Prediction of aortoiliac stent-graft length: comparison of measurement methods

CT angiography after 20 years: a transformation in cardiovascular disease characterization continues to advance

Through a marriage of spiral computed tomography (CT) and graphical volumetric image processing, CT angiography was born 20 years ago. Fueled by a series of technical innovations in CT and image processing, over the next 5-15 years, CT angiography toppled conventional angiography, the undisputed diagnostic reference standard for vascular disease for the prior 70 years, as the preferred modality for the diagnosis and characterization of most cardiovascular abnormalities. This review recounts the evolution of CT angiography from its development and early challenges to a maturing modality that has provided unique insights into cardiovascular disease characterization and management. Selected clinical challenges, which include acute aortic syndromes, peripheral vascular disease, aortic stent-graft and transcatheter aortic valve assessment, and coronary artery disease, are presented as contrasting examples of how CT angiography is changing our approach to cardiovascular disease diagnosis and management. Finally, the recently introduced capabilities for multispectral imaging, tissue perfusion imaging, and radiation dose reduction through iterative reconstruction are explored with consideration toward the continued refinement and advancement of CT angiography.

Three-dimensional blood vessel quantification via centerline deformation

It is clinically important to quantify the geometric parameters of an abnormal vessel, as this information can aid radiologists in choosing appropriate treatments or apparatuses. Centerline and cross-sectional diameters are commonly used to characterize the morphology of vessel in various clinical applications. Due to the existence of stenosis or aneurysm, the associated vessel centerline is unable to truly portray the original, healthy vessel shape and may result in inaccurate quantitative measurement. To remedy such a problem, a novel method using an active tube model is proposed. In the method, a smoothened centerline is determined as the axis of a deformable tube model that is registered onto the vessel lumen. Three types of regions, normal, stenotic, and aneurysmal regions, are defined to classify the vessel segment under-analyzed by use of the algorithm of a cross-sectional-based distance field. The registration process used on the tube model is governed by different region-adaptive energy functionals associated with the classified vessel regions. The proposed algorithm is validated on the 3-D computer-generated phantoms and 3-D rotational digital subtraction angiography (DSA) datasets. Experimental results show that the deformed centerline provides better vessel quantification results compared with the original centerline. It is also shown that the registered model is useful for measuring the volume of aneurysmal regions.

Methods for Three-Dimensional Geometric Characterization of the Arterial Vasculature

Complex vascular anatomy often affects endovascular procedural outcome. Accurate quantitative assessment of three-dimensional (3D) in-vivo arterial morphology is therefore vital for endovascular device design, and preoperative planning of percutaneous interventions. The aim of this work was to establish geometric parameters describing arterial branch origin, trajectory, and vessel curvature in 3D space that eliminate the errors implicit in planar measurements. 3D branching parameters at visceral and aortic bifurcation sites, as well as arterial tortuosity were determined from vessel centerlines derived from magnetic resonance angiography data for three subjects. Errors in coronal measurements of 3D branching angles for the right and left renal arteries were 3.1 +/- 3.4 degrees and 7.5 +/- 3.7 degrees , respectively. Distortion of the anterior visceral branching angles from sagittal measurements was less pronounced. Asymmetry in branching and planarity of the common iliac arteries was observed at aortic bifurcations. The renal arteries possessed considerably greater 3D curvature than the abdominal aorta and common iliac vessels with mean average values of 0.114 +/- 0.015 and 0.070 +/- 0.019 mm(-1) for the left and right, respectively. In conclusion, planar projections misrepresented branch trajectory, vessel length, and tortuosity proving the importance of 3D geometric characterization for possible applications in planning of endovascular interventional procedures and providing parameters for endovascular device design.

Correlation between the Measurement of Transverse Diameter in the Proximal Neck on Computed Tomography and on Aortography before Endovascular Treatment of Infrarenal Aortic Aneurysm

The aim of this study was to determine the correlation between the measurement of transverse diameter of the proximal neck on computed tomographic angiography (CTA) and graduated catheter aortography in patients who are candidates for endovascular graft placement in order to replace, if both measurements are equivalent, aortography for CTA alone. Preoperative dual-slice CTA and graduated catheter aortography were performed in 35 consecutive patients with infrarenal aortic aneurysm within 10 days. Transverse proximal neck diameters were measured on a true axial section on CTA reconstructions and on aortographic images, always 6 mm distal from the most inferior main renal artery. Mean, median, and standard deviation were obtained and the measurements correlated for each patient using Pearson's correlation and linear regression analysis. A significant difference in proximal neck transverse diameter measurements was found between graduated catheter aortography and CTA in all cases. CTA values were a mean of 1.74 mm higher than aortography values. Pearson's correlation indicates a strong correlation between both techniques, and a regression equation determines the predictive value of aortography on the basis of CTA values. Estimation of the transverse diameter of the proximal neck on aortography on the basis of that obtained on CTA allows us to affirm that CTA could be used as the sole method for the preoperative selection of appropriate endograft size in patients with infrarenal aortic aneurysm.

