Veterinary Anaesthesia and Analgesia, 2010, 37, 144–153 doi:10.1111/j.1467-2995.2009.00518.

RESEARCH PAPER

Ultrasound-guided approach for axillary brachial plexus,

femoral nerve, and sciatic nerve blocks in dogs

Luis Campoy*, Abraham J Bezuidenhout , Robin D Gleed*, Manuel Martin-Flores*, Robert M Rawà,
Carrie L Santare§, Ariane R Jay§ & Annie L Wang§
*Department of Clinical Sciences. College of Veterinary Medicine. Cornell University, Ithaca, NY, USA
 Department of Biomedical Sciences. College of Veterinary Medicine. Cornell University, Ithaca, NY, USA
àDepartment of Anesthesia. University of Iowa Hospital and Clinics, Iowa City, IA, USA
§College of Veterinary Medicine. Cornell University, Ithaca, NY, USA

Correspondence: Luis Campoy, Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853-6401,
USA. E-mail: luis.campoy@cornell.edu

nous nerves and the medial portion of the rectus

Abstract
Objective To describe an ultrasound-guided tech-
nique and the anatomical basis for three clinically
useful nerve blocks in dogs.
Study design Prospective experimental trial.
Animals Four hound-cross dogs aged 2 ± 0 years
(mean ± SD) weighing 30 ± 5 kg and four Beagles
aged 2 ± 0 years and weighing 8.5 ± 0.5 kg.
Methods Axillary brachial plexus, femoral, and
sciatic combined ultrasound/electrolocation-guided
nerve blocks were performed sequentially and
bilaterally using a lidocaine solution mixed with
methylene blue. Sciatic nerve blocks were not
performed in the hounds. After the blocks, the dogs
were euthanatized and each relevant site dissected.
Results Axillary brachial plexus block Landmark
blood vessels and the roots of the brachial plexus
were identified by ultrasound in all eight dogs.
Anatomical examination confirmed the relationship
between the four ventral nerve roots (C6, C7, C8,
and T1) and the axillary vessels. Three roots (C7,
C8, and T1) were adequately stained bilaterally in
all dogs.
Femoral nerve block Landmark blood vessels (femoral
artery and femoral vein), the femoral and saphe-
femoris muscle were identified by ultrasound in all
dogs. Anatomical examination confirmed the rela-
tionship between the femoral vessels, femoral nerve,
and the rectus femoris muscle. The femoral nerves
were adequately stained bilaterally in all dogs.
Sciatic nerve block. Ultrasound landmarks (semi-
membranosus muscle, the fascia of the biceps
femoris muscle and the sciatic nerve) could be
identified in all of the dogs. In the four Beagles,
anatomical examination confirmed the relationship
between the biceps femoris muscle, the semimembr-
anosus muscle, and the sciatic nerve. In the Beagles,
all but one of the sciatic nerves were stained
adequately.
Conclusions and clinical relevance Ultrasound-
guided needle insertion is an accurate method for
depositing localanesthetic for axillary brachial plexus,
femoral, and sciatic nerve blocks.
Keywords brachial plexus block, electrostimulation,
femoral nerve block, local anesthesia, sciatic nerve
block, ultrasound-guided techniques.
Introduction
Ultrasound-guided techniques are becoming
increasingly popular for facilitating certain periph-
eral nerve blocks that are used commonly in people

