This study compared information obtained from standard computed tomographic angiography (s-CTA) scans and modified CTA (m-CTA) scans of the deep circumflex iliac artery (DCIA) flap to cadaver dissections. The m-CTA scans showed longer visible DCIA lengths, better visualization of branching patterns, and more detail on vessel course compared to s-CTA scans. However, s-CTA scans allowed bilateral evaluation while m-CTA only showed the injected side. Both CTA methods provided more information than cadaver dissections for preoperative planning of DCIA flaps.

1. SURGICAL ONCOLOGY AND RECONSTRUCTION
Patient-Specific Topographic Anatomy
of the Deep Circumflex Iliac Artery
Flap: Comparing Standard and
Modified Computed Tomographic
Angiography
Victoria Behrens, DMD,* Ali Modabber, MD, DDS, PhD,y Christina Loberg, MD,z
Andreas Herrler, MD, PhD,x Andreas Prescher, MD, PhD,k and
Alireza Ghassemi, MD, DDS, PhD{
Purpose: Computed tomographic angiography (CTA) is reported to give insight into patient-specific
anatomy of the flap pedicle preoperatively. We compared information available from standard CTA (s-
CTA) with that gained by modifying the conventional CTA technique (modified CTA [m-CTA]). Dissected
cadavers served as the control group.
Materials and Methods: We evaluated 16 s-CTA scans (32 deep circumflex iliac arteries [DCIAs]) and
12 m-CTA scans (17 DCIAs) using 3-dimensional software (Vesalius; ps-medtech, Amsterdam, The
Netherlands). We dissected 17 cadavers (n = 34 DCIAs) to serve as the control group. The positions of
4 landmarks (anterior superior iliac spine, origin of DCIA, origin of ascending branch, and crossing of hor-
izontal branch and iliac crest) were defined in a 3-dimensional coordinate system.
Results: We found significant differences concerning the distances from the origin of the DCIA to the
femoral bifurcation (P < .05) and the anterior superior iliac spine to the crossing point of the horizontal
branch with the iliac crest (P < .05) between CTA scans and cadaveric studies. The imaging quality of
the m-CTA scans was shown to be more consistent than and superior to that of the s-CTA scans. The visible
length of the DCIA was longer on m-CTA scans (mean, 134.32 mm) than on s-CTA scans (mean, 73.62 mm).
We could evaluate the branching off of perforators and the relation of the pedicle to the surrounding bone
and soft tissue in more detail on m-CTA scans. Standard CTA allowed the bilateral evaluation of the pedicle,
whereas m-CTA allowed the evaluation of the injected side only.
Conclusions: The quality and quantity of information available from CTA could be improved by modifying
the s-CTA examination by injection as close as possible to the target vessel. Standard CTA delivered informa-
tion about both sides, whereas m-CTA may need an additional injection for contralateral-side imaging.
Ó 2018 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg -:1-7, 2018
*Dentist, Private Dental Office, Viersen, Germany.
yConsultant, Department of Oral and Maxillofacial Surgery,
University Hospital, RWTH Aachen, Aachen, Germany.
zConsultant, Department of Diagnostic and Interventional
Radiology, University Hospital, RWTH Aachen, Aachen, Germany.
xAssociate Professor, Department of Anatomy and Embryology,
FHML Maastricht University, Maastricht, The Netherlands.
kAssociate Professor, Department of Molecular and Cellular
Anatomy, University Hospital, RWTH Aachen, Aachen, Germany.
{Oral and Maxillofacial Surgery Consultant, Klinikum-Lippe,
Detmold, Germany; Teaching Hospital of Georg-August-University
G€ottingen, G€ottingen, Germany; and Medical Faculty, University
RWTH Aachen, Aachen, Germany.
Conflict of Interest Disclosures: None of the authors have any
relevant financial relationship(s) with a commercial interest.
