NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Meenan RT, Saha S, Chou R, et al. Effectiveness and Cost-Effectiveness of Echocardiography and Carotid Imaging in the Management of Stroke. Rockville (MD): Agency for Healthcare Research and Quality (US); 2002 Jul. (Evidence Reports/Technology Assessments, No. 49.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Effectiveness and Cost-Effectiveness of Echocardiography and Carotid Imaging in the Management of Stroke

Effectiveness and Cost-Effectiveness of Echocardiography and Carotid Imaging in the Management of Stroke.

Show details



Links in the chain of evidence required to establish the effectiveness of echocardiography in the management of patients with stroke include the following assertions (Table 20):

  • Clinically inapparent abnormalities identified by echocardiography convey increased risk of recurrent stroke;
  • The prevalence of these abnormalities is not inconsequential;
  • Echocardiography is accurate in diagnosing these abnormalities;
  • Adverse events associated with echocardiography are small or infrequent compared to its benefits;
  • Efficacious treatments exist that reduce morbidity and mortality associated with potential sources of cardioembolic stroke identified by echocardiography;
  • Adverse events associated with these treatments are small or infrequent compared to their benefits.

Table 20. Overall Quality and Summary of Evidence for Each Key Question.


Table 20. Overall Quality and Summary of Evidence for Each Key Question.

Existing studies provide conflicting information or are of insufficient quality to firmly establish the presence and degree of stroke risk associated with most echocardiographic lesions in patients without clinically apparent conditions that indicate the need for therapy regardless of echocardiographic findings. Scant data, for instance, have been collected examining the risk of stroke associated with intracardiac thrombus in patients without AF or recent MI, conditions for which anticoagulation is usually indicated to reduce recurrent stroke risk. Lesions for which there is at least fair evidence of an association with stroke include complex (mobile, ulcerated, or > 4 mm in thickness) aortic atheromas and atrial septal aneurysm. ASA appears to convey the greatest risk in the presence of coexisting patent foramen ovale. Evidence regarding the risk of stroke with PFO alone is conflicting and suggests that any increased risk is likely to be most important in young patients. Data from case series suggest that the prevalence of stroke in patients with atrial myxoma, who are often younger than 50, is substantially higher than the prevalence in the general population. There is fair evidence that mitral valve prolapse and mitral annular calcification, previously thought to be important risk factors, are not independently associated with stroke. The association between MVP and stroke was likely due to inaccurate and biased measurement using early echocardiographic techniques. Mitral annular calcification appears most likely to be a marker of atherosclerotic disease, rather than an independent risk factor for stroke.

If one assumes that certain echocardiographic lesions are true risk factors for cardioembolic stroke, the usefulness of identifying those lesions depends on 1) the availability of efficacious treatments to reduce the risk of future stroke attributable to the lesions, and 2) the lack of ability to clinically (rather than echocardiographically) identify patients who would benefit from such treatments. Anticoagulant drugs represent the most promising treatment, because thromboembolism is felt to be the pathophysiologic mechanism by which most echocardiographic lesions convey increased stroke risk. Consistent with previous systematic reviews, we found evidence that any benefit of anticoagulation in unselected patients with stroke is offset by an increase in complications, particularly intracranial hemorrhage. Evidence for the efficacy of anticoagulation in stroke patients with clinically apparent cardiac disease was generally of poor quality and provided insufficient information to draw firm conclusions. Data on the primary prevention of stroke with anticoagulants among patients with clinically manifest heart disease other than AF (e.g., recent MI, DCM) did not allow conclusions as to whether patients with these conditions who experience stroke benefit from anticoagulants. Thus, it does not appear that clinical criteria, other than atrial fibrillation, are helpful in selecting patients that might benefit from anticoagulation, leaving room for the possibility that echocardiography may be useful in selecting such patients.

There is little evidence, however, regarding the efficacy of anticoagulation for patients identified by echocardiography as having potential sources of cardioembolic stroke. Studies of anticoagulation for specific echocardiographic lesions have been small and generally of poor quality. One poor-quality meta-analysis of primarily observational studies of stroke prevention in patients with intraventricular thrombus suggested a benefit of anticoagulation but was limited by significant heterogeneity across pooled studies. No studies have examined the efficacy of anticoagulation for patients with left atrial thrombus in patients without AF. One small observational study suggested a possible benefit of anticoagulation in patients with mobile aortic atheromas, but the authors did not account for baseline differences between those receiving and those not receiving anticoagulants that may have accounted for differences in outcome. While treatments other than anticoagulation may reduce the risk of stroke associated with certain echocardiographically identified lesions (e.g., surgical resection of left atrial myxoma, surgical or catheter-based closure of PFO), data on the impact of these interventions on recurrent stroke are limited.