144

to provide anesthesia for surgery (Marhofer & Chan
2007). Such nerve blocks are used either as
principle anesthetics or as adjuncts to general
anesthesia. Ultrasound guidance uses anatomical
landmarks, including the target nerves themselves,
rather than a neurophysiologic end point (as in
electroneurostimulation). Sonographic guidance for
locoregional anesthesia offers several advantages
when compared with electrolocation; direct visual-
ization of the target nerves may reduce the need for
multiple needle passes and thus reduce tissue dam-
age (Sites & Brull 2006), a reduced risk of vascular
laceration (Gray 2006) and minimized block per-
formance time (Williams et al. 2003). Additionally,
direct visualization of the spread of local anesthetic
during injection, with the possibility of repositioning
the needle in case of maldistribution of local
anesthetic or intravascular needle placement, may
augment the precision and thus reduce the volume
of local anesthetic solution necessary compared
with the use of conventional blind or electrolocation
techniques (Sandhu et al. 2006; Casati et al. 2007;
Oberndorfer et al. 2007).
The objective of this study was to describe an
ultrasound-guided technique and the anatomical
basis for three clinically useful blocks in dogs,
namely the axillary brachial plexus block in the
thoracic limb, the femoral, and the sciatic nerve
blocks in the pelvic limb. Specifically, each block
was performed under general anesthesia using
bidimensional ultrasound imaging in real time to
locate the tip of the needle close to the sonographic
structure thought to be the target nerve. Location
was confirmed using electroneurostimulation. Local
anesthetic solution containing a dye was injected
and its distribution assessed during subsequent post-
mortem anatomical dissection.
Materials and methods
After obtaining approval from the University Insti-
tutional Animal Care and Use Committee (Cornell
University, IACUC number 2007–0029), four
hound-cross dogs aged 2 ± 0 years (mean ± SD)
weighing 30 ± 5 kg and four Beagles aged
2 ± 0 years and weighing 8.5 ± 0.5 kg were used.
The dogs were part of unrelated studies that
required euthanasia.
The hounds were pre-medicated with a combina-
tion of 2 lg kg)1 IM dexmedetomidine (Dexdomitor
hydrochloride; Pfizer Animal Heath, Div. of Pfizer
Inc, NY, USA) plus 0.1 mg kg)1 IM butorphanol
 2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
Ultrasound-guided locoregional techniques L Campoy et al.
(Torbugesic; Fort Dodge Animal Health, IA, USA)
and then thiopental (Pentothal sodium; Hospira
Inc., IL, USA) was administered IV to effect, until
optimal conditions for orotracheal intubation were
achieved. Isoflurane (Isoflurane, USP; Phoenix
Pharmaceutical Inc., MO, USA) in oxygen was
administered via the orotracheal tube and a circle
breathing system to maintain general anesthesia.
For the Beagles, anesthesia was induced and main-
tained with thiopental. All animals were allowed to
breathe spontaneously during injections of local
anesthetic solution.
Axillary brachial plexus, femoral, and sciatic
nerve blocks were performed sequentially and
bilaterally in each dog except that the sciatic block
was not carried out in the hounds because of
experimental exigencies of other investigators. The
skin where echolocation and local anesthetic injec-
tion was to be performed was clipped and prepared
with chlorhexidine gluconate 2% scrub (Chlohexi-
derm plus, IVX, MO, USA). Echolocation was
performed using a high frequency 7–12 mHz linear
array transducer (LA523, Universal Medical Sys-
tems, Inc., NY, USA) and an ultrasound system
(Universal MyLab30, Universal Medical Systems,
Inc, NY, USA). Insulated needles (Stimuplex Insu-
lated Needle; BBraun Medical Inc., PA, USA) con-
nected to a peripheral nerve locator (Innervator
232; Fischer & Paykel, Healthcare, New Zealand)
were used to inject the local anesthetic. Electrosti-
mulation was used to confirm that the needle tip
was proximate to the relevant nerve. In each case, a
square wave stimulating current (frequency 2 Hz;
plateau duration 0.15 ms, plateau current 0.4 mA)
was used (Shanahan & Edmonson 2004). Injectate
for the nerve blocks was 2% lidocaine solution
(Lidocaine Hydrochloride Injectable 2%; Phoenix,
Inc., MO, USA) mixed with 10 mg mL)1 methylene
blue (Methylene Blue; Sigma-Aldrich, Inc., MO,
USA). After completion of the blocks, the dogs were
euthanatized with an IV overdose of pentobarbitone
(Fatal-Plus Solution; Vortech Pharmaceuticals Ltd,
MI, USA) while still under general anesthesia. Each
injection site was dissected and its anatomy
described. Staining of ‡2 cm along the relevant
nerve(s) was considered adequate.
For this study, we have adopted the convention of
presenting all ultrasound images in the orientation
usually seen by a right-handed operator who holds
the ultrasound transducer in their left hand and the
injection needle in their right hand. Under these
circumstances, the needle consistently moves in real
145
Ultrasound-guided locoregional techniques L Campoy et al.
time from the right of the screen towards the center
of the image, thus aiding hand–eye coordination.
The consequence of this is that the head is to the
right of the image for the brachial plexus and
femoral nerve blocks; the head is to the left of the
image for the sciatic nerve block. It should be noted
that this is the case regardless of whether the block
is being performed on the right or the left limb of the
dog.
Axillary brachial plexus block
Location of the transducer in the axilla produced
images of the axillary blood vessels and hyperechoic
structures just dorsal to the vessels that likely were
the brachial plexus, or the nerve roots that serve it.
The dogs in this study were placed in dorsal
recumbency with the thoracic limbs naturally
flexed. In this position, the pectoralis superficialis
muscle, manubrium of the sternum with the
sternocephalicus muscles attached to it, and the
brachiocephalicus muscle can be identified and
palpated. Just cranial to the thoracic inlet, the
external jugular vein lies immediately lateral to the
sternocephalicus muscles (Fig. 1). The axillary area
was then scanned with the transducer orientated in
a parasagittal plane; the transducer was glided,
Figure 1 Dog in dorsal recumbency. The following land-
marks were drawn on the skin: Jugular vein, cranial
border of pectorales muscles, medial border of brachio-
cephalicus muscle and cranial border of sternum. The
puncture site was also marked with an ‘x’. Note that the
stimulating needle is being advanced in a cranial to caudal
direction in-plane with respect to the ultrasound trans-
ducer.
146  2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
Caudal Ventral Cranial
Pectorales superficiales
Pectorales profundus
a
v
Stimulating needle
Dorsal
Figure 2 Ultrasound image of the axillary region. The
stimulating needle has been advanced to the dorsal aspect
of the axillary artery. The solid arrows indicate C6, C7, C8,
and T1 roots.
rotated or tilted until an optimal short axis (trans-
verse) view of the axillary vessels (axillary vein,
axillary artery) was obtained (Fig. 2). The axillary
artery was identified by its characteristic anechoic
pulsatile ultrasound image. Three rounded hyper-
echoic structures were observed dorsal and close to
the axillary vessels; these were presumed to be the
C7, C8, and T1 roots of the brachial plexus (Fig. 2).
The structure of the human brachial plexus differs
substantially from that in the dog. At the intersca-
lene and supraclavicular levels, humans have three
‘trunks’ (superior, middle, and inferior). At the
infraclavicular level, they are referred as ‘cords’
(lateral, medial, and posterior). In the dog, this
anatomical nomenclature has not been adopted;
therefore, we will refer to ‘nerve roots’.
The needle puncture site was identified dorsal to
the cranial edge of the pectoralis superficialis muscle
and lateral to the jugular vein. A 100-mm long 21-
gauge insulated needle was used for the hounds,
and a 50-mm long 22-gauge insulated needle was
used for the Beagles. The long axis of the needle was
placed beneath the long axis of the ultrasound beam
(in-plane technique); this allowed the needle shaft
and tip to be seen while it was being advanced
craniocaudally to the area just dorsal to the axillary
artery and proximate to those hyperechoic struc-
tures presumed to be the brachial plexus roots
(Fig. 2) until triceps brachii muscle twitch (C8
response) and consequent extension of the elbow
was observed. After confirming that blood could not
be aspirated and that there was minimal resistance
 Ultrasound-guided locoregional techniques L Campoy et al.

