Coronary CTA - SEICATseicat.org/repo/static/public/documentos/top10-cardiaca/... · 2012-01-24 · Key Words: coronary CTA, multidetector-row CT (J Thorac Imaging 2007;22:22–34)

Coronary CTAImage Acquisition and Interpretation

Josef Matthias Kerl, MD,*w Lars K. Hofmann, MD,z Christian Thilo, MD,*yThomas J. Vogl, MD,w Philip Costello, MD,* and U. Joseph Schoepf, MD*y

Abstract: Computed tomography (CT) of the heart, because of

ongoing technical refinement and intense scientific and clinical

evaluation, has left the research realm and has matured into a

clinical application that is about to fulfill its promise to replace

invasive cardiac catheterization in some patient populations. By

nature of its target, the continuously moving heart, CT coronary

angiography is technically more challenging than other CT

applications. Also, rapid technical development requires con-

stant adaptation of acquisition protocols. Those challenges,

however, are in no way insurmountable for users with knowl-

edge of general CT technique. The intent of this communication

is to provide for those interested in and involved with coronary

CT angiography a step-by-step manual, introducing our

approach to performing coronary CT angiography. Included

are considerations regarding appropriate patient selection,

patient medication, radiation protection, contrast enhancement,

acquisition and reconstruction parameters, image display and

analysis techniques and also the radiology report. Our

recommendations are based on our experience which spans the

evolution of multidetector-row CT for cardiac applications from

its beginnings to the most current iterations of advanced

acquisition modalities, which we believe herald the entrance of

this test into routine clinical practice.

Key Words: coronary CTA, multidetector-row CT

(J Thorac Imaging 2007;22:22–34)

W ith the ongoing evolution of ever faster and moresophisticated multidetector-row computed tomo-

graphy (MDCT) technology, CT of the heart has evolvedinto an examination that is applied to a broad variety ofclinical indications.1 With the advent of the latestiterations of MDCT technology, both temporal and

spatial resolution of coronary CT angiography (CTA)have improved to a point where the threshold to routinenoninvasive assessment of the coronary arteries foratherosclerotic disease may have been crossed.

Imaging of the heart has always been technicallychallenging, because of the heart’s continuous motion.The development of electrocardiography (ECG)-synchro-nized MDCT scanning and reconstruction techniques2–4

has provided fast volume coverage and high spatial andtemporal resolution as a prerequisite for successfulcardiac imaging.5 The exceedingly powerful technologywhich enables performance of coronary CTA, however,transcends routine CT applications and needs to be usedin a manner to provide optimized results with the leastdegree of invasiveness for the patient. Our experienceperforming contrast enhanced coronary CTA has evolvedover the years since the introduction of MDCT and hasencompassed every single evolutionary step of thistechnology up to the most current generation of dual-source CT (DSCT). The purpose of this communication isto share our experience with those interested in andinvolved with coronary CTA to facilitate the beneficialapplication of this procedure.

PATIENT PREPARATION

IV AccessIntravenous access is preferably established in a

cubital vein. Especially, if a left internal mammaryarterial bypass graft is to be assessed, an access site inthe right arm should be chosen to prevent streak artifactsarising from dense contrast material in the left subclavianvein from interfering with the evaluation of left internalmammary arterial origin. Because of the fairly fastinjection rates of 4 to 5mL/s, an 18-G catheter or largershould be used, whenever possible.

ECG Lead AttachmentOrdinarily, a 3-lead ECG is used for cardiac CT. A

stable reading of the patient’s ECG with clear identifica-tion of the QRS-Complex is a prerequisite for successfulretrospective ECG-gating. To establish good electricalcontact and prevent lead detachment with consecutivesignal loss during scan acquisition, use of additionalconductive gel and shaving of overly hairy attachmentsites are recommended.Copyright r 2007 by Lippincott Williams & Wilkins

From the *Department of Radiology; yDivision of Cardiology,Department of Medicine, Medical University of South Carolina,SC; wInstitute for Diagnostic and Interventional Radiology, JohannWolfgang Goethe-University, Frankfurt; and zSiemens MedicalSolutions, Division CT, Forchheim, Germany.

U. Joseph Schoepf is a medical consultant to Berlex, Bracco, GeneralElectric, Siemens, and TeraRecon and receives research support fromBerlex, Bracco, General Electric, Medrad, and Siemens.

Reprints: U. Joseph Schoepf, MD, Department of Radiology, MedicalUniversity of South Carolina, 169 Ashley Avenue, Charleston, SC29425 (e-mail: [email protected]).

SYMPOSIA

22 J Thorac Imaging � Volume 22, Number 1, February 2007

Rate Control

Rate Control for Improving Image QualityFor a variety of reasons, slow heart rates are highly

desirable for cardiac CT using 4-slice to 64-slice scanners:Slow heart rates relatively prolong the cardiac phaseswith little cardiac motion, that is, end-diastolic relaxationand end-systolic contraction, so that a reconstructioninterval of a defined duration can be safely placed withinthese phases without inclusion of preceding or consecu-tive portions of the heart cycle that contain motion.

