Redefined Duplex Ultrasonographic Criteria for Diagnosis of Carotid Artery Stenosis

Redefined Duplex Ultrasonographic Criteria for Diagnosis of Carotid Artery Stenosis JOHN HUSTON III, MD; E. MEREDITH JAMES, MD; ROBERT D. BROWN, JR, MD; ROBERT D. LEFSRUD, MD; DUANE M. ILSTRUP, MS; ELLIS F. ROBERTSON, MD; FREDRIC B. MEYER, MD; AND JOHN W. HALLETT, MD velocity of 230 cmls or more resulted in a sensitivity of 86.4%, a specificity of 90.1 %, a positive predictive value of 82.7%, a negative predictive value of 92.3%, and an accuracy of 88.8%. An end diastolic velocity of 70 cmls or more and an internal carotid artery:common carotid artery ratio of 3.2 or more yielded similar values. For an internal carotid artery stenosis of 50% or more, a peak systolic velocity of 130 cmls or more resulted in a sensitivity of 92.1%, a specificity of 89.5%, a positive predictive value of 90.3%, a negative predictive value of91.3%, and an overall accuracy of 90.8%. An internal carotid artery:common carotid artery ratio of 1.6 or more yielded similar values. • Conclusion: In our ultrasonography laboratory, a carotid artery stenosis of 70% or more (for which carotid endarterectomy is typically recommended in symptomatic patients) is diagnosed reliably with the following duplex ultrasonographic criteria: a peak systolic velocity of 230 cmls or more, an end diastolic velocity of 70 cmls or more, or an internal carotid artery:common carotid artery ratio of 3.2 or more. Mayo Clin Proc. 2000;75:1133-1140

• Objective: To evaluate duplex ultrasonographic criteria for the determination of 50% or more and 70% or more stenosis of the diameter of the internal carotid artery based on conventional angiography in order to align ultrasonographic diagnostic categories with current clinical management schemes. • Patients and Methods: Between January 1, 1995, and June 30, 1999,915 patients underwent both carotid duplex ultrasonography and cerebral angiography within 30 days at Mayo Clinic, Rochester, Minn. Of these patients, 294 were excluded from this study because of occlusion of one or both of the internal carotid arteries or atypical flow characteristics. In the remaining 621 patients (61% male, 39% female; mean age, 67.7 years [range, 14-88 years]), 1218 vessels were available for correlation. Several Doppler ultrasonographic velocity variables were compared with the angiographic findings by use of receiver operating characteristic curve analysis. The primary end point was verification of optimal ultrasonographic criteria to diagnose 70% or more internal carotid artery stenosis. The secondary end point was establishment of threshold values to detect stenosis of 50% or more. • Results: At angiography, 382 patients had internal carotid arteries with 70 % or more stenosis. Peak systolic and end diastolic velocities of the internal carotid artery and internal carotid artery:common carotid artery peak systolic velocity ratios were measured. For an internal carotid artery stenosis of 70% or more, a peak systolic

CCA = common carotid artery; CT = computedtomographic; ECST = European Carotid SurgeryTrial; EDV = end diastolic velocity; ICA = internal carotid artery; NASCET = North American Symptomatic Carotid Endarterectomy Trial; PSV = peak systolic velocity; VA = VeteransAffairs

D

uring the past decade, 2 large multicenter trials-the North American Symptomatic Carotid Endarterectomy Trial (NASCET)I and the Veterans Affairs (VA) triaF-reported a reduction in the risk of cerebral infarction for symptomatic patients with 70% or more stenosis of the carotid artery who underwent carotid endarterectomy compared with medical management. For patients with symp-

tomatic stenosis of moderate severity (50%-69%), carotid endarterectomy has shown some degree of efficacy for prevention of cerebral infarction,' but the results make this treatment option controversial in these patients. Currently, duplex ultrasonography of the carotid artery is the primary noninvasive screening procedure for evaluating extracerebral carotid artery stenosis. The 2 most commonly used diagnostic schemes for carotid artery duplex sonography, developed by Zwiebel" and Strandness' on the basis of angiographic measurements, do not include criteria for identifying stenosis of 70% or more, which was the threshold value used in the largest North American trial.I Some vascular surgeons and neurosurgeons have advocated endarterectomy on the basis of ultrasonographic findings alone or in combination with magnetic resonance angiography or computed tomographic (CT) angiography

From the Department of Radiology (J.H., E.M.J., R.D.L., E.F.R.), Department of Neurology (R.D.B.), Section of Biostatistics (D.M.I.), Department of Neurologic Surgery (F.B.M.), and Division of Vascular Surgery (J'w.H.), Mayo Clinic, Rochester, Minn. Dr Lefsrud is now with St Paul Radiology, St Paul, Minn. Dr Robertson is now in Dallas, Tex. Address reprint requests and correspondence to John Huston III, MD, Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (e-mail: [email protected]).