Multidetector CT Evaluation in Arterial Stenting

BACKGROUND

Multidetector CT (MDCT) represents breakthrough in CT technology, significantly improving CT Angiography applications.

METHODS

Twenty one patients with aortoiliac & branch aneurysms or stenosis were evaluated by Digital Subtraction Angiography (DSA) and Multidetector CT (MDCT) before and after endovascular repair.

RESULTS

There were eight cases of aortic & branch aneurysms and 13 with stenosis. Four cases had aortic aneurysms, while one case had left subclavian artery aneurysm, thoracic aneurysm, femoral and popliteal artery pseudoaneurysms. Of the 13 cases with stenotic lesions, iliac stenosis was seen in eight patients. The others included carotid, vertebral, aortic, renal and aortic bifurcation stenotic. MDCT offered accurate information on shape and size of aneurysm, shape and patency of graft, the presence or absence of perigraft thrombosis or endoleaks, while in stenotic lesions it provided useful information on shape of graft, its location, its patency and the presence and quantity of distal flow.

CONCLUSION

MDCT was found to be a potentially useful modality during initial evaluation and follow up of patient undergoing endovascular repair.

Prediction of altered endograft path during endovascular abdominal aortic aneurysm repair with the Gore Excluder

OBJECTIVE

During endovascular abdominal aortic aneurysm (AAA) repair (EVAR), the rapid deployment of the Gore Excluder endograft may be associated with anatomic shortening of the endograft path. This shortened path may result in coverage of the hypogastric artery origin or overly conservative graft length selection that may lead to unnecessary extensions. We quantified the degree of path alteration with this endograft and developed an algorithm to predict it.

METHODS

Preoperative and postoperative three-dimensional (3D) computed tomographic (CT) scans were evaluated for 50 consecutive patients with Gore Excluder endografts by using 21 anatomic measurements and 6 calculated indices. Measurements were evaluated as if only 3D lumen centerline measurements were available, rather than complete 3D computer-aided measurement and "virtual graft" simulation. Tortuosity was quantitated from the renal artery to the hypogastric origin, using the difference between a straight line and the lumen centerline.

RESULTS

The endograft was deployed successfully in all cases. The graft end points were typically quite close to the preoperative plan: mean renal artery-to-graft distance was within 2.0 +/- .5 mm, and the limb end point-to-hypogastric origin differed by an average of only 1.8 +/- 1.6 mm. Although accurate in most cases, the actual graft path shortened 1 cm or more relative to the centerline in 11% of limbs. On univariate analysis, determinants of alteration of >1 cm in the graft deployment path were (1) aortoiliac tortuosity (renal-to-hypogastric artery, P < .002), (2) the degree of planned graft rotation (73% of cases altered >10 mm were in the rotated position, P < .05), and (3) the insertion side (73% of alterations >or=10 mm were ipsilateral to the main device, P < .05). On multivariate analysis, the renal-to-hypogastric artery tortuosity index (RHTI) was significant ( P < .004), and device type and rotation approached significance ( P < .08). We developed a classification scheme based on RHTI to predict the risk of alteration of the graft path >or=1 cm (low risk, 0%; medium risk, 10%; high risk, 25%) and an algorithm to predict the degree of alteration of the anatomy that reduced the number of cases shortening >or=1 cm to zero.

CONCLUSIONS

The graft deployment path will be altered significantly in a minority of cases with the Gore Excluder endograft, but this can cause hypogastric occlusion or other problems. Anatomic shortening is predictable from morphologic features such as tortuosity, graft insertion side, and rotation. We developed an algorithm based on a tortuosity index that quantitates the risk and degree of shortening associated with endograft deployment.

