ª Springer Science+Business Media, Inc.

2005 Cardiovasc Intervent Radiol (2005) 28:646–648

CardioVascular Published Online: 25 July 2005 DOI: 10.1007/s00270-004-0282-3
and Interventional
Radiology

CASE REPORTS

Brachial Plexus Injury from CT-Guided RF Ablation Under General

Anesthesia

Sridhar Shankar,1,2,5 Eric vanSonnenberg,1,2 Stuart G. Silverman,1,2 Kemal Tuncali,1,2

Hugh L. Flanagan Jr.,3 Edward E. Whang4
1
Department of Radiology, Brigham and WomenÕs Hospital, Harvard Medical School, Boston, MA, USA
2
Dana-Farber Cancer Institute, Brigham and WomenÕs Hospital, Harvard Medical School, Boston, MA, USA
3
Department of Anesthesia, Brigham and WomenÕs Hospital, Harvard Medical School, Boston, MA, USA
4
Department of Surgery, Brigham and WomenÕs Hospital, Harvard Medical School, Boston, MA, USA
5
Department of Radiology, University of Massachussets, Worcester, MA, USA

Abstract We report a case of profound brachial plexus palsy following a

Brachial plexus injury in a patient under general anesthesia (GA) is
not uncommon, despite careful positioning and, particularly,
awareness of the possibility. The mechanism of injury is stretching
and compression of the brachial plexus over a prolonged period.
Positioning the patient within the computed tomography (CT)
gantry for abdominal or chest procedures can simulate a surgical
procedure, particularly when GA is used. The potential for brachial
plexus injury is increased if the case is prolonged and the patientÕs
arms are raised above the head to avoid CT image degradation
from streak artifacts. We report a case of profound brachial plexus
palsy following a CT-guided radiofrequency ablation procedure
under GA. Fortunately, the patient recovered completely. We
emphasize the mechanism of injury and detail measures to combat
this problem, such that radiologists are aware of this potentially
serious complication.
Key words: Brachial plexus injury—General anesthesia—CT—
RF ablation
Injuries to the brachial plexus are among the most common nerve
injuries. They occur in approximately 0.2% of patients who un-
dergo general anesthesia (GA), despite careful positioning and
awareness of the possibility [1, 2]. The mechanism of injury is
stretching and compression of the brachial plexus over a prolonged
period. Stationing the patient within the computed tomography
(CT) gantry for abdominal or chest procedures can simulate a
surgical procedure, especially if GA is used for a CT-guided in-
terventional radiology procedure. The potential for brachial plexus
injury is increased if the case is prolonged and the patientÕs arms
are raised above the head to avoid CT image degradation from
streak artifacts [1–4].
Correspondence to: Sridhar Shankar, MD, Department of Radiology,
University of Massachussets, Worcester, MA 01655, USA; email: shank-
ars@ummhc.org
CT-guided RF ablation procedure under GA. Fortunately, the pa-
tient recovered completely. The mechanism of injury, and mea-
sures to combat this problem are emphasized to alert radiologists to
this potentially serious complication.
Case Report
A 54-year-old man was referred for radiofrequency (RF) ablation of multiple
gastrointestinal stromal tumor (GIST) metastases to the liver. The primary
tumor in the stomach had been resected 6 years prior, followed by several
other surgical procedures to remove metastatic lesions in the liver, left kidney,
left adrenal, left colon, spleen, and the distal pancreas. The patient was 5 ft 11
in. in height and weighed 140 lbs (65 kg) at presentation to our institution. He
had four lesions in the liver by CT, the largest near the surgical resection
margin, close to the dome of the right hemi-diaphragm. Lesion diameters
ranged from 1.5 to 5 cm. His other significant history included multiple
sclerosis for 15 years, from which he had minimal right-sided paresis.
The patient was treated with RF ablation in four sessions, each
approximately 6 weeks apart. Both cluster and single-pronged probes
(Radionics, Burlington, MA) were utilized. All procedures were performed
under GA with guidance by spiral CT (Somatom Plus 4, Siemens Medical,
Erlangen, Germany) and ultrasound (Sequoia, Acuson, Mountain View,
CA). From onset of preparation and induction of GA to postprocedure
reversal of the GA, the mean inclusive time was approximately 4.5 h per
procedure; the preprocedure evaluation, induction of GA, and positioning