Mayo Clin Proc. 2000;75:1133-1140

1133

© 2000Mayo Foundation for Medical Education and Research

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1134 Diagnostic Criteria for Carotid Artery Stenosis

Mayo Clin Proc, November 2000,Vol 75

Table 1. Duplex Ultrasonographic Criteria for Diagnosis of Carotid Artery Stenosis of 70% or More (NASCET)*

Source Robinson et al,22 1988

Polak et al,23 1991

No. of patients/ vessels 105/205

96/129

Hunink et al," 1993

96/138

Moneta et al," 1993

100/184

PSV >225 crn/s EDV >75 cm/s ICA:CCA PSV ratio >3.0 PSV >215 cm/s PSV

~230

Sensitivity

Specificity

Accuracy

NA NA NA

NA NA NA

NA NA NA

90 80 (40)t

86

NA

(%)

Criteria

cm/s

PSV ~325 cm/s ICA:CCA PSV ratio

(%)

(%)

90 (99)t

NA

91 87 94 85 98

88 88 93 80 95

Faught et al,2o 1994

4051770

PSV >210 cm/s EDV >100 cm/s PSV >130 and EDV >100 cm/s

83 91 89 77 81

Neale et al," 1994

60/120

PSV >270 cm/s EDV >110 cm/s PSV >270 and EDV > 110 cm/s

96 91 96

86 93 91

88 93 93

Browman et al," 1995 Van Leersum et al,25 1995

75/145

PSV <40 cm/s or > 175 cm/s

91

60

NA

74/115

PSV >200 cm/s

97

86

NA

PSV >210 cm/s EDV >70 cm/s ICA:CCA PSV ratio >3.0 ICA:CCA EDV ratio >3.3

94 91 100

77 60 78 65

83 77 83 79

87

97

95

85

95

92

96 100

93 92

93 93

Carpenter et al,26 1996

110/210

Hood et al,27 1996 AbuRahma et al,28 1998

231/462

PSV

Grant et al,29 2000

NA/201:j:

PSV ~175 cm/s ICA:CCA PSV ratio

248/457

~4.0

PSV >130 and EDV >100 cm/s ~150

and EDV ;:::90 cm/s ~2.5

92

*CCA:= common carotid artery; EDV:= end diastolic velocity; ICA:= internal carotid artery; NA:= not available; NASCET:= North American Symptomatic Carotid Endarterectomy Trial; PSV:= peak systolic velocity. tValues corrected for angiographic selection bias. :j:Symptomatic patients only.

to avoid the cost and potential complications of conventional angiography.v" Because endarterectomies are performed increasingly on the basis of noninvasive testing only, alignment of the ultrasonographic interpretive schemes with the categories of stenosis used in clinical management and decision making is imperative. Several different ultrasonographic criteria have been proposed for diagnosing 70% or more stenosis of the carotid artery, but no consensus exists on optimal or preferred criteria'F" (Table 1). This situation may be attributable to differences in equipment, sonographic practices, patient populations, and techniques for measuring stenosis.v-" These discrepancies underscore the need for each laboratory to validate internally the criteria it uses to diagnose carotid stenosis, particularly if the local practice relies on ultrasonography as the primary preoperative test. The purpose of this study was to verify the optimal ultrasonographic criteria in our practice for diagnosing 70% ~r more internal carotid artery (lCA) stenosis with use of the NASCET measurement techniques on conventional

angiograms as the reference standard. In addition, threshold values for detecting stenosis of 50% or more were identified to use as a marker for moderate carotid atherosclerotic disease. PATIENTS AND METHODS Study Population Between January 1, 1995, and June 30, 1999, 23,965 carotid ultrasonographic examinations and 3465 carotid angiographic studies were performed at Mayo Clinic, Rochester, Minn. Among the patients who underwent these tests, 915 had ultrasonography and angiography performed within 30 days. Of these patients, 294 were excluded from the study because of occlusion of 1 ICA or both, atypical flow patterns within the vessels (such as low velocities in a preocclusive state), extensive calcified plaque resulting in long segments of acoustic shadowing, or severe cardiac arrhythmia. These findings can invalidate the use of velocity measurements as the sole criterion for the determination of stenosis." Therefore, the study population in-

