Accepted Manuscript Somatosensory evoked potential monitoring during instrumented scoliosis corrective procedures: validity revisited Parthasarathy Thirumala, MD Lance Bodily, BS Derrick Tint, BS W. Timothy Ward, MD Vincent F. Deeney, MD Donald Crammond, PhD Miguel E. Habeych, MD, MPH Jeffrey Blazer, PhD PII:
S1529-9430(13)01575-1
DOI:
10.1016/j.spinee.2013.09.035
Reference:
SPINEE 55580
To appear in:
The Spine Journal
Received Date: 26 October 2012 Revised Date:
16 August 2013
Accepted Date: 19 September 2013
Please cite this article as: Thirumala P, Bodily L, Tint D, Ward WT, Deeney VF, Crammond D, Habeych ME, Blazer J, Somatosensory evoked potential monitoring during instrumented scoliosis corrective procedures: validity revisited, The Spine Journal (2013), doi: 10.1016/j.spinee.2013.09.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Somatosensory evoked potential monitoring during instrumented scoliosis corrective procedures: validity revisited.
RI PT
Parthasarathy Thirumala, MD Department of Neurological Surgery UPMC Presbyterian Suite B-400 200 Lothrop Street Pittsburgh, PA 15213 Phone: (412) 648-2570 [email protected]
W. Timothy Ward, MD UPMC, Orthopedic Surgery Pittsburgh, PA
Donald Crammond, PhD UPMC, Neurosurgey Pittsburgh, PA
EP
Miguel Habeych, MD, MPH UPMC, Neurosurgey Pittsburgh, PA
TE D
Vincent F. Deeney, MD UPMC, Orthopedic Surgery Pittsburgh, PA
M AN U
Derrick Tint, BS University of Pittsburgh School of Medicine Pittsburgh, PA
SC
Lance Bodily, BS University of Pittsburgh School of Medicine Pittsburgh, PA
AC C
Jeffrey Blazer, PhD UPMC, Neurosurgey Pittsburgh, PA
Key Words: Scoliosis; Somatosensory evoked potential; Transcranial motor evoked potential Word Count: 4,355
ACCEPTED MANUSCRIPT
Somatosensory evoked potential monitoring during instrumented scoliosis
2
corrective procedures: Utility revisited.
3
Abstract:
4
Background: Intraoperative monitoring (IOM) using somatosensory evoked
5
potentials (SSEP's) plays an important role in reducing iatrogenic neurological
6
deficits during corrective pediatric idiopathic scoliosis procedures. However, for
7
unknown reasons recent reports have cited that the sensitivity of SSEPs to detect
8
neurological deficits has decreased, in some to be less than 50%. This current trend,
9
which is coincident with the addition of transcranial electrical motor evoked
10
potentials (tceMEP's), is surprising given that SSEPs are robust, reproducible
11
responses that were previously shown to have sensitivity and specificity of >90%.
12
Purpose: Our primary aim was to assess whether SSEPs alone can detect impending
13
neurological deficits with similar sensitivity and specificity as originally reported.
14
Our secondary aim was to estimate the potential predictive value of adding
15
tceMEP's to SSEP's monitoring in idiopathic scoliosis procedures.
16
Design: This was a retrospective review to analyze the efficacy of SSEP monitoring
17
in the group of pediatric instrumented scoliosis cases.
18
Patient sample: We retrospectively reviewed all consecutive cases of patients who
19
underwent idiopathic scoliosis surgery between 1999 and 2009 at Children's
20
Hospital of Pittsburgh. We identified 477 patients who had the surgery with SSEP
21
monitoring alone. Exclusion criteria included any patients with neuromuscular
AC C
EP
TE D
M AN U
SC
RI PT
1
1|Page
ACCEPTED MANUSCRIPT
disorders or unreliable SSEP monitoring. Patients who had incomplete
2
neurophysiology data or incomplete postoperative records were also excluded.
