Benefits of 3D Rotational DSA Compared with 2D DSA in the Evaluation of Intracranial Aneurysm

P annotatePDF v9.4b

INSTRUCTIONS ON THE ANNOTATION OF PDF FILES

To view, print and annotate your article you will need Adobe Reader version 8 (or higher). This program is freelyavailable for a whole series of platforms that include PC, Mac, and UNIX and can be downloaded fromhttp://get.adobe.com/reader/. The exact system requirements are given at the Adobe site:http://www.adobe.com/products/reader/tech specs.html.

Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriateplace in the PDF file.

PDF ANNOTATIONS

Adobe Reader version 8 or 9 Adobe Reader version X

To make annotations in the PDF file, go to the main Adobetool bar and change the cursor from a hand symbol to the

normal cursor by clicking on the ‘Select’ button inthe menu bar at the top (version 8).

When you open the PDF file using Adobe Reader, theCommenting tool bar should be displayed automatically; ifnot, click on ‘Tools’, select ‘Comment & Markup’, then clickon ‘Show Comment & Markup tool bar’ (or ‘ShowCommenting bar’ on the Mac). If these options are notavailable in your Adobe Reader menus then it is possiblethat your Adobe Acrobat version is lower than 8 or the PDFhas not been prepared properly.

(PC, Adobe Reader version 8, right click on titlebar (Comment & Markup) to show additional

icons)

(Mac)PDF ANNOTATIONS (Adobe Reader version 9)

The default for the Commenting tool bar is set to ‘off’ inversion 9. To change this setting select ‘Edit | Preferences’,then ‘Documents’ (at left under ‘Categories’), then selectthe option ‘Never’ for ‘PDF/A View Mode’ (the Commentingtool bar is the same as in version 8).

(Changing the default setting, Adobe version 9)

To make annotations in the PDF file, open the PDF file usingAdobe Reader X, click on ‘Comment’.

If this option is not available in your Adobe Reader menusthen it is possible that your Adobe Acrobat version is lowerthan X or the PDF has not been prepared properly.

This opens a task pane and, below that, a list of allComments in the text. These comments initially show allthe changes made by our copyeditor to your file.

HOW TO...

Action Adobe Reader version 8 or 9 Adobe Reader version XInsert text

Click the ‘Text Edits’ button on theCommenting tool bar. Click to set the cursorlocation in the text and simply start typing. Thetext will appear in a commenting box. You mayalso cut and paste text from another file into thecommenting box. Close the box by clicking on ‘x’ inthe top right hand corner.

Click the ‘Insert Text’ icon on the Commenttool bar. Click to set the cursor location in the textand simply start typing. The text will appear in acommenting box. You may also cut and paste textfrom another file into the commenting box. Close

the box by clicking on ‘_’ in the top right handcorner.

Replace textClick the ‘Text Edits’ button on theCommenting tool bar. To highlight the text to bereplaced, click and drag the cursor over the text.Then simply type in the replacement text. Thereplacement text will appear in a commenting box.You may also cut and paste text from another fileinto this box. To replace formatted text (anequation for example) please Attach a file (seebelow).

Click the ‘Replace (Ins)’ icon on theComment tool bar. To highlight the text to bereplaced, click and drag the cursor over the text.Then simply type in the replacement text. Thereplacement text will appear in a commenting box.You may also cut and paste text from another fileinto this box. To replace formatted text (anequation for example) please Attach a file (seebelow).

Remove textClick the ‘Text Edits’ button on theCommenting tool bar. Click and drag over the textto be deleted. Then press the delete button onyour keyboard. The text to be deleted will then bestruck through.

Click the ‘Strikethrough (Del)’ icon on theComment tool bar. Click and drag over the text tobe deleted. Then press the delete button on yourkeyboard. The text to be deleted will then bestruck through.

Highlight text/make acomment

Click on the ‘Highlight’ button on theCommenting tool bar. Click and drag over the text.To make a comment, double click on thehighlighted text and simply start typing.

Click on the ‘Highlight Text’ icon on theComment tool bar. Click and drag over the text. Tomake a comment, double click on the highlightedtext and simply start typing.

Attach a file

Click on the ‘Attach a File’ button on theCommenting tool bar. Click on the figure, table orformatted text to be replaced. A window willautomatically open allowing you to attach the file.To make a comment, go to ‘General’ in the‘Properties’ window, and then ‘Description’. Agraphic will appear in the PDF file indicating theinsertion of a file.

Click on the ‘Attach File’ icon on theComment tool bar. Click on the figure, table orformatted text to be replaced. A window willautomatically open allowing you to attach the file.A graphic will appear indicating the insertion of afile.

Leave a note/comment Click on the ‘Note Tool’ button on

the Commenting tool bar. Click to set the locationof the note on the document and simply starttyping. Do not use this feature to make text edits.

