| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular Imaging |
1 Laurie Imaging Center, University Radiology Group, University of Medicine and
Dentistry of New Jersey, 141 French St., New Brunswick, NJ 08901.
2 Department of Diagnostic Radiology, Mayo Clinic and Foundation, 200 First St.
SW, Rochester, MN 55905.
Received August 18, 2003; accepted after revision March 29, 2004.
Address correspondence to J. K. DeMarco.
OBJECTIVE. The purpose of this study is to determine the sensitivity, specificity, and clinical utility of classic 2D time-of-flight MR angiography (acquired with derated gradients) as an aid to predicting severe carotid stenosis.
SUBJECTS AND METHODS. Our study population was composed of 68 patients, yielding 133 carotid bifurcations for analysis. A 2D time-of-flight MR angiography pulse sequence was modified to provide greater sensitivity for carotid stenosis, which resulted in visualization of a carotid stenosis with a 70% or greater diameter as a signal void. Contrast-enhanced MR angiography was performed with the elliptical centric view order. Multiple overlapping thin-slab acquisition (MOTSA) MR angiography was performed in select patients. Digital subtraction angiography was performed in 51 patients, and the findings were used as the gold standard. In the remaining patients, findings on carotid duplex Doppler sonography and at surgery and clinical follow-up were used as the gold standard.
RESULTS. In 51 patients for whom a digital subtraction angiogram was available, we found that the sensitivity of classic 2D time-of-flight MR angiography for prediction of carotid stenosis with a 70% or greater diameter was 94%, and the specificity of the technique was 97%. In three patients with severe carotid stenosis, the stenoses that appeared as signal voids on the classic 2D time-of-flight MR angiography were underestimated on contrast-enhanced MR angiography. Severe stenosis was confirmed by subsequent digital subtraction angiography, surgical results, or both. Discrepancies between findings on MOTSA MR angiography and contrast-enhanced MR angiography were resolved with classic 2D time-of-flight MR angiography. Classic 2D time-of-flight MR angiography increased diagnostic confidence of a severe stenosis in three patients with focal internal carotid artery stenosis.
CONCLUSION. Classic 2D time-of-flight MR angiography has a high sensitivity and specificity for predicting carotid bifurcation stenosis of 70% or greater diameter. These probability measures allowed the detection of three significant stenoses that would have been missed on contrast-enhanced MR angiography and provided greater diagnostic confidence than contrast-enhanced or MOTSA MR angiography alone.
Since its introduction in the late 1980s, carotid MR angiography has developed into a robust clinical tool. Numerous articles have detailed the ability of time-of-flight MR angiography to reveal the presence of severe carotid stenosis depicted on the invasive digital subtraction angiography. Most researchers reported using multiple overlapping thin-slab acquisition (MOTSA) to directly measure stenosis of the carotid artery [1–6]. Two-dimensional time-of-flight MR angiography was shown to be sensitive for the detection of severe carotid stenosis but did not depict the carotid bifurcation lumen with the same spatial resolution as MOTSA MR angiography. Subsequent development of first-pass contrast-enhanced MR angiography resulted in more rapid image acquisitions that are physiologically analogous to those of digital subtraction angiography and are less prone to motion artifacts than standard time-of-flight MR angiography [7]. This improvement increased the acceptance at many centers of carotid MR angiography as a replacement for invasive digital subtraction angiography in the preoperative evaluation of carotid stenosis [8–11].
Recent increases in gradient performance have led to a decrease in the TE and gradient moments of the second order and higher, resulting in a marked decrease in the intravoxel dephasing (i.e., intravoxel phase dispersion) on 2D time-of-flight and MOTSA MR angiography. Concern has recently been raised that this technologic advance may paradoxically lead to an underestimation of carotid stenosis: when high-performance gradients and short TE are used, the carotid pulsations could blur the lumen, making the residual carotid diameter appear larger than its actual size [3]. Previously, this effect may have been masked by intravoxel dephasing that eliminated the lumen signal on MOTSA MR angiography. The American Society of Neuroradiology funded a prospective study at seven imaging centers that compared the performance of digital subtraction angiography and MOTSA MR angiography using high-performance gradients. The results of the comparison showed that the severity of stenosis was consistently underestimated with MOTSA MR angiography compared with the severity of stenosis determined with digital subtraction angiography (Korosec FR et al., presented at the 12th International Workshop on Magnetic Resonance Angiography, October 2000).