Address correspondence and reprint requests to Dr Ghassemi:
Department of Oral and Maxillofacial Surgery, Teaching Hospital
Klinikum-Lippe, R€ontgenstrasse 18, 32756 Detmold, Germany;
e-mail: aghassemi@ukaachen.de
Received December 21 2017
Accepted January 20 2018
Ó 2018 American Association of Oral and Maxillofacial Surgeons
0278-2391/18/30091-0
https://doi.org/10.1016/j.joms.2018.01.025
1
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO

2. Microvascular bone flaps are the gold standard for the
reconstruction of extensive bony defects. The selected
flap should restore the function and provide an
optimal esthetic outcome. Vascularized iliac bone
flap, nourished by the deep circumflex iliac artery
(DCIA), offers excellent bone stock to achieve these
goals.1
The exact knowledge of the anatomy of the
flap is essential. So far, several studies, mostly based
on cadaveric dissection, have emphasized the high
variability of the vessel anatomy. The variable anatomy
of the DCIA makes the harvest of this flap difficult and
time consuming.2
Recently,preoperative imaging provedto be helpfulto
deliver insight into the anatomy and topography of the
flap pedicle.3,4
An ideal preoperative imaging modality
should show individual variations; the location; the
caliber and branching pattern of the pedicle; and the
existence, quality, and quantity of perforating vessels.
Fabricating preoperative templates helps to design the
flap more accurately.5,6
Different imaging techniques
are available including digital subtraction angiography,
computed tomographic angiography (CTA), magnetic
resonance angiography, and Doppler ultrasound. CTA
is noninvasive and was reported to provide accurate
anatomic information,3,7
allowing fast imaging of larger
body areas with the option of 3-dimensional (3D) recon-
struction. However, it involves radiation exposure and
the injection of an iodine-rich contrast agent with the
risk of an allergic reaction.
The aim of this study was to evaluate the informa-
tion attainable from standard CTA (s-CTA) and CTA im-
ages taken according to a modified CTA (m-CTA)
protocol. We evaluated and compared the visible
course of the DCIA, its branches, and its relationship
to key landmarks. We defined the position of impor-
tant landmarks in a 3D coordinate system to transfer
the obtained virtual 3D data to the intraoperative situ-
ation. Results from cadaveric studies (CSs) served as
the control group.
Materials and Methods
The cadaveric specimens for this study were ob-
tained after institutional approval was received from
the Institute of Anatomy and Cell Biology of our univer-
sity hospital. Furthermore, 16 multiple-row s-CTA
scans (n = 32 DCIAs) obtained after intravenous injec-
tion of the nonionic contrast agent iopromide were
selected. In addition, we evaluated m-CTA images of
12 individuals (n = 17 DCIAs). The common femoral
artery was punctured in the inguinal region for injec-
tion of 40 mL of a nonionic iodine-containing contrast
medium (Ultravist 370; Bayer HealthCare, Leverkusen,
Germany) into the external iliac artery. A multislice
computed tomography (CT) scan with 1-mm slices
(Somatom Definition [128-slice CT system with Dual
Energy Body CT program]; Siemens, Erlangen, Ger-
many) was performed. It included bone and
soft tissue windows, maximum intensity projection,
and a multiplanar re-formation–volume rendering
technique. These CTA scans were performed for diag-
nostic and therapeutic purposes. They were provided
anonymized by the Department of Diagnostic and In-
terventional Radiology of the Medical Faculty, Univer-
sity RWTH Aachen, for further analysis. We evaluated
the visible length of the DCIA and analyzed the infor-
mation gained about several measured distances (Fig
1, Table 1). To display the CT images, we used 3D soft-
ware (Vesalius; ps-medtech, Amsterdam, The
Netherlands) and measured the distances between
landmarks and the lengths of vessels virtually. In addi-
tion, we defined the positions of the 4 following points
to visualize the course of the DCIA in a 3D coordinate
system: anterior superior iliac spine (ASIS), origin of
DCIA, origin of ascending branch (AB), and crossing
point of horizontal branch (HB) with iliac crest.
We defined a sagittal plane through the pubic sym-
physis, a frontal plane through the anterior rim of
the pubic symphysis, and a transverse plane through
the upper rim of the pubic symphysis. We measured
the perpendicular distances between each of the 4
chosen points and the 3 planes to obtain 3 coordinates
each, describing the position of the points in a 3D
coordinate system (Table 2). We dissected 17
formalin-preserved cadavers (n = 34 DCIAs) to serve
as the control group.
DATA ANALYSIS
We determined the mean, standard deviation, and
maximum and minimum values using Excel (Microsoft
Office 2007; Microsoft, Redmond, WA). The normal
distribution of the groups was determined, and
because of the non-normal distribution of the CTA
scans, the Wilcoxon test was performed using RStudio
(version 1.044; FOAS, Boston, MA) to determine the
significance of differences between the CTA and CS re-
sults, setting the border of significance as P < .05.
