This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Howard, G. Articles by Toole, J. F. Search for Related Content PubMed PubMed Citation Articles by Howard, G. Articles by Toole, J. F.

(Stroke. 1996;27:1951-1957.)
© 1996 American Heart Association, Inc.

Articles

An Approach for the Use of Doppler Ultrasound as a Screening Tool for Hemodynamically Significant Stenosis (Despite Heterogeneity of Doppler Performance)

A Multicenter Experience

George Howard, DrPH; William H. Baker, MD; Lloyd E. Chambless, PhD; Virginia J. Howard, MSPH; Anne M. Jones, MS; James F. Toole, MD for the Asymptomatic Carotid Atherosclerosis Study Investigators

the Department of Public Health Sciences (G.H., J.F.T.) and the Stroke Center (G.H., V.J.H., A.M.J., J.F.T.), Bowman Gray School of Medicine of Wake Forest University, Winston-Salem, NC; Section of Vascular Surgery, Loyola University Medical School, Maywood, Ill (W.H.B.); and the Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC (L.E.C.).

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Background and Purpose The Asymptomatic Carotid Atherosclerosis Study (ACAS) Doppler validation study assessed the performance of individual Doppler machines across a spectrum of laboratories. We attempted to establish a threshold specific to individual machines to predict angiographically defined hemodynamic stenosis. The reliability of these Doppler ultrasound criteria was prospectively and independently evaluated among patients screened with ultrasound in the ACAS trial.

Methods Regression techniques were used to establish the relationship between Doppler velocity and percent stenosis by angiography for 63 specific Doppler machines. This relationship was used to establish a Doppler threshold to provide a 90% positive predictive value (PPV) of a 60% stenosis by angiography. The sensitivity of each Doppler machine to detect a 60% stenosis (at the 90% PPV threshold) was estimated. The efficacy of these Doppler thresholds was then prospectively evaluated by calculating the PPV among ACAS participants eligible by ultrasound.

Results Of the 63 machines, 13 (21%) had an excellent sensitivity (80%+) at 90% PPV. In 32 devices (51%) only a marginal sensitivity (50% to 80%) could be achieved. In 9 devices (14%) the sensitivity was poor (0% to 50%), and in 9 (14%) no threshold could be established. Despite the heterogeneity of Doppler performance, the standardization program worked as designed in the ACAS trial. Of 825 surgical patients, 399 were eligible by Doppler and 395 subsequently underwent angiography. Of these, 32 (8.1%; 95% confidence interval, 5.4% to 10.8%) did not have hemodynamically significant stenosis by arteriography, a proportion nonsignificantly lower than the planned 10% by the PPV.

Conclusions The performance of Doppler ultrasound was highly variable. This suggests that Doppler performance is likely overstated in the literature, but specific devices may perform satisfactorily to detect individuals with hemodynamically significant stenosis. Because performance differs substantially among devices, local investigators are strongly urged to maintain local standardization series. With such standardization, ultrasound performance is sufficient for admission to clinical trials and as the basis for carotid surgery. However, without quality control many ultrasound machines are not adequate to accurately predict the degree of carotid stenosis and should not be the only test to decide whether surgery is warranted.

Key Words: angiography • carotid stenosis • clinical trials • ultrasonics

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
After over a decade of widespread clinical use, carotid ultrasonography has recently become a subject of considerable debate.¹ Recent articles extol the virtues of ultrasonography as the most predictive noninvasive screening test for carotid atherosclerosis,² while other investigators greatly discount the value of ultrasonography.³ Despite this debate, carotid ultrasonography continues in widespread clinical use, and it has been embraced as a measure of systemic atherosclerosis by clinical trialists^{4 5 6} and epidemiologists.^{7 8 9 10 11} It is critical to understand the basis for these divergent opinions so that carotid duplex ultrasonography can be applied to future clinical studies and in everyday patient care.

The divergence of the ultrasound literature may reflect heterogeneous ultrasound performance among laboratories. Blood flow is more rapid through the stenotic portion of a carotid artery. Doppler ultrasound provides measurement of the frequency shift of sound returning from the moving blood, where more rapidly flowing blood provides a greater frequency shift. We did not assume that the performance of Doppler devices with the same manufacturer and model number would perform similarly; rather, we evaluated the performance of each specific Doppler machine individually. Hence, when we refer to a "Doppler device" or a "device-specific threshold," we are referring to a specific physical machine in a particular laboratory. A hemodynamically significant stenosis (60% diameter reduction) was an eligibility criterion for the ACAS.¹² By protocol, the presence of such a stenosis could be established by noninvasive testing (including ultrasonography) or angiography. As part of the study-wide quality-control efforts, a validation study was conducted in which each center was required to validate its local ultrasonography against angiography. This effort produced correlative data on ultrasonography and angiography on 63 separate Doppler ultrasound devices used in the ACAS trial. In the first part of this report, we describe the distribution of performance for these 63 devices and update a previous article assessing the performance of 30 Doppler ultrasound devices.¹³

