Content uploaded by Hilali M Noordeen
Author content
All content in this area was uploaded by Hilali M Noordeen on Nov 05, 2015
Content may be subject to copyright.
Journal of Pediatric Orthopaedics Part B
10
:1᎐5䊚2001 Lippincott Williams & Wilkins
Spinal Cord Monitoring in Neuromuscular Scoliosis
Ž.
U
Ž.
U
Stuart K. Tucker, F.R.C.S. Orth. , M. H. H. Noordeen, F.R.C.S. Orth. , † and
Matthew C. Pitt, M.D., F.R.C.P.†
U
The Royal National Orthopaedic Hospital Trust, Stanmore; and †The Hospital for Sick Children, Great Ormond
Street, London, UK
Summary: This article reviews the use of spinal cord
monitoring in neuromuscular scoliosis, a condition having
a higher incidence of true positive results than idiopathic
scoliosis. While somatosensory cortical evoked potentials
Ž.
SCEP are unreliable, somatosensory spinal evoked po-
Ž.
tentials SSEP are possible to obtain in most cases and a
method using an epidural electrode is described. The ‘50%
rule’ is satisfactory having good specificity and sensitivity
with it rare for post-operative paralysis to have occurred
undetected. The spinal cord in these cases appears to have
increased susceptibility particularly during the passage of
sublaminar wires with the incidence of complications re-
duced using modern instrumentation. Key Words: Spinal
cord monitoringᎏScoliosisᎏNeuromuscular Scoliosis.
Intraoperative monitoring of spinal cord function
is now a routine accompaniment to complex spinal
surgery, providing a continuous means to assess the
integrity of the spinal cord. The recovery of
postoperative neurologic deficits shows a directly
proportional relationship to the speed of removal of
Ž.
instrumentation 11 . Theoretically, earlier detection
of spinal cord dysfunction allows immediate inter-
vention and the possible prevention of any neuro-
logic deficit.
In most spinal surgery units, spinal cord moni-
toring has replaced the Stagnara wake-up test. The
latter, however, is still of clinical importance should
there be a significant change in intraoperative moni-
toring. However, the wake-up test is not always
practical in patients with neuromuscular scoliosis,
owing to muscle weakness and mental retardation
Ž.
18 . Ž.
Somatosensory cortical evoked potentials SCEP
have been found to be unreliable and nonspecific in
neuromuscular scoliosis. Somatosensory spinal
Ž.
evoked potential SSEP monitoring, however, is
useful for the detection of neurologic injury. The
detection of true-positive results is higher in neuro-
muscular scoliosis, when compared with idiopathic
scoliosis. In neuromuscular patients, SSEP moni-
toring is less sensitive and specific, and there are
more false-negative results. An amplified loss, less
Address correspondence to M.H.H. Noordeen FRCS, 107
Harley Street, London, WIN 1DG.
than 50% of baseline, may still be associated with
significant neurologic deficit.
METHOD
The recording technique currently in use at The
Hospital for Sick Children is described. Evoked po-
tential recordings are made, at the time of opera-
Ž.
tion, using a commercial electromyography EMG
and evoked potential machine. Two disposable, ad-
hesive and pre-gelled electrodes are applied to each
popliteal fossa with the cathode proximal, and an
earth electrode placed on the posterior thigh. These
are connected to the stimulator outputs of the
preamplifier.
The evoked potentials are recorded with a 4 F
bipolar electrode, inserted into the epidural space,
cephalad to the highest proposed level of fusion, by
either direct incision of the ligamentum flavum,
through a large bore intravenous cannula or via an
epidural needle placed by the anaesthetist before
skin incision.
The legs are stimulated alternately with the stimu-
lus intensity increased until a reproducible signal
could be identified. If no signal is obtained, technical
failure must be excluded before concluding that the
response is absent. The stimulation duration is 0.5
ms and the rate 20 Hz, with 100 stimuli per test. The
recording parameters are: low-frequency filter, 200
Hz; high frequency, 2 kHz; input sensitivity, 50
Vrcm; and average sensitivity, 0.5
Vrcm.
