ww.sciencedirect.com

J o u r n a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0

Available online at w

ScienceDirect

journal homepage: www.JournalofCardiovascularCT.com

Original Research Article

Image quality and radiation exposure of coronaryCT angiography in patients after coronary arterybypass graft surgery: Influence of imaging directionwith 64-slice dual-source CT

Seul Ki Lee MDa, Jung Im Jung MD, PhDa,*, Jeong Min Ko MD, PhDb,Hae Giu Lee MD, PhDa

aDepartment of Radiology, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea,

222, Banpo-daero, Seocho-gu, Seoul, 137-701, South KoreabDepartment of Radiology, St. Vincent Hospital, College of Medicine, The Catholic University of Korea,

93, Jungbu-daero, Paldal-gu, Suwon, Gyeonggi-do, 442-723, South Korea

a r t i c l e i n f o

Article history:

Received 3 August 2013

Received in revised form

23 December 2013

Accepted 26 December 2013

Keywords:

Coronary CT

CABG

Scan direction

Image quality

Radiation dose

Conflict of interest: The authors report no c* Corresponding author.E-mail address: jijung@catholic.ac.kr (J.I.

1934-5925/$ e see front matter ª 2014 Sociehttp://dx.doi.org/10.1016/j.jcct.2013.12.011

a b s t r a c t

Background: The evaluation of native coronary arteries (NCAs) as well as coronary artery

bypass graft (CABG) patency after surgery is essential. However, NCAs are often blurred in

the craniocaudal scan direction because of long scan time with 64-slice CT.

Objective: The purpose of the study was to determine the effect of scan direction on image

quality and radiation exposure in assessment of NCAs and CABGs.

Methods: Retrospective analysis of 191 consecutive individuals undergoing coronary CT

angiography to evaluate CABG patency using 64-slice dual source CT. A retrospectively ECG

gated spiral acquisition protocol with ECG based tube current modulation and automatic

adjustment of tube current to a reference of 320 mAs (“CareDose 4D”) was used. Tube

current was 120 kVp. Scan direction was either cranio-caudal (CRC, n ¼ 98) or caudo-cranial

(CRC, n ¼ 93) and the scan volume covered the entire course of all bypass grafts. Inde-

pendent investigators determined quantitative image quality of the coronary arteries by

evaluating contrast-to-noise ratio (CNR), radiation exposure by comparing the effective

dose, and qualitative image quality through a 5 point rating scale.

Results: Quantitative image quality was not significantly different for the two groups except

for the CNR of the right coronary artery which was significantly higher in patients with

caudio-cranial scan direction (P ¼ .0007). The qualitative image quality of the CaC group

also was better for both NCAs and CABGs (P ¼ .002 for NCAs and <.001 for CABGs), mostly

because of the lower frequency of respiration artifacts on coronary arteries of the CaC

group (P ¼ .005). As an effect of automatic tube current adjustment, radiation dose was

lower in patients with caudo-cranial scan direction (6.8 mSv vs. 9.6 mSv, p < 0.0001).

onflict of interest.

Jung).ty of Cardiovascular Computed Tomography. All rights reserved.

J o u rn a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0 125

Conclusion: In patients with coronary bypass grafts imaged by 64-slice dual source CT with

spiral acquisition and automated tube current adjustment, a caudo-cranial scan direction

results in improved image quality and reduced radiation exposure.

ª 2014 Society of Cardiovascular Computed Tomography. All rights reserved.

1. Introduction patients were scanned craniocaudally (CrC group), and 93

After coronary artery bypass surgery, graft failure is common,

with 10% of grafts becoming occluded shortly after surgery

and 59% of venous grafts and 27% of arterial grafts occluded at

10 years after surgery. At 6 years after surgery, angina reap-

pears in 50% of patients. Symptoms can be the result of graft

failure, but in most patients the cause is disease progression

in native coronary arteries (NCAs).1,2 Hence, the evaluation of

NCAs as well as coronary artery bypass grafts (CABGs) is

important in following patients with CABGs.

