7/28/2019 Ajr 2012 Pulm.bta 30ml

1/5

AJR:199 , December 2012 1247

We have developed a technique to perorm a

pulmonary CTA study on a 64-MDCT scan-

ner with only 30 mL o IV contrast agent. The

smaller amount o contrast material may allow

pulmonary CTA to be perormed in patients

with compromised renal unction. Our hypoth-

esis was that contrast dose reduction to 30 mL

is easible using a 64-MDCT scanner without

compromising image quality in a selected pa-

tient population. This hypothesis was tested us-ing this new low-contrast-dose pulmonary CTA

technique in a population o patients deemed to

be at a higher risk o having incidental pulmo-

nary emboli because o possible recurrent or

underlying malignancy. In our study, thereore,

we retrospectively compared images obtained

using a regular dose (100 mL) and a low dose

(30 mL) o IV contrast material to establish that

image quality using the 30-mL protocol was

not inerior to that using the 100-mL protocol.

Pulmonary 64-MDCT Angiography

With 30 mL of IV ContrastMaterial: Vascular Enhancementand Image Quality

Carol C. Wu1

Edward W. Lee2

Robert D. Suh3

Barton S. Levine4

Bruce M. Barack5

Wu CC, Lee EW, Suh RD, Levine BS, Barack BM

1Department o Radiology, Division o Thoracic Imaging

and Interventions, Massachusetts General Hospital,

Boston, MA.

2Department o Radiology, David Geen School o

Medicine at The University o Caliornia Los Angeles,

Interventional Radiology, Los Angeles, CA .

3Department o Radiology, David Geen School o

Medicine at The University o Caliornia Los Angeles,

Thoracic Imaging, Los Angeles, CA.

4

Department o Medicine, Division o Nephrology, VAGreater Los Angeles Healthcare System, Los Angeles, CA.

5Department o Imaging, Thoracic Radiology Section, VA

Greater Los Angeles Healthcare System, 11301 Wilshire

Blvd, Los Angeles, CA 90 073. Address correspondence

to B. M. Barack (bruce.barack@va.gov).

Cardiopulmonary Imaging Original Research

AJR2012; 199:12471251

0361803X/12/19961247

American Roentgen Ray Society

Pulmonary CT angiography (CTA)

has become the diagnostic test o

choice or the evaluation o pul-

monary thromboembolic disease

over the past 10 years [13]. The current stan-

dard practice in most institutions is to perorm

the study using 80150 mL o IV contrast

agent [4]. The technical advances in both

MDCT scanners and power injectors have

made it possible to signicantly decrease theamount o IV contrast agent used to perorm

pulmonary CTA. With the 64-MDCT scanner,

1-mm contiguous helical axial images o the

chest can be obtained in less than 10 seconds.

IV contrast material can be inused at 5 mL/s

using a power injector and antecubital venous

access. Thus, a more concentrated, smaller

total volume o IV contrast agent can be de-

livered to opaciy the entire pulmonary arte-

rial system or the shorter scanning time.

Keywords: contrast dose, image quality, pulmonary

MDCT angiography, vascular enhancement

DOI:10.2214/AJR.12.8739

Received February 14, 2012; accepted ater revision

May 8, 2012.

OBJECTIVE. The objective o our study was to determine whether vascular enhance-

ment and image quality can be preserved in pulmonary CT angiography (CTA) perormed on

a 64-MDCT scanner with 30 mL o IV contrast material.

MATERIALS AND METHODS. This retrospective matched-cohort study compared

image quality o pulmonary CTA perormed using 30 mL o IV contrast material versus 100

mL o IV contrast material. CT images o 50 patients (46 men, our women; mean age, 66

years) who underwent pulmonary CTA on a 64-MDCT scanner using a low dose (30 mL)o iodixanol 320 and another 50 patients (49 men, one woman; mean age, 65 years) who un-

derwent pulmonary CTA using a regular dose (100 mL) o contrast material during the same

time period were selected or review. The 30- and 100-mL pulmonary CTA studies were ret-

rospectively evaluated by two thoracic radiologists in random order. Attenuation values were

recorded over the main, right main, selected lobar, segmental, and subsegmental pulmonary

arteries. Image quality was also subjectively assessed using visual scores on a scale rom 1

(nondiagnostic) to 5 (excellent).