In our practice, we aim at a target heart rate ofbetween 50 and 65 beats per minute (bpm) for our 64-sliceCT scanners. With DSCT, the newest scanner generation,heart rate control might not be necessary anymore. In ourexperience, DSCT allows scanning patients with high (ie,120 to 140 bpm) and irregular heart rates with diagnosticresults (Fig. 1). These scanners employ 2 x-ray sourcesand 2 corresponding detectors offset by 90 degrees, thusproviding a temporal resolution equivalent to a quarter ofthe gantry rotation time (ie, 83ms), independent of thepatient’s heart rate.6

Rate Control for Single-segment Reconstruction inSingle-source CT Scanners

As a direct consequence of this high temporalresolution, image reconstruction can be ordinarily per-formed using single-segment reconstruction, that is, basedon the projections that are acquired during a single

heartbeat. Single-segment reconstruction has theoreticaladvantages over multisegment reconstruction algorithmsthat are implemented on all available cardiac CTscanners. At multisegment reconstruction, the projectionsthat are needed to form a single section are sampled over2 to 4 consecutive heart beats.5 Although this approachimproves temporal resolution, which is beneficial forfaster heart rates, it requires that the heart follows theexact same motion pattern for each of the 2 to 4 heartbeats during which projection sampling for reconstruc-tion of a single section occurs. This, however, isunreasonable to expect, given the variability of cardiacmotion patterns even under physiologic conditions sothat spatial inconsistencies within the data inevitablyoccur with the use of multisegment reconstruction. In ourpractice, we strive to avoid the use of multisegmentreconstruction algorithms by appropriately reducing theheart rate with 64-slice CT and use it exclusively inpatients with heart rates >80 bpm, above which, in ourexperience, the benefits of improved temporal resolutionoutweigh the risk of spatial inconsistency.

Rate Control for Radiation ProtectionUp to the generation of 64-slice CT, rate control has

been directly related to patient radiation exposure atcardiac CT. Significant dose savings can be realized withECG-gated dose-modulation (‘‘ECG-pulsing’’).5,7 Withthis approach, the nominal tube output is only applied

FIGURE 1. Contrast enhanced, retro-spectively ECG-gated dual-source coro-nary CTA in a 62-year-old man withpersistent atrial fibrillation and tachyar-rhythmia after pulmonary vein ablationtherapy. The heart rate during the CTscan varied between 67 and 150 bpm(A). Despite this high and irregular heartrate, the stable temporal resolution of83 msec enabled by this technologyresulted in a data set of sufficientdiagnostic quality. Volume rendereddisplays both, during diastole (B) andsystole (C) are of comparable imagequality and do not show motion ormisregistration artifacts.

J Thorac Imaging � Volume 22, Number 1, February 2007 Coronary CTA

r 2007 Lippincott Williams & Wilkins 23

during the particular phase of the cardiac cycle at whichimages are most likely to be reconstructed. Over the restof the cardiac cycle, the tube output is reduced. However,the use of ECG-gated dose-modulation is limited topatients with slow and steady heart rates, in whom theoptimal time-point for image reconstruction predictablyoccurs during diastole (Fig. 2). At faster heart rates,ECG-gated dose-modulation increasingly looses its effi-cacy, as the period of reduced tube output becomesprogressively shorter relative to the cardiac cycle. Moreimportantly, with faster heart rates, the optimal time-point for image reconstruction becomes more difficult topredict and, in the case of 64-slice CT frequently occursduring the end-systolic phase of total myocardialcontraction (Fig. 3). Because with ECG-gated dose-modulation the radiation level is typically reduced duringsystole, diagnostic quality will inevitably be compromisedduring image reconstruction so that use of ECG-gateddose-modulation is not recommendable for faster heartrates. This way, rate control has become a crucial factorin the endeavor of keeping radiation exposure at cardiacCT within reasonable limits. In our practice, with 64-sliceCT, we used ECG-gated dose-modulation in patients withsteady heart rates <65 bpm. We feel that in patients withfaster and more irregular heart rates ECG-gated dose-modulation limits, our options with regard to theselection of the optimal reconstruction time-point toomuch as to recommend its general use.

These rules do not apply for the use of DSCT.Because of a temporal resolution window of only 83ms,ECG dose-modulation can be applied much moreeffectively. Full energy windows covering only a 10%window of the cardiac cycle are feasible. Additionally,

DSCT allows the adaptation of the pitch to the patient’sheart rate, because multisegment reconstruction forhigher heart rates will not be required. The table feedcan be efficiently adapted to the patient’s heart rate andsignificantly increased at higher heart rates.6 As shown byMcCollough et al,8 this may reduce radiation dose incardiac CT by up to 50% compared with single-sourceCT.

