the year in cardiac imaging · 2017-10-25 · mi cardiac mri: prevalence of unrecognized...

Journal of the American College of Cardiology Vol. 61, No. 24, 2013� 2013 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2013.02.080

YEAR IN CARDIOLOGY SERIES

The Year in Cardiac Imaging

Raymond J. Gibbons, MD,* Philip A. Araoz, MD,y Thomas C. Gerber, MD, PHD*yRochester, Minnesota

This review is a sequel to our 8 previous reports summa-rizing the most important published research for single-photon emission computed tomography (SPECT), cardiacpositron emission tomography (PET), cardiac computedtomography (CT), and cardiac magnetic resonance imaging(MRI). This report generally covers the English-languagepublished data between July 1, 2011, and September 30,2012.

We have again organized our summary around topicalthemes in the belief that an integrated, multi-modalityimaging approach is ideal for the solution of most clinicalproblems.

Technical Developments

PET/CT. Plaque biology remains of great interest.Dweck et al. (1) reported the use of combined PET andCT to investigate uptake of fluorine-18-sodium fluoride(18F-NaF) as a marker of plaque calcification and fluorine-18-fluorodeoxyglucose (FDG) as a marker of plaqueinflammation. They found that, in 119 volunteers, 18F-NaFuptake was higher in patients with coronary atherosclerosisand correlated (r ¼ 0.65) with the calcium score, although40% of patients with very high calcium scores (>1,000) hadnormal uptake. In contrast, FDG uptake in the coronaryarteries was confounded by myocardial activity and notincreased in patients with atherosclerosis.

In a second study on the same patients, Dweck et al. (2)reported the use of 18F-NaF and FDG to examine calcifi-cation and inflammation in aortic valve disease. Patientswith aortic stenosis had higher activity of both tracers thancontrol subjects. The 18F-NaF uptake increased with valveseverity (R2 ¼ 0.54) and had a much closer relationship thanFDG (R2 ¼ 0.22). Of patients with aortic stenosis, 91% hadincreased 18F-NaF uptake; only 35% had increased FDGuptake. As noted by Aikawa and Otto (3), many morestudies are needed, but these early results suggest the

From the *Division of Cardiovascular Diseases and Internal Medicine, Department

of Medicine, Mayo Clinic and Mayo Foundation, Rochester, Minnesota; and the

yDepartment of Radiology,Mayo Clinic andMayo Foundation, Rochester, Minnesota.

Dr. Gibbons is a consultant for Lantheus Medical Imaging. Dr. Araoz is a consultant

for Medtronic. Dr. Gerber has reported that he has no relationships relevant to the

contents of this paper to disclose.

Manuscript received December 14, 2012; revised manuscript received February 4,

2013, accepted February 19, 2013.

possibility of PET valve imaging as a measurement tool infuture drug trials.

Patient Safety

Cardiac CT radiation dose reduction. Efforts at balancingradiation dose and image quality in CT continue. Radiationdose varies exponentially with changes of the tube potential,but so does image noise. The decision to use tube voltageof 100 kVp versus 120 kVp is sometimes based on bodymass index (BMI). However, BMI often reflects abdominalthickness more than chest thickness. In a series of106 consecutive cardiac CT angiograms, acquired at 120 kVp(n ¼ 64) or 100 kVp (n ¼ 42), Ghafourian et al. (4) deter-mined both subjective and objective measures of image qualityand compared these measures with both BMI and scout viewattenuation. Scout view attenuation was the best predictor ofimage noise and “low noise” images (area under the curve,0.73). After adjustment for the scout view attenuation, BMIwas not statistically significant. This suggests that the scoutview can help decide which tube potential to use.

The findings of prospective, randomized studies ofimage quality and radiation exposure including more than100 patients are summarized in Table 1 (5,6).

Diagnosis: Coronary Artery Disease

CT: coronary stenoses. With the maturation of technicalaspects of coronary computed tomography angiography(CTA), the research focus is increasingly shifting towardidentifying the patient groups in which coronary CTAprovides the most value.

In a retrospective analysis of 14,048 patients enrolled inthe CONFIRM (Coronary CT Angiography Evaluation forClinical Outcomes: An International Multicenter Registry)(7), pre-test probability was estimated on the basis of sex,age, and chest pain. The observed prevalences for 50% and70% diameter coronary stenoses on coronary CTA weresubstantially lower than those predicted by clinical parame-ters (18% vs. 51% for>50% stenoses, 10% vs. 42% for>70%stenoses; p < 0.001), particularly for patients with typicalangina (19% vs. 71% for >70% stenosis) (Fig. 1). In anaccompanying editorial, Diamond (8) attributed some ofthese differences to the use of self-administered question-naires and verification bias.

http://dx.doi.org/10.1016/j.jacc.2013.02.080

Table1

Effec

tof

Cardiac

CTRad

iation

Red

uction

Tech

niqu

eson

Imag

eQua

lity

FirstAutho

r(R

ef.#)

Sca

nTe

chniqu

esInve

stigated

nMainOutco

me

Variable

Major

Find

ings

Mea

nEffec

tive

Dos

eCom

men

ts

Hau

sleiter(5)

ProspectiveEC

Gtrigge

ring(axial)vs.

retrospe

ctiveEC

Gga

ting(helical)

400

Imag

equ

ality

(ordinal,1–4)

Imag

equ

ality

noninferiorin

axialv

s.he

lical

scan

ning

(3.36�

0.59vs.3.37�

0.56)

3.5

�2.1

vs.11.2

�5.9

mSv

Beta-blocke

ruse77%

vs.79%

HR54�

6.1

vs.56�

5.5

Nee

fjes(6)

High-pitchscan

ning

vs.prospe

ctiveEC

Gtrigge

ring(narrow

windo

w)(HR<65

beats/min;n¼

160)

ProspectiveEC

GTrigge

ring

(widewindo

w)vs.retrospe

ctive

ECGga

tingwith

ECTC

M(HR�6

5be

ats/min;n¼

112)

272

Imag

equ

ality

(ordinal,1–3)

Imag

equ

ality

scores:

2.67�

0.38vs.2.86�

0.21(p

<0.001)

2.81�

0.28vs.2.80�

0.38(p

¼0.54)

Highpitchvs.prospe

ctivetrigge

ring

(narrow)

0.81�

0.03vs.2.74�

1.14mSv

(120kV

p)1.65�

0.69vs.4.21�

1.2

mSv

(100kV

p)Prospectivetrigge

ring(wide)

vs.

retrospe

ctivega

ting

4.07�

1.07vs.5.54�

1.76mSv

(120kV

p)7.50�

1.79vs.9.83�

3.49mSv

(100kV

p)

Metop

rolol1

00mgPO1hbe

fore

scan

unless

contraindicated

Stud

ies>100pa

tients.

CT¼

compu

tedtomog

raph

y;EC

G¼

electrocardiog

raph

y;EC

TCM

¼electrocardiog

raph

y-correlated

tube

curren

tmod

ulation;

ETT¼

exercise

trea

dmill

testing;

FDG¼

fluo

rode

oxyglucose;HR¼

heartrate.