It is critical to determine whether carotid ultrasonography can be effectively used as a screening test for the presence of hemodynamically significant stenosis in clinical trials or for surgery. It would have been ruinous to the ACAS (or other clinical trials) to have eligibility criteria that admitted patients who after randomization were found not to have hemodynamically significant stenosis by angiography. Because of the widely accepted clinical-trial principle of intention-to-treat, the outcome of such patients randomized to surgery would have to be attributed to the surgical treatment, even if the extent of carotid atherosclerosis was not sufficient to permit the operation. Also, patients would be potentially harmed if ultrasonography (without angiography) were used to establish the presence of hemodynamically significant lesions for surgery, but at surgery many patients were found not to have surgical lesions. As such, the primary goal of the Doppler ultrasound eligibility criteria should be to have a high PPV, ie, the probability of there truly being a hemodynamically significant stenosis by angiography among those participants (or patients) in whom the Doppler ultrasound test was positive. By using the results from the validation process, device-specific criteria were set for admission to the ACAS trial. In the second part of this report, we examine whether these criteria performed as designed, ie, whether they successfully identified patients with hemodynamically significant stenosis for admission to the trial.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
This report has two goals: (1) evaluating the performance of Doppler ultrasound across clinics and (2) evaluating the performance of Doppler thresholds to predict hemodynamically significant stenosis. The methods are different and are addressed using data from different sources.

Performance of Doppler Ultrasound Across Clinics
This report provides data on the performance of 63 Doppler devices from 37 ACAS centers and updates a previous report describing the performance of 30 Doppler devices from 19 ACAS centers.¹³ Each of the ACAS centers was required to submit data from approximately 50 patients with both a Doppler examination and angiography conducted within 6 weeks of the Doppler examination. To ensure consistency in these evaluations, study-wide Doppler quality-control measures were adopted including (1) certification of sonographers, (2) quality-control overreading of Doppler ultrasound videotapes and/or studies, (3) certification of angiogram readers, (4) selection of only cut-sheet and selective intra-arterial digital subtraction studies, and (5) adoption of a standard angiogram-reading protocol. The primary Doppler examination measures were the peak systolic frequency/velocity and end diastolic frequency/velocity, measured in either centimeters per second or hertz (as determined by Doppler ultrasound device). The primary measure from the angiogram was the percent diameter stenosis. Details of these are provided in the previous report.¹³

Linear regression was used to model the relationship between Doppler velocity/frequency and percent stenosis by angiography. The residuals of each model were examined to ensure adequacy of the regression model, and a variance stabilization weighting was assigned where required. A fourth-degree polynomial (PS=ß₀+ß₁DF+ß₂DF²+ß₃DF³+ß₄DF⁴) adequately described the relationship between percent stenosis by angiography (PS) and Doppler frequency (DF). Once the model was established, a Doppler threshold associated with a 90% PPV of a 60% stenosis by angiography was determined; that is, the threshold was defined as the flow velocity or frequency associated with a 90% probability of an angiogram showing a 60% stenosis. The sensitivity of the Doppler at that threshold was determined as the percent of arteries with 60% stenosis by angiography with Doppler results greater than the threshold.

Performance of Doppler Thresholds for Hemodynamically Significant Stenosis
The performance of the Doppler threshold criteria established in the validation study was evaluated independently and prospectively in the ACAS study. Eligibility for ACAS required evidence of a hemodynamically significant stenosis, which could be documented by either angiography or noninvasive testing (primarily the Doppler thresholds established in the validation study). Additional eligibility criteria included cerebrovascular asymptomatic status, age between 40 and 79 years, performance of required laboratory and electrocardiographic examinations no earlier than 3 months before randomization, patient accessibility and willingness to be followed up for 5 years, and valid informed consent. This resulted in a population with an average age of 67 years that was 66% male and 95% white. Of those eligible for the ACAS by noninvasive testing, approximately one half were randomized to the surgical arm. Because all surgical patients were required to have confirmed noninvasive results by angiography before surgery, a direct measurement of the percent stenosis by arteriogram followed qualification by Doppler ultrasound. This offers a direct prospective estimate of the PPV of Doppler ultrasound using the device-specific thresholds established in the validation study, estimated as the proportion of patients qualifying by Doppler ultrasound who were found to have 60% stenosis by angiography. If the Doppler thresholds defined by the validation study standardization process work as designed, 10% of those subjects qualifying by Doppler ultrasound should be found to have stenosis <60% by angiography (1-PPV). Because those participants not qualifying by ultrasonography were not similarly systematically evaluated by angiography, sensitivity, specificity, and NPV of Doppler ultrasound cannot be estimated prospectively.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Validity and Reliability of Doppler Ultrasound Across Clinics
The performance of Doppler ultrasound was remarkably heterogeneous. For some devices (approximately 20%), the relationship between percent stenosis and Doppler frequency/velocity fell "nicely" about a polynomial regression line (Fig 1, left). However, for a larger number of devices (approximately 60%) the relationship was not as consistent (Fig 1, middle), and for a few devices (approximately 20%) there was no statistical relationship between Doppler frequency/velocity and percent stenosis (Fig 1, right). When no statistical relationship was observed, there was of course no Doppler threshold.