Rate Control: Practical AspectsIn our practice, we routinely used intravenous

b-blocker (Metoprolol Tartrate, Lopressor, Novartis,East Hanover, NJ) for controlling our patients’ heartrate with very satisfactory results and without complica-tions to date. Contraindications for the use of b-blockersinclude chronic obstructive pulmonary disease, asthma,sensitive to b-agonists, second-degree or third-degreeheart block; hypotension (<100mm Hg systolic). In theabsence of contraindications, we inject an initial bolus of5mg of Metoprolol while the patient is already on thescanner table and preparations for scanning ensue. If theventricular response is unsatisfactory, that is, the averageheart rate remains >70 bpm, we inject up to 2 additionaldoses (15mg total maximum) of Metoprolol. Afteradministration of 3 doses, we commence scanning,regardless of the heart rate that is eventually achievedafter b-blocker administration. Oral administration ofb-blockers is an alternative means of rate control, whichcompared with our intravenous protocol, exerts higherdemands on operational logistics. If oral administration ispreferred, ideally, the regimen should be commenced thenight before the scan with an initial dose of 50 to 100mgof Metoprolol. Thirty to sixty minutes before the scan,

FIGURE 2. Contrast enhanced, retro-spectively ECG-gated 64-slice coronaryCTA in a 60-year-old man, referred forpatency evaluation of a left internalmammary artery bypass graft to the leftanterior descending artery (LAD) and 3saphenous vein grafts to the 3 majorcoronary territories. Display in volumerendering technique (left upper panel)seen from an anterior perspective. Aslow and steady heart rate of E60 bpmenables successful use of ECG-pulsing forthe radiation protection. Full nominaltube current (indicated in red in thepatient’s ECG, lower panel) is onlyapplied during diastole, the cardiacphase which is subsequently used forimage reconstruction at 60% RR, whichresults in full image quality with highsignal-to-noise ratio (right upper panel).During other cardiac phases, which arenot used for image reconstruction, thetube current is lowered to 20% ofthe nominal output (indicated in bluein the patient’s ECG, lower panel).

Kerl et al J Thorac Imaging � Volume 22, Number 1, February 2007

24 r 2007 Lippincott Williams & Wilkins

another dose is given, followed by a third dose in theabsence of adequate ventricular response. There is lessexperience with alternative rate-controlling medications.However, if there are contraindications for use ofb-agonists (see above), an attempt with calcium channelblockers may be worthwhile. Calcium channel blockerscan be administered either intravenously [0.25mg/kg

bodyweight (up to 25mg total) Diltiazem, CardizemMonovial, Hoechst Marion Roussel, Kansas City, MO]or as an oral regimen with 30mg of regular releaseDiltiazem (Cardizem, Biovail, Toronto, Canada). In ourpractice, all patients who receive rate-controlling medica-tions or nitroglycerin (NTG) (see below) are monitoredfor 30 minutes after administration of the last dose. Blood

FIGURE 3. Contrast enhanced, retro-spectively ECG-gated 64-slice coronaryCTA in a 70-year-old woman, referredfor equivocal perfusion abnormalities atcardiac single photon emission CT.Because of the fast and irregular heartrate of E120 bpm (lower panel), theoptimal reconstruction interval cannotreliably be predicted, and ECG-pulsingtechnique was not used. Different fromthe patient in Figure 2, in whom theoptimal reconstruction interval was pre-dictably found during diastole, imagereconstruction at 60% RR in this patient(A) results in considerable cardiac mo-tion, which blurs the RCA on transverseimages (right upper panel in A) andentirely prevents visualization of thevessel on volume rendered images (leftupper panel in A), while the LAD iscontorted. However, as full tube currentis maintained throughout the scan (in-dicated in red in the patient’s ECG,lower panel), flexibility is maintained toreconstruct data during any phase of thecardiac cycle. Subsequent reconstruc-tion of the same data set during fullsystolic contraction (B) at 25% RR resultsin nearly motion-free transverse images(right upper panel in B), which enablesclear display of the LAD and RCA atvolume rendering (left upper panel in B).



pressure measurements are performed before administra-tion of the drug and before discharge. Patients whoreceive more than 1 dose of b-blockers are instructed notto drive or operate machinery for 3 hours after theadministration of the drug.

NTGNo systematic studies are available to support the

use of NTG in the context of cardiac CT. In patientsreferred for coronary CTA for suspected coronary arterydisease, we administer a 0.4mg NTG tablet (NitroQuick,Ethex, St Louis, MO) sublingually 2 minutes before thescan with the theoretical rationale of widening thecoronaries for better visualization and suppressing coro-nary artery spasms that may mimic stenosis at coronaryCTA especially in younger individuals. Sublingual spraycan be used alternatively. Contraindications for NTGcomprise hypotension, early myocardial infarction, severeanemia, increased intracranial pressure, and knownhypersensitivity to NTG. We also do not give NTG topatients who recently took nitrate-based medication(Viagra, Cialis, Levitra) for erectile dysfunction.