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, November 2000, Vol 75

eluded 621 patients with 1218 vessels for correlation and included patients both symptomatic and asymptomatic for carotid stenotic disease. Of the 621 patients, 379 (61%) were male and 242 (39%) were female; their ages ranged from 14 to 88 years (mean, 67.7 years). The patients were tracked through a computer-based carotid vascular registry. Carotid Duplex Ultrasonography Technique

Diagnostic Criteria for Carotid Artery Stenosis

1135

Statistical Analysis

Several velocity measurements from the Doppler spectral display were compared with angiographic findings with use of receiver operating characteristic curve analysis. Estimates of sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated for various cut points of PSV, EDV, and ICA:CCA PSV ratios. The 95% confidence limits were computed for each estimate.

Scanning was performed with 7.5- or 5-MHz transducers (Acuson, Inc, Mountain View, Calif). The Doppler angle was maintained at or below 60°. The common carotid artery (CCA), rCA, and external carotid artery were evaluated. Angle-corrected velocities were determined routinely in the proximal, middle, and distal CCAs, at multiple sites in the visible cervical ICA and the proximal external carotid artery, and in any location where stenosis was suspected on the basis of color flow Doppler ultrasonography. The spectral display was used to measure the peak systolic velocity (PSV) and end diastolic velocity (EDV), and from these factors, the ICA:CCA PSV ratio was calculated. For the ICA:CCA ratio, the ICA value was the highest PSV obtainable in the ICA (presumably at the point of maximal stenosis), and the CCA value was a representative velocity in a nondiseased portion of the mid CCA. Use of this ratio theoretically eliminates cardiac output variability as a confounding factor in the process of predicting luminal stenosis on the basis of measured velocity." The Doppler values were measured prospectively at the time of the examination, and these values were used for the analysis. All ultrasonographic findings were interpreted by I or more of 13 experienced sonologists.

At angiography, 382 patients had an ICA stenosis of 70% or more. For an ICA stenosis of 70% or more, a PSV of 230 cm/s or more resulted in a sensitivity of 86.4%, a specificity of 90.1%, a positive predictive value of 82.7%, a negative predictive value of 92.3%, and an accuracy of 88.8%. An EDV of 70 cm/s or more and an ICA:CCA PSV ratio of 3.2 or more yielded similar values. For an ICA stenosis of 50% or more, a PSV of 130 cm/s or more resulted in a sensitivity of 92.1%, a specificity. of 89.5%, a positive predictive value of 90.3%, a negative predictive value of 91.3%, and an overall accuracy of 90.8%. An ICA:CCA ratio of 1.6 or more yielded similar values. An EDV of 40 cm/s or more was less accurate (Table 2). Combinations of the PSV, EDV, and ICA:CCA ratio did not statistically increase the sensitivity, specificity, or accuracy of the results. The receiver operating characteristic curves for the 70% threshold were not markedly different for the PSV, EDV, or ICA:CCA PSV ratios (Figure 1). At the 50% threshold, the EDV values performed worse than the PSV and ICA:CCA ratios, which yielded similar results.

Angiography Technique

DISCUSSION

All carotid angiography examinations were performed within 30 days of carotid duplex ultrasonography. A standard femoral approach with a modified Seldinger technique was used. Both cut film and digital angiography were performed with at least 2, and up to 4, selected views of each bifurcation. The intracranial vessels were studied routinely, and the aortic arch, carotid origins, and vertebral origins were examined in selected cases. Angiograms were interpreted by I or more of 11 experienced neuroradiologists. The degree of stenosis was calculated with use of the method advocated by the NASCET, in which the most severely stenotic site of the ICA was compared with the normal ICA distal to the carotid bulb." The stenosis measurement was recorded prospectively at the time of angiography, and this value was used for the analysis. Familiarity with the ultrasonographic results at the time of angiography precluded a blinded study.

Symptomatic patients with 70% or more stenosis of the carotid artery have a lower risk of subsequent cerebral infarction if treated with endarterectomy, as demonstrated by the NASCET and the VA trial.'> Several different duplex ultrasonographic criteria based on PSV, EDV, or ratios have been used to quantify this level of ICA stenosis, but the threshold criteria have differed":" (Table 1). The PSV in the rCA seems to be the best single variable for detecting and quantifying stenosis." It is also the easiest value to obtain. In prior studies that have determined optimal sonographic criteria for identifying ICA stenosis of 70% or more, the threshold PSV has ranged from 130 to 325 cm/s. The PSV values in most of these studies, however, have been about 200 cm/s, similar to our selected value of 230 cm/s. Because of inherent cost and potential related morbidity and mortality, the goal has been to attempt to replace conventional contrast angiography with noninvasive tech-

