CCS

Esialaredpsoiaft

ti

C

A

1

ross-Sectional Imaging Studies: Whatan We Learn and What Do We Need to Know?

cott D. Flamm, MD

Rapid, noninvasive imaging approaches can provide novel diagnostic information and,when effectively interpreted and implemented in a therapeutic strategy, can simplifyprocedures. Endovascular therapy of thoracic and abdominal aortic disease represents adramatic shift in treatment of thoracoabdominal aortic disease, but one that requires achange in the knowledge base regarding both the morphology and pathophysiology ofaortic disease and the interaction with interventional devices. As a result, the demands oncross-sectional imaging have increased commensurately with the complexity of the ther-apeutic options, but advances in cross-sectional imaging have kept pace. Current com-puted tomography (CT) and magnetic resonance imaging (MRI) technologies providedetailed morphologic assessment, and are advancing rapidly into more sophisticatedphysiologic evaluation of aortic disease. These advances may more effectively triagepatients to appropriate therapy, or exclude patients from unnecessary invasive procedures.The information gleaned from CT and MRI studies is critical for the vascular surgeon whowants to identify appropriate vascular territories for intervention, plan a detailed approach,and develop sophisticated surveillance strategies.Semin Vasc Surg 20:108-114 © 2007 Elsevier Inc. All rights reserved.

bdmt

cpamg

iaosttf

TNbot

NDOVASCULAR THERAPY OF thoracic and abdominalaortic disease has presented a paradigm shift in diagno-

is, assessment, and treatment of aortic aneurysms since itsnception in 1991.1 Prior to endovascular stent placement,ssessment of abdominal aortic aneurysms may have beenimited to maximum aneurysm sac diameter, length of theneurysm, and potential involvement of branch vessels. Cur-ent assessment now requires knowledge of the presence andxtent of intraluminal thrombus and atherosclerotic plaque,iameter and length of the neck(s), position, number, andrecise measurements of renal and mesenteric arteries fortent fenestrations and/or side stents, and quality and caliberf access vessels (namely, the pelvic arterial system). Forntervention in abdominal aortic aneurysms, this list servess the beginning, but is not necessarily all-inclusive. Evenurther considerations must be raised for stents that involvehe thoracic aorta.

Previously, postsurgical assessment of graft placement wasypically limited to the assessment of anastomotic stenosis,nfection, or leak. Follow-up of endovascular stent therapy

ardiovascular Imaging, Department of Radiology, Cleveland Clinic, Cleve-land, OH.

ddress reprint requests to Scott D. Flamm, Cardiovascular Imaging, De-partment of Radiology, Hb-6, Cleveland Clinic, Cleveland, OH 44195.

ME-mail: flamms@ccf.org

08 0895-7967/07/$-see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1053/j.semvascsurg.2007.04.004

rings with it additional considerations: a multiplicity of en-ostent leaks, endostent migration, limitations on imagingodality compatibilities, and more recently pressure inves-

igations, to name just a few.Clearly, the demands on cross-sectional imaging have in-

reased commensurately with the complexity of the thera-eutic options. Fortunately, advances in cross-sectional im-ging have kept pace with these demands for detailedorphologic assessment, and are advancing rapidly into

reater physiologic evaluation of aortic disease.This article will address the current state of the art in non-

nvasive vascular imaging using computed tomography (CT)nd magnetic resonance imaging (MRI) relative to the needsf vascular surgeons. Three areas will be highlighted: (1)tate-of-the-art CT and MRI technologies and their capabili-ies; (2) intrinsic advantages and limitations of CT and MRIechnology; and (3) a look at applications for the near futureor vascular diseases amenable to interventional approaches.

echniquesoninvasive imaging of the aorta is preferred, when feasible,ecause of the cost, inconvenience, and associated morbidityf conventional catheter x-ray angiography. It is also an areahat highlights differences in successful application of CT and

RI technology to a cardiovascular diagnostic problem. CT

harpnehsisce

CCdasittTa

ts1irsmoC3pttatiscri

ttqevaioqp

p

Fm(racoabdominal aorta.