SCANNING PARAMETERS

Gantry Rotation Speed and Collimated SectionWidth

At coronary CTA, the target anatomy is minute,tortuous, and moving rapidly. Accordingly, as a universalrule that applies to MDCT scanners of all manufacturers,for optimally performing coronary CTA one can neverslice too thin or spin too fast. Thus, at coronary CTA,regardless of the scanner type used, the thinnest possiblecollimation and the fastest gantry rotation time should bechosen. An exception is the rare situation where thepatient’s heart rate is so slow (typically <50 bpm), thatuse of a somewhat slower gantry rotation with slowerpitch is required to avoid gaps in the acquired data set.9

Another scenario where choice of a somewhat slowergantry rotation speed may be considered is the severelyobese patient, where slower gantry rotation may behelpful to enhance the photon flux and improve thesignal-to-noise ratio at coronary CTA. DSCT, with its 2x-ray tubes, is considerably better suited than single-source CT for suppressing image noise in obese patients.A reduction in gantry rotation speed with this lattertechnology is generally not necessary. Rather, a heart raterange can be selected that is lower than the patient’sactual heart rate which results in an artificially lower tablespeed and accumulation of dose to suppress image noise.

Tube Current and VoltageIssuing general recommendations for the selection

of tube current settings is challenging, as the appropriatetube current level depends on several scanner specificvariables, such as the collimated section width and thegantry rotation speed. As with all CT applications, theALARA (As Low As Reasonably Achievable) principleapplies, that calls for patient specific adjustment of the

scanner settings to the patient’s body habitus so that thelowest possible tube current setting is chosen that stillresults in a diagnostic study. When scanning normal sizedadults for suspected coronary artery disease with 0.6mmcollimation and 330msec gantry rotation, we ordinarilyadjust the tube current to 750 to 800mAseff w with single-source CT, always employing all means available forreducing radiation exposure to the patient (ie, ECG-pulsing, see above). With DSCT, scanning parameters of120 kV per tube with a current of 560mA and the use ofanatomic tube current modulation is ordinarily chosen.The gantry rotation time is usually set to 0.330 secondsand the pitch ranges from 0.2 to 0.43 depending on heartrate.

As the x-ray absorption of iodine is inverselyproportional to the tube potential, very high tube voltagesettings (ie, 140 kV) are generally not recommendable forcontrast enhanced CTA. When scanning normal sized orlarger adults for suspected coronary artery disease, weordinarily employ a tube potential of 120 kV. Substantialdose savings can be realized by lowering the tube voltageto 100 or 80 kV, while the level of vascular attenuationincreases. Thus, in slim adults the tube voltage can besafely lowered to 100 kV with very satisfactory results inour experience. Similarly, depending on the size of thepatient, we routinely use 100 or 80 kV, when performingcardiac CT in adolescent or pediatric patients forsuspected coronary artery anomalies or other congenitalcardiovascular disorders.10

CONTRAST ENHANCEMENT

Level of EnhancementHigh and consistent vascular enhancement within

the vessel lumen is a prerequisite for successful coronaryCTA. Adequate enhancement is needed for visualizationof the vessel wall, of small side branches of the coronarytree and, especially in obese patients, to compensate forincreased levels of image noise that limit contrastresolution at thin-section coronary CTA. In addition,high and homogenous enhancement serves as the basis forthreshold-dependent 3-dimensional visualization techni-ques. To achieve the desired high vascular enhancement,we use high concentration 350 to 370mgI/mL nonioniccontrast media with a fast injection rate of 5 to 6mL/s.It has been argued that exceedingly high intralumenalcontrast may interfere with the detection of calcifiedatherosclerotic plaque. In our experience, this pitfall canbe easily avoided by appropriately adjusting windowcenter and width (CE100 HU; WE700 HU) according tothe level of enhancement hat was achieved in theindividual patient.

Saline Chasing TechniqueSignificant scientific effort is currently directed at

understanding and optimizing contrast dynamics atCTA.11,12 Insights gained from ongoing investigationsare continuously implemented in commercially availableautomated injectors. One such implementation that has



become imperative for coronary CTA is the use of a salinechaser enabled by dual-syringe injection systems. Use of asaline chaser aims at better utilization of the injectedcontrast media by prolonging the plateau-phase ofcontrast. More importantly, use of a saline chaser reducesthe occurrence of streak artifact from dense contrastmedia in the superior vena cava and the right heartchambers. At coronary CTA, such streak artifacts have

the potential to significantly limit the evaluation of theright coronary artery (RCA), and may simulate stenosisespecially at 2-dimensional and 3-dimensional imagepostprocessing (Fig. 4). However, the occurrence ofstreak artifacts may be reduced or entirely avoided, if atthe time of image acquisition the contrast material isflushed from the right heart by use of saline chasertechnique. On the other hand, a complete void of contrast

FIGURE 4. Contrast enhanced, retro-spectively ECG-gated CT coronaryangiography without saline chasingtechnique. Display as transverse section(A) and volume rendering (B), seenfrom a left anterior oblique perspective.A streak artifact emanating from densecontrast material in the right heart(arrow in A) overlies the RCA and causesartifactual stenosis (arrow in B) of theproximal RCA.



in the right heart is undesirable, as right heart analysis isprecluded. We aim at improving right heart visualizationby using the 3 phases of injection where the initial iodinebolus is followed by a saline/contrast mixture enabled bysimultaneous injection from both syringes (Dual FlowTechnology, Medrad, Pittsburgh, PA) and a final salinechaser (see below). This strategy provides sufficientenhancement for assessment of the right heart (Fig. 5)for detection of thrombo-emboli, tumors, etc, whilestreak artifacts from dense contrast material can generallybe avoided.