If one assumes that there are efficacious treatments to reduce the risk of stroke associated with echocardiographic lesions, the usefulness of echocardiography still depends on the prevalence of such lesions in patients with stroke. If the prevalence is low, then the usefulness of echocardiography may be limited, as large numbers of patients would have to undergo echocardiography to detect few lesions. We found the evidence regarding the yield of echocardiography in patients without AF to be insufficient to derive valid estimates of the prevalence of potential cardioembolic sources of stroke in community-based settings. Studies from primarily university hospitals in several different countries suggest that in unselected stroke patients without AF, the prevalence of intracardiac thrombus is between 1 and 2 percent, indicating that 50 to 100 echocardiograms would need to be performed to detect one thrombus.

The likelihood of detecting intracardiac thrombus is thought to be higher in certain patient subgroups -- those without carotid artery stenosis and those with manifest heart disease. Studies in these subgroups were limited in both quantity and quality. Small numbers of patients limited the ability to derive estimates of intracardiac thrombus prevalence in patients without carotid artery disease. Our finding that thrombus was more prevalent in unselected patients than in patients without heart disease indirectly suggests that patients with heart disease may have a higher prevalence of intracardiac thrombus than others. In young patients (aged 15 to 45), the prevalence of thrombus was similar to that in unselected patients.

The prevalence of complex aortic atheroma was variable across studies, ranging from 2 to 22 percent. The highest rates were reported by studies in which patients underwent TEE for the expressed purpose of detecting aortic atheroma. In studies of patients undergoing TEE to search for any source of cardioembolic stroke, the prevalence of complex atheromas ranged from 2 to 6 percent, suggesting that when conducted routinely, TEE may not yield the majority of atheromas present.

The prevalence of ASA also varied substantially across studies, from 4 to 22 percent, and was similar in studies including and excluding patients with carotid artery disease. In one series of patients under 60 without carotid artery disease, intracardiac thrombus, or AF, the prevalence of coexisting ASA and PFO was 15 percent. Atrial myxoma occurred in one of 721 patients across eight studies enrolling consecutive patients with stroke. In two studies including 180 patients aged 15 to 45, no left atrial myxomas were detected. One right atrial myxoma was detected and was thought to have caused the patient's stroke through embolism through a PFO.

Studies of the accuracy of TTE and TEE in diagnosing intracardiac thrombus have been undertaken primarily in patients undergoing surgery for rheumatic mitral valve disease and left ventricular aneurysm, conditions conveying high risk of LAT and left ventricular thrombus, respectively. The average sensitivity and specificity for diagnosing LAT were 93 and 97 percent, respectively, for transesophageal echocardiography and 42 and 99 percent for transthoracic echocardiography. The low sensitivity of TEE was primarily attributable to missed thrombi in the left atrial appendage. The estimated sensitivity and specificity of TTE for diagnosing LVT were 77 and 95 percent.

Under these estimates of echocardiographic accuracy, using TTE to detect LVT would result in more false positive than true positive tests unless the prevalence of LVT exceeded 6 percent. To achieve a positive predictive value of 90 percent, thrombus would have to be present in 37 percent of patients. TEE would produce as many false positive as true positive diagnoses of LAT at a prevalence of 2 percent. To reach a positive predictive value of 90 percent, the frequency of LAT would need to exceed 15 percent.

We did not identify any studies of echocardiographic accuracy in diagnosing ASA or aortic atheroma. However, since the echocardiographic rather than anatomic definitions and characteristics of these lesions have been found to be associated with stroke, TEE may be considered to be the gold standard for diagnosing these abnormalities. The few studies that have compared the echocardiographic diagnosis of myxoma to surgical or pathological examination suggest 100 percent accuracy, though one study found disagreement between TTE and TEE in 2 of 11 cases.

Although TTE is noninvasive and generally safe, TEE is a semi-invasive procedure and is associated with patient discomfort as well as cardiac, pulmonary, and gastrointestinal complications. Approximately 2 percent of TEE procedures are unsuccessful, mainly due to patient intolerance. About seven in 1,000 patients undergoing TEE experience major complications requiring treatment. The mortality associated with TEE is approximately one in 10,000.

Taken as a whole, our findings indicate that the links in the chain of evidence for the effectiveness of echocardiography in the management of patients with stroke are weak. The risk of recurrent stroke associated with most echocardiographic lesions and the efficacy of treatment in reducing that risk are unclear. The estimated yield and accuracy of echocardiography in detecting intracardiac thrombus -- the lesion typically considered most likely to convey modifiable risk of recurrent stroke -- indicate that for unselected patients, TTE and TEE will produce at least as many false positive as true positive diagnoses. While TEE is generally more accurate than TTE, it is also more invasive and is associated with a small but quantifiable risk of major complications.