Multidetector row CT allows noninvasive evaluation of

CABG patency. Several studies have reported high diagnostic

accuracy of CT for the detection of graft occlusion and ste-

nosis when using 64-slice CT. However, evaluation of the

NCAs can sometimes be difficult for various reasons, such as

motion artifacts, advanced atherosclerosis within small and

tortuous vessels, stents, and calcifications, especially in pa-

tientswith CABGs.3 Among these reasons,motion artifacts are

the main cause of inappropriate NCA evaluation because pa-

tients with cardiovascular diseases are generally older, and

coronary CT angiography (CTA)4 for CABGs requires longer

durations of breathholding.2

The duration of breathholding is proportional to the length

of the scanned area. In patients with internal mammary ar-

tery (IMA) grafts, the scanned area between the IMA origin and

the cardiac apex is longer than that of a routine coronary CTA,

which is between the aortic root and the cardiac apex.

Consequently, patients with CABGs have a longer duration of

breathholding. When scanning in the craniocaudal direction,

the patient may have difficulty with the long duration of the

breathhold at the end of the scan. Because run-off vessels of

NCAs and CABGs are located near the diaphragm, the rela-

tionship between breathholding and vessels may lead to

respiration artifacts that may result in improper assessment

of the run-off vessels.5 Hence, we hypothesized that scanning

in the caudocranial direction could prevent the respiration

artifacts at the end of the breathhold. The purpose of the study

was to determine the effect of scan direction on image quality

of NCAs and CABGs and on radiation exposure in 64-slice

dual-source CT.

2. Methods

2.1. Study population

Our institutional review board approved this retrospective

study and waived the requirement to obtain informed con-

sent. We retrospectively evaluated 191 consecutive patients

with CABG who underwent coronary CTA to assess CABG

patency between April 2009 and December 2010. Ninety-eight

patients were scanned caudocranially (CaC group). Standard

exclusion criteria for contrast-enhancedmultidetector row CT

coronary angiography were applied (previous allergic reaction

to iodinated contrast, atrial fibrillation or other arrhythmias,

renal disease with creatinine serum level >1.5 mg/mL). Most

CT scans were performed for immediate postoperative

exploration, and a few CT scans were performed for clinical

suspicion of vascular insufficiency of CABGs or NCAs. No pa-

tient was excluded.

2.2. Scan protocols

Coronary CTA scans were performed with a dual-source CT

system (Somatom Definition; Siemens Healthcare, For-

chheim, Germany). The parameters were as follows: slice

collimation of 2 � 32 � 0.6 mm by means of a z-flying focal

spot, gantry rotation time of 330 milliseconds, pitch of 0.2 to

0.5, fixed tube voltage of 120 kVp, and reference tube current

of 320 mAs. In each patient, 80 mL of iopromide (Ultravist 370;

370 mg/mL; Bayer Schering Pharma, Berlin, Germany) was

injected at a flow rate of 5 mL/s, followed by 50 mL of contrast

mixture (15% contrast medium and 85% saline solution) at the

same rate. Contrastmaterial administrationwas controlled by

bolus tracking in the ascending aorta (AAO; signal attenuation

threshold, 120 HU). The scan delay was 9 seconds. The patient

maintained an end-inspiration breathhold. In the absence of

contraindications, patients with a heart rate > 80 beats/min

received an intravenous selective b1-blocker, esmolol (Brevi-

bloc; Jeil Phama Co, Ltd, Seoul, Korea) 1 hour before the scan,

and a 0.3-mg sublingual dose of nitroglycerin was adminis-

tered just before the scan. Scans were performed with

electrocardiogram-controlled tube current modulation with

combined automatic exposure control (AEC; CareDose 4D;

Siemens Healthcare).6,7 Scanning was conducted in either the

craniocaudal or the caudocranial direction that covered from

the clavicle to the diaphragm. Imageswere reconstructedwith

a slice thickness of 0.6 mm, a reconstruction increment of

0.5 mm, a medium soft-tissue convolution kernel (B26 F), and

reconstructed matrix size of 512 � 512. The field of view was

manually adjusted to cover the whole heart and CABGs. All

images were transferred to a separate workstation equipped

with the image processing software (Advantage Window 4.3;

GE Healthcare, Milwaukee, WI; Syngo Multimodality Work-

place, version 2008; Siemens Healthcare).

