December 7, 2009 — A new randomized trial--albeit a small one--suggests that continuing aspirin in patients with cardiovascular disease who develop peptic-ulcer bleeding will, not surprisingly, double their risk of bleeding but may also radically reduce their risk of all-cause mortality [1]. The increased bleeding was seen despite all patients receiving a 72-hour infusion of the proton-pump inhibitor (PPI) pantoprazole, followed by oral pantoprazole, after undergoing endoscopic therapy.

The paper, by Dr Joseph JY Sung and colleagues, is published online December 1, 2009 in the Annals of Internal Medicine and concludes that "early resumption" of low-dose aspirin "should be considered" in patients with bleeding ulcers to minimize the mortality risk of stopping aspirin.

To heartwire , Sung emphasized that the study was not powered to detect a difference in mortality: "Therefore, the possibility that the difference in all-cause mortality is merely [due to] chance cannot be excluded. However, cardiologists and gastroenterologists should be aware that there is a likelihood of increasing risk of cardiovascular death if aspirin is withheld for a prolonged period. A balance of risk and benefit of restarting aspirin should be considered."

To Stop or Not Stop

The study enrolled 156 patients already taking aspirin for secondary prevention of cardiovascular or cerebrovascular events who developed peptic-ulcer bleeding; after diagnosis and treatment (hemostasis achieved) of a bleeding ulcer, patients were randomized to receive low-dose aspirin or placebo for eight weeks. (Of note, patients taking clopidogrel were included in the study, but clopidogrel was stopped until the ulcer was completely healed.) Three patients withdrew during the course of the study.

In an intention-to-treat analysis, recurrent ulcer bleeding at 30 days--the study's primary end point--was nearly twice as high in the aspirin group as the placebo group. For the secondary end point of all-cause mortality, placebo-treated patients had a more than 10-fold increase in events, a statistically significant difference. Aspirin-treated patients also had lower rates of mortality due to cardiovascular, cerebrovascular, or gastrointestinal events.

Bleeding and Mortality Outcomes

Event	Aspirin (%)	Placebo (%)
Recurrent ulcer bleeding, 30 d	10.3	5.4
All-cause mortality, 8 wk	1.3	12.9
Death due to cardiovascular, cerebrovascular, or gastrointestinal complications, 8 wk	1.3	10.3

"Among low-dose-aspirin recipients who had peptic-ulcer bleeding, continuous aspirin therapy may increase the risk for recurrent bleeding but potentially reduces mortality rates," the authors cautiously conclude. "Larger trials are needed to confirm these findings."

Commenting on the findings for heartwire , Dr Dominick Angiolillo (University of Florida, Jacksonville) emphasized the small size of the study, pointing out that, typically, much larger randomized controlled trials have had "difficulty" showing differences in mortality; as such, the mortality differences seen by Sung et al "need to be taken with a grain of salt."

"Especially since this [mortality finding] is a secondary end point, this needs to be considered with caution," Angiolillo warned. "The other thing to keep in mind is that the study outcomes were evaluated out to eight weeks, and we do know that with bleeding, there can be potential clinical implications even long-term, beyond the eight weeks, which obviously are not reported in this study."

Long-term effects of bleeding include future ischemic events and CV mortality, he added.

"I would not make any definitive statements on what should be done [on the basis of this study]," Angiolillo told heartwire . "What we clearly know is that if you continue patients on aspirin, the patient is likely going to bleed, and the PPI, even given under infusion, is not going to prevent that."

What the study underscores is the need to evaluate patients on a case-by-case basis, he continued. Some patients on antiplatelet medication who develop bleeding may have good indications for continuing on aspirin.

"One indication may be a patient with a recent MI, who got a DES placed in the proximal LAD a week ago." he said. "That's very different from another patient who is on aspirin for a CV event [that took place] two years ago."

