Asymmetric aortic expansion of the aneurysm neck: Analysis and visualization of shape changes with electrocardiogram-gated magnetic resonance imaging
Article Outline
Objective
Electrocardiogram (ECG)-gated imaging offers insight into aortic shape changes throughout the cardiac cycle. Morphologic changes of the anchoring zones may influence stent graft fixation and sealing and may have serious implications for endograft design and durability. We used multiphase magnetic resonance imaging (MRI) scans to evaluate the asymmetric aspect of aortic shape changes in the aneurysm neck before and after endovascular aneurysm repair (EVAR).
Methods
Eleven patients were scanned before and after EVAR using ECG-gated balanced gradient-echo MRI with 16 reconstructed phases. Transverse scan planes were planned perpendicular to the aorta in the coronal and sagittal planes. Three levels were studied: 3 cm above the lowest renal artery, between the renal arteries, and 1 cm below the lowest renal artery. After segmentation of the aortic area in the images, aortic radius changes during the cardiac cycle were determined over 360 axes and plotted. Radii were measured from the center of mass of the aortic lumen to the vessel wall. An ellipse was fitted over the plots allowing determination of radius changes over the major and minor axis, and the most prominent direction of distention.
Results
The difference between radius change over the major and minor axis was significant preoperatively and postoperatively (P ≤ .002) at all levels, indicating asymmetric expansion. The pre-EVAR mean radius change over the major vs minor axis was infrarenal, 0.9 ± 0.2 vs 0.6 ± 0.1 mm; juxtarenal, 1.0 ± 0.2 vs 0.8 ± 0.1 mm; and suprarenal, 1.3 ± 0.4 vs 0.9 ± 0.2 mm. At all levels, there was no significant difference (P > .05) between pre-EVAR and post-EVAR radius changes. Pre-EVAR, the ratio of the radius change over the major vs minor axis ranged from 1.10 to 1.82. The pre-EVAR and post-EVAR asymmetry ratios did not differ significantly (P > .1). Preoperatively, the suprarenal direction of distention showed a tendency to right-anterior; for infrarenal, the tendency was to left-anterior.
Conclusions
We measured the asymmetric aspect of earlier reported pulsatile aortic shape changes. The rate of asymmetric distention varied by patient and level. Asymmetric aortic expansion may have consequences for endograft design because it probably affects endograft sealing, especially in patients with high radius changes and asymmetry ratios. Asymmetric expansion remained preserved after stent graft placement. The stent grafts with Z-stent rings used in the study participants seem to adapt to the aortic shape changes well.
The purpose of endovascular aneurysm repair (EVAR) is successful exclusion and depressurization of the aneurysm sac.1 Application of EVAR is not limited to aneurysm repair, but has been extended to the repair of aortic ruptures and dissections in the abdominal and thoracic aortas.2, 3 An important element of successful and durable EVAR is an adequate proximal fixation and seal. To accomplish this, an appropriate stent graft size and design is required.
Pulsatile cardiac contraction and aortic compliance naturally result in shape changes of the aorta throughout the cardiac cycle.4, 5, 6, 7 Knowledge of aortic shape and radius changes throughout the cardiac cycle is important because it has consequences for endograft sizing.6 Endograft sizing decisions are most commonly made by using static computed tomography angiography (CTA) images, but with the current high-speed CTA acquisition, the acquired images may be anywhere in systole, diastole, or somewhere in between. This could lead to improper endograft sizing, with subsequent compromised durability in fixation and sealing. Intermittent type I endoleaks may lead to sac enlargement.
An aortic expansion that is also asymmetric will complicate the accomplishment of an adequate and durable proximal fixation and seal. To prevent migration and (intermittent) type I endoleaks, the endograft should be able to adapt to the pulsatile aortic shape changes during the cardiac cycle. Asymmetry in the pulsatile aortic shape changes will further complicate this and therefore could have a crucial affect on endograft design.
The introduction of electrocardiogram (ECG)-gated and triggered imaging modalities has provided more insight into dynamic aortic behavior under the influence of cardiac pulsatility. ECG-triggered or gated CTA or magnetic resonance imaging (MRI) offers insight into aortic expansion throughout the cardiac cycle.6, 7, 8, 9, 10, 11, 12, 13 Pulsatile wall motion has also been observed with ultrasound (US) and intravascular US (IVUS).14, 15 In previous dynamic imaging studies, we noted that the aortic expansion was not evenly distributed.6 In the current study, we aimed to further investigate this observation because it may have serious implications for endograft design and durability.