View larger version (42K): [in this window] [in a new window]   Figure 1. Relationship between Doppler frequency/velocity and percent stenosis by angiography for three specific devices: one with a device with a "strong" relationship (left), one with a "moderate" relationship (middle), and one with a "poor" relationship (right).

The Table provides a summary of the thresholds and associated sensitivities by center and device. The heterogeneity of the strength of the relationship between Doppler frequency/velocity and percent stenosis as evaluated by angiography is apparent in the dispersion of sensitivities (plotted in Fig 2). Of the 63 devices, 13 (21%) had an excellent sensitivity (80%+) at 90% PPV. In 32 devices (51%), only a marginal sensitivity (50% to 80%) could be achieved. In 9 devices (14%), the sensitivity was poor (0% to 50%), and in 9 (14%) no threshold could be established. Also showing remarkable heterogeneity was the threshold value associated with the 90% PPV that was determined in centimeters per second (Fig 3, top) or hertz (Fig 3, bottom).

View this table: [in this window] [in a new window]   Table 1. Descriptive Statistics by Device

View larger version (44K): [in this window] [in a new window]   Figure 2. Distribution of sensitivity of 63 ultrasound devices at 90% PPV standardized against a 60% stenosis by angiography.

View larger version (113K): [in this window] [in a new window]   Figure 3. Top, Threshold of 24 Doppler devices reporting flow in centimeters per second; bottom, threshold of 30 Doppler devices reporting flow in hertz shift (3k to 4k=3001 Hz to 4000 Hz).

Performance of Doppler Thresholds for Hemodynamically Significant Stenosis
Of the 1662 ACAS participants, 825 were randomized to surgery.¹⁴ Of the 825 surgical patients, presence of a hemodynamically significant stenosis was documented by Doppler in 411 patients (the remaining 414 patients were qualified primarily by angiography). The eligibility of these 411 patients was established using 37 of the 63 devices (Table). One device, which was used to randomize 12 patients, was standardized versus another Doppler (not versus angiography), and these patients were omitted from these analyses. The other 26 devices were not used in the randomization process for several reasons: (1) their performance was not acceptable by ACAS standards, (2) the device-specific criterion was set after randomization was closed, and (3) the device was replaced by newer equipment before the documentation of eligible patients. Angiography results were not available in 4 patients: 1 who had a stroke during angiography and 3 who underwent angiography but whose results were declared unavailable by the center. The remaining 395 patients subsequently underwent angiography before their surgery where the percent stenosis was determined. Of these, 363 patients (91.9%) proved to have hemodynamically significant stenosis >60% by angiography. The 95% confidence limits of the estimated 91.9% range from 89.2% to 94.6%, which includes the planned 90%. Hence, the performance of the Doppler in the main ACAS trial was slightly better than the predicted 90% PPV.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Despite the remarkable heterogeneity of Doppler performance, the system to establish Doppler criteria in the validation study worked as designed in the ACAS trial. In fact, one could argue that the average performance of Doppler ultrasound in the ACAS was marginal at best; however, the ACAS approach to the use of Doppler ultrasound in the study overcame these potential shortcomings and allowed Doppler ultrasound to make a valuable contribution to the study. The observed PPV based on the independent prospective series of patients admitted to the ACAS was 91.9%, slightly above the planned 90% threshold. These efforts show that a system can be designed to allow Doppler ultrasound to be used successfully to screen for hemodynamically significant stenosis of the carotid system. However, heterogeneity of performance implies that a considerable quality-control effort must be made, and more important, the criteria set must be specific to particular Doppler ultrasound devices. As such, previously proposed criteria to identify stenosis^{15 16 17 18 19 20 21} are likely appropriate and valid in the laboratory that is the source of the report, but they may not be valid elsewhere. The variation in thresholds within a particular device is surprising. ACAS investigators decided to not introduce a "tight" ultrasound protocol, since this could potentially disrupt the established operation of the ultrasound laboratory. As such, information was not systematically collected to account for these large observed differences in thresholds for similar machines. However, these differences could be introduced by differences in the angle of insonation and other sonographer-dependent factors (when we refer to an ultrasound device, we refer to a specific machine in the hands of a particular staff). These results show the absolute necessity of each noninvasive laboratory maintaining device-specific standardization statistics.