1
S. K. TUCKER ET AL.2
FIG. 1. Typical visual display.
Cursors are applied manually and the maximum
amplitude difference of the polyphasic waveform
measured, usually of the second or third waveform.
The latency is measured to the start of the response.
Recordings are made repeatedly, with the stimulus
Ž.
alternating between the two lower limbs Fig. 1 .
A failure of monitoring is represented by no
recordable responses after all known cases of techni-
cal error have been excluded. Achievement of re-
sponses from only one lower limb is not considered
a technical failure.
INTERPRETATION OF SPINAL CORD
MONITORING
Approximately one-quarter of monitored patients
show a fall of 10% to 50% in sensory evoked poten-
Ž.
tial SEP amplitude during surgery. A 60% fall in
peak-to-peak amplitude of the evoked potential wave
carried a 10-fold increase in the risk of neurologic
Ž.
complication 10 . The latency of the evoked poten-
tial and the duration of the recorded waveform are
Ž.
TABLE 1. Classification of sensory e¨oked potentials 15
Classification Definition
True-negative No change in trace
No neurologic sequelae
False-negative No change in trace
Neurologic sequelae
True-positive Change in trace that can be related to surgery;
does not necessarily mean neurologic sequelae
False-positve Change in trace cannot be related to surgical
or anaesthetic events
No neurologic sequelae
of no significance in the prediction of neurologic
Ž.
outcome 8 .
The present rule is that an amplitude fall of 50%
from the initial baseline should be considered sig-
Ž.
nificant 3,9,16 and carries a risk of spinal cord
Ž.
injury Table 1 .
False-positive SSEPs are not infrequent during
intraoperative monitoring and are usually caused by
technical problems or environmental factors unre-
lated to surgical manipulation. These include equip-
ment failure, electrode movement, alterations in
blood pressure, depth of anaesthesia and tempera-
ture changes.
Hypotension
Hypotensive anaesthesia has significantly reduced
operative blood loss and blood-product requirement.
It does, however, have a deleterious effect, espe-
cially on SCEPs, but may affect SSEPs to a lesser
degree. Anaesthesia appears to have a significant
effect on the amplitude of SCEPs; especially those
recorded from patients undergoing surgery for neu-
romuscular scoliosis.
Decreased body temperature will typically in-
crease the latency of a response because of slowing
in conduction velocity. A gradual increase in latency
TABLE 2. Modes of neurologic injury
Direct
Sublaminar wirerhook passage
Indirect
Distraction instrumentation
Derotation instrumentation
Ischaemia
J Pediatr Orthop Part B, Vol. 10, No. 1, 2001
SPINAL CORD MONITORING IN NEUROMUSCULAR SCOLIOSIS 3
is common during an operative procedure owing to
Ž.
cooling of the patient 7 . Latency increase without
significant loss of SEP amplitude is not associated
Ž.
with neurologic complications Table 2 .
DISCUSSION
SSEPS are primarily an indication of dorsal
column integrity and do not provide information
about the more clinically relevant motor pathways.
Ž.
However, motor evoked potential MEP monitoring
has not yet been refined sufficiently to facilitate its
use routinely. SSEP monitoring therefore relies on
intraoperative events affecting descending motor
pathways producing concurrent changes in dorsal
column activity.
Postoperative paralysis has been reported despite
Ž.
normal SCEP monitoring perioperatively 2,6 . In
the majority of these cases, the operated deformity
was principally kyphosis, which accounts for 30% of
Ž.
cases of major neurologic complication 11 . It is
likely that the cause of neurologic injury in such
cases is vascular, with infarction, initially of the
anterior portion of the cord, explaining the immedi-
ate postoperative finding of motor paralysis despite
normal SEPs during monitoring. Experimentally,
motor cordotomy has been performed without SEP
Ž.
change 13 .
