The Impact of Aspirin, Non Steroidal Anti-Inflammatory Drugs and Warfarin on Acute Lower Gastrointestinal Bleeding: Ten Year Study From a Dedicated Bleeding Unit

Abstract

Background: Aspirin, nonsteroidal anti-inflammatory drugs (NSAIDS) and warfarin have been associated with bleeding from both upper and lower gastrointestinal tract. However, there are no long-term prospective community-based studies addressing what impact these medications have on the natural history of acute lower gastrointestinal bleeding (ALGIB). Aims: To assess the incidence, severity, complications and mortality of ALGIB in patients taking aspirin, NSAIDS and warfarin. Methods: All patients presenting with ALGIB to a dedicated bleeding unit between October 1991 to October 2001 were included. Data were collected prospectively on aspirin, NSAIDS and warfarin usage and stored on a Microsoft Access database for later analysis. Severity of bleeding on admission was categorised: Significant bleeding was defined as one or more of the following features collapse, shock, haemodynamic disturbance or haemoglobin<10 g/dl; Trivial bleeding defined as no haemodynamic disturbance or acute drop in haemoglobin < 10 g/dl. Rebleeding was defined as evidence of further fresh rectal bleeding and a drop in haemoglobin > 2 g/dl after initial stabilisation of bleeding for a period of 24 hours. Results: 1379 patients (47% male) with mean age of 65.5 ± 0.5 years had 1472 episodes of confirmed ALGIB. Regular intake of aspirin, NSAIDS or warfarin on admission was identified in 665 patients (45%). Although the number admitted with confirmed ALGIB doubled, this did not correlate with drug usage. Significant bleeding occured in 57% patients on the above drugs versus 49% of those who were not (p<0.002). Patients on the above medications had increased risk of rebleeding compared with those not taking any (18.4% v 13.1%) p< 0.01. The mean transfusion requirement was higher in the medication than the non medication group (2.5 units v 1.9 units). Overall 30 day mortality was greater in the medication group than the non medication group (7.5% v 4.8%) p<0.05. There was no significant difference in the surgical referral rates between two groups 11% v 12%. Conclusions: The use of aspirin, NSAIDS and warfarin contributes to an 1) Increase in the number of patients admitted with significant bleeds. 2) Increase in Blood transfusion requirements. 3) Increased rebleeding and mortality rates.