Because of the lack of solid evidence for important components of effectiveness, it is difficult to accurately estimate the cost-effectiveness of echocardiography in the management of stroke. Where evidence was lacking or insufficient, we made informed assumptions to enable estimating cost-effectiveness. Assumptions include that intracardiac thrombus conveys increased stroke risk for the first year after the initial stroke; thrombus prevalence is 2 percent in unselected patients and 5 percent in patients with heart disease; and anticoagulant drugs reduce the risk of recurrent stroke by one-third. Using those assumptions, one quality-adjusted life year (QALY) can be saved for an approximate incremental cost of $300,000, using TEE only in patients with heart disease. Other strategies were less cost-effective, though TTE in patients with heart disease was the preferred strategy under some plausible assumptions. The cost-effectiveness ratio for either echocardiographic procedure fell below $50,000 per QALY if the assumed relative risk reduction with anticoagulation was increased to 86 percent and the prevalence of thrombus was simultaneously increased to 6 percent. The cost per QALY of all strategies increased as average life expectancy diminished (e.g., with increasing age or comorbidity). These results suggest that the cost-effectiveness of the routine use of echocardiography to detect intracardiac thrombi in patients with stroke compares unfavorably with other commonly endorsed health care interventions. If one can assume that anticoagulants substantially diminish the risk of recurrent stroke associated with intracardiac thrombus (or other lesions), the cost-effectiveness of echocardiography in the management of stroke may be reasonable when used in selected patients likely to have high pre-test probabilities of thrombus (or other remediable sources of cardioembolic stroke), e.g., patients with dilated cardiomyopathy or other structural heart disease.

Our results differed from the single cost-utility analysis of echocardiography in stroke that we were able to identify in the published literature. 4 In that study, selective or universal application of TEE had incremental cost-effectiveness ratios in the range of $10,000 to $20,000 per QALY. These favorable findings likely resulted from several assumptions that were not consistent with the findings of our systematic review. First, the prevalence of thrombus among unselected patients was assumed to be 8 percent (as opposed to 2 percent in our analysis), which likely stemmed from the authors' collecting prevalence information from several studies that included patients with AF. Second, TEE was assumed in the base-case analysis to have perfect accuracy. Third, the recurrent stroke rate among patients with thrombus was assumed to be 40 percent. Fourth, all thrombi were assumed to be left atrial. Fifth, the duration of benefit from anticoagulation was not specified; presumably it lasted throughout the lifetime time horizon of the model. Sixth, the cost of TEE was assumed to be $360, substantially lower than in our analysis. Finally, complications of TEE were not modeled. All of these assumptions favor the use of TEE to select patients with thrombus for treatment with anticoagulants. (The authors assumed a relative risk reduction with anticoagulation of 33 percent. We used this estimate in our analyses to provide comparability with this study.)

Limitations of our review and cost-effectiveness analysis should be noted. First, we did not systematically review the literature on therapies other than anticoagulation for echocardiographic lesions. Such therapies include surgical and catheter-based interventions for intracardiac tumors, valve disease, and PFO and have generally not been extensively studied. Second, we did not incorporate into our analyses the value of echocardiographic information other than identifying potential sources of cardioembolic stroke. Echocardiography provides information on cardiac structure and function that may not directly assist in reducing recurrent stroke risk but may be helpful in guiding other aspects of clinical care. The extent to which this information positively affects clinical outcomes was not assessed in our review and if substantial, may improve the estimated effectiveness and cost-effectiveness of echocardiography.

Carotid Imaging

The role of carotid imaging is better established than that of echocardiography in patients with stroke. It is clear that carotid artery stenosis conveys increased risk of stroke and that efficacious treatment exists to reduce that risk. However, the most effective imaging strategy for diagnosing carotid artery stenosis is controversial. The most widely used tests include two noninvasive tests, carotid ultrasound and magnetic resonance angiography, and one invasive test, cerebral angiography. These tests may be used alone or in various combinations. Although the noninvasive tests are not associated with significant complications, their effectiveness in predicting who will benefit from surgical intervention has not been directly established, as it has for angiography. The noninvasive tests therefore carry the potential for false positive and false negative diagnoses and the consequent risk of selecting patients without significant carotid stenosis for ineffective and potentially harmful surgery, or excluding patients with significant stenosis from beneficial treatment. In order to compare the effectiveness of various strategies for carotid imaging, we examined evidence related to the following (Table 20):

  • Operating characteristics (sensitivities, specificities, and likelihood ratios) of available tests for measuring carotid stenosis;
  • Harms associated with these tests;
  • Efficacy of treatment for varying degrees of carotid stenosis; and
  • Harms associated with these treatments.

