Skip to main content
PMC home page
Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease logoLink to Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease
. 2020 Apr 20;9(8):e014766. doi: 10.1161/JAHA.119.014766

Validation of Risk Prediction Models to Detect Asymptomatic Carotid Stenosis

Michiel H F Poorthuis 1,2,3,†, Alison Halliday 4,†, M Sofia Massa 1,†, Paul Sherliker 1,2,†, Rachel Clack 1, Dylan R Morris 1,2, Robert Clarke 1, Gert J de Borst 3, Richard Bulbulia 1,2,‡,✉, Sarah Lewington 1,2,‡
PMCID: PMC7428515  PMID: 32310014

Abstract

Background

Significant asymptomatic carotid stenosis (ACS) is associated with higher risk of strokes. While the prevalence of moderate and severe ACS is low in the general population, prediction models may allow identification of individuals at increased risk, thereby enabling targeted screening. We identified established prediction models for ACS and externally validated them in a large screening population.

Methods and Results

Prediction models for prevalent cases with ≥50% ACS were identified in a systematic review (975 studies reviewed and 6 prediction models identified [3 for moderate and 3 for severe ACS]) and then validated using data from 596 469 individuals who attended commercial vascular screening clinics in the United States and United Kingdom. We assessed discrimination and calibration. In the validation cohort, 11 178 (1.87%) participants had ≥50% ACS and 2033 (0.34%) had ≥70% ACS. The best model included age, sex, smoking, hypertension, hypercholesterolemia, diabetes mellitus, vascular and cerebrovascular disease, measured blood pressure, and blood lipids. The area under the receiver operating characteristic curve for this model was 0.75 (95% CI, 0.74–0.75) for ≥50% ACS and 0.78 (95% CI, 0.77–0.79) for ≥70% ACS. The prevalence of ≥50% ACS in the highest decile of risk was 6.51%, and 1.42% for ≥70% ACS. Targeted screening of the 10% highest risk identified 35% of cases with ≥50% ACS and 42% of cases with ≥70% ACS.

Conclusions

Individuals at high risk of significant ACS can be selected reliably using a prediction model. The best‐performing prediction models identified over one third of all cases by targeted screening of individuals in the highest decile of risk only.

Keywords: atherosclerosis, carotid artery stenosis, external validation, ischemic stroke, prevention, risk prediction model, targeted screening

Subject Categories: Ischemic Stroke, Risk Factors, Primary Prevention, Stenosis, Atherosclerosis

Clinical Perspective

What Is New?

  • Established risk prediction models to detect cases at high risk of asymptomatic carotid stenosis were validated in a contemporary screening population in the United States and United Kingdom.

What Are the Clinical Implications?

  • Risk prediction models can be used for targeted screening for asymptomatic carotid stenosis, and cardiovascular risk management can be initiated or intensified to prevent complications of asymptomatic carotid stenosis.

Nonstandard Abbreviations and Acronyms.

ACS asymptomatic carotid stenosis

DBP diastolic blood pressure

HDL‐C high‐density lipoprotein cholesterol

LDL‐C low‐density lipoprotein cholesterol

SBP systolic blood pressure

TC total cholesterol

TIA transient ischemic attack

Introduction

Transient ischemic attack (TIA) or ischemic stroke is the first presentation of cardiovascular disease in about 25% of the cases,1, 2 and 15% to 20% of ischemic stroke cases are associated with extracranial carotid artery stenosis.3, 4, 5 Carotid stenosis is also a predictor for coronary events and vascular death.6 The prevalence of moderate (≥50%) and severe (≥70%) asymptomatic carotid stenosis (ACS) in the general population has been estimated to be 2.0% and 0.5%, respectively.7

Because of this low overall prevalence, population‐level screening for ACS with duplex ultrasound is not recommended in current guidelines.8, 9, 10, 11 However, targeted screening of high‐risk individuals might be worthwhile,11 and risk stratification tools or prediction models have been developed to provide individualized risk estimation for ACS. Before recommending targeted screening, risk prediction tools should be assessed for discrimination, calibration, and likely ability to detect false‐positive and false‐negative cases in an independent external population. We conducted a systematic review of published studies of prediction models for ACS and then externally validated these models in a large contemporary population of screenees in the United States and United Kingdom.

Methods

Systematic review according to a predefined protocol to identify established risk prediction models. This protocol has been registered in an international registry for systematic reviews (PROSPERO [International Prospective Register of Systematic Reviews]): CRD42019108136. The study adhered to the Preferred Reporting Items for Systematic reviews and Meta‐Analyses (PRISMA) recommendations (Table S1) and the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS).12, 13

Data Sharing

Data from large population‐based studies conducted by the Nuffield Department of Population Health can be shared with bona fide researchers on application to the principal investigators of this study. Details of the departmental data access policy can be found at https://www.ndph.ox.ac.uk/data-access.

Search Strategy and Eligibility Criteria

We used comprehensive electronic strategies and incorporated a validated research search filter to search Medline (via PubMed interface) and EMBASE (via OVID EMBASE interface) on March 1, 2019, for studies reporting on development and validation of prediction models for risk of significant ACS in general or screened populations (Data S1).14 We included studies that (1) addressed development and/or validation of diagnostic prediction models to detect ACS of 50% or greater, (2) assessed prediction models in both general and high‐risk populations but not in diseased populations at higher risk of ACS, (3) involved a cross‐sectional study design, and (4) were published in peer‐reviewed journals without any language restrictions.

Screening Process and Data Extraction

Two authors (M.H.F.P. and M.S.M.) independently screened all titles and abstracts of the retrieved references and subsequently independently reviewed full‐text copies for final inclusion in this study. We performed backward citation searching using the bibliographies of included studies.

Two authors (M.H.F.P. and M.S.M.) independently extracted the following data from the included studies reporting the development of a prediction model, based on the CHARMS checklist: source of data, setting study, geographic area (country and continent), study years, sample size, modeling method (eg, logistic model), number of participants with missing data, handling of missing data, investigation of satisfaction of modeling assumptions, selection methods for predictor selection, shrinkage of predictor weights, number of outcome events, number of participants, degree of stenosis, number and type of predictors (diagnostic variables) used in the final model, number of outcome events per variable, presentation of model, and model performance (calibration and validation). In studies that reported internal validation of prediction models, we extracted the following additional data: method of internal validation (eg, cross‐validation, bootstrap), and whether the model was adjusted or updated after internal validation. In studies reporting external validation of a prediction model, we extracted the following additional data: type of external validation (eg, geographical and/or temporal distinct population), whether authors of the external validation also developed the original model, and performance of the model before or after model recalibration.

Critical Appraisal

Prediction modeling studies were assessed for risk of bias and applicability using the Prediction model Risk Of Bias Assessment Tool (PROBAST).15 The assessment of risk of bias involved 4 domains: participants, predictors, outcome, and analysis. Risk of bias was judged as low, high, or uncertain for each domain. The assessment of applicability involved 3 domains: participants, predictors, and outcome. Applicability was judged as low, high, or uncertain for each domain. Each distinct model included in the article was evaluated separately.16

External Validation Cohort

A cohort of 0.6 million self‐referred and self‐funded individuals who attended commercial vascular screening clinics between 2008 and 2013 in the United States and the United Kingdom was used for external validation. All individuals completed a standardized questionnaire including questions about their age; sex; height and weight; history of vascular disease (peripheral arterial disease, TIA, stroke, coronary artery disease, and congestive heart failure); history of hypertension; history of diabetes mellitus; smoking history; and use of antiplatelet, antihypertensive, and lipid‐lowering medication. Standard blood pressure cuffs and sphygmomanometers were used, with systolic pressure measured using a Doppler probe, and peripheral arterial disease was assessed with ankle‐brachial pressure index assessment.

Most participants underwent carotid duplex screening, conducted by trained staff using dedicated vascular ultrasound instruments (GE LOGIQ e). The highest peak systolic velocity and end‐diastolic velocity of both the common carotid arteries and the internal carotid arteries were measured.

A blood sample was collected from a subset of participants for selected plasma biochemical measurements using point‐of‐care testing methods (Alere Cholestech LDX System, Alere Inc, Waltham, MA). Plasma levels of total cholesterol, high‐density lipoprotein‐cholesterol, and triglycerides were measured by enzymatic methods. Low‐density lipoprotein‐cholesterol was estimated using the Friedewald formula (low‐density lipoprotein=total cholesterol−high‐density lipoprotein−triglycerides / 5).