Results
We evaluated the course of the DCIA and its relation-
ship to the key landmarks (Fig 1) in 17 cadavers (n = 34
DCIAs; 11 women and 6 men), on 16 s-CTA scans
(n = 32 DCIAs; 7 women and 9 men), and on
12 m-CTA scans (n = 17 DCIAs; 7 women and
5 men). The distance between bone and vessel, as
well as the number and location of perforators to the
bone, could only be examined on the CTA
scans (Table 1).
The course of the DCIA could be evaluated bilater-
ally on all s-CTA scans (Fig 2) but only 5 of 12 m-CTA
scans (Fig 3). We could identify the division of the
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
2 MODIFIED CT-ANGIOGRAPHY OF THE DCIA

3. DCIA into the HB and AB on 11 of 12 m-CTA scans
(92%). However, this was only possible on 4 of 16
s-CTA scans (25%). In addition, we observed a higher
standard deviation of the visible length of the DCIA
on s-CTA scans (Æ35.62 mm) than on m-CTA scans
(Æ30.95 mm) (Table 1).
The origin of the DCIA was as high as the inferior
epigastric artery in all groups. On 54% of the CTA
print&web4C=FPO
FIGURE 1. Artistic illustration of deep circumflex iliac artery (DCIA) and vein and their relationship to important key landmarks: origin of DCIA
and vein (a), anterior superior iliac spine (b), crossing point of DCIA and iliac crest (c), origin of superficial circumflex iliac artery and vein (d),
bifurcation of femoral vessel (e), and location of horizontal branch along medial iliac surface (f).
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral Maxillofac Surg 2018.
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
BEHRENS ET AL 3

4. scans, the HB was shown to be the same size as the AB;
it was larger than the AB on 31% and smaller on 15%.
On a level with the ASIS, a horizontal distance of
26.24 mm (vs 15.05 mm in CSs) was measured be-
tween the HB and ASIS. At this point, the HB showed
a minimum distance of 11.78 mm to the inner surface
of the iliac bone (Table 1).
We observed a similar distribution of vessel types
based on the classification of Ghassemi et al.8
(2013)
(Table 3). On m-CTA scans, we could identify up to
5 perforators running to the medial bone surface.
However, we could not detect any perforators on the
s-CTA scans.
Comparing the results of CTA analysis with CSs, we
observed that the horizontal distance between the
ASIS and HB was significantly higher on CTA scans
(P < .0001). The distance between the origin of the
DCIA and the bifurcation of the femoral artery was
significantly longer in the CSs compared with the
CTA scans (P < .05). The distance between the ASIS
and crossing point of the HB with the iliac crest was
significantly longer on CTA scans compared with CSs
(P < .05) (Table 1).
The positioning of the 4 landmarks in a 3D coordi-
nate system (triangulation) showed that the DCIA
had its origin at a distance of 68.49 mm from the
sagittal plane, 39.77 mm from the transverse plane,
and 6.29 mm posterior to the frontal plane (Table 2).
On the m-CTA scans, we observed detailed anatomy
including anastomosis of the DCIA and adjacent ves-
sels such as the iliolumbar artery (n = 4).