Click on the ‘Add Sticky Note’ icon on theComment tool bar. Click to set the location of thenote on the document and simply start typing. Donot use this feature to make text edits.

HOW TO...

Action Adobe Reader version 8 or 9 Adobe Reader version XReview To review your changes, click on the ‘Show’

button on the Commenting toolbar. Choose ‘Show Comments List’. Navigate byclicking on a correction in the list. Alternatively,double click on any mark up to open thecommenting box.

Your changes will appear automatically in a listbelow the Comment tool bar. Navigate byclicking on a correction in the list. Alternatively,double click on any mark up to open thecommenting box.

Undo/deletechange

To undo any changes made, use the right clickbutton on your mouse (for PCs, Ctrl Click for theMac). Alternatively click on ‘Edit’ in the mainAdobe menu and then ‘Undo’. You can alsodelete edits using the right click (Ctrl click onthe Mac) and selecting ‘Delete’.

To undo any changes made, use the right clickbutton on your mouse (for PCs, Ctrl Click for theMac). Alternatively click on ‘Edit’ in the mainAdobe menu and then ‘Undo’. You can alsodelete edits using the right click (Ctrl click onthe Mac) and selecting ‘Delete’.

SEND YOUR ANNOTATED PDF FILE BACK TO ELSEVIER

Save the annotations to your file and return as instructed by Elsevier. Before returning, please ensure you haveanswered any questions raised on the Query Form and that you have inserted all corrections: later inclusion of anysubsequent corrections cannot be guaranteed.

FURTHER POINTS

Any (grey) halftones (photographs, micrographs, etc.) are best viewed on screen, for which they are optimized,and your local printer may not be able to output the greys correctly.

If the PDF files contain colour images, and if you do have a local colour printer available, then it will be likely thatyou will not be able to correctly reproduce the colours on it, as local variations can occur.

If you print the PDF file attached, and notice some ‘non standard’ output, please check if the problem is alsopresent on screen. If the correct printer driver for your printer is not installed on your PC, the printed output willbe distorted.

Original Investigation

Benefits of 3D Rotational DSACompared with 2D DSA in the

Evaluation of Intracranial AneurysmQ2 Siong Chuong Wong, MD, MRad, Ouzreiah Nawawi, MBBS, MRad,

Norlisah Ramli, MBBS, Khairul Azmi Abd Kadir, MBBS, MRad

Rationale and Objectives: The aims of this study were to compare conventional two-dimensional (2D) digital subtraction angiography

(DSA) with three-dimensional (3D) rotational DSA in the investigation of intracranial aneurysm in terms of detection, size measurement,neck diameter, neck delineation, and relationship with surrounding vessels. A further aimwas to compare radiation dose, contrast volume,

and procedural time between the two protocols.

Materials and Methods: Thirty-five patients who presented with subarachnoid bleeds on computed tomography and were suspected ofhaving intracranial aneurysms underwent conventional 2D DSA followed by 3D DSA. The 3D digital subtraction angiographic images were

displayed as surface shaded display images. Aneurysm detection, sac size, neck diameter, neck delineation, and relationship of aneurysm

to the surrounding vessels analyzed from the two protocols were compared. Radiation dose, contrast volume, and procedural time for both

examinations were also compared.

Results: Three-dimensional DSA detected 44 aneurysms in 31 patients, with negative findings seen in four patients. A false-negative

detection rate of 6.8% (three of 44) for 2D DSA was noted. There was no significant difference in aneurysm size between 3D and 2D

DSA. The sizes of aneurysm necks were found to be significantly larger in 3D DSA than on 2D DSA. The aneurysm neck and relationshipto surrounding vessels were significantly better demonstrated on 3D DSA than on 2D DSA. Radiation dose (entrance surface dose),

contrast use, and procedural time with 3D DSA were significantly less than with 2D DSA.

Conclusions: Three-dimensional DSA improves the detection and delineation of intracranial aneurysms, with lower radiation dose, lesscontrast use, and shorter procedural time compared to 2D DSA. The size of the aneurysm neck on 3D DSA tended to be larger than on

2D DSA.

Key Words: Imaging; intracranial aneurysm; cerebral angiography; 3D DSA.

ªAUR, 2012

Cerebral aneurysm is a potentially life threatening

disorder, which may result in spontaneous subarach-

noid hemorrhage and is further complicated by

hydrocephalus, vasospasm, and brain infarction. Apart from

localizing the aneurysm, the aim of imaging is to measure

the size and neckof the aneurysm, aswell as determine the rela-

tionship of the aneurysm to the surrounding vessels. Imaging

a cerebral aneurysm can be done using several imaging

methods, with the noninvasive techniques being computed

tomographic angiography and magnetic resonance angiog-

raphy. The introduction of three-dimensional (3D) recon-

struction of the rotational angiographic images has given

reviewers the advantage of viewing the vascular anatomy in

any angle and plane, thus making it useful for viewing small

aneurysms or aneurysms in areas of arterial branching that

may be missed on two-dimensional (2D) angiography.