3
Outcome measures:Major outcomes measured were clinically significant
4
postoperative sensory or motor deficits, as well as significant intraoperative SSEP
5
changes.
6
Methods: Continuous interleaved upper and lower extremity SSEP's were obtained
7
throughout the duration of all procedures. We considered a persistent 50%
8
reduction in primary somatosensory cortical amplitude or a prolongation of
9
response latency by >10% from baseline to be significant. Persistent changes
10
represent significant deviation in SSEP amplitude or latency in more than two
11
consecutive averaged trials. Patients were classified into one of four categories with
12
respect to SSEP monitoring: true positive, false positive, true negative, and false
13
negative. The sensitivity, specificity, positive predictive value (PPV) and negative
14
predictive value (NPV) were then calculated accordingly.
15
Results: Our review of 477 idiopathic scoliosis surgeries monitored using SSEP's
16
alone revealed a new deficit rate of 0.63% with no cases of permanent injury.
17
Sensitivity=95.0%, specificity=99.8%, PPV=95%, NPV=99.8%. Using evidence-
18
based epidemiological measures, we calculated that the number needed to treat
19
(NNT) was 1,587 patients for one intervention to be performed that would have
20
been missed by SSEP monitoring alone. Additionally, the number needed to harm,
21
which represents the increase in false positives with the addition of tceMEP's, was
22
200.
AC C
EP
TE D
M AN U
SC
RI PT
1
2|Page
ACCEPTED MANUSCRIPT
Conclusion: SSEP monitoring alone during idiopathic scoliosis continues to be a
2
highly reliable method for the detection and prevention of iatrogenic injury. Our
3
results confirm the high sensitivity and specificity of SSEP monitoring alone
4
published in earlier literature. As such, we suggest the continued use of SSEP alone
5
in idiopathic scoliosis surgeries. At this time we do not believe there is sufficient
6
data to support the addition of MEP monitoring, although further studies and
7
revised criteria for the use of MEP may provide added value for its use in the future.
SC
RI PT
1
M AN U
8 9 10
TE D
11 12
15 16 17
AC C
14
EP
13
18 19 3|Page
ACCEPTED MANUSCRIPT
1
3
Intro: For pediatric patients with severe scoliosis, instrumented spinal fusions are
RI PT
2
essential to halt progression and ameliorate symptoms. Iatrogenic injury during
5
scoliosis procedures can occur, most often related to derotation of the spinal column
6
causing direct injury to the spinal cord or impinging on the vasculature1. While the
7
incidence of iatrogenic spinal cord injury resulting in paraplegia is low, this is a
8
feared complication as the patients are generally young and otherwise healthy.
9
Intraoperative monitoring (IOM) using somatosensory evoked potentials (SSEP)
10
plays an important role in reducing iatrogenic neurological deficits during these
11
corrective scoliosis procedures. A large multi center study reported a reduction of
12
persistent and major persistent neurological deficits to decrease from 0.72%
13
without monitoring to 0.55% with SSEP monitoring2. The most recent reports cite
14
total neurological deficits from scoliosis surgery, the sum of transient and persistent
15
deficits, to be 0.99% and 1.84% for pediatric and adult patients respectively. Lower
16
incidence rates were reported in idiopathic cases having rates of 0.73% and 1.45%
17
for pediatric and adult patients respectively3.