These findings were the impetus for the development and use of an MR angiography pulse sequence with increased sensitivity for the detection of severe carotid stenosis. Results of 2D time-of-flight MR angiography using modern high-performance gradients (e.g., 23–50 mT/m and 120–200 T/m per second) and shorter minimum TE have shown much less intravoxel dephasing than results reported with gradients (e.g., 10 mT/m and 16.7 T/m per second) typically used before these hardware upgrades. An MR angiography sequence was modified to apply this "classic" gradient strength of 10 mT/m and 16.7 T/m per second to selected imaging lobes of the 2D time-of-flight pulse sequence, regardless of the actual gradient hardware. We have termed this sequence "classic 2D time-of-flight MR angiography."
Our hypothesis was that classic 2D time-of-flight MR angiography would have the "old" sensitivity of 2D time-of-flight MR angiography, in which a signal void indicated a carotid stenosis of 70% or greater diameter, according to North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria [12]. The purpose of our study was to test this hypothesis and to evaluate the clinical utility of classic 2D time-of-flight MR angiography in patients who were being scanned for suspected carotid stenosis.
Subjects and Methods
We enrolled 68 patients (with a total of 133 carotid bifurcations) into our study after our institutional review board approved the protocol. The classic 2D time-of-flight MR angiography series was added to the routine clinical imaging protocol, and the results were compared with those of contrast-enhanced MR angiography in all patients. For 33 patients, findings of MOTSA MR angiography were available, and comparisons were also made with these results. All patients were referred for additional imaging either because the results of duplex Doppler sonography suggested the presence of severe carotid stenosis or because clinical symptoms suggested stroke or transient ischemic attack.
Of these 68 patients, 51 also underwent invasive digital subtraction angiography as part of their routine clinical imaging. We measured the carotid stenosis in each patient on the digital subtraction angiography using the standard NASCET criteria in a double-blinded fashion (without knowledge of the results from MR angiograms). The absence or presence of a signal void on the classic 2D time-of-flight MR angiograms was assessed by another radiologist who had no knowledge of the results of digital subtraction angiography. On the basis of previous reports, we assumed that a signal void on the classic 2D time-of-flight MR angiogram would be seen in patients with severe carotid stenosis of 70% or greater diameter, according to NASCET criteria [12]. We also assumed that the absence of a signal void on the classic 2D time-of-flight MR angiograms would be seen in patients with a carotid stenosis of less than 70% on digital subtraction angiograms. We tested these hypotheses by calculating the sensitivity and specificity of the presence of a signal void on classic 2D time-of-flight MR angiography to predict the presence of a severe carotid stenosis on the digital subtraction angiography.
We also examined the association between the presence (or absence) of a signal void on the classic 2D time-of-flight MR angiograms and the percentage diameter of the stenosis seen on the contrast-enhanced MR angiograms and the MOTSA MR angiograms when these images were available. We measured the stenosis on these images according to the NASCET criteria. A combination of targeted maximum intensity projections, source axial MR images, and oblique axial reformations obtained perpendicular to the lumen of the internal carotid artery was provided for measurement of the carotid stenosis. For purposes of comparison, we classified the degree of the carotid stenoses into the following broad categories: no stenosis (< 30%), mild stenosis (31–49%), moderate stenosis (50–69%), severe stenosis (70–99%), and occluded. Any disagreements between classifications of carotid stenoses as measured on the classic 2D time-of-flight MR angiograms and as measured on the contrast-enhanced MR angiograms, MOTSA MR angiograms, or both were noted. Discrepancies between the classifications of carotid stenoses as measured on MOTSA MR angiograms and as measured on contrast-enhanced MR angiograms were also noted and compared with the classification as measured on the classic 2D time-of-flight MR angiograms. The digital subtraction angiograms, when available, were used as the gold standard to settle these disagreements. For the 17 patients without digital subtraction angiograms, the final clinical outcome was used, including the intraoperative results and the findings on carotid duplex Doppler sonography and at clinical follow-up.
