Gastrointestinal Endoscopy
Volume 70, Issue 2 , Pages 225-228, August 2009

GI bleeding: problems that persist

Raymond G. Murphy VA Medical Center, Division of Gastroenterology, University of New Mexico, Albuquerque, New Mexico, USA

Article Outline

Abbreviations: ASA, acetylsalicylic acid, COX, cyclooxygenase, NSAID, nonsteroidal anti-inflammatory drug, PPI, proton pump inhibitor, RR, relative risk, SSRI, selective serotonin reuptake inhibitor, USFDA, US Food and Drug Administration

 

In GI bleeding, increasing age, the presence of comorbid illnesses, and iatrogenic impaired hemostasis together encompass the highest risks of a poor outcome.

Despite a half century of serious attention, managing GI bleeding still occupies center stage among the daily challenges faced by physicians. In this issue of Gastrointestinal Endoscopy, Loperfideo et al,1 in a large population study from Italy, report a very significant reduction in mortality from GI bleeding over a 20-year interval, ending in 2004. As might be predicted, much of the reduction arises from improved outcomes of variceal bleeding and, to a lesser extent, peptic ulcer disease. That these benefits result principally, but not solely, from therapeutic endoscopic interventions, is generally accepted and has been accompanied by the publication of evidence-based but still-evolving guidelines for managing variceal and ulcer bleeding.2, 3, 4 Data are still incomplete on a number of additional questions that largely pertain to these types of bleeding. This includes final details on the optimal use of endoscopic clips and several other newer approaches (eg, suturing, stapling, loops, cryotherapy, Doppler US), the use of intravenous erythromycin as an aid to endoscopy, the removal of a clot from bleeding vessels by nonexperts, the role of the interventional radiologist in arresting hemorrhage, and the use of transjugular intrahepatic portosystemic shunts in refractory variceal bleeding. These modifications are likely to produce only marginal gains where progress has already been achieved. Despite wide use, based largely on evidence from pooled data,5, 6 no proton pump inhibitor (PPI) has yet been approved by the US Food and Drug Administration (USFDA) as effective in decreasing GI bleeding. Available data5, 6, 7 best support intravenous PPI use as an adjunct to endoscopic therapy in actively bleeding ulcers or those with evidence of recent hemorrhage (Forrest classes 1 and 2). Furthermore, the effects of oral or intravenous PPI therapy before endoscopy in achieving hemostasis or in preventing rebleeding or bleeding associated with acetylsalicylic acid (ASA) or nonsteroidal anti-inflammatory drug (NSAID) use await further clarification. Currently, the use of PPI infusions for GERD is approved by the USFDA.

Beyond all this, what lies at the remaining hard core of nonvariceal, nonulcer GI bleeding? The study from Italy reveals much of interest, as discussed by the authors, but the heart of what emerges is that increasing age, the presence of comorbid illnesses, and iatrogenic impaired hemostasis together encompass the highest risks of a poor outcome. We have little control over the risks attending old age or the prevalence of comorbid illnesses, but regulation of hemostasis in GI bleeding merits serious attention. Whether it is caused by anticoagulants or antiplatelet agents, many commentators continue to try to explain all GI bleeding in the context of ulceration caused by drugs instead of placing more emphasis on the abilities of numerous drugs (by their effects on platelet function or coagulation) to cause GI and other bleeding, from either preexisting tissue lesions (often clinically unrecognized, trivial, or asymptomatic) or disturbances of hemostasis so severe as to cause spontaneous bleeding or to prevent the repair of what might be termed physiologic injuries from food, hypoxia, ischemia, minor inflammation, or other common processes. Similar bleeding, without identifiable lesions, is also seen during therapy with antibodies directed solely against angiogenesis factors such as vascular endothelial growth factor, antibodies that in themselves do not cause ulcers.8

This kind of bleeding is particularly seen during therapy with either Coumadin (warfarin)-type or heparin-type anticoagulant drugs; evidence that these drugs cause ulcers is completely lacking. Yet almost 12% of bleeds in the current study (6% outpatient, 39% inpatient) were associated with anticoagulant use (particularly heparin), regardless of the presence or absence of a lesion. Many of the patients were older than 70 years old and harbored Helicobacter pylori.1 Because of age-cohort associated decreases in both peptic ulcer mortality and prevalence of H pylori infection, true peptic ulcer-associated bleeding is likely to diminish further with time.9, 10, 11 Among recent outpatients in this report, 46.4% were using ASA, NSAIDs, or both. Similarly, in both Europe and the United States, aspirin is the precipitator of hospital admissions for GI bleeding in more than 50% of cases, accounting for more bleeding than all other drugs combined,12, 13 and more than 70% of all peptic ulcer bleeds occur in patients using ASA or NSAIDs.14 The number of users of low-dose aspirin has risen dramatically, accounting for the use of most currently manufactured aspirin.

