Screening for Asymptomatic Carotid Artery Stenosis: A Systematic Review and Meta-analysis for the US Preventive Services Task Force

Background: Approximately 10% of ischemic strokes are caused by carotid artery stenosis (CAS). Estimated prevalence of asymptomatic CAS is 1%. Purpose: To evaluate evidence on screening and treating asymptomatic adults for CAS. Data Sources: MEDLINE, the Cochrane Library, EMBASE, and trial registries through September 2013; MEDLINE through March 2014 for trials. Study Selection: Good-or fair-quality trials of screening, carotid endarterectomy (CEA), or stenting compared with medical therapy or of intensification of medical therapy; systematic reviews; multi-institution studies reporting harms; and externally validated risk-stratification tools. Data Extraction: Dual extraction and quality assessment. Data Synthesis: No trials compared screening with no screening or stenting with medical therapy or assessed intensification of medical therapy, and no externally validated, reliable risk-stratification tools were found. Given the specificity of ultrasonography (range, 88% to 94% for CAS >= 50% to >= 70%), its use in low-prevalence populations would yield many false-positive results. Absolute reduction of non-perioperative strokes was 5.5% (95% CI, 3.9% to 7.0%; 3 trials; 5223 participants) over approximately 5 years for CEA compared with medical therapy. The 30-day rates of stroke or death after CEA in trials and cohort studies were 2.4% (CI, 1.7% to 3.1%; 6 trials; 3435 participants) and 3.3% (CI, 2.7% to 3.9%; 7 studies; 17 474 participants), respectively. Other harms of interventions included myocardial infarction, nerve injury, and hematoma. Limitations: Trials may have overestimated benefits and used highly selected surgeons. Medical therapy used in trials was outdated, and stroke rates have declined in recent decades. Harms may have been underreported. Conclusion: Current evidence does not establish incremental overall benefit of CEA, stenting, or intensification of medical therapy. Potential for overall benefit is limited by low prevalence and harms.