Increasing evidence shows that the human aortic distention due to cardiac pulsatility is anisotropic. Increased anterior wall motion compared with posterior wall motion has been observed using M-mode US in both animals and humans, however, not specifically in abdominal aortic aneurysm (AAA) patients.15 Cine phase contrast MRI (PC-MRI) studies have also observed nonuniform circumferential aortic expansion in the porcine and human aorta, but also not specifically in AAA patients.16, 17, 18
Recently, IVUS was used to measure wall displacement in the proximal aneurysm neck of patients before EVAR. Distention in the anterior-posterior (AP) direction and 90° perpendicular to this direction was measured, and it was concluded that the infrarenal aneurysm neck deforms anisotropically. When anisotropic aortic expansion is measured, however, measurements should be made exactly from the epicenter of the aorta (the center of mass) and perpendicular to the aortic lumen, which seems difficult to accomplish with IVUS because the catheter moves with every cardiac contraction. Noncoaxial IVUS imaging can have a significant influence on diameter measurements.19 Invasive methods also could actually influence motion patterns by vasospasms and alternation of hemodynamic patterns. To our knowledge, no reports have been published on the influence of stent graft placement on the asymmetric expansion of the aorta in vivo.
ECG-gated MRI imaging offers the possibility to give a complete mapping of aortic expansion in any direction in relation to the anatomy. The advantage of MRI compared with US modalities is that it is not operator-dependent and is noninvasive. Moreover, MRI postprocessing allows for center of mass analysis, which is the sum of aortic in-plane movement and can mimic radial expansion, and expansion. Using ECG-gated MRI, we recently demonstrated that the pulsatile aortic distention is asymmetric in healthy volunteers.20 This study used ECG-gated MRI scans to visualize and quantify the magnitude and asymmetric aspect of aortic shape at different levels in the aneurysm neck in patients with AAA and to study the influence of stent graft placement on these shape changes.
Methods
Patients
The data from a previous reported study were used to further analyze and quantify asymmetric expansion.6 The study comprised 11 men (median age, 74; range, 63-78 years) who underwent EVAR for an AAA. The median aneurysm size was 56.5 mm (range, 47.4-71.5 mm). The study design and protocol were approved by the institutional medical ethics committee. Informed consent was obtained from all participants.
Imaging
All scans were performed on a clinical 1.5 Tesla MR scanner (Gyroscan Intera, Philips Medical Systems, Best, The Netherlands). After initial multistack abdominal survey scans, a coronal balanced gradient-echo survey scan was performed to localize the renal arteries and the aortic aneurysm. A transverse balanced gradient-echo MRI scan with ECG gating (16 reconstructed phases over the cardiac cycle) was made perpendicular to the aorta, both in the coronal and in the sagittal planes. Three levels were studied: 1 cm below the lowest renal artery (level A), between the renal arteries (level B), and 3 cm above the lowest renal artery (level C). Scan parameters were echo time, 2.0 milliseconds; repetition time, 4.0 milliseconds, and flip angle, 50°. The acquired voxel size was 2.1 × 0.8 × 6.0 mm3, with a field of view of 400 × 320 mm2, using a scan percentage of 267% in the AP direction. The reconstructed voxel size was 0.8 × 0.8 × 6.0 mm3, reconstructed to a matrix of 512 × 410 pixels. The scan duration for obtaining a data set of 16 heart phases was approximately 4 minutes at each level.
Analysis
Dedicated in-house software (Dynamix, Image Sciences Institute, Utrecht, The Netherlands) was developed to analyze the multiphase scans. A region of interest was manually defined in the 16 images before the data were analyzed. To obtain a smoother segmentation of the aortic lumen and decrease discretization artifacts, we applied in-plane super sampling to the image series. This was done by using linear interpolation in both left–right (x) and AP (y) directions (Fig 1). An up-sampling factor of eight in each direction was found to be satisfactory for both visual and quantitative analysis of the aortic shape changes. The segmentation of the aortic lumen was performed semi-automatically. A single seed point was placed inside the lumen, followed by a region-growing algorithm. To this end, a minimum intensity value of the lumen pixels was defined for all 16 images. All segmentations were reviewed by two blinded independent observers, and minor corrections were mainly caused by branching vasculature or intraluminal thrombus.