Although we found numerous studies assessing the accuracy of CUS, many were hampered by significant biases. The vast majority of studies were limited by verification bias, which occurs when only part of the sample of subjects undergoing CUS also undergo the reference standard test, in this case angiography. When studies without verification bias were reviewed, the estimated sensitivities of CUS based on summary receiver operating characteristic (SROC) curve analysis were 81 and 90 percent, respectively, for moderate (> 50 percent) stenosis, and 75 and 87 percent for severe (> 70 percent) stenosis. These estimates of accuracy were lower than those published in a meta-analysis of the diagnostic accuracy of carotid imaging tests, which did not exclude studies with verification bias. Notably, the accuracy of CUS in our review varied substantially across studies. One large study collecting accuracy statistics from 50 academic centers across North America found an overall sensitivity and specificity of 69 and 68 percent. Other studies from single centers reported sensitivities and specificities above 90 percent. These discrepant findings suggest that at certain centers, high accuracy can be achieved with CUS, but that high estimates of accuracy from published studies may not be generalizable to all settings.

Studies of MRA, even those that were not affected by verification bias, were generally of only fair to poor quality. Large, multicenter studies of MRA accuracy have not been conducted. Data from single-center studies suggest a sensitivity and specificity of 92 and 97 percent, respectively, for detecting severe stenosis. These estimates, however, should be interpreted with caution, as those reporting MRA accuracy from their own centers may be able to achieve a higher sensitivity and specificity than is true for other settings.

We found no studies of combined CUS and MRA that were not affected by verification bias. Most studies were of poor quality. Approximately 18 percent were found to have discrepant results between the two noninvasive tests. The estimated sensitivity and specificity in patients for whom both tests were in agreement were 95 and 98 percent.

Although isolated cases of complications have been reported for CUS and MRA, we did not identify studies that would allow quantification of complication rates for these noninvasive tests. Studies of complications from cerebral angiography indicated an overall mortality rate of 2 per 10,000 patients. Rates of stroke after angiography varied by study; in the highest quality study, stroke occurred in 1.3 percent of patients.

Two large randomized controlled trials have established the efficacy of carotid endarterectomy in reducing the risk of future stroke in patients presenting with symptomatic carotid artery stenosis. For patients with severe (70 to 99 percent) stenosis, one disabling stroke over a 2- to 6-year period is prevented for every 15 patients undergoing CEA. For patients with moderate (50 to 69 percent) stenosis, the benefit is smaller, with a "number needed to treat" of 21. Patients with lesser degrees of stenosis do not appear to benefit from CEA.

Although CEA is clearly efficacious for select patient subgroups, its efficacy as derived from randomized trials must be weighed against the complications incurred by surgery. Studies of good quality indicated that 1.6 percent of patients undergoing CEA for symptomatic carotid stenosis die perioperatively (within 30 days of surgery), and 6.8 percent experience perioperative stroke or death. In the NASCET, 2.4 percent of patients with moderate stenosis and 3.3 percent with severe stenosis treated non-surgically died or experienced stroke within 30 days of randomization, suggesting that CEA conveys an absolute risk of perioperative stroke or death of 3.5 to 4.8 percent. This complication rate does not appear to be higher when CEA is performed early (within 4 to 6 weeks of stroke presentation) than when it is performed later. Whether it is equally safe to perform surgery within one week of presenting symptoms, which would imply the need for carotid imaging shortly after presentation, is unclear.

The lack of good or consistent evidence regarding the accuracy of noninvasive carotid imaging strategies makes it difficult to accurately determine the most cost-effective strategy for selecting patients with stroke for CEA. Assuming the accuracy statistics derived from our review, two testing strategies provide the most benefit when compared to no testing: MRA with direct referral to CEA of patients with severe (70-99 percent) stenosis, and joint CUS and MRA, with direct referral to CEA of patients for whom both tests demonstrate moderate to severe (50-99 percent) stenosis, and angiographic confirmation when the two tests disagree. The incremental cost-effectiveness ratios for these two strategies are approximately $250,000 and $700,000 per QALY, respectively.

