thus the Tsoi analysis) adopted the PPI regimen that has yielded the greatest efficacy in the postendoscopic therapy setting.10Bardou M. Toubouti Y. Benhaberou-Brun D. et al.Meta-analysis: proton-pump inhibition in high-risk patients with acute peptic ulcer bleeding.Aliment Pharmacol Ther. 2005; 21: 677-686Crossref PubMed Scopus (120) Google Scholar, 11Leontiadis G.I. Sharma V.K. Howden C.W. Proton pump inhibitor treatment for acute peptic ulcer bleeding.Cochrane Database Syst Rev. 2006; (CD002094)Google Scholar Only direct costs are assessed, as is often the case; the inclusion of indirect costs would allow for a societal perspective if informing broader decision-takers. Because of the cost data sources, the conclusions are most valid for a Hong Kong practice because converting costs into U.S. dollars does not address any variations in relative cost structure that may exist across different health care systems. These types of considerations are true for any cost analysis and are outweighed by the robustness of the conclusions. The conclusions of an economic analysis can vary according to the choice of unit of effectiveness or utility that in turn usually reflects a clinically meaningful outcome. In the current case, the unusual unit chosen was that of "averted endoscopic therapies." This outcome is only indirectly linked to the end point of subsequent rebleeding and may introduce the risk of some "double counting" because this unit of outcome appears with its cost implications as a result of resource utilization in the denominator while its related expenditures are already counted as part of the cost numerator. The robustness of the findings, however, is reassuring in this context. Understandably, the authors probably chose this outcome because the clinical trial yielded no differences in more traditional end points such as rebleeding, surgery, or mortality rates. Other units of effectiveness could be considered. The Lau trial reported significant differences in proportions of patients hospitalized less than 3 days, but adopting this outcome would have resulted in an even greater risk of double counting; furthermore, this outcome is more subject to local practices and patient population characteristics. The use of utilities and quality-adjusted life-years, although more traditional because they facilitate decisions across disparate programs of health care delivery but are poorly adapted to the brevity of the episode of bleeding, yield disproportionately large incremental cost-effectiveness ratios.12Bala M. Zarkin G. Are QALYs an appropriate measure for valuing morbidity in acute diseases?.Health Econ. 2000; 9: 177-180Crossref PubMed Scopus (48) Google Scholar, 13Al-Sabah S. Barkun A.N. Herba K. et al.Cost-effectiveness of proton pump inhibition before endoscopy in upper gastrointestinal bleeding.Clin Gastroenterol Hepatol. 2008; 6: 418-425Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar Another approach could be to adopt rebleeding as a unit of effectiveness on the basis of endoscopic treatment rates. Although such a model requires some assumptions for patients with nonulcer bleeding, it provides the opportunity to model for increased patient numbers and may bring out the economic impact of small differences in efficacy for which the clinical trial may have been underpowered. Our own group completed such an assessment using North American costs and pooled effectiveness data from the Cochrane review by Dorward et al9Dorward S. Sreedharan A. Leontiadis G. et al.Proton pump inhibitor treatment initiated prior to endoscopic diagnosis in upper gastrointestinal bleeding.Cochrane Database Syst Rev. 2006; (CD005415)PubMed Google Scholar (more conservative and heterogeneous, yet broadly generalizable). We concluded that the PPI approach was more effective but at a slightly higher cost and became a dominant strategy when varying assumptions in durations of hospital stay for high- and low-risk ulcer patients.13Al-Sabah S. Barkun A.N. Herba K. et al.Cost-effectiveness of proton pump inhibition before endoscopy in upper gastrointestinal bleeding.Clin Gastroenterol Hepatol. 2008; 6: 418-425Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar Last, a more traditional approach could have been to carry out a cost-minimization analysis where only the costs attributable to each of the strategies are compared because between-group outcomes are similar. Thus it can be concluded that even with expert groups using sophisticated modeling tools, the final conclusion on the use of PPI pre-endoscopy is not a "slam dunk" either way. It is important at this point to contrast the pre-endoscopy with postendoscopic hemostasis uses of a PPI (particularly high-dose intravenous administration) in patients with bleeding ulcers. Indeed, the latter is more effective and less costly in most settings,3Barkun A. Bardou M. Marshall J.K. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding.Ann Intern Med. 2003; 139: 843-857Crossref PubMed Scopus (498) Google Scholar, 14Barkun A.N. Herba K. Adam V. et al.High-dose intravenous proton pump inhibition following endoscopic therapy in the acute management of patients with bleeding peptic ulcers in the USA and Canada: a cost-effectiveness analysis.Aliment Pharmacol Ther. 2004; 19: 591-600Crossref PubMed Scopus (75) Google Scholar, 15Spiegel B.M. Ofman J.J. Woods K. et al.Minimizing recurrent peptic ulcer hemorrhage after endoscopic hemostasis: the cost-effectiveness of competing strategies.Am J Gastroenterol. 2003; 98: 86-97Crossref PubMed Scopus (87) Google Scholar, 16Lee K.K. You J.H. Wong I.C. et al.Cost-effectiveness analysis of high-dose omeprazole infusion as adjuvant therapy to endoscopic treatment of bleeding peptic ulcer.Gastrointest Endosc. 2003; 57: 160-164Abstract Full Text PDF PubMed Scopus (65) Google Scholar although in real life significant overuse has lessened the forecast third-party savings.17Enns R. Andrews C.N. Fishman M. et al.Description of prescribing practices in patients with upper gastrointestinal bleeding receiving intravenous proton pump inhibitors: a multicentre evaluation.Can J Gastroenterol. 2004; 18: 567-571PubMed Google Scholar, 18Guda NM, Noonan M, Kreiner MJ, Pet al. Use of intravenous proton pump inhibitors in community practice: an explanation for the shortage? Am J Gastroenterol 2004;99:1233-7.Google Scholar, 19Nardino R.J. Vender R.J. Herbert P.N. Overuse of acid-suppressive therapy in hospitalized patients.Am J Gastroenterol. 2000; 95: 3118-3122PubMed Google Scholar, 20Kaplan G.G. Bates D. McDonald D. et al.Inappropriate use of intravenous pantoprazole: extent of the problem and successful solutions.Clin Gastroenterol Hepatol. 2005; 3: 1207-1214Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar, 21Afif W. Alsulaiman R. Martel M. et al.The utilization of IV proton pump inhibitors (IV PPI): which patient population and prescribing behavior to target for a more appropriate practice?.Alim Pharmacol Ther. 2007; 25: 609-615Crossref PubMed Scopus (26) Google Scholar, 22Heidelbaugh J. Inadomi J. Magnitude and economic impact of inappropriate use of stress ulcer prophylaxis in non-ICU hospitalized patients.Am J Gastroenterol. 2006; 101: 2200-2205Crossref PubMed Scopus (136) Google Scholar Moreover, although there is some indirect support for the biologic rationale of clot stabilization attributable to profound acid suppression23Barkun A.N. Cockeram A.W. Plourde V. et al.Review article: acid suppression in non-variceal acute upper gastrointestinal bleeding.Aliment Pharmacol Ther. 1999; 13: 1565-1584Crossref PubMed Scopus (98) Google Scholar in the postendoscopic hemostasis setting, the underlying mechanisms responsible for the reproducible enhanced healing of lesions with PPI administration before endoscopy2Lau J.Y. Leung W.K. Wu J.C. et al.Omeprazole before endoscopy in patients with gastrointestinal bleeding.N Engl J Med. 2007; 356: 1631-1640Crossref PubMed Scopus (254) Google Scholar, 9Dorward S. Sreedharan A. Leontiadis G. et al.Proton pump inhibitor treatment initiated prior to endoscopic diagnosis in upper gastrointestinal bleeding.Cochrane Database Syst Rev. 2006; (CD005415)PubMed Google Scholar remain less well formulated. One key variable for which there exist almost no data is the time elapsed until endoscopy (ie, the duration of intravenous PPI administration before the gastroscopy, which in the case of the Tsoi et al analysis is the duration of intravenous PPI infusion). Indeed, it can be intuitively postulated that if all patients are to undergo endoscopy almost immediately, PPI administration may likely only generate costs for a few benefits. Alternately, if no patients undergo endoscopy within the first 48 to 72 hours of the onset of bleeding, the clinical benefits of widespread PPI use may be more likely and they approach cost saving. Some intermediate threshold value must therefore exist (less than 14-15 hours on the basis of Lau et al2Lau J.Y. Leung W.K. Wu J.C. et al.Omeprazole before endoscopy in patients with gastrointestinal bleeding.N Engl J Med. 2007; 356: 1631-1640Crossref PubMed Scopus (254) Google Scholar); unfortunately, available data do not allow us to be more precise. Qualitatively, it is reasonable to surmise that the earlier the endoscopy is performed, the less the cost-effectiveness of high-dose intravenous PPI. Furthermore, in keeping with the findings of Sung et al,24Sung J.J. Chan F.K. Lau J.Y. et al.The effect of endoscopic therapy in patients receiving omeprazole for bleeding ulcers with nonbleeding visible vessels or adherent clots: a randomized comparison.Ann Intern Med. 2003; 139: 237-243Crossref PubMed Scopus (164) Google Scholar who have generated the most compelling data comparing endoscopy to modern high-dose acid suppression in patients with Forrest lesions IIa and IIb, preemptive PPI use should not replace the performance of early endoscopy. The optimal dose and route of administration of the PPI represent another uncertainty.3Barkun A. Bardou M. Marshall J.K. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding.Ann Intern Med. 2003; 139: 843-857Crossref PubMed Scopus (498) Google Scholar, 25Leontiadis G.I. Sharma V.K. Howden C.W. Systematic review and meta-analysis: enhanced efficacy of proton-pump inhibitor therapy for peptic ulcer bleeding in Asia—a post hoc analysis from the Cochrane Collaboration.Aliment Pharmacol Ther. 2005; 21: 1055-1061Crossref PubMed Scopus (85) Google Scholar The most efficacious and widely generalizable data (thought to be a class effect) relate to high-dose intravenous use, as modeled for in the analysis of Tsoi et al.3Barkun A. Bardou M. Marshall J.K. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding.Ann Intern Med. 2003; 139: 843-857Crossref PubMed Scopus (498) Google Scholar, 10Bardou M. Toubouti Y. Benhaberou-Brun D. et al.Meta-analysis: proton-pump inhibition in high-risk patients with acute peptic ulcer bleeding.Aliment Pharmacol Ther. 2005; 21: 677-686Crossref PubMed Scopus (120) Google Scholar, 11Leontiadis G.I. Sharma V.K. Howden C.W. Proton pump inhibitor treatment for acute peptic ulcer bleeding.Cochrane Database Syst Rev. 2006; (CD002094)Google Scholar, 26Lau J.Y. Sung J.J. Lee K.K. et al.Effect of intravenous omeprazole on recurrent bleeding after endoscopic treatment of bleeding peptic ulcers.N Engl J Med. 2000; 343: 310-316Crossref PubMed Scopus (604) Google Scholar Uses of a lesser dose, or of an oral route, remain grounded in weaker evidence (at least with regard to beneficial effects on mortality rates) or may be hampered by decreased generalizability across different geographic areas.25Leontiadis G.I. Sharma V.K. Howden C.W. Systematic review and meta-analysis: enhanced efficacy of proton-pump inhibitor therapy for peptic ulcer bleeding in Asia—a post hoc analysis from the Cochrane Collaboration.Aliment Pharmacol Ther. 2005; 21: 1055-1061Crossref PubMed Scopus (85) Google Scholar Finally, it could be presumed that the downside of indiscriminant PPI use pre-endoscopy only pertains to cost considerations, but recent data have suggested an association linking in-hospital PPI use to the rare development of Clostridium difficile–associated disease.27Dial S. Alrasadi K. Manoukian C. et al.Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies.Can Med Assoc J. 2004; 171: 33-38Crossref PubMed Scopus (480) Google Scholar None of the cost analyses have modeled for this rare but possible negative outcome, which would be especially relevant to the subgroup of patients receiving no benefit from the PPI administration. So what is a clinician, a hospital pharmacist, or a hospital administrator to do about the use of PPI for patients with suspected upper GI bleeding before endoscopy in an era where needle-to-door times have become the paradigm for many therapies administered by emergency care providers? First and foremost, the existing data cannot lure clinicians into a false sense of security when administering a PPI while awaiting endoscopy. Adequate resuscitation and early endoscopy must be provided because any benefit attributable to PPI administration will only be modest on an individual basis in this setting. Next, the administration of a PPI before endoscopy is appropriate on the basis of current evidence, but probably an effort should be made to better identify subgroups of patients or situations in which this use will be more likely to be cost-effective, situations such as when the endoscopy is delayed or when the patient is more likely to be bleeding from a nonvariceal source, especially if the lesion is expected to exhibit a high-risk stigma (for example, in the case of a patient with a bloody nasogastric aspirate28Aljebreen A.M. Fallone C.A. Barkun A.N. Nasogastric aspirate predicts high-risk endoscopic lesions in patients with acute upper-GI bleeding.Gastrointest Endosc. 2004; 59: 172-178Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar). A cost-effective PPI use is also more probable when the practice setting is one in which the difference of hospital stays between low- and high-risk patients is greater and where the magnitude of the PPI costs is more modest relative to the expenses generated by the endoscopy, endoscopic hemostasis, and hospitalization. It is humbling to realize that, despite currently available data on the use of PPI in upper GI bleeding, many gaps persist in our knowledge of best practice. Clinicians must remain abreast of emerging data, and consensus groups must strive to diffuse and implement updated recommendations in a timely fashion. The author reports the following conflicts: A. N. Barkun is a consultant and has lectured for AstraZeneca and Abbott Canada Inc, and has received at arms length support for research from AstraZeneca Canada, Abbott Canada, and Olympus Canada. A. N. Barkun is a research scholar (Chercheur National) of the Fonds de la Recherche en Santé du Québec and is the current holder of the Doug G. Kinnear Chair in Gastroenterology at McGill University.