University of Groningen
The significance of preoperative vascular mapping of donor- and acceptor vessels in
free flap surgery
Klein, Steven
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to
cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2013
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Klein, S. (2013). The significance of preoperative vascular mapping of donor- and acceptor vessels in free
flap surgery Groningen: s.n.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the
author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.
Download date: 15-06-2017
The significance of preoperative vascular
mapping of donor- and recipient vessels in free
flap surgery.
S.Klein
Paranimfen
Dr. Martin W. Stenekes
Dr. Saskia K. Klein
Financial support for printing this dissertation was obtained by kind contributions
from: Department of Plastic and Reconstructive Surgery of UMCG, Maatschap
Plastische Chirurgie Zuidoost-Brabant, Allergan, BlooMEDical Benelux NV, Dalton
Medical, ERBE, Johnson & Johnson Medical BV, van Wijngaarden Medical.
ISBN: 978-90-367-6377-6
Titel: The significance of preoperative vascular mapping of donor- and
acceptor vessels in free flap surgery.
Author: S. Klein
Cover:
A.J. Klein and S. Klein
Lay-out: S. Klein
Pringing: Gildeprint Drukkerijen, Enschede
© Steven Klein, 2013
No part of this book may be reproduced or transmitted in any form or by any means,
electronically or mechanically, including photocopying, recording or using any
information storage and retrieval system, without the written permission of the author,
or, when appropriate, of the publisher of the publication.
RIJKSUNIVERSITEIT GRONINGEN
The significance of preoperative vascular
mapping of donor- and acceptor vessels in free
flap surgery
Proefschrift
ter verkrijging van het doctoraat in de
Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. E. Sterken,
in het openbaar te verdedigen op
woensdag 25 september 2013
om 12:45 uur
door
Steven Klein
geboren op 20 mei 1977
te Neuss, Duitsland
Promotor:
Prof. dr. P.M.N. Werker
Copromotor:
Dr. J.M. Smit
Beoordelingscommissie:
Prof. dr. R.R.W.J. van der Hulst
Prof. dr. D.J.O. Ulrich
Prof. dr. C.J.A.M. Zeebregts
CONTENTS
1
Design and rationale of the thesis
p.
3
2
Perforatorflaps – the evolution of a reconstructive surgical technique.
Ned Tijdschr Geneeskd 2005; 149: 2392-2398
p.
11
3
Donor site necrosis following fibula free flap transplantation.
p.
21
p.
33
p.
51
p.
67
p.
87
Microsurgery 2005; 25: 538-542
4
Review: An overview of methods for vascular mapping in the planning
of free flaps.
J Plast Reconstr Aesthet Surg 2010; 63: e674-682
5
Ankle-arm index versus angiography for the preassessment of the
fibula free flap.
Plast Reconstr Surg 2003; 111: 735-743
6
General review: Measurement, calculation, and normal range of the ankle
arm index: a bibliometric analysis and recommendation for standardization.
Ann Vasc Surg 2006; 20: 282-292
7
Evaluation of the lower limb vasculature before free fibula flap transfer.
A prospective blinded comparison between magnetic resonance angiography
and digital subtraction angiography.
Accepted for publication in Microsurgery
8
Is there an indication for digital subtraction angiography in the assessment
p.
101
of irradiation-induced vascular damage prior to free flap surgery by the means
of the internal mammary vessels?
Accepted for publication in Journal of Reconstructive Microsurgery
9
Summary and general discussion
p.
113
10
Summary and general discussion (Dutch)
p.
125
11
Explanation of used abbreviations
p.
135
12
Acknowledgements (Dutch)
p.
139
13
Curriculum vitae
p.
145
14
List of publications
p.
149
2
Chapter
1
Design and rationale of the thesis
3
4
Design and rationale of the thesis
Trauma, oncological resections and pressure sores may lead to major soft tissue defects, for which
ultimately a free flap is the only option. By definition, a free flap is an organ-like piece of tissue, which
can be isolated on a vascular pedicle and moved from one site on the body to another site. During this
transfer its blood supply is clamped and divided from its donor site and as soon as possible thereafter
attached or anastomosed at or near the recipient site. As the artery and vein(-s) of such a flap are of
relatively small caliber (usually smaller than 2-3mm), the anastomes are usually made with
microsurgical techniques.
The origin of microvascular surgery can be traced back to the early 1900’s. In 1906 the first kidney
transplantation was done, but lost due to a graft versus host reaction. Alexis Carrel(1) described
reproducible methods of suturing large vessels with high patency rates, and in the 1920’s the operating
microscope was introduced by ENT-doctor Nylen.(2) It took until 1957 before free vascularized human
tissue transfer with smaller vascular anastomoses was accomplished when an esophagus was
reconstructed with a free jejunal segment.(3) However due to the popularity of regional flap techniques
and lack of fine sutures and instruments, it took until the 1970s before the real microvascular break
through occurred. During that period multiple reports of free fasciocutaneous flaps(4-7) as well as other
flaps(8,9) were published. In the decennia thereafter more and more flaps were developed with different
constitution and characteristics.
For a successful free flap reconstruction three stages can be distinguished; the preoperative planning of
the flap, the surgery itself and the postoperative care. All advancements within these three different
stages may contribute to a successfull, safer, faster and easier to perform free flap procedure of higher
quality.
This thesis will focus on the pre-operative planning of flap transfer. A decent preoperative planning of
the flap can improve surgery by giving the surgeon insight into the vascular anatomy and its surrounding
beforehand. This is especially the case in flaps in which the vascular anatomy is not constant, as in for
example perforators flaps. Perforator flaps are based on perforating vessels that branch off and can be
traced back to well-known vessels, thereby limiting donor-site morbidity. Preoperative vascular mapping
in these flaps can help to identify the dominant perforator and its course. With this information at least in
theory the surgeon may be able to select the best perforator prior to surgery and thereby make the
surgery more straightforward.
Preoperative vascular mapping can also be performed to prevent donorsite complications, like in the
case of the free fibula flap. Prior to the harvest of a free fibula flap the arteries of the lower legs are often
mapped in order to exclude peripheral arterial occlusive disease or congenital anomalies in any of the
crural vessels. If present, it may hamper the use of the fibula free flap because of potential insufficiency
of the flap’s peroneal vascular pedicle or impairment of the remaining anterior and posterior tibial
vascular supply after harvest of the peroneal vessels.(10-12) In flaps with a straightforward vascular
anatomy vascular mapping is only of importance in cases where previous surgery or trauma might have
damaged the vascular pedicle.
5
A number of vascular mapping methods exist. Some methods are simple and can be performed as part
of the physical exam by the surgeon him or herself, with or without the help of easy to handle
equipment, and some methods require specific equipment and/or radiological expertise. The easiest
method of assessment of vessels is probably palpation of the vessel, which can be done during physical
examination. An example of an easy to handle adjunct that can be used during physical examination is
the hand-held Doppler, while the ankle-arm index is also a method used to gain insight into the vascular
status of the crural vessels. The hand-held Doppler (HHD) uses a pencil-type probe to register moving
erythrocytes by sending out and detecting reflected ultrasound. Depending on the depth and the
diameter of the vessels to be investigated, various probes with different frequencies can be used. The
two most commonly used frequencies are 8 and 10 MHz.(13) The ankle-arm index (AAI), introduced by
Winsor in 1950,(14) is the ratio of systolic blood pressure at the level of the ankle to that at the level of
the arm. The AAI is often used as a first step to assess peripheral arterial occlusive disease (PAOD).
The role of the AAI in the preoperative assessment of flaps needs to be further investigated.
More advanced mapping methods are colour duplex sonography (CDS), conventional digital subtraction
angiography (DSA), computed tomography angiography (CTA) and magnetic resonance angiography
(MRA). CDS relies on the same working principle as HHD. Blood flow in vessels is detected by the
physical principle of a direct relationship between the recorded Doppler frequency shift and blood-flow
velocity. In addition, different velocities and directions of moving blood streams can be displayed on a
screen in the colour duplex mode. As such, CDS does not only offer information about the internal
vessel diameter and their course, but also depicts the 3D footprint of the perivascular anatomy.(15) DSA
used in interventional radiology, depends on fluoroscopy to visualize blood vessels in a bony or dense
soft tissue environment. Injecting contrast medium and subtracting a ‘pre-contrast image’ produce
images. CTA combines the use of X-rays with computerized 3D analysis of the images. The number of
detector rows decides how fast a scan can be performed and to what extent details can be revealed. A
great variety of CT scanners and software is currently available, enabling the generation of slices of
approximately 1 mm or even thinner, depending on the CT scanner used. MRA imaging uses a powerful
magnetic field to align the nuclear magnetization of hydrogen nuclei in the body. Radio-frequency
pulses are used to tip the alignment of the hydrogen nuclei away from the main magnetic field, causing
the hydrogen nuclei to produce a radio-frequency signal that is detectable by the scanner. Analyzed by
a computer, this yields detailed pictures of organs, soft tissues, bone and virtually all other internal body
structures.
The overall aim of this thesis is to shed light on preoperative mapping in different type of reconstructions
and to investigate which preoperative mapping method is the most valuable for which type of
reconstructions.
6
The specific aims are:
1. To describe the evolution of reconstructive flap surgery.
2. To illustrate the necessity of vascular mapping prior to the harvest of the free fibular flap.
3. To scrutinize the various methods used to assess ankle arm index and its normal range, and to
recommend a standardized method to assess ankle arm index based on that analysis
4. To create an overview of the methods used in the pre-operative planning of free flaps.
5. To compare the ankle-arm index to the current golden standard, digital subtraction angiography
for the pre-operative assessment of the fibula free flap.
6. To compare magnetic resonance angiography to the current golden standard, the digital
subtraction angiography for the pre-operative assessment of the fibula free flap
7. To relate pre-operative angiography of the internal mammary artery to the degree of vascular
damage found during the operation, the clinical course of a free flap connection to it and to the
histology of segments of the recipient artery.
7
Outline of this thesis
In chapter two the evolution of reconstructive flap surgery is described. The increased understanding of
the vascular anatomy of tissues enabled this evolution.
In chapter three the “condito sine qua non” of a good vascular supply to a free flap and to the remaining
donor site will be illustrated with the example of the complicated case histories of three patients in whom
a free fibular flap was harvested. The severity of flap failure or occurrence of severe complications at
the donor site stresses the importance of a proper vascular assessment as part of the preoperative
work-up.
Since its introduction in 1950, a variety of methods of measurement and calculation have been used to
establish an ankle-arm index (AAI). This has resulted in variations of its normal range and difficulty in
comparing study results. Hence, the objective of our study, described in chapter four, is to analyze the
various methods used to assess AAI and its normal range and to come with a standardized method to
assess AAI based on that analysis.
The fifth chapter will give an overview of the various methods for vascular mapping of flaps together
with their advantages and disadvantages. The pro’s and con’s of the hand-held Doppler, colour duplex,
digital subtraction angiography (DSA), computed tomographic angiography (CTA) and magnetic
resonance angiography (MRA) will be reviewed and discussed.
As peripheral arterial occlusive disease or congenital anomalies of the major crural arteries may limit the
use of the fibula free flap, these conditions should be detected preoperatively. In chapter six the AAI is
compared to the current golden standard, the digital subtraction angiography for the pre-assessment of
the fibula free flap. We will test the hypothesis that the ankle AAI of each of the three crural arteries,
combined with pencil Doppler examination of the peroneal skin perforators, would provide sufficient
information to make the use of angiography superfluous.
In Chapter seven 3D-Time Of Flight Magnetic Resonance Angiography (MRA) is compared to the
current golden standard, the Digital Subtraction Angiography (DSA) for the pre-operative assessment of
the fibula free flap.
A successful transplantation of tissue is not only dependent on the good vascular supply of the flap, but
also on the condition of the vessels at the recipient site. In chapter eight the ability to preoperatively
assess the presence of atherosclerosis or irradiation damage to the vessels is studied in the internal
mammary artery used during breast reconstructions using free flaps. For this purpose, pre-operative
angiography was related to the degree of vascular damage found during the operation, the clinical
course following flap transfer and the histology of segments of the recipient artery.
8
The findings of the preceding chapters are summarized and discussed in chapter nine, and suggestions
for future research are given.
References
1 Carrel A. Results of the transplantation of blood vessels, organs and limbs. JAMA.1908;L(20):1662-7
2 Nylen CO. The microscope in aural surgery: Its first use and later development. Acta Otolaryngol
Suppl. 1954;116:226-40.
3 Seidenberg B, Rosenak SS, Hurwitt ES, Som ML. Immediate reconstruction of the cervical
esophagus by a vascularized isolated jejunal segment Ann Surg 1959;149:162-171.
4 Rigg BM. Transfer of a free groin flap to the heel by microvascular anastomoses. Plast Reconstr
Surg. 1975;55:36-40.
5 Harii K, Ohmori K. Free groin flaps in children. Plast Reconstr Surg. 1975;55:588-92.
6 Soutar DS, Scheker LR, Tanner NS, McGregor IA. The radial forearm flap: a versatile method for
intra-oral reconstruction. Br J Plast Surg. 1983;36:1-8.
7 Robinson DW. Microsurgical transfer of the dorsalis pedis neurovascular island flap. Br J Plast Surg.
1976;29:209-13.
8 Taylor GI, Miller GD, Ham FJ. The free vascularized bone graft. A clinical extension of microvascular
techniques. Plast Reconstr Surg. 1975;55:533-44.
9 Taylor GI, Daniel RK. The free flap: composite tissue transfer by vascular anastomosis. Aust N Z J
Surg. 1973;43:1-3.
10 Seres L, Csaszar J, Voros E, Borbely L. Donor site angiography befor mandibular reconstruction with
fibula free flap. J Craniofac Surg 2001; 12:608-613.
11 Futran ND, Stack BC Jr., Payne LP. Use of color Doppler flow imaging for preoperative assessment
in fibular osteoseptocutaneous free tissue transfer. Otolaryngol. Head and Neck Surg. 1997; 117:
660-663.
12 Rosson GD, Singh NK. Devascularizing complications of free fibula harvest: Peronea arteria magna.
J Reconstr Mircosurg 2005; 21: 533-538.
13 Mun GH, Jeon BJ. An efficient method to increase specificity of acoustic Doppler sonography for
planning perforator flap: perforator compression test. Plast Reconstr Surg 2006;118:296-297.
14 Winsor T. Influence of arterial disease on the systolic blood pressure gradients of the extremity. Am J
Med Sci 1950;220:117-126.
15 Tsukino A, Kurachi K, Inamiya T, et al. Preoperative color Doppler assessment in planning of
anterolateral thigh flaps. Plast Reconstr Surg 2004;113:241-246.
9
10
Chapter
2
Perforatorflaps – the evolution of a
reconstructive surgical technique
Klein S
Hage JJ
de Weerd L
Translated from Ned Tijdschr Geneeskd 2005; 149: 2392-2398
11
12
Summary
Trauma, oncological resections and pressure sores can cause major soft tissue defects. The evolution
of cutaneous, myocutaneous, and fasciocutaneous flaps gives the possibility for the reconstruction of
contour and, in many cases, function. This evolution was closely related to the increased understanding
of the vascular anatomy of the skin and subcutis and has led to the development of perforator flaps.
A free perforator flap is a (large) flap of skin that survives on a single artery with concomitant vein(s) that
perforates the muscle and/or fascia, referred to in brief as a ‘perforator’. This has its origin in a larger
vascular pedicle that runs in a deeper plane. The pedicle of such a perforator flap can be lengthened by
dissecting the perforator down to the main vessels running underneath the muscle. Moreover, the larger
diameter of these vessels facilitates the creation of a vascular anastomosis in the receptor area.
By the use of such perforator flaps, proper innervation and a good blood supply to the flap can be
combined with less morbidity at the donor site. Important possibilities include the filling of a dorsal
(pressure sore) defect and breast reconstruction.
Introduction
Innovation is the essence of plastic surgery. Over the years plastic surgeons have introduced
techniques that are now also applied by other surgical specialists. Before a new technique can be
applied routinely and more widely, other doctors are already confronted with the results. Such a
confrontation often causes questions about the new technique and other possibilities to apply it. At
present there is another development in the field of plastic surgery that needs explanation. There is an
increase in so-called perforator flaps, which are used for various reconstructive operations. In this article
we will describe the evolution of plastic surgical transplantation techniques that have resulted in the
development of perforator flaps. Then the anatomical and surgical principles of perforator flaps will be
illustrated with two examples.
Evolution of the pedicled and free tissue transposition
Pedicled cutaneous flaps. Tissue defects that have developed as a result of trauma, oncological
resection or pressure sores can be covered and closed by means of tissue transposition. The choice
and type of tissue transposition is largely depending on the nature, extent, the status and the
localization of the defect (1). Besides this the general health of the patient and the preference of patient
and plastic surgeon play a role. The oldest and most simple solution for small skin and subcutaneous
defects is the transposition of a pedicled local skin flap. Hereby the adjacent, healthy skin and subcutis
will be rotated on one pedicle into the defect or like a visor connected to two pedicles, moved towards
the defect.
If up to the seventies of the last century, sufficient healthy tissue was not available next to the defect for
transposition, the wound was usually closed with the pedicled transposition of a flap from the same
anatomical region (a so called regional flap) or by a staged transplantation of flaps from elsewhere on
the body. For this last technique the flaps would remain connected to the donor site with one pedicle for
its vascularization, while the most distal part of the flap was stitched into the area that needed
reconstruction. The intermediate part was closed into a tube, because it took weeks or months before
the distal part of the flap had sufficiently grown into the receptor area. Only after there was
neovascularization of this distal part, this so-called tubed flap could be separated from its donor site. By
13
repeating this principle several times with a tubed flap it was even possible to use healthy tissue from
areas elsewhere on the body.(2)
Initially the vascular anatomy of this type of pedicled skin flaps was defined insufficiently and they
remained vascularized at random. This often resulted in circulation problems in the flap. To limit the risk
of necrosis of the most distal part of such “at random” flaps, the plastic surgeon was restricted to a ratio
between length and width of maximum 1.5 to 1 (figure 1a) (1,3) Increase of anatomical knowledge about
the vascularization of the skin, resulted in the development of skin flaps in which a pedicle followed an
axial pattern, and consisted of one artery and a vein (the so-called ‘axial pattern’ or arterial flaps) (2,4).
Due to a more reliable circulation it became possible to realize a much bigger ratio between length and
width (see figure 1b)(5,6). Further modification of the axial flap consequently resulted into the
development of the pedicled “island” flap (see figure 1c).(3) For such island flaps the rotation range of
the skin island is enlarged by dissecting the axial pedicle from its origin up to the edge of the skin island.
Figure 1. Evolution of an ‘at random’ flap to an ‘axial pattern’ and ‘island’ flap. (a) Originally no attention
was paid to the pattern of vascularization of pedicled skin flaps, and potential vascular pedicles were
unconsciously cut. For that reason the length-width ratio remained restricted to 1.5 to 1. (b) Then the
‘axial pattern’ or ‘arterial’ skin flaps were developed. By including a vessel pedicle in the centre of the
flap it became possible to realize a greater ratio between length and width. (c) In an ‘island’ flap the
rotation range was increased by dissecting the pedicle from its origin to the edge of the skin island.
Pedicled myocutaneous flaps. At the end of the seventies of the last century the increased interest in
the vascular anatomy of the skin resulted also in the development of many myocutaneous flaps. (7,8).
Hereby the muscle with the subcutaneous tissue and skin that covers the muscle on top of it is
dissected (figure 2a). Then the so composed flap can be transposed. By using this technique it was
often possible to transpose a bigger skin island than with the technique of the island flap. Consequently
it was initially hypothesized that the better vascularization was caused by including the muscle. Besides
the increased reliability of the blood circulation the bigger volume of the flap is an advantage for
reconstruction of big defects. This bigger volume can however also lead to a functional or esthetical
problem in the receptor area. On top of this, annoying muscle contractions can take place if the muscle
is not denervated during the transposition. Furthermore it is difficult to predict the degree of atrophy of
the denervated muscle and therefore the esthetic result of the transposition. For these reasons further
development of skin transpositions was strived for.
14
Pedicled fasciocutaneous flaps. From the anatomical studies it was also concluded that the
circulation of many skin areas is not per se depending on the circulation of the underlying muscle, but
more so on the vascular plexus which is superficial compared to the fascia over this muscle. This plexus
is often fed by a vascular pedicle that runs over the muscle from deeper lying vessels to the subcutis
(1,9)
The skin and subcutis can therefore be mobilized together with the muscle fascia and a vascular pedicle
as a so-called fascio-cutaneous flap. If the pedicle of such a fascio cutaneous flap runs in one of the
intermuscular septa from deep to superficial it is called a “septo-cutaneous” flap (see figure 2b) (1).
Such flaps were developed especially during the eighties of the previous century. (9,10) By using the
fascio-cutaneous or septo-cutaneous flaps the morbidity of the donor area remains limited because
there is hardly any loss of function of the muscle. Moreover these flaps are more pliable and often
slimmer than the myocutaneous flaps and therefore easier in use (9)
Revascularized free flaps. If the defect however is located in an area where it is preferable not to use
local or regional flaps, the tissue needed for covering the defect should be removed from elsewhere on
the body.
Since the end of the seventies of the last century it is no longer necessary to use tubed flaps. At that
time the developments of the operation microscope and microsurgical instruments and the ever
increasing knowledge about the vascular anatomy of skin and muscles all came together. (11,12). This
created the possibility to cut the pedicle of a growing number of axial flaps and to revascularize the flaps
with microanastomoses at the reconstructive site. Moreover these ‘free’ flaps can be composed
according to need with skin, fat, fascia, muscles, bone and other tissues. Through this free
transplantation it not only became possible to choose the flap with the best properties suited for the area
that needs reconstruction, but also to choose that flap that causes the least morbidity and loss in
function in the donor area.
Muscle perforator flaps. Developments went even further to combine the reliable circulation of the
myocutaneous flap with the lesser donor morbidity of the fasciocutaneous flap.
At the end of the eighties of the last century it became clear that the circulation of the subcutis and the
skin of a myocutaneous flap is not per se depending on the inclusion of the muscle in the flap. (13,14).
A big skin flap can in some cases also survive on only one run off vessel, that perforates the muscle.
This pedicle, briefly called ‘perforator’ springs from a bigger pedicle that runs underneath the muscle
(see figure 2c)(15)
By dissecting the perforator from the muscle in continuity with the vessels lying underneath the muscle,
the pedicle of such a perforator flap can be extended considerably. Moreover the inclusion of those
vessels implies that the diameter of the vessels becomes bigger, which makes it easier to realize a
vessel anastomoses in the receptor area.
15
a)
b)
c)
Figure 2. Evolution from myocutaneous flap to perforator flap. (a) For the use of a myocutaneous flap a
muscle together with the fascia, subcutaneous fat and the skin, all attached on top of that muscle, is
dissected and, pedicled on a vessel of the muscle, transplanted as a whole (NB different layers were
partly drawn separately to create a better understanding of the vascularization). (b) The circulation of a
‘septo-cutaneous’ flap does not depend on the inclusion of the underlying muscle, but rather on the
vascular plexus, which is more superficial in relation to the fascie over that muscle. This plexus is in
connection with a deeper vascular system that runs in between two muscles from deep to superficial. (c)
A big skin flap may in some cases also survive on one transmusclar pedicle that runs off a bigger
pedicle underneath the muscle. This principle lead to the development of the perforator flap (see figure
3)
Seven examples of the nomenclature of perforator flaps, based on the vessels from where the
perforating pedicle originates (15)
Abbreviation
Full name of flap
Origin of the perforating
pedicle
DIEP
‘deep inferior epigastric perforator’
deep inferior epigastric artery and
vein
SGAP
‘superior gluteal artery perforator’
superior gluteal artery and vein
IGAP
‘inferior gluteal artery perforator’
inferior gluteal artery and vein
IMAP
‘internal mammary artery
internal mammary artery and vein
perforator’
LAP
‘lumbal artery perforator’
lumbar artery and vein
TDAP
‘thoracodorsal artery perforator’
thoracodorsal artery and vein
ALT
‘anterolateral thigh perforator’
descending branch of lateral
circumflex femoral artery and vein
16
a)
b)
c)
d)
Figure 3. Clinical example of a ‘lumbar artery perforator’ flap (LAP) (a) The doppler signal of the
paravertebral perforators of the lumbal artery is marked preoperatively (b) After dissecting the
paravertebral pedicle, the flap is ready for transposition to the sacral pressure sore defect. (c) Detail of
the pedicle and the adjacent sensory nerve (d) Result of the bilateral transposition of the LAP flaps after
some weeks. The donor defects in both sides are closed primarily.
Pre- operatively the location of the perforating vessels can usually be traced with a pencil Doppler, but
the manipulation of the fragile vessels can be quite difficult peroperatively. A prerequisite to be able to
apply the technique of perforator flaps is that dissection is performed using microsurgical techniques.
The advantage of this type of flaps is that preserving the muscle will cause little loss of function in the
donor area and allows for a faster postoperative recovery of the patient. Together with the quality,
consistency and the quantity of the perforator flaps this advantage can outweigh the technical difficulty.
Examples of perforator flaps.
‘Lumbal artery perforator’(LAP) flap. The LAP flap consists of the subcutis and skin of the left or right
half of the back and is vascularized by one of the paravertebral perforators, originating from the. lumbar
Artery and vein and runs through the m.erector spinae or in between the erector spinae muscle and the
quadratus lumborum muscle (16). These perforators are preoperatively marked on the skin with the help
17
of Doppler examination (figure 3a). The paralumbar skin and subcutis are incised in the shape of a
transverse or longitudinal oriented flap and dissected from the underlying muscles. As soon as the
perforating pedicle has been recognized it is dissected out between the muscle fibres. The flap is now
ready for local transposition, for example for covering a radiation ulcer or a pressure sore at the level of
the sacrum (see figure 3b) (17)
Hereby keeping the normal sensibility of the flap is a great advantage (see figure 3c). If however the flap
is used as a free flap, the pedicle will be cut at the level of its origin on the vasa lumbales and taken out
of the muscle, to be anastomozed at the recipient area. (18)
The LAP flap can be used for many different reconstructions and is preferably used for the
reconstruction of dorsal defects, because the patient does not need to be turned on the operating table
during the operation. The donor defect on the back can easily be closed and hardly ever leads to
morbidity or a problematic scar. (see fig 3d)
‘Deep inferior epigastric perforator (DIEP) flap. The DIEP-flap consists of the subcutis and cutis of
the lower abdomen and is vascularized by the perforators of the deep inferior epigastric artery and vein,
running through and underneath the rectus abdominis muscle. These perforators can usually be marked
on the skin pre-operatively with aid of Echo-Doppler examination (figure 4a). During the operation first
the skin and subcutis of the lower abdomen are incised elliptical and dissected from the fascia of the
external oblique muscles and the rectus sheath. One to three appropriate perforators, crossing from the
rectus abdominis muscle into the subcutis, are chosen and the anterior rectus sheath is opened around
these perforators. By splitting the muscle fibers the perforators are dissected down to the deep inferior
epigastric vessels. In this step the motor innervation of the muscle is preserved as much as possible.
After the deep inferior epigastric vessels are ligated and cut cranially of the perforator, the pedicle is
dissected caudally till its origin from the external iliac vessels (see figure 4b). If the flap is used as a free
flap, the pedicle is ligated cut at their origin and taken out through the rectus muscle. The flap is now
ready for revascularization by means of microsurgical anastomoses in the recipient area.(19) The flap
can be reinnervated by microsurgical neuroraphy of the sensory nerve branches to the sensory nerves
in the receptor area. After the flap harvest the anterior rectus sheath is closed over the muscle and the
skin is closed like in a standard abdominoplasty.
Although the first clinical applications of the DIEP flap were for the reconstruction of a groin defect and
an oral defect, nowadays this flap is mainly used for breast reconstructions (14,20) Consistency and
mass of the fat tissue of the lower abdomen are very similar to the breast tissue. The epigastric vessels
are often anastomosed to the internal mammary or the thoracodorsal vessels (see figure 4c)(19). In
comparison to the ‘transverse rectus abdominis myocutaneous flap’ (TRAM) flap, in which a part of the
de m.rectus abdominis is also harvested with the flap, the use of the DIEP flap leads less often to
abdominal herniations and less loss of muscle power in the donor area (see figure 4d). Furthermore the
DIEP flap leads to less morbidity and a shorter hospital admission (21)
Conclusion
Today there is a big and diverse range of pedicle and free tissue transpositions available for
reconstructive purposes. For an optimal recovery of shape and function the most suitable flap can be
chosen for reconstruction of defects. By using the so-called perforator flaps, a reliable vascularization
18
and the possibility of sensory reinnervation are combined with less morbidity and less loss of function in
the donor area. This application demands microsurgical expertise and in surgical technical sense there
is need for more additional experience with perforator flaps. After this development it seems only a
matter of time before perforator flaps can be applied widely and routinely in other surgical area.
a)
b)
c)
d)
Figure 4. Clinical example of a ‘deep inferior epigastric perforator’ (DIEP)-flap. (a) Status after a left
lateral skin saving mastectomy in combination with direct implantation of a temporary prosthesis. The
implant will be replaced with autologous abdominal tissue. The perforators that run from the inferior
epigastric artery and vein through the rectus abdominis muscle to the subcutaneous and cutaneous
tissue of the lower abdomen are marked preoperatively.
(b) After dissecting the pedicle from the
underlying rectus abdominis muscle up to the origin of the external iliacal vessels the flap will be ready
for transplantation. (c) Results of a breast reconstruction by means of a ‘free’ microsurgical
transplantation of a DIEP flap after the left lateral skin saving mastectomy. (d) The scar from side to
side as a consequence of closing the abdominal donor defect.
19
References
1 Mathes SJ, Nahai F. Reconstructive surgery. New York: Churchill Livingstone; 1997. p. 9-36.
2 Webster JP. Thoracoepigastric tubed pedicles. Surg Clin North Am. 1937;17:145-84.
3 Daniel RK, Kerrigan CL. Principles and physiology of skin flap surgery. In: McCarthy JG, editor.
Plastic surgery. New York: Saunders; 1990. p. 275-328.
4 McGregor IA, Morgan G. Axial and random pattern flaps. Br J Plast Surg. 1973;26:202-13.
5 Milton SH. Pedicled skin-flaps: the fallacy of the length : width ratio. Br J Surg. 1970;57:502-8.
6 Daniel RK, Kerrigan CL. Skin flaps: an anatomical and hemodynamic approach. Clin Plast Surg.
1979;6:181-200.
7 Orticochea M. The musculo-cutaneous flap method: an immediate and heroic substitute for the
method of delay. Br J Plast Surg. 1972; 25:106-10.
8 McCraw JB. The recent history of myocutaneous flaps. Clin Plast Surg.1980;7:3-7.
9 Tolhurst DE, Haeseker B, Zeeman RJ. Fasciocutaneous flaps. Chir Plastica. 1982;7:11-21.
10 Ponten B. The fasciocutaneous flap: its use in soft tissue defects of the lower leg. Br J Plast Surg.
1981;34:215-20.
11 Daniel RK, Taylor GI. Distant transfer of an island flap by microvascular anastomoses. A clinical
technique. Plast Reconstr Surg. 1973; 52:111-7.
12 Haeseker B. Microchirurgie, de ‘kleine’ chirurgische revolutie uit de medische geschiedenis van de
afgelopen eeuw. Ned Tijdschr Geneeskd. 1999;143:858-64.
13 Kroll SS, Rosenfield L. Perforator-based flaps for low posterior midline defects. Plast Reconstr Surg.
1988;81:561-6.
14 Koshima I, Soeda S. Inferior epigastric artery skin flap without rectus abdominal muscle. Br J Plast
Surg. 1989;42:645-8.
15 Blondeel PN, Landuyt KHI van, Monstrey SJ, Hamdi M, Matton GE, Allen RJ, et al. The ‘Gent’
consensus
on
perforator
flap
terminology:
preliminary
definitions.
Plast
Reconstr
Surg.
2003;112:1378-83.
16 Kato H, Hasegawa M, Takada T, Torii S. The lumbar artery perforator based island flap: anatomical
study and case reports. Br J Plast Surg. 1999;52:541-6.
17 Weerd L de, Weum S. The butterfly design: coverage of a large sacral defect with two pedicled
lumbar artery perforator flaps. Br J Plast Surg. 2002;55:251-3.
18 Weerd L de, Elvenes OP, Strandenes E, Weum S. Autologous breast reconstruction with a free
lumbar artery perforator flaps. Br J Plast Surg. 2003;56:180-3.
19 Blondeel PN, Boeckx WD. Refinements in free flap breast reconstruction: the free bilateral deep
inferior epigastric perforator flap anastomosed to the internal mammary artery. Br J Plast Surg. 1994;
47:495-501.
20 Jonasse Y, Werker PMN. Vijftig jaar plastische chirurgie in Nederland. IX. Reconstructieve chirurgie
geïllustreerd aan de mammareconstructie. Ned Tijdschr Geneeskd. 2000;144:1152-6.
21 Blondeel PN, Vanderstraeten GG, Monstrey SJ, Landuyt K van, Tonnard P, Lysens R, et al. The
donor site morbidity of free DIEP flaps and free TRAM flaps for breast reconstruction. Br J Plast
Surg. 1997;50: 322-30.
20
Chapter
3
Donor site necrosis following
fibula free flap transplantation
Klein S
Hage JJ
Woerdeman LAE
Microsurgery 2005; 25: 538-542
21
22
Summary
The free fibula flap is the microsurgeon’s workhorse for the reconstruction of osseous or
osteocutaneous defects. Donor-site morbidity of this flap is reported to occur infrequently, and is
generally considered minor and transient. We present the case histories of three patients with necrosis
at the fibula flap donor site to stress the risks and explain the possible mechanisms of such severe
complications. The small risk of debilitating donor-site necrosis should be considered and discussed
preoperatively with the patient.
Introduction
The fibula free flap has become the microsurgeon’s workhorse for the reconstruction of osseus or
osteocutaneous defects of the head and neck region,(1-3) the extremities,(3-5) and the trunk.(3,6) Flap
survival and receptor site function after fibula flap transplantation have been evaluated extensively, but
donor-site morbidity has been assessed less often. Donor-site morbidity includes wound healing
disturbance and cellulitis,(1-3) transient peroneal sensory loss or cold intolerance,(1-4,7,8) motor
weakness of lower leg muscles,(1,4-9) pain or occasional cramp,(1-4,7-9) impingement of the peroneal
muscle,(10) edema after prolonged standing,(2-4,8) and ugly scarring.(4) Even though valgus deformity,
instability, or stiffness of the ankle,(1,2,4,7-9,11) and walking restriction or gait disturbances,(2,3,5,6)
can be considered more serious donor-site complications, most of the complications mentioned above
are generally rated as acceptable by patients.(1,2,4,6,8,9)
Major donor-site complications with extensive wound breakdown or necrosis are generally considered to
occur infrequently after fibula free-flap transfer. To date, only five cases of such debilitating
complications due to necrosis have been reported to lead to deep necrotectomy. Among these five were
two cases of acute compartment syndrome,(3,12) two cases of late necrosis,(13) and one case of
necrotizing infection.(4) To stress that such donor-site necrosis can lead to severe postoperative
disability, we present three more patients who suffered from extensive necrosis of the lower leg, and we
discuss how to prevent these complications.
23
Case reports
Patient 1
A 36-year-old woman underwent commando resection and supraomohyoidal lymphadenectomy for
T4N0M0 squamous-cell carcinoma of the floor of the mouth and left hemimandible. Her medical history
included systemic lupus erythematosis, tobacco abuse, and a mild myocardial infarct sustained 2
months prior to surgery. Because preoperative angiography showed no acquired or congenital
anomalies of the crural arteries, the defect was reconstructed using a fibula free flap, including a
significant part of the lateral hemisoleus muscle and a skin segment of 6 x 14 cm (Fig.1). Per our
routine, no tourniquet was used during surgery. The donor defect was closed primarily over a suction
drain after loose approximation of the muscles of the lateral compartment. No attempt was made to
approximate the lateral fascia. The lower leg and foot were splinted on a dorsal cast to prevent pes
equinus, and the patient was kept on subcutaneous heparin prophylaxis. No pressure or tensor
bandage was applied to the leg, but the patient was required to keep the leg elevated until mobilization,
one week postoperatively.