Cross-sectional imaging studies 109

as the advantage of speed of application, spatial resolution,nd relative simplicity, but can be limited in patients withenal dysfunction. From a theoretical standpoint MRI is aarticularly powerful tool as a result of the lack of potentiallyephrotoxic contrast agent and radiation exposure. How-ver, a significant proportion of patients with aortic diseaseave contraindications to MRI, and some stent materials (eg,tainless steel) pose particular problems with artifacts render-ng MR images nondiagnostic.2 In addition, most vascularurgeons and other clinicians have a greater familiarity andomfort level with CT imaging, making it a de facto prefer-nce.

Tomputed axial tomographic imaging began with a singleetector acquiring one thick image slice per rotation, withcquisition of the entire chest, abdomen, and pelvis requiringeveral minutes. The speed of acquisition improved with thencorporation of spiral acquisition techniques where the de-ector gantry rotated and acquired data continuously whilehe table moved slowly through and past the detector system.his advance allowed for thinner slice acquisition, as well asreduction in the time of acquisition.Today, multidetector computed tomography (MDCT) is

he standard for CT angiography (CTA) of the cardiovascularystem.3,4 MDCT uses a rotating gantry typically containing6 to 64 closely spaced detector rings with an x-ray generat-

ng source placed 180 degrees opposite in the same rotatinging. The in-plane resolution is approximately 0.5 mm, withlice thicknesses, depending on the manufacturer, from 0.4m to 0.6 mm. As a result of thinner slices, spatial resolution

f MDCT is substantially superior to previous generations ofT scans. In addition, faster gantry rotation times (down to30 milliseconds) have improved temporal resolution to ap-roximately 150 to 200 milliseconds. The latest advance inemporal resolution comes from scanners with two detectorshat are positioned 90 degrees perpendicular to each othernd scan the same area simultaneously resulting in an effec-ive temporal resolution of 85 milliseconds.5 These advancesn CT technology have resulted in dramatic improvements inpatial resolution, such that isotropic voxels are obtained andan be reformatted in unlimited orientations, and temporalesolution, resulting in fewer motion artifacts and improvedmage quality (Fig 1).

Typical protocols for thoracoabdominal endovascularherapy consist of: (1) precontrast images, (2) arterial con-rast-enhanced images, and (3) venous phase imaging ac-uired a few minutes after the arterial phase images.6 Pre-ndovascular therapy examinations need only arterial andenous phase imaging. Approximately 100 to 130 cc iodin-ted contrast material is injected at 3 to 5 cc/second, with themage acquisition timed to follow the contrast bolus through-ut the arterial system. The CT scanner uses spiral data ac-uisition, but electrocardiographic (ECG) gating is not incor-orated.There are few absolute contraindications to CTA, although

atients with a history of previous severe reaction to iodin-

igure 1 Multiplanar reformations of a volumetric computed to-ography data set were performed in the coronal (A) and sagittal

B) orientations, demonstrating a complex dissection of the tho-

aiim

MMpnbqwtsinw

qdit“atpdii

ta

ptticqiTetihtoavf

AoaIpcbacotlipawpm

MBaactmtrpscco

Fcdd

110 S.D. Flamm

ted contrast material should be excluded from repeat exam-nation. Patients with more minor reactions may be examinedf appropriately premedicated and following appropriate

edical clearance.

RIRI and MR angiography of the thoracoabdominal aorta and

elvic arteries are usually performed on 1.5-Tesla MRI scan-ers equipped with cardiac software packages. Imaging maye performed with a body coil alone with adequate imageuality, however, use of a multi-element phased-array coilill improve signal-to-noise ratio and spatial resolution. Pa-

ients require placement of ECG leads, which should be po-itioned according to manufacturer’s recommendations. Thenjection of gadolinium contrast agent requires an intrave-ous line to be placed for thoracoabdominal aortic imagingith typical injection rates of 1.5 to 3.0 cc/second.7

Breath-hold images allow scanning to be performed asuickly as possible and with good resolution and vessel edgeefinition. In brief, standard protocols consist of: (1) scout

mages, (2) axial and coronal “black-blood” spin-echo imageshrough the chest and abdomen, (3) axial and coronalbright-blood” gradient-echo images through the chest andbdomen, and (4) non-ECG�gated, gadolinium-enhancedhree-dimensional MR angiograms (both arterial and venoushase) in optimized orientations through the thoracoab-ominal aorta and pelvic arteries (Fig 2). For additional phys-

ologic information, cine images can be obtained in custom-zed orientations to evaluate and quantify blood flow through

igure 2 Three-dimensional (3D) volumetric reconstruction of a 3Dontrast-enhanced magnetic resonance angiogram. The thoracoab-ominal is remarkable for a fusiform aneurysm at the thoracoab-

tominal junction.