Gibbons et al. JACC Vol. 61, No. 24, 2013Year in Cardiac Imaging June 18, 2013:2471–83

2472

The high negative predictive value (NPV) of coronaryCTAis important for its use to “rule out” coronary artery disease(CAD). Most of the studies showing this high NPV wereperformed in populations with low to intermediate pre-testprobability. In a retrospective analysis of prospectivelyacquired data for the CORE-64 study (Coronary ArteryEvaluation Using 64-Row Multi-Detector Computed To-mography Angiography), Arbab-Zadeh et al. (9) examinedthe impact of disease prevalence and coronary calcium on thediagnostic accuracy of coronary CTA. Although the areasunder the curve for diagnostic accuracy were similar forpatients with intermediate probability of CAD, high proba-bility for CAD, and known CAD (0.93, 0.92, and 0.93,respectively), the NPVs varied substantially (0.90, 0.83, and0.50, respectively [p< 0.001]) and decreased to 0.81when thecalcium score was�600. In patients with high calcium scoresand high pre-test probability for obstructive CAD, the NPVof coronary CTA is lower than previously reported.

Population radiation exposure can be potentially reducedby prospectively identifying patients who are likely to havenondiagnostic image quality of coronary CTA. In a retro-spective analysis of registry data, Vanhecke et al. (10)developed an “uninterpretable risk score” in 8,585 symp-tomatic patients and used it to predict uninterpretablestudies in 915 subsequent symptomatic patients. For each 4-point increase of the score (maximum, 32 points), the rate ofencountering at least 1 uninterpretable coronary segmentincreased approximately 1.5-fold. Increased heart rate andthe coronary calcium score were most predictive of unin-terpretable coronary CTA results. Patients with an unin-terpretable study had an increased frequency of adverseoutcomes over 3-month follow-up.CT myocardial perfusion imaging. The technology andapplication of computed tomography myocardial perfusionimaging (CTMPI) continue to evolve.

Ko et al. (11) compared invasive fractional flow reserve(FFR) measurements and adenosine stress CTMPI in 86coronary perfusion territories that were supplied by vesselswith �50% diameter stenosis. Blinded qualitative assess-ment of segmental CTMPI after triphasic injection ofcontrast during adenosine infusion correctly identified 31 of41 (76%) territories with FFR �0.8 (Fig. 2). The CTMPIwas normal in 38 of 45 (84%) myocardial segments withpreserved FFR. The combination of a defect on CTMPIand a �50% stenosis on coronary CTA (with 320 detectorrows) was 98% specific for identifying abnormal FFR, andthe combination of normal CTMPI and a stenosis <50% orless on coronary CTA was 100% specific for identifyingpreserved FFR (Fig. 2).

Computational fluid dynamics can be applied to estimateFFR from typical coronary CTA protocols, which currentlyrequires proprietary software and approximately 5 h/exam.Koo et al. (12) compared FFR by CT with invasive FFR in159 vessels of 103 patients (Fig. 3). The accuracy ofabnormal FFR by CT to identify vessels with abnormalinvasive FFR (0.8) was 84.3%. The FFR by CT correlated

Figure 1 Prevalence of Coronary Artery Diameter Stenoses

Prevalence of coronary artery diameter stenoses (�50%) as observed (solid bars)

on computed tomography angiography and as expected (open bars) on the basis

of clinical criteria in patients with varying presentations, stratified by age. The

numbers above note the observed/expected ratio of coronary artery stenosis

prevalence. This ratio is increased with age in men with atypical angina

(p < 0.001) and typical angina (p < 0.001) but not in women. CAD ¼ coronary

artery disease. Reprinted, with permission, from Cheng et al. (7).

Figure 2FFR Compared With Combinations ofCoronary CTA and CT Myocardial Perfusion Imaging

Fractional flow reserve (FFR) compared with combinations of coronary computed

tomography angiography (CTA) and computed tomography (CT) myocardial

perfusion imaging findings in 86 myocardial perfusion territories. The FFR is lowest

in patients who have both stenosis �50% on coronary CTA and myocardial

perfusion abnormalities and highest in patients with no significant stenosis and

no myocardial perfusion defects. Among patients with discrepant results, FFR

correlates better with myocardial perfusion abnormalities then with angiographic

stenosis. CTA �50 ¼ diameter stenosis �50% on coronary CTA; CTA <50 ¼diameter stenosis <50% on CTA; CTP þve ¼ CT abnormal myocardial perfusion by

CT; CTP �ve ¼ normal myocardial perfusion by CT. Reprinted, with permission,

from Ko et al. (11).

JACC Vol. 61, No. 24, 2013 Gibbons et al.June 18, 2013:2471–83 Year in Cardiac Imaging

2473

well with invasive FFR (r ¼ 0.717, p < 0.001) with a slightunderestimation (0.022 � 0.116, p ¼ 0.016).MRI and SPECT. In a cardiac MRI examination,multiple parameters can be measured, each of which canseparately be used to detect CAD. However, there have beenno prospective studies comparing a multiparametric cardiacMRI approach with SPECT for the detection of CAD. Ina prospective study (Clinical Evaluation of MAgneticResonance imaging in Coronary heart disease) performed byGreenwood et al. (13), 628 patients with angina underwenta multiparametric cardiac MRI examination, stress SPECT,and subsequent invasive angiography.

The cardiac MRI protocol included rest and adenosinestress perfusion, resting wall motion assessment, delayedenhancement, and coronary magnetic resonance angiog-raphy. The cardiac MRI findings suggesting flow-limitingstenoses (change in perfusion from rest to stress) ormyocardial infarction (MI) (resting wall motion abnor-mality, delayed enhancement) were considered positive. Ona 2-day SPECT protocol, both fixed and reversible defectswere apparently considered positive. Combined cardiac MRIhad a sensitivity of 86%, which was significantly >67.9% forSPECT (p < 0.001). Both cardiac MRI and SPECT had

a specificity of 83% (Fig. 4). The greater spatial resolution ofcardiac MRI might explain the improved sensitivity.SPECT/cardiac MRI/PET. Single-photon emission com-puted tomography, cardiac MRI, and PET have all beenused for rest/stress perfusion imaging for many years.Jaarsma et al. (14) performed a meta-analysis of 143 studies(105 SPECT, 27 cardiac MRI, and 11 PET) publishedbetween 1990 and 2010 in which these modalities were usedto diagnose CAD by invasive angiography. In contrast to theprospective study by Greenwood et al. (13) (described in thepreceding text), Jaarsma et al. (14) found that the pooledsensitivity for all 3 modalities were not significantly differentfrom each other (SPECT ¼ 88%, cardiac MRI ¼ 89%,PET ¼ 84%). However, SPECT had a lower pooledspecificity (61%) than either cardiac MRI (76%) orPET (81%).MI cardiac MRI: prevalence of unrecognized infarction.Delayed enhancement imaging with cardiac MRI is moresensitive than electrocardiography (ECG) for detecting MI.However, previous population studies of the prevalence ofclinically unrecognized MI have used ECG as the referencestandard. Schelbert et al. (15) studied 936 subjects (medianage 76 years, 52% female) as part of a cohort calledICELAND MI. In that cohort, cardiac MRI delayedenhancement detected unrecognized MI in 17% of subjects(95% confidence interval [CI]: 14% to 19%), which wassignificantly greater than the 5% rate of unrecognized MIdetected by ECG (95% CI: 4% to 6%, p < 0.001).MI cardiac MRI: T1 mapping. Most cardiac MRI tissuecharacterization techniques depend on relative signaldifferences and do not give absolute measurements. Over the

Figure 3 FFR Measured Noninvasively From Coronary CTA

Fractional flow reserve measured noninvasively from coronary CTA with computational fluid dynamics in a patient with significant-appearing stenosis on CTA. (A) Noncalcified

plaque of interest (red arrow) on CTA; (B) same stenosis as seen on invasive, selective coronary angiography (red arrow). (C) The FFRCT is 0.85. Vessel portions in blue have

the highest FFR, green shading denotes intermediate FFR. (D) The FFR measured invasively was 0.84, confirming lack of hemodynamic significance. Reprinted, with permission,

from Koo et al. (12). Abbreviations as in Figure 2.


