Screening carotid sonography before elective coronary artery bypass graft surgery: who needs it.

OBJECTIVE: The purpose of this study was to determine whether selection criteria for performing carotid sonographic screening before elective coronary artery bypass graft surgery can decrease the number of negative examinations without overlooking patients with significant carotid disease. SUBJECTS AND METHODS: A history of peripheral vascular disease, a prior cerebrovascular event, smoking, diabetes, hypertension, cervical carotid disease, left main coronary disease, and patient sex were criteria prospectively gathered for 295 consecutive patients undergoing screening carotid sonography before elective coronary artery bypass surgery. Logistic regression modeling was used to determine if any single criterion or combination of criteria could be applied to decrease the number of screening examinations without sacrificing detection of significant (> or = 50% cross-sectional narrowing) carotid stenosis. RESULTS: Smoking, diabetes, hypertension, a previous cerebrovascular event, peripheral vascular disease, left main coronary artery disease, and a history of cervical carotid disease were associated with significant carotid disease (chi-square test) in our subject population. Logistic modeling showed that the probability of detecting significant carotid disease increases 2.98 times for each additional selection criterion present. Possessing at least one selection criterion would still yield 100% examination sensitivity while increasing specificity to 30.0%. CONCLUSION: Selection criteria should be applied when choosing patients for carotid sonographic screening before elective coronary artery bypass surgery. This approach would decrease the number of noncontributory examinations but would have little effect on the detection of significant carotid stenosis in this target population.