2.3. Image analysis

Coronary CTA image analysis was performed by consensus

of 2 observers. First, 3-dimensional volume-rendered re-

constructions were used to obtain general information about

the anatomy of NCAs and CABGs. Subsequently, axial slices

Fig. 2 e Image noise. A round ROI was placed in the

ascending aorta (arrow). The ROI was defined as large as

possible. The image noise was defined as the standard

deviation of the attenuation value (in this data: 27.73). ROI,

region of interest.

Fig. 1 e Vessel contrast. One round ROI was placed in the

proximal left main artery (arrow). The ROI was defined as

large as possible. Another round ROI was placed in the

adjacent perivascular tissue (arrowhead ). The vessel

contrast was calculated as the difference between the

mean density of the vessel lumen (in this data: 339) and

that of the adjacent perivascular tissue (in this data: L42).

The same procedure was used for the proximal right

coronary artery. ROI, region of interest.

J o u r n a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0126

were visually examined for measurements of image quality

parameters. The analysis was assisted by curved multiplanar

reconstructions. Vessels were evaluated, including NCAs and

CABGs. Among the NCAs, the left main artery (LMA) and the

right coronary artery (RCA) were assessed as separate vessels.

We performed a quantitative analysis. In the first step, the

vessel contrast, defined as the difference between the mean

density (MD) (in Hounsfield units, HUs) in the contrast-

enhanced vessel lumen and the MD in the adjacent peri-

vascular tissue was calculated. Densities were measured by

manually placing a round-shaped region of interest (ROI) in

the proximal RCA and the proximal LMA (Fig. 1). ROI mea-

surements were not made within the grafts because their

small size could have potentially created partial volume arti-

facts. The ROI was defined as large as possible, avoiding cal-

cifications, plaques, and stenoses. In the second step, image

noise was defined as the standard deviation (SD) of the

attenuation value in an ROI placed in the AAO (Fig. 2). The

contrast-to-noise ratio (CNR) were defined as the ratios of the

measurement values from the first 2 steps. The signal-to-

noise ratio (SNR) were calculated from the MD of the

contrast-filled left ventricle (LV) by placing a circular ROI of

60 mm2 divided by the SD of these pixel values.6,8

Image noise ¼ SDAAO

CNR ¼ �MDprox: coronary �MDadjacent tissue

��SDAAO

SNR ¼ MDLV=SDLV

Then, the qualitative image quality was rated with a 5-point

ranking scale as follows: a score of 5 ¼ excellent (absence of

motion artifacts or blurring and excellent vessel opacifica-

tion); a score of 4 ¼ good (minor motion artifacts or blurring

and good vessel opacification); a score of 3 ¼ acceptable (some

motion artifacts or blurring and fair vessel opacification); a

score of 2 ¼ suboptimal (marked motion artifacts or blurring

and poor vessel opacification); and a score of 1 ¼ non-

diagnostic (severe artifacts with discontinuity or doubling in

the course).9 Images with a score of 3 or higher were consid-

ered diagnostic. The qualitative assessment included both

NCAs and CABGs.

The presence of respiration artifacts was evaluated on

coronary arteries and thoracic structures. Coronary arteries

were examined whether duplicated or not in the mediastinal

window, and thoracic structures were examined whether

double-imaged or not in the lung window (Figs. 3 and 4).

2.4. Radiation dose

The dose-length product (DLP) was defined as the volume CT

dose index multiplied by scan length and was an indicator of

the integrated radiation dose of an entire CT scan. The esti-

mated effective dose was determined by multiplying the DLP

by a conversion factor, k (0.014 mSv � mGy�1 � cm�1), as

recommended by the American Association of Physicists in

Medicine (AAPM) Report NO. 96.10

2.5. Statistical analysis

The data were presented as the means � SD for continuous

variables and number (%) for categorical variables.

Fig. 3 e The presence of respiration artifacts is only on the thoracic structures. In the lung window (A), there is doubling of

the diaphragm (black arrow) and the left major fissure (double black arrowheads), which suggest the presence of respiration

artifacts on the thoracic structures. In the mediastinal window (B), the right coronary artery (white short arrow), left

ascending artery (white arrowhead ), and left circumflex artery (white long arrow) show distinct contours, which suggests the

respiration artifacts did not affect the coronary arteries.