he thoracic and abdominal aorta,8 as well as function of theortic valve and left ventricle.9-11

As with all MRI studies the standard contraindications ap-ly: patients with pacemakers, implantable cardiodefibrilla-ors, intracranial aneurysm clips, and all other standard con-raindications should be excluded from imaging.12 Mostmplanted devices, such as heart valves, stents, filters, areonsidered relative contraindications, as manufacturers fre-uently recommend a 6-week waiting period prior to imag-

ng for sufficient fibrosis and/or neoepithelialization to occur.hese manufacturer’s recommended waiting periods, how-ver, have not been borne-out by clinical observations, andhe vast majority of such devices can be imaged soon aftermplantation with some notable exceptions.12 In general, theemodynamic forces are greater on the implanted deviceshan the magnetic forces. Heating, within the magnetic field,f devices that are not within the vasculature has also beenconcern. However, for questions regarding specific de-

ices a convenient source for safety information can beound on the internet at: www.mrisafety.com.

dvantages and Limitationsf Cross-Sectional Imagingnd Endovascular Therapy

n general, CT’s primary limitations include the necessity forotentially nephrotoxic contrast administration, as well as aoncerning radiation dose. Alternatively, MRI is unencum-ered by radiation, but requires substantially longer for im-ge acquisition, and is limited by patient size and particularontraindications. MRI examinations typically take on therder of 30 to 45 minutes to complete, as opposed to lesshan 5 minutes for CT scans. Claustrophobia may also be aimitation in 5% to 7% of patients. Obesity continues toncrease in Western society and is particularly common inatients with cardiovascular disease; most MRI scanners canccommodate patients up to approximately 350 pounds,hile standard CT scanners may accommodate up to 450ounds, and newer scanners are being designed to accom-odate patients weighing up to 650 pounds.

orphologyoth CTA and MRI are sensitive and accurate techniques forssessing changes in aneurysm size and morphology, as wells in assessing branch vessel involvement. Both techniquesan assess stability of the aneurysm sac without administra-ion of contrast material, but neither provides reliable assess-ent of endoleaks without intravenous contrast administra-

ion. As a result, patients with renal insufficiency, who mayepresent a significant proportion of endovascular therapyatients, may be problematic for surveillance. In the past,uch patients could be imaged with MRI as the gadoliniumontrast agent was felt to be non-nephrotoxic. With the re-ent recognition of nephrogenic systemic fibrosis imagingptions have become more complex, and will be addressed.Type I and III endoleaks are the most critical to identify, as

he presence of systemic pressures poses the greatest risk for

aeerqair

FCaebdtnActmotvcp

nae

Fcsfaad

FfiiaavIflfl

Cross-sectional imaging studies 111

neurysm rupture. Both of these types of endoleaks are gen-rally readily recognized with both CT and MRI. Type IIndoleaks are more controversial, as their effect on the aneu-ysm sac may be more indolent, and intervention less fre-uently required. Nonetheless, MRI is approximately twices sensitive at detecting these leaks as CT is, though at presentt is uncertain whether this enhanced detection is clinicallyelevant.13

unctionardiac MRI is a highly capable tool for evaluation of cardiacnd valvular function, and is considered the gold standard forvaluation of left ventricular volumes, mass, and functionecause of its high reproducibility and accuracy.14-16 It is notependent on geometric assumptions to determine left ven-ricular volumes and mass, which limit the accuracy of tech-iques such as echocardiography and left ventriculography.ssessment of valvular morphology and function can nowompete with echocardiography: cine MRI now has the spa-ial and temporal resolution to discretely assess valvularorphology, including stenotic and regurgitant valve

rifices.9,17,18 In addition, phase-contrast MRI can quantifyhe jet of turbulent flow in valvular stenosis (Fig 3), or theolume of flow in regurgitant lesions10,19 (Fig 4). The formerapability employs techniques comparable to echocardiogra-hy, while the latter ability is unique to MRI.

igure 3 Cine-magnetic resonance image obtained in the obliqueoronal orientation through the aortic valve. There is severe aortictenosis resulting in a black jet of turbulent blood flow emanatingrom the stenotic aortic valve cusps and extending into the distalscending aorta. Phase-contrast flow quantification imaging, whenpplied in the same orientation, can quantify the peak velocity and

Cetermine the peak and mean gradient of aortic stenosis.