That a PPV of 90% for Doppler ultrasound prediction of hemodynamically significant stenoses was established in the validation study, and confirmed independently and prospectively in the main ACAS trial, again raises the question of whether surgery should be considered on the basis of Doppler ultrasound results without confirmation by angiography. A comparison of predicted events and unneeded operations under the two strategies illustrates the trade-offs (Fig 4). In the ACAS, the angiographic morbidity/mortality was 1.2% (4 strokes and 1 death among 414 patients who underwent angiography), and the surgical morbidity/mortality was 1.5% (10 strokes and 1 death among 724 surgery patients). Consider a hypothetical cohort of 100 patients. The current management strategy would indicate that all 100 should undergo angiography. During angiography, an average of 1.2 patients would have a stroke or die, but 8.1 patients ([1-PPV]*100) would be identified as not having a hemodynamically significant lesion. The remaining 90.7 patients would proceed to surgery, where 1.4 patients would have an event and 89.3 would be effectively treated and benefit from the risk reduction associated with endarterectomy. Thus, the current strategy would result in a total of 2.6 events and 89.3 patients effectively treated. In contrast, if angiography were not performed, then all 100 patients with a positive by Doppler ultrasound examination would continue on to surgery. Assuming the same surgical event rate, 1.5 of these patients would experience an event, 8.1 would undergo surgery to discover a <60% stenosis, and 90.4 patients would be effectively treated. Removing angiography thus reduces the expected number of events from 2.6 to 1.5, a 42% reduction; however, 8.1 patients would undergo surgery to find <60% stenosis ("needless" surgeries). While it is difficult to compare the relative anxiety of angiography versus surgery, these 8.1 "needless" operations would avoid angiography in 100 subjects. If fiscal costs of surgery are less than 12.3 times (100/8.1) greater than angiography, the proposed management system would reduce costs, as well as events. In addition, among the 8.1 patients in whom a "needless" operation was performed (one in which a <60% stenosis was present), if a stenosis of <60% is discovered and removed, it is unlikely to be harmful, and could be helpful, for the long-term prognosis. Finally, some patients may in truth have stenoses >60% that are shown falsely by angiography to be <60%. Removal of such plaques would likely be beneficial, and these patients would be inappropriately treated under the current management schema.

View larger version (31K): [in this window] [in a new window]   Figure 4. Comparison of anticipated complication rates of endarterectomy with and without angiography.

It is possible to argue that the surgical complication rate would be higher without angiography because of intracranial anatomy that is not visible by ultrasound. For simplicity, consider there to be only two types of patients: those with and those without anatomic problems observable only by angiography. Assume that the complication rate in those without anatomic problems is the 1.5% observed in the ACAS and from (unpublished) ACAS data that 5% of the population has anatomic problems revealed only by angiography. The complication rate would have to be 22.9% among those with such anatomic problems for the overall event rate of the proposed management strategy (without angiography) to equal the 2.6% rate in the current management strategy. The ACAS required all patients randomized to surgery to obtain an angiogram and endarterectomy was not performed in patients with ipsilateral intracranial disease. As such, we cannot offer direct data to address the extent of the likely increase in mortality in patients with intracranial disease. Many physicians inherently believe that surgical stroke and death are likely to be higher in patients with ipsilateral intracranial stenosis; however, this belief remains unsupported by data. Subsequent ACAS publications will address the impact on mortality of contralateral intracranial disease. Additional concerns would be introduced concerning carotid arteries falsely declared to be occluded (when in fact they were open) and by those falsely declared patent when in truth they were occluded. Patients with those falsely declared occluded would likely not proceed to angiography under the current treatment strategy; hence, they would be treated identically under either management approach. Those patients with arteries thought to be patent by ultrasound that were occluded would undergo surgery with uncertain (and likely negative) outcome. However, occlusion of carotid arteries has a relatively low prevalence, and this would be unlikely to dramatically affect the average outcome of a patient. Finally, the use of Doppler ultrasound without confirmation by angiography would be practical only for devices in which a 90% PPV can be established with an associated high sensitivity. In other laboratories, confirmation of carotid atherosclerosis would have to be provided by angiography.

The success of the ACAS approach in obtaining a high PPV for Doppler ultrasound implies a relatively low sensitivity for many Doppler ultrasound devices. Thus, there is likely a substantial population with Doppler examinations "suggestive" of hemodynamically significant stenoses but not reaching the high thresholds set by the validation study. Clearly, angiography is required to confirm or eliminate the possibility of stenoses in these patients. Finally, specificity (the ability of a device to accurately identify patients without hemodynamically significant stenosis) needs to be addressed. Should patients be denied further workup and surgical therapy on the basis of a Doppler ultrasound value below the estimated threshold? While others have addressed the specificity of Doppler ultrasound,^{15 16 17 18 19 20 21} our study design does not offer the opportunity to address this issue prospectively. We feel that individual laboratories could set a lower Doppler threshold based on the NPV, below which a patient is reasonably assured of not having a hemodynamically significant stenosis. The clinical decision-making process associated with the results of ultrasound tests would then be one of three choices determined by two thresholds: (1) if the Doppler frequency/velocity is below the lower threshold, exclude the patient from further evaluations; (2) if it is between the two thresholds, refer the patient to angiography; and (3) if it is above the higher threshold, refer the patient to surgery without angiography. Without the establishment of the lower threshold based on a 90% NPV (option 1 above), all patients who may benefit from surgery but whose Doppler values fall below the upper threshold should be considered for angiography. A similar approach could be incorporated in the serial follow-up of patients with asymptomatic stenosis, where the progression of patients could be monitored by comparison with the upper threshold value.