Hochmuth et al (1) reported that compared to biplanar digital

subtraction angiography (DSA), 3D rotational angiography

allows more accurate depiction of anatomic details that are

essential in planning surgical and endovascular treatment for

intracranial aneurysms in terms of improving the delineation

of aneurysmal neck (71%), the parent vessel (45%), and the

relationship to adjacent vessels (50%). In addition, 3D DSA

allows the detection of more aneurysms, especially small aneu-

rysms (<3 mm), which are not detected on DSA (1–3). With

regard to radiation dose, 3D DSA can also reduce the number

of exposures compared to 2D DSA, not only to determine the

working projection for therapy but also for procedures (1,4,5).

However, unlike in previous studies inwhich only the standard

projections of 2D DSA were compared to 3D DSA, in this

study, we included additional 2D digital subtraction

angiographic views in the evaluation and comparison.

In this study, we aimed to confirm the beneficiary role of

3D DSA in the diagnosis and characterization of cerebral

aneurysms and to demonstrate the overall reductions of cost,

Acad Radiol 2012; -:1–7

From the Department of Biomedical Imaging, University Malaya MedicalCentre, 59100 Kuala Lumpur, Malaysia (S.C.W., O.N., K.A.A.K.); and theFaculty of Medicine, University Malaya Research Imaging Centre, UniversityMalaya, Kuala Lumpur, Malaysia (N.R.). Received October 25, 2011;accepted February 16, 2012. Address correspondence to: O.N. e-mail:[email protected]

ªAUR, 2012doi:10.1016/j.acra.2012.02.012

1

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

5657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110

time, and radiation dose. This is especially important in

a developing nation where resources are scarce, so that

procuring expensive equipment must be justified with clear

benefits in terms of cost and time savings.

MATERIALS AND METHODS

Study Design

This was a cross-sectional study including 35 patients who

underwent angiography for suspected intracranial aneurysm

from August 2008 to November 2009. The patients presented

with clinical histories of headache, vomiting, dizziness, neck

pain, altered sensorium, and diplopia, with computed tomog-

raphy showing subarachnoid hemorrhages. The study was

carried out with the approval of the hospital’s medical ethics

committee. Written informed consent was obtained from

patients when possible or from close relatives for patients unfit

to give consent.

Angiography

All patients were imaged using a Siemens angiographic unit

(Axiom-Artis VB30E; Siemens Healthcare, Erlangen,

Germany). The C-arm system used was a single-plane

AXIOM Artis dFa C-Arm Angiography System (Siemens

Healthcare). Image acquisitionwas performedusing a dynamic

flat-panel detector system with a 48-cm diagonal entrance

plane, producing an imagewith a 1024� 1024matrix. Ultrav-

ist 300 mg I/mL (Schering AG, Berlin, Germany) was used for

both 2D DSA and rotational angiography.

All catheterizations were performed with femoral artery

access. Standard 4-Fr or 5-Fr diagnostic catheters were used

to cannulate the intracranial arteries and to perform the angio-

graphic examination. In the first phase of the study (2DDSA),

selective angiography of both internal carotid arteries (ICA)

and vertebral arteries was performed in the Towne’s, lateral,

and oblique views (standard projections). If an aneurysm

was detected, additional projections were performed until

the aneurysm neck was satisfactorily demonstrated.

Following 2DDSA, the second phase of the study (3D rota-

tional DSA) was performed on the vessel that showed or was

suspected of having the aneurysm, regardless of the findings

on 2D DSA. Precontrast data acquisition was performed with

200� of C-arm rotation, at a total rotation time of 5 seconds

at 30 frames/s. This resulted in a C-arm speed of 40�/s. A total

of 266 images were produced. The angiography arm then was

returned to its initial position, after which a postcontrast rota-

tion, taking another 5 seconds, was performed in the same

manner. The images from the contrasted rotational angiogram

were subtracted from its equivalent mask per scanner protocol,

resulting in rotational digital subtraction angiographic images.

Image Analysis

The 2D digital subtraction angiographic images and rotational

digital subtraction angiographic raw data were transferred to

a multimodality workstation (Leonardo; Siemens Healthcare),

on which 3D surface shaded displays (SSDs) of the cerebral

vascular supply were reconstructed from the rotational digital

subtraction angiographic source data. The 2D digital subtrac-

tion angiographic images were reviewed first, followed by the

3D digital subtraction angiographic images. The parameters

analyzed were aneurysm detection, aneurysm sac size, aneu-

rysm neck delineation, aneurysm neck diameter, and relation-

ship of the aneurysm to surrounding arteries.Measurementwas

performed using the built-in automated isocenter calibration

system available on the Siemens angiographic unit. A neurora-

diologist and an interventional radiologist, each with 8 years of

experience, reviewed the images independently. The results

were reviewed, and the final decision in case of any discrepancy

was reached by consensus between the two reviewers.