J o u rn a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0 127

Continuous variables were analyzed with an unpaired, 2-

tailed t test, and categorical variables were analyzed with

the following methods: Pearson c2 test for the comparisons of

patient characteristics and scan parameters, Mann-Whitney

test for the comparisons of quantitative image quality and

radiation dose, and Fisher exact test for the comparisons of

qualitative image quality and frequency of respiration arti-

facts. Values were considered significant when the P value <

.05. Cohen k coefficient was calculated to examine the inter-

observer agreement for image quality assessment. Five levels

of k values were defined as follows: very poor reliability

(�0.20), poor reliability (0.21e0.40), fair reliability (0.41e0.60),

good reliability (0.61e0.80), and excellent reliability (>0.80).

3. Results

All 191 coronary CTA scans were completed without compli-

cations, and no scan had to be repeated because of insufficient

image quality. The mean age of the 191 patients (128 men, 63

women) included in the study was 64.8 � 9.5 years. The pa-

tient characteristics of all 191 patients were summarized in

Fig. 4 e The presence of respiration artifacts both on the thorac

(A), there is doubling of the diaphragm (black arrow), which sug

structures. In the mediastinal window (B), there is a duplicated

respiration artifacts affected the coronary arteries.

Table 1. The 2 groups showed no significant differences in

patient characteristics. Comparison of the details about scan

characteristics (ie, body mass index and scan length) with a

change in the scan directionwas summarized in Table 2. The 2

groups showed no significant differences for body mass index

and scan length (P ¼ .92 and .78, respectively).

3.1. Comparison of objective image quality

The quantitative image quality parameters of this study are

summarized in Table 3. The RCA CNR of the CaC group was

significantly higher (CaC group: 19.2 � 5.5; CrC group: 16.7 �5.2; P ¼ .0007). The LMA CNR and SNR in the CaC group were

higher, although no statistically significant differences were

found between the 2 groups (CaC group: 17.9 � 4.2 and 14.3 �4.1, respectively; CrC group: 17.5 � 5.6 and 13.2 � 3.4,

respectively).

3.2. Comparison of subjective image quality

The overall interobserver agreement for image quality scoring

was excellent for NCAs (k¼ 0.93) and good for CABGs (k¼ 0.73).

ic structures and the coronary arteries. In the lung window

gests the presence of respiration artifacts on the thoracic

right coronary artery (white arrow), which suggests the

Table 1 e Patient characteristics and comparisonbetween the 2 groups.

Allpatients

CrC group CaC group

No. of patients 191 98 93

Age, y, mean � SD 64.8 � 9.5 65.3 � 9.5 64.2 � 9.6

Time interval, d,

mean � SD

459.0 � 897 380.9 � 749.0 547.0 � 1037.3

Cardiovascular risk factors

Family history, n (%) 2 (1.0) 2 (1.8) 0 (0)

Active smoker, n (%) 26 (13.6) 15 (13.6) 11 (10.6)

Hypertension, n (%) 114 (59.6) 57 (51.8) 57 (55.3)

Dyslipidemia, n (%) 13 (6.8) 5 (4.5) 7 (6.7)

Diabetes mellitus,

n (%)

60 (31.4) 31 (28.1) 28 (27.1)

Grafts per patient

Single graft, n (%) 18 (9.4) 7 (7.1) 11 (11.8)

2 grafts, n (%) 72 (37.6) 38 (38.7) 34 (36.5)

3 grafts, n (%) 92 (48.1) 47 (47.9) 45 (48.3)

>3 grafts, n (%) 9 (4.7) 6 (6.1) 3 (3.2)

Arterial and venous,

n (%)

117 (61.2) 62 (63.2) 55 (59.1)

Only arterial grafts,

n (%)

68 (35.6) 33 (33.6) 35 (37.6)

Only venous grafts,

n (%)

6 (3.1) 3 (3.0) 3 (3.2)

CaC, caudocranial; CrC, craniocaudal.

(No statistically significant differences were related to all factors).

Table 3 e Comparison of objective image quality betweenthe 2 groups.