The classic pulse sequence was implemented by modifying selected gradient lobes of a standard 2D time-of-flight pulse sequence available on our 1.5-T system (Signa, GE Healthcare). The target gradient amplitude and slew rate for any gradient lobe between the plateau of the slice selection gradient and the plateau of the readout gradient (inclusive) were derated to 10 mT/m and 16.7 T/m per second, respectively. All other gradient lobes, such as those associated with the spatial presaturation pulse used to reduce venous signal or the end-of-sequence spoiler used to dephase residual transverse magnetization, were not derated. To do so causes an unwanted increase in minimum TR while producing no effect on intravoxel dephasing of the arterial signal. With first-order gradient moment nulling on the slice selection and frequency encoding axes, we used a minimum TE of 8.7 msec for the derated pulse sequence. The other imaging parameters used were TR, 28 msec; matrix, 256 x 128; receiver bandwidth, ± 16 kHz; flip angle, 60°; 40–100 slices with a thickness of 1.8–2.2 mm spaced by 1.5–1.8 mm; signal average, 1; and "traveling" spatial presaturation applied superior to the imaging slice.
The MOTSA MR angiography protocol consisted of two or three overlapping thin slabs obtained with the following parameters: TR/TE, 24/2.5; receiver bandwidth, ± 32 kHz; flip angle, 30°; 1.0-mm-thick partitions interpolated to 0.5-mm intervals with zero-filling; signal average, 1; matrix, 256 x 192 interpolated with zero-filling to 512 x 512; field of view, 20 cm; and superior saturation band. The carotid contrast-enhanced MR angiography consisted of elliptical centric phase reordering with the following parameters: TR range/TE range, 5.9–6.1/1.4–1.6; receiver bandwidth, ± 32 kHz; flip angle, 30°; 40–50 partitions with a thickness of 1.2 mm reformatted at 0.6-mm intervals; signal average, 1; matrix, 256–288 x 192–256; and field of view, 22–26 cm interpolated with zero-filling to 512 x 512.
Results
Classic 2D time-of-flight MR angiography and contrast-enhanced MR angiography of the carotid artery were successfully performed and resulted in diagnostic-quality images for all 68 patients. The increase in imaging time compared with that required for non-derated 2D time-of-flight MR angiography was minimal. The increase in imaging time is the product given by multiplying the increase in TR, the number of acquired slices, and the number of phase-encoding views. For our imaging protocol, the additional imaging time was less than 1 min. In the 51 patients for whom a digital subtraction angiogram was available, the sensitivity of classic 2D time-of-flight MR angiography for prediction of a carotid stenosis of 70% or greater diameter was 94%, and the specificity was 97% (Table 1).
|
Classic 2D time-of-flight MR angiograms displayed a signal void in the region of carotid stenosis of 70% or greater diameter (by NASCET criteria measured on digital subtraction angiograms). The severity of the carotid stenosis was less apparent on the non-derated 2D time-of-flight MR angiography using current high-performance gradients (Figs. 1A, 1B, and 1C).
|
|
|
A discrepancy in the severity of stenosis as depicted on the classic 2D time-of-flight MR angiograms and on carotid contrast-enhanced MR angiograms was noted in three of the 133 carotid bifurcations that were evaluated. In two patients, there was a significant underestimation of stenosis on contrast-enhanced MR angiograms that displayed a signal void on the classic 2D time-of-flight MR angiograms. Subsequent digital subtraction angiograms confirmed the severe stenosis in both patients (Figs. 2A, 2B, 2C, and 2D). In the third patient, the extremely focal severe stenosis was initially misinterpreted as a moderate stenosis on the contrast-enhanced MR angiogram but was obvious as a signal void on the classic 2D time-of-flight MR angiogram. The carotid duplex Doppler sonography suggested a severe stenosis of 70–90% diameter with peak systolic velocity of 396 cm/sec; this finding was confirmed at surgery.