took about 2 h; the RF itself approximately 2 h; and postprocedure reversal
took about 30 min. The patient was positioned on the CT table with his
arms raised over his head (arms extended 90°, abducted 15°, elbows flexed
90°) with adequate cushioning at all pressure points for the first two pro-
cedures; the arms were above the head for the majority of the inclusive
times (Fig. 1, simulated patient position). In constant attendance for all
procedures were interventional radiologists, anesthesiologists, and nurses
experienced in operating room and interventional radiology procedures.
The ablations were successful with complete tumor necrosis in all four
lesions as assessed by follow-up contrast-enhanced magnetic resonance
imaging and positron-emitting tomography scanning.
The patient had mild right arm pain without weakness after the first
procedure that resolved completely after a few days. Following the second
RF procedure, he developed profound weakness of the right arm, but no
S. Shankar et al.: Brachial Plexus Injury from CT-Guided RF Ablation
pain. Neurological consultation at that time revealed 0/5 power in the right
deltoid muscle, 2/5 in the biceps and triceps, and 3/5 in the forearm flexors
and extensors. There was no sensory deficit noted. He was treated with a
sling and physical rehabilitation.
An electromyogram was performed 4 weeks later and revealed a neu-
ropraxic lesion of the brachial plexus with delay in nerve conduction that
confirmed the injury. The initially marked paresis of his right arm improved
gradually over a period of 10 weeks, aided by physical therapy rehabili-
tation. Of note, the last two procedures were performed with modified
positioning when the patient was recovering from his brachial plexus injury.
The patient recovered fully by 3 months.
Discussion
The brachial plexus is particularly vulnerable to injury, because of
both its superficial location in the axilla and its relative fixation.
These fixation points are proximally at the vertebrae and prever-
647
Fig. 1. Incorrect positioning with the patientÕs arm
raised above his head. This position resulted in
brachial plexus injury in our index patient.
Fig. 2. Modified positioning with the patientÕs
arms folded across his chest. This position is an
acceptable compromise to avoid brachial plexus
injury while avoiding artifacts on the CT images.
tebral fascia and distally at the axillary fascia in the arm [1, 3].
Injuries occur from stretching or compression from surrounding
structures such as the adjoining bones and muscles of the shoulder
girdle and vertebrae. The anesthetized patient is susceptible to
brachial plexus injury because he or she is unable to respond to
pain, numbness, paresthesia, or weakness. The susceptibility to
injury is compounded by the use of muscle relaxants during GA
that cause loss of normal muscle tone and sagging of body parts.
The increased vulnerability to pressure point injury in the muscle-
relaxed state is due to abnormal stretching of the plexus. The
longer the stress to the brachial plexus, the more accentuated the
injury is likely to be; in our patient, the time of immobilization that
led to the problem was 4.5 h.
The primary mechanism of injury to nerve plexuses under
anesthesia is ischemia of blood vessels that supply the nerves, the
648 S. Shankar et al.: Brachial Plexus Injury from CT-Guided RF Ablation
vasa nervosa [1–3, 5, 6]. Stretching has been shown to result in
rupture of minute intraneural capillaries with formation of small
hematomas; nerve bundles consequently become compressed. In-
jury is manifested by neural shock that results in numbness, par-
esthesias, and muscular weakness, or even axonal death. These
adverse effects cause complete loss of nerve function until
regeneration of the axon has occurred. Larger hematomas could
result in necrosis of nerve fibers [1, 5, 6].
Percutaneous tumor ablation is a relatively new and attractive
addition to the interventional radiologistÕs armamentarium [7].
As experience grows, complex and larger tumors are being
treated [8–10], necessitating strong anesthetic considerations
such as GA. Patients occasionally request GA to eliminate the
possibility of intraprocedural pain. GA is advantageous if sub-
sequent procedures will be necessary, especially when large
tumor volumes must be eradicated, because patients typically