It is natural to attempt to use these data to determine the "best" Doppler ultrasound device or to recommend thresholds for a particular device. However, when the data were analyzed by specific device maker or device type, no consistent patterns appeared. There were four manufacturers with at least eight devices used in the study: Acuson (n=10), ATL (n=16), Biosound (n=10), and Diasonics (n=8). Each of these manufacturers had at least one device in which a threshold could not be established and at least one device with a sensitivity >85%, and there was no difference (P=.6) in the mean sensitivity level by manufacturer. Furthermore, there was not a consistent threshold for any manufacturer. These results suggest that the sonographer and/or scanning protocol (dependent on local standards) are the primary contributors to "good" performance and that it is difficult to generalize performance even among similar devices.

The present article and our previous study¹³ are (to our knowledge) the only reports describing the distribution of Doppler performance across a spectrum of laboratories. As such, these results also offer important insights into the apparent divergence of results concerning the predictive value of ultrasonography in screening for significant carotid stenosis. The recent meta-analysis of Blakeley and coworkers² regarding alternative methods for noninvasive criteria for establishing carotid stenosis concluded that duplex scanning is the most reliable available. Others have argued that such high sensitivity and specificity are sufficient to perform surgery after ultrasonography without confirmation of stenosis by angiography.^{22 23 24 25 26 27 28 29 30} This is in stark contrast to the recent report of Eliasziw and coworkers,³ who concluded from ultrasound/angiographic relationships from NASCET that ultrasound was only sufficient to exclude patients without disease and that angiography is still required to confirm the presence of disease. Our results can be used to understand these inconsistent results. We have suggested that there may be a considerable publication bias in reports describing the sensitivity and specificity of Doppler ultrasound that is a direct result of its heterogeneous performance.³¹ This publication bias would naturally lead to the positive reports of Blakeley and coworkers.² If our 63 evaluations relating ultrasound to angiography were performed as independent investigations, the 13 comparisons (21%) with excellent sensitivity would be likely to be submitted and accepted for publication. In the 50 comparisons (79%) with moderate, poor, or no correlation of ultrasound and angiography, investigators would be unlikely to submit results, and journals would be even less likely to publish results. As such, the literature is likely to substantially overestimate the "average" performance of ultrasound. Our multicenter, but device-specific, approach provides a more realistic understanding of the Doppler ultrasound performance. ACAS centers were selected partially because of the presence of an established and well-recognized ultrasound laboratory. As such, our results may also still overestimate the performance of ultrasound in the general medical community.

Our results may also help explain the relatively poor relationship between ultrasound and angiography presented by Eliasziw and coworkers.³ Although they do not cite the number of specific ultrasound devices, the data are from the 50 NASCET sites and likely from many devices. Our results suggest two dramatic shortcomings of the report by Eliasziw and coworkers. First, no quality-control efforts were discussed that would identify the likely subset of devices with poor performance that should be excluded (estimated to be 14% of devices by our data). Second, Eliasziw and coworkers assumed that all devices performed identically and applied a common threshold to all devices. This fails to reflect that devices in different hands require different device-specific criteria. Eliasziw et al concluded that Doppler ultrasound was more reliable in detecting the absence than the presence of disease. This result follows from their use of a common threshold. Patients with a low (<6 kHz) Doppler shift will be correctly identified by all machines as not having hemodynamically significant stenosis. However, as the frequency shift increases, the pooled reliability of Doppler ultrasound with a common threshold will decrease as the machines differentially enter the range indicating a hemodynamically significant stenosis. We hypothesize that increased use of quality control and additional efforts to establish device-specific thresholds would have led NASCET to a more successful use of ultrasound.

There are a number of shortcomings in our report. First, the thresholds established here appear appropriate for asymptomatic subjects who are at relatively low risk. However, in higher risk populations, such as symptomatic subjects, these thresholds may need to be adjusted to provide a higher level of protection from potentially missing patients who may greatly benefit from intervention. Also as noted above, the ACAS centers (and their associated patient populations) may not be representative of "general" ultrasound laboratories. Because an established and recognized ultrasound laboratory was a criterion for a center's participation in the ACAS, we caution the reader that our results may be biased to reflect a better-than-average Doppler performance. Finally, lower device-specific thresholds may be established by increasing the sample size of the validation series beyond 50 patients; however, review of these scatters of points for specific machines suggests that this improvement is likely minor.

In conclusion, while the sensitivity and specificity of Doppler ultrasound are likely overstated in the literature, the performance on individual devices can be sufficient to detect individuals with hemodynamically significant stenosis. However, because performance differs substantially across devices, local investigators and caregivers are strongly urged to maintain local standardization series. Technical details of the validation process are available from the ACAS operations office. With such standardization, ultrasound performance is clearly sufficient for admission to clinical trials and as the basis for carotid surgery. However, without this quality control, our data indicate that many ultrasound machines are not adequate to accurately predict the degree of carotid stenosis and should not be the only test used to decide whether surgery is warranted.

	Selected Abbreviations and Acronyms

 

ACAS = Asymptomatic Carotid Atherosclerosis Study NASCET = North American Symptomatic Carotid Endarterectomy Trial NPV = negative predictive value PPV = positive predictive value

	Acknowledgments

 
This study was supported by grant R01 NS22611 from the US Public Health Service National Institute of Neurological Disorders and Stroke. The authors acknowledge the support of all participants in the Asymptomatic Carotid Atherosclerosis Study.