An editorial accompanying the study also urges caution in interpreting the study results, pointing to a number of methodological issues with the study [2]. But Drs Alan N Barkun (Montreal General Hospital, QC) and Marc Bardou (Centre Hospitalier Universitaire de Dijon, France) also urge physicians to "reconsider the all-too-common paradigm" of focusing on the gut at the expense of the heart.

"On the basis of all available data, international consensus recommendations (including the results from Sung and colleagues) concluded that patients with upper gastrointestinal bleeding who require secondary cardiovascular prophylaxis should resume low-dose aspirin therapy as soon as the cardiovascular risks outweigh the gastrointestinal risks (usually within seven days)," they note. "Until additional data become available to better guide management, clinicians will need to rely on limited evidence and appropriate use of common sense that considers the patient as a whole without focusing on a specific organ system to the detriment of another."

Author disclosures are listed in the article.

References

1. Sung JJY, Lau JYW, Ching JYL, et al. Continuation of low-dose aspirin in peptic ulcer bleeding. Ann Intern Med 2009; available at: http://www.annals.org.
2. Barkun AN and Bardou M. Aspirin withdrawal in acute peptic ulcer bleeding: Are we harming patients? Ann Intern Med 2009; available at: http://www.annals.org.

Clinical Context

Study Highlights

• The study was completed at 1 university medical center in Hong Kong and included consecutive patients evaluated for overt gastrointestinal bleeding between 2003 and 2006.
• Patients eligible for the current study protocol had evidence of acute peptic ulcer bleeding and also had indications for continuous therapy with aspirin at a dose of 325 mg daily or less. Patients who used aspirin for primary prophylaxis of cardiovascular disease were not eligible for randomization, nor were those with ulcer perforation or who were receiving concomitant therapy with corticosteroids or other nonsteroidal anti-inflammatory drugs.
• After endoscopy, all study patients received pantoprazole, 80 mg intravenously, followed by an infusion of pantoprazole, 8 mg per hour, for 72 hours. They then were given oral pantoprazole 40 mg daily for 8 weeks.
• Participants were randomly assigned to receive aspirin 80 mg daily or placebo for the 8-week trial period.
• The primary endpoint of the study was recurrent symptomatic peptic ulcer bleeding within 30 days of endoscopic treatment. Researchers also observed all-cause mortality and requirements for blood transfusion and hospital stay.
• 3412 patients received a diagnosis of upper gastrointestinal bleeding, and 156 patients underwent randomization in the current trial. The mean age was 74 years, and 62% of participants were men. None of the study subjects was receiving antiplatelet therapy other than aspirin at baseline.
• Continuing aspirin therapy was associated with a higher 30-day risk for recurrent ulcer bleeding vs placebo (incidence of 10.3% vs 5.4%, respectively). The site of recurrent bleeding was the same as the original bleeding site in all cases.
• However, the total amount of blood transfused was similar between the 2 randomized groups, indicating that recurrent peptic ulcer bleeding may have not been severe. The rate of hospital stay during the first 30 days was also similar between groups.
• All-cause mortality rate at 8 weeks was higher in the placebo group (12.9%) vs the aspirin group (1.3%). Mortality rate from cardiovascular, cerebrovascular, or gastrointestinal causes specifically was also lower in the aspirin group vs the placebo group.
• 4 nonfatal ischemic events occurred in the placebo group vs only 2 in the aspirin group. Other adverse events were disparate in nature and were more common in the aspirin group vs the placebo group.

Clinical Implications

• Aspirin therapy is associated with a 2- to 3-fold increase in the risk for peptic ulcer bleeding, and this risk is even higher among older patients with H pylori infection. Aspirin-induced peptic ulcers are associated with lower rates of dyspepsia, and there appears to be no tolerance with time to the gastropathic effects of aspirin.
• In the current study of older adults with peptic ulcer bleeding treated with proton-pump inhibitors, the use of aspirin after the bleeding event was associated with a higher risk for recurrent ulcer bleeding but a lower risk for overall mortalit