Discussion
The DCIA flap offers excellent bone quality and
quantity required for the reconstruction of extensive
bony defects. However, the anatomy of the DCIA has
been shown to be highly variable.9-11
The knowledge
Table 1. COMPARISON BETWEEN CADAVERIC DISSECTIONS, S-CTA, AND M-CTA
Mean Æ SD (Min-Max)
P ValueCadavers (n = 34) s-CTA (n = 32) m-CTA (n = 17)
Age, yr 83.82 Æ 6.56 (71-92) 63.69 Æ 12.14 (21-78) 50.18 Æ 10.54 (23-66) —
Visible length of DCIA, mm — 73.62 Æ 35.62 (0-151.4) 134.32 Æ 30.95 (53.1-198) —
a to b, mm 68.63 Æ 7.51 (50-90) 72.72 Æ 6.67 (59.3-83.8) P > .05
a to origin HB, mm 47.41 Æ 17.15 (15-105) 33.88 Æ 9.64 (17-62.8) P > .05
a to d, mm 23.73 Æ 9.45 (0-78) 19.34 Æ 3.14 (9.6-23.9) P > .1
a to e, mm 57.08 Æ 11.99 (21-100) 48.44 Æ 12.06 (23.2-90.8) P < .05
b to HB (horizontal), mm 15.05 Æ 6.11 (2-40) 26.24 Æ 4.61 (18.4-36.4) P < .01
b to c, mm 41.49 Æ 18.76 (0-91) 56.73 Æ 15.17 (35.4-85.8) P < .05
DCIA to inner surface of iliac
bone,* mm
Not measurable because
of tissue collapse
11.78 Æ 2.54 (5.9-19) —
Abbreviations: a, origin DCIA; b, anterior superior iliac spine; c, dorsal crossing point of HB with iliac crest, d, origin of super-
ficial circumflex iliac artery; DCIA, deep circumflex iliac artery; e, bifurcation of femoral artery; HB, horizontal branch; m-CTA,
modified computed tomographic angiography; Max, maximum; Min, minimum; n, number of DCIAs; s-CTA, standard computed
tomographic angiography; SD, standard deviation.
* Measured level with anterior superior iliac spine.
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral Maxillofac Surg 2018.
Table 2. TRIANGULATION (N = 24): DISTANCE TO PLANE
Mean Æ SD (Min-Max), mm
Sagittal Transversal Frontal
ASIS 121.70 Æ 9.12 (102.6-154.7) 88.08 Æ 8.42 (63.1-103.1) À2.97 Æ 11.89 (À23.1 to 23)*
Origin of DCIA 68.49 Æ 5.48 (51.7-78) 39.77 Æ 8.29 (20.6-72.5) 6.29 Æ 7.88 (À9.8 to 21.4)
Origin of AB 94.87 Æ 11.42 (65.5-118.1) 77.45 Æ 17.35 (36-117) 4.08 Æ 8.29 (À14.4 to 22.1)
Crossing point with crest 117.73 Æ 7.27 (98.1-132.2) 128.08 Æ 16.09 (78.5-165.4) 31.85 Æ 15.86 (À2.6 to 126.5)
Abbreviations: AB, ascending branch; ASIS, anterior superior iliac spine; DCIA, deep circumflex iliac artery; Max, maximum;
Min, minimum; SD, standard deviation.
* The ASIS lies mainly posterior to the frontal plane.
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral Maxillofac Surg 2018.
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
4 MODIFIED CT-ANGIOGRAPHY OF THE DCIA

5. print&web4C=FPO
FIGURE 2. Standard computed tomographic angiography scan. A, anterior.
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral Maxillofac Surg 2018.
print&web4C=FPO
FIGURE 3. Modified computed tomographic angiography scan. A, anterior.
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral Maxillofac Surg 2018.
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
BEHRENS ET AL 5

6. of pedicle variations and the exact course of the
pedicle and its relationship to intraoperatively
identifiable landmarks can expedite the dissection,
reduce complications, and increase the success rate.
In addition, the location and number of perforators,
the type of vessel, the topographic relationship of
the ASIS and iliac crest, and the crossing point with
the iliac crest and its pathway into the soft tissue are
important for successful flap harvesting.
CSs deliver only general anatomic information and
do not give any information regarding the spatial rela-
tionship of the DCIA to the surrounding structures.
Currently, CTA is reported as the gold standard for
3D topography, and it has been shown to be accurate
and superior in imaging abdominal wall perforators as
compared with ultrasound or magnetic resonance
angiography.12
CTA imaging allows more realistic visu-
alization of the topographic anatomy and is hence
more useful for preoperative planning.
We compared the information gained from the CTA
scans with the extent of information available from
CSs. For m-CTA, the contrast agent was injected as close
as possible to the DCIA pedicle. We analyzed the course
of the DCIA on s- and m-CTA scans and used CSs as the
control group (Table 1). The data from the triangulation
were used to display the course of the DCIA in a 3D co-
ordinate system on preoperative planning. We found
that by modifying the CTA protocol as described, we
could evaluate the DCIA more thoroughly than with
conventional CTA. In addition, using 3D software facili-
tated an extensive study regarding the patients’ specific
topographic anatomy of the DCIA and its relationship to
important landmarks.
