Download AXB / High resolution MRI

Acta Anaesthesiol Scand 2012; ••: ••–••
Printed in Singapore. All rights reserved
© 2012 The Authors
Acta Anaesthesiologica Scandinavica
© 2012 The Acta Anaesthesiologica Scandinavica Foundation
ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/j.1399-6576.2012.02703.x
High-resolution MRI demonstrates detailed anatomy of
the axillary brachial plexus. A pilot study
T. Kjelstrup1, F. Courivaud2, Ø. Klaastad3, H. Breivik3 and P. K. Hol2
1
Department of Anaesthesiology, Diakonhjemmet Hospital, Oslo, Norway, 2The Intervention Centre, Oslo University Hospital, Rikshospitalet,
Norway and 3Department of Anaesthesiology, Oslo University Hospital, Rikshospitalet, Norway
Background: Axillary block is the most commonly performed
brachial plexus block and may be guided by nerve stimulation or
ultrasound. Magnetic resonance imaging (MRI) has proven to be
beneficial in presenting anatomy of interest for regional anaesthesia and in demonstrating spread of local anaesthetic. The aim
of this pilot study was to demonstrate the anatomy as shown by
MRI of the brachial plexus in the axillary region.
Methods: Nine volunteers and nine patients were examined in
a 3.0 Tesla MR. The patients had two different brachial plexus
blocks. Subsequently, they were scanned by MRI and finally
tested clinically for block efficacy before operation. Axial images,
with and without local anaesthetics injected, were viewed in a
sequence loop to identify the anatomy.
Results: With the high-resolution MRI, we obtained images of
good quality, and cords and all terminal nerves could be identified. When local anaesthetics are injected, neurovascular struc-
M
agnetic resonance imaging (MRI) has been
useful for assessing the precision and risks of
landmark-based brachial plexus blocks and for
studying the spread of local anaesthetics (LAs).1–3
Using a 0.5 Tesla open MRI system for an axillary
block study a decade ago, our group studied the
spread of LA as visualized by MRI. However, we
were not able to distinguish the terminal nerves of
the plexus from equally sized vessels. Therefore, we
could not directly determine if the individual main
terminal nerves [nervi axillaris, musculocutaneous
radialis, medianus and ulnaris] were reached or surrounded by LA.3
We are now imaging the anatomy and distribution
of LA after brachial plexus block with a 3.0 Tesla
scanner giving high-resolution images. To facilitate
the interpretation of each MR image, we provide
magnified drawings and ‘link sequence loops (LSL)’
(Supporting Information Links S1–3). The aim of this
ClinicalTrials.gov identifier: NCT01442857
tures are displaced, and the vein is compressed. Viewing the
images in a sequence loop facilitates identification of the different nerves and has high instructive value (links S1–3 to these
loops are enclosed).
Conclusion: Clinical high-field 3.0 Tesla MRI scanner gives
good visualization of brachial plexus in the axilla. The superior
ability to detect local anaesthetics after it has been injected and
the multiplanar imaging capability make MRI a useful tool in
studies of the brachial plexus.
Accepted for publication 11 March 2012
© 2012 The Authors
Acta Anaesthesiologica Scandinavica
© 2012 The Acta Anaesthesiologica Scandinavica Foundation
study was to demonstrate the anatomy of the brachial
plexus as visualized by MRI.
Methods
The study was in accordance with the Helsinki
declaration, and the regional ethical committee
accepted this study as a preliminary investigation in
front of a randomized study with 45 patients. With
informed consent, we first examined nine volunteers not receiving a block and then nine patients
who had an axillary block prior to upper limb
surgery, all in American Society of Anesthesiologists
physical status I–II. The volunteers were primarily
used for optimizing the MRI protocol. The patients
received, under standardized monitoring, an axillary 18-gauge catheter (1.3 ¥ 45 mm, Contiplex®,
B. Braun, Melsungen, Germany) in median nerve
position guided by nerve stimulator (Stimuplex®
HNS11, B. Braun).4 Insertion point was close to the
lateral border of the pectoralis major muscle and
anterior to the brachial artery. The patients were
1
bs_bs_banner
T. Kjelstrup et al.
Fig. 1. The brachial plexus with scanned
regions (rectangles) and the areas of local
anaesthetic deposits (circles). Picture
modified with permission from G. Meier
and J. Büttner.
pre-medicated with paracetamol orally and fentanyl
intravenously before local infiltration. In seven
patients, the total dose of a bupivacain (5 mg/ml
with 5 mcg/ml epinephrine) and mepivacain
(20 mg/ml) mixture (40 ml) was injected through
this catheter (single injection). The two remaining
patients received the same dose by a triple injection
technique: 20 ml lateral (behind) and 10 ml medial
(in front of) to the brachial artery (transarterial
technique) before 10 ml was injected through the
catheter.4 These two methods and a transarterial
technique with two deposits were used in a
follow-up study with 45 patients.