On the first postoperative day, however, the patient complained of pain in the donor leg and sensory
loss of the dorsum of the foot. On physical examination, there was functional loss of the extensor
digitorum muscle, decreased capillary refill of the foot, and lack of a Dopptone signal over the dorsal
tibial artery. These symptoms being indicative of compartment syndrome, the cast and skin sutures
were removed to reduce tension in the calf.(14,15) Angiography subsequently showed arterial filling of
both tibial arteries (Fig.2). Still, necrosis of the extensor digitorum muscle and parts of both peroneal
and long flexor hallucis muscles occurred, and repeated necrotectomy had to be performed surgically,
on postoperative days 11 and 22. After this, the crural wound healed except for its most distal part
(Fig.3). Six months after surgery, part of the distal fibula remnant still protruded from the wound, and Xray examination could not exclude osteomyelitis. Consequently, this part was resected, and the
remaining bone was covered during a fourth and final surgical procedure. When last seen at 44 months
of follow-up, the patient was free of recurrence, and the neomandible was functional. However, her gait
and walking distance were severely impaired.
Figure 1. Design of fibula free flap used on patient 1. Flap included significant part of lateral hemisoleus
muscle, and skin segment measuring 6 x 14 cm.
24
Figure 2. Postoperative angiography of patient 1 showed arterial filling of both tibial arteries.
25
Figure 3. Six months after fibula free-flap transplantation in patient 1, distal part of crural wound had not
closed, and protruding part of distal fibula remnant was resected.
Patient 2
A 74-year-old woman with a history of alcohol and tobacco abuse underwent commando resection and
modified radical lymphadenectomy for a T3N0M0 squamous-cell carcinoma of the floor of the mouth
extending to the left hemimandible. As the preoperative angiography showed no vascular anomalies,
the defect was reconstructed using an osteocutaneous free fibula flap with a skin paddle of 5 x 12 cm.
As no tourniquet was used intraoperatively, it could be confirmed that the manipulated muscles
remained well-vascularized throughout the entire procedure. Following approximation of the donor-site
muscles, the lower leg skin defect was covered with a split-thickness skin graft to avoid undue pressure,
and the lower leg and foot were splinted to prevent pes equinus. Again, no tensor bandage was applied,
and the patient was required to keep the leg elevated until mobilization, 1 week postoperatively.
The postoperative course was uneventful up to day 15, when edema occurred in the donor leg. Even
though wound cultures remained negative for necrotizing species, subsequent necrosis of the skin graft
and surrounding skin and subcutaneous tissue necessitated necrotectomies on postoperative days 23
and 37 (Fig.4). The resulting skin defect measured 10 x 25 cm, and skin grafts were applied
secondarily. Using a vacuum-assisted closure system, the wound completely healed by 10 weeks after
primary surgery. At 24 months of follow-up, the patient was free of recurrence, and her neomandible
functioned adequately. No remaining functional loss of the donor leg was observed.
26
Figure 4. In patient 2, skin defect resulting from initial surgery, postoperative ‘‘pseudo’’-compartment
syndrome, and subsequent necrotectomy measured 10 x 25 cm.
Patient 3
A 56-year-old diabetic man underwent segmental mandibulectomy for residual ameloblastoma, after
local resections had twice been attempted elsewhere. Because the preoperative angiography showed
no atherosclerosis or vascular anomalies, the left hemimandible and gingiva defect was reconstructed
with a fibula osteocutaneous free flap. The skin paddle measured 6 x 11 cm, and was raised on one
mid-lower leg muscular perforator, as no adequate septocutanous vessels were found. This perforator
had a long intramuscular course and a peroneal origin near the crural trifurcation, and several muscular
branches originating from the peroneal artery had to be clipped to allow for its dissection. Again, no
tourniquet was used, allowing for proper hemostasis and assessment of the muscles that were
manipulated during flap dissection. Following microsurgical flap transfer, the lateral lower leg muscles
were loosely approximated, and no attempt was made to close the fascia. The skin was partly closed
primarily, and partly by use of a skin graft. A splint was applied to prevent pes equinus, and the patient
was kept in bed and on heparin prophylaxis for 1 week. No tensor bandage was applied.
The postoperative course was uneventful until the patient became febrile and complained of pain in the
donor leg on postoperative day 11. Wound culture was positive for beta-hemolytic Streptococcus, and
antibiotic therapy was initiated. At surgical exploration, the long peroneal muscle was found to have
necrosed. Following necrotectomy, it took 10 weeks of extensive conservative treatment, including
application of a vacuum-assisted closure system, for the wound to heal. When last seen at 24 months of
follow-up, the patient was free of recurrence of osteoblastoma, and his neomandible was functional, but
exorotation and plantar flexion of the foot were compromised.
27
Discussion
Fibula flap donor-site morbidity and its consequences for daily life function are generally considered to
be rare and minor. Still, we observed severe donor-site necrosis in 3 of 74 patients in whom a fibula flap
transplantation was performed in the Netherlands Cancer Institute from January 1985 until December
2003. The mechanism of necrosis may differ. Our first patient represented an obvious case of acute
compartment syndrome, whereas a delayed ‘‘pseudo’’-compartment syndrome and a necrotizing
infection were the probable causes in our second and third patient, respectively.
Compartment syndrome is allegedly a rare consequence of free fibula flaps, and has been reported only
twice to date.(3,12) This syndrome occurs when intra-compartment pressure builds up in cases where
the fascia surrounding the compartment is intact. Obviously, this is no longer true in cases where the
lateral fascia is not closed after fibula flap transplantation. In these cases, closure of the skin under
excessive tension may be a cause of compartment syndrome, and there- fore it has been advised to
restrict primary closure of the donor site to skin paddles of maximal 6-cm width.(13,16) Alternatively,
compartment pressure may build up as a result of edema, even in cases where skin closure was
performed without undue tension.(13) In both instances, the skin acts analogous to the fascia, and
complete release of the wound may be expected to alleviate the ‘‘skin compartment syndrome.’’
In our first patient, the width of the skin paddle was restricted to the 6-cm maximum. Moreover, part of
the lateral hemisoleus muscle was harvested along with the flap, and this may be expected to reduce
the circumference of the leg and hence the closure tension. Even after adequate response to symptoms
presenting on the first postoperative day,(14,15,17) necrosis of some muscles of the anterior, lateral,
and deep compartments occurred, and hence we feel that avoidance of primary skin closure might have
prevented the compartment syndrome.(12) More careful patient selection might not have prevented the
occurrence of compartment syndrome in this patient. Such occurrence has been associated with age
under 35 years, male sex, diabetes mellitus, obesity, hypertension, traumatic rather than surgical injury,
bleeding disorders or the intake of anticoagulant drugs, low preoperative ankle-arm index, extended
duration of surgery or surgery in the lithotomy position, intraoperative hypothermia or hypotension, and
the postoperative use of regional nerve blockades.(18-22) Furthermore, patients suffering from
peripheral vascular disease were reported to run a higher risk of compartment syndrome when a
tourniquet was used during surgery.(18,20) Except for the duration of surgery, none of these
predisposing factors applied to this patient.
As for our second patient, we feel that the tension that developed in the lower leg after some delay was
too high, notwithstanding the use of a skin graft. The physical mechanism underlying the development
of the delayed necrosis seems to have been a vicious cycle of excessive tension caused by edema
leading to tissue damage and increased capillary permeability.(13) The resulting intercellular pressure
and soft-tissue edema led to more tissue damage and an increase of venous, lymphatic, and arterial
vessel compression, thus perpetuating the cycle. This cycle of gradually increasing ischemia has been
termed ‘‘pseudo’’-compartment syndrome, and may explain why the symptoms in this patient only
presented after 15 days.(13)
Paradoxically, fibulectomy used to be the treatment modality for lower leg compartment
syndrome.(14,17) Still, fibulectomy may not prevent occurrence of the complication. Rather than closing
the wound primarily, we currently apply a skin graft, irrespective of the width of the skin paddle included
in the flap. Often, this graft can safely be resected after the period of edema has passed. Still, a delayed
28
‘‘pseudo’’-compartment syndrome may develop, even in cases where a skin graft was used.(13) To
minimize the damage secondary to (‘‘pseudo’’-) compartment syndrome, we advise the surgeon to be
postoperatively alert to clinical symptoms such as pain that seems disproportionate to the clinical
situation, pain on passive stretch of the muscles, or decreased sensibility in the distribution of the
nerves running through the corresponding compartment.
The necrosis at the donor site in our third patient was likely caused by the combined occurrence of betahemolytic Streptococcus and long-term muscle ischemia as a result of the obligatory clipping of multiple
vascular branches to the long peroneal muscle. Additional pre- disposing risk factors for such local
ischemia include atherosclerosis or vasculitis, endothelial damage, abnormal blood constituents due to
coagulation, fibrinolytic changes or hyperviscosity, and decreased blood flow as a result of
immobilization or vascular compression.(23) The risk factors for atherosclerosis are age over 65 years,
diabetes mellitus, obesity, tobacco abuse, hypertension, hypercholesterolemia, and a sedentary
lifestyle.(24) Our patient was a smoker, suffered from diabetes mellitus, and was expected to be
immobile for some days after surgery, but the need for extensive devascularization of the peroneal
muscle could not have been foreseen preoperatively. As was true for our other two patients, more
careful patient selection might not have prevented the donor-site necrosis in this patient.
We conclude that donor-site morbidity after fibula transfer is generally minor, but there is a small risk of
donor-site necrosis as a result of acute compartment syndrome, delayed ‘‘pseudo’’-compartment
syndrome, or necrotizing infection. This risk and its predisposing factors should be preoperatively
discussed with the patient, as donor-site necrosis may result in debilitating long-term morbidity.
Perioperatively, the risk of such necrosis should be minimized by closure of the wound with a skin graft,
and postoperatively, necrosis should be considered upon presentation of any disproportional local
symptom.
29
References
1 Anthony JP, Rawnsley JD, Benhaim P, Ritter EF, Sadowsky SH, Singer MI. Donor leg morbidity and
function after fibula free flap mandibula reconstruction. Plast Reconstr Surg 1995;96:146-152.
2 Hidalgo DA, Rekow A. A review of 60 consecutive fibula free flap mandible reconstructions. Plast
Reconstr Surg 1995;96:585-596.
3 Zimmermann CE, Börner BI, Hasse A, Sieg P. Donor site morbidity after microvascular fibula
transfer. Clin Oral Invest 2001;5:214-219.
4 Shpitzer T, Neligan P, Boyd B, Gullane P, Gur E, Freeman J. Leg morbidity and function following
fibular free flap harvest. Aesthetic Plast Surg 1997;38:460-464.
5 Babovic S, Johnson CH, Finical SJ. Free fibula donor site morbidity: the Mayo experience with 100
consecutive harvests. J Reconstr Microsurg 2000;16:107-110.
6 Youdas JW, Wood MB, Cahalan TD, Chao EYS. A quantitative analysis of donor site morbidity after
vascularized fibula transfer. J Orthop Res 1988;6:621-629.
7 Parker VT. Donor-site morbidity with use of vascularized autogenous fibular grafts. J Bone Joint
Surg [Am] 1996;78:204-211.
8 Papadopulos NA, Schaff J, Bucher H, Groener R, Geishauser M, Biemer E. Donor site morbidity
after harvest of free fibular flaps with an extended skin island. Aesthetic Plast Surg 2002;49:138-144.
9 Gore DR, Gardner GM, Sepic SB, Mollinger LA, Murray MP. Function following partial fibulectomy.
Clin Orthop 1987;220:206-210.
10 Meagher PJ, Morrison WA. Free fibula flap donor site morbidity: case report and review of the
literature. J Reconstr Microsurg 2002;18:465-467.
11 Wiltse LL. Valgus deformity of the ankle - a sequel to acquired or congenital abnormalities of the
fibula. J Bone Joint Surg [Am] 1972;54:595-606.
12 Saleem M, Hashim F, Babu Manohar M. Compartment syndrome in a free fibula osteocutaneous
flap donor site. Br J Plast Surg 1998;51:405-407.
13 Shindo M, Fong BP, Funk GF, Karnell LH. The fibula osteocutaneous flap in head and neck
reconstruction - a critical evaluation of donor site morbidity. Arch Otolaryngol Head Neck Surg
2000;126:1467-1472.
14 Rollins DL, Bernhard VM, Towne JB. Fasciotomy: an appraisal of controversial issues. Arch Surg
1981;116:1474-1481.
15 Weiner G, Styf J, Nakhostine M, Gershuni DH. Effect of ankle position and a plaster cast on
intramuscular pressure in the human leg. J Bone Joint Surg [Am] 1994;76:1476-1481.
16 Hayden RE. Harvest and clinical application of the fibula flap - editorial review. Curr Opin Otolaryngol
Head Neck Surg 1995;3:257-260.
17 Hyde GL, Peck D, Powell DC. Compartment syndromes. Early diagnosis and a bedside operation.
Am Surg 1983;49:563-568.
18 Ullrich W, Biermann E, Kienzle F, Krier C. Lagerungsschäden in Anästhesie und operativer Medizin.
Anasth Intensivmed Notfallmed Schmertzther 1997;32:4-20.
19 McQueen MM, Gaston P, Court-Brown CM. Acute compartment syndrome. Who is at risk? J Bone
Joint Surg [Br] 2000;82:200-203.
20 Tison C, Perigaud C, Vrignaud S, Capelli M, Lehur PA. Syndrome bilatéral des loges de jambe aprés
chirurgie colorectale en position àdouble équipe. Ann Chir 2002;127:535-538.
30
21 Thonse R, Ashford RU, Williams TI, Harrington P. Differences in attitudes to analgesia in
postoperative limb surgery put patients at risk of compartment syndrome. Injury 2004;35:290-295
22 Modrall JG, Sadjadi J, Ali AT, Anthony T, Welborn MB 3rd, Valentine RJ, Hynan LS, Clagett GP.
Deep vein harvest: Predicting need for fasciotomy. J Vasc Surg 2004;39:387-394.
23 Cotton DWK. Ischaemia, infarction and shock. In: Underwood JCE, editor. General and systematic
pathology. 2nd ed. New York: Churchill Livingstone; 1996. p 165-177.
24 Hooi JD, Kester ADM, Stoffers HEJH, Overdijk MM, van Ree JW, Knottnerus JA. Incidence of and
risk factors for asymptomatic peripheral arterial occlusive disease: a longitudinal study. Am J
Epidemiol 2001;153:666-672.
31
32
Chapter
4
Review:
An overview of methods for vascular
mapping in the planning of free flaps
Smit JM
Klein S
Werker PMN
J Plast Reconstr Aesthet Surg 2010; 63: 674-682
33
34
Summary
Introduction: The aim of this overview is to describe the various methods for vascular mapping of flaps
together with their advantages and drawbacks.
Materials and methods: The PubMed database was used. Relevant search terms included ‘flap’ in
combination with ‘hand-held Doppler’ (HHD), ‘colour duplex sonography’ (CDS), ‘digital subtraction
angiography’ (DSA), ‘computed tomography angiography’ (CTA) and ‘magneticresonance angiography’
(MRA). All studies found between January 2000 and January 2010 was evaluated.
Results: A total of 72 articles were found. Of these, 62 were usable for this overview. Recommendations
could not be found for all types of flaps. Therefore, no uniform guidelines can be provided; some
findings are, however, unequivocal. In general, HHD is cheap and easy to use, but relatively unreliable
in determining the exact site of emergence at fascia level of perforators. CTA and MRA provide the best
three-dimensional images. CTA offers more detailed images, MRA has the advantage however of not
using radiation. CDS can be of value to offer information about the amount of flow in vessels or in cases
in which CTA or MRA are contraindicated. DSA appears to be fading out slowly.
Conclusion: CTA and MRA are currently the best methods available to map the vasculature of donor
sites of perforator flaps with variable anatomy such as anterolateral thigh (ALT) and deep inferior
epigastric perforator (DIEP). In flaps with standard anatomy and superficial vasculature, HHD or no
mapping at all remains the method of choice.
Introduction
Reconstructive surgery has seen great development since the early 1960s, when the concept of axial
vessels became mainstay.(1-10) The first generation of axial pattern flaps was based on well-known
vessels from the anatomy book, such as the radial artery for the radial forearm flap and the
thoracodorsal vessels for the latissimus dorsi flap. The harvest of these flaps, although in that time
revolutionary, is nowadays looked upon as relatively straightforward, due to their constant anatomy.
Only in instances where previous surgery or trauma might have damaged the vascular pedicle, further
investigation of the vasculature is deemed indicated. In the past 20 years, enormous progress has been
made in flap design and more and more flaps are based on perforating vessels that branch off and are
traced back to well-known vessels, thereby limiting donor-site morbidity. The exact location of
perforators, however, varies significantly, and preoperative vascular mapping has been introduced to
help identify the dominant perforator and its course and, as such, speed up flap harvest.
A variety of methods is available for this purpose, the most commonly used being hand-held Doppler
(HHD),(2-4) colour duplex sonography (CDS),(5,11-13) digital subtraction angiography (DSA),(14,15)
computed
tomography
angiography
(CTA)(6-8,16)
and
magnetic
resonance
angiography
(MRA).(8,9,17) HHD was already reported in the planning of flaps in 1975, when it was described for the
localization of the donor and recipient vessels in facial reconstructions.(18) Around the same period,
angiography, later replaced by DSA, was introduced to assess the vascular anatomy of flaps, mainly in
the lower extremities.(19,20) The use of CDS in the planning of flaps was first described in the 1980s
and became a common mapping method in the 1990s.(21,22) In the last decade, major progress has
been made in the applicability of, especially, CTA and MRA. The purpose of this overview is to describe
the various vascular mapping systems in detail, together with their advantages and drawbacks to assist
beginning micro-surgeons in their planning of free flaps.
35
Although other mapping methods such as indocyanine green(23) and near infrared imaging(24) exist,
these can only be used for intra-operative flap design and are therefore beyond the scope of this
overview.
Materials and methods
A literature search was conducted, using the PubMed database. The following search terms were used:
‘flap’ in combination with ‘hand-held Doppler’, ‘colour duplex sonography’, ‘digital subtraction
angiography’, ‘computed tomography angiography’ and ‘magnetic resonance angiography’. All studies
found between January 2000 and January 2010 was evaluated. Only studies written in the English
language were included. Manual cross-referencing was also performed. For historical and technical
backgrounds, reports prior to 2000 were used whenever necessary.
In selecting data regarding the reliability of the mapping method, studies with the highest level of
evidence were preferred over others.
Results
Our query led to 18 studies on HHD, eight on CDS, 10 on DSA, 31 on CTA and 12 on MRA.
Hand-held Doppler sonography (HHD)
A pencil-type Doppler probe registers moving erythrocytes by sending out and detecting reflected
ultrasound. Depending on the depth and the diameter of the vessels to be investigated, various probes
with different frequencies can be used. The two most commonly used frequencies, 8 and 10 MHz, have
a peak sensitivity of only 20 and 15 mm, respectively. Gel must be used at the interface of skin and the
probe to improve ultrasound conduction. When searching for perforators, it has been advised to vary the
amount of pressure applied with the Doppler probe to the skin surface. If the detected sound comes
from a perforator that runs directly towards the HHD, the loudness of the pulsating sound will reduce
with increasing pressure.(25) Reported advantages of the HDD are its non-invasiveness, small size, low
costs, portability and the ease to perform the examination. In addition, there are special probes
available, which can be sterilized, making the technique available intra-operatively to finalize the
planning and check the pulsation of a vessel during the surgical procedure.(2,4,25,26)
The main disadvantage of the most widely used Doppler probe (8 MHz) is, that it only detects vessels to
a depth of 20 mm. This makes the technique less reliable for the detection of the site of emergence of
perforators through the fascia, whenever the thickness of skin and subcutaneous tissue exceeds this
amount.(2) Besides, one can never know for sure what vessel is producing the Doppler signal picked up
by the HHD. Furthermore, this technique does not create a three-dimensional (3D) image of the
vasculature and its surrounding anatomy than can be stored and retrieved later.
The use of HHD has predominantly been reported to locate perforators on the trunk and extremities.(24,28) A relatively new field in reconstructive surgery, in which the HHD is being used, is in free-style
perforator free flaps(27,29-33) and in pedicled perforator flaps.(34-36)
The use of HDD has been reported with variable success: In the study by Yu and Youssef(2) of 2006, in
which they included 100 patients undergoing an anterolateral thigh (ALT) reconstruction, the locations of
HHD signals of an 8-MHz and a 10-MHz probe were compared to the intra-operative findings. The
positive predictive value for the 8-MHz probe in detecting the perforator was 89%, while no false-
36
negative results were found. For the 10-MHz probe, a positive predictive value of 94% and negative
predictive value of 43% were found. The study by Shaw et al.(28) compared HHD to intra-operative
findings in 30 patients undergoing an ALT reconstruction. They found a large underestimate (30%) to an
overestimate (150%) for the HHD findings. In a study of 32 deep inferior epigastric perforator (DIEP)
and eight superior gluteal artery perforator (SGAP) flaps in which HHD findings were compared with
intraoperative findings, a positive predictive value of only 52.4% was found for DIEP and SGAP flaps
combined.(4)
Colour duplex sonography (CDS)
CDS relies on the same working principle as HHD. Blood flow in vessels is detected by the physical
principle of a direct relationship between the recorded Doppler frequency shift and blood-flow velocity.
In addition, different velocities and directions of moving blood streams can be displayed on a screen in
the colour duplex mode. As such, CDS does not only offer information about the internal vessel
diameter and their course, but also depicts the 3D footprint of the perivascular anatomy.(37) As with
HHD, gel must be used at the interface of skin and the probe to improve ultrasound conduction.
Similar to HHD, CDS is non-invasive. An advantage compared with the HHD is its ability to offer more
information about anatomy of the vessel and its perforators in reference to its surrounding tissues, and it
can quantitatively analyze which perforator is the dominant one.(38)
The disadvantage of CDS is, however, that the investigation can only be performed by skilled
personnel, who also have knowledge of free-flap anatomy. In addition, it is less reproducible because of
its real-life dynamics. Another disadvantage in comparison to CTA, MRA and DSA is that CDS - just as
HHD - does not reproduce a 2D or 3D image of the complete vascular anatomy, which can be used by
the surgeon during flap design or flap elevation.(7,8,16)
CDS has been successfully used to preoperatively assess flaps for reconstructions in the head and
neck, trunk(5,11,13,38,40) and extremities.(12,37) Tsukino et al.(37,39) investigated the reliability of
CDS in 10 patients prior to ALT flap harvesting. Comparison of CDS findings to intra-operative findings
showed a concordance of 100%. This was confirmed for DIEP flaps in a report of six cases.11 CDS has
been found to be of value in cases in which the perforating vessels might have been damaged, for
example, after liposuction, as it can give information about the amount of flow in vessels or in cases
where the radiation dose of CTA is undesirable.(11)
37
Digital subtraction angiography (DSA)
In traditional angiography, an iodine-containing contrast medium is administered intra-arterially while Xrays are taken. In DSA, these contrast-enhanced pictures are digitally subtracted from the pre-contrast
mask X-ray to depict the vascular anatomy (Figure 1). This investigation technique generates 2D
images and, therefore, generally has to be performed in two directions.
The reported advantages of DSA include the facts that it gives an image of the intraluminal vascular
anatomy and information about atherosclerotic changes. A disadvantage of DSA is that it is a timeconsuming, invasive technique necessitating the use of iodinate-contrast medium, which may cause
vascular or renal damage as well as allergic reactions.(41)
In addition, there is a radiation dose to be considered.(6,42-45) Another disadvantage is the
vasoconstricting effect of the contrast medium, making exact measurement of the vascular diameter
and the assessment of small-caliber vessels unreliable. Furthermore, the patient has to stay in supine
position after the angiography for several hours, to allow the puncture site to heal. This makes hospital
admission often mandatory and this imaging modality relatively expensive. Finally, there is a 4.5%
chance for the development of false aneurysms at the puncture site.(46)
DSA in free-flap planning is predominantly reported in fibula flaps, where it is essential to be informed
about the continuity of the three lower leg vessels, the level of bifurcation of the tibiofibular trunk and
about arteriosclerotic plaques.(15,42,43) Angiography has been found to provide a more accurate
assessment of the patency of vessels compared with conventional tests such as ankle-arm index and
HHD examination.(15,42,43,47)
The use of DSA has also been reported in the planning of transverse(45) and oblique(44) rectus
abdominis musculocutaneous flaps. It has been reportedly been used during surgery to visualize the
vascular architecture of a flap after its harvest. This can show its perforator and its connection to the
axial flap vessel, which can help the surgeon to safely thin and separate the flap during secondary
procedures.(44,45) This is, however, not a commonly reported technique.
Figure
1
Digital
subtraction
angiography
imaging of the left lower leg, showing a normal
branching pattern of the vessels.
38
Computed tomography angiography (CTA)
CTA combines the use of X-rays with computerized 3D analysis of the images. The number of detector
rows decide how fast a scan can be performed and to what extent details can be revealed. A great
variety of CT scanners and software is currently available, making a comparison of the results of various
studies difficult. The number of multidetector rows used in different studies varies from 4 to 64,(16,48)
enabling the generation of slices of approximately 1 mm or thinner, depending on the CT scanner used.
The actual scan is performed in concert with a high-speed venous contrast-medium injection to enhance
the staining of vessels. A bolus of 80-100 ml of contrast medium is given intravenously at a rate of 4 ml
-1
s . After the scanning, the data need to be processed into maximum intensity projection and 3D
volume-rendered reconstructions. A great variety of software packages are available for this purpose
(e.g., Siemens InSpace,(48) Vitrea version 3.0.1,(7) VoNavix(49) VirSSPA(50) and Virtual Place
21(51)). Based on the software used, 2D pictures in three planes or 3D reconstruction in multiple planes
can be provided. The processing of the images by the radiologist and the preoperative selection of the
right vessels/perforators by the surgeon has been reported to take up to 30 min in perforator
flaps.(16,49)
The advantage CTA offers is that it provides an image with accurate visual details on the caliber and
course of the vessels and their relationships with other anatomic structures (Figures 2 and 3). This
allows surgeons to develop a dissection strategy and opt for a certain perforator prior to surgery, making
the actual dissection safer and swifter.(7,8,16,49,52-54)
The disadvantages of CTA are its radiation dose, which is reported to be 5.6 mSv,(7) and the necessity
to use iodinated contrast medium with its previously listed disadvantages.(7,16) Especially, the
vasospastic action is a serious drawback, because it can make the accurate assessment of smallcalibre vessels difficult.(55) CTA has been predominantly used in the planning of perforator flaps in
breast reconstructions (DIEP and SGAP),(7,8,16,48,49,56,57) and has also been reported to assess the
vasculature of ALT flaps(54,58) and fibula flaps.(6)
Furthermore, it has been described to map the internal mammary artery perforator(59) and the deep
circumflex iliac artery perforator flap(60) in cadaver studies. CTA may also have additive value to
preoperatively assess the recipient site.(6,61,62) However, the spurting test should be used in addition
to confirm blood flow in the recipient site. In the initial reports of Masia et al.(7) and Alonso-Burgos et
al.(16) regarding CTA in DIEP breast reconstructions, a 100% correlation was found between the CTA
findings and the intra-operative findings. This included the location of the perforators, their estimated
size, the course of the pedicle and its relationship with other anatomic structures. This was later
confirmed by other studies.(8,48,56,57) Compared with MRA, the depiction of smaller perforators is
more accurate with CTA (vessels up to 0.3 mm in CTA vs. 1.0 mm in MRA).(63,64) The introduction of
CTA in DIEP in breast reconstructions led to a reduction of operating times of 90-100 min in flaps
previously examined by HHD,(8,57,65) and 76 min in flaps previously mapped by CDS.(56) In the
planning of ALT flaps, a 100% correlation was found between CTA findings and intra-operative findings
as well.(58)
39
Figure 2 Reconstructions of CTA images in the sagittal, coronal and axial planes (right), and the VRT
coronal image of a deep inferior epigastric perforator flap.
Figure 3 3D CTA image of a deep inferior epigastric perforator flap through which can be scrolled, which
enables the user to view the vasculature from any direction. The direction of viewing is predominantly
cranial as indicated by the green cube on the bottom left. The trunk of the umbilicus and the rectus
abdominis muscles can be seen. On the left sight of the umbilicus a perforator can be seen going
medially around the rectus muscle and branching off towards the skin.
40
Magnetic resonance angiography (MRA)
MRA imaging uses a powerful magnetic field to align the nuclear magnetization of hydrogen nuclei in
the body. Radio-frequency pulses are used to tip the alignment of the hydrogen nuclei away from the
main magnetic field, causing the hydrogen nuclei to produce a radio-frequency signal that is detectable
by the scanner. Analyzed by a computer, this yields detailed pictures of organs, soft tissues, bone and
virtually all other internal body structures.
By injection of a paramagnetic contrast agent (gadolinium), the vessels enhance. Since magnetic
resonance imaging (MRI) has no radiation exposure, a picture of the vessels is typically obtained in the
arterial phase and subsequently in the blood-pool phase to selectively visualize only arteries and to also
see the larger artery/vein combination. Recently, a new blood-pool gadolinium contrast agent,
gadofosveset trisodium, has become available that is optimized for imaging both the arterial and bloodpool phases of contrast enhancement without interference from soft-tissue enhancement. There is also
the possibility to visualize vessels without the injection of contrast, albeit with lower resolution.(66)
Just as with CTA, a great variety of magnetic resonance scanners and software are being used, making
comparison of the results of different studies once again difficult. The scanners used varied from 1.5
(63,68-70) to 3-T.(9,17) However, in recent perforator flap studies, the 1.5-T scanners are preferred
over 3-T because of the better image quality. The 1.5-T scanners suppress signals from fat more
homogeneously than the 3-T machines.(66,69,70) The contrast medium administered varied from
none(66) to 60 ml(67) per scan. With the introduction of newer devices, the procedure time of an MRA
scan has been reduced to approximately 20 min, and with actual scan acquisition times of about 20
seconds.(68)
There are several advantages of MRA. It works with magnetism instead of radiation and, depending on
the software used, can be used without a non-iodine contrast medium, making it a relatively safe
procedure for the patient.(67) MRA produces a 3D image, which allows surgeons to accurately assess
the course and diameter of the vessels and their relation to other surrounding structures, prior to
surgery (Figure 4).(9,10,67,71) MRA can be obtained both prone and supine to have images in which
the normal contours of the abdominal and buttock fat are not distorted by the pressure of lying against a
flat surface.
The reported disadvantages of MRA are its relatively high costs. Besides, it cannot be used in
claustrophobic patients or patients with implants containing ferrous metals because these produce
scattering or can cause severe damage to the scanner, if they are magnetic. Further, compared with
CTA, the depiction of smaller perforators is less accurate with MRA (vessels up to 1.0 mm in MRA vs.
0.3 mm in CTA).(63,64) MRA has predominantly been used in the planning for free fibula
flaps(9,10,67,72) but more recently for the mapping of perforator flaps in breast reconstructions as
well.(17,63,66,68-70) MRA, as a single test in the planning of fibula flaps, has been reported to provide
all the goals addressed by Doppler, combined with conventional angiography. Furthermore, it adds
important data concerning the septocutaneous perforators that neither test can provide.(10)
For DIEP flaps, Masia et al.(66) found a 100% correlation between MRA and intra-operative findings in
a study including 56 patients (Figure 5). Greenspun et al.(70) reported a positive predictive value of
100% but a negative predictive value of 96% in a series including 31 patients and using MRA with
contrast. For fibula flaps, it is unfortunately more difficult to draw conclusions about the positive and
negative predictive values based on currently available literature. The reports that correlate the
41
preoperative images to operative findings consist of only small populations and are not
unambiguous.(10,63,64,67,68,71)
Table 1 summarizes all the characteristics found in the literature of the individual mapping methods
such as use of radiation, invasive nature of the investigation and the type of vessels they are able to
image.
Figure 4 MRA image of the lower legs, showing a normal branching pattern of the vessels.
42
Figure 5 MRA image in the axial plane of a deep inferior epigastric perforator flap. Showing a perforator
left from the centre and its course through the muscle.
Discussion
This study was undertaken to create a contemporary overview of the preoperative mapping methods for
(free) flap planning. A serious limitation of this study is that the use of the discussed methods of
mapping has not been reported for all currently used flaps. Therefore, only an extrapolation of the
findings in this study to other flaps can be given. Furthermore, the studies found were mostly large case
series and have low levels of evidence. For this reason, it is not possible to draw concrete conclusions
based on the current literature; nevertheless, some findings are unequivocal and guidelines for clinical
practice can be distilled.
43
While interpreting these data, it is important to realize that anatomical knowledge still remains the
cornerstone to successful flap harvest and that preoperative mapping methods only serve as an adjunct
to surgery. Furthermore, with some of the methods described above, it might be difficult to detect
discrepancies between the perforating artery and the committante vein. Therefore, it is always good to
have a back-up plan. In addition, readers should be aware that most studies took place in specialized
units and the achieved results might not be obtainable in less specialized units in which a specific
mapping method might not be available. Finally, it is important to realize that collective collaboration
between radiologist and surgeon is key in order to maximally exploit the more advanced preoperative
mapping possibilities.
This report shows that, in such settings, CTA(7,8,16,48,49,56,57) and MRA(9,10,41,67,71) can produce
the best 3D image of the vessels and their surrounding structures.
The advantage of CTA over MRA is that it is able to depict small vessels and especially perforators
more accurately.(63,64) This is why CTA, at present, is predominantly being used in the scanning of
perforator flaps, in which the size and the course of the perforators are of great importance for
surgery.(7,8,16,48,49,56,57) The main advantages of MRA over CTA are its lack of exposure of the
patient to ionizing radiation and the necessity to use a potentially nephrotoxic contrast medium (if used
at all).(66,67)
Although HHD has proven to be less accurate in locating perforators compared with CDS(2,3) and
CTA,(7,16,26,48) it will probably remain of importance in clinical practice. This is because the device is
portable and the examination inexpensive and relatively easy to perform and interpret. In contrast to
most other mapping methods, it can be made available during surgery.(2,4,25,26) Furthermore, it can
be used as primary mapping device in thin flaps and in flaps, which do not rely on a specific perforator
such as the radial forearm flap or as a complementary device to CTA or MRA in the planning of, for
example, DIEP and fibula flaps.
CDS is a readily available non-invasive mapping device in most vascular units that, to some extent, also
offers information about the 3D anatomy around the vasculature.37 The disadvantage of CDS is that the
investigation can only be performed by skilled personnel, who also need to have knowledge of flap
surgery. In addition, it is less reproducible because of its real-life dynamics.(8,9) This is why we believe
CDS should only be used as primary mapping modality in selected cases, as for instance, in cases in
which CTA or MRA are contraindicated or in which there is a special interest in flow within the vessels.
With the introduction of CTA, MRA and CDS, the need for DSA in free-flap planning as a primary
mapping device seems to fade. Compared with CDS and MRA, it is more invasive, necessitates a
radiation dose and the use of iodinatecontrast medium. Furthermore, the images it produces are only
2D and finally, false aneurysms may occur.
Future developments in the planning of free flaps should focus on the refining of 3D reconstructions and
in a further elimination of the use of invasive diagnostic tests that rely on contrast media, which may
cause co-morbidity. Because MRA has most of these characteristics, we believe that MRA has the
greatest potential for the future, especially if future developments make it as accurate as CTA.
Nevertheless, HHD will always keep a place in (free) flap planning because of its ease of use and intraoperative availability.
44
Future research should also focus on the comparison between the different mapping methods and
should prospectively compare the findings of the diagnostic tests with intra-operative findings.
CTA and MRA are currently the best methods available to map the vasculature of flaps that rely on
perforators and their surrounding anatomy. In the planning of thin pedicled flaps that are planned close
to a defect, in flaps with a more straightforward anatomy and for intra-operative use, the HHD remains
to be mapping method of choice. DSA is slowly fading out and CDS can be used as an alternative,
whenever there are contra-indications to the use of the other methods of investigation.
45
References
1 Smit JM, Acosta R, Zeebregts CJ, et al. Early reintervention of compromised free flaps improves
success rate. Microsurgery 2007;27:612-616.
2 Yu P, Youssef A. Efficacy of the handheld Doppler in preoperative identification of the cutaneous
perforators in the anterolateral thigh flap. Plast Reconstr Surg 2006;118:928-935.
3 Khan UD, Miller JG. Reliability of handheld Doppler in planning local perforator-based flaps for
extremities. Aesthetic Plast Surg 2007;31:521-525.
4 Giunta RE, Geisweid A, Feller AM. The value of preoperative Doppler sonography for planning free
perforator flaps. Plast Reconstr Surg 2000;105:2381-2386.
