Acetylsalicylic acid use in patients with acute myocardial infarction and peptic ulcer bleeding

1Department of Medicine, University of Alberta Hospital, University of Alberta; 2Integrated Centre for Care Advancement through Research (iCARE), Edmonton, Alberta Correspondence: Dr Justin Cheung, Zeidler Ledcor Centre, 130 University Campus, Edmonton, Alberta T6G 2X8. Telephone 780-492-8243, fax 780-492-1455, e-mail justin.cheung@ualberta.ca Received for publication January 3, 2009. Accepted April 7, 2009 Antiplatelet agents, such as acetylsalicylic acid (ASA), are important for the treatment of acute myocardial infarction (AMI) (1). A subset of patients with cardiovascular disease is also on maintenance ASA therapy for primary or secondary prevention (2-4). However, ASA is a well-known risk factor for peptic ulcer disease (PUD), with the potential for bleeding (5). The optimal early use of ASA in the acute setting of an AMI and concurrent PUD bleeding is unknown. Withholding ASA may prevent optimal treatment of the AMI, whereas continuing ASA may increase ulcer rebleeding. The real-life practice of ASA use in the setting of AMI complicated by PUD bleeding is unknown (6-8). The objectives of the present study were to determine the current practices and predictors of continued ASA use in patients with AMI and concurrent PUD bleeding. METHODS A retrospective review of patient medical records between January 1999 and December 2006 was undertaken at the University of Alberta and Royal Alexandra Hospitals (Edmonton, Alberta), two tertiary care hospitals affiliated with the University of Alberta. Patient records were considered eligible for inclusion if they had a hospitalization diagnosis of AMI and gastrointestinal bleeding or PUD bleeding based on the International Classification of Diseases – Ninth Revision coding. Patients were included if PUD bleeding (hematemesis, melena, rectal bleeding or new-onset anemia) started at least 24 h after the diagnosis of AMI. Patients with potential demand-ischemia AMI due to PUD bleeding (presenting with PUD bleeding before or within 24 h of AMI) were excluded. AMI cases were identified based on a diagnosis of myocardial originAl Article


Outcome measures
The 30-day rebleeding rate (9,10) was defined as the need for endoscopy based on clinical evidence of rebleeding (eg, hematemesis or melena, or a drop in hemoglobin of more than 20 g/L) within 30 days after stabilization and a diagnosis of PUD rebleeding (ie, visualization of blood in the upper gastrointestinal tract with high-risk PUD stigmata present) made by the endoscopist for an ulcer previously documented during the same hospital stay, with or without stigmata of recent hemorrhage (11,12).Other outcome measures included total blood transfusions and in-hospital 30-day mortality.

Ethics
The study was approved by the University of Alberta and the institutional ethics review boards of the respective hospitals.

Statistical analysis
Statistical analysis was performed using SAS version 9.1 software (SAS Institute Inc, USA).Bivariate testing with the Fisher's exact test and Student's t test were used for comparing categorical and continuous variables, respectively.A P<0.05 (based on a two-tailed test) was considered to be statistically significant.Continuous variables were described as mean ± SD.Multivariate logistic regression using a stepwise selection approach was used for the analysis of C-ASA versus D-ASA subjects, and the results are expressed with ORs and 95% CIs.Covariates included in the regression analysis included age, sex, ulcer stigmata, Rockall score, endoscopic treatment, ST segment elevation myocardial infarction versus non-ST segment elevation myocardial infarction, cardiac Killip score, thrombolysis, angioplasty, coronary stent, coronary bypass grafting and cardiac drug treatment only.

RESULTS
The initial search for patients with diagnoses of both AMI and gastrointestinal bleeding produced 710 records.After manually reviewing these records, 102 patients were identified to have experienced an AMI complicated by acute PUD bleeding.Patient characteristics are presented in Table 1.Patients in the C-ASA group had a significantly lower Rockall score (P=0.02) and a greater proportion with low-risk ulcer stigmata than subjects in the D-ASA group (P=0.02).
The management of AMI is presented in Table 2.There was a higher percentage of patients in the C-ASA group than in the the D-ASA group receiving medications such as heparin (87% Management strategies for treating PUD bleeding are shown in Table 3. Oral or intravenous proton pump inhibitors (PPIs) were used in all patients initially.Patients with high-risk stigmata ulcers were treated with intravenous pantoprazole infusion 80 mg, then 8 mg/h for 72 h.PPI therapy was  4. Two cases of rebleeding occurred in 28 ulcers with low-risk stigmata in the C-ASA group.Two cases of rebleeding in 31 ulcers with low-risk stigmata occurred in the D-ASA group.