In the aorta cine-MRI easily defines rapid blood flow in aormal caliber aorta, or stagnant, slowly swirling flow in anneurysmal segment. In patients with aortic dissection differ-ntial blood flow may occur in the true and false lumens.

igure 4 On the left are two images from a phase-contrast flow quanti-cation imaging dataset (A) (upper: magnitude image; lower: phase

mage). The region of interest (black circles) outline the ascendingorta. From the phase data quantitative blood flow can be determinednd integrated with the cross-sectional area resulting in quantitativeolumetric flow throughout the cardiac cycle expressed as a graph (B).n this example, retrograde flow is seen below the baseline (backwardow) during diastole. The retrograde flow is divided by the progradeow to determine the aortic valve regurgitant volume and fraction (40%).

ine MRI can identify the transit of blood flow between lu-

ma

vhrtsmacctctpmopa

RAtttf

cnD

edrtrccggsCmiedftpdid1sts2

Mtm

R(aCsnWa8pdqdtai

toAbmqyt

Fwt

112 S.D. Flamm

ens via fenestrations and the mobility of the intimal flap inny portion of the aorta.20

In contrast, CT has had only a modest ability to evaluateentricular and valvular function, although this landscapeas altered with the introduction of MDCT. ECG-gating withetrospective techniques allows MDCT to extract both ven-ricular and valvular functional data. Four-dimensional dataets are acquired of the heart and valves and may be refor-atted in standard orthogonal orientations. A stack of short

xis reformatted images obtained through the left ventriclean be analyzed using standardized software on commer-ially available workstations, and ventricular volumes, func-ion, and mass determined. Preliminary data suggests a closeorrelation with cardiac MRI values; however, the limitedemporal resolution of CT allows only 6 to 12 cardiac phaseser beat depending on heart rate.21,22 These same data setsay be reformatted along the plane of the aortic valve (or

thers) and valve morphology investigated (Fig 5). In theresence of aortic stenosis or regurgitation, the valve orificerea can be directly planimetered.23

adiation Exposures noted previously, MRI has no ionizing radiation and,

herefore, is a particularly compelling noninvasive imagingechnique for assessment of the cardiovascular system, par-icularly in younger individuals who may require multipleollow-up imaging procedures.

In contrast, radiation exposure has only continued to in-rease with the increasing numbers of detectors in CT scan-ers and the increased capacity of the x-ray generation tubes.

igure 5 A volumetric computed tomography dataset of the heartas reconstructed using an oblique multiplanar reformat to analyze

he morphology of this normal aortic valve.

ose modulation, a means by which the amount of radiation l

xposure is altered depending on the area of interest andensity of the tissue, has been used to diminish the radiationisk. However, this is applied only to gated studies. Newerechnologies have been developed that will further decreaseadiation dose, and are expected to be employed with up-oming CT scanners. Currently, the relatively high doses fororonary CTA (8 to 20 millisieverts)4 has also brought toreater attention the radiation dose from CT examinations ineneral. The average amount of background radiation ab-orbed per person is approximately 3.5 millisieverts/year.oronary CTA examinations therefore impose approxi-ately 3 to 6 years worth of background radiation per exam-

nation. CT examinations of the thoracoabdominal aorta forndovascular therapy also impose a substantial radiation bur-en ranging from 18 to 45 millisieverts (unpublished datarom our own CT laboratory) depending on a variety of fac-ors, including the number of acquisitions, slice thickness,atient body habitus (larger patients require higher radiationoses for adequate tissue penetration to achieve diagnostic

mage quality), and radiation parameters. Thus, thoracoab-ominal aortic CTA examinations impose approximately 5 to3 years worth of background radiation per examination. Ithould be noted that thoracoabdominal aortic CT examina-ions are typically not ECG-gated, and doing so would sub-tantially further increase the radiation burden by a factor ofor greater.As a result of the heightened radiation doses imposed byDCT, greater care must be focused on the cumulative pa-

ient exposure from repeated CT scans, and attempts made toinimize that exposure.