5 Heitland AS, Markowicz M, Koellensperger E, et al. Duplex ultrasound imaging in free transverse
rectus abdominis muscle, deep inferior epigastric artery perforator, and superior gluteal artery
perforator flaps: early and long-term comparison of perfusion changes in free flaps following breast
reconstruction. Ann Plast Surg 2005;55:117-121.
6 Klein MB, Karanas YL, Chow LC, et al. Early experience with computed tomographic angiography in
microsurgical reconstruction. Plast Reconstr Surg 2003;112:498-503.
7 Masia J, Clavero JA, Larran˜aga JR, et al. Multidetector-row computed tomography in the planning
of abdominal perforator flaps. J Plast Reconstr Aesthet Surg 2006;59:594-599.
8 Smit JM, Dimopoulou A, Liss AG, et al. Preoperative CT angiography reduces surgery time in
perforator flap reconstruction. J Plast Reconstr Aesthet Surg 2009;62:1112-1117.
9 Lohan DG, Tomasian A, Krishnam M, et al. MR angiography of lower extremities at 3T: presurgical
planning of fibular free flap transfer for facial reconstruction. Am J Roentgenol 2008; 190:770-776.
10 Fukaya E, Grossman RF, Saloner D, et al. Magnetic resonance angiography for free fibula flap
transfer. J Reconstr Microsurg 2007;23:205-211.
11 De Frene B, Van Landuyt K, Hamdi M, et al. Free DIEAP and SGAP flap breast reconstruction after
abdominal/gluteal liposuction. J Plast Reconstr Aesthet Surg 2006;59:1031-1036.
12 Futran ND, Stack Jr BC, Payne LP. Use of color Doppler flow imaging for preoperative assessment
in fibular osteoseptocutaneous free tissue transfer. Otolaryngol Head and Neck Surg 1997;117:660663.
13 Yano K, Hosokawa K, Nakai K, et al. A rare variant of the deep inferior epigastric perforator:
importance of preoperative color-flow duplex scanning assessment. Plast Reconstr Surg
2003;111:1578-1579.
14 Smith RB, Thomas RD, Funk GF. Fibula free flaps: the role of angiography in patients with abnormal
results on preoperative color flow Doppler studies. Arch Otolaryngol Head Neck Surg 2003;129:712715.
15 Klein S, Hage JJ, van der Horst CM, et al. Ankle-arm index versus angiography for the
preassessment of the fibula free flap. Plast Reconstr Surg 2003;111:735-743.
16 Alonso-Burgos A, García-Tutor E, Bastarrika G, et al. Preoperative planning of deep inferior
epigastric artery perforator flap reconstruction with multislice-CT angiography: imaging findings and
initial experience. J Plast Reconstr Aesthet Surg 2006; 59:585-593.
17 Alonso-Burgos A, García-Tutor E, Bastarrika G, et al. Preoperative planning of DIEP and SGAP
flaps: preliminary experience with magnetic resonance angiography using 3-Tesla equipment and
blood-pool contrast medium. J Plast Reconstr Aesthet Surg 2010;63:298-304.
46
18. Aoyagi F, Fujino T, Ohshiro T. Detection of small vessels for microsurgery by a Doppler flowmeter.
Plast Reconstr Surg 1975;55:372-373.
19 Gerlock Jr AJ, Perry PE, Goncharenko V, et al. Evaluation of the dorsalis pedis free flap donor site
by angiography. Radiology 1979;130:341-343.
20 May Jr JW, Athanasoulis CA, Donelan MB. Preoperative magnification angiography of donor and
recipient sites for clinical free transfer of flaps or digits. Plast Reconstr Surg1979;64:483-490.
21 Merritt CR. Doppler color flow imaging. J Clin Ultrasound 1987; 15:591-597.
22 Hutchinson DT. Color duplex imaging. Applications to upper extremity and microvascular surgery.
Hand Clin 1993;9:47-57.
23 Komorowska-Timek E, Gurtner GC. Intraoperative perfusion mapping with laser-assisted
indocyanine green imaging can predict and prevent complications in immediate breast
reconstruction. Plast Reconstr Surg 2010;125:1065-1073.
24 Matsui A, Lee BT, Winer JH, et al. Image-guided perforator flap design using invisible near-infrared
light and validation with x-ray angiography. Ann Plast Surg 2009;63:327-330.
25 Mun GH, Jeon BJ. An efficient method to increase specificity of acoustic Doppler sonography for
planning a perforator flap: perforator compression test. Plast Reconstr Surg 2006;118:296-297.
26 Puri V, Mahendru S, Rana R. Posterior interosseous artery flap, fasciosubcutaneous pedicle
technique: a study of 25 cases. J Plast Reconstr Aesthet Surg 2007;60:1331-1337.
27 Ahmad N, Kordestani R, Panchal J, et al. The role of donor site angiography before mandibular
reconstruction utilizing free flap. J Reconstr Microsurg 2007 May;23:199-204.
28 Shaw RJ, Batstone MD, Blackburn TK, et al. Preoperative Doppler assessment of perforator
anatomy in the anterolateral thigh flap. Br J Oral Maxillofac Surg; 2009 Sep 15 [Epub ahead of print].
29 Chang CC, Wong CH, Wei FC. Free-style free flap. Injury 2008; 39:S57-61.
30 D’Arpa S, Cordova A, Pirrello R, et al. Free style facial artery perforator flap for one stage
reconstruction of the nasal ala. J Plast Reconstr Aesthet Surg 2009;62:36-42.
31 Hamdi M, Van Landuyt K, Monstrey S, et al. Pedicled perforator flaps in breast reconstruction: a new
concept. Br J Plast Surg 2004 Sep;57:531-539.
32 Niranjan NS, Price RD, Govilkar P. Fascial feeder and perforator- based VeY advancement flaps in
the reconstruction of lower limb defects. Br J Plast Surg 2000;53:679-689.
33 Demirtas Y, Ozturk N, Kelahmetoglu O, et al. Pedicled perforator flaps. Ann Plast Surg 2009;63:179183.
34 Pignatti M, Pasqualini M, Governa M, et al. Propeller flaps for leg reconstruction. J Plast Reconstr
Aesthet Surg 2008;61:777-783.
35 Jakubietz RG, Jakubietz MG, Gruenert JG, et al. The 180-degree perforator-based propeller flap for
soft tissue coverage of the distal, lower extremity: a new method to achieve reliable coverage of the
distal lower extremity with a local, fasciocutaneous perforator flap. Ann Plast Surg 2007;59:667-671.
36 Misra A, Niranjan NS. Fasciocutaneous flaps based on fascial feeder and perforator vessels for
defects in the patellar and peripatellar regions. Plast Reconstr Surg 2005;115:1625-1632.
37 Tsukino A, Kurachi K, Inamiya T, et al. Preoperative color Doppler assessment in planning of
anterolateral thigh flaps. Plast Reconstr Surg 2004;113:241-246.
38 Ogawa R, Hyakusoku H, Murakami M. Color Doppler ultrasonography in the planning of
microvascular augmented “superthin” flaps. Plast Reconstr Surg 2003;112:822-828.
47
39 Michlits W, Papp C, Hörmann M, et al. Nose reconstruction by chondrocutaneous preauricular free
flaps: anatomical basis and clinical results. Plast Reconstr Surg 2004;113:839-846.
40 Chen JJ, Giese S, Jeffrey RB, et al. Treatment and stabilization of complex wounds involving the
pelvic bone, groin, and femur with the inferiorly based rectus abdominis musculocutaneous flap and
the use of power color Doppler imaging in preoperative evaluation. Ann Plast Surg 1999 Nov;
43:494-498.
41 Valentini V, Agrillo A, Battisti A, et al. Surgical planning in reconstruction of mandibular defect with
fibula free flap: 15 patients. J Craniofac Surg 2005;16:601-607.
42 Monaghan AM, Dover MS. Assessment of free fibula flaps: a cautionary note. Br J Oral Maxillofac
Surg 2002;40:258-259.
43 Seres L, Csaszar J, Voros E, et al. Donor site angiography before mandibular reconstruction with
fibula free flap. J Craniofac Surg 2001;12:608-613.
44 Ohjimi H, Era K, Tanahashi S, et al. Ex vivo intraoperative angiography for rectus abdominis
musculocutaneous free flaps. Plast Reconstr Surg 2002;109:2247-2256.
45 Ohjimi H, Era K, Fujita T, et al. Analyzing the vascular architecture of the free TRAM flap using
intraoperative ex vivo angiography. Plast Reconstr Surg 2005;116:106-113.
46 Gabriel M, Pawlaczyk K, Waliszewski K, et al. Location of femoral artery puncture site and the risk of
postcatheterization pseudoaneurysm formation. Int J Cardiol 2007;120:167-171.
47 Monstrey SJ. Ankle-arm index versus angiography for preassessment of the fibula free flap. Plast
Reconstr Surg 2003;112:710-711.
48 Rozen WM, Phillips TJ, Ashton MW, et al. Preoperative imaging for DIEA perforator flaps: a
comparative study of computed tomographic angiography and Doppler ultrasound. Plast Reconstr
Surg 2008;121:9-16.
49 Pacifico MD, See MS, Cavale N, et al. Preoperative planning for DIEP breast reconstruction: early
experience of the use of computerised tomography angiography with VoNavix 3D software for
perforator navigation. J Plast Reconstr Aesthet Surg 2009;62:1464-1469.
50 Gacto-Sa´nchez P, Sicilia-Castro D, Go´mez-Cı´a T, et al. Use of a three-dimensional virtual reality
model for preoperative imaging in DIEP flap breast reconstruction. J Surg Res; 2009 Feb 21 [Epub
ahead of print].
51 Imai R, Matsumura H, Tanaka K, et al. Comparison of Doppler sonography and multidetector-row
computed tomography in the imaging findings of the deep inferior epigastric perforator artery. Ann
Plast Surg 2008;61:94-98.
52 Rozen WM, Ashton MW, Grinsell D, et al. Establishing the case for CT angiography in the
preoperative imaging of abdominal wall perforators. Microsurgery 2008;28:306-313.
53 Rozen WM, Ashton MW. The “limited rectus sheath incisions” technique for DIEP flaps using
preoperative CT angiography. Microsurgery 2009;29:525-528.
54 Ribuffo D, Atzeni M, Saba L, et al. Angio computed tomography preoperative evaluation for
anterolateral thigh flap harvesting. Ann Plast Surg 2009;62:368-371.
55 Singh J, Daftary A. Iodinated contrast media and their adverse reactions. J Nucl Med Technol
2008;36:69-77.
48
56 Uppal RS, Casaer B, Van Landuyt K, et al. The efficacy of preoperative mapping of perforators in
reducing operative times and complications in perforator flap breast reconstruction. J Plast Reconstr
Aesthet Surg 2009;62:859-864.
57 Clavero JA, Masia J, Larrañaga J, et al. MDCT in the preoperative planning of abdominal perforator
surgery for postmastectomy breast reconstruction. Am J Roentgenol 2008;191:670-676.
58 Rozen WM, Ashton MW, Pan WR, et al. Anatomical variations in the harvest of anterolateral thigh
flap perforators: a cadaveric and clinical study. Microsurgery 2009;29:16-23.
59 Wong C, Saint-Cyr M, Rasko Y, et al. Three- and four-dimensional arterial and venous perforasomes
of the internal mammary artery perforator flap. Plast Reconstr Surg 2009;124:1759-1769.
60 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 2009;29:326-329.
61 Demirtas Y, Cifci M, Kelahmetoglu O, et al. Three-dimensional multislice spiral computed
tomographic angiography: a potentially useful tool for safer free tissue transfer to complicated
regions. Microsurgery 2009;29:536-540.
62 Duymaz A, Karabekmez FE, Vrtiska TJ, et al. Free tissue transfer for lower extremity reconstruction:
a study of the role of computed angiography in the planning of free tissue transfer in the
posttraumatic setting. Plast Reconstr Surg 2009;124:523-529.
63 Rozen WM, Stella DL, Bowden J, et al. Advances in the pre-operative planning of deep inferior
epigastric artery perforator flaps: magnetic resonance angiography. Microsurgery 2009;29:119-123.
64 Rozen WM, Ashton MW, Stella DL, et al. Magnetic resonance angiography and computed
tomographic angiography for free fibular flap transfer. J Reconstr Microsurg 2008;24:457-458.
65 Casey 3rd WJ, Chew RT, Rebecca AM, et al. Advantages of preoperative computed tomography in
deep inferior epigastric artery perforator flap breast reconstruction. Plast Reconstr Surg
2009;123:1148-1155.
66 Masia J, Kosutic D, Cervelli D, et al. In search of the ideal method in perforator mapping:
noncontrast magnetic resonance imaging. J Reconstr Microsurg 2010;26:29-35.
67 Kelly AM, Cronin P, Hussain HK, et al. Preoperative MR angiography in free fibula flap transfer for
head and neck cancer: clinical application and influence on surgical decision making. Am J
Roentgenol 2007;188:268-274.
68 Neil-Dwyer JG, Ludman CN, Schaverien M, et al. Magnetic resonance angiography in preoperative
planning of deep inferior epigastric artery perforator flaps. J Plast Reconstr Aesthet Surg
2009;62:1661-1665.
69 Vasile JV, Newman T, Rusch DG, et al. Anatomic imaging of gluteal perforator flaps without ionizing
radiation: seeing is believing with magnetic resonance angiography. J Reconstr Microsurg
2010;26:45-57.
70 Greenspun D, Vasile J, Levine JL, et al. Anatomic imaging of abdominal perforator flaps without
ionizing radiation: seeing is believing with magnetic resonance imaging angiography. J Reconstr
Microsurg 2010;26:37-44.
71 Mast BA. Comparison of magnetic resonance angiography and digital subtraction angiography for
visualization of lower extremity arteries. Ann Plast Surg 2001;46:261-264.
72 Lorenz RR, Esclamado R. Preoperative magnetic resonance angiography in fibular-free flap
reconstruction of head and neck defects. Head Neck 2001;23:844-850.
49
50
Chapter
5
Ankle-arm index versus angiography
for the preassessment of the
fibula free flap
Klein S
Hage JJ
van der Horst CMAM
Lagerweij M
Plast Reconstr Surg 2003; 111: 735-743
51
52
Summary
Peripheral arterial occlusive disease or congenital anomalies of the major crural arteries may limit the
use of the fibula free flap and should be detected preoperatively.
Conventional selective angiography is the definitive standard imaging method for making this diagnosis,
but it has drawbacks. A safer, cheaper, more accurate, and noninvasive alternative is desirable. The
authors sought to test the hypothesis that the ankle-arm index of each of the three crural arteries,
combined with pencil Doppler examination of the peroneal skin perforators, would provide adequate
information to restrict the use of angiography to cases in which the outcomes of either or both of these
options are insufficient. The ankle-arm index data of each of the three crural arteries, as well as pencil
Doppler examination of the peroneal skin perforators of both legs of nine prospectively included patients
and the nonoperated legs of 13 retrospectively included patients, were compared statistically in four
different ways with the preoperative angiographic findings. A combined ankle-arm index and pencil
Doppler examination is not accurate enough to detect legs or arteries with subclinical peripheral arterial
occlusive disease or vascular variation and, hence, is not a sufficient basis on which to develop the
surgical plan for a fibula free flap.
Introduction
Peripheral arterial occlusive disease or congenital anomalies of any of the crural vessels may limit the
use of the fibula osseous or the osteocutaneous free flap because of possible insufficiency of the
peroneal pedicle of the intended flap or possible impairment of the remaining vascular supply to the
extremity after iatrogenic loss of these peroneal vessels.(1) The congenital anomalies and branching
patterns of the popliteal artery of influence on the use of the fibula flap were classified by Lippert and
Pabst(2) and Kim et al.(3) (Table 1 and Fig. 1).
Table 1
Classification of branching patterns of the popliteal artery (2,3)
Class
Vascular pattern*
I-A
AT arises first below the knee joint, followed by bifurcation of TPT in PT and PR
I-B
AT, PT, and PR arise at same point below the knee joint within 0.5 cm
II-A1
AT arises above the knee joint with straight course in its proximal segment
II-A2
AT arises above the knee joint with medial swing in its proximal course
II-B
PT arises at or above the knee joint, common trunk of AT and PR
II-C
PR arises at or above the knee joint, common trunk of AT and PT
III-A
Hypoplastic or aplastic PT, distal replaced by PR
III-B
Hypoplastic or aplastic AT, dorsalis pedis replaced by PR
III-C
Hypoplastic or aplastic AT and PT, dorsalis pedis, and PT replaced by PR
(so-called peronea magna artery)
IV
Hypoplastic or aplastic PR
* AT, anterior tibial artery; TPT, tibioperoneal trunk; PT, posterior tibial artery; PR, peroneal artery.
53
Figure 1. Variations of crural vascular anatomy according to Lippert and Pabst(2) and Kim et al.(3)
Left to right: normal crural vascular anatomy with patent anterior tibial (AT), posterior tibial (PT), and
peroneal (PR) arteries; class III-A vascular anomaly with hypoplastic or aplastic PT, distal replaced by
PR; class III-B anomaly with hypoplastic or aplastic AT; class III-C anomaly with hypoplastic or aplastic
AT and PT (so-called peronea magna artery); and class IV anomaly with hypoplastic or aplastic PR. In
class III-C legs, the use of the peroneal artery for fibula free flap transplantation would lead to an
ischemic foot, whereas in class IV, the fibula free flap lacks a vascular pedicle.
Lippert and Pabst(2) found the peroneal artery to be congenitally insufficient (class IV) in 0.1% of their
dissections. In such class IV cases, dissection would result in a nonviable fibula free flap. In 0.2 to 3.5%
of cases, however, both the anterior and posterior tibial vessels are congenitally insufficient (class IIIC),
and only the so-called peronea magna artery vascularizes the foot.(3,4) Hence, this vessel may not be
used for fibula free flap transfer in these cases.(5) In the class III-A cases in which only the posterior
tibial artery is hypoplastic or aplastic (0.9 to 4.0 %),(3,6) the lower leg and foot are still vascularized by
communicative and perforating branches of the anterior tibial artery remaining after the use of the
peroneal pedicle for fibula flap transplantation. Likewise, the foot is vascularized by communicative
branches of the remaining posterior tibial artery after fibula transplantation in cases of isolated
hypoplasia or aplasia of the anterior tibial vessels (1.6 to 6%)(3); hence, there is no absolute
contraindication to the of use the fibula flap in these class III-B cases. Nevertheless, Blackwell(7)
advocated choosing an alternative flap donor site even in patients with class III-A or III-B popliteal
branching patterns, and most authors agree that the presence and patency of at least two of the three
crural arteries and of the peroneal perforators to any intended skin paddle must be confirmed before
undertaking transplantation of the fibula flap.(4,6–8)
Conventional highly selective angiography remains the generally accepted standard for preoperative
assessment of the crural vessels.(4,6,7) Still, angiography is an expensive and invasive procedure that
features a complication rate of 3 to 5%, a morbidity rate of 0.5 to 3.9%, and a mortality rate of
0.03%.(1,9–11) Even the long-term morbidity and mortality of angiography are not negligible.(9–11)
54
Because of these disadvantages, a safer, cheaper, and accurate alternative to routine angiographic
preoperative examination would be desirable. Contrary to conventional angiography, an alternative
procedure should also allow the assessment of the presence and quantity of peroneal septocutaneous
perforators, because these characteristics determine the feasibility of the use of the osteocutaneous
variation of the fibula flap.(1)
Some researchers have advocated that normal palpatory pulsations of the dorsalis pedis and tibial
posterior arteries are sufficient to ensure the adequacy of the vascularization of the donor lower
extremity and have advised restricting further angiographic vascular assessment to the minority of
patients in whom these pulsations are diminished or abnormal or to those with a history of intermittent
claudication or overt symptoms of vascular insufficiency.(5,12,13) Detection of the pulsations by pencil
Doppler examination improved the accuracy of this limited physical examination, and, contrary to sole
palpation, pencil Doppler can be used to assess the quantity and quality of peroneal skin
perforators.(14,15) Moreover, pencil
Doppler examination of the skin perforators might provide additional information on the patency of the
peroneal artery. Compared with sole ultrasonographic detection of the pulsatile flow, however,
measurement of the ankle-arm index with pencil Doppler is again more accurate in the assessment of
the adequacy of vascularization of the donor lower extremity.(14,16,17)
To date, no one has examined the possibility of detecting congenital or acquired vascular anomalies
that would alter the surgical plan by combining the ultrasonographically obtained ankle-arm index with
pencil Doppler evaluation of the peroneal skin perforators. Futran et al.(8) compared the value of
preoperative vascular assessment of the free fibula flap with the ankle-arm index with that of color-flow
Doppler imaging and found an index of less than 1.0 to be predictive of lower-leg vascular disease that
might jeopardize the flap or the donor leg.
They observed that anomalous vascular patterns could not be detected by using the index alone.(8)
Still, the value of combined ultrasonographic ankle-arm index and peroneal perforator examination for
the preassessment of a fibula free flap has never been checked against the definitive standard of
angiography.
We set out to test the hypothesis that the ankle-arm index of each of the three major crural arteries (the
anterior and posterior tibial arteries and the peroneal artery), combined with the pencil Doppler
examination of the peroneal skin perforators, provides adequate information in patients without
intermittent claudication that the use of angiography can be restricted to cases in which the outcome of
either or both methods is insufficient. For this study, data on the ankle-arm index and pencil Doppler
examination were compared with the findings based on preoperative angiography in a retrospective and
prospective setting.
55
Patients and methods
Patients
Retrospective setting.
Using the databases of the Departments of Plastic Surgery and Radiology of two hospitals in
Amsterdam (Academisch Medisch Centrum and Antoni van Leeuwenhoek Hospital), we gathered data
recorded for 63 patients for whom fibula free flap transplantation was considered between January of
1985 and January of 2001. Of the 63 patients, we traced 35 who were still alive and living in the
Netherlands. Thirteen of these 35 patients could not be included in the study, because only unilateral
angiography of the donor leg was available or because their angiograms had been destroyed. Fourteen
of the remaining 22 patients were willing to support the study after being informed of it by letter or during
routine postoperative follow-up visits. One of these 14 patients had to be excluded because of previous
vascular surgery in the lower leg that was not used as a donor site. Of the remaining 13 patients, none
had a history of diabetes mellitus or intermittent claudication or symptoms of vascular insufficiency, and
they were retrospectively included in this study. Two had not undergone transplantation of the fibula free
flap; hence, the information regarding both legs of these two patients could be used. Because the data
regarding the nonoperated leg in the remaining 11 patients could be used, information on a total of 15
lower legs in 13 retrospectively included patients (seven men and six women) with a mean age of 58
years (range, 16 to 80 years) was available for the study.
Prospective setting.
Nine of the 10 patients (four men and five women) with a mean age of 50 years (range, 28 to 67 years)
who were considered for fibula free flap transplantation between
January of 2001 and October of 2001 agreed preoperatively to enter this study by providing informed
consent. The remaining patient had to be excluded because of a history of intermittent claudication of
the right leg. For the comparison of angiography versus ankle-arm index combined with pencil Doppler,
we used the data of both legs of these nine prospectively included patients who did not have diabetes
mellitus. With these patients added to the data of the nonoperated legs of the 13 retrospectively
included patients, we studied a total of 99 arteries in 33 legs.
Methods
Assessment of ankle-arm index.
The ankle-arm indexes of all 22 patients were obtained by one of the authors (S.K.), who was blinded to
the outcome of angiography to prevent interobserver variation and bias. Brachial and crural systolic
blood pressure was measured bilaterally after a period of rest with the patient supine with the use of an
aneroid sphygmomanometer with a small (14-cm) cuff placed around the patient’s upper arm for the
assessment of the brachial blood pressure and a broad (17-cm) cuff placed just proximally to the ankle
to obtain the crural blood pressure.18 The systolic blood pressure of the radial artery at the level of the
wrist, of the peroneal artery superior to the lateral malleolus, of the dorsalis pedis artery at the dorsum
of the foot, and of the posterior tibial artery at the medial malleolus were measured twice with the use of
a pencil Doppler device (Mini Dopplex D900; Huntleigh Healthcare, Ltd., Cardiff, U.K.) to detect the
arterial pulse.(8) For this procedure, all vessels were located by palpation of the arterial pulses. The
ankle-arm index was calculated separately for each of the three crural arteries in each of both lower
56
legs of the prospectively included patients, and for the nondonor leg of the retrospectively included
patients, according to the following formula: crural systolic pressure ÷ brachial systolic pressure. For the
numerator, the highest of both measurements of each crural artery was used; for the denominator, the
highest of the four bilaterally obtained radial measurements was used.(19) For each of the 33 legs, the
average of the ankle-arm indexes of the three crural arteries was considered to represent the index of
that leg. To evaluate a possible influence of the choice of representative ankle-arm index on the
accuracy of the ultrasonographic assessment, the highest and lowest indexes of each leg were also
scored.(20)
Classification of ankle-arm index.
On the basis of the ankle-arm index, we classified all arteries of the lower legs as adequate (index >1.0)
or inadequate (index ≤1.0).(8,16,21) Because none of our patients had diabetes mellitus and none had
a systolic ankle pressure greater than 300 mmHg, any ankle-arm index higher than 1.3 was also
classified as adequate.(22,23) Based on the findings of Futran et al.,(8) we expected an inadequate
index to be predictive of an aberrant angiogram. Normal angiographic findings were expected for
arteries classified with an adequate index. Likewise, the average ankle-arm index of the legs was
classified as either adequate (index >1.0) or inadequate (index ≤1.0). In patients with an inadequate
index, we expected to find an aberrant angiogram, whereas a normal angiogram was expected in all
patients with an adequate ankle-arm index.(8,18)
Assessment and classification of peroneal skin perforators.
The peroneal skin perforators were assessed with the use of a pencil Doppler device ascending in the
line of the posterior intermuscular septum between the soleus and peroneus muscles at the level of the
lower two-thirds of the fibula, this being the skin part included for osteocutaneous fibula free flap
transplantation.(15) We accepted adequate pulsatile flow in these perforators as additional proof of the
patency of the peroneal artery. On the basis of the results of ultrasonographic detection, the peroneal
arteries were classified as detectable or nondetectable.
Assessment and classification of angiographic information.
All angiograms were reviewed for peripheral arterial occlusive disease and congenital anomalies of
classes III and IV in consultation with a radiologist who was blinded to the outcome of ultrasonographic
ankle-arm index and perforator evaluation.(2,3) Based on the angiographic findings, we classified each
of the 99 arteries and all 33 legs as either angiographically normal or angiographically aberrant (signs of
peripheral arterial occlusive disease or class III or IV vascular pattern). Occlusive disease and class III
or IV vascular patterns were considered in combination, because both may jeopardize the flap or the
donor extremity. Any leg was classified as angiographically aberrant if one or more of its crural arteries
showed angiographic signs of peripheral arterial occlusive disease or if it had a class III or IV vascular
pattern (Fig. 2).
57
Figure 2. Appearance of subclinical peripheral arterial occlusive disease on angiography of crural
arteries (arrows). Such peripheral arterial occlusive disease does not cause intermittent claudication
and proved to be nondetectable with the use of the ankle-arm index.
Checking data of the skin perforators against the peroneal ankle-arm index.
We intended to compare the mean and standard deviation of the peroneal ankle-arm index of all
detectable peroneal arteries with that of all nondetectable peroneal arteries to evaluate the
complementary value of the detectability of the skin perforators in relation to the peroneal index.
Checking data of ankle-arm index against the angiographic information.
To test our hypothesis, the correlation between the data obtained with the ankle-arm index and the data
obtained from the preoperative angiograms was assessed statistically in two different ways for each
individual crural artery and in two different ways for each examined leg:
58
1. The distribution of the 99 arteries regarding inadequate versus adequate ankle-arm indexes was
matched against the classification of these arteries as angiographically aberrant or normal to assess the
accuracy of an inadequate index in discriminating an angiographically aberrant artery.
2. With the use of the two-sample t test, the mean and standard deviation of the anklearm indexes of all
angiographically aberrant arteries were compared with those of all angiographically normal arteries to
evaluate the reliability of the index in tracing angiographically aberrant arteries.
3. The distribution of the 33 legs regarding inadequate versus adequate ankle-arm indexes was
matched against the classification of these legs as angiographically aberrant or normal to assess the
accuracy of an inadequate average index in detecting an angiographically aberrant leg.
4. With the use of the two-sample t test, the mean and standard deviation of the anklearm indexes of all
angiographically aberrant legs were compared with those of the angiographically normal legs to
evaluate the possible correlation between the index of a leg and its angiographic appearance.
Because the mean of the ankle-arm indexes obtained in the three retrospectively included patients in
whom angiography was performed more than 3 years earlier was within the mean and standard
deviation of the other 10 retrospectively included patients, we accepted the data as representative of a
previous state.(24)
Results
Results of ankle-arm index assessment of 99 arteries in 33 legs
The ankle-arm indexes of 51 of the 99 arteries were classified as adequate (>1.0). Eight of these 51
arteries had an index of 1.3 or more, but none of the patients had diabetes mellitus, and none showed
angiographic signs of calcification of the vascular tunica media. The remaining 48 arteries showed
inadequate indexes. The average ankle-arm index in 20 of the 33 legs was classified as adequate
(>1.0). If the highest ankle-arm index were regarded as representative of the entire lower leg, the index
could be classified adequate in 25 of the 33 legs, whereas eight of the 33 legs would show an adequate
index if the lowest of the indexes in one leg were regarded as representative of that leg.
Angiographic findings in the 33 examined legs
No angiographic aberrance of the crural arteries was observed in 27 of the 33 examined legs. A class IV
vascular variation was found in two legs (Table 2).(2,3) Furthermore, eight arteries in five legs (one with
the class IV variation) showed angiographic signs of peripheral arterial occlusive disease. Repeated
interviews of the patients confirmed that the disease was subclinical and had not resulted in intermittent
claudication. In all, we could compare the ankle-arm index data regarding 10 angiographically aberrant
arteries with that of 89 angiographically normal arteries. The information on combined index and pencil
Doppler examination of four aberrant peroneal pedicles (two with a vascular anomaly and two with
peripheral arterial occlusive disease) could be compared with that of 29 angiographically normal
peroneal pedicles.
59
Table 2
Ankle-arm indexes of the 10 arteries with vascular variations or peripheral arterial occlusive disease in
six of the 33 lower legs compared with those of the angiographically normal arteries in the same legs*
* PR, peroneal artery; AT, anterior tibial artery (dorsalis pedis); PT, posterior tibial artery; PAOD,
peripheral arterial occlusive disease. † Value observed in the anatomically aberrant artery that does not
differ significantly from those observed in the anatomically normal arteries. ‡ Value observed in the
artery with signs of PAOD that does not differ significantly from those observed in the anatomically
normal arteries.
Value of detectability of peroneal skin perforators
Seemingly normal pulsatile flow was detected by pencil Doppler imaging in the peroneal skin perforators
of all legs, including the four legs that showed angiographic signs of peripheral arterial occlusive
disease or vascular anomaly of the peroneal artery. Because we found no legs with nondetectable
peroneal skin perforators, a comparison with legs with detectable perforators was impossible.
Ultrasonographic detectability of the peroneal skin perforators seems to offer no complementary
information with regard to patency of the peroneal artery.
Accuracy of ankle-arm index of an artery in discriminating angiographically aberrant arteries
After matching the distribution of the inadequate and adequate ankle-arm indexes in the 99 arteries
against the distribution in aberrant and normal angiographic imaging results, the sensitivity and
specificity for the correlation of aberrant angiographic findings with an inadequate index were 7/10 and
48/89, respectively (Table 3). The accuracy of an inadequate index in discriminating aberrant
angiographic results was 55/99, or 56%. The positive and negative predictive values of an inadequate
index were 7/48 and 48/51. In other words, when restricting the indication for preoperative angiography
to those arteries with an inadequate index, 30% of the arteries with a congenital or acquired vascular
anomaly were not found preoperatively (sensitivity, 0.70), whereas no less than 46% of the
angiographically normal arteries were angiographically examined (specificity, 0.54). Moreover, because
85% of the arteries with an inadequate index show no angiographic aberrance (positive predictive value,
0.15), an inadequate ankle-arm index cannot be considered a good screening test for discriminating
angiographically aberrant arteries.
60
Table 3
Accuracy of inadequate ankle-arm index (<1.0) in discriminating
angiographically aberrant arteries (n = 99)
Comparing the mean ankle-arm index of angiographically normal and aberrant arteries
The ankle-arm index of each of the 10 angiographically aberrant arteries found in six legs did not differ
significantly from the indexes found in the other arteries in the same leg (Table 2). Likewise, the mean
and standard deviation of the indexes of these 10 angiographically aberrant arteries (mean, 1.02; SD,
0.14) did not differ significantly from those of the 89 angiographically normal arteries (mean, 1.05; SD,
0.14) (Table 4).
Table 4
Mean and SD of the ankle-arm indexes of 89 angiographically normal arteries compared with those of
10 angiographically aberrant arteries*
* AAI, ankle-arm index; PR, peroneal artery; AT, anterior tibial artery (dorsalis pedis); PT, posterior tibial
artery.
Accuracy of the ankle-arm index of a leg in detecting an angiographically aberrant leg
After matching the distribution of the 33 legs in inadequate versus adequate average ankle- arm index
against the distribution in aberrant and normal angiography, the sensitivity and specificity for aberrant
angiography of an inadequate average index of the leg were 3/6 and 17/27, respectively (Table 5). The
accuracy of an inadequate index in detecting an angiographically aberrant leg was 20/33, or 61%. The
positive and negative predictive values of an inadequate index were 3/13 and 17/20, respectively.
61
Hence, when restricting the indication for preoperative angiography to those legs with an inadequate
index, no less than 50% of the legs with a congenital or acquired vascular anomaly are not found
preoperatively (sensitivity, 0.50), whereas no less than 37% of the angiographically normal legs are
angiographically examined (specificity, 0.63). Moreover, because 77% of the legs with an inadequate
index show no angiographic aberrance (positive predictive value, 0.23), an inadequate ankle-arm index
cannot be used for discriminating an angiographically aberrant leg.
Table 5
Accuracy of an inadequate ankle-arm index in discriminating an angiographically abnormal leg (n = 33)
Comparing the mean ankle-arm index of angiographically normal and aberrant legs
The mean and standard deviation of the average ankle-arm indexes found in the six angiographically
aberrant legs (mean, 1.05; SD, 0.16) did not differ significantly from those of the 27 angiographically
normal legs (mean, 1.05; SD, 0.12) (Table 6). Because we think that legs with acquired or congenital
class III or IV vascular anomalies ought to be recognized before free fibula flap transplantation, we
conclude that a preoperative ankle-arm index alone or in combination with pencil Doppler assessment
of the peroneal perforators does not provide a sufficient basis for developing the surgical plan.
Table 6
Mean and SD of the ankle-arm index of 27 angiographically normal legs
compared with those of six angiographically aberrant legs*
* AAI, ankle-arm index. Values are expressed as mean AAI ± SD.
62
Discussion
In this study, subclinical congenital or acquired vascular anomalies did not consistently result in
inadequate ankle-arm indexes, and ultrasonographic examination of the peroneal skin perforators
offered no information complementary to that of the peroneal index. In one patient, the preoperatively
detected location of skin perforators was even normal, whereas the perforating branches were
intraoperatively found to branch off from the anterior tibial artery, resulting in a change of surgical plan.
Hence, the detection of pulsatile flow over these skin perforators does not imply that the peroneal
pedicle takes a favorable course for free flap transfer. Still, conventional highly selective angiography
would not have preoperatively discriminated this anomaly, either.
We observed a class IV vascular anomaly in two of 33 legs. This prevalence seems not to be in
accordance with the observations of Lippert and Pabst(2) in a cadaveric specimen. Nevertheless, our
series was limited, and both recorded prevalences may be considered statistically comparable.