RESULTS. The average attenuation measurements o the main, right main, selected lo-

bar, segmental, and subsegmental pulmonary arteries were 260, 262, 280, 316, and 338 HU,

respectively, on the 30-mL studies and 313, 301, 316, 344, and 349 HU, respectively, on the

100-mL studies. The average visual score was 4.0 or both the 30- and 100-mL groups. A vi-

sual score o 4 or 5 was given to 82% o studies in the 30-mL group and 78% o studies in

the 100-mL group.

CONCLUSION. Contrast agent dose or pulmonary CTA using a 64-MDCT scannercan be signicantly reduced without compromising diagnostic image quality.

Wu et al.Pulmonary CTA Using 30 mL o Contrast Material

Cardiopulmonary ImagingOriginal Research

7/28/2019 Ajr 2012 Pulm.bta 30ml

2/5

1248 AJR:199, December 2012

Wu et al.

Materials and Methods

Patients

The study was approved and a waiver o in-

ormed consent was granted by the institutional

review board or this retrospective, matched-co-

hort, HIPAA-compliant study. The CT scans and

medical records o 50 patients (Table 1) who un-

derwent low-contrast-dose pulmonary CTA rom

January 2007 to April 2010 were selected or re-

view. These patients were at risk or incidental

pulmonary emboli and had been reerred or con-

trast-enhanced CT or indications such as ollow-

up o resected lung cancer or other cancers and

ollow-up o parenchymal opacities and nodules

suspicious or primary or metastatic malignan-

cies. Another 50 CT studies rom a matched co-

hort o patients similar to the study group in age

and weight (Table 1) who had undergone regu-

lar-contrast-dose pulmonary CTA studies during

the same time period were selected as the control

group. Because no pulmonary emboli were de-

tected in the 30-mL group, regular-contrast-dose

pulmonary CTA studies negative or pulmonary

emboli were also selected. Patients with class III

or IV congestive heart ailure or supraventricular

tachycardia were also excluded rom both groups.

Medical records were retrospectively reviewed.

Patient sex, age, body weight, and BMI at the time

o scanning were recorded.

Contrast Material Administration

A 20-gauge IV catheter was placed in an ante-

cubital vein or contrast injection. A commercial-

ly available contrast agent (iodixanol [Visipaque

320, GE Healthcare Ireland]) was inused IV us-ing an automated power injector (Stellant Dual

Head Injector, Medrad Healthcare). For the low-

contrast-dose studies, 30 mL o Visipaque 320

was administered. For the regular-contrast-dose

studies, 100 mL o Visipaque 320 was adminis-

tered. The inusion rate was 5 mL/s or both the

30- and 100-mL studies. Thirty milliliters o nor-

mal saline was inused at 5 mL/s ater the inusion

o contrast material in both groups.

Image Acquisition

All studies were perormed using a 64-MDCT

scanner (Somatom Sensation 64, Siemens Health-

care). The mean CT dose index (CTDI) was 7.65

1.24 (SD) mGy or patients weighing 200 lb (90.9

kg) or less and 17.20 3.36 mGy or patients

weighing more than 200 lb (90.9 kg). The dose-

length product (DLP) was 284.28 46.93 mGy

cm or patients weighing 200 lb (90.9 kg) or less

and 654.00 75.03 mGy cm or patients weigh-

ing more than 200 lb (90.9 kg). The monitoring

scan time was 0.5 second and monitoring scans

were obtained at one per second. Scans were ob-

tained ater the IV administration o contrast ma-

terial. The scan was triggered at 75 HU over the

superior vena cava (SVC) at the level o the aor-

tic arch or the low-contrast-dose studies and at

120 HU over the main pulmonary artery (MPA)

or the 100-mL studies. All scans were obtained

in the craniocaudal direction at end inspiration.

The CTA examinations o both groups were

perormed using a proprietary algorithm (Com-

bined Application to Reduce Exposure [CARE]

Dose4D, Siemens Healthcare) and the ollowing

parameters: slice thickness, 0.6 mm; gantry rota-

tion time, 0.5 second; pitch, 1.0; and 180 mAse

.