Caudal
Cranial
Figure 3 Ultrasonographic image of
the axillary region after injection of v a
0.15 mL kg)1 local anesthetic solu-
le
need
tion. The solid arrows indicate C6, lating
Stimu
C7, C8, and T1 roots. Note that the
Local anaesthetic
hypoechoic bleb produced by local
anesthetic extends from root C6 to
root T1.

to injection, 0.15 mL kg)1 of lidocaine/methylene

blue solution was injected. The distribution and
circumferential spread of the solution around the
presumptive nerve roots was observed in real time
by ultrasound (Fig. 3). The same was repeated in
the contralateral limb. The distance from the
transducer to the dorsal wall of the axillary artery
and the length of needle inserted were measured
from the ultrasound images stored prior to injection.

Femoral nerve block

With the dogs in lateral recumbency, the pelvic limb
was abducted 90 and extended caudally (Fig. 4). In
this position, the cranial and caudal bellies of the
sartorius, rectus femoris, vastus medialis, pectineus,
and iliopsoas muscles can be palpated on the medial
(inner) aspect of the thigh. The pulse of the femoral
artery can be palpated in the femoral triangle
between the pectineus and caudal belly of the
sartorius muscle.
The inguinal region of the uppermost leg was
clipped and prepared as noted above. The area was
then scanned ultrasonographically. A hyperechoic
nodular structure presumed to be the femoral nerve
was identified deep and cranial to the femoral artery
and caudal to the fascia of the rectus femoris muscle
(Fig. 5). An in-plane technique was used with the
puncture site located in the quadriceps femoris
muscle (Fig. 4). A 50-mm long 22-gauge insulated
needle was carefully advanced towards the femoral
nerve until there was sonographic evidence of close
Figure 4 Inguinal area of a dog in right lateral recum-
bency with right pelvic limb abducted 90 and extended
caudally. The transducer is placed in the femoral triangle
and the stimulating needle is being introduced through the
quadriceps femoris muscle in-plane with respect to the
ultrasound transducer.
proximity of the tip of the needle to the nerve and
the characteristic twitch of the quadriceps femoris
muscle and consequent extension of the stifle could
be observed. After confirming that blood could not
be aspirated and that there was minimal resistance
to injection, 0.10 mL kg)1 of lidocaine/methylene
blue solution was injected. The distribution and
circumferential spread of the solution around the
femoral nerve was observed in real time by ultra-

 2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153 147
Ultrasound-guided locoregional techniques L Campoy et al.

v Sartorius m.