Fig 1.
These transverse magnetic resonance images perpendicular to the aorta show one cardiac phase of the aortic distention (A) before and (B) after automatic segmentation of the aortic lumen. C, Super-sampling the image results in smoother segmentation.
These final segmented images were used to characterize aortic distention as the aortic radius change during the cardiac cycle. Radius change (difference between minimum and maximum radius during the cardiac cycle) was measured from the center of the mass of the aortic lumen to the inside of the aortic wall. The aortic movement during the cardiac cycle is complex and consists of both aortic in-plane translation and aortic expansion. These movements are summed in the aortic center of mass displacement over the cardiac cycle (Fig 2, A). Thus, to measure asymmetry it is necessary to take radius measurements from the center of mass of each cardiac phase. Ignoring the center of mass displacement effect could lead to incorrect measurement of radius change and a false assumption of asymmetry, as can easily be shown by an experiment of an artificial image where a circle expands to a larger circle (Fig 2, B). According to our definition of asymmetry, the expansion of one circle to another is purely symmetric, which seems intuitively correct.
Fig 2.
A, An example of typical aortic center of mass (COM) displacement is shown in which the COM coordinates are plotted for each cardiac phase. Radii are measured from these coordinates for each cardiac phase. B, Schematic image shows one circle expanding to a larger circle. Neglecting COM movement results in incorrect measurement of radial expansion. Radii during the maximum expansion should not be measured from point A but from point B.
Radius changes were measured over 360 axes, with an angular increment of 1°. The radius change during the cardiac cycle was plotted as a function of angle (Fig 3). An ellipse was fitted over this distention plot using the direct least square fitting of ellipses in Matlab 6.5 computing software (The Mathworks Inc, Natick, Mass; Fig 3).21 Radii (Ra and Rb) and angulation of the ellipse were automatically calculated.21 The radial change over the major (Ra) and minor (Rb) axis was calculated, as well as the orientation of the major axis, expressed as the angle (θ) by which the axis deviated from the AP direction. The AP direction was defined as 0°, with left corresponding to a deviation of 90°, and right to −90°. All anterior orientations are also mirrored (+180° or −180°) in the posterior direction due to the periodicity of the elliptical radial system. The mean orientation of all patients is calculated with the orientation anteriorly (domain −90° to 90). A standard AP line with the spinal foramen as anatomic orientation was used as reference for the baseline AP direction.
Fig 3.
Plots show (A) radius change and (B) corresponding minimum and maximum diameter. An example of a preoperative measurement is shown. A, Radius change is measured over 360 axes and plotted. An ellipsis is fitted and the magnitude and direction of the maximum (Ra) and minimum radius change (Rb) are calculated. B, Illustration shows the corresponding maximum and minimum radii in mm. A, Anterior; L, left; P, posterior; R, right.
The total aortic shape changes over 360 axes is depicted by an ellipse, which is described by the radial change over the major axis (Ra) and the radial change over the minor axis (Rb). The asymmetry ratio was calculated by dividing the radius change over the major axis (Ra) by the radius change over the minor axis (Rb). The radius changes over the major and minor axis and differences between pre-EVAR and post-EVAR measurements were compared using the t test for paired data. Statistical significant differences were assumed at P < .05. Intraobserver and interobserver repeatability was calculated according to Bland and Altman.22
Results
The results are summarized in the Table. Plots and visualization of a typical example of asymmetric distention with a high percentage of distention, at the suprarenal level, is shown in Fig 3. At all levels studied, there was a significant difference (P ≤ .002) between the radius change over the major and minor axis (Fig 4).