18
Multiple research studies have reported that addition of transcranial motor evoked
19
potential (TcMEP) monitoring to SSEPs increases the sensitivity of IOM significantly,
20
during scoliosis surgery. Many of these studies conclude that TcMEP is a better
21
assay of the motor tracts, and suggest that it has greater efficacy in preventing
22
motor deficits4,5. Moreover, for reasons that are unclear, the reported sensitivity of
AC C
EP
TE D
M AN U
SC
4
4|Page
ACCEPTED MANUSCRIPT
SSEPs to detect a neurological deficit has decreased from close to 100% in large
2
studies 2,6 to less than 50%4,5,7 in more recent studies over the last decade. This new
3
development is surprising given that SSEPs provide robust responses that were
4
previously shown to be highly sensitive and specific1,2_ENREF_1. In addition,
5
TcMEPs lack well-defined and accepted alarm criteria to alert the surgeon about an
6
impending neurological deficit. Our primary aim is to assess whether SSEP alone
7
can detect impending neurological deficits with similar sensitivity and specificity as
8
previously reported. We also try to determine the potential need and value of
9
additional monitoring with TcMEPs, considering the current alarm criteria being
M AN U
SC
RI PT
1
used by the field.
11
Materials and Methods
12
Study Design
13
We retrospectively reviewed consecutive cases of patients who underwent scoliosis
14
surgery for idiopathic and congenital scoliosis between 1999 and 2009. We
15
identified 477 patients who had the surgery with SSEP monitoring. These
16
procedures were performed by the same surgeons (WTW, VD, JSG, SAM) at UPMC
17
Children’s Hospital.
18
Inclusion criteria included all patients with idiopathic and congenital scoliosis
19
undergoing instrumented spinal fusions for scoliosis correction. Exclusion criteria
20
included any patients with neuromuscular disorders (neurogenic scoliosis) or
21
abnormal baseline SSEP signals that did not allow monitoring. Patients who had
AC C
EP
TE D
10
5|Page
ACCEPTED MANUSCRIPT
incomplete neurophysiology data or incomplete postoperative records were also
2
excluded.
3
Neurophysiologic Monitoring:
4
Intraoperative SSEP monitoring is standard at our institution for all instrumented
5
spinal fusions of scoliosis patients based on its documented efficacy1. Baseline SSEP
6
values were obtained after the induction of anesthesia but before patient
7
positioning in all cases. Continuous upper and lower extremity interleaved SSEPs
8
were obtained throughout the procedure. Physician oversight and interpretation
9
was performed using a combined on-site and remote model implemented at UPMC.
M AN U
SC
RI PT
1
In all cases, a board-certified Neurophysiologist (American Board of
11
Neurophysiological Monitoring) was on-site, in person available for interpretation
12
while physician (neurologists) oversight, supervision, and interpretation were
13
performed in-person on-site or remotely. The oversight physician provided
14
supervision to 4-5 cases simultaneously on average with a maximum of 8 cases.
EP
15
TE D
10
Upper Extremity SSEPs:
17
Median or ulnar nerve stimulation was performed bilaterally in an alternating
18
fashion at the wrist with sub dermal needle electrode pairs. P4/Fz and P3/Fz scalp
19
electrodes montages were used (per the international 10–20 system) for recording
20
cortically generated responses. A cervical electrode was localized at the C7 spinous
21
process or mastoid (M) and referenced to Fz was used for subcortical recordings.
AC C
16
6|Page
ACCEPTED MANUSCRIPT
Stimulation frequency was 2.45 Hz with duration of 0.2 milliseconds. Band pass
2
filters were set at 10 to 300 Hz with a gain of 20k for cortical recordings and 30 to
3
1,000 Hz for subcortical recordings with a gain of 50K. Averages were computed for
4
either 128 or 256 trials, depending on the signal to noise ratio.
5
RI PT
1
Lower Extremity SSEPs.
7
For the lower extremities, bilateral alternating tibial nerve stimulation was used.
8
Peroneal nerve stimulation was used when high quality tibial nerve responses could
9
not be elicited. Tibial nerve stimulation was performed at the medial malleolus of
10
the ankle with sub dermal needle electrode pairs with a proximally placed cathode
11
and an anode placed approximately 1 cm distally. The peroneal nerve was
12
stimulated using pairs of sub dermal needles located at the head of the fibula and
13
medially in the popliteal fossa. Recordings were obtained from the scalp and
14
cervical region with subdermal electrodes. Pz/Fz and P4/P3 scalp electrodes
15
montages were used (per the international 10–20 system) for recording cortical
16
potentials. A cervical electrode was localized at the C7 spinous process or at the
17
level of the mastoid and referenced to Fz was used for subcortical potentials.