Rectus femoris m.

Caudal Figure 5 Ultrasonographic image of

Cranial
the femoral triangle region. Note the
femoral artery (a), sartorius muscle,
Femur
rectus femoris muscle and the femo-
ral periosteum. The solid arrow indi-
cates the femoral nerve.

sound. The same was repeated in the contralateral

limb. The distance from the transducer to the center
of the nerve and the length of needle inserted were
measured from ultrasound images stored just prior
to injection.

Sciatic nerve block

The dogs were placed in lateral recumbency with
the leg to be blocked uppermost and extended in a
natural position. The gluteal area and the proximal
caudo-lateral aspect of the thigh was clipped and
prepared as noted above. The ischiatic tuberosity
and greater trochanter were identified and an area
immediately distal to these two landmarks was
scanned ultrasonographically (Fig. 6). The trans-
ducer position was optimized until a hyperechoic
double ellipsoid shape was seen just medial to the
fascia of the biceps femoris muscle and cranial to the
fascia of the semimembranosus muscle. This was
presumed to be a short axis view of the sciatic nerve
(Fig. 7). An in-plane technique was used with the
puncture site located immediately distal to the
ischiatic tuberosity in the caudal aspect of the thigh.
For the Beagles, a 50-mm long 22-gauge insulated
needle was then advanced in a cranial direction
through the semimembranosus and abductor mus-
cles, immediately medial to the fascia of the biceps
femoris muscle, towards the hyperechoic shape
presumed to be the sciatic nerve until a character-
istic motor response was elicited (either dorsiflexion
or plantar extension of the foot). After confirming
that blood could not be aspirated and that there was
Figure 6 Pelvic limb of a dog in right lateral recumbency.
Note the ultrasound transducer placed just distal to a line
between the greater trochanter and the ischiatic tuberos-
ity. The stimulating needle is being introduced through the
semimembranosus muscle in a cranial direction in-plane
with respect to the transducer.
minimal resistance to injection, 0.05 mL kg)1 of
lidocaine/methylene blue solution was injected. The
distribution and circumferential spread of the solu-
tion around the sciatic nerve was observed in real
time by ultrasound. The distance from the trans-
ducer to the center of the nerve and the length of
needle insertion were measured from ultrasound
images stored prior to injection. Because of con-
straints imposed by the other investigations, sciatic
injections could not be carried out in the hounds. In
the hounds, distances were measured from the
ultrasound image with no needle inserted. Results
are reported as mean ± SD.

148  2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
 Ultrasound-guided locoregional techniques L Campoy et al.

Cranial Caudal

Semitendinosus m.
Biceps femoris m.

Figure 7 Ultrasonographic image of

the lateral aspect of the pelvic limb. Semimembranosus m.
Note the two bellies of the semimem-
branosus muscle. The solid arrow
indicates the sciatic nerve. Note the Semimembranosus m.
tibial (caudal) and peroneal (cranial)
components. Stimulating needle

Abductor m.

Results

Axillary brachial plexus block

The landmark blood vessels (axillary artery, axillary
vein) were identified bilaterally by ultrasound in all
eight dogs. The ultrasound images showed the
dorsal wall of the axillary artery was 2.5 ± 0.3-cm
deep in the hounds and 1.7 ± 0.1-cm deep in the
Beagles. When its tip was located immediately dor-
sal to the axillary artery (Fig. 2), the needle had
been inserted 2.2 ± 0.3 cm in a craniocaudal
direction in the hounds and 1.7 ± 0.1 cm in the
Beagles. Electrostimulation consistently produced
contraction of the triceps brachii muscle and Figure 8 Dissection of the brachial plexus of a dog in
extension of the elbow without repositioning the dorsal recumbency. The view and positioning are similar
needle. to that in Fig. 1. Note the intimate relationship of the
brachial plexus roots located immediately dorsal to the
Post-mortem anatomical dissection was carried
axillary vessels.
out with the dogs in dorsal recumbency. The four
ventral nerve roots (C6, C7, C8, and T1) that
contribute to the brachial plexus were identified and was either not stained or insufficiently stained in all
their dorsal relationship to the axillary vessels was dogs. No evidence of trauma to the roots or any
confirmed (Fig. 8). The nerve roots were distinct adjacent structures was noted.
structures (7-cm long in hounds and 5-cm long
in beagles) before converging to form the brachial
Femoral nerve block
plexus. Methylene blue staining was confined to the
roots in the area between where they crossed the The landmark blood vessels (femoral artery and
ventro-lateral border of the scalenus muscle and the femoral vein), the femoral and saphenous nerves
proximal aspect of the brachial plexus. Three roots and the medial portion of the rectus femoris muscle
(C7, C8, and T1) were adequately stained bilaterally were identified by ultrasound bilaterally in all dogs.
in all dogs. Pooled values for the length of these The ultrasound image showed the center of the
stained nerve roots was 6.6 ± 0.6 cm in the hounds femoral nerve was 0.9 ± 0.1-cm deep in the hounds
and 4.0 ± 0.6 cm in the beagles. The C6 nerve root and 0.6 ± 0.1-cm deep in the beagles. The length of