Table. Results before and after endovascular aneurysm repair
| Variable | Infrarenal | Juxtarenal | Suprarenal | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Pre-EVAR | Post-EVARa | P | Pre-EVAR | Post-EVAR | P | Pre-EVAR | Post-EVAR | P | |
| Radius change, mm | |||||||||
| Minor axis | |||||||||
 Mean ± SD | 0.6±0.1 | 0.7±0.1 | .35 | 0.8±0.1 | 0.8±0.2 | .79 | 0.9±0.2 | 1.1±0.3 | .09 |
| Range | 0.4-0.8 | 0.4-0.9 | 0.6-1.0 | 0.4-1.1 | 0.5-1.2 | 0.5-1.6 | |||
| Major axis | |||||||||
| Mean ± SD | 0.9±0.2 | 0.9±0.3 | .62 | 1.0±0.2 | 1.1±0.3 | .85 | 1.3±0.4 | 1.4±0.4 | .56 |
| Range | 0.6-1.1 | 0.6-1.4 | 0.8-1.6 | 0.6-1.7 | 0.7-2.0 | 0.7-2.2 | |||
| Orientation major axis, ° | |||||||||
| Mean ± SD | 19.5±42.5 | −15.0±28.5 | .001b | −1.4±32.4 | −25.0±48.5 | .08 | −23.8±41.5 | −16.4±44.1 | .92 |
| Range | −71.8 to 78.1 | −71.7 to 19.9 | −45.5 to 73.1 | −87.3 to 58.8 | −62.5 to 86.6 | −85.3 to 82.0 | |||
| Asymmetry ratio | |||||||||
| Mean ± SD | 1.35±0.18 | 1.30±0.16 | .16 | 1.38±0.20 | 1.38±0.25 | .95 | 1.41±0.20 | 1.32±0.20 | .18 |
| Median | 1.31 | 1.29 | 1.36 | 1.38 | 1.36 | 1.36 | |||
| IQR | 1.23-1.49 | 1.18-1.37 | 1.26-1.41 | 1.17-1.51 | 1.27-1.57 | 1.17-1.38 | |||
| Range | 1.12-1.69 | 1.11-1.60 | 1.10-1.82 | 1.08-1.80 | 1.16-1.73 | 1.05-1.68 | |||
aN = 10. |
bSignificant difference pre-EVAR and post-EVAR. |
Fig 4.
The mean radial distention before and after endovascular aneurysm repair (EVAR) is presented with the standard deviation (error bars). At each level, the radius change over the minor and major axis differed significantly both preoperatively and postoperatively (P ≤ .002). There was no significant difference between pre-EVAR and post-EVAR radius changes (P > .05). The percentages above the bars indicate the percentages of radius change.
No significant difference (P > .05) was noted between pre-EVAR and post-EVAR radius changes at each level. The preoperative ratio of radius change over the major vs minor axis ranged from 1.10 to 1.82. At all levels, the asymmetry ratio did not differ significantly (P > .1) from the postoperative asymmetry ratios (Fig 5). Preoperatively, the suprarenal direction of elliptical expansion (orientation of the major axis) showed a tendency to right–anterior; and for the infrarenal direction, there was a tendency to left–anterior (Fig 6). Postoperatively, only the infrarenal direction of distention changed significantly (P = .001) to right–anterior.
Fig 5.
The asymmetry ratio before and after endovascular aneurysm repair (EVAR) is calculated as radius change over the major axis divided by radius change over the minor axis (Ra/Rb). The dashed line is a ratio of 1.0, which represents symmetric expansion. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively; and the whiskers mark the range. At each level, there was no significant difference (P > .1) between pre-EVAR and post-EVAR asymmetry ratios.
Fig 6.
The major preoperative axis direction of distention is drawn for each patient (lines). The mean axis (dark green line) with −1 and +1 standard deviation (light green) is also shown. The mean axis is calculated with the orientation anteriorly (−90° to 90). Note that all orientations are also mirrored in the posterior direction. A, Anterior, L, left; P, posterior; R, right.
For radius change, the intraobserver mean difference was −0.01 mm and the variability was 0.15 mm, and the interobserver mean difference was −0.03 mm and the variability was 0.19 mm. For the asymmetry ratio, the intraobserver mean difference was 0.005 and the intraobserver variability was 0.23; the interobserver mean difference was 0.01 and the interobserver variability was 0.21. Intraobserver and interobserver variability for the direction of the major axis was 31° and 28°, with mean differences of 0.4° and 1.2°, respectively. The deviation of the baseline AP line ranged from −8.7° to 6.7°.