RESULTS

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1136 Diagnostic Criteria for Carotid Artery Stenosis

Mayo Clin Proc, November 2000, Vol 75

Table 2. Results With Use of Various Criteria for Grading Carotid Artery Stenosis*

Stenosis ~50%

Criteria PSV

~130

cm/s

EDV~40cm/s

~70%

rCACCA PSV ratio ~1.6 PSV ~230 cm/s EDV~70cm/s

rCA:CCA PSV ratio ~3.2

Sensitivity (%)

Specificity (%)

Positive predictive value (%)

Negative predictive value (%)

Accuracy (%)

92.1

89.5 (86.7,91.8) 85.0 (81.9, 87.8) 83.0 (79.7,85.9) 90.1 (87.8,92.1) 89.4 (87.1,91.5) 89.8 (87.5,91.8)

90.3 (87.8,92.5) 86.2 (83.3, 88.8) 85.4 (82.6, 88.0) 82.7 (78.9, 86.1) 81.1 (77.1,84.7) 82.5 (78.7,85.9)

91.3 (88.7,93.5) 86.4 (83.3, 89.0) 91.9 (89.2,94.1) 92.3 (90.2,94.1) 90.1 (87.8, 92.1) 92.6 (90.6,94.4)

90.8 (89.0,92.4) 86.3 (84.2, 88.2) 88.3 (86.3, 90.0) 88.8 (86.8, 90.5) 86.9 (84.8,88.7) 88.8 (86.9, 90.5)

(89.7,94.1) 87.5 (84.6, 89.9) 93.2 (90.9, 95.0) 86.4 (82.8, 89.5) 82.2 (78.3,85.7) 87.1 (83.5,90.1)

*CCA = common carotid artery; EDV = end diastolic velocity; rCA = internal carotid artery; PSV = peak systolic velocity. Numbers in parentheses are 95% confidence limits.

niques, such as duplex ultrasonography, magnetic resonance angiography, and CT angiography, used alone or in combination. Angiography has been regarded as the standard against which other methods for assessing vessel flow and morphologic findings have been judged. Some vascular surgeons and neurosurgeons now advocate endarterectomy on the basis of ultrasonographic studies alone or in combination with magnetic resonance or CT angiography to decrease the costs and morbidity associated with angiography.r" The reported rate of stroke after angiography ranges from about 0.5% to 4.8%, which does not include local or systemic adverse effects." In comparison with angiography, duplex ultrasonography may not be as useful for detecting other factors that may lead to an increased risk of stroke, such as plaque ulcerations, tandem carotid lesions, and intraluminal thrombus. We chose velocity and ratio thresholds that balanced the statistical variables at the highest values. A lower velocity threshold would increase the sensitivity but result in more false-positive values, which could potentially subject some patients to the unnecessary costs and risks of angiography or operation (Figure 2). A higher velocity threshold would decrease the number of false-positive results but eventuate in more missed cases of true stenosis; thus, some patients would be denied a surgical procedure that could potentially reduce their risk of subsequent stroke. The effects of varying the threshold PSV values on sensitivity, specificity, and accuracy for detecting a carotid artery stenosis of 70% or more are shown in Table 3. The threshold values selected were then based on a qualitative review of the data in Table 3 and in similar tables generated for EDV and the ICA:CCA PSV ratios. Of the 915 patients in whom both ultrasonography and angiography were performed within 30 days of each other,

294 were excluded. Of these, 116 were excluded because one ICA or both were occluded, which can result in spuriously increased velocities on the patent side. 32'36 The remaining 178 patients were excluded because of "atypical" imaging or flow conditions that resulted in imprecise or falsely reduced velocity measurements. Examples of such conditions include (I) extensive calcified plaque obscuring the carotid lumen and preventing complete Doppler sampling, (2) severe cardiac irregularities resulting in widely variable velocity readings from one beat to the next, and (3) near-occlusion of an ICA resulting in trickle flow (carotid string sign). In patients with trickle flow, the stenosis is so great or so extensive that blood velocities are actually reduced and no longer bear a direct relationship to the degree of stenosis." Sonographic recognition of these nearly occluded vessels, however, is usually not a problem because of the extensive plaque and/or thrombus evident on gray-scale imaging and the tiny residual flow channel visible on color flow Doppler imaging. The "normal" or low velocities measured in the tiny residual lumen actually belie the severity of the disease and, as such, are misleading. Such cases point out the necessity for analyzing all the ultrasound information available-gray-scale, color flow Doppler, and spectral Doppler-when interpreting carotid ultrasound studies. After all the exclusions described above, 621 patients were available for ultrasonographicangiographic correlation. Of the 1242 potential carotid vessels in these patients, 24 were not imaged by both modalities, leaving 1218 vessels, which constituted the study population. Our PSV criteria for diagnosis of 70% or more stenosis of the carotid artery are similar to those published by Hunink et aF4and Robinson et aF2and within 10% of those published by Polak et al,23 Faught et al," and Carpenter et