The tube current setting was 100 kVp in patients

weighing 200 lb (90.9 kg) or less and 120 kVp in

patients weighing more than 200 lb (90.9 kg). Ad-

ditional 1.5- and 3-mm axial and 1.5-mm sagittal

and coronal reconstructions were obtained.

Image Analysis

Image analysis was retrospectively perormed

in random order by two specialized thoracic radi-

ologists with 3 and 36 years experience, respec-

tively, on a PACS workstation using proprietary

sotware (Impax ES, DS 5000, Aga). Both ob-

servers had access to all the axial source images,

coronal and sagittal reormation images, and stan-

dard window settings (width and level: sot-tissue,

400 and 40 HU, respectively; pulmonary emboli,

450 and 100 HU; lung, 1500 and 600 HU).

Vascular enhancement was evaluated quantita-

tively with attenuation values measured at the lev-

el o the MPA, right MPA, right upper lobar and

let lower lobar pulmonary arteries, and right up-

per lobe apical and let lower lobe posterior basalsegmental and subsegmental arteries.

Overall image quality was subjectively scored

on a 5-point scale as ollows: 1, nondiagnostic, no

diagnosis possible; 2, poor, inadequate or diagno-

sis o the presence or absence o a clot; 3, air, able

to exclude pulmonary emboli to the lobar level; 4,

good, optimal enhancement at least at the segmen-

tal artery level to allow condent diagnosis o the

presence or absence o a clot; or 5, excellent, opti-

mal enhancement at least at the subsegmental ar-

tery level to allow unambiguous diagnosis o the

presence or absence o a clot.

Statistical AnalysisStatistical analysis was perormed using com-

mercially available sotware (SPSS, version 15,

SPSS). Ap value o < 0.05 was considered to indi-

cate a statistically signicant dierence. The Wil-

coxon rank test was applied to determine wheth-

er pulmonary arterial enhancement at each level

and subjective scoring o image quality were sig-

nicantly dierent between the 30- and 100-mL

groups. The Spearman correlation coecient was

used to determine whether patient weight or BMI

has any signicant correlation to the image qual-

ity scores in both groups.

Results

The patient characteristics are summa-

rized in Table 1. There was no statistically

signicant dierence in the age, sex, andweight distributions o patients in the 100-

and 30-mL groups. The male preponderance

in both groups is a refection o the patient

population o our hospital.

Vascular enhancement values and subjec-

tive image quality scores are summarized in

Table 2. A visual score o 4 or 5 was given

or 82% (41/50) o the studies in the 30-mL

group and 78% (39/50) o the studies in the

100-mL group. The 100-mL group had two

more low-quality scans (both rated as 2) than

the 30-mL group. There was no signicant

dierence between the two groups in the en-

hancement o the let lower lobar, right upper,

and let lower lobe segmental and subsegmen-

tal branches. The average enhancement val-

ues o the 30-mL contrast group were lower

than those o the 100-mL group at each o the

pulmonary branches measured, as might be

expected; however, the average values or all

segments were greater than 211 HU, the theo-

retic minimal enhancement required or diag-

nosing pulmonary emboli [4].

There was no dierence between the two

groups in the subjective image quality scores.

The average score or both groups was 4.0,

corresponding to a diagnostic-quality studythat allows condent exclusion o segmental

pulmonary emboli. Neither weight nor BMI

showed a signicant correlation with the sub-

jective image quality scores in the 30-mL

group (correlation coecient = 0.114 and

0.173, respectively; p = 0.429 and 0.229).

However, both patient weight and BMI

showed a signicant correlation with the sub-

jective image quality scores in the 100-mL

group (correlation coecient = 0.434 and

0.361;p = 0.002 andp = 0.010).

No pulmonary emboli were detected in

the 30- or 100-mL group. There was no doc-

umented adverse reaction or complicationrom the CT examinations in either group.