Predicted Outcomes

We externally validated the prediction models for both moderate or severe ACS:

  1. Moderate or severe ACS; estimated stenosis of ≥50% (on the basis of peak systolic velocity ≥125 cm/s at either side or 0 cm/s for occluded arteries); and

  2. Severe ACS, estimated stenosis of ≥70% (on the basis of peak systolic velocity ≥230 cm/s at either side or 0 cm/s for occluded arteries).

Statistical Analysis (External Validation)

Selected characteristics of the external validation cohort were summarized using standard methods. We used the same external validation population for all external validation analyses to enable comparisons between different prediction models. Participants who provided a blood sample and had a duplex ultrasound performed were included in analyses. For most predictors, the percentage of participants with missing data was <12%, except for measured diastolic blood pressure (31.8%) (Table S2). Missing data were imputed using chained equations and we created 20 imputed data sets with 200 iterations.17 Total cholesterol/high‐density lipoprotein cholesterol ratio was calculated before imputation.18 Postimputation rounding was applied for limited‐range variables (systolic blood pressure, diastolic blood pressure, total cholesterol/high‐density lipoprotein cholesterol ratio, high‐density lipoprotein‐cholesterol, low‐density lipoprotein‐cholesterol, and height), if needed.19

The regression formula reported for each model was applied to the external validation cohort to calculate the probability of ≥50% and ≥70% ACS per participant. These individual probabilities were used for assessing the predictive performance. We contacted authors to provide the regression formula if it was not reported. If the authors did not report or could not provide the regression formula, we calculated a sum score (total points) for each participant by summing the scores assigned to each predictor in the original reports (referred to as a “score chart”). We used the sum score to assess the predictive performance.

We examined the performance of discrimination and calibration in the different prediction models. Discrimination is the ability of the prediction model to distinguish between participants with and without the disease outcomes, assessed using the area under the receiver operating characteristic (AUROC) curve. AUROC curve values were calculated per imputed data set and results were subsequently pooled using Rubin's rules.20, 21

Calibration is the agreement between predicted and observed risk and was assessed with calibration plots. For the models that provided the regression formula, we estimated the mean probability per participant across the 20 imputed data sets, and subsequently we split the predicted risks in deciles. We then calculated mean predicted and observed probability with corresponding 95% CIs per decile. In contrast, for the models that did not provide the regression formula, we used the predicted probability per sum score as reported in the original reports, and we calculated the observed probability with corresponding 95% CI in the validation cohort.

Differences between the prevalence of the predicted outcome in the development cohorts and the validation cohort are known to influence calibration. For this reason, we recalibrated the prediction models to the prevalence of the predicted outcome in the validation cohort by reestimating the intercept.22 We fitted a logistic model with a fixed calibration slope and the intercept as the only free parameter.22

STATA version 15.1 was used for all statistical analyses, and R version 3.5.1 was used for constructing the figures.

Clinical Application

Clinical application of the prediction model with the best discrimination was assessed using 2 approaches. The first approach assessed targeted screening of the 10% and 20% cases at highest predicted risk of having significant ACS. For this, we calculated test characteristics for the highest decile and the highest 2 deciles of predicted risk. The second approach assessed targeted screening with a fixed level of sensitivity. For this, test characteristics were calculated for 2 levels of sensitivity (closest to sensitivity 80% and 90%).

Sensitivity Analyses

We performed additional external validation of the prediction models: (1) in complete cases, (2) participants without a history of prior TIA or stroke using imputed data sets, and (3) participants without a history of prior cardiovascular disease (ie, stroke, TIA, myocardial infarction, and peripheral arterial disease) using imputed data sets.

Ethical Approval

The University of Oxford Medical Sciences Inter‐Divisional Research Ethics Committee approved the study. All individuals provided written consent for the data collected at the screening visit to be used for research purposes.

Role of the Funding Source

The study funders had no role in study design, data collection, analysis, or interpretation, drafting the report. The corresponding author had full access to all data in the study and had final responsibility for the decision to publish the report.

Results

We screened 923 unique reports identified by literature searching, assessed the full texts of 102 reports for eligibility, and included 5 studies (Figure 1 and Table S3). Four studies involved model development studies, of which 1 performed additional external validation of an existing prediction model.23, 24, 25, 26 One study was an external validation study.27 Overall, 6 prediction models for the prevalence of significant ACS were developed.23, 24, 25, 26 Characteristics of model development are provided in Table 1 and Table S4.

Figure 1.

Figure 1

Open in a new tab

Flowchart of literature review to identify the included studies.

Table 1.

Selected Characteristics of Studies Assessing Different Risk Prediction Models for Significant ACS

Predicted Outcomes Data Sources Calendar Year of Recruitment No. of Cases/Participants in Derivation Cohort Number of Included Predictors Number of Events Per Predictor First Author, Year of Publication
1. 70%‐100% ACS Renqiu Stroke Screening Study, China 2012 18/3006 (0.6%) 7 2.6 Yan et al, 201826 Model 1
2. 50%‐100% ACS 33/3006 (1.1%) 8 4.1 Model 2
3. >70% ACS Four observational studies: Sweden, Norway, Germany, four communities in the United States

Tromsø: 1994–1995;

MDCS: 1991–1996;

CAPS: NA;

CHS: NA

 127/23 706 (0.5%) 8 15.9 de Weerd et al, 201423 Model 1
4. >50% ACS 465/23 706 (2.0%) 8 58.1 Model 2
5. >50% ACS Screening, NY, USA 2001–2002 38/394 (9.6%) 4 9.5 Jacobowitz et al, 200324
6. ≥60% ACS Screening, NY, USA 1997 239/1331 (18%) 4 59.8 Qureshi et al, 200125
Open in a new tab

ACS indicates asymptomatic carotid stenosis; CAPS, Carotid Atherosclerosis Progression Study; CHS, Cardiovascular Health Study; MDCS, Malmö Diet and Cancer Study; and NA, not available.

Three prediction models were developed to detect ACS ≥50%,23, 24, 26 1 model was developed to detect ACS ≥60%,25 and 2 models were developed to detect ACS ≥70%.23, 26 The risk predictors included age, sex, smoking, hypertension, hypercholesterolemia, diabetes mellitus, myocardial infarction, stroke or TIA, height, measured blood pressure, and blood lipids. The number of predictors included in the prediction models varied from 4 to 8. Two models used clinical characteristics, and 4 models used blood measurements in addition to clinical characteristics. An overview of the predictors used in prediction models is provided in Table S5. The number of cases used to develop the prediction models varied from 394 to 23 706; the number of events varied from 18 to 465, and the number of cases per predictor varied from 2.6 to 59.8.

The overall risk of bias was low in 2 models and high in 4 models. Concerns with the applicability of the prediction models was deemed low in 3 models, unclear in 2 models, and high in 1 model. An overview of the risk of bias and the applicability per model is provided in Table S6.

Predictive Performance

Discriminative performance, as assessed by the AUROC curves varied from 0.81 to 0.88 in the derivation cohorts, and from 0.71 to 0.87 in the internal validation cohorts, respectively (Figure 2).23, 24, 25, 26, 27 Only 1 study provided calibration plots.26

Figure 2.

Figure 2

Open in a new tab

Discriminative performance of risk prediction models.

The symbols represent the AUROC curves of the included prediction models and the vertical bars represent the 95% CIs. The values of the AUROC curves and 95% CIs are provided in Table S6. The models of Jacobowitz et al24 and Qureshi et al25 were originally developed for >50% ACS and ≥60% ACS, respectively. Suri et al, 2008 used ≥50% ACS and ≥75% ACS as outcomes for the external validation.27 The AUROC curves of 2 external validations for ≥50% ACS in the models developed for ≥70% ACS by de Weerd et al23 and Yan et al26 and 2 external validations for ≥70% ACS in the models developed for ≥50% ACS by the same authors are omitted in this figure. ACS indicates asymptomatic carotid stenosis; and AUROC, area under receiver operating characteristic.

In 2 studies, 10 external validation analyses were performed.26, 27 In Yan et al, 6 external validation analyses were performed using both ≥50% and ≥70% ACS as outcomes.26 The number of cases used for external validation in their study was 5010, of which 64 (1.3%) had ≥50% ACS, and 38 (0.8%) had ≥70% ACS. The AUROC curve ranged from 0.63 to 0.68. No (re)calibration was performed. A cohort from China used for external validation was geographically and temporally distinct from the derivation cohorts. In Suri et al, 4 external validation analyses were performed using ≥50% and ≥75% ACS as predicted outcomes.27 The number of cases used for external validation in their study was 5449, of which 227 (4.2%) had ≥50% ACS and 52 (1.0%) had ≥75% ACS. The AUROC curve ranged from 0.56 to 0.60. No (re)calibration was performed. The validation cohort was from the United States, as were the derivation cohorts of the validated models and the data of validation cohort were older than the derivation cohorts.

