Predictors of bleeding risk and long-term mortality in patients with acute coronary syndromes

Objective: The aim of this review was to collate the published evidence on independent predictors of bleeding and late mortality in acute coronary syndrome (ACS) patients, to compare the two sets of risk factors, and to investigate whether bleeding is reported as a predictor of late mortality. Research design and methods: Computerized searches, covering the period from 1999 to July 2004, were performed on MEDLINE, EMBASE, and the Cochrane Library database. Studies were eligible for inclusion in the review if they related to patients with ACS and included an assessment of risk factors for bleeding, mortality, or both. Studies that did not meet these criteria were excluded. Main outcome measures: A total of 937 studies were retrieved, of which 912 were excluded from the review because they did not meet the defined criteria. Results: The available evidence suggests that any impact of bleeding on mortality is confined to the short term. In studies that showed a significant association between bleeding and risk of mortality, this was always related to in-hospital or 30-day mortality. By contrast, studies of long-term mortality consistently showed that bleeding was not an independent predictor. Furthermore, follow-up studies showed that adverse outcomes in hospital and within the first month were not related to 1-year mortality. This may reflect the beneficial impact of anticoagulant therapy on subsequent cardiovascular risk in patients with ACS, which outweighs any short-term detrimental effect of bleeding. Conclusions: The available evidence suggests that any impact of bleeding on mortality in ACS patients appears to be confined to the short term, and long-term outcomes do not reflect the impact of in-hospital bleeding.