2474

past several years, there has been interest in measuring theabsolute value of a cardiac MRI tissue property called T1relaxation time. Dall’Armellina et al. (16) used a 3-T cardiacMRI scanner in 41 patients with acute MI (32 STEMI) tomeasure the utility of T1 mapping for the detection of acuteMI. For each of 674 segments the authors compared theaverage T1 value within the segment with cardiac MRIdelayed enhancement. The segmental mean T1 was associ-ated with (subjective) segmental delayed enhancement(p < 0.01). In 198 segments with abnormal function, T1was associated with recovery of function at 6 months(p < 0.004). The authors concluded that T1 mapping mightbe useful in the evaluation of MI.

CAD Prognosis

PET. There has been a longstanding interest in use ofstress imaging techniques to improve risk stratification.Murthy et al. (17) examined the value of quantitativecoronary flow reserve (CFR) measurements by PET in2,783 patients followed for a median of 1.4 years. Patientswith CFR <1.5 had a 5.6-fold increase in cardiac death,compared with patients with CFR >2.0 (Fig. 5). The CFRcorrectly reclassified 34.8% of intermediate risk patients; thereclassification index in these patients was 0.484.

CT: coronary artery calcium. The value of coronaryartery calcium (CAC) scanning for identifying patients foraggressive, cost-effective pharmacologic risk factor manage-ment has not been well-defined. Blaha et al. (18) performeda retrospective analysis of the 6,814 individuals enrolled inthe MESA (Multi-Ethnic Study of Atherosclerosis) (2000to 2002) and found that 2,083 met the inclusion criteria forthe recent JUPITER (Justification for the Use of Statins inPrimary Prevention: An Intervention Trial EvaluatingRosuvastatin) of high-sensitivity C-reactive protein. Thesepatients were further stratified by CAC. After adjustment forFramingham risk factors, “high” high-sensitivity C-reactiveprotein was not predictive of coronary (hazard ratio [HR]:0.98, 95% CI: 0.62 to 1.57) or cardiovascular (HR: 1.15,95% CI: 0.78 to 1.68) events over a mean follow-up of 5.8years. In contrast, after adjustment for Framingham riskfactors, CAC had a HR of 6.65 (95% CI: 2.99 to 14.78) forcoronary events and an HR of 3.06 (95% CI: 1.82 to 5.13)for cardiovascular events. Reclassification data were notpresented. With the event reduction rates from theJUPITER trial, the predicted number needed to treat toprevent coronary events over 5 years in patients with calciumscores of 0, 1 to 100, and >100 were 549, 94, and 24,respectively; for cardiovascular events, they were 124, 54, and19, respectively.

Figure 4ROC Curves of Summed Stress Scores forCMR and SPECT

Receiver operating characteristic (ROC) curves of summed stress scores for

cardiac magnetic resonance imaging (CMR) and single photon emission computed

tomography (SPECT). LAD ¼ left anterior descending; LCx ¼ left circumflex;

LMS ¼ left main stem; RCA ¼ right coronary artery. Modified, with permission,

from Greenwood et al. (13).


2475

CT: coronary CTA. The number of stenosed coronaryarterial segments on invasive angiography has long beenrecognized to predict poor outcome. In the last year, multiplelarge studies have been performed to determine whethersimilar prognostic information can be obtained from coro-nary CTA (Table 2). In particular, the CONFIRM registry

Figure 5 Kaplan-Meier Plot of Cardiac Mortality

Kaplan-Meier plot of cardiac mortality (A) and after adjustment for age, sex, body mass ind

artery disease, chest pain, dyspnea, early revascularization, last ejection fraction, summ

between coronary flow reserve and cardiac mortality. HR ¼ hazard ratio. Reprinted, with

(19) included 27,125 consecutive patients that were used in3 of the papers cited in the table.Cardiac MRI: acute MI. Recent cardiac MRI studies haveshown that right ventricular (RV) infarction and/or edemaoccurs in 51% of STEMI patients (20). Grothoff et al. (21)performed cardiac MRI to assess the RV in 450 patients 1 to4 days after primary angioplasty for STEMI. Patients werefollowed for an average of 21 months for a compositeendpoint of death, reinfarction, and new congestive heartfailure. The authors defined RV injury as a combination ofedema and associated wall motion impairment. An RVinjury was independently associated with adverse events(HR: 3.36, p < 0.001) in a multivariate model that includedleft ventricular ejection fraction (LVEF) and Thrombosis InMyocardial Infarction risk score.

Chronic CAD: Patient Management

SPECT. Shaw et al. (22) reported on the WOMEN(What Is the Optimal Method for Ischemia Evaluation inWomen) Trial, a randomized trial comparing SPECT andexercise treadmill testing (ETT) in the evaluation ofsymptomatic low- or intermediate-risk women with sus-pected CAD, an interpretable ECG, and the ability toexercise (Table 3). Two-year survival free of major adversecardiac events was similar in the 2 groups (98.0% for ETT,and 97.7% for SPECT, p ¼ 0.59) (Fig. 6). There was nodifference in the percentage of women who were angina-freeduring follow-up between the 2 strategies. Most impor-tantly, the cumulative diagnostic cost savings for ETTcompared with SPECT was 48%.

In a separate report, Shaw et al. (23) reported a post hocsubstudy from the COURAGE (Clinical OutcomesUtilizingRevascularization and Aggressive Drug Evaluation) trial thatexamined the impact of baseline stress SPECT ischemia (as

ex, hypertension, hyperlipidemia, diabetes, family history, tobacco use, prior coronary

ed stress score, and ejection fraction reserve, (B) showing a significant association

permission, from Murthy et al. (17).

Table 2 Coronary CTA Prognosis Studies With >1,000 Patients

First Author(Ref. #) n Setting

Numberof CT

DetectorRows Outcome Follow-Up

Annual EventRate for Patients

With NormalCoronary Arteries Multivariate Model Multivariate Result Reclassification Analysis

Andreini (44) 1,196 Single center 64 SD Cardiac deathor nonfatal MI

4.3 � 1.8 yrs No events Framingham riskscore, anti-anginalmedication, statin,aspirin

1-vessel CAD or greater bycoronary CTA associatedwith increased event rate(HR: 8.76 for 1 vessel CAD,p <0.001)

Not performed

Min (45) 23,854 CONFIRM 64 SD,64 DD,128 DD

All-causemortality

2.3 � 1.1 yrs 0.28% Age, clinical riskfactors

1-vessel CAD or greater bycoronary CTA associatedwith increased all-causemortality (HR: 2.35 for 1vessel CAD, p <0.001)

Not performed

Chow (46) 13,966 CONFIRM 64 SD,64 DD,128 DD

All-causemortality

1.9 � 0.0 yrs 0.4% NCEP/ATP III risk,LVEF

Non-obstructive CAD* orgreater by coronary CTAassociated with increasedall-cause mortality (HR ¼2.57 for non-obstructive),

NRI ¼ 17.8%, p < 0.001

Villines (47) 10,037 CONFIRM 64 SD,64 DD,128 DD

All-causemortality

2.1 yrs,IQR ¼ 2.0 yrs

0.27% Not performed forall-cause mortalityendpoint

NA Among patients with coronarycalcium score of zero, CADwas not associated withincreased all-causemortality

Lin (48) 2,583 patientswith <50%stenosis bycoronary CTA

2 centers 64 SD All-causemortality

3.1 � 0.5 yrs 0.34% Framingham riskscore

Presence of any, non-obstructive plaqueassociated with increasedall-cause mortality (HR:2.74, p ¼ 0.003)