18
Stimulation frequency was 2.45 Hz with duration of 0.2 milliseconds. Band pass
19
filters were set at 30 to 300 Hz for cortical recordings and 30 to 1,000 Hz for
20
cervical recordings. Averages were computed for either 128 or 256 trials depending
21
on the signal quality. IOM for the procedures in this study did not involve
AC C
EP
TE D
M AN U
SC
6
7|Page
ACCEPTED MANUSCRIPT
transcranial motor evoked potentials (TcMEPs) or stimulus triggered EMG
2
recording in response to pedicel screw stimulation.
3
Anesthesia:
4
General anesthesia with inhalation agents (1 MAC or less) and adjunct intravenous
5
agents were used for all patients. Muscle relaxants were used during induction and
6
maintained throughout the procedure.
SC
RI PT
1
M AN U
7 8
Alarm Criteria:
9
The initial recordings made after induction of anesthesia and before positioning were used as baseline potentials. Continuous SSEP signal averages were collected.
11
We considered a persistent, 50% reduction in primary somatosensory cortical
12
amplitude or a prolongation of response latency by >10% from baseline to be
13
significant. “Persistent changes” represent changes in amplitude or latency of the
14
SSEPs in more than two averaged trials. The primary reason for using more than 2
15
trials is to eliminate technical issues like noise as the reason for the change. These
16
criteria are generally agreed on in the literature as being of optimal sensitivity and
17
specificity for detecting iatrogenic spinal cord injury8-10. Henceforth, these
18
threshold signal changes will be referred to as “significant.”
AC C
EP
TE D
10
19 20
Record Review:
8|Page
ACCEPTED MANUSCRIPT
The preoperative and postoperative charts of all patients were examined and
2
compared to assess the presence of a new neurological injury. All motor, sensory,
3
bowel, or bladder symptoms of new onset were considered to be iatrogenic in
4
nature. This assessment was performed blinded to the neurophysiology data. A
5
new motor, sensory, bowel or bladder deficit is defined as a change in the
6
neurological status of the patient as noted in the chart in the progress note,
7
consultant note, and or discharge notes when compared to their preoperative
8
neurological condition. The neurophysiology data was examined in conjunction
9
with neurophysiology technologist notes to assess all cases in which a significant
M AN U
SC
RI PT
1
change in waveforms occurred. This assessment was performed blinded to the
11
medical charts documenting postoperative neurological status. For those cases in
12
which a significant change was observed, examination of the technician notes/data
13
was performed to correlate the change, if possible, with the corresponding
14
procedural cause—instrumentation, hypotension, positioning, correction of the
15
deformity—as well as the intervention performed, if any, to correct the changes in
16
signals and whether or not the signals returned. The medical charts were reviewed
17
to assess the neurological status of the patient post-operatively, upon discharge, and
18
at follow up. Transient deficits were defined as those that were observed post-
19
operatively but had resolved at discharge or subsequent visits with the duration for
20
recovery noted. Permanent deficits were defined as those that were not resolved at
21
discharge or subsequent follow up. The most recent long-term follow-up
22
appointment was also reviewed, with an average follow-up of 4.4 years.
AC C
EP
TE D
10
23 9|Page
ACCEPTED MANUSCRIPT
Data Analysis:
2
To obtain diagnostic test statistics significant changes were classified as true
3
positive (TP), false positive (FP), true negative (TN), or false negative (FN) with the
4
following criteria:
5
TP: Significant SSEP signal changes accompanied by a new postoperative neurologic
6
deficit; or, a case where significant SSEP signal deterioration occurred as the result
7
of a recognized intraoperative cause, event or complication; or, a case where a
8
significant SSEP signal deterioration improved to baseline value after a specific
9
intraoperative intervention.
