| Popis: |
Structured Abstract Background: Stroke is currently the 4th leading cause of death in the United States. After a systematic review in 1996, the USPSTF concluded that the evidence was insufficient to recommend for or against screening of asymptomatic people for carotid artery stenosis (CAS) using auscultation for carotid bruit during physical exam. Using the reference standard of 70-99% stenosis on carotid angiogram, they determined the sensitivity and specificity of a carotid bruit to be 63-76% and 61-76% respectively for identifying CAS of this degree. Purpose: To examine the evidence since the 1996 USPSTF review on accuracy of auscultation for carotid bruit as a screening test for CAS and subsequent stroke outcomes to provide accuracy data and help afford better guidance on the utility of this means of screening for CAS in asymptomatic patients in the general population. Data Sources: MEDLINE search (January 1996-present), recent randomized controlled trials (RCTs), cohort studies, and diagnostic accuracy studies, reference lists of retrieved articles. Study Selection: English language studies were selected to answer the questions: (KQ1) What is the accuracy of auscultation for carotid bruit to predict fatal or nonfatal ischemic stroke or TIA? (KQ2) What is the accuracy of auscultation for carotid bruit to detect potentially clinically important CAS (60% to 99%)? Study types were RCTs or cohort studies of asymptomatic patients at least 19 years of age, from which data on cerebrovascular outcomes could be extracted. Also they could be diagnostic accuracy studies comparing auscultation for carotid bruit to the gold standard of angiography in asymptomatic patients at least 19 years of age. Data Synthesis: There were no RCTs of screening for CAS by auscultation for carotid bruit. Two included cohort studies showed the sensitivity and specificity of carotid bruit to identify subsequent stroke in asymptomatic patients to be 9.1-13.4% and 93.75-96.7% respectively with false-negative rates of 86.6-90.9%. No studies met inclusion for KQ2 due to improper gold standard, but 2 cohort studies and a systematic review using the reference of duplex ultrasound, the sensitivity and specificity for identifying CAS ≥ 60% were 56.25-57.5% and 80-98%. Limitations: There were no RCTs of screening for CAS to answer KQ1 and the included cohort studies were conducted in selected populations with diabetes or isolated systolic hypertension. These only focused on stroke without assessing the outcome of TIA. No accuracy studies met inclusion for KQ2 that used the gold standard of angiography. The quality of included studies was only fair on average. Conclusions: The sensitivity and specificity of auscultation for carotid bruit are low and exhibit poor accuracy for a screening test. Such low values would result in high false-positive and false-negative rates, both for identifying CAS as well as subsequent stroke in the asymptomatic population. Inadequate certainty of the benefit of screening for CAS in the general asymptomatic population, lack of a reliable means to delineate a more distinct group to screen in, and the poor accuracy of auscultation for carotid bruit in the primary care setting makes it difficult to consider widespread use of this screening test in the general population. Until we know with reasonable certainty that such screening leads to reduced strokes and better understand how to delineate the best group of people to screen in, I do not recommend the use of auscultation for carotid bruit in the primary care setting as a screening test for CAS and future cerebrovascular outcomes in asymptomatic patients. |