Discussion

Pulmonary embolism has been recognized

as a signicant cause o morbidity and mor-

tality and has been estimated to cause 24,000

35,000 deaths per year in the United States

[5]. Pulmonary CTA has become the diagnos-

tic study o choice in the evaluation o patients

7/28/2019 Ajr 2012 Pulm.bta 30ml

3/5

AJR:199 , December 2012 1249

Pulmonary CTA Using 30 mL of Contrast Material

with suspected pulmonary emboli. Most re-

ported standard pulmonary CTA studies have

been perormed using 80150 mL o contrast

material [4, 6]. The theoretic minimum atten-

uation o blood required to see all acute and

chronic pulmonary venous thromboemboli

has been calculated to be 93 and 211 HU, re-

spectively [4]. Bae et al. [6] perormed a lin-

ear regression analysis and reported that to

achieve 250 HU o contrast enhancement in

the pulmonary arteries, approximately 1.2 mL/kg o 350 mg I/mL contrast material injected

at 4 mL/s is required. However, given the rap-

id evolution o CT technology and improve-

ment in acquisition speed, achievement o di-

agnostic-quality pulmonary CTA with a lower

dose o IV contrast material has become pos-

sible, as shown by the current study.

With a reduction in the injected IV contrast

dose to 30 mL and the inherent scanning pa-

rameters o the Somatom Sensation 64 scan-

ner, a change in the trigger position rom the

MPA to the SVC was required to obtain re-

producible low-contrast-dose pulmonary CTA

scans. Ater triggering the scan using bolus

tracking, the minimum possible time or the

table to move to the start position and the

scan to begin is 4 seconds. With the trigger

point placed over the MPA and the trigger at-

tenuation set to 75 HU or less, a signicant

portion o the 30-mL contrast bolus wouldhave already passed through the pulmonary

arterial system during this 4-second period.

It thus became apparent that with the contrast

dose reduction to 30 mL, the scan would have

to be triggered proximal to the MPA to make

eective use o the small contrast bolus. The

SVC was chosen as the trigger point because

it was easily recognized and changes in the at-

tenuation setting allowed use o the majority

o the eective contrast bolus previously lost

while the table was moving. A 75-HU trigger

over the SVC proved to yield reproducible di-

agnostic scans in most patients (Fig. 1).

We recognize that the choice o a trig-

ger point proximal to the MPA will result

in premature scan initiation in patients with

a prolonged transit time such as those withcongestive heart ailure or irregular supra-

ventricular tachycardia. In addition, it has

been reported that nondiagnostic pulmonary

CTA examinations tend to occur in patients

with severe congestive heart ailure and ar-

rhythmia even when using a regular dose o

contrast material [7]. Hence, these patients

were excluded rom the 30-mL pulmonary

CTA study group and, subsequently, also the

matched control group.

The enhancement characteristics based on

attenuation values o the pulmonary artery

branches o our 100-mL group are similar to

those values previously reported or pulmo-

nary CTA [8]. In our study, the average en-

hancement values o various pulmonary ar-

tery branches in the 30-mL group were above

the theoretic minimal diagnostic enhance-

ment o 211 HU and also above the 250-HU

threshold considered adequate by other au-

thors reporting a low-contrast-dose pulmo-

nary CTA technique [6, 9]. Furthermore,

subsegmental emboli have been detected in a

study with a central pulmonary artery atten-

uation o 180 HU, and lobar and segmental

emboli have been detected with central pul-

monary artery attenuation as low as 123 HUby other investigators also using a low-con-

trast-dose pulmonary CTA technique [10].

There was no signicant dierence between

the 100- and 30-mL groups in terms o the sub-

jective image quality score and enhancement

value o segmental and subsegmental pulmo-

nary artery branches. These ndings suggest

that pulmonary CTA o diagnostic quality to

detect acute and chronic pulmonary emboli can

be perormed with a signicantly lower dose

o contrast material than customarily used and

previously theorized.

The act that there was a statistically sig-

nicant correlation between weight and BMIand image quality score in the 100-mL group

but not in the 30-mL group implies that other

physiologic parameters such as cardiac out-

put and perhaps mean pulmonary artery pres-

sure become more important in determining

image quality as the volume o contrast in-

jected is reduced. Further studies are neces-

sary to conrm this hypothesis and to urther

elucidate such physiologic parameters.

TABLE 1: Patient Characteristics

CharacteristicsRegular-Contrast-Dosea Group

(n= 50)Low-Contrast-Doseb Group

(n= 50) p

Age (y) 0.602

Mean SD 65.0 12.7 66.2 10.9

Sex, no. o patients 0.359

M 49 46

F 1 4

Weight (kg) 0.950

Mean SD 85.0 26.3 85.3 19.5

BMI (kg/m2) 0.808

Mean SD 27.6 6.0 27.2 8.0

NoteBMI = body mass index.a100 mL.b30 mL.