 2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153 149
Ultrasound-guided locoregional techniques L Campoy et al.
Figure 9 Dissection of the femoral triangle in the pelvic
limb. The caudal belly of the sartorius muscle has been
displaced cranially to allow view of the femoral nerve. Note
the relationship of the femoral vessels, the femoral nerve
and the rectus femoris muscle.
needle insertion was 4.8 ± 1.7 cm in the hounds
and 4.0 ± 1.3 cm in the beagles. Post-mortem
anatomical examination confirmed the relationship
between the femoral vessels, femoral nerve and the
rectus femoris muscle (Fig. 9). Electrostimulation
consistently produced contraction of the quadriceps
femoris muscle and extension of the stifle without
repositioning the needle. The femoral nerves were
adequately stained bilaterally to a length of
4.8 ± 1.7 cm in the hound dogs and 4 ± 1.3 cm in
the Beagles.
Sciatic nerve block
Ultrasound landmarks such as the semimembrano-
sus muscle, the fascia of the biceps femoris muscle
and the sciatic nerve could be identified in all of the
dogs. The ultrasound image showed the sciatic
nerve 1.7 ± 0.1-cm deep in the hounds and
1.5 ± 0.2 cm in the beagles. Length for needle
insertion was 3.1 ± 0.8 cm in the hounds (distance
from theoretical skin puncture site to sciatic nerve)
and 2.7 ± 0.3 cm in the beagles. Post-mortem
anatomical examination confirmed the relationship
between the biceps femoris muscle, the semi-
membranosus muscle and the sciatic nerve
(Fig. 10). Sciatic injections and electrostimulation
were not carried out in the hounds (see above). In
the four beagles, without repositioning the needle
tip, electrostimulation produced plantar extension of
the foot in seven of eight sciatic nerve blocks; in the
remaining case electrostimulation produced con-
tractions of the caudal thigh muscles (semitendi-
nosus and semimembranosus muscles). In the seven
150  2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
Figure 10 Dissection of thigh showing the sciatic nerve.
Lateral view. Note that the biceps femoris muscle has been
lifted to allow view of the sciatic nerve and surrounding
structures.
nerve blocks where plantar extension of the foot
was seen on electrostimulation, the sciatic nerves
were stained adequately to a length of
2.8 ± 0.3 cm. In the single block that produced
caudal thigh muscle contraction on electrostimula-
tion, only the muscular branch of the sciatic nerve
was stained (for 7.6 cm); this was counted as a
failed sciatic nerve block.
Discussion
Success of a nerve block is dependent on placing
local anesthetic solution in close proximity to the
appropriate nerve. Electrostimulation has been
considered the ‘gold standard’ technique for
peripheral nerve localization (Marhofer & Chan
2007). However, ultrasound-guided techniques are
gaining popularity to facilitate peripheral nerve
blocks as this technique provides the ability to both
manipulate the needle under direct guidance and
see the spread of the local anesthetic solution as it is
being injected. In humans, this may increase the
safety and efficacy of the procedure (Perlas et al.
2003) and decrease the time required for block
performance (Williams et al. 2003). Additionally,
lower volumes may be necessary as direct visuali-
zation of the spread of local anesthetic can be
observed (Oberndorfer et al. 2007). Furthermore,
the combination of ultrasound guidance and elec-
trolocation offers the advantage of the anatomical
as well as electrophysiological confirmation of nerve
identification and needle placement. However, in
one study, despite sonographic evidence of the