To our knowledge, only Futran et al.(8) considered the use of the ankle-arm index for preassessment of
the fibula free flap. Because they found all peripheral arterial occlusive disease by performing color-flow
Doppler imaging in six of 12 patients with an inadequate index, they concluded that an alternative donor
site should be considered in patients in whom the ankle-arm index is less than 1.0. Unlike Futran et
al.,(8) we found an inadequate index not to be predictive of angiographically detectable congenital or
acquired vascular anomalies. Some angiographically normal arteries in our series had an ankle-arm
index of 1.0 or less, and the indexes of arteries with angiographic signs of peripheral arterial occlusive
disease in legs without clinical symptoms of intermittent claudication did not differ significantly from
those of angiographically normal arteries. The difference in outcomes in the two studies may have
occurred because color-flow Doppler imaging is not as sensitive as conventional highly selective
angiography in detecting subclinical peripheral arterial occlusive disease of the crural arteries. Another
possible explanation for the difference is the limited number of 99 arteries in 33 legs that were included
in our series, with just 10 congenital or acquired vascular anomalies present in six legs. The different
outcome in our study from that of Futran et al.(8) may be explained by the probable inclusion of patients
with intermittent claudication in the Futran et al. series. In the one patient excluded from our series
because of a history of intermittent claudication and overt symptoms of vascular insufficiency of the right
leg, we found two of three crural pulsations to be nondetectable and the third to have an ankle-arm
index of only 0.53. The patient’s angiogram showed complete occlusion in the proximal part of the right
iliac artery.
Like Futran et al.,(8) we chose 1.0 as the lower limit of the normal range of the ankle-arm index.
Because controversy regarding this normal range still exists, we also evaluated our results by applying
0.95 and 0.90 (22) as the lower limit and found the index to be even less accurate. The same finding
applied when the highest or the lowest of the three arterial indexes was accepted as representative for
that leg: sensitivity, specificity, positive and negative predictive value, and accuracy in detecting
angiographically aberrant legs of the highest and lowest indexes were found to be inferior to that of the
average ankle-arm index.
Likewise, the difference in the mean and deviation between both angiographic classes remained
statistically insignificant, even though the different ways of determining the ankle-arm index of a leg (by
accepting the highest, average, or lowest index as representative) considerably changed the ankle-arm
63
index of both angiographically aberrant and angiographically normal legs (Table 6). We included the
data obtained in both legs of 11 patients for this study. Because the findings in the two legs of each
patient may be dependent, this may imply a limitation on the statistical tests.
For this study, we included only nonsymptomatic patients, all of whom proved to have normal crural
pulsation on the basis of palpation. Hence, such palpatory pulsations did not discriminate possible
subclinical peripheral arterial occlusive disease in our series. The ankle-arm index has, to date, been
proved to reliably detect neither occlusive disease (of less than 50% stenosis) nor vascular anomalies.
That we found the indexes of angiographically aberrant crural arteries not to differ significantly from
those of the normal arteries in the same leg is remarkable (Table 4). It seems that sufficient flow in the
hypoplastic or stenosed artery is maintained through retrograde filling of its distal part by the remaining
normal arteries. This finding is in accordance with the observation of Campbell et al.(25) that 80% of the
220 crural arteries they examined were patent on the basis of Doppler examination, as compared with
only 58% found to be patent on the basis of arteriography. In more than 59% of the 49 vessels judged to
be occluded on arteriograms, a Doppler signal could be detected.
Futran et al.(8) recommended vascular imaging of the donor extremity in all patients with an adequate
ankle-arm index, because an adequate index did not exclude anomalous vascular patterns. For this
reason, conventional highly selective angiography is the generally accepted standard.(4,6,7) Recently,
magnetic resonance angiography(26,27) and color-flow Doppler imaging(28,29) have been advocated
as more cost-effective and safer techniques for the assessment of stenoses in the crural arteries. Still,
further study is needed to determine whether angiography, magnetic resonance angiography, or colorflow Doppler imaging is to be regarded as the method of choice in the preassessment of the fibula free
flap.
Conclusion
Based on our observations in a series of 13 retrospectively and nine prospectively included patients, we
reject the hypothesis that assessment of the ankle-arm index of each of the three major crural arteries,
combined with pencil Doppler examination of the peroneal skin perforators, provides adequate
information to restrict preoperative angiography to cases in which the outcomes of either or both
methods are insufficient. We conclude that examination of the crural arteries before free fibula flap
transplantation solely by the anklearm index and pencil Doppler imaging is not sufficient to detect
congenital vascular anomalies or subclinical peripheral arterial occlusive disease.
64
References
1 Futran ND, Stack BC Jr., Payne LP. Use of color Doppler flow imaging for preoperative assessment
in fibular osteoseptocutaneous free tissue transfer. Otolaryngol Head Neck Surg 1997;117: 660
2 Lippert H, Pabst R. Arteries of the lower leg. In H. Lippert and R. Pabst (Eds.), Arterial Variations in
Man: Classifications and Frequency. Munich: Bergmann Verlag, 1985. Pp. 63–64.
3 Kim D, Orron DE, Skillman JJ. Surgical significance of popliteal artery variants: A unified
angiographic classification. Ann Surg 1989;210: 776
4 Young DM, Trabulsy PP, Anthony JP. The need for preoperative leg angiography in fibula free flaps.
J Reconstr Microsurg 1994;10: 283
5 Jones NF. The need for preoperative leg angiography in fibula free flaps: Invited discussion. J
Reconstr Microsurg 1994;10: 287
6 Carroll WR, Esclamado R. Preoperative vascular imaging for the fibular osteocutaneous flap. Arch
Otolaryngol Head Neck Surg 1996;122: 708
7 Blackwell KE. Donor site evaluation for fibula free flap transfer. Am J Otolaryngol 1998;19: 89
8 Futran ND, Stack BC Jr., Zachariah AP. Ankle arm index as a screening examination for fibula free
tissue transfer. Ann Otol Rhinol Laryngol 1999;108: 777
9 Kadir S. Complications of angiography. In S. Kadir (Ed.), Diagnostic Angiography. Philadelphia:
Saunders, 1986. Pp. 679–689.
10 Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981;138: 273
11 Shehadi WH, Toniolo G. Adverse reactions to contrast media: A report from the Committee on
Safety of Contrast Media of the International Society of Radiology. Radiology 1980;137: 299
12 Disa JJ, Cordeiro PG. The current role of preoperative arteriography in free fibula flaps. Plast
Reconstr Surg 1998;102: 1083
13 Lutz BS, Wei FC, Ng SH, Chen IH, Chen SHT. Routine donor leg angiography before vascularized
free fibula transplantation is not necessary: A prospective study in 120 clinical cases. Plast Reconstr
Surg 1998;103: 121
14 Stoffers HEJH, Kester ADM, Kaiser V, Rinkens PELM, Kitselaar PJEHM, Knottnerus JA. The
diagnostic value of the measurement of the ankle-brachial systolic pressure index in primary health
care. J Clin Epidemiol 1996;49: 1401
15 Jones NF, Monstrey S., Gambier BA. Reliability of the fibular osteocutaneous flap for mandibular
reconstruction: Anatomical and surgical confirmation. Plast Reconstr Surg 1996;97: 707
16 Piecuch T, Jaworski R. Resting ankle-arm pressure index in vascular diseases of the lower
extremities. Angiology 1989;40: 181
17 McDermott MM. Ankle brachial index as a predictor of outcomes in peripheral arterial disease. J Lab
Clin Med 1999;133: 33
18 Zierler RE. Doppler techniques for lower extremity arterial diagnosis. Herz 1989;14: 126
19 Sloan H, Wills EM. Ankle-brachial index: Calculating your patient’s vascular risks. Nursing 1999;29:
58
20 McDermott MM, Criqui MH, Liu K. Lower ankle/brachial index, as calculated by averaging the dorsal
pedis and posterior tibial arterial pressures, and association with leg functioning in peripheral arterial
disease. J Vasc Surg 2000;32: 1164
65
21 Ramaswami G, Al-Kubouti A, Nicolaides AN, Dhanjil S, Coen LD, Belcaro G. The role of duplex
scanning in decision making for patients with claudication. Ann Vasc Surg 1999;13: 606
22 Hämäläinen H, Rönnemaa T, Halonen JP, Toikka T. Factors predicting lower extremity amputation in
patients with type 1 or type 2 diabetes mellitus: A population-based 7-year follow-up study. J Intern
Med 1999;246: 97
23 Orchard TJ, Strandness DE Jr. Assessment of peripheral vascular disease in diabetes: Report and
recommendations of an international workshop sponsored by the American Diabetes Association
and the American Heart Association, New Orleans, La., September 18–20, 1992. Circulation
1993;88: 819
24 Altman DG. Confidence interval for difference between means. In D. G. Altman (Ed.), Practical
Statistics for Medical Research. London: Chapman and Hall, 1991. Pp. 192–193.
25 Campbell WB, Fletcher EL, Hands LJ. Assessment of the distal lower limb arteries: A comparison of
arteriography and Doppler ultrasound. Ann R Coll Surg Engl 1986;68: 37
26 McDermott VG, Meakem TJ, Carpenter JP. Magnetic resonance angiography of the distal lower
extremity. Clin Radiol 1995;50: 741
27 Winchester PA, Lee HM, Khilnani NM. Comparison of two-dimensional MR digital subtraction
angiography of the lower extremity with X-ray angiography. J Vasc Interv Radiol 1998;9: 891
28 Sensier Y, Fishwick G, Owen R, Pemberton M, Bell PRF, London NJM. A comparison between
colour duplex ultrasonography and arteriography for imaging infrapopliteal arterial lesions. Eur J
Vasc Endovasc Surg 1998;15: 44
29 Aly S, Sommerville K, Adiseshiah M, Raphael M, Coleridge Smith PD, Bishop CCR. Comparison of
duplex imaging and arteriography in the evaluation of lower limb arteries. Br J Surg 1998;85: 1099
66
Chapter
6
General review:
Measurement, calculation, and normal
range of the ankle-arm index: a
bibliometric analysis and
recommendation for standardization
Klein S
Hage JJ
Ann Vasc Surg 2006; 20: 282-292
67
68
Summary
Since its introduction in 1950, a variety of methods of measurement and calculation have been used to
establish the ankle-arm index (AAI). This has resulted in variations of its normal range and difficulty in
comparing study results. Hence, the objective of our study was to analyze the disparate methods used
to assess AAI and its normal range and to recommend a standardized method to assess AAI based on
that analysis. We made an inventory of the disparate AAI methods and its normal range reported in 100
randomly selected publications and recommend the means of such standardization. We recommend
that an experienced observer assess AAI with the patient at rest in the supine position. The width of the
sphygmometer cuffs should be 1.5 times that of the extremity to be measured, and brachial and crural
pulses should be detected using a Doppler device. Systolic pressures should be measured at both arms
and over the anterior and posterior arteries of both legs, with the cuff placed just proximally to the
malleoli. The left arm pressure ought to be used as denominator and the mean of pressures of both
crural arteries of each leg ought to be used for the numerator of the AAI for that leg. We advocate 0.90
as the cut-off value to distinguish patients who need further arterial assessment.
Introduction
The ankle-arm index (AAI) is the ratio of systolic blood pressure at the level of the ankle to that at the
level of the arm.(1) Because this noninvasive method is simple, reproducible,(2-6) and accurate at
detecting the decreased blood pressure distal to an arterial stenosis,(7-9) it is often used to assess
peripheral arterial occlusive disease (PAOD). Since the introduction of the concept of the AAI by
Winsor, in 1950 (10) and its popularization by Yao et al. in 1969,(11) a wide variety of methods of AAI
measurement and calculation have been used in studies on its diagnostic and epidemiological value.
Use of these nonuniform and nonstandardized methods has resulted in variations of reportedly ”normal”
versus “abnormal” distribution of AAI. This results in confusion and hampers adequate comparison of
results from one study to another. Moreover, it prohibits the development of an evidence-based
diagnostic approach. Therefore, the objective of our study was to make an analysis of the disparate
methods used to assess AAI and its normal range and to recommend a standardized method to assess
AAI based on that analysis.
Materials and methods
Selection of 100 publications on AAI
Our method of bibliometric analysis has been previously tried and described.(12) Briefly, it seeks to
reproducibly trace and analyze publications on a certain topic, in this case the methodology of AAI
assessment. To do so, we considered all original studies and reviews indexed in Pubmed or the medical
library of the University of Amsterdam, The Netherlands, that featured the term ankle-arm index or
ankle-brachial index in the abstract for inclusion in our analysis. Using the Pubmed search engine
(http://www.ncbi.nlm.nih.gov/ accessed November 18, 2004) and the search engine of the medical
library, 812 medical journal articles and 32 book chapters were traced. As our purpose was to review
the variation of techniques of AAI assessment rather than to calculate quantitative estimates of the
outcome of these assessments, we used simple random sampling to select 100 publications that
mentioned use of AAI as the method to distinguish PAOD.(13) Doing so, we came across 13
69
publications that neither mentioned what methods had been used to assess and calculate the AAI nor
referred to any other report to provide any indication as to what method was used. Hence, these
13 publications were excluded, and instead, 13 other publications were randomly selected.
Assessment of data from the publications
The 100 publications were systematically analyzed for indications as to what method of AAI assessment
had been used by the author(s). As such, we scored on 11 key points of measurement and calculation.
Apart from data provided in each publication on 1) the position of the patient during measurement, we
noted information on 2) the width of the cuff of the sphygmometer used for the arm, 3) the width of the
cuff of the sphygmometer used for the lower leg, 4) the level of placement of this cuff on the lower leg,
5) the method of detection of the pulse in the arm, 6) the method of detection of crural pulses, 7)
whether or not the brachial pressure was measured bilaterally, 8) whether or not crural pressures were
measured bilaterally, 9) which of the crural pulses were assessed to calculate the AAI (anterior tibial,
posterior tibial, or peroneal), 10) which of the brachial and crural pressures were used for the AAI
denominator and numerator, and 11) the cut-off value for the normal AAI. To not further complicate the
grouping of various ranges of distribution of normal AAI, no difference was made between “lower than
(<)” and “equal to or lower than (≤)” or between “higher than (>)” and “equal to or higher than (≥)”.
Likewise, no attention was paid to the mean and range indicated for the different stages of PAOD
because we did not intend to describe these stages.
Whenever the methodology of any of the 11 scored items was not indicated in the text of the
publication, the references provided by the author(s) were searched for such an indication. Still, when
authors described the methods they had used and referred to other studies to motivate their choice of
method, we did not check whether the described method corresponded with the reported source
method.
Results
Information on patient position during assessment of AAI
In 60 of the 100 analyzed reports, the supine position was mentioned as the position used to measure
the brachial and crural blood pressures. Only a reference to other studies indicated the position in nine
of the other 40 reports. Still, two of these nine references failed to mention this position. The remaining
31 reports offered neither information on the position of the patient nor a reference to indicate this
position.
70
Information on sphygmometer cuff width used to assess brachial blood pressure
While describing the method of AAI measurement, a single largest group of 56 (groups of) authors did
not provide any information as to the size of the cuff that was used to assess the brachial blood
pressure.
Twenty-one (groups of) authors stipulated what size of cuff was used, and of these, nine used a cuff of
12cm,(3,4,14-20) two a cuff of 13cm,(21,22) two a cuff of 14cm,(23,24) and three a cuff of 15cm.(25-27)
One author measured the circumference of the arm to determine the proper cuff width,(28) and a cuff of
1.5 times the diameter of the arm was used in two studies.(7,29) One group of authors used either a
14cm or a 17cm cuff, depending on the size of the patient’s arm. The one remaining of these 21 (groups
of) authors reported using a pediatric cuff (8cm) to assess AAI in children.(30)
In 10 studies, the cuff size was reported to have been “appropriate” or “carefully selected”,(31-40)
whereas a “standard” cuff was reportedly used in five.(41-45) Eight (groups of) authors (9,21,46-52)
referred to other studies for their method of AAI assessment, but in seven of these references, no
information on cuff size was provided either.
Information on sphygmometer cuff width used to assess crural blood pressure
A majority of 52 (groups of) authors did not provide any information as to the size of the cuff used to
take the crural blood pressure while describing the method of AAI measurement.
Twenty-six (groups of) authors stipulated what size of cuff was used. Of these, one used a cuff of
10cm,(53)
11 a cuff of 12cm,(1,3,4,15-20,38,54)
two a cuff of 12.5cm,(55,56)
two a cuff of
13cm,(21,22) one a cuff of 13.5cm,(5) two a cuff of 14cm,(23,24) and three a cuff of 15cm.(25-27)
Three (groups of) authors used a cuff of 1.2 or 1.5 times the diameter of the leg,(7,29,57) and again,
the one remaining of these 26 (groups of) authors reported using a pediatric cuff of 8cm width to assess
AAI in children.(30)
In nine studies, an “appropriate” or “carefully selected” sized cuff was used,(31-37,39,40) whereas use
of a “standard” cuff was reported in three.(41,43,44) An arm-cuff size was probably used at the ankle in
two of the studies.(42,45)
Finally, eight (groups of) authors referred to other studies for their method of AAI assessment,(9,21,4652) but again, no information on cuff size was provided in seven of these references.
Information on sphygmometer cuff position on the lower leg
In 54 studies, the cuff position on the lower leg was indicated by either “ankle (or malleolus)” (n=24),
“proximal to malleoli (or ankle)” (n=27), “as distal as possible on the calf” (n=1), “lower 1/3 of lower leg”
(n=1), or the “posterotibial level” (n=1).
Eight (groups of) authors referred solely to other studies for their method of AAI assessment, but in two
of these, no information on the position of the crural cuff was provided in the reference either.
Thirty-eight studies provided no indication as to the position of the crural cuff, while other aspects of the
AAI assessment were described.
71
Information on the method of detection of the pulse in the arm
In 58 of the 100 studies, the brachial pulse was detected with a pencil-Doppler device, whereas
auscultation was used for this in six.(11,15,18,55,57,58)
Other techniques used were the Dynamap (n=7) (15,25,26,32,53,59,60) and photoplethysmography
(n=2).(10,35) The technique of assessment of the arm systolic pressure was not mentioned in 23
reports. In seven reports,(46-52) reference was made to other studies to indicate the method of
assessment, but these references were not always clear on the method used either.
The number of techniques totaled 103 rather than 100 because Jeelani et al.(15) used Dynamap and
auscultatory methods in addition to the pencil-Doppler to compare these three methods of pulse
detection and Yao et al.(11) used both the auscultatory method and the pencil-Doppler.
Information on the method of detection of the pulse in the leg
A vast majority of 77 (groups of) authors reported using a pencil-Doppler to detect the crural pulses. Of
these, Strandness and Sumner(7) used capacitance pulse pick-ups and plethysmography in addition to
the pencil-Doppler, whereas Carter (55) also used capacitance pulse pick-ups, visual flush technique,
and spectroscopy. They did so to compare the influence of these methods on AAI measurements.
One author used solely capacitance pulse pickups for pulse detection,(56) whereas three authors used
solely plethysmography.(10,18,61) Other methods used to detect the crural pulses were the Dynamap
(n=5) (25,26,32,59,62) and the strain-gauge technique (n=4).(21,57,63,64)
Seven (groups of) authors (7%) referred to other studies to indicate their method of assessment of
crural pulses.(46-52) However, three of those seven (groups of) authors referred to more than one other
study and, because different techniques were used in these references, it remained unclear which
technique they had actually used. Finally, three of the 100 studies did not report at all what kind of
device was used for crural pulse detection.(65-67)
Information on whether or not brachial pressure was measured bilaterally
A single largest group of 47 (groups of) authors measured the brachial blood pressure at both arms to
determine the denominator of the AAI formula. Fourteen other reports stipulated the blood pressure to
be measured at the right arm,(2,6,23,38,42,54,58,60,61,68-72) whereas one other indicated the left
arm to be used.(19)
According to five reports, the blood pressure may be taken on either
arm.(9,10,29,32,73)
Twenty-six (groups of) authors did not specify which arm was used, and seven (groups of) authors only
provided a reference for the method they had used.(46-52)
Information on whether or not crural pressures were measured bilaterally
A vast majority of 96 (groups of) authors measured the crural blood pressure at both legs to separately
distinguish
the
presence
of
PAOD
in
each
leg.
Of
these,
eight
(groups
of)
authors
(36,37,53,54,60,61,72,74) reported measuring both legs and using the lowest of the bilateral blood
pressures to determine the presence of PAOD in their patients. Shinozaki et al.(23) and Zheng et al.(32)
reported diagnosing PAOD by measuring the AAI of just one leg. Two authors did not provide any
information about whether one or two legs were measured.(75,76)
72
Information on which crural pulses were assessed for AAI
In 33 studies, both the pressures over the posterior tibial artery (PT) and the dorsal pedal or anterior
tibial artery (DP/AT) were measured before calculating AAI (Table 1). The peroneal artery (PA) pulse
was assessed in addition to that of the PT and DP/ AT in three studies. Twenty-three studies
recommended using the pressures over the PT or DP/AT for the numerator of the AAI. Three of these
23 advocated use of the best audible flow signal.(3,41,69) In 13 studies, the crural pulse was detected
only at the PT.
No specification of the crural artery other than “ankle” was provided to indicate which of the crural or
pedal pulses were detected in 18 reports. In 10 of these 18 reports, this remained unclear because of
the method used for crural pulse detection (strain-gauge, plethysmography, capacitance pulse pick-up,
spectroscopy, visual flush technique, or Dynamap). (10,18,21,25,26,56,57,61,63,64)
Three reports did not clarify which pedal artery was used to detect the crural pressure, whereas seven
(groups of) authors referred to other studies for their method of AAI assessment. Two of these seven
references merely mentioned “ankle” as the location of pulse detection.
Table 1
The 100 publications that were studied divided according to the information
provided on the crural pulses used to assess AAI numerator
Description
Number of reports {References}
PT and DP/AT
n = 33 {2,4,14,16,17,20,28,31,33-35,37,39,40,46,53-55,62,65,74-76,78,82,93100}
PT,DP/AT, and PA
n=3
PT or DP/AT
n = 23 {3,7,19,24,29,30,36,41,43,56,66,67,69,73,77,79,81,90,103-107}
PT only
n = 13 {1,6,22,23,32,38,58-60,71,72,85,108}
“ankle”
n = 18 {5,10,11,18,21,25,26,42,44,45,57,61,63,64,68,70,80,109}
no information
n=3
References for method n = 7
{27,101,102}
{15,84,110}
{9,47-52}
Information on which of the brachial and crural pressures (highest, mean, median, or lowest)
were used for denominator and numerator of AAI
No fewer than 39 different ways to calculate AAI were reported in 77 of the 100 studies. Among these,
the formulas provided most often were as follows: AAI = highest of PT and DP/AT pressures / highest
arm pressure (n=15), AAI = ankle pressure / arm pressure (n=9), and AAI = PT or DP/AT pressure / arm
pressure (n=8). Seven times we found the same (group of) author(s) to have used two different
formulas in two different studies.(19,25,26,28,37,55,56,60,72,77-81) Confusing matters even further,
some authors used more than one formula in a single study.(28,31,36,55)
The remaining 23 (groups of) authors did not report the formula they had used to calculate the AAI.
Eight of these 23 provided a reference for the formula, whereas 15 did not.
73
Of the 47 (groups of) authors who measured systolic pressure at both arms, the highest pressure found
was used for the denominator of the AAI in 38 studies, whereas the mean of both arm pressures was
used in 11 studies. This totals 49 rather than 47 studies because Hiatt et al. twice compared two
different ways of calculating the denominator.(28,78)
The lowest pressure of both arms was never reported to have been used as denominator.
Seven (groups of) authors used both the PT and the DP/AT for the numerator of separate
AAIs.(2,14,20,33,34,78,82)
Twenty-nine more (groups of) authors used more than one detected pressure for the definite calculation
of the numerator. In 23 of these 29 studies, the highest of the two or three crural blood pressures was
used for the numerator, whereas the lower was used in the five and the mean of those pressures in
four. Again, this totals 32 rather than 29 studies because Hiatt et al.(28) compared two methods
(average and lower) while McGrae McDermott et al.(31) compared three methods (average, higher, and
lower) to calculate the numerator of the AAI.
Cut-off value of the normal AAI
Some (groups of) authors defined a lower limit of AAI for the absence of PAOD, whereas others
provided an upper limit felt to be indicative of the presence of PAOD. Still others considered these two
cut-off values to be the same, thus defining one AAI below which PAOD was accepted to be present
and above which it was not.
The lower cut-off value of a normal AAI as offered in 58 of the 100 studies varied from 0.85 to 1.10, but
most often, 1.00 was cited for this (Table 2).
Eight different upper AAI limits indicative of PAOD were cited in 60 of the 100 studies. Although these
varied from 0.80 to 1.00 (Table 3), 0.90 was most often used as this cut-off value.
Rather than one cut-off value, Lennihan and MacKereth44 cited median AAI values for subjects with
and without claudication. Eighteen reports mentioned the method of assessment but lacked any
information on its normal or abnormal limits. Three additional (groups of) authors provided references
for information on these limits.(47,48,50) Neither a description nor a reference was found in one
study.(83)
Table 2
Reports categorized according to information
provided on the lower limits of the range of AAI in subjects without PAOD
Cut-off value
Number of reports {References}
AAI = 0.85
n=1
AAI = 0.90
n = 17 {20,31,32,34,36,40,45,46,53,60,67-69,71,77,97,98}
AAI = 0.92
n=1
{102}
AAI = 0.95
n=6
{1,81,82,84,103,104}
AAI = 0.97
n=3
{4,9,55}
AAI = 1.00
n = 28 {7,10,11,16,18,19,21-23,29,38,39,41,42,54,56,66,70,72,73,75,80,99-
{74}
101,105,106,108}
AAI = 1.10
74
n=2
{59,80}
Table 3
Reports categorized according to information
provided on the upper limits of the range of AAI in subjects with PAOD
Description
Number of reports {References}
AAI = 0.80
n=8
{21,33,52,54,63,70,80,96}
AAI = 0.85
n=2
{24,74}
AAI = 0.90
n = 31 {9,18-20,31-34,36-38,40,45,46,51,53,57,60,64,6769,71,72,76,77,80,85,97,98,107}
AAI = 0.92
n=2
{39,49}
AAI = 0.94
n=1
{28}
AAI = 0.95
n=4
{81,84,103,104}
AAI = 0.97
n=2
{3,4}
AAI = 1.00
n = 10 {10,11,16,22,23,66,101,106,108,109}
Comments and recommendations
Although AAI assessment currently is the most common diagnostic instrument for the detection and
quantification of PAOD,(33,77,84)
the repeatability of assessment continues to be subject to
controversy.(2,33,77,83,85,86) The variability of AAI assessment attributable to observers, timing of
measurement, and repeated measures is considerably less than that attributable to biological factors.
Estimates of intraobserver variability range from 7.3% for experienced observers to 12% for less
experienced observers,(5,6,19,87) and repeated measurements may decrease this variability.(2,19)
Standardized, repeated measurement of AAI by experienced observers is sufficiently accurate to guide
clinical decision making.(77,87) When the art and science of AAI measurement and calculation are
being taught or discussed, moreover, it is important that all involved are speaking of the same standard.
Even when limited to 100 publications, however, our bibliometric analysis yielded a great variety of
methodology. This indicates that there is still need for a consensus on the method of AAI measurement.
Before we present our recommendations to come to such consensus, some potential limitations of our
study need to be addressed. As such, we stress that ours are not evidence-based suggestions.
Because of the variety of methodology, any and all of such suggestions would lack an adequate level of
evidence. Furthermore, we did not score for potential key points such as minimum resting time prior to
measurement, room temperature, or frequency of Doppler probe since these were rarely mentioned in
the 100 reviewed publications.
Hence, ours are merely recommendations provided in an attempt at the standardization that is urgently
needed to allow comparison and meta-analysis of future study results.
75
Position of the patient during measurement
The supine position seems to be the position of choice to assess AAI because the influence of height of
the subject and his or her blood column pressure on AAI may be prevented only in this position. Less
agreement exists on the routine use of premeasurement exercise. Such exercise may be needed since
the ankle pressure may be normal at rest in patients with mild PAOD and there may be adequate
collateral flow around the arterial occlusions.(8) Even though such PAOD may only be detectable after
exercise, exercise influences the heart rate during measurement, which in turn influences AAI.(25)
Hence, in studies where AAI is measured during exercise, this should be mentioned in the Methods
section.
Width and level of the sphygmometer cuff(s)
If the width of the sphygmometer cuff is too narrow in comparison to the extremity (“undercuffing”), the
blood pressure reading will be erroneouslyhigh, whereas the reading may be too low if the cuff is too
wide (“overcuffing”).(88) Ideally, the cuff width should be at least 1.5 times the diameter of that part of
the extremity where the pressure is being measured,(7) and the size of the cuff should be adjusted in
obese patients or in patients with oddshaped arms or ankles.(88) Still, calcification in the arterial wall
can result in spuriously high readings of the systolic pressure,(7,29) and this may not be corrected by
the use of a wider cuff.
As to the position of the cuffs at the extremities, general agreement exists on placement just proximal to
the elbow on the upper arm and just proximal to the malleoli at the lower leg.
Method of detection of the pulse in the arm and leg
From their comparison of three methods of measurement of brachial systolic blood pressure, Jeelani et
al.(15) concluded that the technique of measurement significantly affects the calculation of AAI. As they
found a 20% margin of error, these authors advocated the use of just one technique to limit inter- and
intraobserver errors. For this, use of a pencil-Doppler should be considered the method of choice to
detect the brachial pulse as this was already done in half of the reviewed studies.
Measurements by Doppler device were proven at high, medium, and low blood pressures to correlate
with systolic pressure measurement obtained by conventional methods.(10,11)
Carter (55,56) observed good agreement between the values of systolic blood pressure obtained by
four different methods of monitoring the crural pulses (pulse pick-ups, pencil-Doppler flow detection,
spectroscopy, and visual flush), and capacitance pulse pick-up even allows for detection of nonpalpable
pulses.(7)
Still, allegedly normal crural blood pressures can be recorded with pulse pick-up,
spectroscopic, and visual flush methods as long as only one crural artery is patent, whereas the pencilDoppler may provide information on individual tibial vessels.(55,56) Pencil-Doppler readings are highly
correlated to intra-arterial pressure readings.(7,89) Moreover, ultrasonography is less cumbersome than
plethysmography and more reliable than the auscultation method, which has a 10% failure rate in
obtaining ankle pressures in normal individuals.(90) Although the accuracy of measurement with the
Dynamap may be higher than that with pencil-Doppler, the Dynamap is not available everywhere.
For this reason, we recommend the pencil-Doppler device as the standard instrument to measure both
the brachial and crural blood pressures.
76
Whether or not to bilaterally measure the brachial pressure and which one to use for
denominator
A majority of authors measured the systolic pressures of both arms to assess the denominator of the
AAI. A minority measured only one arm, and of these, only one indicated using the left arm.(19) This is
remarkable as the blood pressure used for the denominator of the AAI should be measured at the left
arm in cases where aortic coarctation results in a difference of blood pressure in the right and left
arms.(28,31,36) In these cases, a difference of 5-10 mmHg warrants further examination, while a
difference of 20 mmHg between the arms indicates severe stenosis.(91) In general, AAI calculated on
the basis of systolic pressure at the left arm was 0.02 lower than that at the right arm.(78) For these
reasons, we advise taking the blood pressure at both arms to rule out serious differences and using that
of the left arm to calculate the AAI denominator.
Whether or not to bilaterally measure crural pressures and which of these to use as numerator
In the healthy population, the differences between the systolic pressures of the DP/AT and PT do not
exceed 10 mmHg.(55) In general, the AAI at the DP/AT was found to be 0.04 lower than that at the
PT.(78) PAOD may, however, separately affect each of the main arteries, and a difference of over 15
mmHg between the DP/AT and the PT may provide a useful clue about the involvement of the individual
crural arteries by the occlusive process.(8,55)
Difference in pressure readings between the DP/AT and PT in the same limb, as well as an abnormal
AAI of both the DP/AT and PT in the same leg, should be considered indicative of PAOD.(8,55)
Because PAOD may also affect both legs separately, moreover, we advise bilaterally measuring the
blood pressure of both the DP/AT and the PT.
How to calculate the AAI numerator and denominator
Numerator.
McGrae McDermott et al.31 performed a multiple linear analysis to identify which one out of three
commonly used formulas to calculate the AAI was most closely associated with objective measures of
leg functioning in PAOD:
AAI = highest of DP/AT and PT/mean of both arms,
AAI = mean of DP/AT and PT/mean of both arms,
and
AAI = lowest of DP/AT and PT/mean of both arms.
The prevalence of PAOD, defined as an AAI 0.9, ranged from a minimum of 47% when the first formula
was used to a maximum of 59% when the third was used.(31) In cases where the right and left legs
showed a difference of AAI, the lowest of these correlated best with leg functioning. Moreover, the lower
AAI determined by “mean of DP/AT and PT/mean both radial artery” was most predictive of walking
endurance and walking velocity in cases of PAOD. The authors offered two potential explanations for
the finding that using the mean of the DP/AT and PT systolic pressures is the optimal way to calculate
AAI when assessing lower extremity functioning.
First, the mean of DP/AT and PT may best reflect total perfusion of the more diseased lower extremity
and, second, when the two pressures are averaged, the random variation and measurement error
77
intrinsic to measures of arterial pressure are minimized, the result being a closer association of the
mean AAI with functioning.(31) Hiatt et al.(28) used two ways to calculate the numerator of the AAI.
Since the difference in systolic blood pressure between the DP/AT and PT was found to have a 95%
range of -21 to +10 mmHg, the “mean of PT and DP/AT” was used if PT and DP/AT differed no more
than -21 to +10 mmHg. Alternatively, the “lower of PT and DP/AT” was used if PT and DP/AT differed
more than -21 to +10 mmHg.
Since AAI may indicate the presence and, more vaguely, the severity of PAOD without a high sensitivity
or specificity, only one standardized formula should be used to calculate it. This may keep its use simple
and reproducible in varying hospital and general practice settings. We strongly advise against
calculating AAI separately for each lower leg artery since the sensitivity and specificity of the AAI for
detecting PAOD per artery is even lower.(92) Still, a large difference in pressure readings between the
DP/AT and PT in the same limb should be considered indicative of PAOD.(8,55) Hence, we advise
calculating AAI for each leg by measuring the systolic pressures over both DP/AT and PT and using
their mean for the numerator as this reduces the measurement bias and gives a good impression of the
total lower leg perfusion. Bias can further be reduced by measuring the AAI twice per leg and using the
mean of both measurements for the numerator.
Denominator.
Furthermore, Hiatt et al.(28,78) advised using the “mean of both arms” for the denominator in cases
where the difference in systolic blood pressure between the right and left arms did not exceed the 95%
range of -9 to +8 mmHg. Alternatively, they advised using the ‘‘higher of both arms’’ for this if the
brachial pressure of both arms differed more than -9 to +8 mmHg because they assumed that there
might be an arterial occlusion on one side causing the difference.(28,78)
For reasons of simplicity, we advise measuring the systolic pressures of both arms to detect possible
pressure differences indicating aortic coarctation or brachial arterial stenosis. Because the systolic
pressure perfusing the body distal of the run-off of the left subclavian artery is equal to that in the left
arm, the left brachial systolic pressure should be used for the denominator, provided no brachial arterial
stenosis is found. Isolated stenosis of the left subclavian artery, however, is very rare; and the chance of
having a left subclavian stenosis without lower extremity involvement can be considered naught.
In short, we advice calculating AAI separately for each leg and using the formula
AAI = mean of DP/AT and PT/left brachial artery.
Cut-off value of the normal AAI
The lower cut-off point of the normal AAI indicative of the absence of PAOD may importantly differ from
the upper cut-off point of an aberrant AAI indicating the presence of PAOD (Tables 2 and 3).
Typically, the AAI cut-off value for presence of disease has been defined between 0.90 and 0.80,(60,74)
but it is obvious that there is – not in the least due to differences in methods of AAI measurement – no
one fixed cut-off point indicating the absence or presence of PAOD.
Varying the cut-off AAI may triple the sensitivity of the test (4,25) and double the estimates of PAOD
prevalence.(28,38,59) Hiatt et al.(78) included an extensive table of lower limits of the normal range
78
subdivided for the left and right legs, the PT and DP, and the two sexes; and some (groups of) authors
even included estimations on the probability of PAOD for a given AAI.(54,63) Still, such differentiation is
hardly applicable in daily clinical practice.
The sensitivity and specificity of AAI cut-off values to detect PAOD are 96% or higher when using
arteriography as the gold standard,(24) and subjects with a resting AAI of 0.94 or higher have no
arteriographic evidence of PAOD.(4,24,55) Carter(55) found 95% of patients without PAOD to have AAI
of 0.97 or higher, whereas all of his patients with arteriographically proven PAOD had AAI lower than
that.