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, November 2000, Vol 75

Diagnostic Criteria for Carotid Artery Stenosis 1137

1.0

,,

0.8

~

.~ :t::

------------------

,-.'

I I I I

0.6

til

c:


Cf)

0.4

-

0.2 0.0

PSV

--- EDV _.- Ratio

0.0

0.4

0.2

0.6

0.8

1.0

1-speeifieity

Figure 1. Receiver operating characteristic curve for internal carotid artery peak systolic velocity (PSV), end diastolic velocity (EDV), and internal carotid artery:common carotid artery PSV ratio for threshold values of 70% stenosis of the internal carotid artery.

aF6 (Table 1). The values reported by Moneta et aP8 and Browman et aP9 of 325 cm/s or more and more than 175 cm/s, respectively, seem to be discordant with our data. Differences in patient population, equipment, or examina- . tion protocol could explain some of these discrepancies. The threshold value of 175 cm/s recommended by Grant et aF9 was derived from a clinical outcome approach (minimal cost to society from stroke and death) rather than from a simple receiver operating characteristic curve analysis and therefore differs from the approach used in the current study and most other investigations. In addition, the

,-,-'----------------------------

1.0 0.8 0


c.

,I ,, ,,

0.6

til

Q)

C/)

," I

,"

.,'

I

I

~

c:

threshold value in the study reported by Grant et al was derived from a population of only symptomatic patients, rather than from a blended population of symptomatic and asymptomatic patients as used in all other studies, including ours. The ICA:CCA PSV ratio is thought to compensate for changes in cardiac output caused by cardiac arrhythmia or interval changes in myocardial function." This ratio had the same overall accuracy as the PSV and was minimally more accurate than EDV in our study. Our ratio of 3.2 was considerably lower than the value of 4.0 reported by

I I I

, ,, ,, ,,, ,,

0.4

I

0.2 0.0

a

Sensitivity --- Specificity

200

400

600

800

em/sec Figure 2. Sensitivity/specificity (Sens/spec) vs peak systolic velocity for a threshold of 70% stenosis of the internal carotid artery.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1138 Diagnostic Criteria for Carotid Artery Stenosis

Mayo Clin Proc, November 2000, Vol 75

Table3. Effect of Variation of Threshold Velocity on Statistical Factors for DetectingCarotid Artery Stenosisof70% or More* PSV (cm/s)

Sensitivity

Specificity (%)

Positive predictive value (%)

Negative predictive value (%)

Accuracy

100

94.9 (92.4, 96.8) 94.0 (91.3,96.0) 92.8 (90.0,95.1) 91.5 (88.4, 93.9) 90.5 (87.4,93.1) 86.4 (82.8,89.5) 81.3 (77.3, 84.9) 73.9 (69.5,78.0) 64.9 (60.2,69.4)

58.2 (54.7,61.7) 68.4 (65.0,71.6) 76.3 (73.2, 79.2) 82.5 (79.7,85.1) 85.9 (83.2, 88.2) 90.1 (87.8,92.1) 92.5 (90.4, 94.2) 95.0 (93.3,96.4) 96.2 (94.6,97.4)

55.6 (51.9, 59.2) 62.1 (58.3, 65.9) 68.4 (64.4,72.1) 74.3 (70.4, 78.0) 77.9 (74.1,81.5) 82.7 (78.9,86.1) 85.6 (81.9, 88.9) 89.1 (85.4, 92.2) 90.4 (86.5,93.4)

95.4 (93.1,97.1) 95.4 (93.3,97.0) 95.1 (93.1,96.6) 94.6 (92.6, 96.2) 94.3 (92.3, 95.9) 92.3 (90.2,94.1) 90.0 (87.7,91.9) 86.8 (84.4, 89.0) 83.2 (80.6, 85.6)

71.3 (68.6,73.8) 77.5 (75.1,79.8) 82.2 (79.9,84.3) 85.7 (83.6,87.6) 87.5 (85.5,89.3) 88.8 (86.8, 90.5) 88.5 (86.6, 90.2) 87.5 (85.5, 89.3) 85.1 (82.9, 87.0)

125 150 175 200 230 250 275 300

(%)

(%)

*PSV = peaksystolic velocity. Numbers in parentheses are 95% confidence limits.

