Risk of myocardial infarction and vascular death after transient ischemic attack and ischemic stroke: a systematic review and meta-analysis.

BACKGROUND: Whether stroke patients should be investigated for asymptomatic coronary artery disease remains matter of debate. Absolute risks of myocardial infarction (MI) and vascular death after a stroke have not been accurately assessed. We performed a systematic review and a meta-analysis to determine the risk of MI and nonstroke vascular death after transient ischemic attack (TIA) and ischemic stroke. Cohort studies of TIA or ischemic stroke patients were included if they were published between 1980 and March 2005, reported risk of MI and nonstroke vascular death, enrolled >100 patients, and had at least 1 year of follow-up. We included 39 studies in a total of 65,996 patients with mean follow-up of 3.5 years. Two reviewers independently carried out data extraction using a standardized form. Absolute annual risks were estimated through weighted meta-regressions with a random effect. To test the predictions of expected event rates derived from our analysis, we used individual patient data. SUMMARY OF REVIEW: The annual risks were 2.1% (CI 95%: 1.9 to 2.4) for nonstroke vascular death, 2.2% (1.7 to 2.7) for total MI, 0.9% (0.7 to 1.2) for nonfatal MI and 1.1% (0.8 to 1.5) for fatal MI. The time course of risk was linear. Estimated risks fitted well with observed risks at the individual level. There was no heterogeneity in the absolute risks according to baseline study characteristics. CONCLUSIONS: Patients with TIA or stroke have a relatively high risk of MI and nonstroke vascular death. Additional research is needed to identify the determinants of coronary artery disease in stroke patients.