NRI ¼ 20.5%, p ¼ 004

Yiu (49) 2,432 4 centers 320 SD,64 SD,

Cardiac deathor nonfatal MI

2.2 yrs(25–75thpercentile:1.3–3.1 yrs)

0.07% for men,0.2% for women

Not performed NA CTA predictive of events inpatients �60 yrs of age andin men <60 (p <0.01 forall) but not in women <60yrs of age (p < 0.45)

*<50%. Reprinted, with permission, from Shreibati JB et al. (27).CAD ¼ coronary artery disease; CONFIRM ¼ coronary computed tomography angiography evaluation for clinical outcomes registry; CTA ¼ computed tomography angiography; DD ¼ dual detector; HR ¼ hazard ratio (note that the reported HRs are for single vessel coronary

disease); IQR ¼ interquartile range; LVEF ¼ left ventricular ejection fraction; MI ¼ myocardial infarction; NCEP/ATP III ¼ National Cholesterol Education Program/Adult Treatment Panel III; NRI ¼ net reclassification index (for Chow the comparison was of CAD þ LVEF modelcompared with a model containing clinical predictors and LVEF; for LIN the comparison was of CAD compared with Framingham risk); SD ¼ single detector.

Gibbonset

al.JACC

Vol.61,No.24,2013Year

inCardiac

Imaging

June18,2013:2471–83

2476

Table 3 Selected Studies of Clinical/Economic Impact of Noninvasive Testing for CAD

First Author (Ref. #) N Modality Endpoint Results

Shaw (22) 824 SPECT vs. ETT Death/MI/adm.Angina-freeDiagnostic costs

2.3% at 2 yrs vs. 1.7% at 2 yrs, p ¼ 0.5964.9% at 2 yrs vs. 60.4% at 2 yrs, p ¼ 0.58$492.67 (median) vs. $174.14 (median), p ¼ 0.001

Hachamovitch (24) 1,703 SPECT, PET, CTA Referral to cath

AspirinLipid-lowering

Cath after mod/sev abn >Cath after mild abn >Cath after nlCath after CTA >cath after SPECT or PET for nl or mildly abnIncreased from 58% to 76% after mod/sev abn studyIncreased from 63% to 77% after mod/sev abn study

Tandon (25) 2,442 CTA vs. SPECT Referral to cathReferral to revasc.

Similar for CTA (10.6%) and SPECT (10.2%)Similar for CTA (6.2%) and SPECT (5.9%)

McEvoy (50) 2,000 CTA vs. no CTA Aspirin use at 18 monthsStatin use at 18 monthsReferral to revasc.

26% for CTA abn vs. 3% for CTA nl vs. 6% for no CTA20% for CTA abn vs. 3% for CTA nl vs. 6% for no CTA1% for CTA vs. 0.1% for no CTA, p < .001

Pazhenkottil (26) 318 CTA and SPECTMatched stenosis vs.unmatched stenosis vs. nl

Referral to revasc. 41% in matched vs. 11% in unmatched vs. 0% in nl,p < 0.0001

Shreibati (27) 282,830 CTA vs. SPECT Referral to cathPCICABGHealthcare spending

Greater with CTA AOR: 2.19 (95% CI: 2.08–2.32), p < 0.001Greater with CTA AOR: 2.49 (95% CI: 2.28–2.72), p < 0.001Greater with CTA AOR: 3.00 (95% CI: 2.63–3.41) p < 0.001Greater with CTA: $4,200 (3,193–5,267), p < 0.001

Abn ¼ abnormal; AOR ¼ adjusted odds ratio; CABG ¼ coronary artery bypass grafting surgery; cath ¼ cardiac catheterization; CI ¼ confidence interval; ETT ¼ exercise treadmill test; Mod ¼ moderately; nl ¼normal; PCI ¼ percutaneous coronary intervention; PET ¼ cardiac positron emission tomography; Revasc ¼ revascularization (PCI or CABG); Sev ¼ severity; SPECT ¼ single-photon emission computedtomography; other abbreviations as in Table 2.


2477

assessed by site investigators) on the effectiveness of optimalmedical therapy and PCI with optimal medical therapy alone.Surprisingly, the extent of ischemia did not predict adverseevents or alter treatment effectiveness, contrary to both priorobservational studies and a prior analysis with quantitativecore laboratory methods in a different COURAGE cohort.SPECT/PET/CTA. Hachamovitch et al. (24) reportedthe results of the large multicenter prospective registry, theSPARC study (Study of Perfusion and Anatomy’s Role inCoronary Artery Disease). They assessed 90-day post-testrates of invasive angiography and medication changes in565 patients studied by SPECT, 548 patients studied byPET, and 590 patients studied by coronary CTA.

Figure 6Kaplan-Meier Plot Showing Major AdverseCardiovascular Event-Free Survival

Kaplan-Meier plot showing major adverse cardiovascular event-free survival,

according to initial randomized assignment to exercise treadmill test (ETT) or

exercise myocardial perfusion imaging by single photon emission computed

tomography (MPI). Reprinted, with permission, from Shaw et al. (22).

Surprisingly, in patients with the most severely abnormaltests, 38% to 61% were not referred to invasive angiography,20% to 30% were not receiving aspirin 90 days after thebaseline test, 35% to 40% were not receiving a beta-blocker,and 20% to 25% were not receiving a lipid-lowering agent.Changes in medication and referral to invasive angiographywere far less frequent than expected after moderately orseverely abnormal studies (Fig. 7). Compared with SPECTand PET, CTA was associated with more changes in aspirinand lipid-lowering agent use and higher 90-day rates ofreferral to invasive angiography after normal, nonobstructive,or mildly abnormal test results.

Figure 7 Post-Test Changes in Patient Management

Frequency of referral for cardiac catheterization (Cath), medication (Med) change,

both, or neither within 90 days after cardiac imaging. Within each test result,

chi-square tests of association are shown. Even when tests were moderately or

severely abnormal, approximately one-half of the patients did not undergo cardiac

Cath or Med change; one-quarter did not undergo either. Reprinted, with

permission, from Hachamovitch et al. (24).

Table 4Adjusted Association Between Noninvasive Cardiac Testing and 180-Day Spending in282,830 Fee-for-Service Medicare Beneficiaries

SPECT CTA Stress Echo ETT

Adjusted Spending 24,136 28,336 19,156 16,687

Difference 0 þ4,200 �4,981 �7,449

(95% CI), $ (Reference) (3,193 to 5,267) (�4,991 to �4,969) (�7,452 to �7,444)

p (multivariable model) <0.001 <0.001 <0.001

Years 2006 to 2008. Modified, with permission, from Shreibati JB et al. (27).Echo ¼ echocardiogram; other abbreviations as in Tables 2 and 3.


2478

SPECT/CTA. Several studies examined the impact ofSPECT and coronary CTA on patient management.Tandon et al. (25) compared 1,221 patients evaluated withcoronary CTA with 1,221 matched patients evaluated bySPECT. Referral to invasive angiography was similar inboth groups (10.6% for coronary CTA; and 10.2% forSPECT). Rates of revascularization were also similar.Coronary CTA had a lower false positive rate than SPECT(9.7% vs. 25.8%, p ¼ 0.009). Pazhenkottil et al. (26)examined the impact of hybrid imaging with SPECT andcoronary CTA in 318 consecutive patients. Referral torevascularization was higher in patients with matchedabnormalities (41%), compared with those with unmatchedabnormalities (11%) or those with normal studies (0%).Shreibati et al. (27) examined subsequent noninvasivetesting and costs after SPECT, coronary CTA, stressechocardiography, and ETT in 282,830 fee-for-serviceMedicare beneficiaries. Compared with SPECT, coronaryCTA was more often followed by cardiac catheterization(adjusted odds ratio [OR]: 2.19), PCI (adjusted OR: 2.49),

Table 5 Randomized Trials of Chest Pain Triage in the ED

First Author(Ref. #) n Pts Random MACE Diff.