M AN U
SC
RI PT
1
TN: Normal intraoperative SSEP signals in the absence of new postoperative deficits.
11
FP: persistent significant SSEP signal deterioration, which did not improve with
12
intraoperative interventions and the patient, woke up neurologically intact.
13
FN: Normal intraoperative SSEP signals with a new postoperative neurologic deficit.
14
Results:
15
Demographic information:
16
Age:
17
The average age of the patients overall was 14 years old, with a range from 1-28
18
years old. 41 patients were between the age of 1-10 years old, 404 patients were
19
between the ages of 11-18 years old, 31 patients were between the ages of 19-25
20
years old, and one patient was 28 years old.
AC C
EP
TE D
10
10 | P a g e
ACCEPTED MANUSCRIPT
Among those with significant IONM changes, the average age was 14.5 years old. Of
2
patients with significant IONM changes, 19 were between the ages of 11-18 and one
3
was between the ages of 19-25 years old.
4
Of the 3 patients with new post-operative deficits the average age was 18 years old
5
with ages of 16, 22, and 16.
6
Sex:
7
There were 352 females and 125 males in this series, for 74% and 26% of the series
8
respectively.
9
Among the 20 patients with significant IONM changes, 14 were female and 6 were
M AN U
SC
RI PT
1
males, 70% and 30% respectively.
11
Among the 3 patients with postoperative deficits 1 was male and 2 were female,
12
33% and 66% respectively.
13
Operative Date:
14
Overall 120 cases were performed between 6/14/99-7/8/03 (group 1), 120 cases
15
between 7/10/03-3/17/05 (group 2), 119 cases between 3/17/05-7/10/07 (group
16
3) and 118 cases between 7/12/07-3/12/09 (group 4).
EP
AC C
17
TE D
10
18
Among the 20 cases with significant changes, 9 correspond to group one, 3 to group
19
2, 6 to group 3, and 2 to group 4.
20
11 | P a g e
ACCEPTED MANUSCRIPT
1
Of the 3 patients with post-operative deficits, 2 correspond to group 1, 0 to group 2,
2
1 to group 3, and 0 to group 4.
RI PT
3 Post operative deficits and SSEP changes (table 1):
5
The incidence of significant SSEP change was 20/477 (4.2%). The incidence of new
6
postoperative deficits was 3/477 (0.63%) (table 1). The incidence of new
7
postoperative deficit in patients who had changes in SSEPs was 2/477 (0.42%).
8
Although patient 3 did have changes in SSEPs, they were not clearly correlated with
9
the deficit. This resulted in 1/477 (0.21%) of patients with a post-operative
10
neurological deficit without changes in SSEPs. The incidence of serious and
11
permanent neurological injury was 0/477 (0.00%).
12
True Positives:
13
19/477 cases were determined to be true positives as they were observed to have
14
significant changes in intraoperative SSEP monitoring associated with a clear
15
surgical cause. This allowed an intervention to be made that resulted in the return
16
of the signals to baseline or resulted in a deficit. As the discussion above indicates,
17
the lack of correlation between the deficit and SSEP changes in patient 3 compels us
18
to classify this patient as both a false positive and false negative rather than a true
19
positive.
20
Clinical context of SSEP changes (table 3):
AC C
EP
TE D
M AN U
SC
4
12 | P a g e
ACCEPTED MANUSCRIPT
As outlined in table 3 below 10/19 of these changes were related to instrumentation,
2
5/19 were related to hypotensive episodes, 3/19 were related to poor positioning,
3
and 1/19 was related to a tight blood pressure cuff.
4
Rod placement (n=5): Patients 2, 4, 7, 8, 9 had SSEP changes associated with rod
5
placement and subsequent derotation. In patients 2,4,7,8, and 9 adjustment of the
6
rods led to return of the SSEP responses and were not associated with a
7
postoperative deficit.