External Validation

The validation cohort consisted of 596 469 participants, of whom 11 178 (1.87%) participants had ≥50% ACS and 2033 (0.34%) participants had ≥70% ACS. Baseline characteristics of the validation cohort are provided in Table 2.

Table 2.

Selected Characteristics of Participants in the External Validation Cohort, by Severity of ACS

Participants With <50% ACS (n=585 291) Participants With 50% to 69% ACS (n=9145) Participants With ≥70% ACS (n=2033)b All Participants (n=596 469)
Age, y 62.0±10.0 68.7±8.9 68.3±8.8 62.2±10.1
Sex (male) 208 285 (35.6) 3442 (37.6) 1009 (49.6) 212 736 (35.7)
Current or former smoker 207 329 (40.0) 4865 (61.0) 1245 (69.2) 213 439 (40.4)
Never smoker 311 192 (60.0) 3112 (39.0) 555 (30.8) 314 859 (59.6)
Hypertension 202 768 (36.0) 5185 (58.9) 1166 (60.6) 209 119 (36.4)
Diabetes mellitus 44 986 (8.2) 1577 (18.3) 312 (16.4) 46 875 (8.4)
Coronary heart diseasea 26 997 (5.1) 1262 (14.9) 344 (18.6) 28 603 (5.3)
Stroke/TIA 17 154 (3.3) 758 (9.0) 274 (15.0) 18 186 (3.4)
Peripheral arterial disease 16 370 (2.8) 1184 (13.4) 424 (21.8) 17 978 (3.1%)
Height, m 1.68±0.1 1.67±0.1 1.69±0.1 1.68±0.1
SBP, mm Hg 132±19.5 142±21.8 146±23.5 132±19.6
DBP, mm Hg 78±9.8 76±10.2 78±11.5 78±9.8
HDL‐C, mmol/L 1.4±0.5 1.3±0.5 1.3±0.4 1.4±0.5
LDL‐C, mmol/L 3.0±0.9 3.0±1.1 3.0±1.1 3.0±0.9
TC/HDL‐ratio 4.0±1.6 4.2±1.7 4.4±2.0 4.0±1.6
Open in a new tab

Values are mean±SD for continuous variables and n (%) for categorical variables. DBP indicates diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; SBP, systolic blood pressure; TC, total cholesterol; and TIA, transient ischemic attack.

a

In this group, 500 participants had a presumed occlusion.

b

Coronary heart disease is defined as previous myocardial infarction or a coronary intervention (bypass, angioplasty, or stenting).

Discrimination for outcome ≥50% ACS

The model with the best discrimination showed an AUROC curve of 0.749 (95% CI, 0.744–0.753).23 The discriminative performance was fair in 3 other models with an AUROC curve of 0.727 (95% CI, 0.722–0.732), 0.704 (95% CI, 0.700–0.709) and 0.703 (95% CI, 0.699–0.708).25, 26 The discriminative performance was poor in 1 model with an AUROC curve of 0.673 (95% CI, 0.668–0.678).24

Discrimination for outcome ≥70% ACS

The model with the best discrimination showed an AUROC curve of 0.779 (95% CI, 0.770–0.789).23 The discriminative performance was fair in 3 other models with an AUROC curve of 0.759 (95% CI, 0.749–0.770), 0.731 (95% CI, 0.721–0.742) and 0.701 (95% CI, 0.690–0.712).25, 26 The discriminative performance was poor in 1 model with an AUROC curve of 0.689 (95% CI, 0.677–0.701)24 (Figure 2 and Table S7).

Calibration

In the model with the best discrimination, predicted probabilities (after recalibration with adjusting the intercept) showed good concordance between the predicted prevalence calculated with the prediction model and the observed prevalence in the external validation cohort. The predicted and observed prevalence of ≥50% ACS in the highest decile was 6.4% and 6.5%, respectively (Figure 3A).23 The predicted and observed prevalence of ≥70% ACS in the highest 2 deciles was 1.7% and 1.4%, respectively (Figure S1). Other calibration plots are provided as Figures S1 and S2 for the outcomes ≥70% ACS and ≥50% ACS, respectively.

Figure 3.

Figure 3

Open in a new tab

Clinical application of the prediction model of de Weerd et al 23 for ≥50% ACS.

A, Calibration plot of external validation of the prediction model developed by de Weerd et al.23 It shows the predicted and observed prevalence of ≥50% ACS (after recalibration with adjusting the intercept). The boxes represent one decile of predicted risk, and the vertical lines represent the 95% CIs. B, Graph showing the sensitivity and specificity and corresponding observed prevalence and number needed to screen to detect 1 participant with ≥50% ACS using the prediction model developed by de Weerd et al.23 The square corresponds to targeted screening of participants in the highest decile of predicted risk. The prevalence in this decile is 6.5% with a number needed to screen of 15, and sensitivity is 34.8%. The circle corresponds to targeted screening of participants in the highest two deciles of predicted risk. The prevalence in these deciles is 4.8% with a number needed to screen of 21 and sensitivity of 55.0%. ACS indicates asymptomatic carotid stenosis; and NNS, number needed to scan.

Application of the Prediction Model With the Best Discrimination

Application for outcome ≥50% ACS

First, we assessed targeted screening in the highest decile and highest 2 deciles of predicted risk. Prevalence of ≥50% ACS in the highest decile of predicted risk was 6.5% with a number needed to scan (NNS) of 15. Targeted screening of the highest decile identified 34.8% of cases with ≥50% ACS. Prevalence in the 2 highest deciles of predicted risk was 4.8% with an NNS of 21. Targeted screening of the 2 highest deciles identified 55.0% of cases with ≥50% ACS (Figure 3B and Table S8).

Second, we assessed targeted screening with fixed levels of sensitivity. For this, test characteristics were calculated for 2 levels of sensitivity (≈80% and 90%). Observed prevalences of ≥50% ACS were 2.78% and 3.38% for the sensitivity of 90.0% and 79.5%. The corresponding specificity was 40.0% and 56.6%, respectively (Table S8).

Application for outcome ≥70% ACS

Prevalence of ≥70% ACS in the highest decile of predicted risk was 1.4% with an NNS of 70. Targeted screening of the highest decile identified 41.7% of cases with ≥70% ACS. Prevalence in the 2 highest deciles of predicted risk was 0.98% with an NNS of 102. Targeted screening of the 2 highest deciles identified 62.1% of cases with ≥70% ACS (Figure S3 and Table S8).

Using fixed levels of sensitivity (≈80% and 90%), observed prevalences of ≥70% ACS were 0.8% and 0.5% for the sensitivity of 76.8% and 92.0%. The corresponding specificity was 65.1% and 40.0%, respectively (Table S8).

Sensitivity Analysis

Validation in subsets with complete cases, cases without a history of TIA or stroke, showed comparable results. Validation in the subset of cases without a history of cardiovascular disease showed a lower AUROC (Figure S4 and Table S9).

Discussion

The present study validated prediction models in an external population to identify a cohort of individuals at high risk of asymptomatic carotid stenosis (ACS). In the model with the best discrimination, the observed prevalence of ACS in the decile at highest risk was 6.5% (≥50% ACS) and 1.4% (≥70% ACS) with an NNS of 15 and 70, respectively. Targeted screening of individuals in the highest decile of risk reliably identified 35% of cases with ≥50% ACS and 42% of cases with ≥70% ACS.

Early identification of ACS cases allows the initiation or intensification of cardiovascular risk management using triple medical therapy (ie, antithrombotic, antihypertensive, and lipid‐lowering medication) to decrease the risk of cardiovascular disease. Carotid intervention might further decrease the risk of stroke in selected cases. Clinical and imaging features associated with an increased risk of stroke in patients with medically treated ACS, such as silent brain infarction, contralateral stroke, or TIA, plaque echolucency, intraplaque hemorrhage, microemboli, and reduced cerebrovascular reserve, have been identified.10, 28 Risk stratification tools, using a wide range of predictors, have been developed to estimate long‐term stroke and cardiovascular disease risk in cases with ACS, but these have not been validated with current medical treatment.29, 30 Reliable and validated risk stratification tools might help further refine the use of targeted screening for ACS by identifying cases at higher risk for stroke and cardiovascular disease.

