In the case of non-ST-segment elevation acute coronary syndromes (NSTE-ACSs), the acute use of certain antiplatelet agents is complicated by concerns about perioperative bleeding risks in patients requiring coronary artery bypass grafting (CABG) during the index hospitalization. As a result, clinicians often withhold potentially useful agents, such as clopidogrel, before determining patients' coronary anatomy. An accurate predictive model could allow for a better balance of this safety concern with the demonstrated benefits of agents such as clopidogrel. To create an accurate decision-making tool that would assess, at hospital presentation, the need for CABG in patients with NSTE-ACSs, we studied 61,974 high-risk patients with NSTE-ACS admitted to 311 CABG-capable hospitals participating in Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines (CRUSADE) from 2001 to 2003. A total of 8,395 patients (14%) underwent CABG during their initial hospital stay. A multivariate model was developed and identified 13 presenting clinical characteristics significantly associated with the likelihood of CABG (previous CABG, male gender, previous heart failure, diabetes, hyperlipidemia, renal insufficiency, ST depression and transient ST elevation, age > or = 75 years, previous percutaneous coronary intervention, family history of coronary artery disease, hypertension, trends in CABG rates, and previous stroke). This model had only modest predictive accuracy and calibration (c-index = 0.67). In conclusion, although certain presenting clinical features are associated with an increased likelihood of CABG in patients with NSTE-ACSs during the index hospitalization, it remains difficult to reliably identify, before diagnostic angiography, those who will subsequently undergo surgical revascularization.
View details for DOI 10.1016/j.amjcard.2006.03.040
View details for Web of Science ID 000240324300012
View details for PubMedID 16923449