Creation of individual ideally shaped stents using multi-slice CT: in vitro results from the semi-automatic virtual stent (SAVS) designer

To plan stent-grafting for thoracic aortic aneurysm with complicated morphology, we created a virtual stent-grafting program [Semi Automatic Virtual Stent (SAVS) designer] using three-dimensional CT data. The usefulness of the SAVS designer was evaluated by measurement of transformed anatomical and straight stents. Curved model images (source, multi-planer reconstruction and volume rendering) were created, and a hollow virtual stent was produced by the SAVS designer. A straight Nitinol stent was transformed to match the curved configuration of the virtual stent. The accuracy of the anatomical stent was evaluated by experimental strain phantom studies in comparison with the straight stent. Mean separation length was 0 mm in the anatomical stent [22 mm outer diameter (OD)] and 5 mm in the straight stent (22 mm OD). The straight stent strain voltage was four times that of the anatomical stent at the stent end. The anatomical stent is useful because it fits the curved structure of the aorta and reduces the strain force compared to the straight stent. The SAVS designer can help to design and produce the anatomical stent.

Endovascular Repair of Thoracic and Abdominal Aortic Aneurysms: Pre- and Postprocedural Imaging

Endovascular repair of thoracic and abdominal aortic aneurysms is a safe alternative to conventional open surgical repair. Clinical success, however, is highly dependent on patient selection. Diagnostic vascular imaging has an essential role for this selection process. Following endovascular aneurysm repair (EVAR), patients require long-term surveillance and again vascular imaging serves an integral function. This article reviews EVAR selection criteria and post-EVAR assessment and then discusses the imaging modalities used to evaluate these patients, namely multi-detector-row computed tomographic angiography, magnetic resonance imaging/angiography, duplex ultrasonography, and catheter angiography.

Abdominal aortic aneurysms: virtual imaging and analysis through a remote web server

The study describes the application of a web-based software in the planning of the endovascular treatment of abdominal aortic aneurysms (AAA). The software has been developed in the framework of a 2-year research project called Aneurysm QUAntification Through an Internet Collaborative System (AQUATICS); it allows to manage remotely Virtual Reality Modeling Language (VRML) models of the abdominal aorta, derived from multirow computed tomography angiography (CTA) data sets, and to obtain measurements of diameters, angles and centerline lengths. To test the reliability of measurements, two radiologists performed a detailed analysis of multiple 3D models generated from a synthetic phantom, mimicking an AAA. The system was tested on 30 patients with AAA; CTA data sets were mailed and the time required for segmentation and measurement were collected for each case. The Bland-Altman plot analysis showed that the mean intra- and inter-observer differences in measures on phantoms were clinically acceptable. The mean time required for segmentation was 1 h (range 45-120 min). The mean time required for measurements on the web was 7 min (range 4-11 min). The AQUATICS web server may provide a rapid, standardized and accurate tool for the evaluation of AAA prior to the endovascular treatment.

Decreased use of iliac extensions and reduced graft junctions with software-assisted centerline measurements in selection of endograft components for endovascular aneurysm repair

OBJECTIVE

The purpose of this study was to determine the impact of using computerized software-assisted centerline measurements for extensions and graft junctions during the selection of endograft components for modular aortic endografts in endovascular repair of abdominal aortic aneurysms.

METHODS

From April 1998 to December 2002, 289 modular aortic endografts were implanted at our institution. These included 248 grafts (prior to 2002, group 1) with components selected on the basis of manual caliper measurements from combined contrast computed tomography (CT) and marker-catheter arteriography data, and 41 grafts (2002, group 2) with components selected with the use of computerized software that allowed for centerline measurements on 3-dimensional reconstructions based on CT data. These 2 groups were compared for the number and type of extensions required per case. Seventeen other relevant variables were analyzed for their potential influence on selection of endograft components. These variables included age, gender, maximum aneurysm size, level of distal fixation, length and diameter at the fixation points, endograft manufacturer (make), and configuration. The significance of the observed differences was analyzed with a multivariate regression model, adjusting for potentially confounding preoperative measures.

RESULTS

Multivariate analysis demonstrated that the number of right iliac extensions, left iliac extensions, total extensions, and total graft junctions was significantly reduced by the use of computerized software-assisted centerline measurements (group 2) compared with caliper measurements (group 1), independent of all other 17 preoperative variables. Notably, the mean number of required right iliac extensions was double in group 1 versus group 2.

CONCLUSIONS

Centerline software-assisted measurements can significantly reduce the need for iliac extensions and, concomitantly, the number of required endograft junctions. On average, twice as many extensions were required for right iliac fixation when the manual caliper measurements were used compared with software-assisted measurements. These findings are highly relevant to issues of total endograft cost and long-term endograft integrity and focus attention on the tools that may need to be considered standards of care rather than optional for selection of endograft components.