We found that discrimination was less for participants without cardiovascular disease, but targeted screening could also include participants with a history of cerebrovascular or cardiovascular disease, since not all of these participants were taking adequate preventive treatments. Annual ipsilateral risk of stroke in ACS cases on medical therapy in previous randomized controlled trials varied between 1.4% and 2.4%.31, 32, 33 More recent studies have reported lower risks attributable to improving risk factor management.29 Annual risk of ipsilateral ischemic stroke and TIA in cases with >50% ACS and a history of TIA or minor stroke in another territory with consequent use of secondary prophylaxis was as low as 0.34% and 1.78%, respectively.34

The discrimination of the best model was fair and calibration good, despite differences between the original derivation and our validation cohort. Differences in duplex protocols, (eg, unilateral or bilateral screening), and differences in the methods of measurement of degree of stenosis between populations may have contributed to lower external performance in this large external validation cohort. Duplex screening does not assess intracranial stenosis, and extracranial calcified vessels can hamper reliable assessment. Different criteria for assessment of stenosis are available, but validity of duplex ultrasound performed by experienced sonographers is good,35 and peak systolic velocity, while it is a simple measurement, may be useful as a screening tool to identify cases for more intensive evaluation.

The present study had several strengths. We conducted an extensive literature search to identify existing models and previous external validation according to a prespecified protocol. We used a large cohort for external validation and all models were validated using the same participants, allowing us to directly compare their predictive performance. Missing data in the validation population were limited for most variables, and our findings were unaffected by missing values. Multiple imputation was used to handle missing data, which is preferred to complete‐case analysis. A direct match between predictors in the models and the external validation cohort was available for all predictors of externally validated models. Bilateral examination of the carotid arteries was performed and stenoses of either side were used as outcome. Our sensitivity analyses showed that exclusion of participants with previous stroke or TIA and exclusion of participants with previous cardiovascular disease did not influence the findings of the main analysis substantially.

The present study also had several limitations. First, even though the external validation data were prospectively collected, it was not primarily designed for research purposes. Second, participants were self‐referred and self‐funded, which may limit the generalizability to other (screened) populations. In addition, some predictors were not included in established risk prediction models, such as social status, possibly hampering reliable prediction in specific groups of patients. Third, data on medical history and height were assessed by self‐reporting and, hence, may be susceptible to recall bias. Fourth, data from duplex measurement of the internal carotid artery and common carotid artery were not recorded separately.

Risk prediction models with good calibration are needed to improve the efficiency of targeted screening programs by identifying those at greatest risk, but future research should determine the long‐term predictors of stroke and cardiovascular disease and determine the number of events that could be prevented by using more intensive medical treatment.

In conclusion, the present study showed that most prediction models had modest discrimination but could reliably identify a cohort of cases at high risk of ACS. The prevalence of ACS in the decile(s) at highest predicted risk of ACS was considerably higher than the overall prevalence, thereby substantially reducing the number of individuals needed to screen to detect ACS. Further research should determine the optimum thresholds required for a targeted screening by considering the number needed to screen, the diagnostic yield, the absolute reduction of stroke risk by prophylactic treatment, and cost‐effectiveness of different approaches.

Sources of Funding

Prof Halliday is funded by the UK Health Research (NIHR) Oxford Biomedical Research Centre (BRC). Prof Lewington reports grants from UK Medical Research Council and from the CDC Foundation (with support from Amgen) outside the submitted work.

Disclosures

None.

Supporting information

Data S1Tables S1–S9Figures S1–S4References 12, 14, 23–27, 36–131

Click here for additional data file. (4.1MB, pdf)

Acknowledgments

The authors acknowledge Life Line Screening for providing data for these analyses. This study was designed, conducted, and reported independently of Life Line Screening and all funding sources. The authors are also grateful to J.A.A.G. Damen for advice on the selection of prediction models and E.C. Brand for advice on statistical analysis.

Author contributions: Prof Halliday and Mr Bulbulia, and Prof Lewington obtained the data and considered the clinical applicability of Life Line Screening. Mr Poorthuis designed the study. Mr Poorthuis, Mr Sherliker and Ms Clack cleaned the data. Mr Poorthuis designed the search strategy, performed literature searches, and removed duplicates. Mr Poorthuis and Dr Massa screened titles and abstracts and assessed full‐text articles and reference lists of included studies. Mr Poorthuis performed the statistical analyses, supervised by Dr Massa, Mr Sherliker, and Prof Lewington. The manuscript was drafted by Mr Poorthuis. All authors interpreted the data, contributed to revision and editing of the manuscript, and approved the final version of the manuscript for submission for publication.

(J Am Heart Assoc. 2020;e014766 DOI: 10.1161/JAHA.119.014766.)

For Sources of Funding and Disclosures, see page 10.