Contrast ProtocolIn our practice, the individual delay time is

determined by injection of a 20mL contrast media testbolus at 5mL/s (64-slice CT) or 6mL/s (DSCT), followedby 50mL of saline using a dual-syringe injector (StellantD, Medrad). Repeated scanning at the same z-position atthe level of the aortic root is performed to monitor thearrival and passage of the test bolus. The peak time of testbolus enhancement is used as the delay time. The contrastvolume for the actual coronary CTA scan is individuallycomputed according to the following formula: Volume(mL)=scan time (s)� 5 (64-slice CT) or � 6 (DSCT).The injector is preprogrammed to deliver 50mL of a 30%contrast media/70% saline mixture during the secondphase of injection, followed by a final 30 to 50mL salinechaser, all injected at 5mL/s (64-slice CT) or 6mL/s(DSCT), respectively.

IMAGE RECONSTRUCTION

Single-segment Versus MultisegmentReconstruction Algorithms

For considerations regarding use of single-segmentversus multisegment reconstruction, please see abovediscussion on rate control (‘‘Rate Control for Single-Segment Reconstruction with Single Source CTScanners’’ section). In general, we avoid the use ofmultisegment reconstruction in patients with heart ratesof less than 80 bpm with 64-slice CT and entirely withDSCT.

Choosing the Optimal Reconstruction IntervalFor the assessment of cardiac morphology, a phase

with minimal cardiac motion is preferably chosen forplacement of the image reconstruction interval. To definethe starting point of the reconstruction interval within thecardiac cycle, absolute and relative approaches areavailable on most cardiac CT scanner types. With anabsolute approach, each image reconstruction interval isplaced in the cardiac cycle with a predefined temporaldistance (eg, 400ms) before or after an R-peak in theECG. With a relative approach, the starting point of theimage reconstruction interval is defined as a certainpercentage (eg, 60%) of the duration of the cardiac cycle.We use the relative, percentage-based approach for64-slice CT and an absolute approach for DSCT. Ifavailable, a preview function is preferably used fordetermining the optimal reconstruction phase with theleast cardiac motion. Typically, the preview series consists

FIGURE 5. Contrast enhanced, retrospectively ECG-gated CT coronary angiography in 3 different patients. Use of a monophasicinjection (left) of iodine only, using a single syringe injector, results in dense streak artifacts in the right heart. Use of a biphasicinjection (middle) with a dual-syringe injector and saline chasing technique flushes residual contrast from the right heart andavoids artifacts; however, the right cardiac chambers can no longer be assessed. A triphasic injection, also using a dual-syringeinjector, with simultaneous flow (right) from 2 syringes administers a mixture of contrast and saline during the second injectionphase. This provides sufficient enhancement for assessment of the right heart while streak artifacts from dense contrast materialcan generally be avoided.



of 20 images (Fig. 6), reconstructed at 20 different RRpositions in 5% increments (0% to 95% RR) at the samez-position at the mid-level of the heart. The phase thatshows the least motion artifacts in both, the left and RCAsystem is chosen for the image reconstruction. In caseswhere the right and left coronary artery show divergingmotion patterns, more than one reconstruction isperformed for optimized visualization of both arterialsystems. If a preview function is not available, a firstimage reconstruction of the data set can be performed at60% RR (Fig. 2), which has been shown to result indiagnostic image quality in most patients,13 especially atslower, regular heart rates. With the improved temporalresolution of newer scanners, late systole with totalcardiac contraction (ie, 30% to 40% RR) has emergedas a second suitable time-point for image reconstruc-tion (Fig. 3), where cardiac motion is at a minimum. Inour experience, image reconstruction during late systoleyields diagnostic results in most patients with a fasterheart rate and is especially well suited for visualization ofthe RCA.

Reconstruction Parameters

Field of ViewTo maximize spatial resolution, the smallest possi-

ble field of view should be chosen that encompasses theentire anatomy of the heart, for performing imagereconstruction at CT coronary angiography. In addition,for each coronary CTA study, we perform a full field ofview reconstruction of the entire chest along the acquiredz-volume with 3-mm section thickness and a lungalgorithm, to assess for incidental lung pathology. Forspecialized applications, such as ‘‘triple rule-out’’ scan-ning, we perform the 2 reconstructions described aboveand a third reconstruction with 1-mm section thickness, avascular algorithm and a field of view that encompassesthe entire chest, to evaluate for vascular pathology of thepulmonary circulation and the thoracic aorta.