enal ToxicityContrast-Induced Nephropathynd Nephrogenic Systemic Fibrosis)ontrast-induced nephropathy remains a risk in most CT

canning, but because of the increased speed of MDCT scan-ers, the contrast requirements have continued to decline.ith today’s CT scanners, it is unusual for the dose of iodin-

ted contrast material to exceed 150 cc, and frequently only0 to 120 cc is used. As a result, in the vast majority ofatients, no renal toxicity occurs. However, as patients un-ergoing endovascular therapy are often older, and fre-uently have multiple comorbidities, the potential for renalysfunction is increased. In patients at risk, standard pre-reatments (careful hydration, acetylcysteine, etc.) are avail-ble and have helped to minimize the likelihood of contrast-nduced nephropathy.

For patients with significant renal dysfunction, obviatinghe use of iodinated contrast material for CT scanning, gad-linium contrast enhancement has been a viable alternative.total injection of 60 cc gadolinium contrast agent followed

y a 40 cc saline flush—as a substitute for iodinated contrastaterial—provides CT images that have diagnostic image

uality, though are less aesthetically pleasing. Over the pastear, this alternative has been complicated by the recognitionhat administration of intravenous gadolinium has been

inked to nephrogenic systemic fibrosis in patients with acute

arfsTdpasA(a

WWCCoCdicsitf

qmdmapcptbe

VCtwtevdasc

eaacsvessrb

wsaddaraosgladmc

CR

Ftwdnv

Cross-sectional imaging studies 113

nd chronic renal insufficiency. Over 200 cases have beeneported in patients with moderate to end-stage renal diseaseollowing gadolinium-based contrast injections with progres-ion potentially fulminant and treatment options limited.he entity was first identified in 1997, with patients noted toevelop thickening and fibrosis of the skin, contractures,ain, and fibrosis in other organs. While a direct causativessociation has not yet been proven and it is unclear if it iseen with all forms of gadolinium, the US Food and Drugdministration recently issued a caution in this situation

available at: www.fda.gov/cder/drug/advisory/gadolinium_gents.htm).

hat Else Doe Want to Know?

an We Image the Coronaries Too?T is an enticing technology in that it has the potential tobviate the need for a conventional coronary angiogram.ombining this with the CT angiogram of the thoracoab-ominal aorta could speed the workup of patients in a prom-

sing and practical manner. However, such an examinationomplicates the performance of the CT and imposes a sub-tantial penalty and radiation exposure. Standard CT exam-nations for endovascular therapy do not require submillime-er slice thickness for data acquisition, while this is standardor coronary CTA4 (Fig 6). In addition, coronary CTA re-

igure 6 A volume-rendered three-dimensional cardiac computedomography (CT) dataset was constructed using an independentorkstation (Vital Images, Inc., Minnetonka, MN). Modern multi-etector CT scanners are now able to produce images of the coro-ary arteries with diagnostic detail that obviates the need for con-

sentional coronary angiography in some subpopulations.

uires ECG-gating with retrospective data acquisition, whichay increase radiation exposure by a factor of five. Finally,iagnostic coronary CTAs require a heart rate �65 beats perinute; for patients with higher heart rates, �-blockade (oral

nd/or intravenous) is required, which may extend patientreparation by 10 minutes to 2 hours. All in all, while theoncept for simultaneous coronary CTA acquisition is ap-ealing, the disadvantages in substantially increased radia-ion burden and lengthening patient preparation time areurdensome and not within an appropriate clinical risk-ben-fit boundary.

ascular ComplianceT and MRI technologies have focused on anatomic evalua-

ion of the vascular system. This approach has served patientsell as operative complications and mortality have continued

o decline, and this trend has only further extended withndovascular therapeutic options. However, further ad-ances will come not only from the development of newevices and technologies, but from optimal patient selectionnd the stratification of specific technologies to appropriateubpopulations. To that end, the determination of vascularompliance is a likely critical strategic component.