References

  • 1. George J, Rapsomaniki E, Pujades‐Rodriguez M, Shah AD, Denaxas S, Herrett E, Smeeth L, Timmis A, Hemingway H. How does cardiovascular disease first present in women and men? Circulation. 2015;132:1320–1328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Dorresteijn JAN, Visseren FLJ, Wassink AMJ, Gondrie MJA, Steyerberg EW, Ridker PM, Cook NR, van der Graaf Y. Development and validation of a prediction rule for recurrent vascular events based on a cohort study of patients with arterial disease: the SMART risk score. Heart. 2013;99:866–872. [DOI] [PubMed] [Google Scholar]
  • 3. Petty GW, Brown RD Jr, Whisnant JP, Sicks JD, O'Fallon WM, Wiebers DO. Ischemic stroke subtypes: a population‐based study of functional outcome, survival, and recurrence. Stroke. 2000;31:1062–1068. [DOI] [PubMed] [Google Scholar]
  • 4. Schulz UG, Rothwell PM. Differences in vascular risk factors between etiological subtypes of ischemic stroke: importance of population‐based studies. Stroke. 2003;34:2050–2059. [DOI] [PubMed] [Google Scholar]
  • 5. Bogiatzi C, Wannarong T, McLeod AI, Heisel M, Hackam D, Spence JD. SPARKLE (Subtypes of Ischaemic Stroke Classification System), incorporating measurement of carotid plaque burden: a new validated tool for the classification of ischemic stroke subtypes. Neuroepidemiology. 2014;42:243–251. [DOI] [PubMed] [Google Scholar]
  • 6. Goessens BM, Visseren FL, Kappelle LJ, Algra A, van der Graaf Y. Asymptomatic carotid artery stenosis and the risk of new vascular events in patients with manifest arterial disease: the SMART study. Stroke. 2007;38:1470–1475. [DOI] [PubMed] [Google Scholar]
  • 7. de Weerd M, Greving JP, Hedblad B, Lorenz MW, Mathiesen EB, O'Leary DH, Rosvall M, Sitzer M, Buskens E, Bots ML. Prevalence of asymptomatic carotid artery stenosis in the general population: an individual participant data meta‐analysis. Stroke. 2010;41:1294–1297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Meschia JF, Bushnell C, Boden‐Albala B, Braun LT, Bravata DM, Chaturvedi S, Creager MA, Eckel RH, Elkind MS, Fornage M, et al. Guidelines for the primary prevention of stroke: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2014;45:3754–3832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Ricotta JJ, Aburahma A, Ascher E, Eskandari M, Faries P, Lal BK; Society for Vascular Surgery . Updated Society for Vascular Surgery guidelines for management of extracranial carotid disease. J Vasc Surg. 2011;54:e1–e31. [DOI] [PubMed] [Google Scholar]
  • 10. Naylor AR, Ricco JB, de Borst GJ, Debus S, de Haro J, Halliday A, Hamilton G, Kakisis J, Kakkos S, Lepidi S, et al. Editor's Choice—management of atherosclerotic carotid and vertebral artery disease: 2017 clinical practice guidelines of the European Society for Vascular Surgery (ESVS). Eur J Vasc Endovasc Surg. 2018;55:3–81. [DOI] [PubMed] [Google Scholar]
  • 11. Brott TG, Halperin JL, Abbara S, Bacharach JM, Barr JD, Bush RL, Cates CU, Creager MA, Fowler SB, Friday G, et al. 2011 ASA/ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/SAIP/SCAI/SIR/SNIS/SVM/SVS guideline on the management of patients with extracranial carotid and vertebral artery disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American Stroke Association, American Association of Neuroscience Nurses, American Association of Neurological Surgeons, American College of Radiology, American Society of Neuroradiology, Congress of Neurological Surgeons, Society of Atherosclerosis Imaging and Prevention, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of NeuroInterventional Surgery, Society for Vascular Medicine, and Society for Vascular Surgery. J Am Coll Cardiol. 2011;57:e16–e94. [DOI] [PubMed] [Google Scholar]
  • 12. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta‐analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med. 2009;6:e1000100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Moons KG, de Groot JA, Bouwmeester W, Vergouwe Y, Mallett S, Altman DG, Reitsma JB, Collins GS. Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLoS Med. 2014;11:e1001744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Ingui BJ, Rogers MA. Searching for clinical prediction rules in MEDLINE. J Am Med Inform Assoc. 2001;8:391–397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Wolff RF, Moons KGM, Riley RD, Whiting PF, Westwood M, Collins GS, Reitsma JB, Kleijnen J, Mallett S. PROBAST: a tool to assess the risk of bias and applicability of prediction model studies. Ann Intern Med. 2019;170:51–58. [DOI] [PubMed] [Google Scholar]
  • 16. Moons KGM, Wolff RF, Riley RD, Whiting PF, Westwood M, Collins GS, Reitsma JB, Kleijnen J, Mallett S. PROBAST: a tool to assess risk of bias and applicability of prediction model studies: explanation and elaboration. Ann Intern Med. 2019;170:W1–W33. [DOI] [PubMed] [Google Scholar]
  • 17. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30:377–399. [DOI] [PubMed] [Google Scholar]
  • 18. Morris TP, White IR, Royston P, Seaman SR, Wood AM. Multiple imputation for an incomplete covariate that is a ratio. Stat Med. 2014;33:88–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Rodwell L, Lee KJ, Romaniuk H, Carlin JB. Comparison of methods for imputing limited‐range variables: a simulation study. BMC Med Res Methodol. 2014;14:57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Marshall A, Altman DG, Holder RL, Royston P. Combining estimates of interest in prognostic modelling studies after multiple imputation: current practice and guidelines. BMC Med Res Methodol. 2009;9:57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Rubin DB. Inference and missing data. Biometrika. 1976;63:581–592. [Google Scholar]
  • 22. Steyerberg EW. Clinical Prediction Models: A Practical Approach to Development, Validation, and Updating. New York: Springer‐Verlag; 2009. [Google Scholar]
  • 23. de Weerd M, Greving JP, Hedblad B, Lorenz MW, Mathiesen EB, O'Leary DH, Rosvall M, Sitzer M, de Borst GJ, Buskens E, et al. Prediction of asymptomatic carotid artery stenosis in the general population: identification of high‐risk groups. Stroke. 2014;45:2366–2371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Jacobowitz GR, Rockman CB, Gagne PJ, Adelman MA, Lamparello PJ, Landis R, Riles TS. A model for predicting occult carotid artery stenosis: screening is justified in a selected population. J Vasc Surg. 2003;38:705–709. [DOI] [PubMed] [Google Scholar]
  • 25. Qureshi AI, Janardhan V, Bennett SE, Luft AR, Hopkins LN, Guterman LR. Who should be screened for asymptomatic carotid artery stenosis? Experience from the Western New York Stroke Screening Program. J Neuroimaging. 2001;11:105–111. [DOI] [PubMed] [Google Scholar]
  • 26. Yan Y, Gao S, Yang H, Qin S, Li F, Zhang G, Yang B, He Y, Zhao Y, Li E, et al. ECAS score: a web‐based risk model to predict moderate and severe extracranial carotid artery stenosis. Neurol Res. 2018;40:249–257. [DOI] [PubMed] [Google Scholar]
  • 27. Suri MFK, Ezzeddine MA, Lakshminarayan K, Divani AA, Qureshi AI. Validation of two different grading schemes to identify patients with asymptomatic carotid artery stenosis in general population. J Neuroimaging. 2008;18:142–147. [DOI] [PubMed] [Google Scholar]
  • 28. Aboyans V, Ricco JB, Bartelink MEL, Bjorck M, Brodmann M, Cohnert T, Collet JP, Czerny M, De Carlo M, Debus S, et al. 2017 ESC guidelines on the diagnosis and treatment of peripheral arterial diseases, in collaboration with the European Society for Vascular Surgery (ESVS): document covering atherosclerotic disease of extracranial carotid and vertebral, mesenteric, renal, upper and lower extremity arteriesEndorsed by: the European Stroke Organization (ESO) The Task Force for the Diagnosis and Treatment of Peripheral Arterial Diseases of the European Society of Cardiology (ESC) and of the European Society for Vascular Surgery (ESVS). Eur Heart J. 2018;39:763–816. [DOI] [PubMed] [Google Scholar]
  • 29. Nicolaides AN, Kakkos SK, Griffin M, Sabetai M, Dhanjil S, Tegos T, Thomas DJ, Giannoukas A, Geroulakos G, Georgiou N, et al. Severity of asymptomatic carotid stenosis and risk of ipsilateral hemispheric ischaemic events: results from the ACSRS study. Eur J Vasc Endovasc Surg. 2005;30:275–284. [DOI] [PubMed] [Google Scholar]