Measurement of Radiation Dose

The parameters measured were entrance surface dose (ESD),

which is the radiation dose to skin at the entrance point of the

radiation beam, and total dose-area product (DAP). Both

parameters were calculated using the built-in electronic

dose-measuring chamber available on the C-arm system.

The radiation doses for total 2D DSA (standard and additional

views) and for additional views on 2D DSA (only additional

views) were measured and compared to the measured 3D

DSA radiation dose.

Measurement of Contrast Volume

The contrast volumes used for total (standard and additional

views) 2D DSA and for additional views on 2D DSA (only

additional views) were recorded and compared to the contrast

volume used for 3D DSA.

Measurement of Procedural Time

Procedural time for 2D DSA was calculated from the time

taken to perform the first acquisition until the last acquisition,

including additional views. As for 3D DSA, procedural time

started from the setup position of the mask rotation until

the end of 3D computer subtraction.

Statistical Analysis

Statistical analysis of differences between the two techniques

in aneurysm sac diameter, neck diameter, radiation dose,

contrast volume and procedural time were performed using

Mann-Whitney U tests. The difference in aneurysm neck

delineation and the relationship of the aneurysm to

surrounding vessels was determined using Wilcoxon’s test.

Statistical significance for all analyses was taken as P < .05.

RESULTS

Four patients (three men, one woman) had no aneurysms

detected on either 2D DSA or 3D DSA. There were 22

WONG ET AL Academic Radiology, Vol -, No -, - 2012

2

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165

166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220

women and 13 men recruited in our study, for a ratio of about

2:1. The age range was 16 to 71 years, with a mean age of 50.5

� 13.5 years.

Aneurysm Detection

Forty-four aneurysms were detected on 3D DSA, while 41

aneurysms were detected on 2D DSA, for a false-negative

detection rate of three of 44 (6.8%) for 2D DSA. The three

aneurysms that were not detected on 2D DSA measured

<3 mm in size; one was obscured by a larger aneurysm

(Fig 1), and two were hidden by bends in the intracavernous

portion of the ICA (Fig 2).

Number and Sites of Aneurysms

The majority of patients had single aneurysms (26 patients

[83.9%] on 2D DSA and 25 patients [80.6%] on 3D DSA).

Four patients (12.9%) and three patients (9.7%) had two aneu-

rysms on 2D and 3D DSA, respectively. Two patients (6.5%)

had three aneurysms detected only on 3D DSA. The largest

number of aneurysms detected in a patient on 2D and 3D

DSA was seven, in a patient with hypertension who had

undergone renal transplantation. The patient had five aneu-

rysms measuring <5 mm and two aneurysms measuring 5 to

10 mm. Most of the aneurysms were found to arise from

the ICA (25%), followed by the anterior communicating

artery (23%) and posterior communicating artery (23%).

Only one aneurysm originated from the basilar artery.

The largest aneurysm measured in our study was a left

supraclinoid ICA aneurysm, measuring 23.1 mm on 2D

DSA and 22.9 cm on 3D DSA in a patient with hypertension.

The smallest aneurysm on 2D DSA measured 1.5 mm but

measured 1.1 mm in 3D DSA. Twenty-four aneurysms

(58.5%) were found to be larger on 3D DSA than on 2D

DSA. Four aneurysms (9.8%) had the same measurements

with both techniques, while 13 of 41 aneurysms measured

smaller on 3D DSA than on 2D DSA. However, these differ-

ences were not statistically significant.

The majority of aneurysms were <5 mm in size. Table 1

summarizes the frequency of aneurysms according to sac

size measured on 2D DSA and 3D DSA.

View of the Aneurysm Neck

There was a statistically significant difference between 2D

and 3D DSA in the capability to delineate the necks of

the aneurysms, with better delineation seen on 3D DSA

(P < .0005).

Three-dimensional DSA was able to clearly delineate the

necks of all the aneurysms except one. The same aneurysm

was also not well visualized on standard 2D DSA or on the

additional acquisitions. The reason for the poor neck visuali-

zation was that the aneurysm arose between or straddled across

a bifurcation in the distal left middle cerebral artery branches

(Fig 3). Table 2 summarizes the delineation of aneurysm necks

on 2D DSA and 3D DSA.

Diameter of the Aneurysm Neck

The diameters of the aneurysm necks were significantly larger

(P = .02) on 3D DSA, averaging 3.0 mm, than on 2D DSA,

averaging 2.6 mm. Comparing 3D DSA to 2D DSA in 31

patients, aneurysm neck diameters were found to be larger

in 22 patients (71%), similar in three (10%), and smaller in

six (19%). The largest difference was 2.5 mm, measured in

a patient with a bilobar anterior communicating artery aneu-

rysm (Fig 4).