Discussion
Aortic shape changes throughout the cardiac cycle were visualized in this study. With advanced postprocessing methods, we measured and quantified asymmetric aortic expansion. In most of the scans we studied, the shape of the distention plots in aneurysm neck was clearly elliptical as judged by two independent observers (J. P., J. H.). Therefore, an elliptic distention model was applied to the data. This fitting procedure allowed us to describe the ellipse by radius change over a major and minor axis and to assess the orientation of the most prominent axis in case of asymmetry. A major advantage of the fitting procedure is that the influence of outliers, segmentation inaccuracies, and small manual corrections is minimized as the ellipse reflects the general trend in the plots.
In addition, we were able to measure subvoxel changes (changes within the spatial resolution), with an intraobserver variability of 0.15 mm for radius change. This was accomplished by up-sampling the images: smaller voxels were created in the y and x direction (factor 8). Subvoxel measurements were further aided by measuring the radius change from the center of mass and application of an elliptical model, which takes into account partial volume effects along the total aortic circumference. Using digital simulations of pulsatile aortas, we found that a radius change of 5% could still be accurately measured, and that the accuracy in detecting asymmetry decreases below a 5% radius change. The observed radius changes were significant because they exceeded the intraobserver variability.
The levels we studied are relevant landing zones for EVAR. The zone along which most endografts will appose is 1 cm below the lowest renal artery. The juxtarenal and suprarenal levels have become relevant with the extension to juxtarenal and suprarenal repair. Furthermore, we aimed to study the procession of the aortic distention proximal to the most common sealing zone.
We have observed a variation in the rate of asymmetry and magnitude of distention per patient and per level. Asymmetric expansion was observed at all levels of the aneurysm neck. Asymmetric expansion appears to be present throughout the aorta, which was similar to our observations in young healthy volunteers.20 An important observation is that some patients have pronounced asymmetry and high radial changes, and the asymmetry in these patients is likely to be more relevant. For example, for one patient, a maximum radius change of 2.0 mm vs a minimum radius change of 1.1 mm was calculated. When interpreting the presented radius changes over the major and minor axis, one must realize that the diameter changes are even higher.
The focus of our study was to detect asymmetry in the aortic expansion during the cardiac cycle. However, this study also emphasizes the importance of over-sizing to reach an appropriate and durable proximal seal with the pulsatile aorta. Our observation may have consequences for endograft design because asymmetric distention may affect endograft sealing and fixation and may pose different forces on different parts of the endograft. Obviously, the proximal part of the aortic endograft should be able to adapt to conformational changes of the aorta throughout the cardiac cycle. An asymmetric expansion will complicate this adaptation. The proximal part of the endograft should probably have a high radial force to press itself to the vessel wall and maintain in close contact with the wall throughout the cardiac cycle. Additional barbs will help to maintain the close graft–vessel contact. Furthermore, to adapt to the pulsatile aortic shape changes, the proximal part of the endograft should probably be flexible and compliant. Therefore, Z- or M-shaped stent rings seem preferable. Inability of the graft to adapt to the aortic morphology changes throughout the cardiac cycle could lead to intermittent pressurization of the aneurysm sac, endoleaks, and compromised durability.
It is of interest that at all three levels studied, including the infrarenal proximal fixation zone, the distention and asymmetry remained preserved after EVAR. All patients in this study were treated with the Talent (Medtronic, Minneapolis, Minn) or Excluder (W. L. Gore & Associates, Flagstaff, Ariz) device. These devices are composed of self-expandable nitinol stents with Z-shaped stent rings. This design of the stent rings, combined with proximal oversizing, probably allows the stent graft to follow the aortic wall closely during the cardiac cycle. This explanation is in line with the in vitro observations of Schurink et al,23 who demonstrated with IVUS that the Gianturco Z-stent (William Cook Europe A/S, Bjaeverskov, Denmark) was able to follow the aortic wall closely during the cardiac cycle and that the Palmaz stent graft (Cordis/Johnson & Johnson Co, Warren, NJ), unlike the aorta, did not expand during the systolic phase.
Our findings raise interesting questions, such as what effect placement of stent grafts with different designed (circular) stent rings would have on asymmetric aortic expansion and what the long-term consequences of endograft placement on aortic distention would be. We hypothesize that there is a correlation between the degree of distention and asymmetry and the incidence of complications related to graft fixation and seal such as proximal (intermittent) endoleaks and graft migration. The influence of asymmetric aortic distention on long-term EVAR outcome and durability has to be awaited.