|
|
|
|
MOTSA MR angiograms and contrast-enhanced MR angiograms were obtained in 33 patients. A discrepancy in the severity of stenosis as depicted on the MOTSA MR angiograms and contrast-enhanced MR angiograms was noted in four of the 66 carotid bifurcations that were evaluated on both types of images. In three cases, the degree of stenosis predicted on the basis of MOTSA MR angiograms was less severe than the degree predicted on the basis of the contrast-enhanced MR angiograms. In one case, the MOTSA MR angiogram led to a prediction of a higher-grade stenosis than the prediction based on the contrast-enhanced MR angiogram. The absence or presence of the signal void on the classic 2D time-of-flight MR angiogram correctly resolved the discrepancy and was used to accurately predict the final clinical outcome. In two cases, the presence of a signal void on the classic 2D time-of-flight MR angiogram suggested a severe carotid stenosis as noted on the contrast-enhanced MR angiogram and confirmed by the results of digital subtraction angiography, duplex Doppler sonography, or surgery (Figs. 3A, 3B, 3C, and 3D). Using MOTSA MR angiography alone would have resulted in incorrectly categorizing the stenoses as having less than 70% diameter by NASCET criteria. In two patients, the lack of a signal void implied the carotid stenoses had a less than 70% diameter. The carotid stenosis was overestimated on the MOTSA MR angiogram in one patient, whereas in the other, the stenosis was overestimated on the contrast-enhanced MR angiogram (Figs. 4A, 4B, 4C, and 4D).
|
|
|
|
|
|
|
|
In an additional three patients with extremely focal plaques, carotid stenosis was difficult to measure on contrast-enhanced MR angiography alone. The presence of a signal void on classic 2D time-of-flight MR angiograms suggested a high-grade stenosis that was confirmed by detailed measurements on multiplanar reformations and results of duplex Doppler sonography and surgery (Figs. 5A, 5B, and 5C).
|
|
|
Discussion
Classic 2D time-of-flight MR angiography has a high sensitivity and specificity for predicting carotid bifurcation stenosis of 70% or greater diameter as depicted on digital subtraction angiography. Elliptical centric contrast-enhanced MR angiography has been shown to offer high-quality images of the carotid bifurcation with lumen-filling characteristics that make the technique physiologically analogous to conventional angiography [7]. Combining the two sequences has evolved into the preferred preoperative MR angiographic evaluation of patients with suspected carotid bifurcation stenosis in our practice. Although concern has recently been raised about the possibility of overestimating stenosis on carotid contrast-enhanced MR angiography [13], our clinical experience as well as the results of a different double-blinded study comparing carotid MR angiography and digital subtraction angiography suggests a much more serious possibility of underestimating carotid bifurcation stenosis (Korosec FR et al., presented at the 12th International Workshop on Magnetic Resonance Angiography, October 2000). As pointed out by Kuntz et al. [14], noninvasive imaging techniques need to have a high sensitivity for detecting severe carotid stenosis to maintain the lowest long-term morbidity and mortality rates as well as have a favorable cost-effectiveness ratio.
The clinical utility of classic 2D time-of-flight MR angiography has been proven in our study. In three of the 51 patients studied, severe carotid stenosis would have been misclassified as less significant as measured on contrast-enhanced MR angiography. Without classic 2D time-of-flight MR angiography, the underestimation of carotid stenosis based on MOTSA MR angiography and contrast-enhanced MR angiography may have resulted in the three patients being denied surgery. In addition, classic 2D time-of-flight MR angiography helped resolve discrepancies between MOTSA MR angiography and contrast-enhanced MR angiography. Measurement of carotid stenosis associated with extremely focal plaques can be problematic on contrast-enhanced MR angiography. The presence of a signal void on the classic 2D time-of-flight MR angiography can increase confidence in diagnosing the presence of severe stenosis in those patients with extremely focal plaques.