Figure 1. Two-dimensional (2D) digital

subtraction angiographic image from lateral

acquisition of left internal carotid artery(ICA) run (a) showing a large aneurysm

(open black arrow) arising from the supracli-

noid part of the left ICA. There was another

smaller aneurysm (open white arrows) de-tected on three-dimensional digital subtrac-

tion angiographic lateral (b) and oblique

lateral (c) views, close to the large aneurysm,which was not seen on 2D digital subtraction

angiography.

Q1Academic Radiology, Vol -, No -, - 2012 ---

3

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275

276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330

Relationship of the Aneurysm to Surrounding Vessels

There was a statistically significant difference between 3D

DSA and 2D DSA in the capability to delineate aneurysms

from the surrounding vessels (P < .0005). On 3D DSA, the

relationship of the surrounding vessels to the aneurysm was

well delineated for all aneurysms. However, on standard

conventional 2D digital subtraction angiographic views,

only one aneurysm could be clearly delineated from the

surrounding arteries. Using additional 2D views, 23 of the

41 aneurysms became well delineated, but 17 remained

obscured by overlying vessels.

Radiation Dose and Contrast Media Volume

The ESDs for total 2D DSA (mean, 287.14 mGy;

maximum, 882.6 mGy) and for additional views on 2D

DSA (mean, 173.31 mGy; maximum, 607.60 mGy) signifi-

cantly exceeded the ESD for 3D DSA (mean, 93.25 mGy;

maximum, 165.20 mGy). The differences in ESDs between

the two study protocols were statistically significant

(P < .0005 for total 2D DSA total vs 3D DSA, P = .011

for additional views on 2D DSA vs 3D DSA).

We also found that the DAP for total 2D DSA (mean,

2797.29mGy $m2;maximum, 6428.20mGy $m2)was slightly

larger than that for 3D DSA (mean, 2632.23 mGy $ m2;

maximum, 4666.20 mGy $ m2). However, this difference

was not statistically significant. On the other hand, the DAP

for 3D DSA was significantly larger (P < .0005) than

the DAP for additional views on 2D DSA (mean, 1053.57

mGy $ m2; maximum, 2911.4 mGy $ m2).

Significantly (P < .0005) more contrast was used for total

2D acquisitions (mean, 59.46 mL; maximum, 130 mL) and

additional views (mean, 33.00 mL; maximum, 90 mL)

compared to 3D DSA (mean, 21.1 mL; maximum, 36 mL).

Procedural Time

The times needed to perform total 2D DSA (mean, 189.2

seconds; maximum, 1345 seconds) and additional views on

2D DSA (mean, 163.10 seconds; maximum, 1325 seconds)

were longer than for 3D DSA (mean, 101.7 seconds;

maximum, 110 seconds). The difference was statistically

significant (P = .028) for total 2D DSA compared to 3D

DSA but was not statistically significant for additional views

on 2D DSA compared to 3D DSA.

DISCUSSION

In this study, we considered 3D DSA to be the relative ‘‘refer-

ence standard’’ compared to 2DDSA. This is because previous

studies have stated that there is substantial benefit of the

former technique, especially in aneurysm detection, depiction

of aneurysm shape, neck location, and relationship to parent

artery (1,2,4). The overall incidence of aneurysm detection

from spontaneous subarachnoid hemorrhage in our study

was 88.6%, which is within the range of previous studies

Figure 2. Two-dimensional digital subtrac-tion angiography (DSA) of the left internal

carotid artery (ICA) in an anteroposterior

(AP) projection (a) did not show any aneu-

rysm. However, the AP (b) and oblique AP(c) projections on three-dimensional DSA

showed an aneurysm (open white arrow)

arising from the intracavernous part of the

left ICA.

TABLE 1. Frequency of Aneurysms According to Sac SizeMeasured on 2D DSA and 3D DSA

Size Category (mm)

2D DSA 3D DSA

n % n %

1–5 26 63.4 31 70.5

5–10 10 24.4 10 22.7

10–15 3 7.3 1 2.3

15–20 1 2.4 1 2.3

20–25 1 2.4 1 2.3

Total 41 100.0 44 100.0

DSA, digital subtraction angiography; 3D, three-dimensional; 2D,

two-dimensional.

WONG ET AL Academic Radiology, Vol -, No -, - 2012

4

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385

386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440

(6,7). The average sizes of aneurysms in our study were

5.5 mm on 2D DSA and 5.3 mm on 3D DSA, similar to

a study done in Germany (8). Almost all the aneurysms in

our study arose from the anterior circulation, in particular

the distal ICA, anterior communicating artery, and posterior

communicating artery.

The recent adoption of 3D rotational angiography has

enabled better understanding and discussion of complex intra-

cranial vascular anatomy surrounding aneurysms among inter-

ventionists, neuroradiologists, and neurosurgeons (9,10). In

addition, patients and their relatives can also be shown the

manipulation of SSD images in the explanation of the

aneurysm, methods of treatment, and the risks of

endovascular therapy. This can promote better understanding

of the intracranial vascular anatomy so that patients can be

better informed in the decision-making process.