Moneta et al" but was within 10% of the value of 3.0 determined by Carpenter et aF6 and Robinson et aU2 Moneta et al" also found that the ICA:CCA PSV ratio was more sensitive, specific, and accurate than the PSV value. Nevertheless, the PSV is often regarded as the variable of choice because it is the easiest measurement to obtain, necessitates no additional calculation, and has shown slightly greater sensitivity and specificity in different studies. Combinations of the PSV, EDV, and ICA:CCA ratio did not increase the sensitivity, specificity, or accuracy in our current study, as they did in the studies by Faught et al," AbuRahma et al,2s and Hood et al." Receiver operating characteristic curves are a common method for comparison of the performance of diagnostic tests and measurements." The areas under the curves showed no statistically significant difference among PSV, EDV, or systolic velocity ratios for the prediction of ICA stenosis of 70% or more in this study. The EDV for the 50% stenosis threshold was less accurate than the corresponding PSV or systolic velocity ratio, and we do not routinely use or recommend it. It should be noted that, in addition to the previously cited NASCET and VA studies, another large trial, the European Carotid Surgery Trial (ECST), reported a reduction in the risk of infarction for symptomatic patients with 70% or more stenosis of the carotid artery who underwent carotid endarterectomy." However, the ECST used a technique for calculating stenosis that differed from both the

NASCET and the VA trial. Specifically, the NASCET and . the VA trial measured the maximal stenosis as a ratio to the more distal normal ICA, whereas the ECST used the presumed but unseen carotid bulb as the denominator." Therefore, a stenosis determined with the NASCET and the VA trial method would always be less than that determined by the ECST method. For example, a stenosis of 70% based on the ECST criteria is equivalent to a stenosis of 45% to 50% with use of the NASCET and the VA trial criteria. We elected to use the NASCET criteria for the comparisons in this report because this is the technique we use clinically at our institution. However, at other institutions, other angiographic techniques may be used to quantify stenosis. For this reason, it is imperative that the sonologist and the surgeon be aware of how the diagnostic and therapeutic criteria at their institution were developed. The computer-based carotid vascular registry used in our study provides several advantages. First, it allows cases to be added continuously and consecutively. Second, it provides the flexibility to reexamine the data from time to time, as clinical practice evolves. For example, as the NASCET and other trials continue to define the minimal notable stenosis appropriate for endarterectomy, new stenosis categories may emerge; hence, new velocity thresholds that correspond to these new stenosis categories will need to be established. Finally, the large number of cases in the registry results in narrower 95% confidence intervals when statistical analysis is performed.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, November 2000, Vol7S

Diagnostic Criteria for Carotid Artery Stenosis

1139

Table 4. Carotid Ultrasonography Criteria for Evaluating Carotid Artery Stenosis* Criteriafor grading stenosis Category, % stenosis

PSV

EDV

(cmls)

(cmls)

ICACCA PSV ratio

Normal or mild disease, <50% Moderate disease, 50%-69% Severedisease, 70%-99%

<130 130-229

... t <70

<1.6 1.6-3.1

~230

~70

~3.2

*CCA = common carotid artery; EDV = end diastolic velocity; ICA = internal carotidartery; NASCET = North American Symptomatic Carotid EndarterectomyTrial; PSV =peak systolic velocity. tNo reliableEDVfor stenosis <50%. The current study provides data regarding the ultrasonographic criteria for determining a 50% stenosis of the ICA. Recently available results from a randomized trial have documented the relative efficacy of carotid endarterectomy for reducing the risk of subsequent cerebral infarction in selected patients with an ICA stenosis of 50% to 69% and recent minor cerebral infarction or transient ischemic attack.' Although the results are somewhat controversial and a larger number of patients must be treated in order to prevent a stroke at 2 years after carotid endarterectomy in comparison with patients who have carotid artery stenosis of 70% or more, the finding of 50% stenosis is now being used as an important cutoff for consideration for performance of a carotid endarterectomy. In our current study, patients were selected to have angiography partially on the basis of their ultrasonographic findings. Consequently, the calculated sensitivity and specificity are higher than expected in the general population because of selection bias." Both verification and test interpretation biases are limitations of this study. CONCLUSION Duplex ultrasonography is an accurate, noninvasive diagnostic method for evaluating carotid artery stenosis. At our institution, the criterion currently used to detect 70% or more stenosis of the carotid artery (for which carotid endarterectomy is typically recommended in symptomatic patients) is a PSV of 230 cm/s or more, an EDV of 70 cm/s or more, or an ICACCA PSV ratio of 3.2 or more. For a 50% stenosis threshold, we use a PSV of 130 cm/s or an ICA:CCA ratio of 1.6 (Table 4). Preoperatively, most patients undergo confirmatory magnetic resonance angiography. For borderline, questionable, or discrepant cases, conventional angiography is performed.