Section ThicknessGenerally, to avoid artifacts, thin-section MDCT

data should be reconstructed with a section width that is

FIGURE 6. Contrast enhanced, retrospectively ECG-gated DSCT coronary angiography. Same patient as in Figure 9. Use of apreview function identifies the most suitable cardiac phase for the image reconstruction. Shown are 6 representative transversesections at the same z-position at the level of the aortic valve, out of a series of 20 reconstructions at different RR positions in 5%increments (0% to 95% RR), covering the entire cardiac cycle. Image reconstructions during full cardiac contraction at end systole(45% RR) and during full relaxation during diastole (60% RR) are suitable to sufficiently suppress cardiac motion and enable clearassessment of the stenosed RCA origin, whereas this area is blurred during other phases of the cardiac cycle. Image reconstructionduring early (15%) and late (90%) phases of the cardiac cycle also show higher levels of image noise, owing to our routine use ofECG-pulsing techniques during these phases with DSCT.



slightly wider than the collimated section width.14 Forexample, if the scan was acquired with 0.6-mm collimatedsection width, the next higher available reconstructionthickness (eg, 0.75mm) should be chosen for the imagereconstruction. Forty to sixty percent increment isordinarily used for the image reconstruction at coronaryCTA, which in our experience results in a somewhatcrisper and sharper delineation of the coronary artery treebut does not necessarily improve diagnostic accuracycompared to contiguous image reconstruction withoutoverlap.

Reconstruction Algorithm (‘‘Kernel’’)Most CT scanners used for coronary CTA offer a

dedicated reconstruction filter (kernel) for the imagereconstruction of cardiac CT studies. Typically, thesekernels maintain a degree of edge enhancement, toprovide the spatial resolution necessary to visualize smallvascular detail. Ideally, the kernels are also optimized tosuppress image noise as much as possible, to improve the

visual impression and maintain contrast resolution forevaluation of the myocardium and the vessel wall. For theevaluation of coronary artery stents, it is recommendableto use a kernel with even stronger edge enhancingcharacteristics and greater spatial resolution. Thisapproach suppresses beam hardening artifacts to someextent and provides better delineation of metallic stentstruts (Fig. 7) than the algorithms which are routinelyused at coronary CTA. This approach may also some-what increase the diagnostic yield in the presence of heavycalcifications, which pose a similar problem as densestent struts for the evaluating luminal integrity. Althoughdedicated reconstruction algorithms improve visualiza-tion of coronary artery stents, our ability to assessfor stent patency with CT is extremely variable anddepends on the overall quality of the data set and the sizeand material of the stent. Because of this variability,reliable assessment of stents cannot be expected on aroutine basis and we discourage use of CT for dedicatedstent follow-up.

FIGURE 7. Contrast enhanced, retro-spectively ECG-gated 64-slice coronaryCTA in a 27-year-old woman statuspostmyocardial infarction and subse-quent LAD stent placement referred fornoninvasive disease surveillance becauseof desired pregnancy. Three-dimen-sional volume rendered display of theentire cardiac volume (A) and of theautomatically extracted coronary arterytree (B) show the location of the stent(arrow) in the proximal LAD. Reconstruc-tion of the same data set with a routinecoronary CTA algorithm (C) and a moreedge enhancing algorithm (D) dedi-cated to the assessment of coronaryartery stents and segments with densecalcification. Display of the LAD ascurved MPR (C, D) show the stent(arrow) in the proximal portion of thevessel. The dedicated stent reconstruc-tion algorithm (D) enables somewhatmore detailed assessment of the metallicstent struts and the patent lumen,particularly in the more distal portionof the stent.



IMAGE DISPLAY FOR LESION DETECTION ANDGRADING

Diagnostic StrategiesFor cardiac applications, the role of advanced,

dedicated image display, and analysis tools is consider-ably greater than for general CT applications. Review ofthe individual transverse source images, however, cannotbe abandoned and must be a part of the diagnosticprocess in each case. The transverse source images arerichest in information regarding incidental mediastinalfindings, artifacts (Fig. 4), and the overall atheroscleroticplaque burden within the coronary artery tree. Everypostprocessing step necessarily and by design reduces theavailable information for the sake of more intuitive imagevisualization.

Depending on the particular indication for perform-ing cardiac CT, we employ slightly different strategies forour diagnostic approach. When assessing bypass grafts orthe left atrium and pulmonary veins in the context ofablation therapy for cardiac arrhythmia, a 3-dimensionalvolume rendered model is used for quick initial orienta-tion, for example regarding the type and course of bypassgrafts or the general configuration of the pulmonaryvenous return. This is followed by the review of transversesource images for the detection and grading of graftlesions or pulmonary vein stenosis and also additional oralternative findings in the chest. For suspected coronaryartery stenosis, the transverse source images are initially

reviewed, to obtain general information on the presence,location, and composition (calcified vs. noncalcified)of atherosclerotic lesions15 and also consequences ofischemic disease, such as myocardial perfusion deficits orscarring (Fig. 8). Once lesions are detected, stenosisseverity is evaluated by using simple visualization toolsthat enable a more comprehensive, condensed display ofthe data set. Multiplanar reformats (MPRs, see below)are easy to use basic tools and are available on most CTscanners. For improved detection and grading ofcoronary artery lesions, we use dedicated visualizationand analysis tools (see below), whenever interpreting ascan performed for suspected stenotic disease. Differentfrom the evaluation of bypass grafts and pulmonaryveins, there is a little diagnostic value in performing3-dimensional volume rendered displays for suspectedcoronary artery disease, as lesions are frequently obscuredor overestimated, depending on the rendering parameters.In our practice, 3-dimensional rendering is exclusivelyused for intuitive communication of our findings toreferring physicians and patients.