DISCUSSION
The present study evaluated the practices of ASA use in the management of AMI accompanied by PUD bleeding at two tertiary care hospitals.We found that 37% of patients with AMI and concurrent PUD bleeding were immediately continued on ASA therapy.The continuation of ASA was associated with lower-risk ulcers and patients who received coronary intervention.
The endoscopic finding of an ulcer with low-risk stigmata, such as a clean base or a flat, pigmented spot, was found to be an independent predictor for the immediate continuation of ASA therapy despite PUD bleeding.Previous studies (11,12) in the non-AMI setting have reported that patients with lowrisk stigmata have minimal risk (less than 1% to 3%) of rebleeding while on acid-suppressive therapy.Although it is generally recommended to discontinue ASA to allow for PUD healing, there is evidence to suggest that PUD healing continues while on PPI therapy, despite continued ASA use (13,14).However, it remains unknown what the early risk of rebleeding is in the setting of AMI with immediate ongoing ASA use.There may be potential for a higher rebleeding rate due to the possible impairment of hemostasis caused by concurrent antiplatelet and anticoagulant therapy after an AMI.Additionally, AMI treatments such as clopidogrel   Data presented as n (%) unless specified otherwise may be associated with ulcer rebleeding (15).The current study suggests that in AMI patients, the endoscopic appearance of an ulcer is an important factor for physicians to consider when deciding whether to immediately continue ASA despite peptic ulcer bleeding.An additional predictor for continued ASA use was the need for AMI coronary interventions such as balloon angioplasty, coronary stent placement or coronary bypass surgery.ASA is recommended for secondary prevention in patients with AMI (1).Additionally, patients who have received coronary interventions also have an additional risk of future rethrombosis at the site of intervention (16)(17)(18).Therefore, it is currently recommended that patients who undergo coronary stenting should initially receive double antiplatelet therapy with ASA and clopidogrel for at least one month, depending on the stent type (16).Results of the current study support the high concern for rethrombosis in the absence of ASA therapy in patients who received coronary intervention.It is important for gastroenterologists involved in the care of PUD bleeding to be aware of this competing cardiac concern after coronary stenting.The optimal use and timing of ASA in AMI and PUD bleeding remains unknown.
An unpublished interim report (14) has suggested that continued ASA therapy in patients with cardiovascular disease and PUD bleeding may not increase rebleeding and may also decrease mortality.The need for ASA in AMI patients with PUD bleeding may also have more clinical significance because a recent study (19) suggested that concurrent PPI and clopidogrel (another recommended AMI antiplatelet agent) may increase recurrent hospitalizations for coronary events.Additionally, ASA plus PPI has been shown to be superior to clopidogrel in preventing rebleeding, but the study was limited to patients who already had confirmed ulcer healing after a period in which ASA was withheld (15).Therefore, the outcomes of immediate ASA use during an AMI and PUD bleeding remains to be clarified.Although the present study recorded overall rebleeding and mortality outcomes of 14% and 14%, respectively, it was not the primary aim of the study.Additionally, despite reviewing the records of two tertiary care centres, the present study was not sufficiently powered to compare rebleeding or mortality outcomes between patients who immediately continued ASA versus those who did not.A comparison of outcomes is also limited by the demonstrated selection bias between these patient groups.Finally, the rebleeding and mortality outcomes recorded in the present study represent the early 30-day outcomes that are generally reported in acute PUD bleeding studies (9,10).However, outcomes after 30 days may need to be assessed to determine the longer-term effects of temporary or prolonged ASA discontinuation.
The current study investigated the practices of two tertiary care centres with acute coronary care units in which both gastroenterologists and cardiologists are involved in the management of AMI patients with PUD bleeding.In addition, emergency endoscopy and percutaneous coronary therapy was available 24 h/day.Practices of ASA use may be different in centres where emergency facilities are not readily available to manage potentially acute complications related to withholding or continuing ASA.Additionally, the current study may not be generalizeable to cases of demand-ischemia precipitated by anemia from acute PUD bleeding.Myocardial infarctions secondary to demand-ischemia were not included in the present study because they may have different pathophysiology, outcomes and cardiac management strategies than patients with thrombosis-related myocardial infarctions.

CONCLUSION
Immediate continuation of ASA therapy in AMI patients with concurrent PUD bleeding was variable.However, the continuation of ASA therapy during PUD bleeding was more likely in patients with low-risk ulcer stigmata who received coronary intervention.Further studies are needed to guide physicians regarding the gastrointestinal risk of immediate ASA use during treatment of AMI accompanied by PUD bleeding.
This study was presented in full, or in part, as a poster at the Annual Meeting of the Canadian Association of Gastroenterology Canadian Digestive Diseases Week, February 16-20, 2007, Banff, Alberta; and The Annual Meeting of the American College of Gastroenterology, October 12-17, 2007, Philadelphia, Pennsylvania, USA.

TABLE 1 Baseline patient characteristics
None of the patients with low-risk ulcers underwent endoscopic therapy.Endoscopic therapy was performed in 40 of 43 patients (93%) with high-risk ulcers.Endoscopic therapy was not performed in three patients with adherent clots based on the gastroenterologist's discretion at the time of the endoscopy.The clinical patient outcomes are shown in Table