Farkouh (51) 424 IR CPU vs. Adm 6 months NS

Udelson (52) 2,475 ACS SPECT vs. UC 30 days NS

Goldstein (53) 197 CP CTA vs. SPECT 6 months NS

Goldstein (54) 699 LR CTA vs. SPECT 6 months NS

Litt (55) 1,370 LIR CTA vs. UC 30 days NS

Hoffmann (56) 1,000 ACS CTA vs. UC 28 days NS

Miller (57) 109 IR MR vs. Adm 12 months NS

Miller (58) 120 LR MR vs. MD 30 days NS

ACS ¼ acute coronary syndrome; Adm ¼ admission; CP ¼ chest pain; CPU ¼ chest pain unit; d ¼ dayscoronary angiography; IR ¼ intermediate risk; LR ¼ low risk; MD ¼ physician choice for testing; mo ¼ mrandomization; Random ¼ randomization; UC ¼ usual care; w/o ¼ without; Y ¼ yes.

and coronary artery bypass surgery (adjusted OR: 3.0). Thisled to higher total healthcare spending (Table 4). Incontrast, compared with SPECT, both stress echocardiog-raphy and ETT were associated with lower healthcarespending (Table 4).SPECT/cardiac MRI. Van der Veen et al. (28) demon-strated the feasibility of an automated software tool usinggeometrical algorithms to quantify ventricular dyssynchronywith gated SPECT in 24 heart failure patients with provendyssynchrony by cardiac MRI and compared the results witha normal population of 80 patients. Several parametersdemonstrated highly significant differences between thenormal and abnormal populations.

Acute CAD: Patient Management

Chest pain triage in the emergency room remains a majorchallenge. Several randomized trials in the past year haveevaluated coronary CTA and cardiac MRI. Table 5compares them with the previous randomized trials on this

Admissions Cost

ED Time ResultsH CPU ED Total

R R N Y N Lower adm rate and totalcosts with CPU

Y Y N N Y Lower adm rate withSPECT in pts w/oischemia

Y E Y N Y Lower ED costs with CTA

ICA only E Y N DX only Lower ED costs with CT;shorter DX time withCTA

Combined N N Y Lower rate of H/CPUadm, shorter ED timewith CTA

Y Y N Y Y Shorter stay, fewer admwith CTA

No difference in totalcosts

R R N Y N Lower adm rate and totalcosts with MR

Y E Y N Y No diff. in ED time withMR

ED costs higher with MRVery low adm rate in bothgroups

; Dx ¼ diagnostic; E ¼ required for enrollment; ED ¼ Emergency Department; H ¼ hospital; ICA ¼onths; MR ¼ stress cardiac magnetic resonance imaging; N ¼ no; Pts ¼ patients; R ¼ required by

Figure 8 CT Angiography of Heart, Aortic Root, and Ascending Aorta

Computed tomography (CT) angiography of heart, aortic root and ascending aorta, reformatted in oblique-sagittal (A) and oblique-coronal (B) planes. Diameter measurements

are taken at the sinotubular junction (*), the aortic annulus (y), and the left ventricular outflow tract (z). Reprinted, with permission, from Jabbour et al. (33).


2479

subject. Gibbons (29) highlighted the strengths and weak-nesses of the trials of coronary CTA. Prasad et al. (30) calledfor randomized trials to determine whether any testing isjustified in low-risk patients.

Table 6 Cardiovascular CT Imaging Before TAVI

First Author(Ref. #)

ImagingTechniques n Outcomes Variables

Tamborini (32) 3D TEE64-detectorrow CT

122 LCC lengthLM-AA distance (pre-op)Min-D and Max-D, aortic

annulus

LCC lenLM-AA d0.001

Min-D: 2Max-D:

Jabbour (33) Cardiac MRI128-detectorrow dualsource CT

TTE

202 Largest diameterAortic valve annulusSinus of ValsalvaSinotubular junctionAscending aorta

Bias (mAortic vSinus oSinotubAscendiBias (mAortic vSinus oSinotubAscendiBias (mAortic vSinus oSinotubAscendi

Willson (59) 64-detectorrow CT

109 Difference betweenprosthesis size andannular sizemeasured by CT

Presence and severity ofpost-procedural PAR

Moderasizingfor tr

Differenpredicircum

Studies >100 patients.AR ¼ aortic regurgitation; AUC ¼ area under the curve; CT ¼ computed tomography; LCC ¼ left corona

minimal diameter; MRI¼magnetic resonance imaging; PAR¼ paravalvular aortic regurgitation; pre-op ¼echocardiography; TTE ¼ transthoracic echocardiogram; 3D ¼ 3-dimensional.

Structural Heart Disease

PET/CT. Sarrazin et al. (31) reported on the use of FDGPET/CT imaging in patients with suspected infections of

Major Findings Comments

gth: 13.7 � 1.8 vs. 13.6 � 1.8 (r ¼ 0.69, p < 0.001)istance: 13.5 � 2.2 vs. 13.9 � 2.2 (r ¼ 0.83; p <

)3.3 � 2.1 vs. 21.4 � 2.2 (r ¼ 0.70; p < 0.001)24.6 � 2.3 vs. 25.3 � 2.2 (r ¼ 0.68, p < 0.001)

m) cardiac MRI vs. CTalve annulus: 0.39 � 2.54f Valsalva: �0.24 � 2.25ular junction: �0.81 � 2.01ng aorta: �0.1 � 2.21m) cardiac MRI vs. TTEalve annulus: 4.52 � 3.29f Valsalva: �0.45 � 3.45ular junction: �0.73 � 3.94ng aorta: 1.78 � 4.21m) CT vs. TTEalve annulus: 4.1 � 3.25f Valsalva �0.33 � 3.43ular junction: 0.06 � 4.6ng aorta: 1.56 � 4.04

Presence and severity ofAR after TAVI wereassociated with largeraortic annulusmeasurements byboth cardiac MRI andCT but not TTE

te to severe PAR in 13/102 pts associated with under-of prosthesis (�0.7 � 1.4 mm vs. a 0.9 � 1.8 mm)

ivial to mild PARce between prosthesis size and annular size by CTctive of PAR (AUCs: mean diameter, 0.81; area, 0.8;ference, 0.76)

ry cusp; LM-AA ¼ left main coronary ostium to aortic annulus; Max-D ¼ maximal diameter; Min-D ¼pre-operative; pts ¼ patients; TAVI ¼ transcatheter aortic valve implantation; TEE¼ transesophageal

Figure 9Kaplan-Meier Analysis of 143 Patients WithAortic Stenosis

Kaplan-Meier analysis of 143 patients with aortic stenosis shows a significant

increase in cardiac mortality with midwall fibrosis (orange dashed line) or

infarction (green solid line) compared with the no late enhancement (LGE) group

(blue dotted lines). Modified, with permission, from Dweck et al. (35).