	Footnotes

 
Reprint requests to Dr G. Howard, c/o ACAS Operations Center, Stroke Research Center, Bowman Gray School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157-1068. E-mail vjhoward@bgsm.edu.

Received March 18, 1996; revision received July 22, 1996; accepted July 24, 1996.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Toole JF, Castaldo JE. Accurate measurement of carotid stenosis. J Neuroimaging. 1994;4:222-230.[Medline] [Order article via Infotrieve]

2. Blakeley DD, Oddone EZ, Hasselbland V, Simel DL, Matchar D. Noninvasive carotid artery testing: a meta-analytic review. Ann Intern Med. 1995;122:360-367.[Abstract/Free Full Text]

3. Eliasziw M, Rankin RN, Fox AJ, Haynes RB, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial Group. Accuracy and prognostic consequences of ultrasonography in identifying severe carotid artery stenosis. Stroke. 1995;26:1747-1752.[Abstract/Free Full Text]

4. Asymptomatic Carotid Artery Plaque Study Group. Rationale and design for the Asymptomatic Carotid Artery Plaque Study (ACAPS). Control Clin Trials. 1992;13:293-314.[Medline] [Order article via Infotrieve]

5. Crouse JF, Byington RP, Bond MG, Espeland MA, Sprinkle JW, McGovern M, Furberg CD. Pravastatin, lipids, and atherosclerosis in the carotid arteries: design features of a clinical trial with carotid atherosclerosis outcome. Control Clin Trials. 1992;13:495-506.[Medline] [Order article via Infotrieve]

6. Mack WJ, Selzer RH, Hodis HN, Erickson JK, Liu CR, Liu CH, Crawford DW, Blankenhorn DH. One-year reduction and longitudinal analysis of carotid intima-media thickness associated with colestipol/niacin therapy. Stroke. 1993;24:1779-1783.[Abstract/Free Full Text]

7. Bond MG, Barnes RW, Riley WA, Wilmoth SK, Chambless LE, Howard G, Owens B, for the ARIC Study Group. High-resolution B-mode ultrasound scanning methods in the Atherosclerosis Risk in Communities (ARIC) Study. J Neuroimaging. 1991;1:68-73.[Medline] [Order article via Infotrieve]

8. Bots ML, Hofman A, de Bruyn AM, de Jong PTVM, Grobbee DE. Isolated systolic hypertension and vessel wall thickness of the carotid artery: the Rotterdam Elderly Study. Arterioscler Thromb. 1993;13:64-69.[Abstract/Free Full Text]

9. O'Leary DH, Polak JF, Wolfson SK Jr, Bond MG, Bommer W, Sheth S, Psaty BM, Sharrett AR, Manolio TA, on behalf of the Cardiovascular Health Study Collaborative Research Group. Use of sonography to evaluate carotid atherosclerosis in the elderly: the Cardiovascular Health Study. Stroke. 1991;22:1155-1163.[Abstract/Free Full Text]

10. Riley WA, Barnes RW, Bond MG, Evans GW, Chambless LE, Heiss G. High-resolution B-mode ultrasound reading methods in the Atherosclerosis Risk in Communities (ARIC) cohort. J Neuroimaging. 1991;1:168-172.[Medline] [Order article via Infotrieve]

11. Salonen R, Salonen JT. Determinants of carotid intima-media thickness: a population-based ultrasonography study in Eastern Finnish men. J Intern Med. 1991;229:225-231.[Medline] [Order article via Infotrieve]

12. Asymptomatic Carotid Atherosclerosis Study Group. Study design for randomized prospective trial of carotid endarterectomy for asymptomatic atherosclerosis. Stroke. 1989;20:844-849.[Abstract/Free Full Text]

13. Howard G, Chambless LE, Baker WH, Ricotta JJ, Jones AM, O'Leary D, Howard VJ, Elliott TJ, Lefkowitz DS, Toole JF. A multicenter validation study of Doppler ultrasound versus angiography. J Stroke Cerebrovasc Dis. 1991;1:166-173.

14. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA. 1995;273:1421-1428.[Abstract/Free Full Text]

15. Blackshear WM, Phillips DJ, Chikos PM, Harley JD, Thiele BL, Strandness DE. Carotid artery velocity patterns in normal and stenotic vessels. Stroke. 1980;11:67-71.[Abstract/Free Full Text]

16. Robinson ML, Sacks D, Perlmutter GS, Marinelli DL. Diagnostic criteria for carotid duplex sonography. AJR Am J Roentgenol. 1988;151:1045-1049.[Abstract/Free Full Text]

17. Ranke C, Creutiz A, Alexander K. Duplex scanning of the peripheral arteries: correlation of the peak velocity ratio with angiographic diameter reduction. Ultrasound Med Biol. 1992;18:433-440.[Medline] [Order article via Infotrieve]

18. Hunink MGM, Polak JF, Barlan MM, O'Leary HD. Detection and quantification of carotid artery stenosis: efficacy of various Doppler velocity parameters. AJR Am J Roentgenol. 1993;160:619-625.[Abstract/Free Full Text]