Assessment of Automatic Vessel Tracking Techniques in Preoperative Planning of Transluminal Aortic Stent Graft Implantation

OBJECTIVE

To evaluate automatic vessel tracking techniques in the course of preoperative planning prior to transluminal aortic endograft implantation by comparing accuracy, reproducibility, and postprocessing time with source image and volume-rendered analysis methods.

METHODS

Multislice computed tomography datasets of 5 patients with abdominal aortic aneurysms were preoperatively examined, performing volumetric analysis of diameter and position of renal artery orifices, aneurysmal neck, maximal aneurysmal extension, aortic bifurcation, and iliac arteries and bifurcation. Analysis was realized by utilizing transverse datasets, volume rendering, and automated vessel tracking strategies (MxView, Philips, Best, The Netherlands). Measurement techniques were evaluated by 2 independent readers 3 times for each patient and measurement modality. Statistical analysis evaluated accuracy of the measurements and intra- and interobserver reliability. Postprocessing time was documented.

RESULTS

Using transverse source datasets, intraobserver reliability ranged from 0.49 to 0.58. Intraobserver reliability improved to 0.7 to 0.98 when volume-rendered datasets were evaluated. Interobserver variability for transverse and volume-rendered datasets ranged from 0.49 to 0.76 and 0.70 to 0.96, respectively. Automated vessel tracking datasets did not demonstrate any intra- or interobserver variability. Based on transverse datasets, the length and diameter of iliac arteries and location and diameter of the aneurysmal neck were measured as statistically different in all cases in contrast to volume rendering and automated segmentation techniques. Postprocessing time consumption for measurements based on transverse, volume-rendered, and automated tracking segmentation datasets averaged 3.32 minutes, 25.43 minutes, and 2.24 minutes, respectively.

CONCLUSIONS

Preoperative measurements improve significantly if datasets are evaluated based on volume-rendering techniques. This time-consuming procedure can be shortened, while further reducing observer variability, with automatic segmentation techniques.

Two-dimensional versus three-dimensional CT scan for aortic measurement

PURPOSE

To examine if 3-dimensional (3D) reconstructions of computed tomographic (CT) data, by imaging perpendicular to blood flow, can improve aortic diameter measurement accuracy over axial (2D) CT.

METHODS

Two independent, blinded observers used electronic calipers to measure the minor axis and the line perpendicular to it on 40 2.5-mm 2D CT scans from 31 patients. A circular electronic tool was used to estimate diameters on 3D reconstructions from the same 40 scans. Measurements of the aortic neck were obtained 5 mm below the renal arteries and the widest slice of the aneurysm was used to measure sac diameter. Only the minor axis was measured at the iliac arteries immediately above the left (LI) and right (RI) iliac bifurcations. Datasets were compared with an intraclass correlation coefficient (ICC), Bland and Altman variation assessments, and absolute differences.

RESULTS

ICC between 2D and 3D scans demonstrated high correlation with 2D minor axis measurements (neck=0.9282, sac=0.8956, RI=0.8755, LI=0.7381). 3D to 2D major axis correlation was lower (neck=0.6388, sac=0.8995). Variation between 3D and 2D minor axis measurements was low (0.51-mm average variation from the mean for the minor axis and 1.30-mm variation for the major axis). Average absolute difference between 3D and 2D diameters was 1.01 mm (minor axis) versus 2.61 mm (major axis). Interobserver correlation was highest for sac measurements both in 2D minor axis (ICC=0.8990) and 3D (ICC=0.9518).

CONCLUSIONS

Minor axis measurements on axial CT scan can substitute for diameters obtained from 3D reconstructions in most clinical situations.

Prediction of aortoiliac stent graft length: comparison of a semiautomated computed tomography angiography method and calibrated aortography

OBJECTIVE

The aim of this study is to compare multislice computed tomography (MSCT) in combination with a newly developed semiautomated software program with calibrated aortography in patients who are scheduled for endovascular aortic stent graft placement.

METHODS

From November 2000 until December 2001, seven patients with an abdominal aortic aneurysm (AAA) underwent both calibrated aortography and MSCT for preoperative endovascular stent graft planning. Both studies were performed within 14 days. Further, length measurements were performed with a semiautomated computerized tomographic angiography (CTA) calibration method and a conventional calibrated aortography technique using three differently configured tubes with variable tortuosity. The AAA length measurements of the semiautomated CTA calibration method and the calibrated aortography were compared.

RESULTS

Statistical analysis included linear regression analysis and revealed a probability value of 0.000381 and an r2 value of 0.93. Using phantoms, it is proven by the authors that the accuracy of the semiautomated CTA calibration method increases with increasing tortuosity when compared with the conventional calibrated aortography technique.

