Introduction Antiplatelet and antithrombotic therapy for acute coronary syndromes may increase the risk of gastrointestinal and other haemorrhagic events. The aim of this study was to compare the frequency of bleeding events, and consequent morbidity and mortality among patients treated for acute coronary syndromes and a control group of hospital inpatients.

Methods Design: Observational cross-sectional study performed in Ealing Hospital NHS Trust, UK, used from 1 December 2004 to 28 February 2005. The treatment group consisted of 111 inpatients diagnosed with acute coronary syndrome and treated with standard anti-thrombotic medication. A control group of 106 patients was admitted during the same period. The main outcome measure was the incidence of bleeding events within 30 days. Haemoglobin concentration was monitored.

Results A significant increase in haemorrhage was found in the group treated for acute coronary syndrome (OR 10.6, 95% CI 4.3 to 26.4), predominantly upper GI haemorrhage (OR 9.3, 95% CI 2.7 to 32.0). Bleeding events occurred in 42 patients in the treatment group (∼38%) and 6 patients in the control group. The treatment group were also more likely to develop anaemia as inpatients (OR 4.93, 95% CI 2.65 to 9.2). Examination of serial haemoglobins indicated that 56 patients became newly anaemic and that there was an overall drop in haemoglobin of over 12.3% (p=0.0005). Eleven patients died with major bleeds, contributing as a major cause of mortality in 3 patients (2 gastrointestinal and 1 intracranial), although this difference was not significant.

Conclusion We have demonstrated a greater incidence of GI and other system haemorrhage than that reported in many of the major trials for treatment of acute coronary syndromes. This calls in to question the current recommendations for the use of proton pump inhibitors in such patients, a recent matter of much controversy due to the interactions of these drugs with clopidogrel.