It is generally assumed that, in the absence of a
50% fall of epidural SEP amplitude, which is not
rapidly reversible, the risk of clinically significant
motor tract damage must be exceedingly low. How-
ever, the need for concurrent, reliable MEP and
SSEP monitoring is clear.
Ž.
Ashkenaze et al. 1 found SCEP monitoring to be
unreliable and nonspecific in 101 patients under-
going neuromuscular scoliosis surgery, and con-
sidered it nonefficacious in preventing or detecting
spinal cord injury when used alone. Of note in their
study was the inability to obtain adequate baseline
tracings in 28% of their patients.
Ž.
Owen et al. 14 found that, with the use of
multiple recording sites, a reliable cortical or sub-
cortical SEP could be obtained in 90% of neuromus-
cular patients. Furthermore, with the addition of
MEPs, a reliable response was present in 96% of
patients. It was possible to obtain reliable myogenic
or neurogenic MEPs in all patients with some motor
function demonstrable before surgery. Owen et al.
Ž.
14 considered that the combined use of multiple
recording sites and multiple modalities improved
their capability to monitor spinal cord function suc-
cessfully in patients with neuromuscular scoliosis.
Ž.
Williamson and Galasko 18 documented 60 neu-
w
romuscular patients 30 Duchenne muscular dystro-
Ž.
phy DMD , 6 myelomeningocele, 6 spinal muscular
Ž.
atrophy SMA , 7 cerebral palsy, 11 miscellaneous
x
conditions . They obtained satisfactory SSEP traces
Ž
in all but two patients one FA, one myelo-
.Ž.
meningocele . Using the classification of Szalay 15 ,
true negative responses occurred in 43 patients and
true-positive responses in 15 patients, invariably as-
sociated with the tightening of a sublaminar wire. In
11 of these patients, the SEP trace returned to its
control latency and amplitude after loosening of the
wire, which was later tightened uneventfully. In one
patient, the SEP trace did not recover until all the
sublaminar wires had been loosened. Instrumenta-
tion was therefore abandoned and an in situ fusion
Ž.
performed, without consequence. Szalay 15 found
no false-positive and no false-negative results, and in
no case did the postoperative neurologic state differ
from the preoperative state.
The largest reported series is that of Forbes et al.
Ž.
5 . They detailed 1168 consecutive cases with SSEP
monitoring in scoliosis surgery, of which 21% were
congenital or neuromuscular curves. No technically
adequate tracing could be obtained in 26 patients
Ž.
2.2% .
In neuromuscular scoliosis and, particularly, the
cerebral palsy and Charcot᎐Marie᎐Tooth groups,
altered neural pathways between the applied stimu-
lus and the sensory cortex may account for the high
rate of nonreproducible tracings. The higher success
rate of obtaining a reproducible waveform in DMD
and polio patients reflects the underlying integrity of
the sensory pathway being tested. Ž.
One hundred and nineteen 10.5% patients
showed a drop in amplitude of greater than 50% of
the initial SEP on one or both sides at some time
during the operation. In 35 of these patients, the
loss of trace resolved spontaneously or when the
recording electrode was repositioned. None of this
group suffered neurologic complications. The re-
maining 84 patients with persistent depression of
more than 50% included all but one of the 33
Ž.
patients 2.7% who suffered neurologic complica-
tions. Of the group with persistent SSEP decre-
ments, 38% had neuromuscular scolioses.
The overall incidence of SSEP fall was 7.2% and
that of neurologic complication was 2.7%. These
incidences were significantly higher in patients with
neuromuscular scoliosis, most due to DMD or SMA,
than for patients with idiopathic and osteogenic
Ž.
curves P-0.01 .
All patients who sustained neurologic complica-
tions at operation recovered to some degree. Most
of the serious episodes were related to distraction of
the spine or to wire placement and tightening. Of
the 84 patients with significant SEP decrements, 53
received instrumentation with an applied distraction
force. All the neuromuscular patients received Luque
instrumentation.