Two of the seven patients with single deposit
received the injection via the catheter while positioned in the opening of the closed-bore MRI
scanner. They were scanned just before and after the
injection. The other patients were injected outside
the scanner and examined just after the block. All
patients had a block assessment after the scanning.
MR images were achieved by a 3.0 Tesla scanner
(Achieva 3T, Philips Medical Systems, Best, the
Netherlands). Having entered the scanner, each volunteer and patient was in the supine position and
with adducted arms. MR images were obtained
from the proximal humerus to the supraclavicular
region. The scanned region included the terminal
nerves and the cords, as seen in Fig. 1.
The volunteers, not receiving LA, were scanned
by several MRI protocols without fat suppression.
2
This resulted in the optimized MR protocol used for
patient imaging but now with fat suppression. The
final scanner setting for the patients was:
1. Distal cross-sectional images were acquired using
a multislice, angulated, transversal, multishot
rapid acquisition with relaxation enhancement
(RARE), with echo time (TE)/repetition time
(TR) = 65/3193 ms, turbo spin-echo (TSE) factor
= 16 (shot length 122 ms, echo spacing = 7.6 ms),
acquired 1.11 ¥ 1.11 ¥ 2.5 mm, 20 slices, field of
view (FOV) 200 [right-left (RL)] ¥ 242 [anteriorposterior (AP)], bandwidth (BW) = 232.6 Hz
and spectral selection attenuated inversion
recovery (SPAIR) fat saturation with inversion
time (TI) = 100 ms protocol. The scan time was
05:58.
2. Cross-sectional images at the cord level were
acquired using a multislice, angulated, sagital,
multishot RARE, with TE/TR = 100/3078 ms
(two packages), TSE factor = 15 (shot length
122 ms, echo spacing = 8.1 ms), acquired voxel
size 1.10 ¥ 1.13 ¥ 2.5, 20 slices, FOV 220 [height of
FOV (FH)] ¥ 245 (AP), BW = 217.7 Hz and SPAIR
fat saturation with TI = 100 ms protocol. The scan
time was 06:09.
3. Coronal images were acquired using a multislice,
angulated, coronal, multishot RARE, with
TE/TR = 65/7983 ms, TSE factor = 16 (shot length
122 ms, echo spacing = 7.6 ms), acquired 1.20 ¥
Axillary anatomy and brachial plexus block. MRI
Fig. 2. Illustrations of the two crosssectional scanned regions (A) at cord level
and (B) at the level of the terminal nerves.
The brachial plexus is scanned perpendicular to the neurovascular bundle. Local
anaesthetics appear white.
Fig. 3. Magnetic resonance imaging at
cord level, T2-weighted image of a volunteer without local anaesthetics (A) and
T2-weighted image with fat suppression
after local anaesthetics injection of the brachial plexus (B). The components of the
neurovascular bundle are identifiable: the
axillary artery (short arrow), the axillary
vein (long arrow) and the surrounding
cords. The cords appear as black dots. The
lateral cord (LC) is anterior, and the posterior cord (PC) is posterior to the artery.
The medial cord (MC) is located between
the axillary artery and the axillary vein. In
(B), the local anaesthetics has compressed
the axillary vein and displaced the cords.
The cords still remain in periarterial
positions.
1.26 ¥ 2 mm, 50 slices FOV 220 (FH) ¥ 202 (RL),
BW = 221.7 Hz and SPAIR fat saturation with
TI = 100 ms protocol. The scan time was 05:35.
Initial coronal slices were used for determination
of the region to be cross-sectionally scanned and to
examine the extent of the longitudinal LA spread. A
cross-sectional view of the neurovascular bundle
was obtained through scanning of two sections perpendicular to the bundle of nerves and vessels
(Figs 1 and 2). To confirm the identity of the terminal
nerves, the cross-sectional slices were viewed in a
sequence loop.