MPRsFor visualization of the coronary artery tree at

contrast enhanced CT coronary angiography, MPRs16

are widely used and recommended as a robust and easyto perform secondary visualization tool for dataviewing. Because of the isotropic (equal voxel dimensionsin x-axis, y-axis and z-axis) or near isotropic nature of

FIGURE 8. Contrast enhanced, retro-spectively ECG-gated dual-source CTAin a 69-year-old man after LAD infarc-tion. The transverse section (A) showsmyocardial thinning and calcification(arrow) of the LAD myocardial perfusionterritory predominantly in the anteriorwall of the left ventricle. Three-dimen-sional volume rendered display (B)shows the full extent of dystrophiccalcification owing to chronic ischemia.CT-based assessment of cardiac function(C) shows akinesis of the anterior wall,indicated by dark cold spectrum colorsin an automatically generated functionalmodel of the left ventricle.



high-resolution CT acquisitions, image data can berearranged in arbitrary imaging planes with comparableimage quality as in the original transverse section.

MPRs serve the purpose of enabling views ofcoronary artery lesions from different angles and per-spectives, which enables better assessment of stenosisseverity and residual perfused lumen than can beappreciated by only a single projection. This is of

particular importance in the presence of severe calcifica-tions, where a single view often fails to display residuallumen in the vicinity of a heavily calcified plaque (Fig. 9).

Advanced Visualization ToolsAdvanced software tools have become available and

are being continuously refined that facilitate viewing andanalysis of large volume data sets. The common rationale

FIGURE 9. Contrast enhanced, retro-spectively ECG-gated dual-source CTAin a 72 year old woman with atypicalchest pain. Same patient as in Figure 6.The axial source images (left in A) andthe sagittal MPR (middle in A) clearlydemonstrate a moderate ostial stenosis(arrow) of the RCA, adjacent to densecalcifications in the right coronary sinusof the aorta. The lesion is more difficultto detect in the coronal reformat (rightin A). Display as automatically generatedcurved MPR (B) is best suited for stenosisdetection and grading, whereas displayas 3-dimensional volume rendering (C)fails to show the full extent of the lesion.



of most of these software platforms is to provide a meansfor rapid analysis of the coronary artery tree for thedetection and grading of stenosis. Typically, the first stepof postprocessing after review of the transverse sectionsand MPRs (Fig. 10A) consists in automated sculpting ofthe chest wall to enable an unobstructed view of the heart(Figs. 9C, 10B). Threshold-dependent or contour-depen-dent extraction of the coronary arteries from the contrastenhanced data set is then performed (Fig. 10C). Mostsoftware applications enable unraveling of the tortuouscourse of the extracted coronary artery, which affordsintuitive visualization of the entire vessel, typically as anautomatically generated MPR (Figs. 9B, 10D). Lastly,most available software platforms provide tools forquantitative evaluation of stenosis severity (Fig. 10E)based on cross-sectional measurements of vessel diameteror area. Naturally, the accuracy of such tools for stenosisgrading is directly related to the image quality and spatialresolution of the original acquisition and subject to theinherent limitations of coronary CTA for assessingstenosis severity. Therefore, as with any automatedassessment in medicine, the measurement results of vesselanalysis tools should not be trusted blindly, but theexperience and acumen of the physician is still required tovalidate the results in the appropriate clinical context.

THE REPORTWe use a standardized template for reporting

findings at coronary CTA. With the exception of a moredetailed discussion of the coronary arteries and othercardiac structures, our coronary CTA reports are notfundamentally different from general radiology reportsand include all pathology, variations, and changes thatare visible on the different reconstruction series. In theProcedure section, we include the items that are pertinentto appropriate billing in our local health care environ-ment. These items may be different in other geographicalareas, but generally include the section thickness, use ofretrospective ECG-gating, contrast volume and injectionspeed, medications used, and image postprocessingmethods, such as MPRs or 3-dimensional reconstruc-tions. In the Findings section, we begin with describinggeneral cardiac and great vessel anatomy, commenting onthe myocardium (thickness, areas of infarction, scars etc),the cardiac chambers, valves, pericardium, pulmonaryveins, pulmonary arteries, and aorta. A section isdedicated to incidental findings in the chest wall,mediastinum, and lung, for example, the descriptionand classification of lung nodules including recommenda-tions for the follow-up according to standard clinicalpractice.17 In the cardiac section, we report on the

FIGURE 10. Contrast enhanced, retro-spectively ECG-gated 64-slice coronaryCTA in a 59-year-old man with atypicalchest pain. Illustrated is the workflow ofcoronary CTA analysis, using dedicatedvisualization platforms. A circumscribed,short segment high degree stenosis(arrow) in the proximal LAD is difficultto detect on transverse, sagittal, orcoronal MPRs (A). Postprocessing con-sists in automated sculpting of the chestwall to enable an unobstructed view ofthe heart (B), displayed as volumerendering. Automated extraction of thecoronary arteries from the contrast en-hanced data set is performed (C). Theextracted coronary artery is displayed asan automatically generated MPR (D),which affords intuitive visualization ofthe entire vessel and the circumscribedlesion. Automated assessment of stenosisseverity (E) at the level of the lesionshows 78% luminal obstruction (dia-meter measurement). Catheter angio-graphy (F) later confirms the site andseverity of the LAD lesion (arrow).