CrC group CaC group Pvalue

No. of patients 98 93

Image noise*, HU, mean � SD 23.6 � 6.0 26.5 � 6.2 .0004

Vessel contrast of the LMA,

HU, mean � SD

356.4 � 81.6 441.4 � 89.4 <.0001

Vessel contrast of the RCA,

HU, mean � SD

326.3 � 78.7 446.8 � 106.0 <.0001

CNR in the LMA, mean � SD 17.5 � 5.6 17.9 � 4.2 .26

CNR in the RCA, mean � SD 16.7 � 5.2 19.2 � 5.5 .0007

SNR, mean � SD 13.2 � 3.4 14.3 � 4.1 .08

CaC, caudocranial; CNR, contrast-to-noise ratio; CrC, craniocaudal;

LMA, left main artery; RCA, right coronary artery; SNR, signal-to-

noise ratio.

* Image noise ¼ standard deviation of aorta.

Table 4 e Comparison of subjective image quality scores

J o u r n a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0128

The image quality scores of the 2 groups are presented in

Table 4. None of the NCAs or CABGswere nondiagnostic (score

1 or 2) in the 2 groups. Overall, 72.5% and 91.4% of NCAs

received a score of 5 for the CrC group and the CaC group,

respectively. The CaC group had significantly better subjective

image quality of NCAs (P ¼ .002). For CABGs, 76.5% and 95.7%

received a score of 5 for the CrC group and the CaC group,

respectively. The CaC group had significantly better subjective

image quality of CABGs (P � .0001).

3.3. Comparison of respiration artifacts

Respiration artifacts on coronary arteries and thoracic

structures were analyzed separately. This analysis is

Table 2 e Comparison of scan characteristics andradiation dose between the 2 groups.

CrC group CaC group P value

No. of patients 98 93

Body mass index,

mean � SD

25.0 � 3.2 25.0 � 3.3 .92

Scan length, cm,

mean � SD

234.5 � 37.1 249.0 � 35.4 .78

Total DLP, mGy/cm,

mean � SD

692.1 � 331.2 488.7 � 285.1 <.0001

Effective dose*, mSv,

mean � SD

9.6 � 4.7 6.8 � 4.0 <.0001

CaC, caudocranial; CrC, craniocaudal; DLP, dose-length product.

* Effective dose ¼ DLP � 0.014.

summarized in Table 5. The frequency of respiration arti-

facts on thoracic structures did not differ significantly be-

tween the 2 groups (CaC group: n ¼ 41 [44.1%]; CrC group:

n ¼ 34 [34.7%]; P ¼ .236). The frequency of respiration arti-

facts on coronary arteries was significantly lower in the CaC

group (CaC group: n ¼ 4 [4.3%]; CrC group: n ¼ 17 [17.4%];

P ¼ .005).

3.4. Comparison of effective radiation dose

DLP and effective dose in the CrC group were significantly

greater (CaC group: 6.8 � 4.0 mSv; CrC group: 9.6 � 4.7 mSv;

P < .001; Table 2).

4. Discussion

In coronary CTA of patients with CABG, the scanned area is

longer than that of routine coronary CTA because the scanned

area includes the entire course of grafts. Consequently, the

breathholding time is longer during the scan for patients with

for NCAs and CABGs between the 2 groups.

CrC group CaC group

No. of patients 98 93

NCA (P ¼ .002)

Score 5, n (%) 71 (72.5) 85 (91.4)

Score 4, n (%) 12 (12.2) 5 (5.4)

Score 3, n (%) 15 (15.3) 3 (3.2)

Score 2, n (%) 0 (0) 0 (0)

Score 1, n (%) 0 (0) 0 (0)

CABG (P < .0001)

Score 5, n (%) 75 (76.5) 89 (95.7)

Score 4, n (%) 11 (11.2) 4 (4.3)

Score 3, n (%) 12 (12.3) 0 (0)

Score 2, n (%) 0 (0) 0 (0)

Score 1, n (%) 0 (0) 0 (0)

CABG, coronary artery bypass graft; CaC, caudocranial; CrC, cra-

niocaudal; NCA, native coronary artery.

Table 5eComparison of frequency of respiration artifactsfor thorax and coronary arteries between the 2 groups.

CrC group CaC group P value

No. of patients 98 93

Thorax, n (%) 34 (34.7) 41 (44.1) .24

Coronary arteries, n (%) 17 (17.4) 4 (4.3) .005

CaC, caudocranial; CrC, craniocaudal.