needle tip in close proximity to the nerves, a motor
response at or below 0.5 mA could only be elicited
in 42% of otherwise successful blocks in people
(Sinha et al. 2007). In the dogs presented here,
motor response was one of the end points used to
confirm a correct location of the needle; a motor
response with a current of 0.4 mA was obtained in
all blocks. The initial current used in this study
(0.4 mA) is much less than the initial current used
when electrolocation only is performed (1 mA)
(Shanahan & Edmonson 2004). This suggests that
ultrasound location, confirmed by low current
electrolocation, as described in this study, could be
tolerated by selected canine patients that are
sedated, whereas the higher initial current used in
conventional electrolocation usually requires more
extensive CNS depression. The absence of any nee-
dle adjustment after ultrasound location in this
study suggests that confirmation by electrical stim-
ulation may become unnecessary with experience.
The brachial plexus in the dog is formed by the
ventral branches of the last three cervical (C6, C7,
C8) and the first thoracic (T1) spinal nerves. In
some individuals, C5 and T2 may also contribute to
the brachial plexus (Redding et al. 1982; Sharp
et al. 1990, 1991). After the roots of C6, C7, C8,
and T1 exit the cervical and thoracic spinal column
through the intervertebral foramina and the inter-
transversarius musculature, there is exchange of
nerve fibers between them. The four roots (C6-T1)
emerge through and cross the ventro-lateral border
of the scalenus muscle (Fig. 8). Then, they divide to
form the brachial plexus and, after the roots cross
the axillary space, they form the individual nerves
that provide the sensory and motor supply of the
thoracic limb. The ventral root of C6 (with some
input from C7) is the main contributor to the
suprascapular nerve; C7 (with some input from C6)
is the main contributor to the musculocutaneous
and subscapular nerves; C8 (with some input from
T1) serves the radialis nerve and T1 (with some
input from C8) is the main contributor to the
median and ulnar nerves. Dissection of the axillary
area consistently showed adequate staining of the
roots of C7, C8, and T1 but not C6. Based on this,
we anticipate that the axillary brachial plexus block,
as described here, will produce local anesthesia of
the thoracic limb distal to the shoulder. To achieve a
complete blockade of the shoulder joint, the needle
would need to be repositioned during the course of
the injection towards a more cranial location to
block the root of C6.
 2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
Ultrasound-guided locoregional techniques L Campoy et al.
The ultrasonographic anatomy of the brachial
plexus in dogs in dorsal recumbency has been
described by Guilherme & Benigni (2008). However,
the axillary approach for conduction blockade of the
brachial plexus has not been described previously in
veterinary medicine. An infraclavicular nerve block
in people is performed at a corresponding location
producing a similar ultrasonographic image (Tran de
et al. 2008). In dorsal recumbency, the weight of
the thoracic limb passively opens the axillary space
allowing the nerve roots (C6-T1) to be identified
ultrasonographically as they traverse this space
dorsal to the axillary artery and axillary vein (Figs 2
& 8). The ease of the technique and the apparent
reliability and repeatability of the injection suggests
that this approach is likely to be useful clinically,
particularly in light of the frequent failure rate of the
conventional blind approach. It should be pointed
out that the target nerve roots are very close to the
axillary vessels when performing the axillary
approach to the brachial plexus (Fig. 8); this sug-
gests that needle placement without the benefit of
imaging might increase the risk for lacerating a
vessel and producing an axillary hematoma.
The electrostimulation needles used in this study
have a 30 bevel and are designed to be ‘atraumat-
ic’. This adds a measure of security because they do
not readily penetrate blood vessels or epineurium.
Using the axillary approach with ultrasound,
electrostimulation produced extension of the elbow
in all cases. Extension of the elbow is caused by
triceps brachii contraction and is characteristic of
radial nerve stimulation; the latter nerve is princi-
pally served by the C8 vertebral nerve root.
In this study, 0.15 mL kg)1 of local anesthetic
was used for the axillary brachial plexus block; this
volume is substantially less than previously pub-
lished for the conventional approach in lateral
recumbency (0.25–0.4 mL kg)1) (Duke et al.
1998; Duke 2000; Futema et al. 2002; Wenger
2004; Wenger et al. 2005). If the axillary approach
with the lower dose proves clinically effective, it will
probably reduce the risk of overdose and the
incidence of side effects, particularly when several
nerve blocks need to be carried out in the same
patient.
The femoral nerve enters the pelvic limb from the
iliopsoas muscle through the femoral canal. It then