  • 30. Nicolaides AN, Kakkos SK, Kyriacou E, Griffin M, Sabetai M, Thomas DJ, Tegos T, Geroulakos G, Labropoulos N, Dore CJ, et al. Asymptomatic internal carotid artery stenosis and cerebrovascular risk stratification. J Vasc Surg. 2010;52:1486–1496. [DOI] [PubMed] [Google Scholar]
  • 31. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA. 1995;273:1421–1428. [PubMed] [Google Scholar]
  • 32. Halliday A, Harrison M, Hayter E, Kong X, Mansfield A, Marro J, Pan H, Peto R, Potter J, Rahimi K, et al. 10‐year stroke prevention after successful carotid endarterectomy for asymptomatic stenosis (ACST‐1): a multicentre randomised trial. Lancet. 2010;376:1074–1084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Hobson RW II, Weiss DG, Fields WS, Goldstone J, Moore WS, Towne JB, Wright CB. Efficacy of carotid endarterectomy for asymptomatic carotid stenosis. The Veterans Affairs Cooperative Study Group. N Engl J Med. 1993;328:221–227. [DOI] [PubMed] [Google Scholar]
  • 34. Marquardt L, Geraghty OC, Mehta Z, Rothwell PM. Low risk of ipsilateral stroke in patients with asymptomatic carotid stenosis on best medical treatment: a prospective, population‐based study. Stroke. 2010;41:e11–e17. [DOI] [PubMed] [Google Scholar]
  • 35. Jahromi AS, Cina CS, Liu Y, Clase CM. Sensitivity and specificity of color duplex ultrasound measurement in the estimation of internal carotid artery stenosis: a systematic review and meta‐analysis. J Vasc Surg. 2005;41:962–972. [DOI] [PubMed] [Google Scholar]
  • 36. Abd Allah F, Baligh E, Ibrahim M. Clinical relevance of carotid atherosclerosis among Egyptians: a 5‐year retrospective analysis of 4,733 subjects. Neuroepidemiology. 2010;35:275–279. [DOI] [PubMed] [Google Scholar]
  • 37. Aboyans V, Lacroix P. Indications for carotid screening in patients with coronary artery disease. Presse Med. 2009;38:977–986. [DOI] [PubMed] [Google Scholar]
  • 38. Admani AK, Mangion DM, Naik DR. Extracranial carotid artery stenosis: prevalence and associated risk factors in elderly stroke patients. Atherosclerosis. 1991;86:31–37. [DOI] [PubMed] [Google Scholar]
  • 39. Ahn SS, Baker JD, Walden K, Moore WS. Which asymptomatic patients should undergo routine screening carotid duplex scan? Am J Surg. 1991;162:180–183; discussion 183‐4. [DOI] [PubMed] [Google Scholar]
  • 40. Aizenberg DJ. Cardiovascular testing in asymptomatic patients: carotid duplex, cardiac stress testing, screen for peripheral arterial disease. Med Clin North Am. 2016;100:971–979. [DOI] [PubMed] [Google Scholar]
  • 41. Alexandrov AV. Ultrasound and angiography in the selection of patients for carotid endarterectomy. Curr Cardiol Rep. 2003;5:141–147. [DOI] [PubMed] [Google Scholar]
  • 42. Ansari S, Tan JY, Larcos GS, Paterson H. Low prevalence of significant carotid artery disease on ultrasound in patients proceeding to coronary artery bypass surgery. Intern Med J. 2011;41:658–661. [DOI] [PubMed] [Google Scholar]
  • 43. Archbold RA, Barakat K, Magee P, Curzen N. Screening for carotid artery disease before cardiac surgery: is current clinical practice evidence based? Clin Cardiol. 2001;24:26–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Ascher E, Hingorani A, Yorkovich W, Ramsey PJ, Salles‐Cunha S. Routine preoperative carotid duplex scanning in patients undergoing open heart surgery: is it worthwhile? Ann Vasc Surg. 2001;15:669–678. [DOI] [PubMed] [Google Scholar]
  • 45. Ballard JL, Mazeroll R, Weitzman R, Harward TR, Flanigan DP. Medical benefits of a peripheral vascular screening program. Ann Vasc Surg. 2007;21:159–162. [DOI] [PubMed] [Google Scholar]
  • 46. Barvalia M, Silber D, DiVita M, Joshi A, Wasty N, Cohen M. Utility of carotid duplex ultrasonography in a general inner‐city hospital. Cardiovasc Ultrasound. 2014;12:48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Belcaro G, Nicolaides AN, Cesarone MR, De Sanctis MT, Geroulakos G, Lennox A, Griffin M, Incandela L. Vascular screening in andropause: non‐invasive investigations in vascular disease screening. Curr Med Res Opin. 2000;16(suppl 1):s68–s71. [DOI] [PubMed] [Google Scholar]
  • 48. Berens ES, Kouchoukos NT, Murphy SF, Wareing TH. Preoperative carotid artery screening in elderly patients undergoing cardiac surgery. J Vasc Surg. 1992;15:313–321; discussion 322‐3. [PubMed] [Google Scholar]
  • 49. Berger JS, Hochman J, Lobach I, Adelman MA, Riles TS, Rockman CB. Modifiable risk factor burden and the prevalence of peripheral artery disease in different vascular territories. J Vasc Surg. 2013;58:673–681.e1. [DOI] [PubMed] [Google Scholar]
  • 50. Bosevski M, Borozanov V, Georgievska‐Ismail L. Influence of metabolic risk factors on the presence of carotid artery disease in patients with type 2 diabetes and coronary artery disease. Diab Vasc Dis Res. 2007;4:49–52. [DOI] [PubMed] [Google Scholar]
  • 51. Bosevski M, Lazarova‐Trajkovska E. Carotid artery disease in patients with type 2 diabetes. Angiol Vasc Surg. 2015;21:17–26. [PubMed] [Google Scholar]
  • 52. Carnicelli AP, Stone JJ, Doyle A, Chowdhry A, Gillespie DL, Chandra A. Predictive multivariate regression to increase the specificity of carotid duplex ultrasound for high‐grade stenosis in asymptomatic patients. Ann Vasc Surg. 2014;28:1548–1555. [DOI] [PubMed] [Google Scholar]
  • 53. Carnicelli AP, Stone JJ, Doyle A, Chowdhry AK, Mix D, Ellis J, Gillespie DL, Chandra A. Cross‐sectional area for the calculation of carotid artery stenosis on computed tomographic angiography. J Vasc Surg. 2013;58:659–665. [DOI] [PubMed] [Google Scholar]
  • 54. Carsten CG III, Elmore JR, Franklin DP, Thomas DD, Mordan F, Wood GC. Use of limited color‐flow duplex for a carotid screening project. Am J Surg. 1999;178:173–176. [DOI] [PubMed] [Google Scholar]
  • 55. Chiquete E, Torres‐Octavo B, Cano‐Nigenda V, Valle‐Rojas D, Dominguez‐Moreno R, Tolosa‐Tort P, Florez‐Cardona JA, Flores‐Silva F, Reyes‐Melo I, Higuera‐Calleja J, et al. Characterisation of factors associated with carotid stenosis in a population at high risk. Rev Neurol. 2014;58:541–547. [PubMed] [Google Scholar]
  • 56. Chua HC, Sitoh YY, Earnest A, Venketasubramanian N. Detection of internal carotid artery stenosis with duplex velocity criteria using receiver operating characteristic analysis. Ann Acad Med Singapore. 2007;36:247–252. [PubMed] [Google Scholar]
  • 57. Chou CL, Wu YJ, Hung CL, Liu CC, Wang SD, Wu TW, Wang LY, Yeh HI. Segment‐specific prevalence of carotid artery plaque and stenosis in middle‐aged adults and elders in Taiwan: a community‐based study. J Formos Med Assoc. 2019;118:64–71. [DOI] [PubMed] [Google Scholar]
  • 58. Colgan MP, Strode GR, Sommer JD, Gibbs JL, Sumner DS. Prevalence of asymptomatic carotid disease: results of duplex scanning in 348 unselected volunteers. J Vasc Surg. 1988;8:674–678. [DOI] [PubMed] [Google Scholar]
  • 59. Cull DL, Cole T, Miller B, Johnson B, Rawlinson D, Walker E, Taylor SM. The value of a carotid duplex surveillance program for stroke prevention. Ann Vasc Surg. 2011;25:887–894. [DOI] [PubMed] [Google Scholar]
  • 60. Derdeyn CP, Powers WJ. Cost‐effectiveness of screening for asymptomatic carotid atherosclerotic disease. Stroke. 1996;27:1944–1950. [DOI] [PubMed] [Google Scholar]
  • 61. Derdeyn CP, Powers WJ, Moran CJ, DeWitte ITC, Allen BT. Role of Doppler US in screening for carotid atherosclerotic disease. Radiology. 1995;197:635–643. [DOI] [PubMed] [Google Scholar]
  • 62. Di Carli MF, Hachamovitch R. Should we screen for occult coronary artery disease among asymptomatic patients with diabetes? J Am Coll Cardiol. 2005;45:50–53. [DOI] [PubMed] [Google Scholar]
  • 63. Duval M, Altman JJ. When should ultrasonography be used to detect asymptomatic carotid atheroma in diabetic patients? Diabetes Metab. 2006;32:638–642. [DOI] [PubMed] [Google Scholar]
  • 64. Ellis MR, Franks PJ, Cuming R, Powell JT, Greenhalgh RM. Prevalence, progression and natural history of asymptomatic carotid stenosis: is there a place for carotid endarterectomy? Eur J Vasc Surg. 1992;6:172–177. [DOI] [PubMed] [Google Scholar]
  • 65. Elmore EM, Mosquera A, Weinberger J. The prevalence of asymptomatic intracranial large‐vessel occlusive disease: the role of diabetes. J Neuroimaging. 2003;13:224–227. [PubMed] [Google Scholar]
  • 66. Engelhardt M, Bruijnen H, Schnur C, Pfadenhauer K, Bohndorf K, Wolfe KD. Duplex scanning criteria for selection of patients for internal carotid artery endarterectomy. Vasa. 2005;34:36–40. [DOI] [PubMed] [Google Scholar]