Among NSAIDs, associations with GI bleeding vary widely as do their effects on mucosal damage and platelet function15; their effects on hemostasis have been little studied. Recently, it also emerged that with cotherapy with low-dose aspirin, the risk of hospitalization from also using an NSAID varies considerably with the specific NSAID and may not increase.16, 17 With ASA use, in healthy normal volunteers, platelet function is severely impaired by as little as 10 mg/d.18 In large epidemiologic studies, an increased risk of GI bleeding is significantly associated with ASA use at doses of less than 100 mg/d, increasing with dose, whereas a significantly increased risk of clinically apparent gastric ulcer is associated only with doses starting at approximately 1000 mg/d.19 Furthermore, although the risk of ASA causing GI hemorrhage persists for as long as therapy continues, it is markedly decreased within 5 to 7 days of withdrawing the drug, a time very comparable to the life of the platelet.19 Thus, the key effect of ASA therapy is to cause GI bleeding (regardless of the site in the GI tract) and not just to cause ulcers that then bleed, although this too can occur at higher doses.

For ulcer bleeding with ASA/NSAID use, a history of peptic ulcer, particularly one that has bled, seems to be the most important risk factor for bleeding. Other true ulcer complications occur far less commonly. Related to this, the majority of ulcers that bleed are in the duodenum, not in the stomach, where most mucosal injury by drug occurs.1, 20 Furthermore, the argument that most ASA/NSAID-associated GI bleeding is from ulcers is questionable. The VIGOR outcome study of naproxen versus rofecoxib showed that only a minority of GI bleeds originated from lesions in acidic areas proximal to the ligament of Treitz (rofecoxib, 13/31 bleeds; naproxen, 39/82 bleeds).21, 22 Relevant to this issue, the meta-analysis by Huang et al23 indicated that H pylori infection, not ASA/NSAID use, was the principal risk factor for ulceration, whereas ASA/NSAID use was the principal risk factor for GI bleeding. Together these data support the view that inhibiting the action of platelets in hemostasis is critical to bleeding; any GI protective effects of selective cyclooxygenase (COX)-2 inhibitor drugs may be attributed as much to sparing COX-1 activity in platelets as to gastroprotection. COX-2 selective drugs may well protect in nonacidic sites.24

Although bleeding is by far the most common GI complication of ASA/NSAID therapy, the view that most GI bleeding is an ulcer complication and somehow related to acid continues to be widely proposed, although that is now attracting close scrutiny by the USFDA.25 In addition, an explosion of research on the various mechanisms of platelet action reveals that, whether antiplatelet drugs inhibit platelet adhesion, platelet activation, or platelet thrombus propagation, all of them cause bleeding in the gut, brain, and other sites.26 Furthermore, both severe acquired thrombocytopenia and hereditary platelet defects also cause spontaneous bleeding. How much inhibition of platelets by drugs is associated with bleeding is not well studied. Nevertheless, the hemorrhagic effects of NSAIDs are additive with those of ASA, selective COX-2 inhibitors, and other drugs such as clopidogrel; this last drug has little or no ulcerogenic effect27 but causes GI bleeding at rates only slightly lower than those of ASA. This suggests that platelet inhibitory, not ulcerogenic, properties of such drugs are integral to their causing bleeding.

