The primary prevention of cardiovascular diseases (CVD) is based on the individual risk assessment by using Framingham and similar algorithms derived mainly from US or European male populations. The purpose of this prospective study was to compare the appropriateness of three different common guidelines for cardiovascular risk stratification when using subclinical-atherosclerosis as a surrogate of CV risk. The study population included 560 women (W), aged from 20 to 65 years, non diabetic and without established CVD, who were consecutively were referred to our Lipid Clinic for therapeutic advices. CV-risk factors were systematically screened for each subject, including medical history, physical examination and clinical chemistry. Their estimated 10-year CV-risk assessment was obtained by using the guidelines of Swiss-AGLA (AGLA), NCEP-ATP-3 (ATP-3) and 3th JESESC (ESC) guidelines (GL). B-mode ultrasounds on carotid and femoral arteries was performed to detect atherosclerotic plaques (focal thickening of intimamedia >1.2 mm). W with plaques on >2 carotid and/or femoral sites were considered as high CV risk (HR). The proportions of W stratified as high 10-year CV-risk by the GL were the following: 5% (AGLA), 2% (ESC) and 24% (ESC). The comparative values (area under the ROC-curve, 95% IC) of GL to detect the 160 HR-W with subclinical atherosclerosis (29%) was significantly higher for AGLA-GL (0.7869, 0.75-0.82) than for ATP-3-GL (0.75, 0.71-0.79) or than for ESC-GL (0.73, 0.68-0.77). In conclusion, the results suggest that current AGLA-GL recommended by the Swiss Society of Cardiology are the most appropriate to identify HR-W requiring more intensive therapy for the primary prevention of CVD.