The advantage of this study over other studies is that we were able to measure the radial changes over 360 axes, yielding a complete mapping of the aortic expansion instead of only measuring in one or two directions.14, 15 Although PC-MRI studies have studied circumferential wall displacement and found a nonuniform expansion, these studies focused on circumferential strain and deformation rather then radius changes.16, 17, 18 Moreover, these studies did not study the aneurysm neck and the influence of stent graft placement.
The purpose of our study was to assess the changes of the geometric appearance of the aorta and demonstrate that the pulsatile aortic expansion in the aneurysm neck is asymmetric. By expressing the asymmetric aspect with a major and a minor axis, and their ratios, a logical and comprehensible measure for asymmetry is offered. Measuring radii from the aortic center of mass at each cardiac phase is essential when determining asymmetry. The center of mass displaces during each cardiac phase as a result of in-plane translation and expansion. Such analysis is very hard to accomplish using IVUS because the center of the US catheter, which moves with every heartbeat, should be placed exactly in the center of the aorta. ECG-gated MRI is noninvasive, not operator-dependent, and offers a clear overview of the aortic distention in relation to the anatomy.
One limitation is that we only measured the in-plane movement of the aorta. Although we adjusted our measurements for aortic translation in the transverse plane, we are not able to correct the out of plane (craniocaudal) movement of the aorta at this moment. Any possible longitudinal distention could not be measured and may also influence endograft fixation and seal. A three-dimensional approach is theoretically possible using a clinical MR scanner, although this is not feasible because of the extremely long scanning time. When coronal and sagittal ECG-gated images are observed, however, the out-of-plane movement seems to be minimal in the abdominal aortic region.
Further, juxtarenal distention must be interpreted with care. We observed that the circumferential regions of low distention throughout the cardiac cycle corresponded with the regions of debranching renal arteries. This might be due to manual segmentation necessary at those sites. We did not study the expansion in the aneurysm sac because very little wall motion occurs at that level, which would make asymmetry measurements difficult and less reliable.24
In general, the direction of maximum aortic distention is AP oriented, ranging from right–anterior to left–anterior. We were surprised to see that preoperatively, the direction of maximum distention above the renal arteries showed a tendency to right–anterior and that below the renal arteries there was a tendency to left–anterior. Spiralling and retrograde flow patterns have both been described in the ascending thoracic aorta, and the flow might also be spiralling in the abdominal aorta.25, 26, 27 These flow dynamics may pose different forces on different parts and levels of the aorta, with subsequent different degrees and directions of aortic distention during the cardiac cycle. Further, the direction of aortic distention might be related to the surrounding tissues and debranching arteries. The direction of the debranching celiac trunk and superior mesenteric artery (most commonly slightly right–anteriorly), and the inferior mesenteric artery (most commonly slightly left–anteriorly) might not be based on coincidence but related to flow and distention patterns.
Different forces on the aortic wall and anisotropic expansion may explain the circumferential differences in wall thickness and architecture.14 The observed magnitude and direction of distention varied by patient and might be a function of aneurysmatic disease progression. Flow, circumferential distention, shear stresses, wall thickness, and tissue histology are closely related. Ideally in the future, all these elements will be synthesized in a model that explains aneurysm growth and is able to predict aneurysm rupture. Future studies, integrating the mentioned characteristics, are anticipated.
Conclusions
We used ECG-gated MRI with sophisticated postprocessing to measure the asymmetric aspect of pulsatile aortic shape changes throughout the cardiac cycle. We observed that the rate of asymmetric distention varied by patient and level. The presence of asymmetric aortic expansion in the aneurysm neck is an important observation. It may have consequences for endograft design because it probably affects endograft sealing, especially in those patients with high radius changes and asymmetry ratios. Asymmetric expansion remained preserved after stent graft placement. The stent grafts with Z stent rings used in the study participants seem to adapt to the aortic shape changes well.
Author contributions
We would like to thank Sara M. Sprinkhuizen and Gijsbrecht K. W. Barwegen for their help with the analysis.