TABLE 2: Vascular Enhancement and Image Quality in Regular-Contrast-Dose and Low-Contrast-Dose Groups

Pulmonary Artery Branches

Vascular Enhancement (HU), Mean SD

pRegular-Contrast-

Dosea Group (n= 50)Low-Contrast-Doseb

Group (n= 50)

Main 312.9 116.4 260.0 79.0 0.009

Right main 301.2 111.7 261.6 79.4 0.044

Right upper lobar 320.2 116.8 277.1 82.2 0.031

Right upper lobe apical segmental 349.6 126.9 318.6 98.4 0.175

Right upper lobe apical subsegmental 353.8 123.0 352.7 108.7 0.963

Let lower lobar 311.4 120.6 282.7 88.2 0.177

Let lower lobe posterior basal segmental 338.0 133.0 313.1 107.2 0.305

Let lower lobe posterior basal subsegmental 342.5 138.3 323.4 106.5 0.441

Image quality score 4.0 0.8 4.0 0.6 0.780a100 mL.b30 mL.

7/28/2019 Ajr 2012 Pulm.bta 30ml

4/5

1250 AJR:199, December 2012

Wu et al.

The average body weight o patients in the

30-mL group was 85.3 kg. Using the 1.2 mL/

kg estimation o Bae et al. [6], 100.2 mL o

contrast material would be the average dose

used in these patients. In addition, a contrast

material with lower iodine concentration, Vis-

ipaque 320, was used in this study compared

with 350 mg I/mL used in the study by Baeand colleagues. The ability to achieve ade-

quate enhancement using a smaller volume o

contrast material with lower iodine concentra-

tion may be attributable to both the higher rate

o contrast inusion (5 mL/s) and the 30-mL

saline fush used in our study. Both o these

actors were suggested by Bae et al. as addi-

tional mechanisms by which the injected con-

trast volume could be urther reduced.

There are several theoretic advantages to

a reduction o injected IV contrast material.

Although the true incidence and risk o con-

trast-induced kidney injury associated with

IV contrast material or CT remain contro-

versial [11], the volume o contrast material

has been suggested as a possible contributo-

ry actor [12, 13]. Thereore, by decreasing

the contrast volume, the risk o inducing

contrast nephropathy may be reduced. It has

been recommended that the lowest volume

and lowest dose o contrast agent needed to

achieve a diagnostic result should be used,

particularly in patients at risk o developing

contrast-induced acute kidney injury such as

patients with chronic kidney disease and old-

er patients [14]. Reducing the contrast vol-

ume can also reduce the amount o contrastmaterial that inltrates the tissue in the event

o extravasations and, hence, decrease the

risk o compartment syndrome or tissue ne-

crosis [15, 16]. There is also potential health

care cost savings associated with the use o a

smaller amount o contrast material.

Because the ecacy o the 30-mL pulmo-

nary CTA technique had not yet been estab-

lished, the 30-mL protocol was used in a se-

lected group o patients with no overt clinical

suspicion o pulmonary emboli, although they

were at increased risk o incidental pulmonaryemboli. These patients had been reerred or

contrast-enhanced CT or indications such as

ollow-up o resected lung or other cancers

and ollow-up o parenchymal opacities and

nodules suspicious or primary or metastatic

cancer. In one study, 70% o 20 incidental em-

boli ound in 581 patients occurred in patients

with malignancy [17]. In another reported

study o 403 oncology patients, 4% o patients

were ound to have incidental pulmonary em-

boli [18]. In another study o 407 oncology pa-

tients, 4.4% o patients had unsuspected em-

boli but seven o the 18 emboli, or 39%, were

identiable only on a pulmonary CTA study;

these results led the authors to suggest modi-

ying the CT protocol in oncology patients to

include a pulmonary CTA examination [19].

Other benets o the new protocol include

opacication o the hilar vessels or better

evaluation o central structures in patients

who have undergone thoracic surgeries and

improved detection o developing signi-

cant central adenopathy compared with un-

enhanced scans. No incidental pulmonary

emboli were detected in the 30-mL group,

which is likely related both to the low-pre-

test probability (complete absence o clinicalsuspicion or pulmonary emboli) and to the

small sample size.