awaken with minimal to no pain and are receptive to subsequent
procedures.
Specific to CT-guided procedures in the abdomen or chest is the
necessity to raise the arms above the head to avoid beam-hardening
artifact from upper extremity bones. Almost assuredly, the cause of
the brachial plexus injury during our patientÕs CT-guided procedure
under GA was positioning within the constraining CT gantry, with
the hands and arms above the head in the supine position. The
exaggerated abduction of the arms likely caused stretching of the
plexus that led to the injury. GA was used in our patient both to
avoid pain during the extensive ablation procedure and to honor the
patientÕs request.
Brachial plexus injuries secondary to improper positioning have
heretofore been described only in surgical cases, and although
recovery is universal, the time could vary from hours to months
[4]. All of the injuries reported consist of either neuropraxia (a
temporary nerve dysfunction) or axonotmesis (axonal nerve dam-
age, as from compression or crushing, that does not completely
sever the surrounding endoneurial sheath so that regeneration can
occur) [1–6]. There is a broad spectrum of injuries that varies from
those that are subtle and hardly noticed by the patient, to profound
paresis that takes months to recover completely.
We have observed subtler degrees of brachial plexus injury in
two additional patients prior to our patient in this Case Report. The
severity of injury in our patient in this report emphasizes and
warrants the importance of precautions.
Strategies to prevent brachial plexus injuries include the fol-
lowing: (1) positioning the arms crossed across the chest and
slightly raised from the chest itself, with adequate padding, as was
done for our patient for his last two procedures (Fig. 2); (2)
positioning of arms above the head only during the actual imaging
stage, (3) frequent changing of arm position; (4) monitoring nerve
conduction during the procedure for early detection of conduction
abnormalities that might predict neuronal injury; (5) using anes-
thesia other than GA; (6) discussing in detail the positioning,
treatment plans, and time estimation with the anesthesiologists and
nurses prior to the procedure. Although intraprocedural monitoring
of nerve conduction might be helpful (somatic sensory evoked
potentials), adherence to the aforementioned guidelines should be
adequate to prevent injury. Nonetheless, we now include nerve
injury in our informed consent list of possible complications.
To conclude, injury to the brachial plexus under GA during CT-
guided procedures is a hitherto unreported intraprocedural posi-
tional injury. A variety of maneuvers can be used to lessen or
eliminate the risk. Communication among the radiologist, anes-
thesiologist, and nursing staff is essential to optimize care of the
patient by avoiding the injury.
References
1. Cooper DE, Jenkins RS, Bready L, et al. (1998) The prevention of
injuries of the brachial plexus secondary to malposition of the patient
during surgery. Clin Orthop 228:33–41
2. Dawson DM, Kraup C (1989) Perioperative nerve lesions. Arch Neurol
46:1355–1360
3. Cansen EJ (1942) Postoperative paralysis of the brachial plexus. Sur-
gery 12:933–937
4. Kwaan JH, Rappaport I (1970) Postoperative brachial plexus palsy. A
study on the mechanism. Arch Surg 101:612–615
5. Denny-Brown D, Brenner C (1945) Paralysis of nerve induced by di-
rect pressure and tourniquet. Arch Neurol Psychol 5:1–5
6. Denny-Brown D, Doherty MM (1945) Effects of transient stretching of
peripheral nerves. Arch Neurol Psychol 54:116–119
7. Friedman M, Mikityansky I, Kam A, et al. (2004) Radiofrequency
ablation of cancer. Cardiovasc Intervent Radiol 27(5):427–434. Epub
2004 June 03
8. Livraghi T, Goldberg SN, Lazzaroni S, et al. (2000) Hepatocellular
carcinoma: Radio-frequency ablation of medium and large lesions.
Radiology 214(3):761–768
9. Shankar S, vanSonnenberg E, Silverman SG, et al. (2001) Large and
multiple hepatic lesions: Strategies and rationale for radiologic tumor
ablation. In Programs and abstracts, Meeting of The Society of Gas-
trointestinal Radiologists and the Society of Uroradiology, Abdominal
Radiology Postgraduate Course, Scottsdale, AZ, 2001
10. Gervais DA, Arellano RS, Mueller PR (2002) Percutaneous radiofre-
quency ablation of nodal metastases. Cardiovasc Intervent Radiol
25(6):547–549. Epub 2002 Oct 24