References
- Initial results of wireless pressure sensing for endovascular aneurysm repair: the APEX Trial–Acute Pressure Measurement to Confirm Aneurysm Sac EXclusion. J Vasc Surg. 2007;45:236–242
- Endovascular repair of acute type B aortic dissection: early and mid-term results. J Vasc Surg. 2006;43:1090–1095
- Endovascular management of traumatic ruptures of the thoracic aorta: a retrospective multicenter analysis of 28 cases in The Netherlands. J Vasc Surg. 2006;43:1096–1102discussion 1102
- . Reduced pulsatile wall motion of abdominal aortic aneurysms after endovascular repair. J Vasc Surg. 1998;27:624–631
- . Aortic motion: a potential pitfall in CT imaging of dissection in the ascending aorta. J Comput Assist Tomogr. 1995;19:569–572
- Dynamic magnetic resonance angiography of the aneurysm neck: conformational changes during the cardiac cycle with possible consequences for endograft sizing and future design. J Vasc Surg. 2006;44:22–28
- Use of cine magnetic resonance angiography in quantifying aneurysm pulsatility associated with endoleak. J Vasc Surg. 2003;38:652–656
- . Aorta wall motion monitoring by 1-d MRI of perpendicular diameters. J Magn Reson Imaging. 1999;10:833–840
- Dynamic cine-CT angiography for the evaluation of the thoracic aorta; insight in dynamic changes with implications for thoracic endograft treatment. Eur J Vasc Endovasc Surg. 2006;32:532–536
- Toward endografting of the ascending aorta: insight into dynamics using dynamic cine-CTA. J Endovasc Ther. 2007;4:551–560
- Use of dynamic computed tomography to evaluate pre- and postoperative aortic changes in AAA patients undergoing endovascular aneurysm repair. J Endovasc Ther. 2007;14:44–49
- Dynamic aortic changes in patients with thoracic aortic aneurysms evaluated with electrocardiography-triggered computed tomographic angiography before and after thoracic endovascular aneurysm repair: preliminary results. Ann Vasc Surg. 2008;[Epub ahead of print]
- Toward endografting of the ascending aorta: insight into dynamics using dynamic cine-CTA. J Endovasc Ther. 2007;14:551–560
- . Dynamic geometry and wall thickness of the aortic neck of abdominal aortic aneurysms with intravascular ultrasonography. J Vasc Surg. 2007;46:891–897
- . Increased anterior abdominal aortic wall motion: possible role in aneurysm pathogenesis and design of endovascular devices. J Endovasc Ther. 2007;14:574–584
- Quantification of vessel wall cyclic strain using cine phase contrast magnetic resonance imaging. Ann Biomed Eng. 2002;30:1033–1045
- Quantification of vessel wall motion and cyclic strain using cine phase contrast MRI: in vivo validation in the porcine aorta. Magn Reson Med. 2004;52:286–295
- . Measurement of vessel wall strain using cine phase contrast MRI. J Magn Reson Imaging. 2002;15:418–428
- The effect of angulation on intravascular ultrasound imaging observed in vascular phantoms. J Endovasc Surg. 1998;5:126–133
- Aortic pulsatile distention in young healthy volunteers is asymmetric: analysis with ECG-gated MRI. Eur J Vasc Endovasc Surg. 2009;37:168–174
- . Direct least square fitting of ellipses. IEEE Trans Pattern Anal Mach Intell. 1999;21:476–480
- . Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–310
- . Stent attachment site-related endoleakage after stent graft treatment: an in vitro study of the effects of graft size, stent type, and atherosclerotic wall changes. J Vasc Surg. 1999;30:658–667
- . Aortic aneurysm pulsatile wall motion imaged by cine MRI: a tool to evaluate efficacy of endovascular aneurysm repair?. Eur J Vasc Endovasc Surg. 2002;23:158–161
- . Spiral laminar flow in the abdominal aorta: a predictor of renal impairment deterioration in patients with renal artery stenosis?. Nephrol Dial Transplant. 2004;19:1786–1791
- . Comparison of flow patterns in ascending aortic aneurysms and volunteers using four-dimensional magnetic resonance velocity mapping. J Magn Reson Imaging. 2007;26:1471–1479
- . Two-dimensional flow quantitative MRI of aortic arch blood flow patterns: effect of age, sex, and presence of carotid atheromatous disease on prevalence of spiral blood flow. J Magn Reson Imaging. 2003;18:169–174
Competition of interest: none.
PII: S0741-5214(09)00499-6
doi:10.1016/j.jvs.2009.02.216
© 2009 Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