2480

implantable devices (ID) (i.e., pacemakers and implantablecardioverter-defibrillators). They compared FDG uptake in42 patients with suspected ID infection with 24 patientswith ID without infection (12 with recent IDs, 12 witholder IDs). Of the 42 patients, 32 had positive FDG scans;24 underwent extraction, and 6 patients were treated assuperficial infection with clinical improvement. Ten patientswith negative scans were treated with antibiotics onlywithout recurrence. Patients with ID without infection hadeither no FDG uptake or mild uptake at the level of theconnector.CT transcatheter aortic valve implantation: planning.The success of transcatheter aortic valve implantation

Figure 10 CT Characteristics of Culprit Lesion in a Patient With Pos

Coronary computed tomography (CT) angiography plaque imaging in a patient who had ca

coronary intervention (PCI) of a high-grade distal right coronary artery stenosis (arrow). (A

coronary artery. Cross sectional reformations in (C) a proximal reference segment, (D) a

(remodeling index: 1.28), spotty calcification (D) (red circle), and low CT density (16 Ho

depends on various patient-related factors (Fig. 8).Table 6 summarizes studies that examined the value ofcardiac CT compared with other imaging modalities foroptimizing and predicting the outcome of transcatheteraortic valve implantation in more than 100 patients(32,33). The aortic annular dimensions are larger whenmeasured by cardiac CT rather than by echocardiography.Sizing aortic valve prostheses on the basis of echocardio-graphic measurements alone is associated with a higherprevalence and severity of post-procedural paravalvularaortic regurgitation.Cardiac MRI: aortic stenosis. In aortic stenosis, cardiacMRI might also be used to detect a midwall pattern ofmyocardial fibrosis (34). This pattern is distinct from thefibrosis seen in MI from CAD. However, it is not knownwhether midwall fibrosis portends poor prognosis. Dwecket al. (35) studied 143 patients with moderate or severeaortic stenosis who underwent gadolinium-enhanced car-diac MRI for evaluation of delayed enhancement. Patientswere followed for 2.0 � 1.4 years for all-cause mortality.Midwall fibrosis was independently associated withincreased all-cause mortality (HR: 5.35, p ¼ 0.03) afteradjustment for LVEF, infarct-pattern fibrosis, and subse-quent aortic valve replacement. Kaplan-Meier survivalestimates showed a significant increase in cardiac mortalityin both the midwall fibrosis and the infarct groups (Fig. 9),although at 4 years the number of patients in each groupwas small. Midwall fibrosis seems to be an independentprognostic marker in patients with aortic stenosis.Cardiac MRI. In hypertrophic cardiomyopathy somestudies have found that myocardial fibrosis, as measured withdelayed-enhancement cardiac MRI is associated with cardiacdeath, but others have not. These studies have generally

t-PCI in a TnT Elevation �3� Upper Limit of Normal

rdiac troponin elevation �3� upper limit of normal after percutaneous

) Selective coronary angiogram. (B) Multi-planar reconstruction of the right

t the stenosis, and (E) a distal reference segment show positive remodeling

unsfield units). Reprinted, with permission, from Watabe et al. (41).


2481

been underpowered for outcomes. Therefore, the presence ofdelayed enhancement is currently only recommended for“resolving clinical decision making” when risk stratification“is inconclusive after documentation of the conventional riskfactors” (36). Green et al. (37) performed a meta-analysis of4 single-center studies. Pooled together, the 4 studiesincluded 1,063 patients followed an average of 3.1 years with36 cardiac deaths. In their analysis, delayed enhancement wasassociated with cardiac death (pooled OR: 2.92, p¼ 0.0047).However, because no multivariate analysis was performed, itis not clear whether delayed enhancement adds prognosticinformation that is incremental to clinical risk factors.

Cardiac MRI can risk-stratify patients undergoing cardiacresynchronization therapy (CRT), by using delayed en-hancement imaging as a marker for poor response to CRTand/or poor overall prognosis. Leyva et al. (38) studied 97patients with dilated cardiomyopathy and 161 patients withischemic cardiomyopathy undergoing CRT. Patientsunderwent a cardiac MRI examination before CRT place-ment and then follow-up for up to 8.7 years. Among thepatients with dilated cardiomyopathy, those with midwallfibrosis had increased cardiovascular mortality, even afteraccounting for New York Heart Association functional class,QRS duration, and LVEF (p ¼ 0.008). Patients with dilatedcardiomyopathy and midwall delayed enhancement hada similarly poor prognosis as patients with ischemiccardiomyopathy, but patients with dilated cardiomyopathyand no delayed enhancement had a much better outcome.

By measuring a cardiac MRI property know as TI beforeand after administration of intravenous contrast material,extracellular volume can be calculated. Wong et al. (39) usedthis technique in 793 patients and followed them for a meanof 0.8 years. Extracellular volume fraction was associatedwith all-cause mortality, even after accounting for LVEF,MI size, and patient age.

Vascular Structure

CT: plaque imaging. Efforts continue to define the clinicalimplications of plaque composition by coronary CTA.

Ozaki et al. (40) compared the coronary CTA charac-teristics of 75 culprit lesions in 75 patients with chest painand acute coronary syndromes or stable angina with thepresence of intra-luminal thrombus assessed on opticalcoherence tomography or angioscopy and integrity of thefibrous cap on optical coherence tomography. On coronaryCTA, low attenuation plaques were more frequentlyobserved in acute coronary syndromes with ruptured fibrouscaps than with intact fibrous caps (88% vs. 58%). Positiveremodeling was more frequent in lesions with rupturedfibrous caps than with intact fibrous caps (96% vs. 34%).However, coronary CTA could not clearly distinguishbetween lesions with intact fibrous caps that had thrombusand those that did not.

Watabe et al. (41) studied 107 patients scheduled toundergo elective PCI for stable angina with coronary CTA

to predict post-procedural cardiac troponin T elevations(cTnT). In patients who had cTnT elevation �3� the upperlimit of normal, compared with patients who did not havesignificant post-procedural cTnT elevations (Fig. 10), theattenuation value of the culprit plaque was significantly lower(43.0 [26.5 to 75.7] Hounsfield units vs. 94.0 [65.0 to109.0] Hounsfield units), the remodeling index was signif-icantly greater (1.20 � 0.18 vs. 1.04 � 0.15), and spottycalcification was significantly more frequent (50% vs. 11%)(p < 0.001 for all comparisons) (Fig. 10). Positive remod-eling and spotty calcification were independent predictorsfor cTnT elevation in multivariate analysis. Plaque attenu-ation value, remodeling index, and spotty calcification takentogether had a positive predictive value of 94% and an NPVof 90% for post-procedural cTnT elevation �3� the upperlimit of normal.

Appropriateness

Shah et al. (42) used an administrative database to study theincidence of stress imaging (SPECT and echocardiography)exams within 1 month of a cardiac outpatient visit afterrevascularization. The cumulative incidence was 12.2%within 30 days. The adjusted ORs among patients treatedby physicians who billed for both technical and profes-sional fees, compared with patients treated by physicianswho did not bill for either, was 2.3 (95% CI: 1.8 to 2.9)for SPECT and 12.8 (95% CI: 7.6 to 21.6) for echocardi-ography. These data strongly suggest that financial incen-tives influence the decision to perform stress imaging aftercoronary revascularization.

The American College of Radiology published appro-priateness criteria for patients with chronic chest pain anda high probability of CAD (43).

Conclusions

We hope that this review will inspire readers to examine theindividual publications in more detail.

Reprint requests and correspondence: Dr. Raymond J. Gibbons,Mayo Clinic, Gonda 6, 200 First Street Southwest, Rochester,Minnesota 55905. E-mail: [email protected].

REFERENCES

1. Dweck MR, Chow MW, Joshi NV, et al. Coronary arterial 18F-sodium fluoride uptake: a novel marker of plaque biology. J Am CollCardiol 2012;59:1539–48.

