Asymptomatic carotid arterial stenosis - population based screening

Abstract

Screening for asymptomatic carotid artery stenosis in the general population is discussed in many countries because of the benefits of carotid endarterectomy in the three trials. Many factors influence the cost-effectiveness of screening. These factors are the prevalence of carotid stenosis, the costs of the screening tool, the sensitivity and the specificity of the screening tool and the benefits of the treatments. We found that prevalence-numbers are not easy to extract from literature, and after performing an IPD-meta-analysis, we found that severe asymptomatic carotid artery stenosis is uncommon in the general population. This prevalence ranges from zero percent to 1.7%. Another factor that influences whether screening is worthwhile is the screening tool. Screening performed by DUS in the general population followed by endarterectomy when severe stenosis was found is worthwhile in 65-year old men, when the prevalence is at least 3%. In 75-year-old women screening by Duplex followed by endarterectomy when a severe stenosis was found is was worthwhile when the prevalence of severe asymptomatic carotid artery stenosis was at least 5%. When screening is followed by cardiovascular risk factor management when a moderate stenosis was found and followed by endarterectomy when a severe stenosis was found, screening is worthwhile in men aged 45 when the prevalence of severe asymptomatic carotid artery stenosis is at least 0.01%. In 50-year old women screening followed by cardiovascular risk factor management when moderate stenosis was found was worthwhile, when the prevalence was at least 0.01%. Additionally, we found that endarterectomy decreased the QALY gain and thus increased the incremental cost-effectiveness ratio. Finally we intended to identify individuals for whom one-time screening for ACAS might be cost-effective. We developed a prediction rule for the presence of severe (>70%) stenosis and moderate (>50%) stenosis. We used individual participant data from four population-based studies (MDCS, Tromso, CAPS and CHS). The presence of moderate stenosis was related to age, sex, HDL cholesterol, LDL cholesterol, systolic and diastolic blood pressure, body mass index, waist to hip ratio and smoking. The presence of severe stenosis was related to age, sex, HDL cholesterol, LDL cholesterol, systolic and diastolic blood pressure and smoking. Both models discriminated and calibrated well. We were able to identify subgroups for which screening followed by endarterectomy was worthwhile. Population based screening for asymptomatic carotis stenosis is not cost-effective considering detection of severe stenosis followed by endarterectomy only. We can however identify people at high risk of having severe stenosis for which screening might be cost-effective. In further research our prediction rule should be validated and evaluated in a cost-effectiveness-analysis. When also cardiovascular risk factor management is offered, carotid endarterectomy is no longer appropriate, i.e., carotid endarterectomy made the cost-effectiveness even worse. For now, population based screening for carotid artery stenosis should not be implemented.