presence and location of cardiac devices, catheters etc.When coronary artery bypass grafts are present, wedescribe the type, origin, course, site of anastomosis, thepresence, location, and degree of graft stenosis and alsothe quality of the run-off within the grafted vessel distal tothe anastomosis. The presence and course of anomalouscoronary arteries is noted, and also the coronary supplytype (right-dominant, left-dominant or codominant).Finally, each coronary artery (left main, left anteriordescending artery, circumflex, RCA) is separately com-mented on with regards to the presence and type ofatherosclerotic plaque burden (calcified vs. noncalcified).For reporting the site of stenosis, use of the AmericanHeart Association/American College of Cardiology seg-mental model, which is widely employed for researchpurposes, has proved less useful for routine clinicalinterpretation. We rather use the common terminologyfound in routine catheter reports, that describe lesions aslocated in the proximal, mid, or distal portion of therespective main coronary arteries or their side branches(left anterior descending artery: diagonals and septals,circumflex: obtuse marginals, RCA: acute marginals). Weuse our visualization methods (see above) to determinethe severity of stenosis as the percentage of luminalobstruction, on the basis of cross-sectional measurementsof vessel diameter or area.

REFERENCES1. Schoepf U. CT of the Heart: Principles and Applications. Totowa,

NJ: Humana Press; 2004.2. Ohnesorge B, Flohr T, Becker C, et al. Cardiac imaging by means of

electrocardiographically gated multisection spiral CT: initial experi-ence. Radiology. 2000;217:564–571.

3. Flohr T, Schoepf U, Kuettner A, et al. Advances in cardiac imagingwith 16-section CT systems. Acad Radiol. 2003;10:386–401.

4. Ohnesorge BM, Hofmann LK, Flohr TG, et al. CT for imagingcoronary artery disease: defining the paradigm for its application.Int J Cardiovasc Imaging. 2005;21:85–104.

5. Schoepf UJ, Becker CR, Ohnesorge BM, et al. CT of coronaryartery disease. Radiology. 2004;232:18–37.

6. Flohr TG, McCollough CH, Bruder H, et al. First performanceevaluation of a dual-source CT (DSCT) system. Eur Radiol.2006;16:256–268.

7. Jakobs TF, Becker CR, Ohnesorge B, et al. Multislice helical CT ofthe heart with retrospective ECG gating: reduction of radiationexposure by ECG-controlled tube current modulation. Eur Radiol.2002;12:1081–1086.

8. McCollough CH, Primak A, Saba O, et al. Hot topic: doseperformance of a new 64-channel dual-source (DSCT) scanner.RSNA Annual Meeting. 2005:SSE16–06.

9. Ohnesorge B, Becker C, Flohr T, et al. Multislice CT in CardiacImaging: Technical Principles, Clincal Application and FutureDevelopments. Berlin, Heidelberg, London, New York: Springer;2002.

10. Siegel MJ, Schmidt B, Bradley D, et al. Radiation dose and imagequality in pediatric CT: effect of technical factors and phantom sizeand shape. Radiology. 2004;233:515–522.

11. Fleischmann D, Hittmair K. Mathematical analysis of arterialenhancement and optimization of bolus geometry for CT angio-graphy using the discrete fourier transform. J Comput AssistTomogr. 1999;23:474–484.

12. Bae KT, Tran HQ, Heiken JP. Multiphasic injection method foruniform prolonged vascular enhancement at CT angiography:pharmacokinetic analysis and experimental porcine model.Radiology. 2000;216:872–880.

13. Hofmann LK, Zou KH, Costello P, et al. Electrocardiographicallygated 16-section CT of the thorax: cardiac motion suppression.Radiology. 2004;233:927–933.

14. Flohr TG, Schaller S, Stierstorfer K, et al. Multi-detector row CTsystems and image-reconstruction techniques. Radiology. 2005;235:756–773.

15. Vogl T, Abolmaali N, Diebold T, et al. Techniques for the detectionof coronary atherosclerosis: multi-detector row CT coronaryangiography. Radiology. 2002;223:212–220.

16. Ropers D, Baum U, Pohle K, et al. Detection of coronary arterystenoses with thin-slice multi-detector row spiral computed tomo-graphy and multiplanar reconstruction. Circulation. 2003;107:664–666.

17. MacMahon H, Austin JH, Gamsu G, et al. Guidelines formanagement of small pulmonary nodules detected on CT scans: astatement from the Fleischner Society. Radiology. 2005;237:395–400.



Documents

Coronary CTA - SEICATseicat.org/repo/static/public/documentos/top10-cardiaca/... · 2012-01-24 · Key Words: coronary CTA, multidetector-row CT (J Thorac Imaging 2007;22:22–34)