2. Dweck MR, Jones C, Joshi NV, et al. Assessment of valvular calcifi-cation and inflammation by positron emission tomography in patientswith aortic stenosis. Circulation 2012;125:76–86.

3. Aikawa E, Otto CM. Look more closely at the valve: imaging calcificaortic valve disease. Circulation 2012;125:9–11.

4. Ghafourian K, Younes D, Simprini LA, Weigold WG, Weissman G,Taylor AJ. Scout view X-ray attenuation versus weight-based selectionof reduced peak tube voltage in cardiac CT angiography. J Am CollCardiol Img 2012;5:589–95.

mailto:[email protected]


2482

5. Hausleiter J, Meyer TS, Martuscelli E, et al. Image quality andradiation exposure with prospectively ECG-triggered axial scanningfor coronary CT angiography: the multicenter, multivendor,randomized PROTECTION-III study. J Am Coll Cardiol Img 2012;5:484–93.

6. Neefjes LA, Dharampal AS, Rossi A, et al. Image quality and radiationexposure using different low-dose scan protocols in dual-source CTcoronary angiography: randomized study. Radiology 2011;261:779–86.

7. Cheng VY, Berman DS, Rozanski A, et al. Performance of thetraditional age, sex, and angina typicality-based approach for estimatingpretest probability of angiographically significant coronary artery diseasein patients undergoing coronary computed tomographic angiography:results from the multinational coronary CT angiography evaluation forclinical outcomes: an international multicenter registry (CONFIRM).Circulation 2011;124:2423–32, 1–8.

8. Diamond GA. Right answer, wrong question: on the clinical relevanceof the cardiovascular history. Circulation 2011;124:2377.

9. Arbab-Zadeh A, Miller JM, Rochitte CE, et al. Diagnostic accuracy ofcomputed tomography coronary angiography according to pre-testprobability of coronary artery disease and severity of coronary arterialcalcification. The CORE-64 (Coronary Artery Evaluation Using64-Row Multidetector Computed Tomography Angiography) Inter-national Multicenter Study. J Am Coll Cardiol 2012;59:379–87.

10. Vanhecke TE, Madder RD, Weber JE, Bielak LF, Peyser PA,Chinnaiyan KM. Development and validation of a predictive screeningtool for uninterpretable coronary CT angiography results. Circ Car-diovasc Imaging 2011;4:490–7.

11. Ko BS, Cameron JD, Meredith IT, et al. Computed tomography stressmyocardial perfusion imaging in patients considered for revascularization:a comparison with fractional flow reserve. Eur Heart J 2012;33:67–77.

12. Koo BK, Erglis A, Doh JH, et al. Diagnosis of ischemia-causingcoronary stenoses by noninvasive fractional flow reserve computedfrom coronary computed tomographic angiograms. Results from theprospective multicenter DISCOVER-FLOW (Diagnosis of Ischemia-Causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve)study. J Am Coll Cardiol 2011;58:1989–97.

13. Greenwood JP, Maredia N, Younger JF, et al. Cardiovascular magneticresonance and single-photon emission computed tomography fordiagnosis of coronary heart disease (CE-MARC): a prospective trial.Lancet 2012;379:453–60.

14. Jaarsma C, Leiner T, Bekkers SC, et al. Diagnostic performance ofnoninvasive myocardial perfusion imaging using single-photon emis-sion computed tomography, cardiac magnetic resonance, and positronemission tomography imaging for the detection of obstructive coronaryartery disease: a meta-analysis. J Am Coll Cardiol 2012;59:1719–28.

15. Schelbert EB, Cao JJ, Sigurdsson S, et al. Prevalence and prognosis ofunrecognized myocardial infarction determined by cardiac magneticresonance in older adults. JAMA 2012;308:890–7.

16. Dall’Armellina E, Piechnik SK, Ferreira VM, et al. Cardiovascularmagnetic resonance by non contrast T1-mapping allows assessment ofseverity of injury in acute myocardial infarction. J Cardiovasc MagnReson 2012;14:15.

17. Murthy VL, Naya M, Foster CR, et al. Improved cardiac risk assess-ment with noninvasive measures of coronary flow reserve clinicalperspective. Circulation 2011;124:2215–24.

18. Blaha MJ, Budoff MJ, Blumenthal RS, Nasir K. Coronary arterycalcium for guiding statin treatment Reply. Lancet 2012;379:312–3.

19. Min JK, Dunning A, Lin FY, et al. Age- and sex-related differences inall-cause mortality risk based on coronary computed tomographyangiography findings results from the International MulticenterCONFIRM (Coronary CT Angiography Evaluation for ClinicalOutcomes: An International Multicenter Registry) of 23,854 patientswithout known coronary artery disease. J Am Coll Cardiol 2011;58:849–60.

20. Masci PG, Francone M, Desmet W, et al. Right ventricular ischemicinjury in patients with acute ST-segment elevation myocardial infarc-tion: characterization with cardiovascular magnetic resonance. Circu-lation 2010;122:1405–12.

21. Grothoff M, Elpert C, Hoffmann J, et al. Right ventricular injury inST-elevation myocardial infarction: risk stratification by visualization ofwall motion, edema, and delayed-enhancement cardiac magneticresonance. Circ Cardiovasc Imaging 2012;5:60–8.

22. Shaw LJ, Mieres JH, Hendel RH, et al. Comparative effectiveness ofexercise electrocardiography with or without myocardial perfusion

single photon emission computed tomography in women with sus-pected coronary artery disease: results from the What Is the OptimalMethod for Ischemia Evaluation in Women (WOMEN) Trial.Circulation 2011;124:1239–49.

23. Shaw LJ, Weintraub WS, Maron DJ, et al. Baseline stress myocardialperfusion imaging results and outcomes in patients with stableischemic heart disease randomized to optimal medical therapy with orwithout percutaneous coronary intervention. Am Heart J 2012;164:243–50.

24. Hachamovitch R, Nutter B, Hlatky MA, et al. Patient managementafter noninvasive cardiac imaging. J Am Coll Cardiol 2012;59:462–74.

25. Tandon V, Hall D, Yeung Y, et al. Rates of downstream invasivecoronary angiography and revascularization: computed tomographiccoronary angiography vs. Tc-99m single photon emission computedtomography. European Heart J 2012;33:776–82.

26. Pazhenkottil AP, Nkoulou RN, Ghadri JR, et al. Impact of cardiachybrid single-photon emission computed tomography/computedtomography imaging on choice of treatment strategy in coronary arterydisease. Eur Heart J 2011;32:2824–9.

27. Shreibati JB, Cowper PA, O’Brien SM, et al. Association of coronaryCT angiography or stress testing with subsequent utilization andspending among Medicare beneficiaries. JAMA 2011;306:2128–36.

28. van der Veen BJ, Younis IA, Ajmone-Marsan N, et al. Ventriculardyssynchrony assessed by gated myocardial perfusion SPECT usinga geometrical approach: a feasibility study. Eur J Nucl Med MolImaging 2012;39:421–9.

29. Gibbons RJ. Chest pain triage in the ED: is CT coronary angiographythe answer? J Nucl Cardiol 2012;19:404–6.

30. Prasad V, Cheung M, Cifu A. Chest pain in the emergency depart-ment. Arch Intern Med 2012;172:1506–9.

31. Sarrazin JF, Philippon F, Tessier M, et al. Usefulness of fluorine-18positron emission tomography/computed tomography for identifica-tion of cardiovascular implantable electronic device infections. J AmColl Cardiol 2012;59:1616–25.