Our results show that the main benefit of 3D DSA in

neuroradiology is the 3D visualization of complex anatomic

vascular patterns, especially aneurysms in difficult locations,

complex configurations (multilobulation), and obscuring of

the aneurysm neck in standard projections. The visualization

of the neck and delineation from surrounding vascular struc-

tures are vastly improved on 3D DSA compared to 2D DSA.

This benefits both interventionists and neurosurgeons in

terms of improved visualization of aneurysms to their parent

vessels and surrounding vascular structure and planning safe

treatment. In particular, these benefits enable a more

informed decision on whether an aneurysm is better suited

for coil placement or surgical clipping. More important, in

endovascular treatment, knowledge of detailed anatomy can

help in selecting the shape of the coil, configuration of the

aneurysm, accurate neck localization, and angiographic

working position during coil placement and deployment (11).

As documented in previous studies (2,12), our data also

show that there is a greater tendency to detect aneurysms on

3D DSA compared to 2D DSA. In our case, this is especially

true for aneurysms <4 mm in size. In the standard 2D digital

subtraction angiographic projections, these small aneurysms

are often obscured by surrounding vessels. In contrast, the

multiplanar capability of 3D DSA enables the vessels to be

virtually rotated out of the view of small aneurysms,

resulting in a higher confidence level for detecting small

aneurysms. If 3D DSA were not performed, patients with

undetected aneurysms would fall into the angiographically

negative subarachnoid hemorrhage group and would

routinely be subjected to further investigations and possibly

surgical exploration.

We used 3D SSD to detect, localize, and measure aneurysm

size and the aneurysm neck. Display of aneurysms on 3D SSD

involves the selection of a threshold of Hounsfield units

(HUs), above which the contrast density within the artery is

shaded. Although the threshold of HUs is predetermined by

a preset, it may also be changed manually to optimize vessel

visualization. The threshold of HUs selected may affect the

size of the measured aneurysm, whereby lower a HU

threshold may artificially make vessels appear larger and vice

versa. Another 3D data set display technique is maximum-

Figure 3. Two-dimensional digital subtrac-

tion angiography of the left internal carotidartery in an anteroposterior acquisition (a)and oblique projection (b) showing an aneu-

rysm arising from the distal middle cerebralartery, with the neck straddled between

a bifurcation (open black arrow). The three-

dimensional digital subtraction angiographic

manipulated image (c) of the same vesselalso could not clearly delineate the aneu-

rysm’s neck (open white arrow) from the

bifurcation.

TABLE 2. Delineation of Aneurysm Necks on 2D DSA and 3DDSA

Neck View

2D DSA 3D DSA

n % n %

Not well delineated 19 46.34 1 2.27

Well delineated 1 2.44 43 97.73

Well delineated in

additional view

21 51.22 NA NA

Total 41 100.00 44 100.00

NA, not applicable.

Academic Radiology, Vol -, No -, - 2012 ---

5

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495

496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550

intensity projections (1,4,5). However, maximum-intensity

projections are less superior at displaying the morphologic

characteristics of aneurysms, although they have similar detec-

tion and neck visualization rates to SSD (4).

We found that there was no significant difference between

the two protocols in the size of the aneurysm sac. This is in

contrast to a study done by Kawashima et al (12), who found

that aneurysm size was larger on 3D DSA than on 2D DSA.

However, we did find a significant difference in the sizes aneu-

rysm necks. The sizes of aneurysm necks that could be

measured on 2D DSA tended to be smaller compared to 3D

DSA. Previous authors have also reported larger measurements

of aneurysm necks on 3D SSD than on 2DDSA, leading them

to conclude that 3DDSAmay be inferior to 2DDSA for triage

of aneurysms to or from endovascular therapy. However, they

acknowledged that results may be affected by the type of

equipment used to perform 3D DSA and the threshold values

selected to analyze the 3DSSD images. Their study also did not

show if their findings would necessarily affect patient manage-

ment (13). In our study, although the mean neck diameter

measured larger on 3D SSD than on 2D DSA, the difference

was only in the submillimeter range, which is unlikely to affect

management decisions. Furthermore, the success of aneurysm

embolization does not depend solely on neck size but more

importantly on other factors, such as the angle between the

sac and the parent artery, the configuration of the aneurysm

sac, and differentiation of the neck from adjacent branches

(14). These factors are shown to be better evaluated by 3D

DSA than 2D DSA (2,4,11).

In trying to explain the discrepancy in neck diameters

between the two techniques, we offer these contributing

factors. First, the angle of rotation may have some effect on

the reconstruction and measurement of the neck. Second,

the mathematical algorithm used to display the aneurysm

may not have been optimized to display the angle of the aneu-

rysm neck. Third, the aneurysm neck, usually being of smaller

structure and caliber than the aneurysm, is more affected by

the HU threshold selected for the display of SSD images. A

phenomenon related to the angle of rotation beam is also

known to affect aneurysm neck measurement in 3D images.