Aortic compliance is a reflection of the biophysical prop-rties of the aortic wall. Disruption of elastin fibers results indecrease in aortic compliance, an increase in aortic stiffness,nd an elevation in pulse pressure. Aortic compliance typi-ally has been measured in one of two ways: (1) pressure-train elastic modulus measurements, or (2) the pulse waveelocity method.24-26 The former technique requires knowl-dge of pressures, which are sometimes unknown, requiringurrogate estimates. The latter requires no estimations, as-uming flow profiles and distances can be measured accu-ately. Neither technique can be performed using CT, whileoth can be accomplished with MRI.Changes in compliance of the thoracoabdominal aortic

all are seen in diseases such as atherosclerosis, hyperten-ion, and normal aging.25 Taken to an extreme these alter-tions result in aneurysm formation. Early detection of aecrease in the aortic compliance could help to identify earlyisease in asymptomatic patients, particularly in those whore at greatest risk, such as in Marfan syndrome. In fact,ecent work suggests MRI detectable altered aortic compli-nce in Marfan patients who have more rapid aortic dilationr develop dissections over and above simple caliber mea-urements.27-29 Such early detection could lead to more tar-eted interventions, thereby avoiding procedures in patientsess likely to have complications. These measurements canlso be used to monitor the results of both surgical and en-ovascular interventions and may provide a useful measure-ent for differentiating clinically significant from insignifi-

ant endoleaks.30

onclusionapid, noninvasive imaging approaches have the potential to

implify procedures and make them safer. They may also

meitcatwtc

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

114 S.D. Flamm

ore effectively triage patients to appropriate therapy, orxclude patients from unnecessary invasive procedures. Thenformation gleaned from CT and MRI studies is critical forhe vascular surgeon who wants to identify appropriate vas-ular territories for intervention, plan a detailed approach,nd develop sophisticated surveillance strategies. The twoechniques, both alone or in combination, can provide aealth of information to enhance operative and interven-

ional procedures and improve patient outcomes, and willontinue to grow in importance with interventional options.

eferences1. Parodi JC, Palmaz JC, Barone HD: Transfemoral intraluminal graft im-

plantation for abdominal aortic aneurysms. Ann Vasc Surg 5:491-499,1991

2. van der Laan MJ, Bartels LW, Bakker CJ, et al: Suitability of 7 aorticstent-graft models for MRI-based surveillance. J Endovasc Ther 11:366-371, 2004

3. Budoff MJ, Cohen MC, Garcia MJ, et al: ACCF/AHA clinical compe-tence statement on cardiac imaging with computed tomography andmagnetic resonance: a report of the American College of CardiologyFoundation/American Heart Association/American College of Physi-cians Task Force on Clinical Competence and Training. J Am CollCardiol 46:383-402, 2005

4. Budoff MJ, Achenbach S, Blumenthal RS, et al: Assessment of coronaryartery disease by cardiac computed tomography: a scientific statementfrom the American Heart Association Committee on CardiovascularImaging and Intervention, Council on Cardiovascular Radiology andIntervention, and Committee on Cardiac Imaging, Council on ClinicalCardiology. Circulation 114:1761-1791, 2006

5. Achenbach S, Ropers D, Kuettner A, et al: Contrast-enhanced coronaryartery visualization by dual-source computed tomography—initial ex-perience. Eur J Radiol 57:331-335, 2006

6. Hellinger JC: Endovascular repair of thoracic and abdominal aorticaneurysms: pre- and postprocedural imaging. Tech Vasc Interv Radiol8:2-15, 2005

7. Rajagopalan S, Prince M: Magnetic resonance angiographic techniquesfor the diagnosis of arterial disease. Cardiol Clin 20:501-12, v, 2002

8. Kvitting JP, Ebbers T, Wigstrom L, et al: Flow patterns in the aorticroot and the aorta studied with time-resolved, 3-dimensional,phase-contrast magnetic resonance imaging: implications for aorticvalve-sparing surgery. J Thorac Cardiovasc Surg 127:1602-1607,2004

9. Kupfahl C, Honold M, Meinhardt G, et al: Evaluation of aortic stenosisby cardiovascular magnetic resonance imaging: comparison with estab-lished routine clinical techniques. Heart 90:893-901, 2004

0. Krombach GA, Kuhl H, Bucker A, et al: Cine MR imaging of heart valvedysfunction with segmented true fast imaging with steady state freeprecession. J Magn Reson Imaging 19:59-67, 2004

1. Alfakih K, Plein S, Thiele H, et al: Normal human left and right ven-tricular dimensions for MRI as assessed by turbo gradient echo andsteady-state free precession imaging sequences. J Magn Reson Imaging17:323-329, 2003