This series has a highly significant false-negative
Ž.
rate P-0.001 , suggesting that, in the absence of
J Pediatr Orthop Part B, Vol. 10, No. 1, 2001
S. K. TUCKER ET AL.4
technical failures, SSEP traces that remain above
50% of their initial value will not be associated with
clinically detectable neurologic complications.
An SSEP fall of 50% or greater in initial SSEP
amplitude should be considered significant. Greater
falls in SSEP amplitude are generally associated
with a higher incidence of neurologic complications.
Although recovery of amplitude is encouraging, it
does not guarantee that there will be no neurologic
complications. Ž.
Noordeen et al. 12 retrospectively reviewed 99
consecutive patients who underwent reconstructive
Ž
spinal surgery for neuromuscular scoliosis 55 DMD,
.
30 SMA, 14 miscellaneous conditions , with SSEP
intraoperative monitoring. Traces could not be ob-
Ž.
tained in only two patients 2% , one with DMD and
the other with myelomeningocele and paraplegia.
They studied the significance of any diminution in
amplitude with three levels; namely, 25%, 50% and
75% of the initial baseline. In the DMD group, 25%
loss of amplitude showed a 90% sensitivity, with 295
specificity and 2 false-negative results; 50% loss of
amplitude had 87 sensitivity with 44% specificity and
3 false-negative results; and 75% loss of amplitude
had a 70% sensitivity with 52% specificity and 7
false-negative results. Results were comparable in
the SMA and miscellaneous groups.
Noordeen et al. concluded that a loss of amplitude
of greater or equal to 50% of the initial baseline
should be used to define abnormality in neuromus-
cular scoliosis, as with idiopathic scoliosis.
The incidence of neurologic injury in neuromuscu-
lar scoliosis is much higher than in idiopathic scolio-
sis. In the neuromuscular group, an amplitude re-
covery greater than 50% may still be associated with
significant, although temporary, neurologic sequelae.
Both these findings suggest an increased sensitivity
of the spinal cord to injury in the neuromuscular
patient.
An amplitude loss of greater than 50% in the
wŽ.x
DMD group seen in 35 patients 69% was at-
Ž
tributable to sublaminar wiring in all cases 34
.
during wire tightening, 1 during wire passage . Neu-
rologic injury was detected in 22 cases but was
temporary in all.
Within the SMA group, 73% had a loss of ampli-
tude greater than 50%, attributable to sublaminar
Ž
wiring in 95% 81% wire tightening, 14% wire pas-
.Ž.
sage . Fifteen of 22 patients had some degree of
neurologic impairment.
In the authors’ experience with third-generation
Ž
instrumentation pedicle screwrhookrrod con-
.
structs , there has been no loss of amplitude in 20
consecutive patients operated on for Duchenne
muscular dystrophy. This is additional evidence to
the previous reporting of the sensitivity of neuro-
muscular patients to the passage of sublaminar wires
Ž.
4,17 .
CONCLUSION
Spinal cord monitoring can reliably be obtained in
approximately 98% of patients with neuromuscular
scoliosis, using SSEPs with epidural placed elec-
trodes rather than SCEPs. A decline in amplitude of
50% of the initial baseline reading should be con-
sidered significant.
The occurrence of postoperative motor paralysis,
despite normal intraoperative SEPs, although re-
ported, is extremely low. However, this reinforces
the urgency for the development of reliable MEP
monitoring in accompanied of SSEP monitoring.
The spinal cord appears to have an increased
susceptibility to injury in neuromuscular patients,
especially during the passage and tightening of sub-
laminar wires. The rates of diminution in SSEP
amplitudes previously reported are much lower us-
ing modern pedicle screwrhookrrod constructs.
REFERENCES
1. Ashkenaze D, Mudiyam R, Boachie-Adjei O, Gilbert C.
Efficacy of spinal cord monitoring in neuromuscular scolio-
sis. Spine 18:1627᎐33.