32. Tamborini G, Fusini L, Gripari P, et al. Feasibility and accuracy of3DTEE versus CT for the evaluation of aortic valve annulus to leftmain ostium distance before transcatheter aortic valve implantation.J Am Coll Cardiol Img 2012;5:579–88.

33. Jabbour A, Ismail TF, Moat N, et al. Multimodality imaging intranscatheter aortic valve implantation and post-procedural aorticregurgitation: comparison among cardiovascular magnetic resonance,cardiac computed tomography, and echocardiography. J Am CollCardiol 2011;58:2165–73.

34. Rudolph A, Abdel-Aty H, Bohl S, et al. Noninvasive detection offibrosis applying contrast-enhanced cardiac magnetic resonance indifferent forms of left ventricular hypertrophy relation to remodeling.J Am Coll Cardiol 2009;53:284–91.

35. Dweck MR, Joshi S, Murigu T, et al. Midwall fibrosis is an inde-pendent predictor of mortality in patients with aortic stenosis. J AmColl Cardiol 2011;58:1271–9.

36. Gersh BJ, Maron BJ, Bonow RO, et al. 2011 ACCF/AHA guidelinefor the diagnosis and treatment of hypertrophic cardiomyopathy:a report of the American College of Cardiology Foundation/AmericanHeart Association Task Force on Practice Guidelines. Developed incollaboration with the American Association for Thoracic Surgery,American Society of Echocardiography, American Society of NuclearCardiology, Heart Failure Society of America, Heart Rhythm Society,Society for Cardiovascular Angiography and Interventions, and Societyof Thoracic Surgeons. J Am Coll Cardiol 2011;58:e212–60.

37. Green JJ, Berger JS, Kramer CM, Salerno M. Prognostic value of lategadolinium enhancement in clinical outcomes for hypertrophiccardiomyopathy. J Am Coll Cardiol Cardiovasc Imaging 2012;5:370–7.

38. Leyva F, Taylor RJ, Foley PW, et al. Left ventricular midwall fibrosis asa predictor of mortality and morbidity after cardiac resynchronizationtherapy in patients with nonischemic cardiomyopathy. J Am CollCardiol 2012;60:1659–67.

39. Wong TC, Piehler K, Meier CG, et al. Association between extra-cellular matrix expansion quantified by cardiovascular magnetic reso-nance and short-term mortality. Circulation 2012;126:1206–16.

40. Ozaki Y, Okumura M, Ismail TF, et al. Coronary CT angiographiccharacteristics of culprit lesions in acute coronary syndromes not relatedto plaque rupture as defined by optical coherence tomography andangioscopy. Eur Heart J 2011;32:2814–23.


2483

41. Watabe H, Sato A, Akiyama D, et al. Impact of coronary plaquecomposition on cardiac troponin elevation after percutaneous coronaryintervention in stable angina pectoris: a computed tomography analysis.J Am Coll Cardiol 2012;59:1881–8.

42. Shah BR, Cowper PA, O’Brien SM, et al. Association betweenphysician billing and cardiac stress testing patterns following coronaryrevascularization. JAMA 2011;306:1993–2000.

43. Earls JP, White RD, Woodard PK, et al. ACR Appropriateness criteriachronic chest paindhigh probability of coronary artery disease. J AmColl Radiol 2011;8:679–86.

44. Andreini D, Pontone G, Bartorelli AL, et al. High diagnostic accuracyof prospective ECG-gating 64-slice computed tomography coronaryangiography for the detection of in-stent restenosis: in-stent restenosisassessment by low-dose MDCT. Eur Radiol 2011;21:1430–8.

45. Min JK, Dunning A, Lin FY, et al. Age- and sex-related differences inall-cause mortality risk based on coronary computed tomography angi-ography findings results from the InternationalMulticenter CONFIRM(Coronary CT Angiography Evaluation for Clinical Outcomes: AnInternational Multicenter Registry) of 23,854 patients without knowncoronary artery disease. J Am Coll Cardiol 2011;58:849–60.

46. Chow BJ, Small G, Yam Y, et al. Incremental prognostic value ofcardiac computed tomography in coronary artery disease usingCONFIRM: COroNary computed tomography angiography evalua-tion for clinical outcomes: an InteRnational Multicenter registry. CircCardiovasc Imaging 2011;4:463–72.

47. Villines TC, Hulten EA, Shaw LJ, et al. Prevalence and severity ofcoronary artery disease and adverse events among symptomatic patientswith coronary artery calcification scores of zero undergoing coronarycomputed tomography angiography: results from the CONFIRM(Coronary CT Angiography Evaluation for Clinical Outcomes: AnInternationalMulticenter) registry. JAmCollCardiol 2011;58:2533–40.

48. Lin FY, Shaw LJ, Dunning AM, et al. Mortality risk in symptomaticpatients with nonobstructive coronary artery disease: a prospective2-center study of 2,583 patients undergoing 64-detector row coronarycomputed tomographic angiography. J Am Coll Cardiol 2011;58:510–9.

49. Yiu KH, de Graaf FR, Schuijf JD, et al. Age- and gender-specificdifferences in the prognostic value of CT coronary angiography.Heart 2012;98:232–7.

50. McEvoy JW, Blaha MJ, Nasir K, et al. Impact of coronary computedtomographic angiography results on patient and physician behavior ina low-risk population. Arch of Intern Med 2011;171:1260–8.

51. Farkouh ME, Smars PA, Reeder GS, et al. A clinical trial of a chest-pain observation unit for patients with unstable angina. N Engl J Med1998;339:1882–8.

52. Udelson JE, Bechansky JR, Ballin DS. Myocardial perfusion imagingfor evaluation and triage of patients with suspected acute cardiacischemia: a randomized controlled trial. JAMA 2002;288:2693–700.

53. Goldstein JA, Gallagher MJ, O’Neill WW, Ross MA, O’Neil BJ,Raff GL. A randomized controlled trial of multi-slice coronarycomputed tomography for evaluation of acute chest pain. J Am CollCardiol 2007;49:863–71.

54. Goldstein JA, Chinnaiyan KM, Abidov A, et al. The CT-STAT(Coronary Computed Tomographic Angiography for SystematicTriage of Acute Chest Pain Patients to Treatment) trial. J Am CollCardiol 2011;58:1414–22.

55. Litt HI, Gatsonis C, Snyder B, et al. CT Angiography for safedischarge of patients with possible acute coronary syndromes. N Engl JMed 2012;366:1393–403.

56. Hoffmann U, Truong QA, Schoenfeld DA, et al. Coronary CTangiography versus standard evaluation in acute chest pain. N Engl JMed 2012;367:299–308.

57. Miller CD, Hwang W, Case D, et al. Stress CMR imaging observationunit in the emergency department reduces 1-year medical care costs inpatients with acute chest pain. J Am Coll Cardiol Img 2011;4:862–70.

58. Miller CD, Hoekstra JW, Lefebvre C, et al. Provider-directed imagingstress testing reduces health care expenditures in lower-risk chest painpatients presenting to the emergency department. Circ CardiovascImaging 2012;5:111–8.

59. Willson AB, Webb JG, Labounty TM, et al. 3-dimensional aorticannular assessment by multidetector computed tomography predictsmoderate or severe paravalvular regurgitation after transcatheter aorticvalve replacement: a multicenter retrospective analysis. J Am CollCardiol 2012;59:1287–94.

Key Words: cardiac magnetic resonance imaging - coronary CTangiography - radionuclide imaging.

the year in cardiac imaging · 2017-10-25 · mi cardiac mri: prevalence of unrecognized...

Documents