The Use Of Antiplatelet Agents Does Not Impact the Hospital Course Of Patients Admitted With GI Bleed
Bijo K. John, M.D., Sushma Arramraju, M.D., Albert Shalomov, M.D., Moshe Rubin, M.D.
The use of antiplatelet agents (APA) has been shown to reduce morbidity and mortality in patients with cardiovascular risk factors. They also increase the risk of GI bleeding by a factor of 2 to 3. In a recent French study, 30% of patients admitted for GI bleeding were on APA. We sought to evaluate the prevalence of APA use in patients admitted with GI bleeding to our hospital and to determine whether the use of these agents leads to an increase in morbidity and mortality. Methods: 104 sequential patients with a diagnosis of GI bleeding admitted to New York Hospital Queens, were retrospectively studied. Demographics, clinical presentation, treatment, endoscopic findings and outcomes were reviewed. Patients with a known bleeding disorder, end stage liver disease, on warfarin or thrombocytopenia were excluded. Results: There were 71 males and 33 females, mean age of 71. Caucasians and Asians comprised 52 & 31% respectively. 30 of 104 (29%) patients were on long-term aspirin and/or clopidogrel on admission. Concomitant use of NSAIDS increased the number to 35 (34%). 56 of 104 (54%) had a history of prior GI disease, 22 of which were admitted for recurrent bleeding. Clinical presentation was hematemesis, melena and bright red blood in 38, 20 and 38% respectively. During the first 24 hours of admission, an average of 3 units of blood were transfused. An identifiable source was found in nearly 80% of cases. The average length of stay was 5 days. There was no difference between patients using APA and those not on APA with regard to the admission hemoglobin, age, presentation, source of bleed, number of units transfused emergently, ICU stay and overall length of stay. APA use did not worsen the outcome in patients with a prior GI history compared to those not on APA. There was, however, a significant difference in the presence of hemodynamic compromise on initial presentation, with a higher proportion of APA users being orthostatic (51.4% versus 26% in non users, with p=0.02, by Fishers exact). Conclusion: The use of APA in patients admitted with GI bleeding did not significantly alter the hospital course, severity, findings or outcome in our study population. These agents did, however, increase the risk of hemodynamic compromise on admission, suggesting a more rapid bleed acutely. Furthermore, the majority of patients admitted with GI bleeding were not on APA despite their increasing use in this population to prevent cardiovascular events. This study suggests that factors other than the use of APA contribute to the risk of major GI bleeding.