Results. MRI visualization of cords
and nerves
In the T2-weighted images without fat suppression
of the volunteers (without LA), fat appears grey-
white and blood vessels black (Figs 3A, 4A, 5A, 6A
and 7A). The patients were examined with T2weighted fat suppression, where the signals from fat
are suppressed and the areas containing LA or
liquid are standing out as white (Figs 3B, 4B, 5B, 6B
and 7B). To facilitate the visualization of the
anatomy, images from one volunteer (with the best
overall images) and different patients are arranged
side by side, demonstrating approximately the same
axillary level and nerves.
In the most proximal images, the relation between
the artery, the vein and the three cords is seen
(Fig. 3A). The lateral cord is located anterior to the
artery, the posterior cord is posterior to the artery,
and the medial cord is located between the axillary
artery and vein. The arterial diameter is smaller than
that of the vein. After injection of LA, the vein is
compressed (Fig. 3B). Generally, LA injection displaces the neurovascular structures (Figs 3B, 4B, 5B,
3
T. Kjelstrup et al.
Fig. 4. Magnetic resonance imaging of
the musculocutaneus nerve (MCN),
T2-weighted image (A) and T2-weighted
image with fat suppression after local
anaesthetics injection of the brachial plexus
(B). In (A), the MCN (arrow) is leaving the
lateral cord and enters the coracobrachial
muscle at the level of the humeral head.
From lateral to medial: humeral head,
MCN (arrow), branch of the axillary vein,
axillary artery surrounded by nerves and
axillary vein. In (B), the vascular structures are compressed by local anaesthetics.
The MCN (arrow) is seen on its way into
the coracobrachial muscle. M, medianus;
MC, median cord; PC, posterior cord.
Fig. 5. Magnetic resonance imaging of
the radial, median and ulnar nerve,
T2-weighted image (A) and T2-weighted
image with fat suppression after local
anaesthetics injection of the brachial plexus
(B). In (A), the median nerve (anterior
arrow), ulnar nerve (posterior arrow) and
posterior cord (lateral arrow) are seen
around the brachial artery. In (B), the
structures are displaced by local anaesthetics, which appear as a white sea surrounding the median, ulnar and radial nerve.
PC, posterior cord; M, medianus; R, radialis; U, ulnaris.
6B and 7B) and makes them more difficult to identify than in images without LA.
When viewed from proximal to distal, the MCN
branches off from the lateral cord and enters the
coracobrachial muscle (Fig. 4A). The nerve is clearly
seen also when surrounded by LA (Fig. 4B).
The median nerve consists of one branch from the
lateral cord and one part from the medial cord, and
is located anterior to the axillary artery (Fig. 5A).
The ulnar nerve is a continuation of the medial cord
and is seen medial to the axillary artery (Fig. 5A).
After LA injection (Fig. 5B), the axillary vein is compressed, but the periarterial positions of the nerves
are still traceable.
4
In distal direction, first, the axillary nerve, then
the radial nerve branch off from the posterior cord
(Figs 6A and 7A). The axillary nerve accompanies
the arteria circumflexa humeri posterior at the
humeral head (Fig. 6A). None of the nine patients
had a complete block of the axillary nerve, and as
seen in Fig. 6B, the axillary nerve is not surrounded
by LA.
The radial nerve is positioned lateroposterior to
the axillary artery (Fig. 7A). The nerve leaves the
brachial plexus together with the profunda brachii
artery posterior to humerus and enters the extensor
muscles. The space around the nerve is expanded
and filled with LA (Fig. 7B). In Fig. 7B, the block
Axillary anatomy and brachial plexus block. MRI
Fig. 6. Magnetic resonance imaging of the
axillary nerve, T2-weighted image (A) and
T2-weighted image with fat suppression
after local anaesthetics injection of the brachial plexus (B). In (A), the axillary nerve
(arrow) is seen posterolaterally, branched
off from the posterior cord. It passes to the
back of the arm and enters the quadrangular space accompanied by the posterior circumflex humeral artery. In (B), the axillary
nerve (posterior arrow) is not contacted by
the white local anaesthetics. The brachial
artery (anterior arrow) is surrounded by
three terminal nerves. The brachial vein is
seen compressed between the median and
ulnar nerve. AX, axillaris; M, medianus;
R, radialis; U, ulnaris.