runs deep to the caudal belly of the sartorius muscle
in between the rectus femoris and vastus medialis
muscles. The femoral vein and artery are located
caudal to the nerve (Fig. 9). On ultrasound, the
151
Ultrasound-guided locoregional techniques L Campoy et al.
femoral nerve in the femoral triangle region is
imaged as a nodular hyperechoic structure lying
cranial and deep relative to the femoral artery
(Fig. 5). There are no studies substantiating the
optimal volume to be injected in a femoral nerve
block in the dog. Oberndorfer et al. (2007) reported
a volume of 0.3 mL kg)1 in children when blocking
the femoral nerve using electrolocation. However,
in the same study, when an ultrasound-guided
technique was used, 0.15 mL kg)1 was sufficient to
achieve adequate blockade. In this study, using
ultrasound location in dogs, a volume of 0.1 mL
kg)1 of local anesthetic solution was used; this
stained sufficient nerve length for us to anticipate
that femoral nerve conduction block would be
complete.
In the gluteal region, the sciatic nerve lies
between the superficial gluteal muscle laterally
and the gemelli and quadratus femoris muscles
medially. The sciatic nerve exits the pelvis through
the greater sciatic foramen. It descends between the
greater trochanter and the ischiatic tuberosity. It
then runs between the biceps femoris muscle
laterally and the abductor muscle medially and
semimembranosus muscle medially and caudally.
Its division into tibial and peroneal nerves is
variable and can be anywhere from the level of
the hip joint to just above the stifle (Miller et al.
1993). The muscular branch of the sciatic nerve
gives off branches to the caudal thigh muscles as far
distally as the distal part of the semimembranosus
muscle (Fig. 10). In its proximal portion, the sciatic
nerve is accompanied by the caudal gluteal artery
and vein which lie caudal to the nerve.
On ultrasound, the sciatic nerve is difficult to
image in its short axis since it has a flat structure. It
can be seen as a hyperechoic ellipsoid just medial to
the fascia of the biceps femoris and cranial to the
fascia of the semimembranosus muscle (Fig. 7).
Occasionally, pulsations of the caudal gluteal vessels
can be seen caudal to the nerve in between the
semitendinosus muscle and the distal body of the
semimembranosus muscle. It is also the same
location where the muscular branch of the sciatic
nerve is found.
In seven of eight blocks plantar extension of the
foot was observed with electrostimulation; this is
characteristic of tibial nerve stimulation. In one of
the dogs, the muscular branch and not the sciatic
nerve was imaged erroneously and subsequently
stained. This could have been avoided had the
ultrasound transducer been positioned as proximal
152  2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
as possible, just below the ischiatic tuberosity
where the branches are very close to the sciatic
nerve and the fascias of the biceps femoris and
semimembranosus muscle can be identified sono-
graphically. The sciatic nerve is medial to the
thickest part of the biceps femoris, lateral to the
abductor muscle and cranial to the semimem-
branosus muscle. Based on staining of the nerve,
a volume of 0.05 mL kg)1 of local anesthetic
appeared to be adequate for sciatic blockade; this
is consistent with the volume previously used to
produce blockade of the sciatic nerve using electro-
location (Campoy et al. 2008).
In this study, the assumption is made that the
length of nerve stained relates to the efficacy of
conduction blockade. In order for nerve conduction
to be successfully blocked, a critical length of nerve
must be in contact with local anesthetic solution at
sufficient concentration (Nakamura et al. 2003). In
myelinated nerves, at least three nodes of Ranvier
must be exposed to local anesthetic to ensure that
nerve conduction is halted; this corresponds to
approximately 3–4 mm of nerve (Raymond et al.
1989). Nonmyelinated C nerve fibers are also
responsible for conduction of nociception and are
even more susceptible to conduction blockade than
the myelinated fibers investigated above. Hence, we
consider the standard we chose that (‡2 cm of
nerve staining is evidence of adequate block) to be
conservative.
Conclusion
Ultrasound-guided needle insertion is an accurate
method for depositing local anesthetic for axillary
brachial plexus, femoral, and sciatic nerve blocks.
The axillary approach to the brachial plexus is a
novel approach for providing regional anesthesia of
the thoracic limb which may prove superior to the
conventional approach to the brachial plexus. The
proposed advantages of these three blocks when
they are carried out with sonolocation (e.g.
decreased dose, decreased side effects, decreased
performance time, decreased failure rate) must be
confirmed in future clinical trials.
References
Campoy L, Martin-Flores M, Looney AL et al. (2008) Dis-
tribution of a lidocaine-methylene blue solution staining
in brachial plexus, lumbar plexus and sciatic nerve
blocks in the dog. Vet Anaesth Analg 35, 348–354.