We recommend that 0.9 be accepted as the upper limit of an aberrant AAI and 1.0 as the lower limit of a
normal AAI. An AAI between 0.9 and 1.0 seems to be not conclusive and should lead to further
assessment. Future investigators should state what AAI value was used as a cut-off point since the
sensitivity of the AAI for the presence of PAOD depends on this.
To summarize, we feel that the AAI reported in scientific work should be assessed by experienced
observers. Assessment of AAI at rest in the supine position is acceptable as the standard procedure.
Different sphygmometer cuffs ought to be used for the arms and legs, both of which should have a width
of 1.5 that of the extremity at the level of measurement. The cuff should be placed just proximal to the
malleoli to assess crural blood pressures. A handheld pencil-Doppler device ought to be used for
measurement of both the brachial and crural blood pressures. Measurements should be performed at
both arms and over the DP/AT and PT of both legs, but the left arm pressure is preferred for use as the
denominator. The mean of the systolic pressures of the DP/AT and the PT ought to be used for the
numerator of the AAI for that leg. However, difference in pressure readings between the DP/AT and PT
in the same limb should be considered indicative of PAOD. Measurement bias is further reduced by
measuring AAI twice per leg and using the mean of the two measurements. We recommend accepting
0.9 as the upper limit of an aberrant AAI and further assessing the arterial status of all patients with an
AAI between 0.9 and 1.0.
To allow for comparison of results from one investigator to another without tremendous variations due to
the different methods of assessment, we urge future investigators to adhere to these recommendations
or to mention the circumstances or methods of assessment and calculation of AAI in the Method section
of their report in cases where these differ from those recommended. In these cases, information on why
the authors felt it better to use an alternative method may further enhance the possibilities of
comparison with reports from other research groups.
79
References
1 Sloan H, Wills EM. Ankle-brachial index: calculating your patient’s vascular risks. Nursing
1999;29:58-59.
2 Ray SA, Srodon PD, Taylor RS, Dormandy JA. Reliability of ankle-brachial pressure index
measurement by junior doctors. Br J Surg 1994;81:188-190.
3 Lepäntola M, Lindfors O, Pekkola P. The ankle-arm systolic blood pressure ratio as a screening test
for arterial insufficiency in the lower limb. Ann Chir Gynaecol 1983;72:57-62.
4 Ouriel K, McDonnel AE, Metz CE, Zarins CK. A critical evaluation of stress testing in the diagnosis of
peripheral vascular disease. Surgery 1982;91:686-693.
5 Johnston KW, Hosang MY, Andrews DF. Reproducibility of noninvasive vascular laboratory
measurements in the peripheral circulation. J Vasc Surg 1987;6:147-151.
6 Fowkes FGR, Housley E, Macintyre CCA, Prescott RJ, Ruckley CV. Variability of ankle and brachial
systolic pressures in the measurement of atherosclerotic peripheral disease. Epidemiol Community
Health 1988;42:128-133.
7 Strandness DE, Sumner DS. Haemodynamics for Surgeons. New York: Grune and Stratton, 1975.
8 Carter SA. Respons of ankle systolic pressure to leg exercise in mild or questionable arterial
disease. N Engl J Med 1972;287:578-582.
9 Laing S, Greenhalgh RM. The detection and progression of asymptomatic peripheral arterial
disease. Br J Surg 1983;70:628-630.
10 Winsor T. Influence of arterial disease on the systolic blood pressure gradients of the extremity. Am
J Med Sci 1950;220:117-126.
11 Yao JST, Hobbs JT, Irvine WT. Ankle systolic pressure measurements in arterial diseases affecting
the lower extremities. Br J Surg 1969;56:676-679.
12 Vossen M, Hage JJ, Karim RB. Formulation of trichloroacetic acid peeling solution: a bibliometric
analysis. Plast Reconstr Surg 2000;105:1088-1094.
13 Altman DG, Bland JM. How to randomise. BMJ 1999;319:703-704.
14 Gale SS, Scissons RP, Salles-Cunha SX, Dosick SM, Whalen RC, Rigott JP. Lower extremity
arterial evaluation: are segmental arterial blood pressures worthwhile? J Vasc Surg1998;27:831-839.
15 Jeelani NUO, Braithwaite BD, Tomlin C, MacSweeney ST. Variation of method for measurement of
brachial artery pressure significantly affects ankle-brachial pressure index values. Eur J Vasc
Endovasc Surg 2000;20:25-28.
16 Ramaswami G, Al-Kutoubi A, Nicolaides AN, Dhanjil S, Coen LD, Belcaro G. The role of duplex
scanning in decision making for patients with claudication. Ann Vasc Surg1999;13:606-612.
17 Baker JD, Dix D. Variability of Doppler ankle pressures with arterial occlusive disease: an evaluation
of ankle index and brachial-ankle pressure gradient. Surgery 1981;89:134-137.
18 Hirai M, Schoop W. Clinical significance of Doppler velocity and blood pressure measurements in
peripheral arterial occlusive disease. Angiology 1984;3:45-53.
19 Stoffers HEJH, Kaiser V, Kester A, Schouten H, Knottnerus A. Peripheral arterial occlusive disease
in general practice: the reproducibility of the ankle-arm systolic pressure ratio. Scand J Prim Health
Care 1991;9:109-114.
80
20 Koji Y, Tomiyama H, Ichihashi H, et al. Comparison of ankle-brachial pressure index and pulse wave
velocity as makers of the presence of coronary artery disease in subjects with a high risk of
atherosclerotic cardiovascular disease. Am J Cardiol 2004;94:868-872.
21 Feigelson HS, Criqui MH, Fronek A, Langer RD, Molgaard CA. Screening for peripheral arterial
disease: the sensitivity, specificity, and predictive value of noninvasive tests in a defined population.
Am J Epidemiol 1994;140:526-534.
22 Köhler M, Krüpe M. Untersuchungen über Spezifität und Normalität der peripheren systolischen
Druckmessung mit der Ultraschall-Doppler-Technik an gesunden angiographierten Extremitäten. Z
Kardiol 1985;74:39-45.
23 Shinozaki T, Hasegawa T, Yano E. Ankle-arm index as an indicator of atherosclerosis: its application
as a screening method. J Clin Epidemiol 1998;51:1263-1269.
24 Bernstein EF, Fronek A. Current status of noninvasive tests in the diagnosis of peripheral arterial
disease. Surg Clin North Am 1982;62:473-487.
25 Abraham P, Desvaux B, Colin D, Leftheriotis G, Saumet JL. Heart rate correlated ankle-to-arm index
in the diagnosis of moderate lower extremity arterial disease. Angiology 1995;46:673-677.
26 Abraham P, Bickert S, Vielle B, Chevalier JM, Saumet JL. Pressure measurements at rest and after
heavy exercise to detect moderate arterial lesions in athletes. J Pediatr Surg 2001;33:721-727.
27 Smith FCT, Shearman CP, Simms MH, Gwynn BR. Falsely elevated ankle pressures in severe leg
ischaemia: The pole test - an alternative approach. Eur J Vasc Endovasc Surg 1994;8:408-412.
28 Hiatt WR, Marshall JA, Baxter J, Sandoval R, Hildebrandt W, Kahn LR. Diagnostic methods for
peripheral arterial disease in the San Louis Valley Diabetes Study. J Clin Epidemiol 1990;43:597606.
29 Zierler RE. Doppler techniques for lower extremity arterial diagnosis. Herz 1989;14:126-133.
30 Katz S, Golberman A, Avitzour M, Dolfin T. The anklebrachial index in normal neonates and infants
is significantly lower than in older children and adults. J Pediatr Surg 1997;32:269-271.
31 McGrae McDermott M, Criqui MH, Liu K, Guralnik JM, Greenland P, Martin GJ. Lower ankle-brachial
index, as calculated by averaging the dosalis pedis and posterior tibial arterial pressures, and
association with leg functioning in peripheral arterial disease. J Vasc Surg 2000;32:1164-1171.
32 Zheng ZJ, Sharrett AR, Chambless LE, Rosamond WD, Nieto FJ, Sheps DS. Associations of anklebrachial index with clinical coronary heart disease, stroke and preclinical carotid and popliteal
atherosclerosis:
the
Atherosclerosis
Risk
in
Communities
(ARIC)
Study.
Atherosclerosis
1997;131:115-125.
33 McGrae McDermott M. Ankle brachial index as a predictor of outcomes in peripheral arterial disease.
J Lab Clin Med1999;133:33-40.
34 Orchard TJ, Strandness DE. Assessment of peripheral vascular disease in diabetes. Report and
recommendations of an international workshop sponsored by the American Diabetes Association
and the American Heart Association; September 18-20, 1992, New Orleans, Louisiana. Circulation
1993;88:819–828.
35 Brothers TE, Esteban R, Robinson JG, Elliott BM. Symptoms of chronic arterial insufficiency
correlate with absolute ankle pressure better than with ankle:brachial index. Minerva Cardioangiol
2000;48:103-109.
81
36 McGrae McDermott M, Liu K, Guralnik JM, Mehta S, Criqui MH, Martin GJ. The ankle brachial index
independently predicts walking velocity and walking endurance in peripheral arterial disease. J Am
Geriatr Soc 1998;46:1355-1362.
37 Vogt MT, McKenna M, Anderson SJ, Wolfson SK, Kuller LH. The relationship between ankle-arm
index and mortality in older men and women. J Am Geriatr Soc1993;41:523-530.
38 Newman AB, Sutton-Tyrrell K, Rutan GH, Locher J, Kuller LH. Lower extremity arterial disease in
elderly subjects with systolic hypertension. J Clin Epidemiol 1991;44:15-20.
39 Vowden K, Vowden P. Doppler and the ABPI: how good is our understanding?. J Wound Care
2001;10:197-202.
40 McGrae McDermott M, Greenland P, Liu K, et al. The ankle- brachial index is associated with leg
function and physical activity: the walking and leg circulation study. Ann Intern Med 2002;136:873883.
41 Cutajar CL, Marston A, Newcombe JF. Value of cuff occlusion pressures in assessment of peripheral
vascular disease. B M J 1973;2:392-395.
42 Newman AB, Siscovick DS, Manolio TA, Polak J, Linda PF, Borhani NO. Ankle-arm index as a
marker of atherosclerosis in the Cardiovascular Health Study. Circulation 1993;88:837-845.
43 Brown RS, Hollier LH, Batson RC. Noninvasive evaluation of the diabetic extremity. Am Surg
1980;46:481-484.
44 Lennihan R, MacKereth MA. Ankle blood pressure as a practical aid in vascular practice. Angiology
1975;26:211-224.
45 Fishbane S, Youn S, Kowalski EJ, Frei GL. Ankle-arm blood pressure index as a marker for
atherosclerotic vascular disease in hemodialysis patients. Am J Kidney Dis1995;25:34-39.
46 McGrae McDermott M, Mehta S, Liu K, Guralnik JM, Martin GJ, Criqui MH. Leg symptoms, the
ankle-brachial index, and walking ability in patients with peripheral arterial disease. J Gen Intern Med
1999;14:173-181.
47 Criqui MH, Fronek A, Barrett-Connor E, Klauber MR, Gabriel S, Goodman D. The prevalence of
peripheral arterial disease in a defined population. Circulation 1985;71:510-515.
48 Winter-Warnars HAO, van der Graaf Y, Mali WPTM. Ankle- arm index, angiography, and duplex
ultrasonography after recanalization of occlusions in femoropopliteal arteries: comparison of longterm results. Cardiovasc Intervent Radiol 1996;19:234-238.
49 McGrae McDermott M, Feinglass J, Slavensky R, Pearce WH. The ankle-brachial index as a
predictor of survival in patients with peripheral vascular disease. J Gen Intern Med 1994;9:445-449.
50 Kotzloff L, Collins GJ, Rich NM, McDonald PT, Collins GT, Clagett GP. Fallibility of postoperative
Doppler ankle pressure in determining the adequacy of proximal arterial revascularization. Am J
Surg 1980;139:326-329.
51 Carter SA, Tate RB. Value of toe pulse waves in addition to systolic pressures in the assessment of
the severity of peripheral arterial disease and critical limb ischemia. J Pediatr Surg 1996;24:258-265.
52 Criqui MH, Fronek A, Klauber MR, Barrett-Connor E, Gabriel S. The sensitivity, specificity, and
predictive value of traditional clinical evaluation of peripheral disease: results from noninvasive
testing in a defined population. Circulation 1985;71:516-522.
53 Atkins LAM, Gardner AW. The relationship between lower extremity functional strength and severity
of peripheral arterial disease. Angiology 2004;55:347-355.
82
54 Jönsson B, Skau T. Ankle-brachial index and mortality in a cohort of questionnaire recorded leg pain
on walking. Eur J Vasc Endovasc Surg 2002;24:405-410.
55 Carter SA. Clinical measurements of systolic pressures inlimbs with arterial occlusive disease. JAMA
1969;207: 1869-1873.
56 Carter SA. Indirect systolic pressures and pulse waves in arterial occlusive disease of the lower
extremities. Circulation 1968;37:624-637.
57 Eickhoff JH, Engell HC. Diagnostic correctness of distal blood pressure measurements in patients
with arterial insufficiency. Scand J Clin Lab Invest 1980;40:647-652.
58 Smith FB, Lee AJ, Frice JF, van Wijk MCW, Fowkes FGR. Changes in ankle brachial index in
symptomatic and asymptomatic subjects in the general population. J Vasc Surg 2003;38:1323-1330.
59 Ono K, Tsuchidam A, Kawai H, et al. Ankle-brachial blood pressure index predicts all-cause and
cardiovascular mortality in hemodialysis patients. J Am Soc Nephrol 2003;14:1591-1598.
60 Fowkes FGR, Housley E, Cawood EHH, Macintyre CCA, Ruckley CV, Prescott RJ. Edinburgh artery
study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general
population. Int J Epidemiol 1991;20:384-392.
61 Vaccoro O, Pauciullo P, Rubba P, et al. Peripheral arterial circulation in individuals with impaired
glucose tolerance. Diabetes Care 1985;8:594-597.
62 Long J, Modrall JG, Parker BJ, Swann A, Welborn MB, Anthony T. Correlation between anklebrachial index, symptoms, and health-related quality of life in patients with peripheral vascular
disease. J vasc Surg 2004;39:723-727.
63 Criqui MH, Browner D, Fronek A, Klauber MR, Coughlin SS, Barrett-Connor E. Peripheral arterial
disease in large vessels is epidemiologically distinct from small vessel disease: an analysis of risk
factors. Am J Epidemiol 1989;129:1110-1119.
64 Ogren M, Hedblad B, Jungquist G, Isacsson SO, Lindell SE, Janzon L. Low ankle-brachial pressure
index in 68-year old men: prevalence, risk factors and prognosis. Eur J Vasc Surg 1993;7:500-506.
65 McLafferty RB, Moneta GL, Taylor LM, Porter JM. Ability of ankle-brachial index to detect lowerextremity atherosclerotic disease progression. Arch Surg 1997;132:836-841.
66 Levi M, Hart W, Legemate DA. Fysisch diagnostiek bij perifeer arterieel vaatlijden. Ned Tijdschr
Geneeskd 2001;145:902-905.
67 Aso Y, Tayama K, Takanashi K, Inukai T, Takemura Y. Changes in skin blood flow in type 2 diabetes
induced by prostacyclin: association with ankle brachial index and plasma thrombomodulin levels.
Metabolism 2001;50:568-572.
68 Carmelli D, Fabsitz RR, Swan GE, Reed T, Miller B, Wolf PA. Contribution of genetic and
environmental influences to ankle-brachial blood pressure index in the NHLBI Twin Study. Am J
Epidemiol 2000;151:452-458.
69 Hämäläinen H, Rönnema T, Halonen JP, Toikka T. Factors predicting lower extremity amputations in
patients with type 1 or type 2 diabetes mellitus: a population-based 7- year follow-up study. J Intern
Med 1999;246:97-103.
70 Popper JS. Ankle-brachial blood pressure in men > 70 years of age and the risk of coronary heart
disease. Am J Cardiol 2000;86:280-284.
71 Leng GC, Fowkes FGR, Lee AJ, Dunbar J, Housley E, Ruckley CV. Use of ankle brachial pressure
index to predict cardiovascular events and death: a cohort study. B M J 1996;313:1440-1443.
83
72 Vogt MT, Cauley JA, Newman AB, Kuller LH, Hulley SB. Decreased ankle-arm blood pressure index
and mortality in elderly women. J A M A 1993;270:465-469.
73 Piecuch T, Jaworski R. Resting ankle-arm pressure index in vascular diseases of the lower
extremities. Angiology 1989;40:181-185.
74 McKenna M, Wolfson S, Kuller L. The ratio of ankle and arm arterial pressure as an independent
predictor of martality. Atherosclerosis 1991;87:119-128.
75 Lamerton AJ, Nicolaides AN, Sutton D, Eastcott HHG. The haemodynamic effects of percutaneous
transluminal angioplasty. Int Angiol 1985;4:93-97.
76 Dieter RS, Tomasson J, Gudjonsson T, et al. Lower extremity peripheral arterial disease in
hospitalized patients with coronary artery disease. Vasc Med 2003;8:233-236.
77 Fowkes FGR. The measurement of atherosclerotic peripheral arterial disease in epidemiological
surveys. Int J Epidemiol 1988;17:248-254.
78 Hiatt WR, Hoag S, Hamman RF. Effect of diagnostic criteria on the prevalence of peripheral arterial
disease - the San Luis Valley Diabetes Study. Circulation 1995;91:1472-1479.
79 Kaiser V, Kester ADM, Stoffers HEJH, Kitslaar PJEHM, Knottnerus JA. The influence of experience
on the reproducibility of the ankle-brachial systolic pressure ratio in peripheral arterial occlusive
disease. Eur J Vasc Endovasc Surg 1999;18:25-29.
80 Kaiser V, Hooi JD, Stoffers HEJH, Boutens EJ, van der Laan JR. NHG-Standaard Perifeer arterieel
vaatlijden, http://www.artsennet.nl/data/custom/nhg/standaards/m13/std.htm (accessed Nov 16,
2004).
81 Stoffers HEJH, Rinkens PELM, Kester ADM, Kaiser V, Knottnerus JA. The prevalence of
asymptomatic and unrecognized peripheral arterial occlusive disease. Int JEpidemiol 1996;25:282290.
82 Hiatt WR, Jones DN. The role of hemodynamics and duplex ultrasound in the diagnosis of peripheral
arterial disease. Curr Opin Cardiol 1992;7:805-810.
83 Franzeck UK, Bernstein EF, Fronek A. The effect of sensing site on the limb segmental blood
pressure determination. Arch Surg 1981;116:912-916.
84 Stoffers HEJH, Kester ADM, Kaiser V, Rinkens PELM, Kitslaar PJEHM, Knottnerus A. The
diagnostic value of the measurement of the ankle-brachial systolic pressure index in primary health
care. J Clin Epidemiol 1996;49:1401-1405.
85 Simon A, Papoz L, Ponon A, Segond P, Becker F, Drouet L. Feasibility and reliability of ankle-arm
blood pressure index in preventive medicine. Angiology 2000;51:463-471.
86 Bernstein EF, Witzel TH, Stotts JS, Fronek A. Thigh pressure artifacts with noninvasive techniques
in an experimental model. Surgery 1981;89:319-323.
87 Yao JST. Discussion on: Variability of Doppler ankle pressures with arterial occlusive disease: an
evaluation of ankle index and brachial-ankle pressure gradient. Surgery 1981;89:137.
88 Yao JST. Pressure measurements in the extremity In: Bernstein, EF, Vascular Diagnosis. St. Louis:
Mosby, 1993, pp 169-175.
89 Bollinger A, Barras JP, Mahler F. Measurement of foot artery blood pressure by micromanometry in
normal subjects and in patients with arterial occlusive disease. Circulation 1976;53:506-512.
90 Mätzke S, Franchena M, Albäck A, Railo M, Lepäntalo M. Ankle-brachial index measurements in
critical leg ischaemia - the influence of experience on reproducibility. Scand J Surg 2003;92:144-147.
84
91 Rahiala E, Tikanoja T. Suspicion of aortic coartation in an outpatient clinic: how should blood
pressure measurements be performed?. Clin Physiol 2001;21:100-104.
92 Klein S, Hage JJ, Lagerweij M, van der Horst CMAM. Ankle arm index versus angiography in the
preassessment of the fibula free flap. Plast Reconstr Surg 2003;111:735-743.
93 Radak D, Labs KH, Jäger KA, Bocjic M, Popovic AD. Doppler based diagnosis of restenosis after
femoropopliteal percutaneous translumenal angioplasty: sensitivity and specificity of the ankle
brachial pressure index versus changes in absolute pressure values. Angiology 1999; 50:111-122.
94 Decrinis M, Doder S, Stark G, Pilger E. A prospective evaluation of sensitivitiy and specificity of the
ankle brachial index in the follow-up of superficial femoral artery occlusions treated by angioplasty.
Clin Invest 1994;72:592-597.
95 Delis KT, Nicolaides AN, Wolfe JHN, Stansby G. Improving walking ability and ankle brachial
pressure indices in symptomatic peripheral vascular disease with intermittent pneumatic foot
compression: a prospective controlled study with one-year follow-up. J Vasc Surg 2000;31:650-661.
96 Jirkovska A, Boucek P, Woskova V, Bartos V, Skibova J. Identification of patients at risk for diabetic
foot: a comparison of standardized noninvasive testing with routine practice at community diabetes
clinics. J Diabetes Complications 2001;15:63-68.
97 Papamichael CM, Lekakis JP, Stamatelopoulos KS, Papaioannou TG, Alevizaki MK, Cimponeri AT.
Ankle-brachial index as a predictor of the extent of coronary atherosclerosis and cardiovascular
events in patients with coronary artery disease. Am J Cardiol 2000;86:615-618.
98 Boyko EJ, Ahroni JH, Davignon D, Stensel V, Prigeon RL, Smith DG. Diagnostic utility of the history
and physical examination for peripheral vascular disease among patients with diabetes mellitus. J
Clin Epidemiol 1997;50:659-668.
99 Abu Rahma AF, Boland J, Diethrich EB. Correlation of the resting and exercise Doppler ankle-arm
index to angiographic findings. Angiology 1980;31:331-336.
100 Sumner DS. Noninvasive assessment of peripheral arterial occlusive disease. In: Rutherford RB
(ed), Vascular Surgery. Philadelphia: Saunders, 1989.
101 Futran ND, Stack BC, Zachariah AP. Ankle-arm index as a screening examination for fibula free
tissue transfer. Ann Otol Rhinol Laryngol 1999;108:777-780.
102 Howell MA, Colgan MP, Seeger RW, Ramsey DE, Sumner DS. Relationship of severity of lower
limb peripheral vascular disease to mortality and morbidity: a six-year followup study. J Vasc Surg
1989;9:691-697.
103 Hooi JD, Stoffers HEJH, Kester ADM, van Ree JW, Knottnerus JA. Peripheral arterial occlusive
disease: prognostic value of signs, symptoms, and the ankle-brachial pressure index. Med Decis
Making 2002;22:99-107.
104 Hooi JD, Kester ADM, Stoffers HEJH, Overdijk MM, van Ree JW, Knottnerus A. Incidence of and
risk factors for asymptomatic peripheral arterial occlusive disease: a longitudinal study. Am J
Epidemiol 2001;153:666-672.
105 Yao JST. New techniques in objective arterial evaluation. Arch Surg 1973;106:600-604.
106 Fronek A, Johanson KH, Dilley RB. Noninvasive physiologic tests in the diagnosis and
characterization of peripheral arterial occlusive disease. Am J Surg 1973;126:205-214.
85
107 Stoffers HEJH, Kester ADM, Kaiser V, Rinkens PELM, Knottnerus JA. Diagnostic value of signs
and symptoms associated with peripheral arterial occlusive disease seen in general practice. A
multivariable approach. Med Decis Making 1997;17:61-70.
108 Chamberlain J, Housley E, Macpherson AIS. The relationship between ultrasound assessment and
angiography in occlusive arterial disease of the lower limb. Br J Surg1975;62:64-67.
109 Yao JST. Hemodynamic studies in peripheral arterial disease. Br J Surg 1970;57:761-766.
110 Ray SA, Buckenham TM, Belli AM, Taylor RS, Dormandy JA. The relationship between the delayed
improvement in ankle-brachial pressure index and changes in limb volume following balloon
angioplasty for leg ischaemia. Eur J Vasc Endovasc Surg 1997;14:114-117.
86
Chapter
7
Evaluation of the lower limb vasculature
before free fibula flap transfer.
A prospective blinded comparison
between magnetic resonance
angiography and digital subtraction
angiography
Klein S,
van der Lienden KP
van ’t Veer M
Smit JM
Werker PMN
Accepted for publication in Microsurgery
87
88
Summary
Introduction: The aim of this study was to compare magnetic resonance angiography (MRA) with digital
subtraction angiography (DSA) in the preoperative assessment of crural arteries and their skin
perforators prior to free fibular transfer.
Patients and methods: 15 consecutive patients, scheduled for free vascularized fibular flap transfer,
were subjected to DSA as well as MRA of the crural arteries of both legs (n=30). All DSA and MRA
images were assessed randomly, blindly and independently by two radiologists. Each of the assessors
scored the degree of stenosis of various segments on a 5 point scale from 0 (occlusive) to 4 (no
stenosis). The Cohen's Kappa coefficient was used to assess the agreement between DSA and MRA
scores. In addition, the number of cutaneous perforators were scored and the assessors were asked if
they would advise against fibula harvest and transplantation based on the images.
Results: A Cohen’s Kappa of 0.64, indicating “substantial agreement of stenosis severity scores” was
found between the two imaging techniques. The sensitivity of MRA to detect a stenosis compared to
DSA was 79% (CI95%:60-91), and a specificity of 98% (CI95%: 97-99). In 53 out of 60 assessments,
advice on suitability for transfer were equal between DSA and MRA. The median number of cutaneous
perforators that perfuse the skin overlying the fibula per leg was one for DSA as well as MRA (p =
0.142).
Conclusions: A substantial agreement in the assessment of stenosis severity was found between DSA
and MRA. The results suggest that MRA is a good alternative to DSA in the preoperative planning of
free fibula flap transplantation.
Introduction
The fibula free flap has become the microsurgeon’s workhorse for the reconstruction of osseus or
osteocutaneous defects of the head and neck region,(1,2) trunk,(3,4) and extremities.(3-6) Peripheral
arterial occlusive disease or congenital anomalies of any of the crural vessels may hamper its use
because of potential insufficiency of the flap’s peroneal vascular pedicle or impairment of the remaining
anterior and posterior tibial vascular supply after harvest of these peroneal vessels.(7-9) To prevent
these type of complications, preoperative assessment of the crural vascular supply is essential in free
fibula flap candidates.(10,11) Among the various methods available to asses the vascular supply,
conventional selective digital subtraction angiography is the generally accepted standard.(9,12-14)
Some surgeons still rely on physical examination and the ankle-arm index, previous research has
however shown that both are not accurate enough to detect legs or arteries with subclinical peripheral
arterial occlusive disease or vascular variation.(11) Although angiography does have this ability, it is an
expensive and invasive procedure featuring a morbidity rate of 0.5 to 3.9 percent, a complication rate of
3 tot 5 percent, and a mortality rate of 0.03 percent.(7,15,16) Hence, a safer, and at least equally
accurate alternative for routine preoperative angiographic assessment would be favorable.
Magnetic resonance angiography (MRA) is increasingly being applied in the assessment of crural
arterial disease over the last few years.(17) Improvements of technology and protocols have allowed
MRA to provide high levels of reliability in the detection of such arterial disease, as compared with
digital subtraction angiography (DSA).(17) Still, this reliability has, to date, predominantly been
assessed in patients with vascular disease. MRA has been proposed by some reconstructive surgeons
as a method of vascular assessment prior to free fibular transfer in non-symptomatic patients,(18-20)
89
but these series lacked a comparison to DSA. To date, we could only find two studies comparing DSA
and MRA in free fibular transfer.(21,22) In these studies MRA was, however, made after the harvest of
the fibular flap. Moreover, until now there has not been one study concerning the possibility to visualize
skin perforators with DSA. There have been four studies trying to answer this question for MRA,
however without comparing this method to other imaging technique available.(20,23-25)
The aim of this study was to compare MRA with DSA in the preoperative assessment of the crural
arteries and its skin perforators prior to free fibular transfer.
Patients and methods
Demographics
Over a period of three years 15 consecutive patients scheduled for free fibular transfer participated in
this study after they had provided oral and written consent. The population consisted of 12 men and 3
women with a median age of 49 year (range, 22 - 66 year). In all but one patient, fibula free flap
transplantation was planned for mandibula reconstruction. In the remaining patient, the flap was
planned for reconstruction of a defect in the humerus after resection of osteosarcoma (n = 1). In the
included patients a DSA as well as a MRA was made of the crural arteries of both legs.
The study protocol was approved by the medical ethical committees of the Academic Medical Center
and the Netherlands Cancer Institute-Antoni van Leeuwenhoek Hospital, Amsterdam (MEC 02/239
#03.17.0112).
Digital subtraction angiography
Following application of local anesthesia at the groin using 10 ml of mepiracain 2% without adrenaline
(Scandicain; Astra Zeneca, Sodertalje, Sweden) puncture of the femoral artery was performed through
either the ipsilateral or contralateral femoral artery approach. Via a 5-F introducer sheath (Terumo,
Tokyo, Japan), a 5-F catheter (Berenstein; William A. Cook, Queensland, Australia) was placed at the
level of the proximal popliteal artery. Selective DSA images were obtained from the distal thigh to the
foot with use of a standard angiographic unit (Integris V-5000; Philips, Hamburg, Germany), and a 17to 38-cm field-of-view image intensifier and a 1024 x 1024 display matrix or a comparable angiographic
unit (Axiom Artis; Siemens, Erlangen, Germany). For these selective arteriograms, 15–30 ml of contrast
agent (Accupaque 300; Bracco, Milano/Italy) was injected at a rate of 5–7 ml/sec. The typical total
volume of injected contrast material ranged between 28 and 40 ml. Anteroposterior and lateral views
were chosen to best depict the arteries at the discretion of the interventional radiologist by using a field
of view of 28 cm to 38 cm.
All angiographies were performed at the Department of Interventional Radiology of the Academic
Medical Center in Amsterdam by the same investigator (KvL) to prevent bias.
Magnetic resonance angiography
The MRAs were made using a 1.5 T MR-scanner (Signa echospeed, General Electric Medical Systems,
Milwaukee WI, USA) using a phased array spine coil and the 9.0 software-release. The patients were
positioned feet first in a supine position with the feet positioned in slight plantar flexion. Pillows and
straps were used to prevent motion-artifacts. The legs were positioned in such a way that the field of
90
view contained the entire lower leg from the joint space of the knee to the phalanges of the foot. This
positioning took 10 minutes on average.
A 3D-Time-of-Flight (TOF) fast spoiled gradient sequence was used in a sagittal plane, featuring the
parameters FA 35, TE 1.5 ms., TR 6 ms, FOV 46, slab thickness 3 mm,1.5 mm spacing, matrix 256 x
192, nex 1, rec FOV 0.8, zip 512. First, a non-enhanced basic series needed for the subtractions was
obtained. Subsequently, 0.25 ml/kg of gadopentetate dimeglumine (Magnevist, Schering, Berlin,
Germany) was administered by a power injector into the medial cubital vein with a flow-rate of 3 ml/s
(maximum of 25 ml gadopentetate dimeglumine). Hereafter the cubital vein was flushed using 30 ml of
saline 0.9% administered at the same flow-rate. Six identical consecutive series were made of the same
field-of-view. With this technique no popliteal bolus timing is needed and the series with the most
optimal arterial enhancement, without venous overprojection, can be chosen for evaluation. The actual
scan time, including localizers, measured 6 to 8 minutes.
Subtractions were made of the six phases that each showed a different vascular filling dependent on the
flow velocity and the extent of the proximal stenosis or occlusion. Multiplanar reconstruction (MPR) and
maximum intensity projection (MIP)-reconstructions were made in anteroposterior, anterolateral, and
posterolateral directions at the workstation (General Electric Medical Systems, Milwaukee WI).
Assessment protocol
All DSA and MRA images were assessed randomly, blindly and independently by two interventional
radiologists with at least 6 years of experience with vascular imaging techniques. Each of the assessors
scored the degree of diameter stenosis of the various segments of all three crural arteries as: 0
occlusion/aplasia; 1- severe stenosis (>50%); 2- moderate stenosis (25 - 50%); 3- light stenosis (<25%),
4- no stenosis; or 5- ‘not informative’, e.g. because of movement, artifacts or insufficient supply of
contrast.
For this assessment, 11 segments of the crural arteries were distinguished: (1) the popliteal artery; (2)
the tibioperoneal trunk; (3-11) the proximal, middle, and distal parts of each of the three crural arteries.
The number of skin perforators of the peroneal artery was also scored.
In addition to the segmental scoring and the number of skin perforators, the assessors were asked to
indicate for each leg and each angiography, be it the DSA or MRA, whether or not they advised against
transplantation of a fibula free flap based on their assessment of the angiography. Criteria to advice
against transplantation were a stenosis or occlusion of the anterior and posterior tibial artery or of the
peroneal artery. Furthermore congenital anomalies at the trifurcation of the vessels were a reason to
advice against transplantation.
A total of 330 segments (30 legs x 11 segments) were scored by each assessor. One hundred and fifty
out of those 330 segments were in the proximal section, 90 in the middle section and 90 in the distal
section. Given the two assessors, we compared a total of 660 scores per imaging method.
91
Data analysis
When a segment was scored as ‘not informative’ by an observer in either DSA or MRA or both groups,
this segment was excluded from further analysis. Technique agreement and inter-observer agreement
was assessed with Cohen's Kappa coefficient for the remaining scores. This coefficient can be
interpreted as follow: 0-0.2 as little to no match, 0.2-0.4 as a fair match, 0.4-0.6 as a moderate match,
0.6-0.8 as a substantial match, and 0.8-1.0 as an almost perfect match.(26)
The scores were divided into two groups for calculation of the sensitivity and specificity of detecting a
stenosis. The first group included the no stenosis scores (score 4). The second group included the
occlusion/aplasia, severe, moderate and light stenosis scores (scores 0, 1, 2 and 3). For this calculation
the DSA group was used as the gold standard.
The popliteal artery, the tibioperoneal trunk and the proximal thirds of the three crural arteries were
referred to as “proximal segments”. The middle thirds of the peroneal artery and the anterior and
posterior tibial artery were referred to as “mid segments”, and likewise the distal thirds of these three
crural vessels as “distal segments”.
The Wilcoxon signed-rank test was used to compare the number of skin perforators scored per leg. A pvalue <0.05 was considered to be statistically significant. Statistical analyses were performed using IBM
SPSS statistics version 19 (IBM Corporation, Armonk, NY)
Results
DSA vs MRA
With the use of DSA, 570 segments were scored as no stenosis, 22 as light stenosis, seven as
moderate stenosis, one as severe stenosis. No segments were scored as being occluded. Sixty
segments were judged as not-informative. In the MRA group 612 segments were scored as no stenosis,
37 as light stenosis, three as moderate stenosis. Severe stenosis and occlusion were both scored once,
six segments were judged as not-informative. Cohen’s Kappa for agreement was found to be 0.64
between DSA and MRA implying a substantial agreement of stenosis severity scores. A Cohen’s Kappa
for agreement between the two radiologists was found to be 0.55 implying a moderate match.
Total counts of stenosis severity scores per segment are shown in Table 1 for DSA and MRA. The
counts are categorized by location (proximal, mid, and distal). An example of a DSA is depicted in figure
1. As we scored the MRA by scrolling through the multi-slice pictures its not representative to show a
single-slice picture or a multi-slice reconstruction.
The sensitivity of MRA to detect a stenosis compared to DSA is 79% (CI95%: 60-91). The specificity
was found to be 98% (CI95%: 97-99).
92
Table 1: Overview of the DSA and MRA scores, as well as the scores per segment.
Stenosis severity
DSA
MRA
Occlusion/
Severe
Moderate
Light
No
Not-
aplasia
stenosis
stenosis
stenosis
stenosis
informative
(0)
(1)
(2)
(3)
(4)
(5)
proximal
0
0
1
7
288
4
middle
0
0
3
5
165
7
distal
0
1
3
10
117
49
Total
0
1
7
22
570
60
proximal
0
0
0
14
286
0
middle
0
0
1
11
168
0
distal
1
1
2
12
158
6
Total
1
1
3
37
612
6
Figure 1: DSA image of a lower leg
93
Judgment of suitability for fibula transfer
In 53 out of the 60 assessments, the opinions for suitability for fibula transfer were equal between DSA
and MRA. Six legs were rated as suitable on MRA while unsuitable on DSA. One leg was rated
unsuitable on MRA while suitable on DSA. In all legs that were rated as unsuitable for transfer on either
DSA or MRA or both, fortunately the contra lateral leg was scored suitable for transfer on both DSA and
MRA and was therefore used.