Investigators have reported that nondiag-

nostic pulmonary CTA examinations tend

to occur in patients with severe congestive

heart ailure and ar rhythmia using a regular

dose o contrast material [7]. We theorized

that the 30-mL pulmonary CTA technique

would be even more dependent on a rela-

tively normal circulation time and relative-

ly normal let ventricular unction than the

ull-dose pulmonary CTA study because oincreased susceptibility o a smaller contrast

bolus to dilution by an increased intravascu-

lar volume or to unopacied blood returning

to the heart. Hence, the 30-mL pulmonary

CTA protocol was not used in patients with

documented class III or IV congestive heart

ailure or in those with irregular supraven-

tricular tachycardia. Thereore, in selecting

the matched cohort 100-mL control group,

patients with class III or IV congestive heart

ailure and those with irregular supraven-

tricular tachycardia were not included. Sim-

ilarly, because no pulmonary emboli were

detected in the 30-mL group, patients with

pulmonary emboli were excluded rom the

100-mL control group.

There are several weaknesses to the study

including the retrospective nature o the study

and the potential selection bias in the 30- and

100-mL groups. Despite exclusion o patients

with congestive heart ailure and those with

supraventricular tachycardia in both groups

and the lack o an age dierence in the two

groups, it is possible that the clinical condition

o the patients in the 100-mL group was worse

than in the 30-mL contrast group. The patients

in the 100-mL group were reerred or pulmo-nary CTA because o clinical suspicion or

pulmonary emboli. Although no pulmonary

emboli were detected in these patients, they

most likely had respiratory distress, chest pain,

A

Fig. 1Pulmonary CT angiographic images o 22-year-old man obtained on 64-MDCT using 30 mL o 320 mg I/mL IV contrast material.A, Axial image through level o main pulmonary artery shows adequate contrast opacifcation o main, right main, and let main pulmonary artery branches.B, Coronal reormatted image shows adequate opacifcation o segmental and subsegmental branches.

B

7/28/2019 Ajr 2012 Pulm.bta 30ml

5/5

AJR:199 , December 2012 1251

Pulmonary CTA Using 30 mL of Contrast Material

or other signs and symptoms raising the pos-

sibility o pulmonary emboli. We should also

note that the 30-mL pulmonary CTA scans

could be dierentiated rom the 100-mL pul-

monary CTA scans by the relative absence o

contrast material in the pulmonary veins, the

let heart, and the aorta. Thereore, although

radiologists were blinded to the amount ocontrast material actually used or each ex-

amination during retrospective evaluation, the

possibility o observer bias in the subjective

image scores cannot be entirely excluded.

It is noteworthy that the average enhance-

ment o the segmental and subsegmental

branches is higher than the average enhance-

ment o the more proximal branches using

our 30-mL protocol, suggesting that more o

the contrast material was in the distal portion

o the pulmonary arterial system. Distribu-

tion o such a small volume o contrast mate-

rial theoretically allows both easier and bet-

ter detection o segmental and subsegmental

emboli, which are generally more dicult to

recognize. Detection o central pulmonary

emboli is not usually as dicult because o

the larger size o both the vessel and the em-

boli. Hence, the lower average attenuation

within the main and lobar pulmonary ar-

tery branches o the 30-mL pulmonary CTA

studies compared with 100-mL pulmonary

CTA studies did not result in a signicant

dierence in the subjective diagnostic qual-

ity scores between the two groups.

The successul use o the 30-mL CTA tech-

nique depends on certain actors being keptrelatively stable such as the monitoring scan

interval, the region o interest and attenuation

setting used to trigger the scan, and the con-

trast inusion rate. Because all these actors are

interrelated, a change in one actor will prob-

ably require changes in one or more actors to

compensate. In addition, the technique may

have to be adjusted or dierent scanners.

In conclusion, this study showed that the

contrast agent dose or pulmonary CTA can

be reduced signicantly without compromis-

ing image quality in careully selected groups

o patients using a 64-MDCT scanner. In

most subjects the image quality and attenu-

ation values obtained were adequate to diag-

nose the presence or absence o pulmonary

embolism. Randomized prospective studies

are required to conrm preservation o im-

age quality and diagnostic accuracy with the30-mL pulmonary CTA technique in a larger,

less selected patient population.