19. Moneta GL, Edwards JM, Chitwood RW, Taylor LM, Lee RW, Cummings CA, Porter JM. Correlation of North American Symptomatic Carotid Endarterectomy Trial (NASCET) angiographic definition of 70% to 99% internal carotid artery stenosis with duplex scanning. J Vasc Surg. 1993;17:152-159.[Medline] [Order article via Infotrieve]

20. Faught WE, Mattos MA, van Bemmelen PS, Hodgson KJ, Barkmeier LD, Ramsey DE, Sumner DS. Color-flow duplex scanning of carotid arteries: new velocity criteria based on receiver operator characteristic analysis for threshold stenosis used in the symptomatic and asymptomatic carotid trials. J Vasc Surg. 1994;19:818-828.[Medline] [Order article via Infotrieve]

21. Neale ML, Chambers JL, Kelly AT, Connard S, Lawton MA, Roche J, Appleberg M. 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 Vasc Surg. 1994;20:642-649.[Medline] [Order article via Infotrieve]

22. Geuder JW, Lamparello PJ, Riles TS, Giangola G, Imparato AM. Is duplex scanning sufficient evaluation before carotid endarterectomy? J Vasc Surg. 1989;9:193-201.[Medline] [Order article via Infotrieve]

23. Kent KC, Kuntz KM, Patel MR, Kim D, Klufas RA, Whittemore AD, Polak J, Skillman JJ, Edelman RR. Perioperative imaging strategies for carotid endarterectomy: an analysis of morbidity and cost-effectiveness in symptomatic patients. JAMA. 1995;274:888-893.[Abstract/Free Full Text]

24. Dawson DL, Zierler E, Strandness E Jr, Clowes AW, Kohler TR. The role of duplex scanning and arteriography before carotid endarterectomy: a prospective study. J Vasc Surg. 1993;18:673-683.[Medline] [Order article via Infotrieve]

25. Goodson SF, Flanigan P, Bishara RA, Kikta MJ, Meyer JP. Can carotid duplex scanning supplant arteriography in patients with focal carotid territory symptoms? J Vasc Surg. 1987;5:551-557.[Medline] [Order article via Infotrieve]

26. McKittrick JE, Cisek PL, Pojunas KW, Blum GM, Ortgiesen P, Lim RA. Are both color-flow duplex scanning and cerebral arteriography required prior to carotid endarterectomy? Ann Vasc Surg. 1993;7:311-316.[Medline] [Order article via Infotrieve]

27. Ricotta JJ, Holen J, Schenk E, Plassche W, Green RM, Gramiak R, DeWeese JA. Is routine angiography necessary prior to carotid endarterectomy? J Vasc Surg. 1984;1:96-102.[Medline] [Order article via Infotrieve]

28. Moore WS, Ziomek S, Quinones-Baldrick WJ, Machleder HI, Busuttil RW, Baker JD. Can clinical evaluation and noninvasive testing substitute for arteriography in the evaluation of carotid artery disease? Ann Surg. 1988;208:91-94.[Medline] [Order article via Infotrieve]

29. Marshall WG, Kouchoukos NT, Murphy SF, Pelate C. Carotid endarterectomy based on duplex scanning without preoperative arteriography. Circulation. 1988;78(pt 2):I-1-I-5.

30. Polak JF, Kalina P, Donaldson MC, O'Leary DH, Whittemore AD, Mannick JA. Carotid endarterectomy: preoperative evaluation of candidates with combined Doppler sonography and MR angiography. Radiology. 1993;186:333-338.[Abstract/Free Full Text]

31. Howard G, Toole JF. Noninvasive carotid artery testing. Ann Int Med. 1995;123:663. Letter.

This article has been cited by other articles:

				M. H. Criqui, M. J. Alberts, F. G. R. Fowkes, A. T. Hirsch, P. T. O'Gara, J. W. Olin, and for Writing Group 2 Atherosclerotic Peripheral Vascular Disease Symposium II: Screening for Atherosclerotic Vascular Diseases:: Should Nationwide Programs Be Instituted? Circulation, December 16, 2008; 118(25): 2830 - 2836. [Full Text] [PDF]

				E. R. Bates, C. J. D. Babb, D. E. Casey, C. U. Cates, G. R. Duckwiler, T. E. Feldman, W. A. Gray, K. Ouriel, E. D. Peterson, K. Rosenfield, et al. ACCF/SCAI/SVMB/SIR/ASITN 2007 Clinical Expert Consensus Document on Carotid Stenting: A Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents (ACCF/SCAI/SVMB/SIR/ASITN Clinical Expert Consensus Document Committee on Carotid Stenting) Vascular Medicine, February 1, 2007; 12(1): 35 - 83. [PDF]

				American Society of Interventional & Therapeutic N, Society for Cardiovascular Angiography and Interve, Society for Vascular Medicine and Biology, Society of Interventional Radiology, E. R. Bates, J. D. Babb, D. E. Casey Jr, C. U. Cates, G. R. Duckwiler, T. E. Feldman, et al. ACCF/SCAI/SVMB/SIR/ASITN 2007 Clinical Expert Consensus Document on Carotid Stenting: A Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents (ACCF/SCAI/SVMB/SIR/ASITN Clinical Expert Consensus Document Committee on Carotid Stenting) J. Am. Coll. Cardiol., January 2, 2007; 49(1): 126 - 170. [Full Text] [PDF]

