A nomogram and a carotid risk score comprising seven predictors achieved an AUC of 0.785 and classification accuracy of 0.797 for predicting moderate or high carotid atherosclerosis.
Observational (n=956)
Can a clinical nomogram and risk score accurately predict moderate or high carotid atherosclerosis in asymptomatic elderly individuals?
A simple clinical risk score can help identify asymptomatic elderly individuals with a high probability of significant carotid atherosclerosis who may benefit from screening ultrasound.
Effect estimate: AUC 0.785
Carotid atherosclerosis is associated with cardiovascular and cerebrovascular events. We explored an appropriate method for selecting participants without ischemic cerebrovascular disease but with various comorbidities eligible for a carotid ultrasound. This was a retrospective subgroup analysis of the carotid plaque burden from a previous study involving a vascular and cognitive survey of 956 elderly recycling volunteers (778 women and 178 men; mean age: 70.8 years). We used carotid ultrasound to detect the carotid plaque and computed the carotid plaque score (CPS). A moderate or high degree of carotid atherosclerosis (MHCA) was defined as CPS > 5 and was observed in 22% of the participants. The CPS had positive linear correlations with age, systolic blood pressure, and fasting glucose. We stratified the participants into four age groups: 60–69, 70–74, 75–79, and ≥80 years. Multivariable analysis revealed that significant predictors for MHCA were age, male sex, hypertension, diabetes mellitus, hyperlipidemia, coronary artery disease, and a nonvegetarian diet. Coronary artery disease and advanced age were the two strongest predictors. We chose the aforementioned seven significant predictors to establish a nomogram for MHCA prediction. The area under the receiver operating characteristic curve in internal validation with 10-fold cross-validation and the classification accuracy of the nomogram were 0.785 and 0.797, respectively. We presumed people who have a ≥50% probability of MHCA warranted a carotid ultrasound. A flowchart table derived from the nomogram addressing the probabilities of all models of combinations of comorbidities was established to identify participants who had a probability of MHCA ≥ 50% (corresponding to a total nomogram score of ≥15 points). We further established a carotid risk score range from 0 to 17 comprising the seven predictors. A carotid risk score ≥ 7 was the most optimal cutoff value associated with a probability of MHCA ≥ 50%. Both total nomogram score ≥ 15 points and carotid risk score ≥ 7 can help in the rapid identification of individuals without stroke but who have a ≥50% probability of MHCA—these individuals should schedule a carotid ultrasound.
Hsiao et al. (Mon,) conducted a observational in Carotid atherosclerosis (n=956). Nomogram and Carotid Risk Score was evaluated on Moderate or high degree of carotid atherosclerosis (MHCA) prediction (AUC 0.785). A nomogram and a carotid risk score comprising seven predictors achieved an AUC of 0.785 and classification accuracy of 0.797 for predicting moderate or high carotid atherosclerosis.