Fig. 7. Magnetic resonance imaging of the
radial nerve, T2-weighted image (A) and
T2-weighted image with fat suppression
after local anaesthetics injection of the brachial plexus (B). In (A), the radial nerve
(arrow), accompanied by the deep brachial
artery, is leaving the plexus sheath lateroposterior to the brachial artery. More distally, they pass the posterior aspect of
humerus. In (B), the radial nerve (arrow) is
seen surrounded by local anaesthetics. The
brachial artery and vein appears as two
black circles. M, medianus; U, ulnaris.
technique includes a transarterial installation of
20 ml lateral to the axillary artery. This explains the
white lateral expansion of LA surrounding the nerve
and corresponds to a complete block of this nerve.
Five of nine patients had complete blocks of the
four terminal nerves. Three of the seven patients in
the catheter group and one of the two patients in the
triple injection group had incomplete blocks of the
radial nerve and received supplementation. One of
the three patients in the catheter group also had an
MCN supplementation.
Conclusion
The high-resolution 3.0 Tesla MRI provides more
detailed visualization of the different cords and
nerves than demonstrated in our previous publication using a 0.5 Tesla MRI.3 Now, identification not
only of cords but also of terminal nerves is possible.
LA injection compresses the veins and displaces the
neurovascular structures making their identification
more difficult. Many anaesthesiologists have, during
the last decade, noticed the ease of recognizing the
5
T. Kjelstrup et al.
terminal nerves by ultrasound.5–7 MRI has, as seen in
this pilot study, a superior ability to detect LA after
it has been injected. This, together with the multiplanar imaging capability, makes MRI a useful tool
for scientific studies of LA distribution in the axilla.
between ultrasound and nerve stimulation guidance for multiple injection axillary brachial plexus block. Anesthesiology
2007; 106: 992–6.
7. Gadsden J, McCally C, Hadzic A. Monitoring during peripheral nerve blockade. Curr Opin Anaesthesiol 2010; 23: 656–61.
Acknowledgement
Address:
Trygve Kjelstrup
Department of Anaesthesiology
Diakonhjemmet sykehus
Pb 23 Vinderen
N 0319 Oslo
Norway
e-mail: [email protected]
We thank Birgitta Kjelstrup for her drawings in Figs 3–7.
Conflicts of interest: The authors report no external funding
and no relevant conflict of interest.
References
1. Brown DL, Cahill DR, Bridenbaugh LD. Supraclavicular
nerve block: anatomic analysis of a method to prevent pneumothorax. Anesth Analg 1993; 76: 530–4.
2. Klaastad O, Lilleas FG, Rotnes JS, Breivik H, Fosse E. Magnetic resonance imaging demonstrates lack of precision in
needle placement by the infraclavicular brachial plexus block
described by Raj et al. Anesth Analg 1999; 88: 593–8.
3. Klaastad O, Smedby O, Thompson GE, Tillung T, Hol PK,
Rotnes JS, Brodal P, Breivik H, Hetland KR, Fosse ET. Distribution of local anesthetic in axillary brachial plexus block: a
clinical and magnetic resonance imaging study. Anesthesiology 2002; 96: 1315–24.
4. Kjelstrup T. Transarterial block as an addition to a conventional catheter technique improves the axillary block. Acta
Anaesthesiol Scand 2006; 50: 112–6.
5. Chan VW, Perlas A, McCartney CJ, Brull R, Xu D, Abbas S.
Ultrasound guidance improves success rate of axillary brachial plexus block. Can J Anaesth 2007; 54: 176–82.
6. Casati A, Danelli G, Baciarello M, Corradi M, Leone S, Di
Cianni S, Fanelli G. A prospective, randomized comparison
6
Supporting information
Additional Supporting Information may be found in
the online version of this article:
Link S1. LSL: Figs 3A, 4A, 5A, 6A, and 7A are all
from volunteer number 7.
Link S2. LSL: Figs 3B, 5B and 6B are from patient 7.
Link S3. LSL: Figs 4B and 7B are from patient 9.
Please note: Wiley-Blackwell are not responsible for
the content or functionality of any supporting materials supplied by the authors. Any queries (other
than missing material) should be directed to the
corresponding author for the article.