2. Ben-David B, Haller G, Taylor P. Anterior spinal fusion
complicated by paraplegia; a case report of a false negative
SEP potential. Spine 1987;12:536᎐9.
3. Dolan EJ, Transfeldt EE, Tator CH, Simmons EH, Hughes
KF. The effect of spinal distraction on regional spinal cord
blood flow in cats. J Neurosurg 1980;53:756᎐64.
4. Dove J. Segmental wiring for spinal deformityᎏa morbidity
report. Spinal 14:229᎐31.
5. Forbes HJ, Allen PW, Waller CS, Jones SJ, Edgar MA,
Webb PJ, Ransford AO. Spinal cord monitoring in scoliosis
surgeryᎏexperience with 1168 cases. J Bone Joint Surg Br
1991;73:487᎐91.
6. Ginsbury HH, Shetter AG, Reudzens PA. Post-operative
paraplegia with preserved intra-operative somato sensory
evoked potentials. J Neurosurg 1985;63:296᎐300.
7. Gollenhon D, Kahanovitz N, Happel LT. Temperature ef-
fects on feline cortical and spinal evoked potentials. Spine
1983;8:443᎐6.
8. Jones SJ, Edgar MA, Ransford AO. Sensory nerve conduc-
tion in human spinal epidural recordings made during scol-
isosis surgery. J Neurol Neurosurg Psychiat 1982;45:445᎐51.
9. Jones SJ, Edgar MA, Ransford AO. A system for the elec-
trophysical monitoring of the spinal cord during operation
for scoliosis. J Bone Joint Surg Br 1983;65:134᎐9.
10. Jones SJ, Carter L, Edgar MA, Morley TR, Ransford AO,
Webb PJ. Experience of epidural spinal cord monitoring in
410 cases. In: Schramm J, Jones SJ, eds. Spinal cord moni-
toring. Berlin: Springer-Verlag, 1985:215᎐20.
11. MacEwen GD, Bunel WP, Sriram K. Acute neurological
complications in the treatment of scoliosisᎏa report of The
Scoliosis Research Society. J Bone Joint Surg Am
1975;57:404᎐8.
12. Noordeen MHH, Lee J, Gibbons CER, Taylor BA, Bentley
G. Spinal cord monitoring in operations for neuromuscular
scoliosis. J Bone Joint Surg Br 1997;79:53᎐7.
13. Owen JH, Jenny AB, Naito M, Weber K, Bridwell KH,
McGhee R. Effects of spinal cord lesioning on somatosen-
sory and neurogeni motor evoked potentials. Spine
1989;14:673᎐82.
J Pediatr Orthop Part B, Vol. 10, No. 1, 2001
SPINAL CORD MONITORING IN NEUROMUSCULAR SCOLIOSIS 5
14. Owen JH, Sponseller PD, Szymanski J, Hurdle M. Efficacy
of multimodality spinal cord monitoring during surgery for
neuromuscular scoliosis. Spine 20:1480᎐8.
15. Szalay EA, Carollo JJ, Roach JW. Sensitivity of spinal cord
monitoring to intra-operative events. J Pediatr Orthop
1986;6:437᎐41.
16. Tamaki T, Noguchi T, Takano H, et al. Spinal cord moni-
toring as a clinical utilization of the spinal evoked potential.
Clin Orthop 1984:58᎐64.
17. Wilber RG, Thompson GH, Shaffer JW, Brown RH, Nash
CL. Post-operative neurological deficits in segmental spinal
instrumentation; a study using spinal cord monitoring. J
Bone Joint Surg Am 1984;66:1178᎐87.
18. Williamson JB, Galasko CSB. Spinal cord monitoring during
operative correction of neuromuscular scoliosis. J Bone Joint
Surg Br 1992;74.
J Pediatr Orthop Part B, Vol. 10, No. 1, 2001