CONCLUSIONS

Our preliminary results show that the semiautomated CTA calibration method has a potentially advantageous role in preoperative stent graft planning regarding the aortic length measurements and seems to be more accurate than calibrated aortography, especially in extremely tortuous vessels. Further studies have to be performed, however.

Evaluation of aortoiliac aneurysm before endovascular repair: comparison of contrast-enhanced magnetic resonance angiography with multidetector row computed tomographic angiography with an automated analysis software tool

PURPOSE

The purpose of this study was to assess accuracy and reliability of a volumetric analysis of abdominal aneurysms on the basis of multidetector row computed tomographic angiography (CTA) and magnetic resonance angiography (MRA) with a commercially available automated vessel analysis software program.

MATERIALS AND METHODS

Twenty patients with abdominal aortic aneurysms underwent preoperative CTA and MRA before endovascular repair. Postdeployment CTA was performed in 15 of these 20 patients (75%). All preoperative CTA and MRA and postdeployment CTA data sets were analyzed with an automated software tool. The length of the stent grafts on postdeployment CTA was measured and compared with the true length of the primary component. Two readers independently evaluated 13 vessel parameters on preoperative CTA and MRA, which are considered to be important in planning stent graft deployment.

RESULTS

With the automated analysis software tool, all measurements could be performed on either CTA or MRA data sets. There was no statistically significant difference between postdeployment measurements of stent graft length on CTA and the true dimensions of the implanted stent grafts. Interobserver agreement for all of the measurements with either CTA or MRA was good to excellent (interclass coefficient, 0.71 to 0.99) with only minimal mean differences of measured dimensions between both readers (range, -2.0 to +2.3 mm, Bland-Altman). Intermodality agreement between CTA and MRA was good to excellent (interclass coefficient, 0.62 to 0.98) with small mean differences of measured dimensions between both methods (range, -4.1 to +2.1 mm, Bland-Altman).

CONCLUSION

Volumetric measurement with an automated analysis software tool allows a fast, precise, and reliable noninvasive preoperative determination of all aortic dimensions on the basis of either CTA or MRA data sets.

Reconstruction and web distribution of measurable arterial models

In this paper a novel framework for the segmentation, 3D reconstruction and web distribution of vessel structures specifically tailored to the assessment of abdominal aortic aneurysms for endovascular surgery planning is presented. Deformable models are used for segmentation, while VRML97 and ECMA scripting are used to obtain models that are not only viewable from any VRML97 enabled browser, but that also allow users to perform, directly from standard web browsers, guided measurements of geometrical parameters, relevant to surgical planning.

Imaging of aortic stent-grafts and endoleaks

Although the technical success of stent-graft implantation is established and relatively safe, data on the long-term safety and efficacy of endovascular repair are just emerging. Because several late complications of aortic stent-graft placement have been observed, life-long follow-up remains essential. Imaging methods form an integral part of every stage of endovascular aortic aneurysm repair. The current imaging strategy should include initial plain films, CT angiography, and color-coded Duplex sonography. Plain films are an excellent means to detect migration, angulation, kinking, and structural changes of the stent mesh, including material fatigue, at follow-up. Helical CT angiography is considered a potentially revolutionary method for the noninvasive complete postprocedural assessment of aortic sten-grafting. Current data justify the use of biphasic C angiography as the postprocedural imaging technique of choice in most patients [118]. Ultrasound offers the advantages of low cost and lack of radiation exposure. High-quality ultrasound reliably excludes endoleaks in patients after stent-grafting of AAAs. There is a substantial variability, however, in measuring the diameter of aneurysm sacs; thus, confirmation using an alternative study is prudent in cases that demonstrate a significant change in size during follow-up. MR angiography serves as an attractive alternative to CT angiography in patients with impaired renal function or known allergic reaction to iodinated contrast media. With current techniques, the visualization of aortic stent-grafts (with the exception of stainless-steel-based devices) is sufficient with MR angiography. There is evidence that MR imaging is superior to CT angiography in detecting small type 2 endoleaks or for excluding retrograde perfusion in patients with suspected endotension. The role of diagnostic catheter angiography is limited to assessment of vascular pathways in equivocal cases or for suspected endotension. Currently, a consensus view about postprocedural management after aortic stent-graft implantation is lacking. The authors propose performing a baseline CT angiography at discharge and a biphasic CT angiography and Duplex ultrasound scan at three months. In patients with no evidence of an endoleak, CT angiography, plain film and Duplex sonography (abdomen) should be repeated every year after endovascular repair. If an endoleak is present at follow-up, immediate appropriate treatment should be initiated.