Evidence to Support Screening for Asymptomatic CAS is Lacking

Data supporting routine screening for asymptomatic carotid artery stenosis (CAS) are lacking, according a recent study [Ann Intern Med. 2014; DOI:10.7326/M14-0530]. “The evidence [showed] that there may be greater harm than benefit from screening the general population for asymptomatic CAS,” said Daniel E. Jonas, MD, MPH, assistant professor of medicine, University of North Carolina, Chapel Hill, in an interview with First Report Managed Care.

Ischemic stroke accounts for almost 90% of all strokes in the United States, and CAS is believed to cause approximately 10% of ischemic strokes. Because many surgeries or interventions for asymptomatic CAS continue to be performed despite the United States Preventive Services Task Force (USPSTF) recommending otherwise in 2007, Dr. Jonas and his colleagues conducted a systematic review and meta-analysis to evaluate whether screening asymptomatic adults for CAS reduces the risk for ipsilateral stroke.

A total of 78 articles reporting on 58 studies published through September 2013 were culled from sources including MEDLINE, the Cochrane Library, and Embase. These included randomized, controlled trials of screening for CAS; randomized, controlled trials and systematic reviews of treatment effectiveness; multi-institution trials or cohort studies that reported harms; and studies that attempted to externally validate risk-stratification tools.

The researchers conducted a meta-analysis of randomized, controlled trials that compared carotid endarterectomy (CEA) with medical therapy for relevant outcomes and used DerSimonian-Laird random-effects models to estimate pooled effects and calculated risk differences between CEA and medical therapy to show the absolute differences between groups. They also conducted meta-analyses of cohort studies that reported perioperative (30-day) stroke or death rates.

The study’s results showed sensitivity and specificity rates of duplex ultrasonography for detecting stenosis of ≥50% were 98% and 88%, respectively, and for detecting stenosis of ≥70%, were 90% and 94%, respectively. These data led the USPSTF to estimate the sensitivity and specificity for detecting stenosis of ≥60% to be 94% and 92%, respectively. “Using ultrasound to screen for asymptomatic CAS in the general population would yield many false-positive results, leading to unnecessary interventions and harm,” Dr. Jonas said.

The researchers found an absolute reduction rate of 5.5% for nonperioperative strokes over approximately 5 years for patients who underwent CEA compared with those who had medical therapy. They also found that 1.9% more patients treated with CEA had perioperative (30-day) stroke or death than those in medical therapy groups. Among studies using Medicare claims data and medical records, the rate of perioperative death or stroke was 3.3% for patients who had CEA, and among all trials that included a CEA group regardless of comparator, the rate was 2.4%. Other harms reported as a result of interventions included myocardial infarction, nerve injury, and hematoma.

The researchers indicated limitations of the study, including that much of the data were from the 1990s and thus do reflect changes in technology and medical therapy.

The researchers concluded that current evidence did not establish incremental overall benefit of CEA beyond current standard medical therapy, mainly because the medical therapy for trials was poorly defined, varied, and, in some cases, lacked treatments that are standard today. “The potential for overall population benefit of screening for and treating asymptomatic CAS is limited by low prevalence and harms,” Dr. Jonas said.—Mary Mihalovic