2. Sawyer-Glover AM, Shellock FG: Pre-MRI procedure screening: recom-mendations and safety considerations for biomedical implants and de-

vices. J Magn Reson Imaging 12:510, 2000

3. van der Laan MJ, Bartels LW, Viergever MA, et al: Computed Tomog-raphy versus Magnetic Resonance Imaging of Endoleaks after EVAR.Eur J Vasc Endovasc Surg 32:361-365, 2006

4. Grothues F, Smith GC, Moon JC, et al: Comparison of interstudy re-producibility of cardiovascular magnetic resonance with two-dimen-sional echocardiography in normal subjects and in patients with heartfailure or left ventricular hypertrophy. Am J Cardiol 90:29-34, 2002

5. Myerson SG, Montgomery HE, Whittingham M, et al: Left ventricularhypertrophy with exercise and ACE gene insertion/deletion polymor-phism: a randomized controlled trial with losartan. Circulation 103:226-230, 2001

6. Bellenger NG, Davies LC, Francis JM, et al: Reduction in sample size forstudies of remodeling in heart failure by the use of cardiovascularmagnetic resonance. J Cardiovasc Magn Reson 2:271-278, 2000

7. Schlosser T, Malyar N, Jochims M, et al: Quantification of aortic valvestenosis in MRI—comparison of steady-state free precession and fastlow-angle shot sequences. Eur Radiol 17:1284-1290, 2007

8. Caruthers SD, Lin SJ, Brown P, et al: Practical value of cardiac magneticresonance imaging for clinical quantification of aortic valve stenosis:comparison with echocardiography. Circulation 108:2236-2243, 2003

9. Kon MW, Myerson SG, Moat NE, et al: Quantification of regurgitantfraction in mitral regurgitation by cardiovascular magnetic resonance:comparison of techniques. J Heart Valve Dis 13:600-607, 2004

0. Strotzer M, Aebert H, Lenhart M, et al: Morphology and hemodynamicsin dissection of the descending aorta. Assessment with MR imaging.Acta Radiol 41:594-600, 2000

1. Schlosser T, Mohrs OK, Magedanz A, et al: Assessment of left ventric-ular function and mass in patients undergoing computed tomography(CT) coronary angiography using 64-detector-row CT: comparison tomagnetic resonance imaging. Acta Radiol 48:30-35, 2007

2. van der Vleuten PA, V, Willems TP, Gotte MJ, et al: Quantification ofglobal left ventricular function: comparison of multidetector computedtomography and magnetic resonance imaging. A meta-analysis andreview of the current literature. Acta Radiol 47:1049-1057, 2006

3. Vogel-Claussen J, Pannu H, Spevak PJ, et al: Cardiac valve assessmentwith MR imaging and 64-section multi-detector row CT. Radiographics26:1769-1784, 2006

4. Murai S, Hamada S, Ueguchi T, et al: Aortic compliance in patients withaortic regurgitation: evaluation with magnetic resonance imaging. Ra-diat Med 23:236-241, 2005

5. Metafratzi ZM, Efremidis SC, Skopelitou AS, et al: The clinical signifi-cance of aortic compliance and its assessment with magnetic resonanceimaging. J Cardiovasc Magn Reson 4:481-491, 2002

6. Lalande A, Khau van KP, Salve N, et al: Automatic determination ofaortic compliance with cine-magnetic resonance imaging: an applica-tion of fuzzy logic theory. Invest Radiol 37:685-691, 2002

7. Lehmann ED, Hopkins KD, Gosling RG: Aortic distensibility measuredby magnetic resonance imaging in patients with Marfan’s syndrome.Heart 75:214, 1996

8. Groenink M, de Roos A, Mulder BJ, et al: Changes in aortic distensibil-ity and pulse wave velocity assessed with magnetic resonance imagingfollowing beta-blocker therapy in the Marfan syndrome. Am J Cardiol82:203-208, 1998

9. Nollen GJ, Groenink M, Tijssen JG, et al: Aortic stiffness and diameterpredict progressive aortic dilatation in patients with Marfan syndrome.Eur Heart J 25:1146-1152, 2004

0. van Herwaarden JA, Muhs BE, Vincken KL, et al: Aortic compliancefollowing EVAR and the influence of different endografts: determina-

tion using dynamic MRA. J Endovasc Ther 13:406-414, 2006