8
Wire Tightening (n=3): Patients 10, 14*, and 16 had SSEP changes associated with
9
tightening of sub laminar wires. In patients 10 and 16 loosening of the wires was
M AN U
SC
RI PT
1
associated with return of the SSEP responses and no postoperative deficit. In
11
patient 14 the SSEP responses improved to a degree but remained poor relative to
12
baseline for an extended period of time before returning to baseline at the end of the
13
procedure. The patient subsequently woke up with left sided weakness of the upper
14
and lower extremities.
15
Hook Placement (n=1): Patient 18 had deterioration of SSEP responses associated
16
with placement of hooks. Removal and replacement resulted in the responses
17
returning to baseline. No new postoperative deficit was observed.
18
Removal of old instrumentation (n=1): Patient 6 was undergoing a PSF revision and
19
had a deterioration of SSEP signals with removal of his old instrumentation.
20
However, the signal soon returned to baseline and no deficit was observed.
AC C
EP
TE D
10
13 | P a g e
ACCEPTED MANUSCRIPT
Poor Patient Positioning (n=3): Patients 17*, 19, and 20 had SSEP changes that were
2
correlated with poor positioning. 19 and 20 had their SSEP signals return fully to
3
baseline with adjustment of positioning and did not have postoperative deficits.
4
While patient 17 had some improvement of SSEP signals with adjustment of
5
positioning, they remained poor relative to his preoperative baseline throughout the
6
procedure.
7
Hypotension (n=5): Patients 1, 11, 12, 13, 15 all experienced hypotensive (less than
8
or equal to 60mmHg MAP) episodes intraoperatively that were associated with
9
changes in SSEP signals. All of these SSEP changes resolved after increasing the
M AN U
SC
RI PT
1
MAP and none were associated with a postoperative deficit.
11
Tight Pressure cuff (n=1): Patient 5 had a poor upper right extremity response that
12
was thought to be associated with a tight blood pressure cuff. Once removed the
13
signals returned to baseline and the patient did not have a deficit postoperatively.
14
Clinical context of patients with SSEP changes and deficits (table 2):
15
Patient 14 was a 23yr old woman with kyphoscoliosis who, during the course of
16
instrumentation, had complete loss of bilateral peroneal nerve SSEPs. These
17
responses improved when the instrumentation wires were loosened and returned
18
to baselines at the end of the procedure. The patient awoke with a motor deficit,
19
which at the time of discharge had improved to 2/5 strength in the left leg and 3/5
20
in the right leg. Her deficit had fully resolved, with 5/5 strength in upper and lower
21
extremities at 3 month follow up.
AC C
EP
TE D
10
14 | P a g e
ACCEPTED MANUSCRIPT
Patient 17 was 14yr old boy who had changes in the right median nerve cortical and
2
subcortical responses. This was believed to be secondary to positioning of the
3
patient. He had right upper extremity weakenss (3/5)in the wrist flexor, extensor,
4
finger flexor, extensor muscles. He improved to a 5/5 power in the right upper
5
extremity before discharge.
6
False Positive and False negative:
7
Patient 3 was a 17yr old girl who showed loss of her left tibial nerve SSEPs during
8
rod placement. A subsequent test of peroneal nerve SSEP indicated a presence of a
9
good cortical response. A wake up test was performed intraoperatively and the
M AN U
SC
RI PT
1
patient moved all four extremities. The tibial response recovered and was at
11
baseline values at closure. The patient however woke up with a post-operative
12
deficit that was observed to be in the distribution of the musculocutaneous nerve.
13
True Negatives:
14
We had a total of 457/477 (95.8 %) patients who had no changes in intraoperative
15
monitoring and no neurological deficit.
16
Statistical analysis:
17
All the 477 patients were classified into one of the four categories (TP, FP, TN, FN)
18
as described in the data analysis and a (2x2) table was thus created. The sensitivity,
19
specificity, positive predictive value and negative predictive value then calculated
20
accordingly (table 4).