However, this phenomenon, known as pseudostenosis, causes

the aneurysm neck to measure smaller on 3DDSA than on 2D

DSA (15). Because of these conflicting findings, further inves-

tigation, such as correlation with surgical or endovascular

findings, should be undertaken.

We noted that 2D DSA resulted in a statistically significant

higher ESD compared to 3DDSA. In terms of biologic effect,

this means that there is a lower risk for developing determin-

istic effects, such as skin erythema and cataracts, with 3D DSA

compared to 2D DSA. The maximum ESD for 2D DSA

recorded in our patients was almost 1 Gy. Fortunately, none

of our patients reported any deterministic effect. This may

be because the radiation was not consistently concentrated

on a single particular area during the procedure. Our results

indicate that the risk for deterministic effects on patients

would be greatly reduced if 3D DSA were to replace the

acquisition of additional views on 2D DSA.

The DAP value or stochastic risk of 3D DSA was signifi-

cantly lower compared to that of total 2D DSA but not signif-

icantly higher compared to that of additional views on 2D

DSA. A previous study also found higher stochastic risk for

3D DSA compared to additional views on 2D DSA (16).

The main consideration for stochastic risk is in children and

young adults, as the risk for developing malignancy increases

in this age group. The DAP is the product of the incident

dose and x-ray field area, which are totaled over the entire

field of view. For 2D DSA, the fluoroscopic field of view

was manually collimated to just within the skull boundary.

In contrast, for 3D rotational angiography, the field of view

extended beyond the skull margin because of the standard

rotational beam. As a result, the cumulated DAP for 3D

DSA will be larger than the corresponding DAP for 2D

DSA, thus overestimating the patient’s exposed area as well

as the risk for stochastic effects.

Figure 4. Two-dimensional (2D) (a) and

three-dimensional (3D) (b) digital subtractionangiographic images showing a bilobar

aneurysmarising from the anterior communi-

cating artery. The neck diameter measured

smaller on 2D digital subtraction angiog-raphy (open black arrow) than on 3D digital

subtraction angiography (open white arrow).

WONG ET AL Academic Radiology, Vol -, No -, - 2012

6

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605

606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660

We found a significant reduction of contrast use on 3DDSA

compared to 2D DSA. This is not surprising, as a single rota-

tional angiographic acquisition requires only a single contrast

injection of 18 mL, while 2D DSA necessitates multiple

contrast injections because of the multiple projections. The

contrast volume reduction would not only provide cost effec-

tiveness but would also be very beneficial in children (16–18).

It is also more time saving to perform 3D DSA instead of

additional projections on 2D DSA, with an average saving

of 1 minute. This time saving is important, especially in crit-

ical patients, in whom angiographic procedure time needs to

be as short as possible. However, depending on the hardware,

3D DSA reconstruction involves additional time for data

transfer and image processing. With more technological

advancement in the future, this duration will be reduced.

It is important to note that because of the necessity of mask

images, both 2D DSA and 3D DSA are susceptible to small

motion artifacts. Motion or registration artifacts are a known

limitation of subtraction studies. Although this may not affect

overall image quality, it may still influence the measurement of

size, especially on 3D DSA.

Apart from diagnosing and characterizing aneurysms, there

are other uses of 3D rotation. Three-dimensional rotation can

be used without subtraction for on-table evaluation of post-

procedural complications (eg, intracranial hematoma, intra-

ventricular hemorrhage, hydrocephalus, and brain edema),

without the need to transfer to a computed tomographic

scan (19). In the setting of aneurysm clipping, 3D DSA can

also show more aneurysm remnant than 2D DSA (20).

Neurosurgeons can also use the information from 3D DSA

for guidance of aneurysm surgery (21).

CONCLUSIONS

In our investigation of intracranial aneurysms, the main bene-

fits of 3D DSA are for visualization of the aneurysm neck and

delineation of the aneurysm from surrounding vessels. Three-

dimensional DSA also detects more aneurysms than 2D DSA,

especially small aneurysms, and can reliably measure aneurysm

size. Three-dimensional DSA has additional advantages over

2D DSA in terms of significantly lower radiation dose, lesser

contrast use, and shorter procedural time.

REFERENCES

1. Hochmuth A, Spetzger U, Schumacher M. Comparison of three-

dimensional rotational angiography with digital subtraction angiography

in the assessment of ruptured cerebral aneurysms. AJNR Am J Neurora-

diol 2002; 23:1199–1205.

2. van Rooij WJ, Sprengers ME, de Gast AN, et al. 3D rotational angiography:

the new gold standard in the detection of additional intracranial aneu-

rysms. AJNR Am J Neuroradiol 2008; 29:976–979.