  • 67. Eugene JR, Abdallah M, Miglietta M, Vernenkar VV, Pascual R, Briones R, Barnes T, Hager J. Carotid occlusive disease: primary care of patients with or without symptoms. Geriatrics. 1999;54:24–41. [PubMed] [Google Scholar]
  • 68. Fabris F, Zanocchi M, Bo M, Scarafiotti C. Carotid plaques, aging, and risk factors for atherosclerosis. Arch Gerontol Geriatr. 1994;18:45–50. [DOI] [PubMed] [Google Scholar]
  • 69. Felberg RA, Christou I, Demchuk AM, Malkoff M, Alexandrov AV. Screening for intracranial stenosis with transcranial Doppler: the accuracy of mean flow velocity thresholds. J Neuroimaging. 2002;12:9–14. [DOI] [PubMed] [Google Scholar]
  • 70. Fernandes M, Keerthiraj B, Mahale AR, Kumar A, Dudekula A. Evaluation of carotid arteries in stroke patients using color Doppler sonography: a prospective study conducted in a tertiary care hospital in South India. Int J Appl Basic Med Res. 2016;6:38–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71. Ghanaati H, Golchin N, Motevalli M, Shakiba M, Jalali AH, Firouznia K. Carotid Doppler ultrasonography in preoperative assessment of coronary artery bypass graft surgery in an Iranian population: association between atherosclerosis risk factors and carotid stenosis. Iran J Radiol. 2009;6:125–129. [Google Scholar]
  • 72. Giral P, Bruckert E, Dairou F, Boubrit K, Drobinski G, Chapman JM, Beucler I, Turpin G. Usefulness in predicting coronary artery disease by ultrasonic evaluation of the carotid arteries in asymptomatic hypercholesterolemic patients with positive exercise stress tests. Am J Cardiol. 1999;84:14–17. [DOI] [PubMed] [Google Scholar]
  • 73. Gao MJ, Hua Y, Jia LY, Ling C, Duan C. Characteristics of large vascular lesions detected by trancranial Doppler in diabetic patients complicated with hypertension. [Chinese]. Chin J Cerebrovasc Dis. 2011;8:298–302. [Google Scholar]
  • 74. Greco G, Egorova NN, Moskowitz AJ, Gelijns AC, Kent KC, Manganaro AJ, Zwolak RM, Riles TS. A model for predicting the risk of carotid artery disease. Ann Surg. 2013;257:1168–1173. [DOI] [PubMed] [Google Scholar]
  • 75. Hedblad B, Janzon L, Jungquist G, Ogren M. Factors modifying the prognosis in men with asymptomatic carotid artery disease. J Intern Med. 1998;243:57–64. [DOI] [PubMed] [Google Scholar]
  • 76. Helfre M, Grange C, Riche B, Maucort‐Boulch D, Thivolet C, Vouillarmet J. Usefulness of a systematic screening of carotid atherosclerosis in asymptomatic people with type 2 diabetes for cardiovascular risk reclassification. Ann Endocrinol (Paris). 2017;78:14–19. [DOI] [PubMed] [Google Scholar]
  • 77. Hogberg D, Kragsterman B, Bjorck M, Tjarnstrom J, Wanhainen A. Carotid artery atherosclerosis among 65‐year‐old Swedish men—a population‐based screening study. Eur J Vasc Endovasc Surg. 2014;48:5–10. [DOI] [PubMed] [Google Scholar]
  • 78. Hoshino T, Sissani L, Labreuche J, Ducrocq G, Lavallee PC, Meseguer E, Guidoux C, Cabrejo L, Hobeanu C, Gongora‐Rivera F, et al. Prevalence of systemic atherosclerosis burdens and overlapping stroke etiologies and their associations with long‐term vascular prognosis in stroke with intracranial atherosclerotic disease. JAMA Neurol. 2018;75:203–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79. Howard G, Baker WH, Chambless LE, Howard VJ, Jones AM, Toole JF. An approach for the use of Doppler ultrasound as a screening tool for hemodynamically significant stenosis (despite heterogeneity of Doppler performance). A multicenter experience. Asymptomatic Carotid Atherosclerosis Study Investigators. Stroke. 1996;27:1951–1957. [DOI] [PubMed] [Google Scholar]
  • 80. Hua Y, Tao YL, Li M, Yong Q, He W, Zhao H, Luo Y, Zhang Y, Peng T, Yu DL, et al. Multicenter ultrasound screening for the results of carotid atherosclerotic lesions in a Chinese population with high‐risk of stroke: a preliminary analysis. [Chinese]. Chin J Cerebrovasc Dis. 2014;11:617–623. [Google Scholar]
  • 81. Hughes JP, Dubin R, Rodriguez‐Wong A, Porreca FJ. Risk focused screening for vascular disease: one University Hospital's experience. J Vasc Ultrasound. 2010;34:118–123. [Google Scholar]
  • 82. Joakimsen O, Bonaa KH, Mathiesen EB, Stensland‐Bugge E, Arnesen E. Prediction of mortality by ultrasound screening of a general population for carotid stenosis: the Tromso Study. Stroke. 2000;31:1871–1876. [DOI] [PubMed] [Google Scholar]
  • 83. Jonas DE, Feltner C, Amick HR, Sheridan S, Zheng ZJ, Watford DJ, Carter JL, Rowe CJ, Harris R. Screening for asymptomatic carotid artery stenosis: a systematic review and meta‐analysis for the U.S. Preventive Services Task Force. Ann Intern Med. 2014;161:336–346. [DOI] [PubMed] [Google Scholar]
  • 84. Kakkos SK, Nicolaides AN, Charalambous I, Thomas D, Giannopoulos A, Naylor AR, Geroulakos G, Abbott AL, Adovasio R, Ziani B, et al. Predictors and clinical significance of progression or regression of asymptomatic carotid stenosis. J Vasc Surg. 2014;59:956–967.e1. [DOI] [PubMed] [Google Scholar]
  • 85. Karnon J, Goyder E, Tappenden P, McPhie S, Towers I, Brazier J, Madan J. A review and critique of modelling in prioritising and designing screening programmes. Health Technol Assess. 2007;11:1–145. [DOI] [PubMed] [Google Scholar]
  • 86. Kazemi‐Bajestani SMR, van der Vlugt M, de Leeuw FE, Blankensteijn JD, Bredie SJH. A high prevalence of carotid artery stenosis in male patients older than 65 years, irrespective of presenting clinical manifestation of atherosclerotic diseases. Angiology. 2013;64:281–286. [DOI] [PubMed] [Google Scholar]
  • 87. Kazum S, Eisen A, Lev EI, Iakobishvili Z, Solodky A, Hasdai D, Kornowski R, Mager A. Prevalence of carotid artery disease among ambulatory patients with coronary artery disease. Isr Med Assoc J. 2016;18:100–103. [PubMed] [Google Scholar]
  • 88. Lacroix P, Aboyans V, Criqui MH, Bertin F, Bouhamed T, Archambeaud F, Laskar M. Type‐2 diabetes and carotid stenosis: a proposal for a screening strategy in asymptomatic patients. Vasc Med. 2006;11:93–99. [DOI] [PubMed] [Google Scholar]
  • 89. Lassila HC, Tyrrell KS, Matthews KA, Wolfson SK, Kuller LH. Prevalence and determinants of carotid atherosclerosis in healthy postmenopausal women. Stroke. 1997;28:513–517. [DOI] [PubMed] [Google Scholar]
  • 90. Lee TT, Solomon NA, Heidenreich PA, Oehlert J, Garber AM. Cost‐effectiveness of screening for carotid stenosis in asymptomatic persons. Ann Intern Med. 1997;126:337–346. [DOI] [PubMed] [Google Scholar]
  • 91. LeFevre ML. Screening for asymptomatic carotid artery stenosis: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2014;161:356–362. [DOI] [PubMed] [Google Scholar]
  • 92. Li Q, Zhang Y, Wang M, Zhu X. Study on the screening program and risk factors of carotid artery stenosis with coronary artery disease. [Chinese]. Zhonghua Liu Xing Bing Xue Za Zhi. 2013;34:626–629. [PubMed] [Google Scholar]
  • 93. Liang Y, Yan Z, Sun B, Cai C, Jiang H, Song A, Qiu C. Cardiovascular risk factor profiles for peripheral artery disease and carotid atherosclerosis among Chinese older people: a population‐based study. PLoS One. 2014;9:e85927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 94. Lim LS, Haq N, Mahmood S, Hoeksema L, Surricchio M, Abraham‐Katz RB, Bergeisen G, Compton MT, Guillory VJ, et al. Atherosclerotic cardiovascular disease screening in adults: American College of Preventive Medicine position statement on preventive practice. Am J Prev Med. 2011;40:381.e1–381.e10. [DOI] [PubMed] [Google Scholar]
  • 95. Lim YJ, Kim YW, Choe YH, Ki CS, Park SK. Risk factor analysis for development of asymptomatic carotid stenosis in Koreans. J Korean Med Sci. 2006;21:15–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 96. Martin MJ, Mullenix PS, Crawford JV, Cuadrado DS, Andersen CA. Focused high‐risk population screening for carotid arterial stenosis after radiation therapy for head and neck cancer. Am J Surg. 2004;187:594–598. [DOI] [PubMed] [Google Scholar]
  • 97. Mathiesen EB, Joakimsen O, Bonaa KH. Prevalence of and risk factors associated with carotid artery stenosis: the Tromso Study. Cerebrovasc Dis. 2001;12:44–51. [DOI] [PubMed] [Google Scholar]
  • 98. Meng X, Liu G, Wang X, Xu Q. Analysis of carotid atherosclerosis and related risk factors in a university physical examination population in Beijing. [Chinese]. Natl Med J China. 2017;97:2620–2624. [DOI] [PubMed] [Google Scholar]