				M. Berg, Z. Zhang, A. Ikonen, P. Sipola, R. Kalviainen, H. Manninen, and R. Vanninen Multi-Detector Row CT Angiography in the Assessment of Carotid Artery Disease in Symptomatic Patients: Comparison with Rotational Angiography and Digital Subtraction Angiography AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1022 - 1034. [Abstract] [Full Text] [PDF]

				M. Kamouchi, K. Kishikawa, Y. Okada, T. Inoue, S. Ibayashi, and M. Iida Reappraisal of Flow Velocity Ratio in Common Carotid Artery to Predict Hemodynamic Change in Carotid Stenosis AJNR Am. J. Neuroradiol., April 1, 2005; 26(4): 957 - 962. [Abstract] [Full Text] [PDF]

				J. Kennedy, H. Quan, W. A. Ghali, and T. E. Feasby Importance of the imaging modality in decision making about carotid endarterectomy Neurology, March 23, 2004; 62(6): 901 - 904. [Abstract] [Full Text] [PDF]

				E. G. Grant, C. B. Benson, G. L. Moneta, A. V. Alexandrov, J. D. Baker, E. I. Bluth, B. A. Carroll, M. Eliasziw, J. Gocke, B. S. Hertzberg, et al. Carotid Artery Stenosis: Gray-Scale and Doppler US Diagnosis--Society of Radiologists in Ultrasound Consensus Conference Radiology, November 1, 2003; 229(2): 340 - 346. [Abstract] [Full Text] [PDF]

				G. Young, P. Humphrey, D. C. C. Johnston, and L. B. Goldstein Clinical carotid endarterectomy decision-making: Noninvasive vascular imaging versus angiography Neurology, December 26, 2001; 57(12): 2324 - 2324. [Full Text] [PDF]

				V. S. Lee, B. S. Hertzberg, M. J. Workman, T. P. Smith, M. A. Kliewer, D. M. DeLong, and B. A. Carroll Variability of Doppler US Measurements along the Common Carotid Artery: Effects on Estimates of Internal Carotid Arterial Stenosis in Patients with Angiographically Proved Disease Radiology, February 1, 2000; 214(2): 387 - 392. [Abstract] [Full Text]

				V. S. Lee, B. S. Hertzberg, M. A. Kliewer, and B. A. Carroll Assessment of Stenosis: Implications of Variability of Doppler Measurements in Normal-appearing Carotid Arteries Radiology, August 1, 1999; 212(2): 493 - 498. [Abstract] [Full Text]

				H. P. Adams Jr., B. H. Bendixen, E. Leira, K. C. Chang, P. H. Davis, R. F. Woolson, W. R. Clarke, and M. D. Hansen Antithrombotic treatment of ischemic stroke among patients with occlusion or severe stenosis of the internal carotid artery: A report of the Trial of Org 10172 in Acute Stroke Treatment (TOAST) Neurology, July 1, 1999; 53(1): 122 - 122. [Abstract] [Full Text]

				P. B. Gorelick, R. L. Sacco, D. B. Smith, M. Alberts, L. Mustone-Alexander, D. Rader, J. L. Ross, E. Raps, M. N. Ozer, L. M. Brass, et al. Prevention of a First Stroke: A Review of Guidelines and a Multidisciplinary Consensus Statement From the National Stroke Association JAMA, March 24, 1999; 281(12): 1112 - 1120. [Abstract] [Full Text] [PDF]

				S. S. Kang and W. H. Baker Current Technique for Carotid Endarterectomy Perspectives in Vascular Surgery and Endovascular Therapy, January 1, 1999; 10(2): 79 - 88. [Abstract] [PDF]

				O. E. H. Elgersma, M. van Leersum, P. C. Buijs, M. S. van Leeuwen, Y. T. van de Schouw, B. C. Eikelboom, and Y. van der Graaf Changes Over Time in Optimal Duplex Threshold for the Identification of Patients Eligible for Carotid Endarterectomy Stroke, November 1, 1998; 29(11): 2352 - 2356. [Abstract] [Full Text] [PDF]

				W. S. Moore, R. F. Kempczinski, J. J. Nelson, and J. F. Toole Recurrent Carotid Stenosis : Results of the Asymptomatic Carotid Atherosclerosis Study Stroke, October 1, 1998; 29(10): 2018 - 2025. [Abstract] [Full Text] [PDF]

				A. V. Alexandrov, D. Vital, D. S. Brodie, P. Hamilton, and J. C. Grotta Grading Carotid Stenosis With Ultrasound : An Interlaboratory Comparison Stroke, June 1, 1997; 28(6): 1208 - 1210. [Abstract] [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Howard, G. Articles by Toole, J. F. Search for Related Content PubMed PubMed Citation Articles by Howard, G. Articles by Toole, J. F.