Skin perforators from peroneal artery
The median number of skin perforators from the peroneal artery per leg was 1 for both DSA (range 0-3,
mean 0.84 and of SD 0.96) and MRA (range 0-2, mean of 0.67 and SD of 0.61) (p = 0.142).
Figure 2: Number of perforators per leg for DSA and MRA
94
Discussion
The aim of this study was to compare MRA with DSA in the preoperative assessment of the crural
arteries and its skin perforators prior to free fibular transfer. For this purpose we prospectively compared
the MRA and DSA images of 15 patients. We found a substantial agreement (Cohen’s Kappa 0.64)
between DSA and MRA concerning the ability to detect and assess arterial stenosis and its degree. A
sensitivity of 79% (CI95%: 60-91) and a specificity of 98% (CI95%: 97-99) were found for MRA, relative
to the gold standard DSA. In most cases there was an agreement for suitability for fibula transfer
between DSA and MRA and the number of septocutaneous perforators scored was found not to differ
significantly between DSA and MRA. The two assessors in this series of non-symptomatic patients
observed no congenital anatomical vascular anomalies. At the same time, the assessors felt that the
image of the crural vascular tree (poplitial, peroneal and anterior and posterior tibial arteries) on MRA
was of such a good quality, that an abnormal branching pattern of the arteries would definitively have
been detected by MRA.
Ideally the sensitivity score should have been higher to safely conclude that MRA could replace DSA in
the work-up of free fibula flap transfer. It should be noted that the majority of the assessed segments in
our study had little or no stenosis, hence the rather wide range of the confidence interval for sensitivity
79% (CI95%: 60-91). Higher sensitivities of over 95% for MRA have been reported in studies comparing
MRA to DSA for peripheral arterial occlusive disease.(27,28) Moreover, the MRI, type of contrast, and
software package used might have been of influence in the detection of stenosis.
The reasons for imaging prior to free fibular flap harvesting are to rule out (congenital) vascular
anomalies and peripheral arterial occlusive disease, which can both jeopardize the viability of the
harvested flap or the donor limb. Apart from DSA and MRA, other techniques such as colour duplex
sonography (CDS)(7) and computed tomographic angiography (CTA)(29-32) have also been described
for this purpose. CDS has been reported to be able to accurately map the crural vessels and cutaneous
perforators. The advantages it offers are its low costs, no morbidity and detailed information about the
flow in vessels.(7) Disadvantages are however that it is less reproducible because of its real life
dynamics and that it does not lead to a 2D or 3D image like the other vascular imaging techniques.(33)
CTA has been reported to accurately predicted the course and location of the peroneal artery and
perforators as well. Like MRA the advantage of CTA is that it provides an accurate 3D image of the
artery, its perforators and the surrounding anatomy. Compared to MRA, CTA is able to visualize vessels
with a smaller diameter of up to 0.3 mm. A major disadvantage of CTA however is its use of radiation
and the necessity to use iodinate contrast medium.(29-32) Especially the vasospastic action of the
contrast medium is a serious drawback, because it can make the accurate assessment of small-calibre
vessels difficult.(34)
A number of authors have reported their findings with the use of solely MRA prior to free fibula flap
transfer.(18,20,23,24) Most were positive about the possibilities that MRA offers although there is
discussion about the ability of MRA to detect skin perforators. In their prospective study Fukaya et
al.(20) preoperatively investigated among others the number of skin perforators found on MRA. Among
the seven patients included (seven legs), a total of 13 perforators were detected on the MRA, of which
12 were confirmed during surgery. Based on these findings they encourage the use of MRA prior to free
fibula flap transfer. In their retrospective report Miller et al.(24) were less positive about the ability of
95
MRA to detect septocutaneous perforators. They reviewed the radiological findings of 123 patients who
underwent preoperative MRA as part of surgical planning for fibula free flap tissue transfer and
compared these to intraoperative finings. Two patients were found to have a single perforator originating
from the posterior tibial artery during surgery, while MRA suggested the perforators to arise from the
peroneal artery. Analysis of the entire cohort demonstrated that agreement between the number of
perforators documented on MRA and the number found intraoperatively approached zero (unweighted κ
= -0.088, P = .04).
The report of Holzle et al.(22) in 2011 also compared DSA and MRA in the evaluation of the vessels of
the lower leg in 15 patients scheduled for microsurgical fibular transfer. While in our study MRA and
DSA was performed prior to surgery, in their study DSA was performed preoperatively and MRA
postoperatively. Both techniques were used to compare vessel size at the trifurcation; hypoplastic or
missing vessels; appreciable stenosis, vascular occlusions, and atherosclerotic malformations; and
overall vascular anatomy of both limbs. With regards to vessel size, the results showed that for the
operated lower leg the anterior and posterior tibial arteries were judged larger on the postoperative MRA
than on the preoperative DSA (p = 0.045). In the non-operated leg there was no difference. The findings
with regards to the other variables were alike in their study. They concluded that high resolution MRA
enabled a reliable judgment of the lower leg vessels equal to DSA. A drawback of this study is however
that the MRA was performed post-operatively, and therefore the comparison of the DSA and MRA
images the operated legs is impossible.
In some studies the findings of MRA were compared to the intraoperative findings. We chose not to do
this, since we only harvested the fibula from one side. Furthermore the main disadvantage of the
comparison of MRA to intraoperative findings is, that the only information to compare the MRA to is the
site, which is surgically assessed. There is no detailed information on the percentage of stenosis and
potential variations in the rest of the vessel or the other crural arteries,
In our study more legs are scored unsuitable for transfers using DSA compared to MRA, two versus
seven respectively. This could not be explained by the difference in scores of stenosis and
occlusion/aplasia (scores 0, 1, 2 and 3 combined) between DSA and MRA as they were about equal, 29
versus 32, respectively. This difference was most likely caused by the higher number of segments
scored as ‘not informative’ in the DSA group, 60 versus 6, respectively. We choose to exclude these
scores, as they were often the result of technical difficulties or errors, for example by movement of the
patient during the imaging. As we wanted to compare the outcome of both techniques and not the
technical difficulties and errors we decided to exclude these scores from further analysis. However
because of the fact that the score ‘not informative’ was given ten times more often in the DSA group,
this could be an argument to favor MRA over DSA.
Strong points of our study were that prospective design as well as the random and blind assessment of
all images, decreasing the chance on bias. Furthermore, DSA as well as MRA investigations were
performed prior to surgery, in contrary to some of the study described above.(22) There are however
also limitations. First of all the outcome of DSA and MRA were not compared to the ‘real’ golden
standard; the anatomy of each scanned individual. Due to obvious reason it is not possible to dissect
the involved legs after performing the scans. As DSA has been the standard for years in the work-up of
free fibula flap transfers, we chose to use DSA as the standard in our study and compare MRA to DSA
instead of the other way around. It is however important to realize that in the comparison of two
96
methods with one of them defined as the “golden standard” the other method can only score as good
as, and definitely not better than the golden standard.
Conclusion
We found a substantial agreement between DSA and MRA concerning the ability to detect and assess
the degree of arterial stenosis. In the majority of the cases there was an agreement for suitability for
fibula transfer between DSA and MRA. The number of skin perforators scored did not differ significantly
between DSA and MRA.
Based on the results of this study we believe that MRA is a good alternative to DSA in the pre-operative
work-up of free fibula flap transfers. It is less invasive, does not use nefrotoxic contrast mediums, and
patients are not exposed to ionizing radiation. Although the sensitivity found in this study for the MRA
should have ideally been higher, we believe that with current developments with regards to MRA this
will only improve over the years.
97
References
1 Lutz BS, Wei FC. Microsurgical workhorse flaps in head and neck reconstruction. Clin Plast Surg.
2005;32:421-30.
2 Wein RO, Lewis AF. Synchronous reconstruction of the floor of mouth and chin with a single skin
island fibular free flap. Microsurgery. 2008;28:223-6.
3 Korompilias AV, Paschos NK, Lykissas MG, Kostas-Agnantis I, Vekris MD, Beris AE. Recent
updates of surgical techniques and applications of free vascularized fibular graft in extremity and
trunk reconstruction. Microsurgery. 2011;31:171-5.
4 Clemens MW, Chang EI, Selber JC, Lewis VO, Oates SD, Chang DW. Composite extremity and
trunk reconstruction with vascularized fibula flap in postoncologic bone defects: a 10-year
experience. Plast Reconstr Surg. 2012;129:170-8.
5 Beris AE, Lykissas MG, Korompilias AV, Vekris MD, Mitsionis GI, Malizos KN, Soucacos PN.
Vascularized fibula transfer for lower limb reconstruction. Microsurgery. 2011;31:205-11.
6 Soucacos PN, Korompilias AV, Vekris MD, Zoubos A, Beris AE. The free vascularized fibular graft
for bridging large skeletal defects of the upper extremity. Microsurgery. 2011;31:190-7.
7 Futran ND, Stack BC Jr., Payne LP. Use of color Doppler flow imaging for preoperative assessment
in fibular osteoseptocutaneous free tissue transfer. Otolaryngol. Head and Neck Surg. 1997;117:6603.
8 Rosson GD, Singh NK. Devascularizing complications of free fibula harvest: Peronea arteria magna.
J Reconstr Mircosurg 2005;21:533-8.
9 Seres L, Csaszar J, Voros E, Borbely L. Donor site angiography befor mandibular reconstruction with
fibula free flap. J Craniofac Surg 2001;12:608-13.
10 Klein S, Hage JJ, Woerdeman LA. Donor-site necrosis following fibula free-flap transplantation: a
report of three cases. Microsurgery. 2005;25:538-42.
11 Klein S, Hage JJ, van der Horst CM, Lagerweij M. Ankle-arm index versus angiography for the
preassessment of the fibula free flap. Plast Reconstr Surg. 2003;111:735-43.
12 Young DM, Trabulsy PP, Anthony JP. The need for preoperative leg angiogrphay in fibula free flaps.
J.Reconstr. Microsurg. 1994;10:283-9.
13 Carroll WR, Esclamado R. Preoperative vascular imaging for the fibular osteocutaneous flap. Arch.
Otolaryngol. Head Neck Surg. 1996;122:708-12.
14 Blackwell KE. Donorsite evaluation for fibula free flap transfer. Am. J. Otolarynogl 1998;19:89-95.
15 Shehadi WH, Toniolo G. Adverse reactions to contrast media: A report from the Committee on
Safety of Contrast Media of the International Society of Radiology. Radiology 1980;137:299-302.
16 Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981;138:273-81.
17 Collins R, Burch J, Cranny G, Aguiar-Ibanez R, Craig D, Wright K, Berry E. Duplex ultrasonography,
magnetic resonance angiography, and computed tomography
angiography for diagnosis and
assessment of symptomatic, lower limb peripheral arterial disease: systematic review. BMJ.
2007;334:1257-65.
18 Mast BA. Comparison of magnetic resonance angiography and digital subtraction angiography for
visualization of lower extremity arteries. Ann Plast Surg 2001;46:261-4.
19 Lorenz RR, Esclamado R. Preoperative magnetic resonance angiography in fibular-free flap
reconstruction of head and neck defects. Head Neck 2001;23:844-50.
98
20 Fukaya E, Saloner D, Leon P, Wintermark M, Grossman RF, Nozaki M. Magnetic resonance
angiography to evaluate septocutaneous perforators in free fibula flap transfer. JPRAS
2010;63:1099-104.
21 Hölzle F, Franz EP, von Diepenbroick VH, Wolff KD. Evaluation der Unterschenkelarterien vor
mikrochirurgischem Fibulatransfer: MRA vs. DSA. Mund Kiefer GesichtsChir 2003;7:246-53.
22 Hölzle F, Rstow O, Rau A, Mücke T, Loeffelbein DJ, Mitchell DA, Stimmer H, Wolff KD, Kesting MR.
Evaluation of the vessels of the lower leg before microsurgical fibular transfer. Part II: Magnetic
resonance angiography for standard preoperative assessment. Br J Oral Maxillofac Surg
2011;49:275-80.
23 Fukaya E, Grossman RF, Saloner D, Leon P, Nozaki M, Mathes SJ. Magnetic resonance
angiography for free fibula flap transfer. J Reconstr Microsurg. 2007;23:205-11.
24 Miller ME, Moriarty JM, Blackwell KE, Finn JP, Yiee JH, Nabili V. Preoperative magnetic resonance
angiography detection of septocutaneous perforators in fibula free flap transfer. Arch Facial Plast
Surg. 2011;13:36-40.
25 Sandhu GS, Rezaee RP, Wright K, Jesberger JA, Griswold MA, Gulani V. Time-resolved and boluschase MR angiography of the leg: branching pattern analysis and identification of septocutaneous
perforators. AJR 2010;195:858-64.
26 Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics
1977;33:159–74.
27 Berg F, Bangard C, Bovenschulte H, Hellmich M, Nijenhuis M, Lackner K, Gossmann A. Feasibility
of peripheral contrast-enhanced magnetic resonance angiography at 3.0 Tesla with a hybrid
technique: comparison with digital subtraction angiography. Invest Radiol. 2008;43:642-9.
28 Wang CC, Liang HL, Hsiao CC, Chen MC, Wu TH, Wu CJ, Huang JS, Lin YH, Pan HB. Single-dose
time-resolved contrast enhanced hybrid MR angiography in diagnosis of peripheral arterial disease:
compared with digital subtraction angiography. J Magn Reson Imaging. 2010;32:935-42.
29 Ribuffo D, Atzeni M, Saba L, Guerra M, Mallarini G, Proto EB, Grinsell D, Ashton MW, Rozen WM.
Clinical study of peroneal artery perforators with computed tomographic angiography: implications for
fibular flap harvest. Surg Radiol Anat. 2010;32:329-34.
30 Garvey PB, Chang EI, Selber JC, Skoracki RJ, Madewell JE, Liu J, Yu P, Hanasono MM. A
Prospective study of preoperative computed tomographic angiographic mapping of free fibula
osteocutaneous flaps for head and neck reconstruction. Plast Reconstr Surg. 2012;130:542e-50e.
31 Rozen WM, Ashton MW, Stella DL, Phillips TJ, Taylor GI. Magnetic resonance angiography and
computed tomographic angiography for free fibular flap transfer. J Reconstr Microsurg. 2008;24:4578.
32 Wang WH, Deng JY, Li M, Zhu J, Xu B. Preoperative three-dimensional reconstruction in
vascularized fibular flap transfer. J Craniomaxillofac Surg. 2012;40:599-603
33 Smit JM, Klein S, Werker PM. An overview of methods for vascular mapping in the planning of free
flaps. J Plast Reconstr Aesthet Surg. 2010;63:e674-82.
34 Singh J, Daftary A. Iodinated contrast media and their adverse reactions. J Nucl Med Technol.
2008;36:69-77.
99
100
Chapter
8
Is there an indication for digital
subtraction angiography in the
assessment of irradiation-induced
vascular damage prior to free flap
surgery by the means of the internal
mammary vessels?
Klein S
Hoving S
Werker PMN
Russel NS
Accepted for publication in Journal of Reconstructive Microsurgery
101
102
Summary
Introduction: Secondary breast reconstruction is increasingly performed after postmastectomy
radiotherapy. Damage to blood vessel walls is one of the adverse effects of irradiation therapy, which
may jeopardize reconstructive free flap surgery. It would be of great value to be informed about the
quality of the recipient vessel prior to reconstructive surgery. The aim of this study was to prospectively
assess the value of pre-operative angiography in the assessment of radiation-induced arterial damage
and to relate the findings to the degree of vascular damage found during the operation and with
histology.
Material & Methods: Women, who had been treated with thoracic radiotherapy and required free flap
breast reconstruction, were included. Pre-operative angiographic, intraoperative vessel quality and
histological findings of vessels were scored and compared, together with the occurrence of
postoperative complications.
Results: In 34 patients a total of 40 free flaps breast reconstruction were performed. Twenty-one internal
mammary arteries had been within the field of irradiation. In only two out of six patients with aberrant
angiographies the internal mammary artery has been within the field of irradiation.
Conclusion: The damage to the internal mammary vessels cannot always be detected pre-operatively
by angiography, or even by intraoperative examination.
Introduction
Free flap reconstructive surgery is often performed after oncologic resections and adjuvant irradiation
therapy.(1) Already in 1899 it was reported that irradiation can lead to arteritis and atherosclerotic
disease,(2-6) and the overall incidence of radiation-related vascular damage in general ranges from
30% to 89%.(7,8)
To date, the accuracy of detection of irradiation damage of arteries with angiography prior to free flap
reconstructions has not been assessed. Hence, the aim of this study was to prospectively assess the
value of pre-operative angiography in the assessment of radiation-induced arterial damage and to relate
the findings to the degree of vascular damage found during the operation and with histology. For this
purpose we choose to study the internal mammary artery (IMA) in patients scheduled for free flap
mammary reconstruction after radiotherapy to the thoracic region.
Materials and Methods
The Institutional Review Board of the Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital,
approved this study. Patients gave their written informed consent for the study.
From October 2003 through March 2007, highly selective digital subtraction angiography of the internal
mammary arteries (IMA) was preoperatively performed in 34 female candidates for free deep inferior
epigastric perforator (DIEP) flap mammary reconstruction. Clinical and treatment data were extracted
from the patients’ medical and radiotherapy chart and radiation dose plan was used to determine the
actual radiation dose delivered to the internal mammary vessels.
A radiologist blinded for the history of the patients scored the angiographies for the anatomy and course
of the IMA and its degree of intra-luminal atherosclerotic changes.
During free flap breast reconstruction the internal mammary arteries and veins (IMV) were dissected by
removing the overlying medial part of the rib. Before performing the anastomosis in the subcostal space
103
the vessels were scored for quality of their wall and lumen diameter. Representative 5 mm specimens of
both the IMA and the deep inferior epigastric arteries of 25 patients were obtained. All specimens were
fixed in 1% paraformaldehyde for 48 hours after which 4 µm thick paraffin embedded cross sections
were made. The sections were stained with hematoxylin and eosin and with Lawson's elastin stain.(9)
Because irradiation may cause a thickening of the media or intima with simultaneous decrease of the
luminal diameter in medium-sized arteries,(10,11) the wall-to-lumen area ratio and the media-to-lumen
area ratio were calculated.(9) In this measurement the vessel wall was defined as intima and media. All
measurements and calculations were done by one investigator (S.H.).
Statistical analysis
Combining the radiological, intra-operative, and histological observations, we calculated the correlation
between the results of the various observations, using the chi-square test to assess the predictive value
of the angiography for radiation-related vascular damage and surgical outcome. Statistical significance
was accepted at p – value ≤ 0.05. Distinction was made between internal mammary vessels that had
received the full therapeutic radiation dose at the position of the internal mammary vessels, and those
that had not.
Patient characteristics
The 34 women had a mean age of 45 years (range 32 - 60 yrs) and all had previously undergone
mediastinal radiotherapy for Hodgkin lymphoma (n = 2) or unilateral thoracic radiotherapy because of
breast cancer (n = 32), a mean of 3.77 years (range 0.5 - 17 yrs) before angiography. The mean total
dose had been 43.69 Gy (range 17.5-50.0 Gy). In 19 of the 32 unilaterally irradiated patients, the
internal mammary vessels had been included within the field of radiation. 30 of our 34 patients (36 out
of 40 flaps) were treated with chemotherapy prior to breast reconstruction.
A bilateral internal mammary angiography was performed in 31 women. Unilateral angiography of only
the irradiated IMA was performed in the remaining three patients. Thus, a total of 65 IMA’s were
depicted angiographically.
In all 34 women, free flap breast reconstruction was performed unilaterally (n = 28) or bilaterally (n = 6).
An in vivo assessment of the vessels at 21 previously irradiated internal mammary recipient sites and
19 previously non-irradiated recipient sites could be made by the surgeon performing the microsurgery.
For the histological assessment technically insufficient slides had to be excluded, leaving a total of 14
irradiated and 11 non-irradiated specimens of the IMA’s and 27 non-irradiated specimens of the deep
inferior epigastric arteries.
Results
1. Correlation between internal mammary angiography and previous radiotherapy
The anatomy, course and contrast-filling diameter of the IMA was scored as normal on 60 of the 65
angiographies. Twenty-one of these 60 IMA’s had previously received the full dose of radiation. Three of
the five remaining arteries that were scored as angiographically aberrant had previously not been
irradiated (Table 1).
104
Because just 40% of the angiographically aberrant arteries and 35% of the angiographically normal
arteries had been irradiated, we conclude that radiation therapy does not correlate with angiographically
obvious damage to the IMA (p > 0.2).
Table 1
Correlation between pre-operative angiographic findings (n = 65) and parasternal irradiation therapy
(number of aberrant angiographies per side)
Angiografic findings
Parasternal
irradiation
No parasternal
irradiation
Narrowing
0
3
Deviation of vessel course
2
0
Normal
21
39
2. Correlation between internal mammary angiography and intra-operative observations
Of the 40 arteries, that could be assessed intra-operatively, 37 did not show macroscopic changes of
the vascular wall or diameter. In one of the remaining three cases there was an evident intra-luminal
plaque and in the other two there were mild fibrotic changes with increased stiffness of the arterial wall
(Table 2). There was no correlation between the preoperative angiographic findings and the intraoperative observations of the arteries (p > 0.2).
Tabel 2
Comparison between intra-operative finding of the internal mammary artery (IMA) and preoperative
angiographic findings and preoperative parasternal irradiation therapy
(Number of reconstructed sides; n = 40)
Operative findings
Aberrant
Normal
Parasternal
No parasternal
angiography
angiography
irradiation
irradiation
Mild fibrosis IMA
1
1
1
1
Extended fibrosis IMA/
0
1
0
1
4
33
20
17
intraluminal plaque
Normal IMA
3. Correlation between radiotherapy and intra-operative observations
We found no significant correlation between the intraoperative findings on the internal mammary
vessels and exposure to previous radiation (p > 0.2). Arterial wall fibrosis was observed in one of the
irradiated internal mammary vessels and two of the non-irradiated receptor arteries (Table 2).
105
The IMV’s showed a macroscopically normal vascular wall in 28 of the 40 recipient sites (Tabel 3).
Because of internal mammary venous insufficiency, the microsurgeons switched to using to the jugular
vein (n = 1) or the cephalic vein (n = 1) for the venous anastomosis in two of the twelve cases, which
exhibited abnormalities. As for the arteries, there was no significant correlation (p > 0.2) between intraoperative venous macroscopical appearance and previous radiation exposure (Table 3).
Table 3
Correlation between intra-operative finding of the internal mammary vein (IMV) and preoperative
parasternal irradiation therapy (number of reconstructed sides; n = 40)
Operative findings
Parasternal irradiation
No parasternal irradiation
Small diameter IMV
0
3
Brittle wall of IMV
4
2
Mild fibrosis IMV
1
2
Normal IMV
16
12
4. Comparison of histological vessel measurements between irradiated and non-irradiated
arteries.
There were no significant differences in mean “wall-to-lumen area ratio” between the irradiated IMA,
non-irradiated IMA and the IEA, and neither in their mean “media-to-lumen area ratio” (Table 4).
Comparing the media-lumen and the wall-lumen area ratio of irradiated and non-irradiated IMA’s
graphically (Figure), it was apparent that both ratios were nearly congruent in both groups. Of the three
patients, who underwent bilateral breast reconstruction with just one of the IMA’s being irradiated, no
significant difference in wall-to-lumen area ratio could be found between the irradiated and the nonirradiated IMA.
Table 4
Wall-to-lumen and media-to-lumen area ratios of irradiated (n = 14) and non-irradiated (n = 11) internal
mammary arteries (IMA) and the inferior epigastric arteries (IEA) in the pedicle of the flaps (n = 27)
Wall-to-lumen area ratio
Media-to-lumen area ratio
Mean
SD
Mean
SD
Irradiated IMA
2.12
0.88
1.99
0.80
Non-irradiated IMA
1.72
0.42
1.61
0.43
IEA of DIEP-flap
2.31
1.23
2.25
1.16
106
Figure
Graph comparing media-lumen and wall-lumen area ratios of irradiated and non-irradiated IMA’s
5. Correlation between internal mammary angiographic findings and outcome of surgery
A total of 40 free flaps were transplanted in 34 patients. Thirty-three breast reconstructions were free of
complications (Table 5). There were seven postoperative flap complications, leading to complete flap
loss in four of them (10%). Five of the seven complicated breast reconstructions were associated with a
normal angiography. Hence, the preoperative angiographic findings did not correlate with the surgical
outcome (0.1> p >0.05).
Of those two patients with an aberrant angiography and a postoperative complication,
one patient had a combined arterial and venous thrombosis resulted in total flap loss after revision. This
patient had shown a medial deviation of the IMA on the preoperative angiography but intraoperative,
macroscopic inspection of the internal mammary vessels had not revealed any vascular disorder. The
other patient had a partial fat necrosis, while the pre-operative angiography had shown a small lumen of
the IMA but this could not be confirmed during surgery.
107
Table 5
Correlation between postoperative complications and preoperative angiographic findings and preoperative parasternal irradiation therapy (number of reconstructed sides, n = 40)
Postoperative complications
Aberrant
Normal
Parasternal
No parasternal
angiography
angiography
irradiation
irradiation
0
2
1
1
0
2
1
1
1
0
1
0
0
1
1
0
Partial fat necrosis
1
0
0
1
None
3
30
17
16
Venous thrombosis with successful reanastomosis
Venous thrombosis with unsuccessful
reanastomosis and flap loss
Arterial and venous thrombosis with
unsuccessful reanastomosis and flap
loss
Too fragile vessels for anastomosis with
flap loss
6. Correlation between previous radiotherapy and outcome of surgery
Seventeen of the 33 uneventful breast reconstructions and four of the seven complicated breast
reconstructions were in the group of previous radiotherapy to the internal mammary vessels (Table 5).
Hence, there was no statistical significant correlation between the two (p > 0.2).
Discussion
Even though irradiation may induce atherosclerotic narrowing in areas unusual for the natural
occurrence of arterial disease,(12) only few authors studied the possible irradiation-related damage of
the IMA as a source of free flap failure.(13) Likewise, reports on the outcome of IMA-grafts in cardiac
revascularisation surgery in patients with a history of thoracic radiation therapy are sparse. Some of
these reports suggest that the irradiated IMA is unsuitable for cardiac grafting,(14,15) whereas other
studies found no adverse effects on revascularization outcome.(16-18)
We found that internal mammary chain radiotherapy does not always cause angiographic or
macroscopically obvious damage to the IMA. Furthermore, it was not possible to preoperatively predict
postoperative complications on the basis of internal mammary angiography. Still, we found some
correlation between previous radiotherapy and postoperative complications, but no significant or specific
correlation between such irradiation and total flap loss.
Potential shortcomings of our study
Because we only performed arterial angiographies, it was not possible to assess the IMV’s
preoperatively. To image the IMV an extra venography via the sternum or a CT-scan would have been
necessary. To depict the veins the patients would have had to be exposed to extra radiation and the risk
108
of complications.(19-20) Therefore in our study a normal angiography does not rule out possible
surgical insufficiency of the IMV, which might be better depicted by duplex sonography.
Second, our series are too small to distinguish between short-term and long-term prevalence of
radiation effects.
Last, we did not reckon with possible vascular damage induced by chemotherapy. Still, because 30 of
our 34 patients (36 out of 40 flaps) were treated with chemotherapy prior to breast reconstruction no
statistically significant observations could have been made from such data.
Quality of internal mammary vessels
In unirradiated IMA’s atherosclerotic changes are rare. The incidence of artherosclerotic changes
detected by angiography in literature ranges from 0% to 11.1%.(21-25) and the incidence of
histologically proven arterial stenosis has been found to vary from 4.2% to 12.4%.(26,27) Still, none of
the 215 specimen of the study by Kay et al.(26) and only one specimen in a series of 160 patients
studied by Sisto and Isola(27) showed more than a 50% reduction in lumen diameter of the internal
mammary artery.
Hardly any data are available on the incidence of atherosclerosis in series of irradiated IMA’s. As such
van Son et al. reported on four patients with radiation-induced coronary artery disease in whom internal
mammary angiography was performed.(16) The IMA was judged not patent in two patients and, in both,
histological examination confirmed dense fibrosis of the IMA-wall with complete obliteration of the
lumen. The arteries of the other two patients showed a slightly thickened adventitia with minimal signs
of fibrosis.
In our study, angiographic changes of the IMA-wall have been found in five out of 65 angiographies
(7.7%). Macroscopic changes of the IMA-wall were intraoperatively observed in three out of 40 patients
(7.5%). Compared to the literature both prevalences are well within the normal range.(21-27)
Histology of irradiation damage
Grassman was the first to describe vascular lesions of radiodermatitis and noted the swelling and
2
proliferation of the endothelium to the point of projection into the lumen of small arteries. Irradiation
may produce arterial damage varying from intimal thickening, fragmentation of the elastic lamina, overproduction of elastic tissue, chronic inflammation and necrosis of the adventitia, hyaline thickening, to
thrombosis and the production of collagen in the larger arteries.(3-5)
Medium-sized arteries such as the IMA are less sensitive to irradiation than blood capillaries,(10-11)
but these arteries may still show prominent adventitial fibrosis, subendothelial and intimal accumulation
of foam cells, and hyaline deposition in the media as a result of irradiation.(10) Russell et al. described
an increase in proteoglycan content of the intima of irradiated IMA vessels compared with unirradiated
IMA, from 65% to 73%. However the collagen content between irradiated an unirradiated IMA’s was not
found to differ significantly.(9)
We were unable to find differences in wall-to-lumen area ratio, or media-to-lumen area ratio in the
irradiated arteries and the non-irradiated arteries. Probably, the dose of radiation in our series was
relatively low compared to the radiation sensitivity of the IMA; or the follow-up interval was too short.
Still, linking the intraoperative observations on the IMV’s with prior irradiation therapy, it was our
impression that irradiated veins were a bit more brittle than non-irradiated veins.
109
Discussion of our observations
We observed a low, statistically insignificant correlation between prior irradiation and surgical outcome.
But there was no correlation between angiographic aberrances and surgical outcome. This may be
explained, as flap survival is not only related to the quality of the IMA and IMV, but is influenced by
many more factors, like skills, technique, ischemia, perforator quality and smoking habits of the patient.
We found a slight predominance of macroscopically brittle walls among the previously irradiated IMV’s.
Other intra-operatively observed venous aberrances such as a smaller diameter or mild fibrosis did not
correlate with previous radiotherapy.
Our series may be too small to allow a statistically warranted conclusion regarding the correlation
between previous radiotherapy exposure of vessels and surgical outcome. Still, it offers a good
impression of the lack of correlation between pre-operative angiography, possible irradiation damage to
the IMA, and the intraoperative finding.
Conclusion and alternative techniques of vascular evaluation
We conclude that there is no value of pre-operative highly selective digital subtraction angiography in
the assessment of the degree of radiation-induced arterial damage and that there is no correlation to the
vascular damage found during the operation and with histology. Thus digital subtraction angiography is
not helpful as a preoperative assessment tool for the selection of potential candidates for free flap
mammary reconstruction. Furthermore, such angiography is expensive and invasive. Although its risk of
complications is small, these risks are not negligible as they include, arterial aneurysm at the puncture
side (0.4-2.0%), anaphylactic shock resulting from allergy to the contrast medium (0.2-0.4%), superficial
phlebitis and edema (17.9%), renal impairment (<2%), and hemorrhagic (11.4%) and thrombotic events
(0-4.0%).(28-33) Because of these risks and because we found the subtraction angiography unhelpful,
color flow doppler sonography, computer tomography angiography, or magnetic resonance angiography
might be better screening methods.(18,22,34,35)
110
References
1 Pomahac B, Recht A, May JW, et al. New trends in breast cancer management: is the era of
immediate breast reconstruction changing? Ann Surg 2006; 244: 282-288
2 Ganssman A. Zur Histologie der Roentgenulcera. Fortschr Roentgenstr 1899; 2: 199-207
3 Lindsay S, Kohn HI, Dakin RL, et al. Aortic arteriosclerosis in the dog after localized aortic Xirradiation. Circ Res 1962; 10: 51-60
4 Sams A. Histological changes in the larger blood vessels of the hind limb of the mouse after xirradiation. Int J Rad Biol 1965; 9: 165-174
5 Fonkalsrub EW, Sanchez M, Zerubavel R, et al. Serial changes in arterial structure following
radiation therapy. Surg Gynecol Obstet 1977; 145: 395-400
6 Hoopes PJ, Gillette EL, Withrow SJ. Intraoperative irradiation of the canine abdominal aorta and
vena cava. Int J Radiat Oncol Biol Phys 1987; 13: 715-722
7 Silverberg GD, Britt RH, Goffinet DR. Radiation-induced carotid artery disease. Cancer 1978; 41:
130-137
8 Moritz MW, Higgins RF, Jacobs JR. Duplex imaging and incidence of carotid radiation injury after
high-dose radiotherapy for tumors of the head and neck. Arch Surg 1990; 125: 1181-1187
9 Russell NS, Hoving S, Heeneman S, et al. Novel insights into pathological changes in muscular
arteries of radiotherapy patients. Radiother Oncol 2009; 92: 477-483.
10 Chuang VP. Radiation-induced arteritis. Semin Roentgenol 1994; 29: 64-69
11 Fajardo LF. The pathology of ionizing radiation as defined by morphologic patterns.
Acta Oncol 2005; 44: 13-22
12 Nylander G, Pettersson F, Swedenborg J. Localized arterial occlusions in patients treated with pelvic
field radiation for cancer. Cancer 1978; 41: 2158-2161
13 Tran NV, Chang DW, Gupta A, et al. Comparison of immediate and delayed free TRAM flap breast
reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg
2001;108: 78-82.
14 Schulman HE, Korr KS, Myers TJ. Left internal thoracic artery graft occlusion following mediastinal
radiation therapy. Chest 1994;105: 1881-1882.
15 Renner SM, Massel D, Moon BC. Mediastinal irradiation: A risk factor for atherosclerosis of the
internal thoracic arteries. Can J Cardiol 1999; 15: 597-600
16 van Son JAM, Noyez L, van Asten WNJC. Use of internal mammary artery in myocardial
revascularization after mediastinal irradiation. J Thorac Cardiovasc Surg 1992; 104: 1539-1544
17 Gansera B, Haschemi A, Angelis I, et al. Cardiac surgery in patients with previous carcinoma of the
breast and mediastinal irradiation: is the internal thoracic artery graft obsolete? Thorac Cardiovasc
Surg 1999; 47: 376-380.
18 Nasso G, Canosa C, de Fillipo CM, et al. Thoracic radiation therapy and suitability of internal
thoracic arteries for myocardial revascularization. Chest 2005; 128: 1587-1592
19 Chasen MH, Charnsangavej C. Venous chest anatomy: clinical implications. Eur J Radiol 1998; 27:
2-14
20 Chiappa S, Coopmans de Yoldi G, Magri M. Trans-sternal phlebography of the internal mammary
veins. Am J Roentgenol Radium Ther Nucl Med 1960; 83: 320-334
111
21 Finci L, Meier B, Steffeniono G, et al. Nonselective preoperative digital subtraction angiography of
internal mammary arteries. Cathet Cardiovasc Diagn 1990; 19: 13-16
22 Sons HJ, Marx R, Godehardt E, et al. Duplex sonography of the internal thoracic artery. J Thorac
Cardiovasc Surg 1994; 108: 549-555
23 Sons HJ, Godehardt E, Kunert J, et al. Internal thoracic artery: prevalence of atherosclerotic
changes. J Thorac Cardiovasc Surg 1993; 106: 1192-1195
24 Ochi M, Yamauchi S, Yajima T, et al. The clinical significance of performing preoperative
angiography of the internal thoracic artery in coronary artery bypass surgery. Surg Today Jpn J Surg
1998; 28: 503-508
25 Bauer EP, Bino MC, von Segesser LK, et al. Internal mammary artery anomalies. Thorac Cardiovasc
Surgeon 1990; 38: 312-315
26 Kay HR, Korns ME, Flemma RJ, et al. Atherosclerosis of the internal mammary artery. Ann Thorac
Surg 1976; 21: 504-507
27 Sisto T, Isola J. Incidence of atherosclerosis in the internal mammary artery. Ann Thorac Surg 1989;
47: 884-886
28 Skinner JS, Jackson MJ, Gholkar A, et al. Cortical blindness during left internal mammary
angiography. Intern J Cardiol 1995; 52: 119-123
29 Feit A, Reddy CV, Cowley C, et al. Internal mammary artery angiography should be a routine
component of diagnostic coronary angiography. Cathet Cardiovasc Diagnos 1992; 25: 85-90
30 Agaba AE, Hardiment K, Burch N, et al. An Audit of Vascular Surgical Intervention for Complications
of Cardiovascular Angiography in 2324 Patients from a Single Center. Ann Vasc Surg 2004; 18: 470473
31 Saklayen MG, Gupta S, Suryaprasad A, et al. Incidence of atheroembolic renal failure after coronary
angiography. A prospective study. Angiology 1997; 48: 609-613.