At present, multiple types of different antiplatelet drugs are being used in the same patient, with consequent increases in GI and other bleeding, particularly in patients with cardiovascular diseases, many of whom are elderly and hospitalized and have comorbid illnesses, the core group. The contribution of antiplatelet drugs to the overall incidence of GI bleeding has prompted the recent publication of consensus recommendations from the American College of Cardiology Foundation, American Heart Association, and American College of Gastroenterology on the combined use of antiplatelet and NSAID agents.28 Their recommendation that “a gastroprotective therapy should be prescribed for at-risk patients” using any NSAID or ASA would protect bleeders with ulcerative or erosive lesions in acid-exposed areas, mainly the esophagus, stomach, and duodenum, regardless of whether the lesions are drug induced or preexisting. Cited studies29, 30 from Hong Kong, however, showing superior safety of ASA plus esomeprazole compared with clopidogrel used alone, relate only to the outcome of GI bleeds arising from endoscopically proven ulcers (ie, documented ulcer bleeding was a study entry requirement) and do not include outcomes of nonulcer or total GI bleeding. Cotherapy with other PPIs, histamine-2 receptor antagonists, or low-dose misoprostol to prevent GI bleeding has not been adequately studied. The effect of impaired hemostasis on bleeding from the rest of the GI tract receives little attention in the guidelines. Meanwhile, research on platelet pharmacotherapy does not discount the possibility of finding drugs that would dissociate platelet effects that oppose thrombosis from those that cause bleeding.26 Most recently, a number of studies have indicated that co-use of clopidogrel and some (but not all) PPIs may increase the risk of heart attack by 50% to 80% compared to use of clopidogrel alone.31, 32 Clopidogrel is a pro-drug requiring activation after absorption by cytochrome P450-2C19. PPI drugs, which inhibit CYP2C19, greatly reduce clopidogrel's effect on platelets, apparently with adverse clinical consequences, although the USFDA, because of possible confounding factors in the studies, has thus far not required any change in clinical practice.33

A final potentially important issue pertaining to platelets, but ignored by the guidelines, concerns the effects of widely used selective serotonin reuptake inhibitor (SSRI) antidepressants. In numerous published studies, the use of SSRIs alone, through effects on platelet function, seems to approximately double the relative risk (RR) of GI bleeding and even more so use of the subclass of these drugs (eg, selective serotonin and norepinephrine reuptake inhibitors).34, 35 Although these drugs do not appear to increase the risk of bleeding when used with other solely antiplatelet agents, they do seem to increase the risk when used with NSAIDs or ASA. In the absence of cotherapy with acid-suppressive drugs, the RR associated with SSRI plus NSAID use was 9.1 (95% CI, 4.8-17.3), but when acid-suppressant therapy was also used, the risk became nonsignificant (RR = 1.3; 95% CI, 0.5-3.3),29 suggesting that SSRIs cause bleeding only in the presence of mucosal injury. Thus, the effect of SSRIs somewhat resembles that of clopidogrel, which is most dangerous in the presence of a mucosal lesion or when used with ASA or another NSAID. Clearly, a need remains for clinically significant decreases in the GI bleeding and associated mortality, which are still high in the vulnerable group (elderly patients with underlying serious diseases who use antihemostatic drugs). However, these reductions seem more likely to result from improved pharmacologic regulation of hemostasis, more cotherapy with antiulcer drugs, and better patient compliance with such therapy than from further advances in therapeutic endoscopy.

Back to Article Outline

Disclosure 

The author disclosed no financial relationships relevant to this publication.