Casati A, Baciarello M, Di Cianni S et al. (2007) Effects of
ultrasound guidance on the minimum effective anaes-
thetic volume required to block the femoral nerve. Br J
Anaesth 98, 823–827.
Duke T (2000) Local and regional anesthetic and analgesic
techniques in the dog and cat: part ii, infiltration and
nerve blocks. Can Vet J 41, 949–952.
Duke T, Cullen CL, Fowler JD (1998) Anesthesia case of the
month. J Am Vet Med Assoc 212, 649–650.
Futema F, Tabacchi D, Costa JO et al. (2002) A new bra-
chial plexus block technique in dogs. Vet Anaesth Analg
29, 133–139.
Gray AT (2006) Ultrasound-guided regional anesthesia:
current state of the art. Anesthesiology 104, 368–373.
Guilherme S, Benigni L (2008) Ultrasonographic anatomy
of the brachial plexus and major nerves of the canine
thoracic limb. Vet Radiol Ultrasound 6, 577–583.
Shanahan J, Edmonson KG (2004) Peripheral Nerve
Stimulators and Nerve Stimulation. In: Peripheral Nerve
Blocks: Principles and Practice. Hadzic A, Vloka JD (eds).
McGraw-Hill, New York, USA. pp. 43–49.
Miller ME, Evans HE, Christensen GC (1993) The spinal
nerves. In: Miller’s Anatomy of the Dog (3rd edn). Evans
HE (ed). W.B. Saunders Company, Philadelphia, USA.
pp. 829–893.
Marhofer P, Chan VW (2007) Ultrasound-guided regional
anesthesia: current concepts and future trends. Anesth
Analg 104, 1265–1269.
Nakamura T, Popitz-Bergez F, Birknes J et al. (2003) The
critical role of concentration for lidocaine block of
peripheral nerve in vivo: studies of function and drug
uptake in the rat. Anesthesiology 99, 1189–1197.
Oberndorfer U, Marhofer P, Bosenberg A et al. (2007)
Ultrasonographic guidance for sciatic and femoral nerve
blocks in children. Br J Anaesth 98, 797–801.
Perlas A, Chan VW, Simons M (2003) Brachial plexus
examination and localization using ultrasound and
electrical stimulation: a volunteer study. Anesthesiology
99, 429–435.
Raymond SA, Steffensen SC, Gugino LD et al. (1989) The
role of length of nerve exposed to local anesthetics in
impulse blocking action. Anesth Analg 68, 563–570.
 2010 The Authors. Journal compilation  2010 Association of Veterinary Anaesthetists, 37, 144–153
Ultrasound-guided locoregional techniques L Campoy et al.
Redding RW, Ingram JT, Colter SB (1982) Sensory nerve
conduction velocity of cutaneous afferents of the radial,
ulnar, peroneal, and tibial nerves of the dog: reference
values. Am J Vet Res 3, 517–521.
Sandhu NS, Bahniwal CS, Capan LM (2006) Feasibility of
an infraclavicular block with a reduced volume of lido-
caine with sonographic guidance. J Ultrasound Med 25,
51–56.
Sharp JW, Bailey CS, Johnson RD et al. (1990) Spinal
nerve root origin of the median, ulnar and musculocu-
taneous nerves and their muscle nerve branches to the
canine forelimb. Anat Histol Embryol 4, 359–368.
Sharp JW, Bailey CS, Johnson RD et al. (1991) Spinal root
origin of the radial nerve and nerves innervating
shoulder muscles of the dog. Anat Histol Embryol 3,
205–214.
Sinha SK, Abrams JH, Weller RS (2007) Ultrasound-gui-
ded interscalene needle placement produces successful
anesthesia regardless of motor stimulation above or
below 0.5 mA. Anesth Analg 105, 848–852.
Sites BD, Brull R (2006) Ultrasound guidance in peripheral
regional anesthesia: philosophy, evidence-based medi-
cine, and techniques. Curr Opin Anaesthesiol 19, 630–
639.
Tran de QH, Clemente A, Tran DQ et al. (2008) A com-
parison between ultrasound-guided infraclavicular block
using the ‘‘double bubble’’ sign and neurostimulation-
guided axillary block. Anesth Analg 107, 1075–1078.
Wenger S (2004) Brachial plexus block using electrolo-
cation for pancarpal arthrodesis is a dog. Vet Anaesth
Analg 31, 272–275.
Wenger S, Moens Y, Jaggin N et al. (2005) Evaluation of
the analgesic effect of lidocaine and bupivacaine used to
provide a brachial plexus block for forelimb surgery in
10 dogs. Vet Rec 156, 639–642.
Williams SR, Chouinard P, Arcand G et al. (2003) Ultra-
sound guidance speeds execution and improves the
quality of supraclavicular block. Anesth Analg 97,
1518–1523.
Received 23 September 2008; accepted 15 March 2009.
153