32 Lensing AW, Prandoni P, Buller HR, et al. Lower extremity venography with iohexol: results and
complications. Radiology 1990; 177: 503-505
33 AbuRahma,AF, Powell M, Robinson PA. Prospective study of safety of lower extremity phlebography
with nonionic contrast medium. Am J Surg 1996; 171: 255-260
34 Evans GRD, David CL, Loyer EM, et al. The long term effects of internal mammary chain irradiation
and its role in the vascular supply of the pedicled transverse rectus abdominis musculocutaneous
flap breast reconstruction. Ann Plast Surg 1995; 35: 342-348
35 Glassberg RM, Sussman SKK, Glickstein MF. CT anatomy of the internal mammary vessels:
Importance in planning percutaneous transthoracic procedures. AJR 1990; 155: 397-400
112
Chapter
9
Summary and general discussion
113
114
Summary and general discussion
Summary
Trauma, oncological resections and pressure sores can lead to major soft tissue defects, which can
create a challenge for surgical closure. Reconstructive surgery has seen great development since the
early 1960s, when the concept of axial vessels became mainstay. In the past 20 years, enormous
progress has been made in flap design and more and more flaps are based on perforating vessels that
branch off and are traced back to well-known vessels, thereby limiting donor-site morbidity. The exact
location of perforators, however, varies significantly, and preoperative vascular mapping has been
introduced to help identify the dominant perforator and its course and, as such, speed up flap harvest.
The aim of this thesis was to investigate the need for and compare various techniques for preoperative
vascular assessment in free flap reconstruction.
In chapter two the evolution of the reconstructive flap surgery is described. This evolution followed the
increased understanding of the vascular anatomy of the skin and subcutis. While the first local flaps for
reconstruction were based on a random vascular pattern, the next generation of flaps was based on an
axial vascular supply. In the Western world, surgeons such as Esser and Machot were the pioneers in
this field around the turn for the nineteenth century. (1,2) Due to the influence of surgeons, like Gillies,
the principle of axial pattern flaps was not expanded until their reinvention in 1969 by Stuart Milton. The
first axial pattern flaps were based on well-known vessels from the anatomy book, such as the radial
artery for the radial forearm flap and the thoracodorsal vessels for the latissimus dorsi flap. The harvest
of these flaps, although in that time revolutionary, is nowadays looked upon as relatively straightforward,
due to their predictable anatomy. In the recent years, more and more flaps have been described that
are based on a single perforating vessel. By the use of such perforator flaps, proper blood supply to the
flap can be combined with less morbidity at the donor site.(3)
In chapter three the “condito sine qua non” of a good vascular supply to a free flap and the donor site is
illustrated with the example of the complicated case histories of three patients in whom a free fibular flap
was harvested. The free fibula flap is the microsurgeon’s workhorse for the reconstruction of osseous or
osteocutaneous defects. Donor-site morbidity of this flap is reported to occur infrequently, and is
generally considered minor and transient. Anatomical variations of vascular patterns, prior vascular
trauma or atherosclerosis can however jeopardize the survival of the free flap and the preservation of
the tissues surrounding the donor site. The drama of flap failure and the occurrence of severe
complications at the donor site stress the importance of decent vascular assessment as part of the
preoperative work-up.
In order to validate a technique and the results of several studies, the method of measurement ought to
be standardized. However this is not always the case, as is presented in the general review in chapter
four about the example of the measurement and calculation of the ankle-arm index (AAI). Since its
introduction in 1950, a variety of methods of measurement and calculation have been used. This has
resulted in variations of its normal range and difficulty in comparing study results. Hence, the objective
of the study depicted in this chapter was to analyze the various methods used to assess AAI and its
115
normal range and to construct a standardized method to assess AAI based on that analysis. This study
resulted in an inventory of the disparate AAI methods and its normal range reported in 100 randomly
selected publications and a recommendation for standardization. We concluded, that the left arm
pressure ought to be used as denominator and the mean of pressures of both tibial arteries of each leg
ought to be used for the numerator of the AAI for that leg. We advocate 0.90 as the cut-off value to
distinguish patients who need further arterial assessment.
The fifth chapter gives an overview of the various methods for vascular mapping of flaps together with
their advantages and disadvantages. The pro’s and con’s of the hand-held Doppler, colour duplex,
digital subtraction angiography (DSA), computed tomographic angiography (CTA) and magnetic
resonance angiography (MRA) are reviewed and discussed. As CTA and MRA are able to produce
detailed 3D images of the vasculature and its surrounding structures, these methods currently are
thought to be the best methods available for mapping the vasculature of donor sites of perforator flaps
with variable anatomy such the upper thigh, the donor site of the Anterolateral Thigh flap (ALT), and the
lower abdomen, donor site of the Deep Inferior Epigastric Perforator (DIEP). In flaps with standard
anatomy and superficial vasculature hand held Doppler remains the method of choice.
In chapter six the ankle-arm index (AAI) is compared to the current golden standard, the Digital
Subtraction Angiography (DSA) for the assessment of the donor site of the fibula free flap. As peripheral
arterial occlusive disease or congenital anomalies of the major crural arteries may limit the use of the
fibula free flap, these conditions should be detected preoperatively. Since DSA has drawbacks, a safer,
cheaper, more accurate and noninvasive alternative is desirable. We tested the hypothesis that AAI of
each of the three crural arteries, combined with pencil Doppler examination of the peroneal skin
perforators, would provide adequate information to restrict the use of angiography to cases in which the
outcomes of either or both of these options are insufficient. The ankle-arm index data of each of the
three crural arteries, as well as pencil Doppler examination of the peroneal skin perforators of both legs
of nine prospectively included patients and the nonoperated legs of 13 retrospectively included patients,
were compared statistically in four different ways with the preoperative angiographic findings. The
conclusion that could be drawn was that combined ankle-arm index and pencil Doppler examination is
not accurate enough to detect legs or arteries with subclinical peripheral arterial occlusive disease or
vascular variation and, hence, is not a sufficient basis on which to develop the surgical plan for a fibula
free flap.
In Chapter seven 3D-TOF Magnetic Resonance Angiography (MRA) is compared to the current golden
standard, the Digital Subtraction Angiography (DSA) for the vascular assessment of the fibula free flap
donor site. Fifteen consecutive patients, scheduled for free vascularized fibular flap transfer, were
subjected to DSA as well as MRA of the crural arteries of both legs (n=30). Two radiologists randomly,
blindly and independently assessed all DSA and MRA images. Each of the assessors scored the
degree of stenosis or hypoplasia of various segments on a 5-point scale from 0 (occlusive) to 4 (no
stenosis). In addition, the number of cutaneous perforators was scored and the assessors were asked if
they would advise against fibula harvest and transplantation based on the images. Substantial
agreement of stenosis severity scores was found between the two imaging techniques. The sensitivity
116
of MRA to detect a stenosis compared to DSA was 0.79, and a specificity of 0.98. In 53 out of 60
assessments, advice on suitability for transfer was equal between DSA and MRA. And the median
number of cutaneous perforators per leg was one for DSA as well as for MRA (p = 0.142). The results of
this study suggest that MRA is a good alternative to DSA in the preoperative planning of free fibula flap
transplantation.
A successful transplantation of tissue is not only dependent on the good vascular supply of the flap, but
also on the condition of the vessels at the recipient site. As presented in chapter eight anatomic
variations, atherosclerosis and irradiation damage to the acceptor vessels can result in a challenging
and troublesome microsurgical procedure. The possibility to preoperatively assess the presence of
atherosclerosis or irradiation damage to the vessels is studied at the example of the internal mammary
artery that is used as recipient vessel during free flap breast reconstructions. Pre-operative angiography
findings were compared to the degree of vascular damage found during the operation, the clinical
course of the reconstruction and the histology of segments of the recipient artery. A total of 34 patients
were included with the intention of free flap breast reconstruction after radiation therapy. In total 40 free
flaps were transplanted for breast reconstruction. Twenty-one internal mammary arteries had been
within the field of irradiation and 19 out of field. In only two out of six patients with aberrant
angiographies the internal mammary artery had been within the field of radiation. Based on this study
the conclusion had to be drawn, that the damage to the internal mammary vessels cannot always be
detected pre-operatively by angiography, nor by intraoperative examination.
General Discussion
As stated before, in flaps with a variable vascular anatomy or with a suboptimal vascular state, for
example due to atherosclerosis, it is important to be informed about course and quality of the vessels
before the flap elevation starts. Not only the flap pedicle needs to be in good shape to prevent serious
complications after transfer, but also the vessels at the recipient site need to be of good quality, while
the remaining vessels at the donor site need to be able to supply the donor site after the harvest of the
flap. This thesis aimed to shed light on preoperative vascular mapping in different type of
reconstructions and to investigate which method is the most valuable for which type of reconstructions.
All vascular mapping methods have their own advantages and disadvantages.(4-6)
The ankle-arm index (AAI), is a blood pressure index, in which the left arm pressure ought to be used as
denominator and the mean of pressures of both tibial arteries of each leg ought to be used for the
numerator of the AAI for that leg. The advantages are its non-invasiveness, small size, low costs,
portability and the ease to perform the examination. But the disadvantages are its insensitivity to detect
moderate or mild stenosis and it’s lack on information on the vascular state of a single crural vessel.
The advantages of the hand-held Doppler (HDD) are comparable to the AAI: its non-invasiveness, small
size, low costs, portability and the ease to perform the examination. In addition, there are special
sterilized probes available for intra-operative use. The main disadvantage of the most widely used
Doppler probe (8 MHz) is, that it only detects vessels to a depth of 20 mm. Besides, one can never
know for sure what vessel is producing the Doppler signal picked up by the HHD. Furthermore, this
technique does not create a three-dimensional (3D) image of the vasculature and its surrounding
anatomy than can be stored and retrieved later.
117
Similar to HHD, Color Duplex Sonography (CDS) is non-invasive. An advantage compared with the
HHD is its ability to offer more information about anatomy of the vessel and its perforators in reference
to its surrounding tissues, and it can quantitatively analyze which perforator is the dominant one. The
disadvantage of CDS however, is the fact that only skilled personnel, who also have knowledge of freeflap anatomy, can perform the investigation. In addition, it is less reproducible because of its real-life
dynamics. Another disadvantage in comparison to CTA, MRA and DSA is that CDS - just as HHD does not reproduce a 2D or 3D image of the complete vascular anatomy, which can be used by the
surgeon during flap design or flap elevation
The reported advantages of Digital Subtraction Angiography (DSA) include the facts that it gives a 2D
image of the intraluminal vascular anatomy and information about atherosclerotic changes. A
disadvantage of DSA is that it is a time-consuming, invasive technique necessitating the use of iodinatecontrast medium, which may cause vascular or renal damage as well as allergic reactions. In addition,
there is a radiation dose to be considered. The vasoconstricting effect of the contrast medium, make
exact measurement of the vascular diameter and the assessment of small-caliber vessels unreliable.
Furthermore, the patient has to stay in supine position after the angiography for several hours, to allow
the puncture site to seal. This makes hospital admission often mandatory and therefor makes this
imaging modality relatively expensive. Finally, there is a risk for the development of false aneurysms at
the puncture site.
The advantage Computed Tomographic Angiography (CTA) offers is that it provides an image with
accurate visual details on the intraluminal calibre and course of the vessels and their relationships with
other anatomic structures in a 3D image. This allows surgeons to develop a dissection strategy and opt
for a certain perforator prior to surgery, making the actual dissection safer and swifter. The
disadvantages of CTA are its radiation dose, which is reported to be 5.6 mSv, and the necessity to use
iodinated contrast medium with its previously listed disadvantages. Especially, the vasospastic action is
a serious drawback, because it can make the accurate assessment of small-calibre vessels difficult.
The big advantage of Magnetic Resonance Angiography (MRA) are that it works with magnetism
instead of radiation and. Depending on the software used, it can be used without a non-iodine contrast
medium, making it a relatively safe procedure for the patient. MRA produces a 3D image, which allows
surgeons to accurately assess the course and diameter of the vessels and their relation to other
surrounding structures. The reported disadvantages of MRA are its relatively high costs. Besides, it
cannot be used in claustrophobic patients or a patient with implants containing ferrous metals because
of the scatter artefacts influencing the image quality.
Apart from the descriptive studies in this thesis (chapter two, three, four and five) we investigated the
AAI, HHD, DSA and MRA in comparative studies (chapter six, seven and eight). We concluded that a
combined ankle-arm index and pencil Doppler examination is not accurate enough to detect arteries
with subclinical peripheral arterial occlusive disease or vascular variation and, hence, gives insufficient
information to develop the surgical plan for a free flap of the lower leg. Furthermore, using the example
of the free fibula flap, we drew the conclusion that MRA is a good alternative to DSA in the preoperative
planning of free flap transplantation of the lower leg. Also this thesis showed (in case of the internal
mammary vessels) that DSA is not sensitive and specific enough to detect the irradiation-induced
damage of vessels.
118
In the most recent literature there is a clear tendency towards MRA being used more over and replacing
techniques such as DSA and CTA. (7-9) In the planning of free fibula flaps, MRA is able (in
concordance to our own results) to detect hypoplastic vessels, stenoses, occlusions, or atherosclerotic
changes of the vessels, and enables both accurate assessment of the quality of the main vessels and
their septo- or musculoccutaneous perforators.(7,10-15) CTA is suggested to be a good imaging
modality for the lower leg arteries as well.(16-18) In perforator flaps the use of both MRA (8,9) and CTA
(19-24) is being described. The advantage of CTA is that it is able to detect vessels with a smaller
diameter (<0.5mm).(25) Compared to CTA the advantages of MRA are, as stated earlier, that it is
performed without radiation and there is no need for iodinate contrast medium, which makes it less
harmful for patients.(8-26) It should however be noted that prospective controlled comparative studies
about the use of MRA and CTA in the planning of free fibula flaps are lacking.
Combining the outcome of our studies and the literature we conclude that MRA and CTA are currently
the best methods available to map the vasculature of crural vessels and perforator flaps in general. (4)
In the planning of thin pedicled flaps that are planned close to a defect, in flaps with a more
straightforward anatomy and for intra-operative use, the HHD remains to be mapping method of
choice.(4) DSA is slowly fading out and CDS can be used as an alternative, whenever there are contraindications to the use of the other methods of investigation.
When evaluating the outcome of a study, especially a comparative study, it is of importance to look at
the methods used. The strong points of our studies are the prospective and comparative study design.
The investigated technique was compared to the gold standard at that moment. There are however also
limitations, especially with regard to the news value of some of this work in 2013. The field of vascular
mapping is developing so rapidly, that during the design, execution, analysis and report process new
developments make some of this work at present already a bit outdated. When we performed our study
the 1.5 Tesla 3D TOF-MRA technique was the currently used method to depict crural vessels. But from
that time MRA-techniques developed very rapidly.(27-30)
In clinically used scanners the magnetic field increased from 1.5 to 3.0 Teslas in strength. For research
purposes already 7.0 Tesla scanners are used. By the use of higher Tesla scanner it is possible to
improve the signal to noise ratio and the spatial and temporal resolution. As we used a 1.5 Tesla
scanner it is obvious, that with the use of a 3.0 Tesla scanner the resolution of the depicted vessels
could have been improved. At present time, the most clinics still use a 1,5 T MRI scanner and therefore
the outcome of this study however would probably be the same.
Parallel to the improvement of the scanners, new scan techniques have beeing developed, to depict
vessels more accurately, than with the TOF-MRA technique we used. With MRA it is possible to depict
vessels by two different techniques, the flow-dependent angiography (FDA) and flow-independent
angiography (FIA).
The FDA MRA-technique can be divided into Time-of-Flight (TOF) and Phase-Contrast (PC). In TOFMRA flowing blood gives a much higher signal than stationary tissue, but areas with slow flow or flow
that is in plane of the image may not be well visualized. With PC-MRA slow flow can be detected much
better and the velocity of moving blood can be detected as well. The disadvantage of PC-MRA is that
the flow can only acquire flow in one direction at a time. To give a complete image of flow, 3 separate
119
image acquisitions in all three directions must be computed. Despite the slowness of this method, the
strength of the technique is that in addition to imaging the flowing blood, quantitative measurements of
blood flow occur at the same time.
In general, slow blood flow is a major challenge in FDA-MRA, because the differences between the
blood signal and the static tissue signal are small. To increase blood signal, which is especially
important for very small vessels or slow flow, contrast agents may be used. Therefore ContrastEnhanced (CE) FIA-MRA has been developed. The use of contrast agent is currently the most common
method of acquiring MRA. The contrast medium is injected into a vein, and images are acquired during
the first pass of the agent through the arteries. If the timing of the scanning is correct, the images are
usually of a very high quality. But if the timing is bad or “blood-pool agents” are used the depiction of the
arteries is disturbed by venous overprojection.
Since the injection of contrast agents may be dangerous for patients with renal failure, non-contrastenhanced techniques have been developed. These methods are based on the differences of T1, T2 and
chemical shift of the different tissues of the voxel.
The acquisition of the images is changing as well from 3D, which we used, to 4D.(31) Three
dimensional data acquisition is helpful when dealing with complex vessel geometries where blood is
flowing in all spatial directions. The big advantage of the new 4D technique is that the arterial and
venous phases can be divided. In order to overcome the venous overlay and to gain dynamic flow
information the most recent development was the Triple-TWIST MRA (Time-resolved angiography With
Interleaved Stochastic Trajectories), which seems to become the new standard as imaging investigation
in patients suffering from peripheral arterial occlusive disease.(31) With this technique multiple fast
series are obtained of the same area during contrast injection. The maximum intensity image of each
series is selected. All these images together form a dynamic MRI angiography series with high
resolution, without the venous over projection.
Furthermore development of new software packages made it possible to not only asses vascular
imaging via the hospital network, but also for authorized external users. (e.g. Siemens IHE Integrating
Healthcare Enterprise – XDS-1b). This increase in accessibility of the imaging data facilitates expert
consults all over the world and preoperative surgical planning from outside the hospital.
With these improvement in technique we expect that, preoperative mapping in free flap surgery and
especially the role of MRA will get an even more dominant role and will enable surgeons to analyse
more key-aspect of the surgery prior to the surgery itself in the future.(32) Therefore, we expect that the
role of a radiologist familiar with free flap surgery will become more prominent in a reconstructive team
and will create a new dynamic within the team. To combine the necessary knowledge of the different
specialists, we believe that a reconstruction should be discussed and performed in a multidisciplinary
team. Depending on the location of an existing or expected defect, the team should consist of an
oncologist, the ablative surgeon (general surgeon, ENT surgeon, maxillofacial surgeon, orthopaedic
surgeon, gynaecologist, or neurosurgeon), a radiologist, the reconstructive plastic surgeon and an
anaesthesiologist.
A good example of how pre-operative planning is evolving is the Rohner technique for mandible or
maxilla reconstruction, in which the bone segment of a fibula or iliac crest flaps are planned prior to
surgery to fit a defect.(33-37) A CTA is made during the workup that not only allows for vascular
120
imaging, but also generates data that enables the 3D reconstruction of the bony component. The
reconstruction is then further optimized, by creating pre-manufactured cutting guides, reducing the
constraints of sometimes unpredictable intraoperative environments, and maximizing bony contact. With
these techniques the implants that are required for dental bridges can even be placed prior to the bone
transfer, creating a reconstruction, which can immediately bare weight post-surgery. (33) This CADCAM (computer-aided design and manufacturing) technology and the use of stereolithographic models
can improve the accuracy of the surgical result and the intraoperative efficiency.(34,36-38) These
techniques seem to offer improved patient outcome and the reduction of complication and non-union
rates due to this CAD-CAM approach. But there are no comparative studies to investigate the long-term
results in a larger patient group, concerning patient outcome and the reduction of complication and nonunion rates due to this CAD-CAM approach.
Suggestions on future research
Based on our finding during this research, we have several suggestions for future research in the field of
free flap surgery.
1) We showed that the Ankle-Arm Index is insufficient in the preoperative work-up for a free fibula
flap, but we believe that MRA is a good alternative to DSA in the pre-operative work-up of free
fibula flap transfers for reasons stated before. Although the sensitivity found in our study for the
MRA should have ideally been higher, we believe that with current developments with regards
to MRA this is only a matter of time. Although DSA is still advocated as the golden standard for
the detection of peripheral arterial occlusive disease in recent literature,(28) we do believe, that
this technique is becoming more and more obsolete.
To prove that CTA and MRA are just as or even more accurate than DSA in the detection of
peripheral arterial occlusive disease, a comparative study should be conducted, using the
newest scanners and scanning protocols. To investigate the resolution power and the accuracy
of these imaging modalities in the detection of vascular stenosis precisely a comparison to an
anatomical dissection is desirable. It is obvious that this can only be performed in an animal
study.
2) With regards to the preoperative mapping of the recipient vessels, DSA did not prove to be
reliable. We however believe that there are selected cases in which preoperative mapping of
the recipient vessels is necessary, for example after previous surgery or irradiation. In these
cases CTA, MRA or colour duplex could be of additional value, depending on the size and
location of the vessels. Future research can focus on the indication for and the necessity of
preoperative imaging and which method should be used.
3) As discussed above, new developments in the pre-operative planning is virtual surgery in which
key-points of the surgery can already be performed in a virtual setting. Thereby the surgery can
become more straightforward and safer and the results can be optimized. Until now CTA has
been predominantly used for these purposes. We believe however, that if MRA can be used for
these purposes this would further add to patient safety. But due to scattering with metal
implants MRI might not be the most suitable imaging method.
121
References
1 Esser JFS. Artery flaps, Antwerpen, 1929
2 Manchot C. Die Hautarterien des menschlichen Körpers, Leipzig, 1889
3 Sinna R, Boloorchi A, Mahajan AL. What should define a "perforator flap"? Plast Reconstr Surg.
2010 Dec;126(6):2258-63.
4 Smit JM, Klein S, Werker PMN. An overview of methods for vascular mapping in the planning of free
flaps. J Plast Reconstr Aesthet Surg 2010; 63: e674-682
5 Klein S, Hage JJ, van der Horst CMAM. Ankle-arm index versus angiography for the preassessment
of the fibula free flap. Plast Reconstr Surg 2003; 111: 735-743
6 Klein S, Hage JJ. General review: Measurement, calculation, and normal range of the ankle-arm
index: a bibliometric analysis and recommendation for standardization. Ann Vasc Surg 2006; 20:
282-292
7 Miller ME, Moriarty JM, Blackwell KE. Preoperative magnetic resonance angiography detection of
septocutaneous perforators in fibula free flap transfer. Arch Facial Plast Surg. 2011 JanFeb;13(1):36-40
8 Masia J, Clavero JA, Larrañaga JR. Multidetector-row computed tomography in the planning of
abdominal perforator flaps. J Plast Reconstr Aesthet Surg. 2006;59:594-9.
9 Newman TM, Vasile J, Levine JL. Perforator flap magnetic resonance angiography for reconstructive
breast surgery: a review of 25 deep inferior epigastric and gluteal perforator artery flap patients. J
Magn Reson Imaging. 2010 May;31(5):1176-84.
10 Hölzle F, Rstow O, Rau,A, Mücke T. Evaluation of the vessels of the lower leg before microsurgical
fibular transfer. Part II: Magnetic resonance angiography for standard preoperative assessment. Br J
Oral Maxillofac Surg 2010; doi:10.1016/j.bjoms.2010.05.003
11 Fukaya E, Saloner D, Leon P. Magnetic resonance angiography to evaluate septocutaneous
perforators in free fibula flap transfer. J Plast Reconstr Aesthet Surg. 2010 Jul;63(7):1099-104
12 Fukaya E, Grossman RF, Saloner D..Magnetic resonance angiography for free fibula flap transfer. J
Reconstr Microsurg. 2007 May;23(4):205-211
13 Hölzle
F,
Franz
EP,
von
Diepenbroick
VH.
Evaluation
der
Unterschenkelarterien
vor
mikrochirurgischem Fibulatransfer: MRA vs. DSA. Mund Kiefer GesichtsChir 2003; 7: 246-253
14 Mast BA. Comparison of magnetic resonance angiography and digital subtraction angiography for
visualization of lower extremity arteries. Ann Plast Surg. 2001 Mar;46(3):261-4
15 Sandhu GS, Rezaee RP, Wright K. Time-resolved and bolus-chase MR angiography of the leg:
branching pattern analysis and identification of septocutaneous perforators. AJR Am J Roentgenol.
2010 Oct;195(4):858-64
16 Chan D, Anderson ME, Domatch BL. Imaging Evaluation of Lower Extremity Infrainguinal Disease:
Role of the Noninvasive Vascular Laboratory, Computed Tomography Angiography, and Magnetic
Resonance Angiography Tech Vasc Interv Radiol. 2010 Mar;13(1):11-22.
17 Foley WD, Stonely T. CT angiography of the lower extremities. Radiol Clin North Am. 2010
Mar;48(2):367-96
18 Ribuffo D, Atzeni M, Saba L. Clinical study of peroneal artery perforators with computed tomographic
angiography: implications for fibular flap harvest. Surg Radiol Anat. 2010 Apr;32(4):329-34.
122
19 Keys KA, Louie O, Said HK, Neligan PC, Mathes DW. Clinical utility of CT angiography in DIEP
breast reconstruction. J Plast Reconstr Aesthet Surg. 2012 Oct 18. pii: S1748-6815(12)
20 Rozen WM, Anavekar NS, Ashton MW. Does the preoperative imaging of perforators with CT
angiography improve operative outcomes in breast reconstruction? Microsurgery. 2008;28(7):51623.
21 Molina AR, Jones ME, Hazari A,. Correlating the deep inferior epigastric artery branching pattern
with type of abdominal free flap performed in a series of 145 breast reconstruction patients. Ann R
Coll Surg Engl. 2012 Oct;94(7):493-5.
22 Chiu WK, Lin WC, Chen SY. Computed tomography angiography imaging for the chimeric
anterolateral thigh flap in reconstruction of full thickness buccal defect. ANZ J Surg. 2011
Mar;81(3):142-7
23 Kim EK, Kang BS, Hong JP. The distribution of the perforators in the anterolateral thigh and the
utility of multidetector row computed tomography angiography in preoperative planning. Ann Plast
Surg. 2010 Aug;65(2):155-60
24 Chen SY, Lin WC, Deng SC. Assessment of the perforators of anterolateral thigh flaps using 64section multidetector computed tomographic angiography in head and neck cancer reconstruction.
Eur J Surg Oncol. 2010 Oct;36(10):1004-11
25 Rozen WM, Whitaker IS, Stella DL. The radiation exposure of Computed Tomographic Angiography
(CTA) in DIEP flap planning: low dose but high impact. J Plast Reconstr Aesthet Surg.
2009;62:e654-5.
26 Rozen WM, Stella DL, Bowden J. Advances in the pre-operative planning of deep inferior epigastric
artery perforator flaps: magnetic resonance angiography. Microsurgery. 2009;29:119-23.
27 Kramer JH, Grist TM. Peripheral MR Angiography. Magn Reson Imaging Clin N Am. 2012
Nov;20(4):761-76.
28 Fischer A, Maderwald S, Orzada S. Nonenhanced Magnetic Resonance Angiography of the Lower
Extremity Vessels at 7 Tesla: Initial Experience. Invest Radiol. 2013 Mar 13
29 Wheaton AJ, Miyazaki M. Non-contrast enhanced MR angiography: physical principles. J Magn
Reson Imaging. 2012 Aug;36(2):286-304.
30 Nielsen YW, Thomsen HS. Contrast-enhanced peripheral MRA: technique and contrast agents. Acta
Radiol. 2012 Sep 1;53(7):769-77
31 Kinner S, Quick HH, Maderwald S. Triple-TWIST MRA: high spatial and temporal resolution MR
angiography of the entire peripheral vascular system using a time-resolved 4D MRA technique. Eur
Radiol. 2013 Jan;23(1):298-306
32 Saba L, Atzeni M, Rozen WM. Non-invasive vascular imaging in perforator flap surgery. Acta Radiol.
2013 Feb 1;54(1):89-98.
33 Schepers RH, Raghoebar GM, Lahoda LU. Full 3D digital planning of implant supported bridges in
secondarily mandibular reconstruction with prefabricated fibula free flaps. Head Neck Oncol. 2012
Sep 9;4(2):44.
34 Haddock NT, Monaco C, Weimer KA. Increasing bony contact and overlap with computer-designed
offset cuts in free fibula mandible reconstruction. J Craniofac Surg. 2012 Nov;23(6):1592-5.
35 Zheng GS, Su YX, Liao GQ. Mandible reconstruction assisted by preoperative virtual surgical
simulation. Oral Surg Oral Med Oral Pathol Oral Radiol. 2012 May;113(5):604-11.
123
36 Foley BD, Thayer WP, Honeybrook A. Mandibular reconstruction using computer-aided design and
computer-aided manufacturing: an analysis of surgical results. J Oral Maxillofac Surg. 2013
Feb;71(2):e111-9.
37 Rohner D, Guijarro-Martínez R, Bucher P. Importance of patient-specific intraoperative guides in
complex maxillofacial reconstruction. J Craniomaxillofac Surg. 2012 Dec 7.pii:S1010-5182(12)
38 Laible M, Schoenberg SO, Weckbach S. Whole-body MRI and MRA for evaluation of the prevalence
of atherosclerosis in a cohort of subjectively healthy individuals. Insights Imaging. 2012 3: 485-493.
124
Chapter
10
Summary and general discussion
(Dutch)
De rol van preoperatieve
radiodiagnostiek van donor- en
acceptorgebied bij vrije
weefseltransplantaties
125
126
Samenvatting en algemene discussie
Traumata, oncologische chirurgie en doorligplekken kunnen grote weke delen defecten veroorzaken,
waarvan de chirurgische sluiting een uitdaging vormt. Sinds de opkomst van de weefseltransposities
met een axiale vaatsteel in de jaren 60 van de vorige eeuw heeft de reconstructieve chirurgie een grote
ontwikkelingen doorgemaakt. In de laatste 20 jaar werd een revolutionaire stap gezet bij het ontwikkelen
van locale en vrije lappen, hun design en manier van oogsten. Daarbij werd met het oog op minder
donorplaats morbiditeit, bij meerdere lappen het principe van de perforatorlap gevolgd, waarbij de
perforator tot een groter, voedend hoofdvat wordt vrij geprepareerd. Per lap kan de precieze locatie van
de perforator echter sterk variëren. Daarom is het verstandig pre-operatief het vaatsysteem in kaart te
brengen, waardoor de dominante perforator en zijn verloop gevisualiseerd worden en sneller oogsten
van de lap mogelijk is. Het doel van dit proefschrift was het onderzoeken van de noodzaak van
preoperatief vaatonderzoek bij vrije weefsel transplantaties en verschillende technieken met elkaar te
vergelijken.
In hoofdstuk twee wordt de evolutie van de reconstructieve lappen chirurgie beschreven. De evolutie
was het gevolg van steeds verder voorscheidende kennis over de vasculaire anatomie van de huid en
het onderhuidse weefsel. Terwijl de eerste locale lappen voor reconstructies op een willekeurig en
toevallig doorbloedingspatroon berustten, was de volgende generatie lappen gekenmerkt door een
axiale vaatsteel. In deze dagen waren chirurgen als Esser en Machot de pionieren.(1,2) Door de
negatieve invloed van andere chirurgen als Gillies, werd het principe van de axiale vaatsteel niet meer
verder vervolgd tot hun herontdekking door Stuard Milton in 1969. De eerste lappen met een dergelijke
axiale vaatsteel waren gebaseerd op goed bekende vaten uit de anatomie boeken, zoals de a.radialis
voor de radialis onderarm lap en de thoracodorsale vaten voor de latissimus dorsi lap. Hoewel de
ontwikkeling en toepassing van dergelijke lappen toentertijd revolutionair was, wordt het oogsten van
deze lappen door de constante anatomie tegenwoordig als vrij makkelijk en onspectaculair beschouwd.
In de laatste jaren worden steeds meer lappen beschreven die alleen gesteeld zijn op één kleine
perforator, die tot zijn oorsprong uit een groter voedend vat in de diepte vervolgd wordt. Door het
gebruik van perforator lappen kan een goede doorbloeding naar de lap en minimale donorsite
morbiditeit gecombineerd worden.
In hoofdstuk drie wordt de “condito sine qua non” van een goede doorbloeding van een vrije lap en zijn
donorgebied geïllustreerd aan de hand van drie patiënten casus met een vrije fibula lap transplantatie.
De vrije fibula lap is de meest gebruikte vrije lap als het gaat om de reconstructies van bot- en bot-enhuid defecten. Donor-plaats morbiditeit wordt in het algemeen als zeldzaam en mineur beschreven.
Anatomische variaties in het vaatpatroon, eerdere chirurgie en atherosclerose kunnen de vitaliteit van
een vrije lap en het donorgebied negatief beïnvloeden. Het verlies van een vrije lap of het optreden van
ernstige complicaties in het donorgebied onderstrepen de noodzaak van een goed preoperatief
vaatonderzoek.
127
Om de techniek en de resultaten van verschillende studies goed met elkaar te kunnen vergelijken, is het
noodzakelijk dat meetmethodes gestandaardiseerd worden. Echter is dit niet altijd het geval, zoals uit
het overzichtsartikel van hoofdstuk vier blijkt, waarin de meetmethodes en berekening van de EnkelArm-Index (EEA) worden onderzocht. Sinds de introductie hiervan in 1950, werden diverse methodes
voor het meten en berekenen gebruikt. Dit resulteerde in variaties van de normaalwaarden en in
moeilijkheden de resultaten van studies met elkaar te kunnen vergelijken. Daarom was het doel van
deze studie de verschillende meet- en berekenmethodes van de EEA uit 100 publicaties in kaart te
brengen, en aan hand daarvan een gestandaardiseerde methode en normaalwaarden te introduceren.
In hoofdstuk vijf wordt een overzicht gegeven over de verschillende methodes om de vascularisatie van
lappen in kaart te brengen. De voor- en nadelen van hand-Doppler, kleuren-duplex, digitale subtractie
angiografie (DSA), computer tomographische angiografie (CTA) en magneet resonantie angiografie
(MRA) worden beschreven en bediscussieerd. Omdat CTA en MRA 3D-beelden van de vaten en het
omliggende weefsel produceren, lijken deze twee methodes het meest geschikt om vaten van het
donorgebied van perforator lappen zoals de anterieure dijbeenlap (ALT) en de deep inferior epigastric
peforator flap (DIEP) weer te geven. In lappen met een standaard anatomie of oppervlakkig gelegen
vaatvoorziening blijft de hand-Doppler het onderzoek van keuze.
In hoofdstuk zes werd de EEA vergeleken met de gouden standaard, de digitale subtractie angiografie,
als preoperatief vaatonderzoek bij de vrije fibula lap. Perifeer vaatlijden en congenitale anomalieën van
de grote onderbeenvaten kunnen het gebruik van de vrije fibula lap onmogelijk maken, waardoor het
belangrijk is deze afwijkingen preoperatief op te sporen. Omdat DSA ook zijn nadelen kent, is het
wenselijk een veilig, goedkoop, nauwkeurig en niet-invasief alternatief daarvoor te hebben. Wij hebben
de hypothese getest, dat een EEA meting van de drie onderbeenvaten gecombineerd met een handDoppler onderzoek van de peroneale huidperforatoren genoeg informatie geeft over de vaatstatus, om
angiografisch onderzoek te beperken tot de patiënten, waarbij één van deze twee of beide onderzoeken
afwijkend zijn. Daarvoor werden de meetgegevens van de EEA van de drie onderbeenvaten en van het
hand-Doppler onderzoek van de peroneale huidperforatoren van beide benen van negen prospectief
geïncludeerde patiënten en de niet-geopereerde benen van 13 retrospectief geïncludeerde patiënten op
vier verschillende manieren vergeleken met de preoperatieve bevindingen tijdens de angiografie. De
conclusie van de studie was dat het gecombineerd onderzoek van enkel arm index en hand-Doppler
onderzoek niet nauwkeurig genoeg is, om benen of arteriën met asymptomatisch vaatlijden of
vaatanomalieën op te sporen. Daarom kan deze onderzoek combinatie niet worden gebruikt, om het
besluit te nemen, of een vrije fibula lap kan worden uitgevoerd.