Back to Article Outline

References 

  1. Loperfido S, Baldo V, Piovesana E, et al. Changing trends in acute upper-GI bleeding: a population-based study. Gastrointest Endosc. 2009;70:212–224
  2. Gralnek IM, Barkun AN, Bardou M. Management of acute bleeding from a peptic ulcer. N Engl J Med. 2008;359:928–937
  3. Standards of Practice Committee. ASGE Guideline: the role of endoscopy in the management of variceal hemorrhage. Gastrointest Endosc. 2005;62:651–655
  4. ASGE Standards of Practice Committee. ASGE guideline: role of endoscopy in acute nonvariceal upper-GI hemorrhage. Gastrointest Endosc. 2004;60:497–503
  5. Bardou 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;21:677–686
  6. Leontiadis GI, Sharma VK, Howden CW. Proton pump inhibitor treatment for acute peptic ulcer bleeding. Cochrane Database Syst Rev. 2006;1:CD002094
  7. McCarthy DM. Management of bleeding peptic ulcer: current status of intravenous proton pump inhibitors. Best Pract Res Clin Gastroenterol. 2004;18(Suppl 1):7–12
  8. Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II. Randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol. 2003;21:60–65
  9. Sonnenberg A. Time trends of ulcer mortality in Europe. Gastroenterology. 2007;132:2320–2327
  10. Sonnenberg A. Time trends of ulcer mortality in non-European countries. Am J Gastroenterol. 2007;102:1101–1107
  11. Rupnow MFT, Shacter RD, Owens DK, et al. A dynamic transmission model for predicting trends in Helicobacter pylori and associated diseases in the United States. Emerging Infect Dis. 2000;6:228–237
  12. Lanas A, Sekar MC, Hirschowitz BI. Objective evidence of aspirin use in both ulcer and non-ulcer upper and lower gastrointestinal bleeding/perforation. Gastroenterology. 1992;103:8629–8637
  13. Stack WA, Atherton JC, Hawkey GM, et al. Interactions between Helicobacter pylori and other risk factors for peptic ulcer bleeding. Aliment Pharmacol Ther. 2002;16:497–506
  14. Pritchard JP, Kitchingman GK, Walt RP, et al. Human gastric mucosal bleeding induced by low-dose aspirin but not by warfarin. Br Med J. 1989;298:493–496
  15. McCarthy DM. The comparative toxicity of non-steroidal anti-inflammatory drugs. Am J Med. 1999;107:37–47
  16. Rahme E, Nedjar H, Bizzi A, et al. Hospitalization for gastrointestinal adverse events attributable to the use of low-dose aspirin among patients 50 years or older also using non-steroidal anti-inflammatory drugs: a retrospective cohort study. Aliment Pharmacol Ther. 2007;26:1387–1398
  17. Gladding PA, Webster MWI, Farrell HB, et al. The antiplatelet effect of six non-steroidal anti-inflammatory drugs and their pharmacodynamic interaction with aspirin in healthy volunteers. Am J Cardiol. 2008;101:1060–1063
  18. Cryer B, Feldman M. Effects of very low dose daily, long term aspiring therapy on gastric, duodenal and rectal prostaglandin levels and on mucosal injury in healthy humans. Gastroenterology. 1999;117:17–25
  19. McCarthy DM. Non-steroidal anti-inflammatory drugs—the clinical dilemmas. Scand J Gastroenterol. 1992;192:9–16
  20. Kurata J, Abbey DE. The effect of chronic aspirin use on duodenal and gastric ulcer hospitalization. J Clin Gastroenterol. 1990;12:260–266
  21. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343:1520–1528
  22. McCarthy DM. Ulcers, Helicobacter pylori infection, platelets and gastrointestinal complications of non-steroidal anti-inflammatory drugs: what are the connections?. Eur J Surg. 2002;587:89–99
  23. Huang J-Q, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic ulcer disease: a meta-analysis. Lancet. 2002;359:14–22
  24. Laine L, Connors LG, Reicin A, et al. Serious lower gastrointestinal adverse clinical events with non-selective NSAID or coxib use. Gastroenterology. 2003;124:288–292
  25. FDA raising approval bar for NSAID/PPI combos: end-points under review. The Pink Sheet. 2008;70:31–32
  26. Barrett NE, Holbrook L, Jones S, et al. Future innovations in anti-platelet therapies. Br J Pharmacol. 2008;154:918–939
  27. Fork FT, Lafolie P, Toth E, et al. Gastro-duodenal tolerance of 75 mg clopidogrel versus 325 mg of aspiring in healthy volunteers. Scand J Gastroenterol. 2000;35:464–469
  28. Bhatt DI, Scheiman J, Abraham NS, et al. ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of anti-platelet therapy and NSAID use. Am J Gastroenterol. 2008;103:2890–2897
  29. Chan FKL, Ching JYL, Hung LCT, et al. Clopidogrel versus aspirin and esomeprazole to prevent recurrent ulcer bleeding. N Engl J Med. 2005;352:238–244
  30. Lai K-C, Chu K-M, Hui W-M, et al. Esomeprazole with aspirin versus clopidogrel for prevention of recurrent gastrointestinal ulcer complications. Clin Gastroenterol Hepatol. 2006;4:860–865
  31. Aubert RE, Epstein RS, Teagarden JR, et al. Proton pump inhibitors effect on clopidogrel effectiveness: the clopidogrel medco outcomes study. [abstract] Circulation. 2008;118(S815):3998
  32. Juurlink DN, Gomes T, Ko DT, et al. Population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ. 2009 Jan 28;[Epub ahead of print]
  33. Early communication about an ongoing safety review of clopidogrel bisulfate (marketed as Plavix). Available at: http://www.fda.gov/cder/drug/early_comm/clopidogrel_bisulfate.htm. Accessed January 27, 2009.
  34. Wessinger S, Kaplan M, Choi L, et al. Increased use of selective serotonin reuptake inhibitors in patients admitted with gastrointestinal hemorrhage: a multicentre retrospective analysis. Aliment Pharmacol Ther. 2006;23:937–944
  35. De Abajo FJ, Garcia-Rodriguez LA. Risk of upper gastrointestinal tract bleeding associated with selective serotonin reuptake inhibitors and venlafaxine therapy. Arch Gen Psychiatry. 2008;65:795–803

PII: S0016-5107(08)03429-9

doi:10.1016/j.gie.2008.12.247

Gastrointestinal Endoscopy
Volume 70, Issue 2 , Pages 225-228, August 2009

Article Tools