REFERENCES I.

2.

3.

4. 5. 6.

7.

8.

9. 10. II. 12.

13. 14.

ACKNOWLEDGMENT We gratefully acknowledge Cynthia P. Rausch for assistance with manuscript preparation.

15.

North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Eng I J Med. 1991; 325:445-453. Mayberg MR, Wilson SE, Yatsu F, et aI, Veterans Affairs Cooperative Studies Program 309 Trialist Group. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. JAMA. 1991;266:3289-3294. Barnett HJM, Taylor DW, Eliasziw M, et aI, North American Symptomatic Carotid Endarterectomy Trial Collaborators. Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. N Engl J Med. 1998;339:1415-1425. Zwiebel WJ. Doppler evaluation of carotid stenosis. In: Zwiebel WJ, ed. Introduction to Vascular Ultrasonography. 3rd ed. Philadelphia, Pa: WB Saunders Co; 1992:123-132. Strandness DE Jr. Duplex Scanning in Vascular Disorders. New York, NY: Raven Press; 1990:92-120. Huston J III, Lewis BD, Wiebers DO, Meyer FB, Riederer SJ, Weaver AL. Carotid artery: prospective blinded comparison of two-dimensional time-of-flight MR angiography with conventional angiography and duplex US. Radiology. 1993;186:339-344. Cartier R, Cartier P, Fontaine A. Carotid endarterectomy without angiography: the reliability of Doppler ultrasonography and duplex scanning in preoperative assessment. Can J Surg. 1993;36:411416. Dawson DL, Zierler RE, Strandness DE Jr, Clowes AW, Kohler TR. The role of duplex scanning and arteriography before carotid endarterectomy: a prospective study. J Vasc Surg. 1993;18:673680. Golledge J, Wright R, Pugh N, Lane IF. Colour-coded duplex assessment alone before carotid endarterectomy. Br J Surg. 1996; 83:1234-1237. Gortler M, Niethammer R, Widder B. Differentiating subtotal carotid artery stenoses from occlusions by colour-coded duplex sonography. J Neurol. 1994;241:301-305. Horn M, Michelini M, Greisler HP, Littooy FN, Baker WHo Carotid endarterectomy without arteriography: the preeminent role of the vascular laboratory. Ann Vasc Surg. 1994;8:221-224. Karacagil S, Bergqvist D. Reevaluation of duplex criteria for diagnosis of high-grade carotid artery stenosis: when do we need angiography and when can we operate without angiography? Int Angiol. 1995;14:410-414. Muto PM, Welch HJ, Mackey WC, O'Donnell TF. Evaluation of carotid artery stenosis: is duplex ultrasonography sufficient? J Vasc Surg. 1996;24:17-22. Nicolaides AN, Shifrin EG, Bradbury A, et al. Angiographic and duplex grading of internal carotid stenosis: can we overcome the confusion? J Endovasc Surg. 1996;3:158-165. Padayachee TS, Cox TC, Modaresi KB, Colchester AC, Taylor PRo The measurement of internal carotid artery stenosis: comparison of