In hoofdstuk zeven werd 3D-TOF magneet resonantie angiografie (MRA) vergeleken met de gouden
standaard, de digitale subtractie angiografie (DSA) in het preoperatief vaatonderzoek van de vrije fibula
lap. Vijftien patiënten, die voor een vrije fibula lap transplantatie in aanmerking kwamen, werden
geïncludeerd en kregen van beide onderbenen een DSA en MRA onderzoek (n = 30). De DSA en MRA
onderzoeken werden door twee radiologen onafhankelijk van elkaar en geblindeerd beoordeeld. Beide
beoordelaars classificeerden de maat van stenose of hypoplasie van de diverse vaatsegmenten op een
5-punten schaal van 0 (occlusief) tot 4 (geen stenose). Tevens werd het aantal cutane perforatoren
128
beoordeeld en werd aan de radiologen gevraagd of zij – gebaseerd op de onderzoeken – een fibula lap
transplantatie zouden afraden. Tussen de twee beeldvormende onderzoeken werd een opmerkelijke
overeenkomst geconstateerd in de scores over de maat van stenose. De sensitiviteit van MRA, om
stenoses te detecteren, was vergeleken met DSA 0.79, en de specificiteit 0.98. In 53 van de 60
beoordelingen, was het oordeel, of een fibula lap transplantatie kon worden uitgevoerd, gelijk tussen
DSA en MRA. Het gemiddeld aantal cutane perforatoren per been was gelijk tussen DSA en MRA (p =
0.142). De resultaten van dit onderzoek suggereren, dat MRA een goed alternatief is voor DSA in het
preoperatief vaatonderzoek voor de vrije fibula lap transplantatie.
Een geslaagde weefseltransplantatie is niet alleen afhankelijk van de goede doorbloeding van de lap,
maar ook van de vaatstatus van het acceptorgebied. Zoals in hoofdstuk acht gepresenteerd wordt,
kunnen anatomische variaties, arteriosclerose en bestralingsschade van de acceptorvaten een
microchirurgische ingreep erg moeilijk maken. Met als voorbeeld de arteria mammaria interna werd
tijdens
borstreconstructies
onderzocht,
of
het
preoperatief
mogelijk
is
arteriosclerose
of
bestralingsschade van de vaten angiografisch op te sporen. Daarvoor werden de bevindingen van het
preoperatieve angiografie onderzoek van de mammaria interna vaten vergeleken met de mate van
vaatschade zoals gevonden tijdens de operatie, het postoperatief klinisch beloop en histologisch
onderzoek van het acceptorvat. In totaal werden 34 patiënten geïncludeerd die na radiotherapie een
borstreconstructie zouden krijgen. In totaal werden bij hun 40 vrije lappen voor een borstreconstructie
getransplanteerd. Eenentwintig mammaria interna arteriën lagen binnen en 19 arteriën buiten het
bestralingsveld. Alleen in twee van de zes patiënten met een afwijkende angiografie lag de arteria
mammaria interna in het bestralingsveld. Op basis van de studieresultaten kan geconcludeerd worden,
dat schade aan de mammaria interna vaten niet altijd door een preoperatieve angiografie of
intraoperatief onderzoek kan worden opgemerkt.
Algemene discussie
Zoals reeds genoemd, is het bij lappen met een variabele vaatanatomie of suboptimale vaatstatus ,
bijvoorbeeld ten gevolge van atherosclerose, belangrijk vóór het oogsten van een lap over de kwaliteit
en het verloop van de vaten geïnformeerd te zijn. Om ernstige complicaties na weefseltransplantatie te
voorkomen is het uitermate belangrijk, dat niet alleen de vaatsteel van de lap een goede kwaliteit heeft,
maar ook de vaten van het acceptorgebied en de achterblijvende vascularisatie van het donorgebied.
Dit proefschrift beoogt de verschillende preoperatieve vaatonderzoeken bij de diverse typen
reconstructies te onderzoeken en de vraag te beantwoorden, welk onderzoek het meest geschikt is bij
welk soort reconstructie.
Alle vaatonderzoeken hebben hun eigen voor- en nadelen.(4-6)
De enkel arm index (EEI), is een bloeddruk index, waarbij de druk van de linker arm als noemer en het
gemiddelde van de drukken van beide tibiaal arterien per been als teller moeten worden gebruikt. De
voordelen zijn het niet invasieve karakter, de lage kosten, de compactheid van de apparatuur, waardoor
het makkelijk draagbaar is, en het gemak waarmee het onderzoek uitgevoerd kan worden. Maar de
nadelen zijn de ongevoeligheid voor de detectie van geringe tot matige stenoses en het onvermogen de
vaatstatus van een enkel vat aan te geven.
129
De voordelen van een handdoppler (HHD) zijn vergelijkbaar met die van de EEI: niet invasief,
goedkoop, klein en handzaam en makkelijk te bedienen. Het grootste nadeel is, dat de meest gangbare
Doppler kop (8 MHz), een detectievermogen tot een diepte van 20mm heeft. Tevens is het niet mogelijk
te bepalen welk vat het signaal produceert, dat met de HHD gedetecteerd wordt. Bovendien verkrijgt
men met deze techniek geen 3D-beeld van het vaatstelsel en het omringende weefsel, dat opgeslagen
zou kunnen worden.
Net als HHD is ook de kleurenduplex echografie (CDS) niet invasief. Vergeleken met de HHD heeft
CDS het grote voordeel, dat daarmee wel informatie over de vaatanatomie kan worden verkregen,
evenals kwantitative analyse naar dominante perforator. Het grote nadeel is echter, dat het onderzoek
alleen uitgevoerd kan worden door personeel, dat in deze specifieke techniek en de anatomie van vrije
lappen geschoold is. Doordat het om een levensecht, dynamisch onderzoek gaat is, zijn de
onderzoeksresultaten van CDS minder goed reproduceerbaar. Vergeleken met CTA, MRA en DSA is
CDS - net als HHD – niet in staat een 2D of 3D beeld van de complete vasculaire anatomie te maken,
dat door de chirurg tijdens de planning en het oogsten van de lap kan worden gebruikt.
De beschreven voordelen voor Digitale Subtractie Angiografie bestaan uit het feit, dat een 2D beeld van
de intraluminale vaat anatomie en informatie over atherosclerotische veranderingen kan worden
verkregen. Nadelen van DSA zijn, dat röntgenstralen worden gebruikt, en dat het een vrij langdurige en
invasieve techniek is, waarbij een jodium-houdend contrastmiddel intravasculair moet worden
toegediend, dat allergische reacties en vaat- en nierschaden kan veroorzaken. Door de vasoconstrictie,
die door het contrastmiddel wordt veroorzaakt, wordt de exacte meting van een vatdoorsnede en de
beoordeling van vaten met een klein kaliber onbetrouwbaar. Tevens moet de patiënt na een angiografie
enkele uren in het ziekenhuis blijven liggen om de plaats, waar het vat voor het onderzoek aangeprikt
werd, te laten genezen. Daarom is vaak een klinische opname noodzakelijk, waardoor het onderzoek in
zijn geheel relatief duur is. Tenslotte bestaat het risico op een vals aneurysma, waar het vat aangeprikt
is.
Het voordeel dat een Computer Tomografische Angiografie (CTA) biedt, is de mogelijkheid een 3D
plaatje te generen, waarop heel nauwkeurig de intraluminale doorsnede en het verloop van een vat in
relatie tot het omliggende weefsel kan worden bepaald. Daardoor is het voor de chirurg mogelijk
preoperatief een operatieplan op te stellen en een bepaalde perforator uit te kiezen, waardoor de
dissectie tijdens de operatie veiliger en sneller kan gebeuren. Een groot nadeel van CTA is, dat een vrij
hoge dosis röntgenstralen nodig is (ca. 5.6mSv) en dat een jodium-houdend contrastmiddel moet
worden gebruikt, met de reeds boven beschreven nadelen. Het vasospastische effect van het
contrastmiddel is een ernstig nadeel, waardoor de nauwkeurige weergave van vaten met een klein
kaliber moeilijk is.
Het grote voordeel van Magneet Resonantie Angiografie (MRA) is, dat het met magnetische velden
werkt in plaats van röntgen straling. Afhankelijk van de gebruikte software, kunnen vaten ook zonder
gebruik van contrastmiddel worden weergegeven, waardoor het een vrij veilig onderzoek voor de
patiënt is. MRA produceert 3D plaatjes, waardoor de chirurg nauwkeurig het verloop en de doorsnede
van de vaten en de relatie tot het omliggende weefsel kan bestuderen. De nadelen van MRA zijn de
hoge kosten van het onderzoek en het feit, dat het niet gebruikt kan worden bij patiënten met
claustrofobie of metalen implantaten, omdat deze strooi-artefacten veroorzaken en de kwaliteit negatief
beïnvloeden.
130
Naast de beschrijvende studies van dit promotieonderzoek (hoofdstuk twee, drie, vier en vijf)
onderzochten wij EEI, HHD, DSA en MRA in vergelijkende studies (hoofdstuk zes, zeven en acht). Wij
concludeerden, dat de combinatie van EEI en HHD niet nauwkeurig genoeg is voor de detectie van
arteriën met subklinisch vaatlijden of vaatanomalieën en dus onvoldoende informatie geeft, om een
chirurgisch plan voor een vrije lap van het onderbeen te kunnen maken. Bovendien trokken wij, uit het
voorbeeld van de vrije fibula lap, de conclusie, dat MRA een goed alternatief is voor DSA voor de
preoperatieve planning van vrije weefsel transplantaties van het onderbeen. Tevens liet dit
promotieonderzoek aan de hand van de mammaria interna vaten zien, dat DSA niet gevoelig en
specifiek genoeg is, om vaatschaden ten gevolge van bestraling zichtbaar te maken.
In de meest recente literatuur is er een duidelijke trend naar MRA in de vaatdiagnostiek en lijkt het DSA
en CTA af te lossen.(7-9) Overeenkomend met onze resultaten is MRA bij de planning van vrije fibula
lappen sensitief genoeg voor de detectie van hypoplastische vaten, stenoses, vaatafsluitingen en
atheorsclerotische veranderingen van de vaten . Het is mogelijk zowel nauwkeurig de kwaliteit van het
hoofdvat als ook de septo- en musculocutane perforatoren weer te geven.(7,10-15) CTA wordt ook als
een goed beeldvormend onderzoek beschreven voor de weergave van de arteriën van de het
onderbeen.(16-18) Bij de preoperatief vaatonderzoek van perforator lappen is het gebruik van zowel
MRA(8,9) als ook CTA(19-24) beschreven. Het voordeel van CTA is, dat het mogelijk is vaten met een
kleinere diameter (<0.5mm) goed weer te geven.(25) Zoals reeds eerder genoemd wordt bij MRA in
tegenstelling tot CTA geen röntgenstraling gebruikt en is het niet nodig jodium-houdend contrastmiddel
te gebruiken, waardoor het minder schadelijk voor de patient is.(8-26) Het moet echter wel genoemd
worden, dat er geen vergelijkende studies tussen MRA en CTA voor de planning van vrije fibula lappen
bestaan.
Als we de uitkomsten uit onze studies en de literatuur met elkaar combineren, kunnen wij concluderen,
dat MRA en CTA op het moment de beste beschikbare methodes zijn, om de vaten van het onderbeen
en perforatorlappen in het algemeen in kaart te brengen.(4) Bij de planning van lappen met een vrij
oppervlakkig gelegen vaatsteel, in de nabijheid van het te bedekken defect, bij lappen met een
makkelijke standaard anatomie en voor het intraoperatieve gebruik, blijft HHD de onderzoeksmethode
van keuze.(4) DSA verliest steeds meer zijn plaats als gouden standaard en CDS kan als alternatief
onderzoek worden gebruikt, als de andere methodes als MRA en CTA niet beschikbaar zijn of er bij een
patiënt contra-indicaties voor bestaan.
Als men de resultaten van een onderzoek weegt, vooral als het om een vergelijkende studie gaat, is het
belangrijk de gebruikte methode kritisch te bekijken. Het sterke punt van ons onderzoek is het, dat zij
als prospectieve en vergelijkende studies zijn opgezet. De onderzochte technieken werden met de
toenmalige gouden standaard vergeleken. Er zijn echter ook beperkingen aan onze studie, zoals de
relevantie van de resultaten voor de hedendaagse kliniek. Het gebied van het beeldvormend onderzoek
ontwikkelt zich zo snel, dat tijdens het opzetten, het uitvoeren, het analyseren en het beschrijven van
ons onderzoek sommige delen van de studie nu reeds als verouderd kunnen worden beschouwd. Toen
wij de studie opzetten was 1.5 Tesla 3D TOF-MRA de gebruikelijke methode om onderbeenvaten weer
te geven. Maar sindsdien heeft zich de MRA techniek zeer snel verder ontwikkeld.(27-30)
131
De sterkte van het magneetveld van MRI’s in de kliniek loopt op van 1.5 naar 3.0 Tesla. Voor
onderzoeksdoeleinden worden zelfs 7.0 Tesla scanners gebruikt. Door het gebruik van een scanner
met een hoger Tesla is het mogelijk de verhouding tussen gewenste en ongewenste signalen te
verbeteren en ruimtelijke en tijdelijke resolutie te vergroten. Gezien wij een 1.5 Tesla scanner hebben
gebruikt is het duidelijk, dat de resolutie van de afgebeelde vaten beter zou zijn geweest als we een 3.0
Tesla scanner hadden gebruikt. Maar aangezien de meeste klinieken nog steeds een 1.5 Tesla scanner
gebruiken, wijkt de door ons gebruikte scanner en onze resultaten niet van hun af.
Naast de verbetering van de scanners werden ook de scan technieken ontwikkeld, om vaten nog
nauwkeuriger af te kunnen beelden, dan met de TOF-MRA techniek die wij gebruikten. Met MRA
kunnen de vaten op twee verschillende manieren worden weegegeven met de stromingsafhankelijke
angiografie (FDA) en de stomingsonafhankelijke angiografie (FIA).
De FDA MRA techniek kan worden onderverdeeld in “time-of-flight” (TOF) en fase-contrast (PC). Bij de
TOF-MRA geeft stromend bloed een hoger signaal dan weefsel, maar gebieden met een trage stroming
worden niet goed weer gegeven. Met PC-MRA kan ook langzame stroming beter worden gedetecteerd
en kan zelfs de stromingssnelheid van het bloed worden bepaald. Het nadeel van PC-MRA is echter,
dat de stroming maar in één richting tegelijk kan worden gedetecteerd. Om een complete weergave van
de stroming te genereren, moeten er in alle drie richtingen losse scanseries worden vervaardigd en
samen worden gevoegd. Ondanks de traagheid van deze methode, is het sterke punt dat zowel het
stromende bloed kan worden weergegeven maar ook tegelijk een kwantitatieve analyse van de
stroomsnelheid kan worden gedaan.
In het algemeen, is het een uitdaging traag stromend bloed goed weer te geven, omdat het verschil
tussen het signaal van het bloed en het statische weefsel erg klein is. Om het signaal van het bloed te
vergroten, wat uitermate belangrijk is voor de weergave van zeer kleine vaten en een trage
stromingssnelheid, kan een contrastmiddel worden gebruikt. Daarom werd contrast-versterkte (CE) FIAMRA techniek ontwikkeld. De techniek met gebruik van contrastmiddel wordt als meeste gebruikt.
Daarbij wordt een contrastmiddel in een vene ingespoten en worden de plaatjes tijdens de eerste
passage van het contrastmiddel door de arteriën opgenomen. Als de timing van de scan correcti is, zijn
de beelden gebruikelijk van een erg hoge kwaliteit. Maar als de timing slecht is of contrastmiddel wordt
gebruikt, dat langer in de bloedvaten blijft hangen, wordt het weergave van de arteriën verstoord door
een veneuze overprojectie.
Omdat het gebruik van contrastmiddel gevaarlijk kan zijn voor patiënten met nierfalen, werden de FIAtechnieken ontwikkeld waarbij geen contrast nodig is. Deze methodes zijn gebaseerd op het verschil
tussen T1, T2 en chemical shift van de verschillende weefseltypes en de voxel.
De vervaardiging van de plaatjes is tevens aan het veranderen van de 3D-techniek, die wij hebben
gebruikt naar de 4D.(31) Het verkrijgen van drie dimensionale data is nuttig bij een complexe
vaatanatomie waarbij het bloed in alle ruimtelijke richtingen vloeit. Het grote voordeel van de nieuwe 4D
techniek is, dat de arteriële en veneuze fase van elkaar kunnen worden gescheiden. Om de veneuze
overprojectie te voorkomen en om dynamische stromingsinformatie te verkrijgen, werd recent de TripleTWIST MRA (Time-resolved angiography With Interleaved Stochastic Trajectories) ontwikkeld, wat de
nieuwe standaard in beeld-vormend onderzoek bij patiënten met perifeer vaatlijden lijkt te worden.(31)
Bij deze techniek worden tijdens contrast injectie van hetzelfde gebied meerdere snelle series gemaakt.
Alle afbeeldingen met de maximale intensiteit van iedere serie worden geselecteerd en gecombineerd,
132
om samen
een dynamisch MRA serie te vormen met een hoge resolutie en zonder veneuze
overprojectie.
Tevens is er inmiddels software op de markt, die het mogelijk maakt, niet alleen via het ziekenhuis
netwerk de plaatjes van beeldvormend onderzoek te bestuderen, maar ook voor geautoriseerde externe
gebruikers
(b.v.
Siemens
IHE
Integrating
Healthecare
Enterprise
–
XDS-1b).
Door
deze
toegankelijkheid van de onderzoeken is het mogelijk externe experts van over de hele wereld in consult
te roepen en de preoperatieve chirurgische planning vanuit buiten het ziekenhuis te verrichten.
Met deze verbeteringen in techniek verwachten wij, dat het preoperatief vaatonderzoek van de vrije
lappen chirurgie en in het bijzonder de rol van MRA in toekomst een steeds belangrijkere rol gaat
spelen.(32) Derhalve denken wij, dat een radioloog, die begrip heeft van vrije weefsel transplantaties,
een steeds prominentere rol in een reconstructief team zal spelen en een nieuwe dynamiek in een team
zal brengen. Afhankelijk van de locatie van een bestaand of te verwachten defect, moet het team
bestaan uit een oncoloog, een oncologisch chirurg (algemeen chirurg, HNO-arts, kaakchirurg,
orthopedisch chirurg, gynaecoloog of neurochirurg) of traumachirurg, een radioloog, een reconstructief
plastisch chirurg en een anesthesist.
Een goed voorbeeld voor de ontwikkeling en rol van een preoperatieve planning is de Rohner
procedure voor de reconstructies van onder- en bovenkaken. Daarbij wordt het te transplanteren
botsegment als vrije lap van het kuitbeen of de bekkenkam preoperatief geplant, om precies in het
defect te passen.(33-37) Tijdens de work-up wordt een CTA vervaardigd, waaruit niet alleen de
informatie over de vascularisatie kan worden verkregen, maar ook data voor d 3D reconstructie van het
botsegment. Daardoor is het mogelijk preoperatief zaagmallen te maken, waardoor wederom de
controle tijdens de ingreep vergroot, de operatietijd verminderd en het botcontact van de nieuw te
vormen kaak geoptimaliseerd wordt. Dankzij deze techniek is het ook mogelijk al vóór de vrije weefsel
transplantatie van het kuitbeen de tandimplantaten voor de kronen te plaatsen, waardoor een direct
belastbare kaakreconstructie gemaakt kan worden.(33) Door het gebruik van de CAD-CAM (computeraided design and manufacturing) technologie en stereolithografisch modellen kan de nauwkeurigheid
van het chirurgisch resultaat en de intraoperatieve efficiëntie worden verhoogd.(34,36-38) Het lijkt erop,
dat deze CAD-CAM techniek tot betere uitkomsten voor de patiënt en tot een risicovermindering op
complicaties en non-union leidt. Echter is dit nog niet in vergelijkende studies met een langere follow-up
en grotere patiënten groepen aangetoond.
133
Voorstel voor vervolgonderzoek
Gebaseerd op de uitkomsten van onze onderzoeken, stellen wij voor in toekomst verder onderzoek op
het gebied van de vrije lappen chirurgie te verrichten.
1) Wij hebben aangetoond, dat de enkel-arm index (EEI) in de pre-operatieve work-up
onvoldoende informatie geeft voor de vrije fibula lap, maar wij denken, met de bovengenoemde
argumenten, dat MRA een goed alternatief is voor DSA. Alhoewel het beter geweest zou zijn
als de sensitiviteit van MRA in onze studie hoger was, denken wij dat door de recente
ontwikkelingen van de MRA techniek, MRA het onderzoek van keuze zal worden. Ook al wordt
DSA nog steeds in de recente literatuur als gouden standaard voor het onderzoek van perifeer
vaaltijden geadviseerd,(28) wij zijn van mening, dat deze techniek meer en meer als obsoleet
beschouwd moet worden.
Om te bewijzen, dat CTA en MRA net zo nauwkeurig of zelfs nauwkeuriger zijn dan DSA voor
het opsporen van perifeer vaatlijden, zou een vergelijkende studie opgezet moeten worden,
waarbij de nieuwste scanners en scanprotocollen worden gebruikt. Om het oplossend
vermogen en de nauwkeurigheid van deze beeldvormende onderzoeken in de detectie van
vaatstenoses te onderzoeken, is het wenselijk CTA en MRA met de resultaten van een
anatomische dissectie te vergelijken. Dit zou alleen in een dierproef kunnen worden uitgevoerd.
2) Wij hebben aangetoond dat DSA onbetrouwbaar is in het preoperatief vaatonderzoek van de
acceptorsite. Echter zijn wij van mening, dat het in bepaalde gevallen nuttig is goed
geïnformeerd te zijn, over de vaatstatus van de acceptorsite, zoals bij patiënten met eerdere
chirurgie of bestraling in dit gebied. Afhankelijk van de locatie van de acceptorvaten, zou in
deze gevallen CTA, MRA of kleuren duplex, van aanvullende waarde kunnen zijn. Toekomstig
onderzoek kan zich richten op de indicatie en noodzaak voor dit preopartieve vaatonderzoek en
tevens aantonen welk beeldvormend onderzoek daarvoor het meest geschikt is.
3) Zoals eerder genoemd, is virtuele chirurgie, waarbij de belangrijke chirurgische stappen virtueel
worden doorlopen, een nieuwe ontwikkeling. Daardoor kan de chirurgie veiliger, sneller en met
betere resultaten worden verricht. Tot op heden werd voor deze planning vooral CTA gebruikt.
Wij denken, dat in dit kader het gebruik van MRA een toevoeging zou leveren voor de
patiëntveiligheid. Maar door strooisignalen van metaal implantaten zal MRI niet altijd het
onderzoek van keuze zijn.
134
Chapter
11
Explanation of used abbreviations
135
136
Explanation of used abbreviations
AAI
Ankle-Arm Index
ALT
Anterolateral Thigh flap
AT
Anterior Tibial artery
CAD-CAM
Computer-Aided Design and Computer-Aided Manufacturing
CDS
Color Duplex Sonography
CE
Contrast Enhanced
CTA
Computed Tomographic Angiography
DIEP
Deep Inferior Epigastric Perforator flap
DSA
Digital Subtraction Angiography
FDA
Flow-Dependent Angiography
FIA
Flow-Independent Angiography
HHD
Hand Held Doppler
IEA
Inferior Epigastric Artery
IGAP
Inferior Gluteal Artery Perforator flap
LAP
Lumbal Artery Perforator flap
IMA
Internal Mammary Artery
IMAP
Internal Mammary Artery Perforator flap
IMV
Internal Mammary Vein
MRA
Magnetic Resonance Angiography
MRI
Magnetic Resonance Imaging
PAOD
Peripheral Arterial Occlusive Disease
PC
Phase Contrast
PR
Peroneal artery
PT
Posterior Tibial artery
SGAP
Superior Gluteal Artery Perforator flap
TDAP
Thoracodorsal Artery Perforator flap
TOF
Time Of Flight
TPT
Tibioperoneal Trunk
TRAM
Transervese Rectus Abdominis Myocutaneous flap
TWIST
Time-resolved angiography With Interleaved Stochastic Trajectories
137
138
Chapter
12
Acknowledgements (Dutch)
139
140
Dankwoord
Hooggeleerde heer, Prof. P. Werker, beste Paul, ik wil jou van harte danken voor de goede opleiding tot
plastisch chirurg, die ik bij jou mocht genieten. Jij hebt mij als opleider gestuurd waar nodig, maar ook
ruimte geboden mijn eigen weg te banen door het doolhof van de opleiding. Doordat ik eerder dan
verwacht met mijn opleiding tot plastisch chirurg kon beginnen, kwam mijn onderzoeksplanning in het
gedrang, maar jij spoorde mij aan om het onderzoek weer op te pakken en het af te maken. Dit heeft
uiteindelijk geresulteerd in voorliggend proefschrift. Daarvoor ben ik jou zeer dankbaar!
Weledelzeergeleerde heer, Dr. J. Hage, beste Joris. Door jou werd het onderzoek dat leidde tot dit
proefschrift geïnitieerd. Jij liet mij zien wat het betekent onderzoek te doen, en dat overal onderwerpen
te vinden zijn, die het waard zijn te onderzoeken. Jouw onderzoekergeest is fascinerend en inspirerend,
jou drijfveer voor de wetenschap bijna onuitputtelijk. Toen mijn opleiding tot plastisch chirurg in
Groningen begon scheidden onze wetenschappelijke wegen. Ik dank jou voor jouw goede enthousiaste
begeleiding tijdens mijn eerste stappen en het vertrouwen dat je in mij had.
Beste Jeroen Smit. Het is niet alleen het onderzoek, dat wij samen voor jouw en mijn promotie hebben
verricht, dat ons verbindt, maar ook een hechte vriendschap. Door jouw ‘drive’ is het onderzoek in de
afrondende fase in een stroomversnelling gekomen. Jouw kritische kijk op de onderzoeksdata en jouw
adviezen over de presentatie ervan hielpen mij kritisch te blijven nadenken over mijn wetenschappelijk
werk. Jij bent een zeer begaafde onderzoeker en zult een net zo goede plastisch chirurg worden. Ik
hoop, dat wij nog vele gezellige uren samen met onze vrouwen zullen doorbrengen. Dank voor jouw
waardevolle ondersteuning, co-promotorschap en vriendschap.
Beste Martin Stenekes, het was jouw stem in de sollicitatiecommissie, die mij naar Groningen heeft
gebracht en het is onze vriendschap, die ook nu mijn band met Groningen versterkt. Ik ben onder de
indruk van jouw gevoel en liefde voor wetenschappelijk onderzoek. De gezellige avonden samen met
onze vrouwen en ook de vrolijke verjaardagen van onze kinderen, waren iets om naar uit te kijken. Ik
mis het ritme van deze plezierige activiteiten nu wij op grotere afstand van elkaar wonen. Dank voor je
inspiratie en vriendschap.
141
Lieve Saskia, vanaf mijn geboorte tot nu heb je altijd een minnend en beschermend oog op mij gehad,
en hebben wij met veel plezier vele leuke dingen met elkaar beleefd. Ook hebben wij in moeilijke
momenten veel steun aan elkaar gehad. Als jouw broertje heb ik onder andere altijd bewondering
gehad voor jouw discipline en vastberadenheid. Door afronding van jouw school, geneeskunde studie,
promotieonderzoek, specialisatie en subspecialisatie heb jij jouw huidige positie als internisthaematoloog bereikt. Jij liet mij zien, dat je door vlijt bijna alles kan bereiken.
Jou hulp in de afrondende fase van dit proefschrift heeft de finishing touch gegeven. Ik vind het leuk, jou
als zus te hebben en hoop, dat we nog vele leuke dingen met elkaar zullen beleven. Ik houd van je.
Geachte leden van de leescommissie, professor Van Der Hulst, professor Ulrich en professor
Zeebregts, ik dank u zeer, dat u in de commissie wilde zitten, om mijn promotieonderzoek op zijn
wetenschappelijke merites te beoordelen.
Geachte leden van de oppositie, ik dank U zeer voor uw moeite en tijd, die u investeert, om zitting te
nemen in de oppositie om met mij van gedachte te wisselen over dit proefschrift en mij aan de tand te
voelen over de inhoud ervan. Ik zie uit naar een inspirerende dialoog met u.
Beste Krijn van Lienden. In de tijd, dat ik het onderzoek in het AMC deed, was jij mijn kundig
aanspreekpunt. Je hebt mij van alles uitgelegd en veel tijd in mij geïnvesteerd. Zoals je ziet, niet zonder
resultaat. Ook buiten het ziekenhuis hebben wij een gezamenlijke passie – het motorrijden. De sound
van mijn motor laat mij altijd aan onze gemeenschappelijke uren denken. Ik hoop, dat wij binnenkort
weer eens kunnen gaan touren.
Beste Marcel in het Veer. De grootste horde van dit promotieonderzoek was het bewerken van mijn
data en de bijbehorende statistiek. Doordat ik jou via Jeroen leerde kennen is het gelukt ook deze
horde, met jouw uitleg, kunde en grote hulp te nemen. Jij hebt mij zeer geholpen dit promotieonderzoek
af te ronden. Hartelijk dank.
Tevens wil ik hier mijn dank uitspreken aan allen, die met mij samen aan de artikelen hebben gewerkt
en hun steentje hebben bijgedragen aan de totstandkoming van mijn proefschrift.
142
De maatschap plastische chirurgie Zuidoost-Brabant wil ik danken voor mijn buitengewoon interessante
en uitdagende baan in Eindhoven in de functie van chef de clinique.
Lieve mama en papa. Jullie hebben mij met liefde opgevoed, waar nodig in de juiste richting gestuurd,
in mij geloofd en mij ondersteund. Door de kans die jullie mij hebben gegeven, om in Amsterdam te
studeren, ben ik geworden wie ik nu ben. Het is leuk, te zien, dat jullie met interesse naar mijn vak
kijken en soms verbaast zijn over de mogelijkheden, die de plastische chirurgie voor bepaalde
problemen biedt. Ik dank jullie van harte voor deze inzet en de kansen die jullie mij hebben gegeven.
Zonder jullie was het niet gelukt. Bedankt, onder andere voor de prachtige cover van dit proefschrift. Ik
houd van jullie.
Liefste Carmen, door jou ben ik wat en wie ik ben. Jij hebt mij met goede adviezen, optimisme, en liefde
bijgestaan in de moeilijke tijden van de specialisatie en de afronding van mijn promotieonderzoek. Je
hebt mij waar mogelijk de rug vrijgehouden, om weer tijd in het onderzoek te kunnen steken. Tevens
heb jij mij als bewijs van onze liefde twee prachtige kinderen gegeven. Naast jouw eigen drukke en
verantwoordelijke baan ben je voor hen een fantastische moeder. Voor al deze dingen dank ik je van
harte. Ik ben trots op je en ik houd van jou!
Liefste Florian en Eveline, jullie zijn de kinderen, waar ik altijd van heb gedroomd. Door een drukke
baan en dit promotieonderzoek was het helaas niet altijd mogelijk nog meer tijd met jullie door te
brengen. Na mijn werk en in het weekend geven jullie mij door jullie glimlach en jullie vrolijke
aanwezigheid altijd een gelukkig en trots gevoel. Jullie zijn echte schatjes! Ik zal na de verdediging van
mijn proefschrift alle tijd maken om een goede vader te zijn. Ik houd van jullie!
143
144
Chapter
13
Curriculum vitae
145
146
Curriculum vitae
Steven Klein was born in Neuss, Germany, on the 20th of Mei 1977. He started his education at the
convent school Norbert Gymnasium in Knechtsteden (Nordrhein-Westfalen) in 1987 and completed his
secondary school at the Albert Einstein Gymnasium in Kehl (Baden-Württemberg) in 1996. After his
military service at the medical corps he started his medical education at the University of Amsterdam in
1997.
Already in secondary school he became interested in plastic and reconstructive surgery. Knowing that it
would be difficult to reach his dream he applied for a researchproject in his second year of his studies at
the department of prof. dr. C.M.A.M. van der Horst in the AMC. Later under the supervision of dr. J.J.
Hage, department of reconstructive surgery of the Dutch Cancer Institute in Amsterdam, he became
more involved in research.. This is where the idea for his PhD-thesis started to rise and where the first
cornerstones were layed. It further developed under the guidance of prof. dr. P.M.N. Werker.
After completing his medical degree in 2004 he worked as a senior house officer at the department of
plastic, reconstructive and hand surgery at the Academic Medical Centre of Amsterdam. Thereafter he
continued his education as a senior house officer at the department of surgery at the Dutch Cancer
Institute in Amsterdam till he was enrolled as a specialist registrar for the specialization of plastic
reconstructive surgery at the University Medical Centre of Groningen.
He followed the first two years of training in general surgery at the Medical Centre of Leeuwarden, and
completed the four-year training in plastic surgery at the University Medical Centre of Groningen under
the guidance of prof. dr. P.M.N. Werker. A part of the training was done at the department of plastic
surgery at the Medical Centre of Leeuwarden and at the Burns Unit of the Martini Hospital in Groningen.
A special training in postbariatric surgery was done in the Dreifaltigheits Krankenhaus Wesseling,
Germany, under the guidance of dr. D.F. Richter. On the 7th of October 2011 he finished his training
and was registered as a plastic surgeon.
He started his work experience as a plastic surgeon in the Martini Hospital in Groningen and now works
at the Catharina Hospital in Eindhoven.
147
148
Chapter
14
List of publications
149
150
List of publications
1)
Klein S, van der Horst CMAM
Verdwenen prothesen – borstimplantaten worden onvoldoende geregistreerd
Medisch Contact 2003; 58: 31-32
2)
Klein S, Hage JJ, van der Horst CMAM, Lagerweij M
Ankle-Arm Index versus Angiography for the Preassessment of the Fibula Free Flap
Plast Reconstr Surg 2003; 111: 735-743
3)
Klein S, Hage JJ, Woerdeman LAE
Donor site necrosis following fibula free flap transplantation
Microsurgery 2005; 25: 538-542
4)
Klein S, Hage JJ, de Weerd L
Capita Selecta: Perforatorlappen - de evolutie van een reconstructief-chirurgische techniek
Ned Tijdschr Geneeskd 2005; 149: 2392-2398
5)
Klein S, Hage JJ
General Review: Measurement, Calculation, and Normal Range of the Ankle-Arm Index:
A Bibliometric Analysis and Recommendation for Standardization.
Ann Vasc Surg 2006; 20: 282-292
6)
Weum S, de Weerd L, Klein S, Hage JJ
Soft tissue defects treated with perforator flaps.
Tidsskr Nor Laegeforen. 2008; 128: 313-315
7)
Werker PMN, Klein S, LU Lahoda, M Rüttermann. Dissection course of flaps in arm and
leg. Groningen: Wenckebach Instituut. 2009; ISBN 978-90-75823-71-4
8)
Smit JM, Klein S, Werker PM
An overview of methods for vascular mapping in the planning of free flaps.
J Plast Reconstr Aesthet Surg 2010; 63: e674-682
9)
Smit JM, Klein S, de Jong EH, de Bock GH, Werker PMN
Value of the implantable Doppler system in free flap monitoring.
J Plast Reconstr Aesthet Surg 2012; 65(9): 1276-1277
10)
Klein S, van Lienden KP, van ’t Veer M, Smit JM, Werker PMN
Evaluation of the lower limb vasculature before free fibula flap transfer. A prospective blinded
comparison between magnetic resonance angiography and digital subtraction angiography.
Accepted for publication in Microsurgery
11)
Klein S, Hoving S, Werker PM, Russel NS
Is there an indication for digital subtraction angiography in the assessment of irradiationinduced vascular damage prior to free flap surgery by the means of the internal mammary
vessels? Running title: Angiography of irradiated internal mammary arteries.
Accepted for publication in Journal of Reconsructive Microsurgery
151