In sensitivity analyses, the variable with the greatest influence on the results of the carotid imaging model was the prevalence of severe carotid stenosis. At severe stenosis prevalences of 0.15 and below, all testing strategies were dominated by the strategy of no testing or had cost-effectiveness ratios exceeding $250,000 per QALY (0.15 was the prevalence assumed in the base-case analysis). However, as this prevalence increased above 0.15, the cost-effectiveness ratios of two strategies -- CUS with angiographic confirmation of severe stenosis (CUS/Angio-70), and MRA with direct CEA referral for severe stenosis (MRA-70) -- fell precipitously, such that at a prevalence of 0.20, these strategies had cost-effectiveness ratios in the range of $60,000 to $75,000 per QALY. At higher prevalences, these ratios fell further. When compared to the strategy of no testing, CUS/Angio-70 had an incremental cost-effectiveness ratio of less than $50,000 per QALY at a prevalence of 0.25, while the incremental cost-effectiveness of MRA-70 fell below $50,000 per QALY as the prevalence of severe stenosis approached 0.30. These results suggest that carotid imaging may compare unfavorably, in terms of cost-effectiveness, with other commonly endorsed health care interventions, when the prevalence of carotid stenosis is low. It may be most efficient, therefore, to refer for carotid imaging those with a high pretest probability of severe stenosis, e.g., patients with peripheral vascular disease or audible carotid bruits.

We identified one previous cost-utility analysis of carotid imaging strategies. 342 In that study, noninvasive testing strategies using either CUS alone or with MRA were found to have incremental cost-effectiveness ratios of approximately $10,000 to $20,000 per QALY. However, in that study single-center estimates of diagnostic accuracy were used, and the prevalence of severe carotid stenosis was 32 percent. 343 At a prevalence of 32 percent in our analysis, CUS/Angio-70 and MRA-70 had cost-effectiveness ratios in the range of $40,000 per QALY.

Varying the cost of testing did not substantively affect the results, except in the case where MRA was assumed to cost $2,500. In this analysis, the strategy of initial CUS with angiographic confirmation of severe stenosis became undominated, with a cost-effectiveness ratio of $280,000 per QALY. Varying the accuracy of the different testing strategies over wide ranges did not have a substantial overall effect on the results. When the perioperative complication rate was assumed to be zero, noninvasive strategies involving direct referral to CEA of patients with moderate or greater stenosis expectedly became the most cost-effective; without risk of complications, angiographic confirmation to avoid false positives was no longer beneficial, and the marginal benefit of CEA among patients with moderate stenosis was no longer counterbalanced by perioperative risk. Varying the duration of risk reduction associated with CEA between 2 and 10 years also did not substantively affect the cost-effectiveness ratios. Likewise, restricting the cohort to only patients with TIA or minor stroke, which reflects the patient populations in the two large carotid endarterectomy trials, did not have a major impact on cost-effectiveness ratios, though it did produce a different set of undominated strategies.

It is noteworthy that strategies in which patients with moderate (50-69 percent) stenosis were referred for CEA provided fewer QALYs than strategies in which such patients were treated non-surgically, despite the fact that our review (and hence our model inputs) reflected an overall benefit from CEA for moderate stenosis. This occurred as a result of the fact that the benefit of CEA over non-surgical management in patients with moderate stenosis is small, such that when a 3 percent discount rate is applied to account for the fact that health benefits incurred or realized in the future are considered to be of lower value than benefits realized in the present, the future benefits are outweighed by the perioperative complications incurred immediately after surgery. When perioperative complication rates were assumed to be zero, or when the discount rate was removed, strategies involving CEA for patients with moderate stenosis became more cost-effective.

Several limitations of our review and cost-effectiveness analysis should be noted. First, we did not address the accuracy of noninvasive carotid imaging strategies other than CUS and MRA. Tests such as computed tomographic angiography are less commonly used than CUS and MRA and are less extensively studied. Second, because of the small number of good- or fair-quality studies, we combined studies of CUS and MRA that used different techniques (e.g., color flow vs. non-color flow imaging for CUS, two-dimensional time-of-flight [2D-TOF] vs. 2D- and 3D-TOF for MRA). These different techniques may have distinct operating characteristics. Third, we did not examine the potential benefits of carotid imaging other than selection of candidates for CEA. Carotid imaging may aid in stratifying risk of both future stroke and other cardiovascular events and may therefore inform further diagnostic and therapeutic interventions other than CEA. Finally, we did not incorporate into our analyses nonsurgical interventions to reduce recurrent stroke risk from carotid artery stenosis. Other interventions, such as carotid artery stenting, have been studied, though experience with these interventions is limited. It should also be noted that our review and analysis explicitly addressed only the issue of symptomatic carotid stenosis. The effectiveness and cost-effectiveness of carotid imaging strategies in patients with asymptomatic stenosis are beyond the scope of this report.

PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...