We observed a higher variability of the measure-
ments, as expressed by the standard deviation, in CSs
compared with CTA studies. One explanation could
be that using the Vesalius equipment in CTA delivered
an unaltered 3D topography as compared with CSs. In
CSs, the horizontal distance between the ASIS and HB
was significantly smaller than that measured on CTA
scans (Table 1). This could be due to the shrinkage
and collapse of the tissue after preservation of the
cadaver.13
However, the distance between the origin
of the DCIA and the bifurcation of the femoral artery
was longer in CSs (57.08 mm) than on CTA scans
(48.44 mm). The distance between the ASIS and the
crossing point of the HB with the iliac crest was longer
on CTA scans than in CSs (56.7 mm vs 41.5 mm). The
reason could be that, using the 3D software, we were
able to follow the vessel course and the curvature of
the iliac crest more precisely with the measuring tool.
Using a cord and ruler in the CSs’ surrounding struc-
tures distorted the measurement. The CTA data seemed
to match the in vivo situation better than the CSs.
Analyzing the m-CTA scans, we could easily measure
the distance between the HB and the medial surface of
the iliac bone and examine the number and location of
perforators to the bone, as well as the diameter of the
HB and AB. The preserved cadavers could not deliver
this information. Information regarding the perfora-
tors was scarce on s-CTA scans (Fig 2). However, we
could observe multiple penetrating periosteal vessels
originating from the HB on the m-CTA scans (Fig 3).
Such information is helpful in selecting the suitable
part of the iliac bone with a reliable blood supply.
Furthermore, the osteotomy can be planned on preop-
erative templates, and donor-site morbidity can be
reduced by harvesting only the necessary bone stock.5
Although Ting et al (2009)14
stated that CTA imaging
offers high-resolution images, we could analyze only
6 of 32 s-CTA scans to obtain the sought information.
In addition, the quality and dimension of visualization
of the DCIA on s-CTA scans are subject to high vari-
ability. On m-CTA scans, we could track the DCIA far
into the periphery with clear illustration of the perfora-
tors to the bone in 24%, as compared with 0% on s-CTA
scans. While analyzing the m-CTA scans, we also could
trace the anastomosis of the DCIA with other vessels
such as the iliolumbar vessel, as described by Taylor
and Watson15
(1978). The application of contrast agent
close to the DCIA, as performed on m-CTA, was shown
to increase the illustration quality.
By using 3D software, we could clearly differentiate
between the HB and AB by virtual 3D rotation of the
CTA images and observe their relationship to neigh-
boring structures more realistically. This reduced
superimposition and false interpretation, which are
disadvantages of 2-dimensional imaging. The 3D visu-
alization also allowed the description of relevant parts
of the DCIA within a 3D coordinate system in relation
to intraoperative identifiable landmarks. This will facil-
itate the intraoperative orientation and identification
of the DCIA’s branches and simplify the dissection.
Nevertheless, not all the m-CTA scans showed the
same good quality. This should be evaluated in further
Table 3. TYPE OF VESSEL OBSERVED ACCORDING TO
CLASSIFICATION OF GHASSEMI ET AL
8
Cadavers CTA (s-CTA or m-CTA)
Type Ia 82% (n = 31) 57% (n = 8)
Type Ib 5% (n = 2) 14% (n = 2)
Type Ic 8% (n = 3) 7% (n = 1)
Type II 5% (n = 2) 7% (n = 1)
Type III 0% (n = 0) 14% (n = 2)
Abbreviation: CTA, computed tomographic angiography;
m-CTA, modified computed tomographic angiography;
n, number of types of classification; s-CTA, standard
computed tomographic angiography.
Behrens et al. MODIFIED CT-ANGIOGRAPHY OF THE DCIA. J Oral
Maxillofac Surg 2018.
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
6 MODIFIED CT-ANGIOGRAPHY OF THE DCIA

7. studies by improving the examination technique.