J o u rn a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0 129

CABG. When CABGs are scanned craniocaudally, the required

time of breathholding may be too long and thus challenge

patients’ limitations toward the end of the scan, especially in

older cardiovascular patients. This tendency may lead to

respiration artifacts at the end of breathholding associated

with diaphragmatic movement. In the related literature,

nearly all studies on coronary CTA of patients with CABG have

been performed in the craniocaudal direction; no study

compared the effect of scan directions on image quality and

radiation exposure.We hypothesized that if the scan direction

is reversed from the heart base to the IMA origin, the distal

CABGs, anastomoses, and NCAs can be scanned first with

fewer respiration artifacts, and the proximal CABGs (specially

IMA grafts), which are less prone to diaphragmaticmotion and

have a larger diameter than NCAs, can be examined suc-

cessfully.5 The results of this study support our hypothesis.

We found that the frequency of respiration artifacts on

coronary arteries was significantly lower in the CaC group.

This finding suggests that image quality scores of NCAs and

CABGs were significantly better in the CaC group. By contrast,

the frequency of respiration artifacts on thoracic structures

was higher in the CaC group, although a significant difference

was not observed. In the CaC group, most respiration artifacts

on the thoracic structures were detected near the diaphragm

at the initiation of the scan. This finding indicated the possi-

bilities of incomplete suppression of end-inspiration breath-

holding at the time of scan initiation or involuntary

diaphragmatic motion, which caused the respiration artifacts

in the lower thorax. Hence, this artifact in the CaC group could

be controlled with patient education on proper end-

inspiration breathholding.

In this study, we found that the LMA CNR and SNR were

higher for the CaC group, although no significant differences

were observed. Moreover, RCA CNR was significantly higher

for the CaC group. The contrast enhancement in the AAO and

coronary arteries were significant predictors of the CNR. Bet-

ter contrast enhancement can be achieved by using a higher

concentration of contrast agents.11 With the use of bolus

tracking, the coronary arteries located lower (ie, RCA) are

scanned earlier in the CaC group. This allowed higher

concentrated contrast agents in coronary arteries with

optimal contrast enhancement.12

Radiation exposure is a main concern of coronary CTA. In

our study, the CaC group significantly reduced radiation dose

by up to 70% compared with the CrC group. In the AEC, 2

attenuation profiles in the anteroposterior and lateral di-

rections are calculated from a single anteroposterior tomo-

gram scan and are combined to an averaged attenuation

profile that reflects patient size and body shape. The averaged

attenuation profile algorithm determines a mean adapted

tube current level that is kept constant during the scan.7 The

differences in the scan direction may cause differences in

attenuation of the introducing body part where AEC starts,

which is the upper abdomen for the CaC group and the

shoulder in the CrC group. The upper abdomen is a less-

attenuating body part than the shoulder. Therefore, when

the scan begins with the less-attenuating body part, tube

current will be constantly lower than that obtained with the

opposite direction through AEC.13 When the tube current in-

creases according to the high-attenuation body part in the CaC

group, a threshold may be reached, and then the penetrating

strength might be insufficient to generate tube current mod-

ulation in the heaviest body part, such as the shoulder. This

limited function provided the underpenetrated scanned

image with a lower SNR but without damaging the image’s

diagnostic value.14 These consequences may result in signifi-

cant differences of radiation dose according to the scan di-

rection. The reduced radiation exposure was the most

important unexpected plus of the reversed scan direction.

Our study has some limitations. First, our study was a

nonrandomized retrospective study, and data were obtained

from a single institution. However, to avoid bias in patient

selection, patients were enrolled consecutively during a spe-

cific period of time, and the image analysis was performed

blindlywithout patient information. Second,we assessed only

respiration artifacts, although other sources of nonrespiration

artifacts may be possible in a coronary CTA scan. Third, a

comparison of diagnostic accuracy between the 2 groups was

not covered in our study because only a small number of pa-

tients underwent supplemental coronary angiography. Inevi-

tably, quantitative and qualitative image qualities were

substituted for diagnostic accuracy. Fourth, reduction of the

radiation dose in our study mainly depended on AEC (ie,

Siemens CareDose 4D), and this could limit the generaliz-

ability of our strategy because different machines use

different methods for adapting the tube current to patient

attenuation. Finally, the conventional 64-slice dual-source CT

was used in our study despite the recent introduction of a new

generation of CT scans. The scan time will no longer be a

major concern with the development of CT scans. However

the conventional 64-slice CT is currently themost widely used

machine in clinical settings. Furthermore, the benefit of

reduced radiation exposure with the caudocranial direction

can be applied to the more-advanced generation of CT scans

in the same manner used in our study.