Classic 2D time-of-flight MR angiograms are designed to evaluate whether severe carotid stenosis is present by displaying a signal void. Motion artifact and spin dephasing inherent in this 2D technique limit accurate measurement of the carotid stenosis. Evaluating all types of stenotic carotid lesions with classic 2D time-of-flight MR angiograms alone is not advisable because of the false-positive findings that would result. Higher-resolution MOTSA and contrast-enhanced MR angiograms are the workhorse sequences to measure carotid stenosis and are the images typically provided to referring clinicians. Instead, classic 2D time-of-flight MR angiography is useful in conjunction with MOTSA and contrast-enhanced MR angiography to reduce false-negative results in the evaluation for the presence of severe carotid stenosis. The clinical utility of this highly sensitive and specific combination of pulse sequences has been shown in this study.
Thirty-four percent of the subjects in our study had a carotid stenosis equal to or greater than 70% diameter, which is representative of our referral pattern: our patients are preselected for a high likelihood of carotid stenosis by the primary physician. Estimating the sensitivity and specificity of classic 2D time-of-flight MR angiography to predict severe carotid stenosis is difficult if the pretest probability of disease was lower than it was in our patients. Possibly, the results would be similar even with a lower incidence of severe stenosis, given that classic 2D time-of-flight MR angiography was shown to be useful in cases in which MOTSA and contrast-enhanced MR angiograms both under- and overestimated the degree of carotid stenosis. This potential design bias is present in most of the literature on carotid MR angiography.
Many centers use 2D time-of-flight MR angiography in their routine clinical protocols for carotid stenosis, especially as a localizing sequence for MOTSA MR angiography or contrast-enhanced MR angiography. Classic 2D time-of-flight MR angiography could be implemented by simply substituting it for conventional 2D time-of-flight MR angiography. The resultant increase in imaging time is less than 1 min.
Classic 2D time-of-flight MR angiography has a high sensitivity and specificity for predicting carotid bifurcation stenosis of 70% or greater diameter as depicted on digital subtraction angiography. Displaying a signal void in the presence of severe stenosis allowed the detection of three significant stenoses that would have been missed on contrast-enhanced MR angiography. In addition, classic 2D time-of-flight MR angiography can increase diagnostic confidence in stenosis assessment as measured on contrast-enhanced or MOTSA MR angiography alone. The increased sensitivity of classic 2D time-of-flight MR angiography to depict severe carotid stenosis was especially important in patients in whom a discrepancy in the degree of carotid stenosis was noted on duplex Doppler sonography, contrast-enhanced MR angiography, and MOTSA MR angiography. Diagnostic confidence in excluding or not excluding the presence of severe carotid stenosis was enhanced because of this increased sensitivity. Likewise, the increased sensitivity was useful in patients in whom an extremely focal plaque made the stenosis difficult to measure on contrast-enhanced MR angiography.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. R Jaff, G. V Goldmakher, M. H Lev, and J. M Romero Imaging of the carotid arteries: the role of duplex ultrasonography, magnetic resonance arteriography, and computerized tomographic arteriography Vascular Medicine, November 1, 2008; 13(4): 281 - 292. [Abstract] [PDF]
|
|
|
|
 |
|
 S. M. Debrey, H. Yu, J. K. Lynch, K.-O. Lovblad, V. L. Wright, S.-J. D. Janket, and A. E. Baird Diagnostic Accuracy of Magnetic Resonance Angiography for Internal Carotid Artery Disease: A Systematic Review and Meta-Analysis Stroke, August 1, 2008; 39(8): 2237 - 2248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. F. Layton, J. Huston 3rd, H. J. Cloft, T. J. Kaufmann, K. N. Krecke, and D. F. Kallmes Specificity of MR Angiography as a Confirmatory Test for Carotid Artery Stenosis: Is It Valid? Am. J. Roentgenol., April 1, 2007; 188(4): 1114 - 1116. [Abstract] [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]
|
|
|
|
 |
|
 J. Huston III, M.A. Bernstein, and S.J. Riederer Feathering: Vertebral artery pseudostenosis with elliptical centric contrast-enhanced MR angiography. AJNR Am. J. Neuroradiol., April 1, 2006; 27(4): 850 - 852. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