A frequently changing time span between injection
and CTA imaging could be one reason. The technique
of injection or catheter positioning needs to be opti-
mized. Nonetheless, we also should consider the inva-
sive nature and potential morbidity caused by catheter
angiography.16
Standard CTA requires a less invasive
peripheral intravenous injection and allows bilateral
examination. If we apply the m-CTA technique, a sec-
ond injection may be needed to select the suitable
side. This can increase the rate of possible complica-
tions such as hematoma and allergic reaction.16
In summary, we performed a retrospective study
comparing the information gained by s-CTA with that
gained by m-CTA. Modified CTA delivered more abun-
dant and detailed information important for flap harvest-
ing and preoperative planning. However, we also should
consider the higher rate of possible complications, as
knownindigital subtraction angiography. Using 3D visu-
alization allowed more accurate and detailed measure-
ments as compared with CSs. Further studies with a
larger sample size and matched groups are needed to
illuminate and refine the technique presented.
Acknowledgment
We are grateful to those who donated their bodies to science. We
express our sincere thanks to Dirk Traufelder for his contribution in
drafting the illustration.
References
1. Jewer DD, Boyd JB, Manktelow RT, et al: Orofacial and mandibular
reconstructionwiththe iliaccrest free flap: Areviewof60 casesand
a new method of classification. Plast Reconstr Surg 84:391, 1989
2. Baliarsing AS, Kumar VV, Malik NA, et al: Reconstruction of
maxillectomy defects using deep circumflex iliac artery-based
composite free flap. Oral Surg Oral Med Oral Pathol Oral Radiol
Endod 109:8, 2010
3. Ribuffo D, Atzeni M, Saba L, et al: Clinical study of peroneal ar-
tery perforators with computed tomographic angiography: Im-
plications for fibular flap harvest. Surg Radiol Anat 32:329, 2010
4. Rozen WM, Ting JW, Leung M, et al: Advancing image-guided sur-
gery in microvascular mandibular reconstruction: Combining
bony and vascular imagingwith computed tomography-guidedster-
eolithographic bone modeling. Plast Reconstr Surg 130:227, 2012
5. Modabber A, Gerressen M, Stiller MB, et al: Computer-assisted
mandibular reconstruction with vascularized iliac crest bone
graft. Aesthetic Plast Surg 36:653, 2012
6. Ayoub N, Ghassemi A, Rana M, et al: Evaluation of computer-
assisted mandibular reconstruction with vascularized iliac crest
bone graft compared to conventional surgery: A randomized
prospective clinical trial. Trials 15:114, 2014
7. Tregaskiss AP, Goodwin AN, Bright LD, et al: Three-dimensional
CT angiography: A new technique for imaging microvascular
anatomy. Clin Anat 20:116, 2007
8. Ghassemi A, Furkert R, Prescher A, et al: Variants of the supply-
ing vessels of the vascularized iliac bone graft and their relation-
ship to important surgical landmarks. Clin Anat 26:509, 2013
9. Taylor GI, Townsend P, Corlett R: Superiority of the deep circum-
flex iliac vessels as the supply for free groin flaps. Plast Reconstr
Surg 64:595, 1979
10. Bitter K, Danai T: The iliac bone or osteocutaneous transplant
pedicled to the deep circumflex iliac artery. I. Anatomical and
technical considerations. J Maxillofac Surg 11:195, 1983
11. Bergeron L, Tang M, Morris SF: The anatomical basis of the
deep circumflex iliac artery perforator flap with iliac crest. Plast
Reconstr Surg 120:252, 2007
12. RozenWM,StellaDL,BowdenJ,etal:Advancesinthepre-operative
planning of deep inferior epigastric artery perforator flaps: Mag-
netic resonance angiography. Microsurgery 29:119, 2009
13. Tran T, Eble JN, Grignon DJ, et al: Correcting the shrinkage
effects of formalin fixation and tissue processing for renal
tumors: Toward standardization of pathological reporting of
tumor size. J Cancer 6:759, 2015
14. Ting JW, Rozen WM, Grinsell D, et al: The in vivo anatomy of the
deep circumflex iliac artery perforators: Defining the role for the
DCIA perforator flap. Microsurgery 29:326, 2009
15. Taylor GI, Watson N: One-stage repair of compound leg defects
with free, revascularized flaps of groin skin and iliac bone. Plast
Reconstr Surg 61:494, 1978
16. Tavakol M, Ashraf S, Brener SJ: Risks and complications of
coronary angiography: A comprehensive review. Glob J Health
Sci 4:65, 2012
FLA 5.5.0 DTD Š YJOMS58142_proof Š 30 March 2018 Š 3:58 am Š CE KO
BEHRENS ET AL 7