5. Conclusion

The use of the caudocranial direction in coronary CTA to

assess NCAs and CABGs is a feasiblemethod to improve image

quality while also reducing radiation exposure.

r e f e r e n c e s

1. Alderman EL, Kip KE, Whitlow PL, et al. Native coronarydisease progression exceeds failed revascularization as causeof angina after five years in the Bypass Angioplasty

J o u r n a l o f C a r d i o v a s c u l a r C om p u t e d T omog r a p h y 8 ( 2 0 1 4 ) 1 2 4e1 3 0130

Revascularization Investigation (BARI). J Am Coll Cardiol.2004;44:766e774.

2. Malagutti P, Nieman K, Meijboom WB, et al. Use of 64-slice CTin symptomatic patients after coronary bypass surgery:evaluation of grafts and coronary arteries. Eur Heart J.2007;28:1879e1885.

3. Goetti R, Leschka S, Baumuller S, et al. Low dose high-pitchspiral acquisition 128-slice dual-source computedtomography for the evaluation of coronary artery bypass graftpatency. Invest Radiol. 2010;45:324e330.

4. Weigold WG, Abbara S, Achenbach S, et al. Standardizedmedical terminology for cardiac computed tomography: areport of the Society of Cardiovascular computedTomography. J Cardiovasc Comput Tomogr. 2011;5:136e144.

5. Hazirolan T, Turkbey B, Karcaaltincaba M, et al. Does 16-MDCT angiography scanning direction affect image quality ofcoronary artery bypass grafts and the native coronaryarteries? Eur Radiol. 2007;17:97e102.

6. Park YJ, Kim YJ, Lee JW, et al. Automatic tube potentialselection with tube current modulation (APSCM) in coronaryCT angiography: comparison of image quality and radiationdose with conventional body mass index-based protocol.J Cardiovasc Comput Tomogr. 2012;6:184e190.

7. Hermann F, Martinoff S, Meyer T, et al. Reduction of radiationdose estimates in cardiac 64-slice CT angiography in patientsafter coronary artery bypass graft surgery. Invest Radiol.2008;43:253e260.

8. Lee YW, Yang CC, Mok GS, Law WY, Su CT, Wu TH.Prospectively versus retrospectively ECG-gated 256-slice CTangiography to assess coronary artery bypassgraftsecomparison of image quality and radiation dose. PloSOne. 2012;7:e49212.

9. Tatsugami F, Matsuki M, Nakai G, et al. The effect of adaptiveiterative dose reduction on image quality in 320-detector rowCT coronary angiography. Br J Radiol. 2012;85:e378ee382.

10. McCollough C, Cody D, Edyvean S. The measurement,reporting, and management of radiation dose in ct: AAPM(American Association of Physicists in Medicine) report no 96.2008.

11. Ferencik M, Nomura CH, Maurovich-Horvat P, et al.Quantitative parameters of image quality in 64-slicecomputed tomography angiography of the coronary arteries.Eur J Radiol. 2006;57:373e379.

12. Weininger M, Barraza JM, Kemper CA, Kalafut JF, Costello P,Schoepf UJ. Cardiothoracic CT angiography: current contrastmedium delivery strategies. AJR Am J Roentgenol.2011;196:W260eW272.

13. Goo HW, Suh DS. Tube current reduction in pediatric non-ECG-gated heart CT by combined tube current modulation.Pediatr Radiol. 2006;36:344e351.

14. Lesser JR, Flygenring BJ, Knickelbine T, Longe T,Schwartz RS. Practical approaches to overcoming artifactsin coronary CT angiography. J Cardiovasc Comput Tomogr.2009;3:4e15.