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1140

Diagnostic Criteria for Carotid Artery Stenosis

duplex with digital subtraction angiography. Eur J Vase Endovase Surg.1997;13:180-185. 16. Turnipseed WD, Kennell TW, Turski PA, Acher CW, Hoch JR. Combined use of duplex imaging and magnetic resonance angiography for evaluation of patients with symptomatic ipsilateral high-grade carotid stenosis. J Vase Surg. 1993;17:832-839. 17. Young GR, Humphrey PR, Shaw MD, Nixon TE, Smith ET. Comparison of magnetic resonance angiography, duplex ultrasound, and digital subtraction angiography in assessment of extracranial internal carotid artery stenosis. J Neurol Neurosurg Psychiatry. 1994;57:1466-1478. 18. Moneta GL, Edwards JM, Chitwood RW, et al. Correlation of North American Symptomatic Carotid Endarterectomy Trial (NASCET) angiographic definition of70% to 99% internal carotid artery stenosis with duplex scanning. J Vase Surg. 1993;17:152157. 19. Browman MW, Cooperberg PL, Harrison PB, Marsh n, Mallek N. Duplex ultrasonography criteria for internal carotid stenosis of more than 70% diameter: angiographic correlation and receiver operating characteristic curve analysis. Can Assoe Radiol J. 1995; 46:291-295. 20. Faught WE, Mattos MA, van Bemme1en PS, et al. Color-flow duplex scanning of carotid arteries: new velocity criteria based on receiver operator characteristic analysis for threshold stenoses used in the symptomatic and asymptomatic carotid trials. J Vase Surg. .1994;19:818-827. 21. Neale ML, Chambers JL, Kelly AT, et al. Reappraisal of duplex criteria to assess significant carotid stenosis with special reference to reports from the North American Symptomatic Carotid Endarterectomy Trial and the European Carotid Surgery Trial. J Vase Surg. 1994;20:642-649. 22. Robinson ML, Sacks D, Perlmutter GS, Marinelli DL. Diagnostic criteria for carotid duplex sonography. AJR Am J Roentgenol. 1988;151:1045-1049. 23. Polak JF, Hunink MG, O'Leary DH. Doppler US for internal carotid stenosis of 70% or more: ROC analysis [abstract]. Radiology. 1991;181(suppl):133. 24. Hunink MG, Polak JF, Barlan MM, O'Leary DH. Detection and quantification of carotid artery stenosis: efficacy of various Doppler velocity parameters. AJR Am J Roentgenol. 1993;160:619-625. 25. Van Leersum M, Van Leeuwen MS, Van der Schouw Y, Mali WPTM, Eike1boom Be. New duplex threshold values for angiographically determined stenosis in the internal carotid artery in the light of the NASCET and ECST [abstract]. Cardiovase Intervent Radiol. 1995;18(suppll):S62.

Mayo Clin Proc, November 2000, Vol7S

26. 27. 28.

29. 30. 31. 32.

33. 34. 35. 36. 37.

38. 39.

Carpenter JP, Lexa FJ, Davis JT. Determination of duplex Doppler ultrasound criteria appropriate to the North American Symptomatic Carotid Endarterectomy Trial. Stroke. 1996;27:695-699. Hood DB, Mattos MA, Mansour A, et al. Prospective evaluation of new duplex criteria to identify 70% internal carotid artery stenosis. J Vase Surg. 1996;23:254-262. AbuRahma AF, Robinson PA, Strickler DL, Alberts S, Young L. Proposed new duplex classification for threshold stenoses used in various symptomatic and asymptomatic carotid endarterectomy trials. Ann Vase Surg. 1998;12:349-358. Grant EG, Duerinckx AJ, EI Saden SM, et al. Ability to use duplex US to quantify internal carotid arterial stenoses: fact or fiction? Radiology. 2000;214:247-252. Fillinger MF, Baker RJ Jr, Zwolak RM, et al. Carotid duplex criteria for a 60% or greater angiographic stenosis: variation according to equipment. J Vase Surg. 1996;24:856-864. Kuntz KM, Polak JF, Whittemore AD, Skillman JJ, Kent xc. Duplex ultrasound criteria for the identification of carotid stenosis should be laboratory specific. Stroke. 1997;28:597-602. van Everdingen KJ, van der Grond J, Kappelle LJ. Overestimation of a stenosis in the internal carotid artery by duplex sonography caused by an increase in volume flow. J Vase Surg. 1998;27:479485. O'Leary DH, Clouse ME, Potter JE, Wheeler HG. The influence of noninvasive tests on the selection of patients for carotid angiography. Stroke. 1985;16:264-267. Fox AJ. How to measure carotid stenosis [editorial]. Radiology. 1993;186:316-318. Earnest F IV, Forbes G, Sandok BA, et al. Complications of cerebral angiography: prospective assessment of risk. AJR Am J Roentgenol. 1984;142:247-253 . Busuttil SJ, Franklin DP, Youkey JR, Elmore JR. Carotid duplex overestimation of stenosis due to severe contralateral disease. Am J Surg. 1996;172:144-147. Mansour MA, Mattos MA, Hood DB, et al. Detection of total occlusion, string sign, and preocclusive stenosis of the internal carotid artery by color-flow duplex scanning. Am J Surg. 1995; 170:154-158. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982; 143:29-36. European Carotid Surgery Tria1ists' Collaborative Group. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70-99%) or with mild (0-29%) carotid stenosis. Lancet. 1991;337:1235-1243.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.