| Author year | Population/setting | Mean/median age (range) | Female (%) | Symptomatic patients (%) | TIA(%) | Stroke subtypes | Reference standard | Method for measuring degree of stenosis | Stenosis defined as positive on reference standard | Prevalence of positive stenosis | Prevalence of occlusion | CUS method | CUS criteria | Cutoff for positive test by NASCET criteria | Number arteries/patients | Indeterminate results | Sensitivity | Specificity | Positive LR | Negative LR | Quality | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Eliasziw 1995 | 1011 patients with stroke/TIA from 50 academic medical centers participating in NASCET | 66 | 31% | 100% | 68% | Hemispheric or retinal | Selective carotid angiography | NASCET | 70-99% | 48% (patients) | Not reported | Duplex | Combinations of PSV, PSF, ICA/CCA PSV ratio | 70% | 1011 patients (ipsilateral arteries studied) | Not reported | 69% | 68% | 2.16 | 0.46 | GOOD. | Exclusion of patients with < 30% stenosis may have inflated sensitivity and decreased specificity. |
| Dippel 1999 | 152 consecutive patients with minor stroke/TIA at a university hospital in the Netherlands | 56 (19-80) | 35% | 100% | 47% | Not reported | Selective carotid angiography | ECST | 70-99% | 16% (patients) | 8% (patients) | Duplex/color flow | PSV>150, ICA/CCA PSV ratio >2, EDV>50, 1+ turbulence | 50% | 321 patients (ipsilateral arteries studied) | 2 uninterpretable duplex exams excluded | 58% | 92% | 7.25 | 0.46 | FAIR. No inter-rater reliability assessment. Discretionary subject selection. Two indeterminate arteries excluded. | Sensitivity and specificity were adjusted for verification bias. |
| Golledge 1999 | 50 patients with recent stroke/TIA at an academic medical center in the UK | 71 (47-84) | 48% | 100% | Not reported | Not reported | Selective carotid DSA | NASCET | 70-99% | 33% (arteries) | 9% (arteries) | Duplex/color flow | EDV > 90 | 70% | 100 arteries | None | 88% | 84% | 5.50 | 0.14 | FAIR. CUS criteria were derived from this sample, not validated. No inter-rater reliability assessment. | |
| Russell 1985 | 66 patients with symptoms suggestive of carotid occlusion at a hospital in Norway | Not reported | Not reported | 100% | Not reported | Not reported | Selective carotid angiography | NASCET | 51-99% 76-99% | 23% (arteries) 10% (arteries) | 9% (arteries) | Doppler spectral analysis | Ratio of spectral spreading at stenosis site vs. distal ICA | 51% 76% | 132 arteries | None None | 86% 72% | 99% 93% | 86.00 9.63 | 0.14 0.30 | FAIR. No inter-rater reliability assessment. Discretionary subject selection. Sample was not well described. | |
| Torvaldsen 1985 | 52 patients undergoing Doppler scanning who had had angiography in the prior month at a hospital in Australia | 59 | 33% | 27% | 23% | Not reported | Carotid angiography | NASCET | 50-99% 75-99% | 29% (arteries) 12% (arteries) | 6% (arteries) | Doppler/color Echoflow | Color codes indicating degree of stenosis | 50% 75% | 101 arteries | None None | 83% 61% | 78% 93% | 3.72 8.17 | 0.22 0.42 | FAIR. No inter-rater reliability assessment. Discretionary subject selection. | |
| Vanninen 1995 | 45 consecutive patients with carotid territory ischemic symptoms at a university hospital in Finland | 58 (34-72) | 16% | 100% | Not reported | Carotid territory only | Carotid DSA | NASCET | 70-99% | 14% (arteries) | Not reported | Duplex/color flow | PSV > 230 | 70% | 90 arteries | None | 93% | 93% | 13.29 | 0.08 | FAIR. No inter-rater reliability assessment. Discretionary subject selection. | |
| Worthy 1997 | 73 consecutive patients with symptomatic carotid territory ischemia at a hospital in the UK | 62 (47-84) | 47% | 100% | 68% | Carotid territory only | Selective carotid DSA | ECST | 70-99% | 31% (arteries) | 16% (arteries) | Duplex/color flow | PSV>300 and EDV>120 | 50% | 143 arteries | 3 arteries excluded due to atypical findings on CUS | 76% (73%) | 95% (91%) | 15.20 (8.44) | 0.25 (0.28) | FAIR. No inter-rater reliability assessment. Three indeterminate arteries excluded. | Figures in parentheses represent sensitivity, specificity, and LRs when 3 indeterminate results are assumed to be all false + or all false - |
| Sillesen 1988 | 68 consecutive patients referred for carotid evaluation to the vascular surgery department of a university hospital in Denmark | 61 (37-78) | Not reported | 79% | Not reported | Not reported | Cerebral angiography | NASCET | 50-99% | 23% (arteries) | 8% (arteries) | Doppler spectral analysis | Systolic spectral broadening index > 0.37 | 50% | 128 arteries | 3 uninterpretable Doppler exams excluded | 92% (86%) | 83% (80%) | 5.41 (4.60) | 0.10 (0.17) | POOR. CUS criteria were derived from this sample, not validated. No inter-rater reliability assessment. Three indeterminate arteries. | Figures in parentheses represent sensitivity, specificity, and LRs when 3 indeterminate results are assumed to be all false + or all false - |
TIA: transient ischemic attack; CUS: carotid ultrasound; LR: likelihood ratio; NASCET: North American Symptomatic Carotid Endarterectomy Trial; ECST: European Carotid Surgery Trial; PSV: peak systolic velocity; PSF:peak systolic frequency; ICA: internal carotid artery; CCA: common carotid artery; EDV: end-diastolic velocity; DSA: digital subtraction angiography
From: Evidence Tables
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.