  • 99. Moneta GL, Taylor DC, Zierler RE, Kazmers A, Beach K, Strandness DE Jr. Asymptomatic high‐grade internal carotid artery stenosis: is stratification according to risk factors or duplex spectral analysis possible? J Vasc Surg. 1989;10:475–483. [DOI] [PubMed] [Google Scholar]
  • 100. Mostaza JM, Gonzalez‐Juanatey JR, Castillo J, Lahoz C, Fernandez‐Villaverde JM, Maestro‐Saavedra FJ. Prevalence of carotid stenosis and silent myocardial ischemia in asymptomatic subjects with a low ankle‐brachial index. J Vasc Surg. 2009;49:104–108. [DOI] [PubMed] [Google Scholar]
  • 101. Niederkorn K, Horner S, Schmidt R, Fazekas F, Lechner H. Ultrasonic assessment of the prevalence of carotid artery stenosis in asymptomatic volunteers. J Neuroimaging. 1991;1:119–122. [Google Scholar]
  • 102. Obuchowski NA, Modic MT, Magdinec M, Masaryk TJ. Assessment of the efficacy of noninvasive screening for patients with asymptomatic neck bruits. Stroke. 1997;28:1330–1339. [DOI] [PubMed] [Google Scholar]
  • 103. O'Leary DH, Kronmal RA, Wolfson SK, Bond MG, Tracy RP, Gardin JM, Kittner SJ, Price TR, Savage PJ. Sonographic evaluation of carotid artery atherosclerosis in the elderly: relationship of disease severity to stroke and transient ischemic attack. Radiology. 1993;188:363–370. [DOI] [PubMed] [Google Scholar]
  • 104. O'Leary DH, Polak JF, Kronmal RA, Kittner SJ, Bond MG, Wolfson SK Jr, Bommer W, Price TR, Gardin JM, Savage PJ. Distribution and correlates of sonographically detected carotid artery disease in the Cardiovascular Health Study. Stroke. 1992;23:1752–1760. [DOI] [PubMed] [Google Scholar]
  • 105. Paprottka KJ, Saam D, Rubenthaler J, Schindler A, Sommer NN, Paprottka PM, Clevert DA, Reiser M, Saam T, Helck A. Prevalence and distribution of calcified nodules in carotid arteries in correlation with clinical symptoms. Radiol Med. 2017;122:449–457. [DOI] [PubMed] [Google Scholar]
  • 106. Park JH, Kim WH, Kim JH, Park TS, Baek HS. Prevalence of and risk factors for extracranial internal carotid artery stenosis in Korean Type 2 diabetic patients. Diabet Med. 2006;23:1377–1380. [DOI] [PubMed] [Google Scholar]
  • 107. Prati P, Vanuzzo D, Casaroli M, Di Chiara A, De Biasi F, Feruglio GA, Touboul PJ. Prevalence and determinants of carotid atherosclerosis in a general population. Stroke. 1992;23:1705–1711. [DOI] [PubMed] [Google Scholar]
  • 108. Prati P, Vanuzzo D, Casaroli M, Bader G, Mos L, Pilotto L, Canciani L, Ruscio M, Touboul PJ. Determinants of carotid plaque occurrence. A long‐term prospective population study: the San Daniele Project. Cerebrovasc Dis. 2006;22:416–422. [DOI] [PubMed] [Google Scholar]
  • 109. Qiu J, Zhou Y, Yang X, Zhang Y, Li Z, Yan N, Wang Y, Ge S, Wu S, Zhao X, et al. The association between ankle‐brachial index and asymptomatic cranial‐carotid stenosis: a population‐based, cross‐sectional study of 5440 Han Chinese. Eur J Neurol. 2016;23:757–762. [DOI] [PubMed] [Google Scholar]
  • 110. Rockman CB, Hoang H, Guo Y, Maldonado TS, Jacobowitz GR, Talishinskiy T, Riles TS, Berger JS. The prevalence of carotid artery stenosis varies significantly by race. J Vasc Surg. 2013;57:327–337. [DOI] [PubMed] [Google Scholar]
  • 111. Rockman CB, Jacobowitz GR, Gagne PJ, Adelman MA, Lamparello PJ, Landis R, Riles TS. Focused screening for occult carotid artery disease: patients with known heart disease are at high risk. J Vasc Surg. 2004;39:44–51. [DOI] [PubMed] [Google Scholar]
  • 112. Roh YN, Woo SY, Kim N, Kim S, Kim YW, Kim DI. Prevalence of asymptomatic carotid stenosis in Korea based on health screening population. J Korean Med Sci. 2011;26:1173–1177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 113. Ryglewicz D, Rozenfeld A, Baranska‐Gieruszczak M, Czyrny M, Lechowicz W. Carotid artery disease in patients with ischemic stroke. Results of year‐long observation conducted within the framework of prospective epidemiological studies, Warsaw, 1991–1992. [Polish]. Neurol Neurochir Pol. 1998;32:255–264. [PubMed] [Google Scholar]
  • 114. Saleem MA, Sadat U, Walsh SR, Young VE, Gillard JH, Cooper DG, Gaunt ME. Role of carotid duplex imaging in carotid screening programmes—an overview. Cardiovasc Ultrasound. 2008;6:1–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 115. Savji N, Rockman CB, Skolnick AH, Guo Y, Adelman MA, Riles T, Berger JS. Association between advanced age and vascular disease in different arterial territories: a population database of over 3.6 million subjects. J Am Coll Cardiol. 2013;61:1736–1743. [DOI] [PubMed] [Google Scholar]
  • 116. Shah B, Rockman CB, Guo Y, Chesner J, Schwartzbard AZ, Weintraub HS, Adelman MA, Riles TS, Berger JS. Diabetes and vascular disease in different arterial territories. Diabetes Care. 2014;37:1636–1642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 117. Silaghi CN, Fodor D, Craciun AM. Circulating matrix Gla protein: a potential tool to identify minor carotid stenosis with calcification in a risk population. Clin Chem Lab Med. 2013;51:1115–1123. [DOI] [PubMed] [Google Scholar]
  • 118. Smolen HJ, Cohen DJ, Samsa GP, Toole JF, Klein RW, Furiak NM, Lorell BH. Development, validation, and application of a microsimulation model to predict stroke and mortality in medically managed asymptomatic patients with significant carotid artery stenosis. Value Health. 2007;10:489–497. [DOI] [PubMed] [Google Scholar]
  • 119. Solomon NA, Heidenreich PA, Oehlert J, Garber AM. Cost‐effectiveness of screening for carotid stenosis in asymptomatic persons. Ann Intern Med. 1997;126:337–346. [DOI] [PubMed] [Google Scholar]
  • 120. Stein RA, Rockman CB, Guo Y, Adelman MA, Riles T, Hiatt WR, Berger JS. Association between physical activity and peripheral artery disease and carotid artery stenosis in a self‐referred population of 3 million adults. Arterioscler Thromb Vasc Biol. 2015;35:206–212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 121. Sutton‐Tyrrell K, Alcorn HG, Wolfson SK Jr, Kelsey SF, Kuller LH. Predictors of carotid stenosis in older adults with and without isolated systolic hypertension. Stroke. 1993;24:355–361. [DOI] [PubMed] [Google Scholar]
  • 122. Touze E. Natural history of asymptomatic carotid artery stenosis. Rev Neurol (Paris). 2008;164:793–800. [DOI] [PubMed] [Google Scholar]
  • 123. Walters GK, Jones CE, Meyd CJ, Cavaluzzi JA, Chachich BM. The role of carotid duplex ultrasonography in the therapeutic algorithm of extracranial carotid disease. J Vasc Technol. 1993;17:177–182. [Google Scholar]
  • 124. Weisman SM, Brooks EA. Costs and benefits of bundled community‐based screening for carotid artery stenosis, peripheral artery disease, and abdominal aortic aneurysm. Manag Care. 2015;24:45–53. [PubMed] [Google Scholar]
  • 125. Whitty CJM, Sudlow CLM, Warlow CP. Investigating individual subjects and screening populations for asymptomatic carotid stenosis can be harmful. J Neurol Neurosurg Psychiatry. 1998;64:619–623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 126. Willeit J, Kiechl S. Prevalence and risk factors of asymptomatic extracranial carotid artery atherosclerosis: a population‐based study. Arterioscler Thromb. 1993;13:661–668. [DOI] [PubMed] [Google Scholar]
  • 127. Woo SY, Joh JH, Han SA, Park HC. Prevalence and risk factors for atherosclerotic carotid stenosis and plaque a population‐based screening study. Medicine (Baltimore). 2017;96:e5999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 128. Wyman RA, Mays ME, McBride PE, Stein JM. Ultrasound‐detected carotid plaque as a predictor of cardiovascular events. Vasc Med. 2006;11:123–130. [DOI] [PubMed] [Google Scholar]
  • 129. Yin D, Carpenter JP. Cost‐effectiveness of screening for asymptomatic carotid stenosis. J Vasc Surg. 1998;27:245–255. [DOI] [PubMed] [Google Scholar]
  • 130. Yu RHY, Ho SC, Ho SSY, Chan SSG, Woo JLF, Ahuja AT. Carotid atherosclerosis and the risk factors in early postmenopausal Chinese women. Maturitas. 2009;63:233–239. [DOI] [PubMed] [Google Scholar]
  • 131. Zorach BB, Arpin PA, Nelson J, MacKey WC. Progression of moderate‐to‐severe carotid disease. J Vasc Surg. 2016;63:1505–1510. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1Tables S1–S9Figures S1–S4References 12, 14, 23–27, 36–131

Click here for additional data file. (4.1MB, pdf)

Articles from Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease are provided here courtesy of Wiley

RESOURCES