References

1. Perrier A, Roy PM, Sanchez O, et al. Multidetector-

row computed tomography in suspected pulmonary

embolism.N Engl J Med2005; 352:17601768

2. Weiss CR, Scatarige JC, Diette GB, Haponik EF,

Merriman B, Fishman EK. CT pulmonary angi-

ography is the rst-line imaging test or acute pul-

monary embolism: a survey o US clinicians.

Acad Radiol 2006; 13:434446

3. British Thoracic Society Standards o Care Com-

mittee Pulmonary Embolism Guideline Develop-

ment Group. British Thoracic Society guidelines

or the management o suspected acute pulmo-

nary embolism. Thorax2003; 58:470483

4. Wittram C. How I do it: CT pulmonary angiogra-

phy.AJR 2007; 188:12551261

5. Horlander KT, Mannino DM, Leeper KV. Pulmo-

nary embolism mortality in the United States,

19791988: an analysis using multiple-cause mor-

tality data.Arch Intern Med2003; 163:17111717

6. Bae KT, Tao C, Gurel S, et al. Eect o patient

weight and scanning duration on contrast en-

hancement during pulmonary multidetector CT

angiography.Radiology 2007; 242:5825897. Kelly AM, Patel S, Carlos RC, Cronin P, Kazer-

ooni EA. Multidetector row CT pulmonary angi-

ography and indirect venography or the diagnosis

o venous thromboembolic disease in intensive

care unit patients.Acad Radiol 2006; 13:486495

8. Matsuoka S, Hunsaker AR, Gill RR, et al. Vascu-

lar enhancement and image qual ity o MDCT pul-

monary angiography in 400 cases: comparison o

standard and low kilovoltage settings.AJR 2009;

192:16511656

9. Ramadan SU, Kosar P, Sonmez I, Karahan S, Ko-

sar S. Optimisation o contrast medium volume

and injection-related actors in CT pulmonary

angiography: 64-slice CT study.Eur Radiol 2010;

20:21002107

10. Radon MR, Kaduthodil MJ, Jagdish J, et al. Po-

tentials and limitations o low-concentration con-

trast medium (150 mg iodine/ml) in CT pulmo-

nary angiography. Clin Radiol 2011; 66:4349

[Erratum in Clin Radiol 2011; 66:692]

11. Bruce RJ, Djamali A, Shinkl K, Michel SJ, Fine

JP, Pozniak MA. Background fuctuation o kid-

ney unction versus contrast-induced nephrotox-

icity.AJR 2009; 192:711718

12. Li J, Solomon RJ. Creatinine increases ater intra-

venous contrast administration: incidence and

impact.Invest Radiol 2010; 45:471476

13. Trivedi H, Foley WD. Contrast-induced nephrop-

athy ater a second contrast exposure. Ren Fail

2010; 32:796801

14. Laville M, Juillard L. Contrast-induced acute kid-

ney injury: how should at-risk patients be identi-

ed and managed?J Nephrol 2010; 23:387398

15. Wang CL, Cohan RH, Ellis JH, Adusumilli S,

Dunnick NR. Frequency, management, and out-

come o extravasation o nonionic iodinated con-

trast medium in 69,657 intravenous injections.

Radiology 2007; 243:8087

16. Bellin MF, Jakobsen JA, Tomassin I, et al.; Con-

trast Media Saety Committee o the European So-

ciety o Urogenital Radiology. Contrast medium

extravasation injury: guidelines or prevention and

management.Eur Radiol 2002; 12:28072812

17. Storto ML, Di Credico AD, Guido F, Larici AR,

Bonomo L. Incidental detection o pulmonary

emboli on routine MDCT o the chest.AJR 2005;184:264267

18. Gladish GW, Choe D, Marom EM, Sablo BS,

Broemeling LD, Munden RF. Incidental pulmo-

nary emboli in oncology patients: prevalence, CT

evaluation, and natural history. Radiology 2006;

240:246255

19. Browne AM, Cronin CG, English C, NiMhuirche-

aetaigh J, Murphy JM, Bruzzi JF. Unsuspected pul-

monary emboli in oncology patients undergoing

routine computed tomography imaging. J Thorac

Oncol 2010; 5:798803