3. White PM, Teasdale EM, Wardlaw JM, et al. Intracranial aneurysms:

CT angiography and MR angiography for detection prospective

blinded comparison in a large patient cohort. Radiology 2001; 219:

739–749.

4. Sugahara T, Korogi Y, Nakashima K, et al. Comparison of 2D and 3D digital

subtraction angiography in evaluation of intracranial aneurysms. AJNR Am

J Neuroradiol 2002; 23:1545–1552.

5. Abe T, Hirohata M, Tanaka N, et al. Clinical benefits of rotational 3D angi-

ography in endovascular treatment of ruptured cerebral aneurysm. AJNR

Am J Neuroradiol 2002; 23:686–688.

6. Kim BJ, Park KJ, Shin IY, et al. Clinical analysis of spontaneous subarach-

noid hemorrhage with negative angiography. Korean J Cerebrovasc Surg

2009; 11:5–11.

7. Suarez JI, Tarr RW, SelmanWR. Aneurysmal subarachnoid hemorrhage. N

Engl J Med 2006; 354:387–396.

8. Beck J, Rohde S, Berkefeld J, et al. Size and location of ruptured and un-

ruptured intracranial aneurysms measured by 3-dimensional rotational

angiography. Surg Neurol 2006; 65:18–25.

9. Tanoue S, Kiyosue H, Kenai H, et al. Three-dimensional reconstructed

images after rotational angiography in the evaluation of intracranial aneu-

rysms: surgical correlation. Neurosurgery 2000; 47:866–871.

10. Casta~no-Duque CH, Guardia-Mas E, de Juan-Delago M, et al. Pre-embo-

lization study of ruptured cerebral aneurysm with rotational and 3D recon-

struction angiography. Int Cong Ser 2002; 1247:39–52.

11. Anxionnat R, Bracard S, Ducrocq X, et al. Intracranial aneurysms: clinical

value of 3D digital subtraction angiography in the therapeutic decision and

endovascular treatment. Radiology 2001; 218:799–808.

12. Kawashima M, Kitahara T, Soma K, et al. Three-dimensional digital

subtraction angiography vs two-dimensional digital subtraction angiog-

raphy for detection of ruptured intracranial aneurysms: a study of 86 aneu-

rysms. Neurol India 2005; 53:287–289.

13. Brinjikji W, Cloft H, Lanzino G, et al. Comparison of 2D digital subtraction

angiography and 3D rotational angiography in the evaluation of dome-to-

neck ratio. AJNR Am J Neuroradiol 2009; 30:831–834.

14. Cognard C,Weill A, Castaings L, et al. Intracranial berry aneurysms: angio-

graphic and clinical results after endovascular treatment. Radiology 1998;

206:499–510.

15. Hirai T, Korogi Y, Ono K, et al. Pseudostenosis phenomenon at volume-

rendered three-dimensional digital angiography of intracranial arteries:

frequency, location, and effect on image evaluation. Radiology 2004;

232:882–887.

16. Racadio JM, Fricke BL, Jones BV, et al. Three-dimensional rotational

angiography of neurovascular lesions in pediatric patients. AJR Am J

Roentgenol 2006; 186:75–84.

17. Schueler BA, Kallmes DF, Cloft HJ. 3D cerebral angiography: radiation

dose comparison with digital subtraction angiography. AJNR Am J Neuro-

radiol 2005; 26:1898–1901.

18. Bridcut RR, Murphy E, Workman A, et al. Patient dose from 3D rotational

neurovascular studies. Br J Radiol 2007; 80:362–366.

19. Brzegowy P, Urbanik A, Popiela TJ. Application 3D DSA and angiographic

CT in interventional neuroradiology. Przegl Lek 2010; 67:300–305.

20. Ahn SS, Kim YD. Three-dimensional digital subtraction angiographic eval-

uation of aneurysm remnants after clip placement. J Korean Neurosurg

Soc 2010; 47:185–190.

21. Raabe A, Beck J, Rohde S, et al. Three-dimensional rotational angi-

ography guidance for aneurysm surgery. J Neurosurg 2006; 105:

406–411.

Academic Radiology, Vol -, No -, - 2012 ---

7

FLA 5.1.0 DTD � XACRA2731_proof � 30 March 2012 � 3:10 pm � ce 68

661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715

716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770

Our reference: XACRA 2731 P-authorquery-v9

AUTHOR QUERY FORM

Journal: XACRA

Article Number: 2731

Dear Author,

Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen

annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than

Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please

return your corrections within 48 hours.

For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions.

Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in

the proof.

Location

in articleQuery / Remark: Click on the Q link to find the query’s location in text

Please insert your reply or correction at the corresponding line in the proof

Q1 Please provide a short title of no more than 6 words for the running head.

Q2 Please confirm that given names and surnames have been identified correctly.

Thank you for your assistance.