AC C
EP
TE D
10
15 | P a g e
ACCEPTED MANUSCRIPT
This results in the following test statistics for SSEP monitoring: sensitivity=
2
19/20=95.0%, specificity= 457/458=99.8%, PPV= 19/20=95%,
3
NPV=457/458=99.8%.
4
True positive rate 19/477= 3.98%; False positive rate 1/477= 0.21%; True negative
5
rate 457/477= 95.8%; False negative rate 1/477= 0.21%.
6
Discussion
7
The results of our instrumented scoliosis procedures monitored with SSEP alone
8
reveal a total injury rate of 0.63% and a permanent deficit rate of 0.00%. SSEP
9
monitoring alone was observed to have 95.0% sensitivity and 99.8% specificity.
10
Our results for iatrogenic injury are slightly lower than, but consistent with, the
11
most recent rates of total injury and persistent/serious injury rate for idiopathic
12
scoliosis procedures in the literature1-3. Of the deficits that occurred during this
13
series all resolved within days to weeks. Significant changes in SSEP responses
14
were observed in 20 cases (4.2%), with 19/20 (95.0%) being true positives. This is
15
similar to other reports in the literature2,3_ENREF_2. As table 3 outlines, 10/20
16
changes were due to instrumentation, 5/20 were the result of hypotension during
17
surgery, 3/20 were the result of poor patient positioning, and one was the result of a
18
tight blood pressure cuff with distal extremity ischemia.
19
The majority of SSEP changes in this series were related to insertion, adjustment, or
20
removal of the instrumentation used during correction of the scoliotic deformity.
21
Placement of instrumentation involves corrective rods, wires, screws, and hooks, as
22
well as manual tightening of these instruments to achieve some correction of the
AC C
EP
TE D
M AN U
SC
RI PT
1
16 | P a g e
ACCEPTED MANUSCRIPT
spinal column and to prevent further deformation1,2. Placement of instrumentation
2
and subsequent derotation can result in impingement on the vasculature of the
3
spinal cord or direct compression of the neural structures, resulting in ischemia1,2.
4
Animal studies have confirmed that a significant loss of SSEP responses results from
5
“electrical failure” of the neural structures. This change precedes “ion pump failure”
6
and ensuing infarction. Timely intervention during the critical window between
7
electrical and ion pump failure can prevent permanent neurological deficits11. In
8
9/10 of the cases involving changes related to instrumentation, immediate
9
interventions resulted in return of the SSEP signals to baseline with no
M AN U
SC
RI PT
1
postoperative deficit in our study. However, in 1 case the patient’s responses were
11
slow to return and the patient experienced a transient deficit that resolved in
12
approximately 1 month. Collectively this data would suggest that SSEP monitoring
13
aided in avoiding several postoperative deficits and may have helped in reducing
14
the severity of the deficits that did occur, as has been suggested in prior studies12.
15
Low MAP (
S1529-9430(13)01575-1
DOI:
10.1016/j.spinee.2013.09.035
Reference:
SPINEE 55580
To appear in:
The Spine Journal
Received Date: 26 October 2012 Revised Date:
16 August 2013
Accepted Date: 19 September 2013
Please cite this article as: Thirumala P, Bodily L, Tint D, Ward WT, Deeney VF, Crammond D, Habeych ME, Blazer J, Somatosensory evoked potential monitoring during instrumented scoliosis corrective procedures: validity revisited, The Spine Journal (2013), doi: 10.1016/j.spinee.2013.09.035. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Somatosensory evoked potential monitoring during instrumented scoliosis corrective procedures: validity revisited.
RI PT
Parthasarathy Thirumala, MD Department of Neurological Surgery UPMC Presbyterian Suite B-400 200 Lothrop Street Pittsburgh, PA 15213 Phone: (412) 648-2570 [email protected]
W. Timothy Ward, MD UPMC, Orthopedic Surgery Pittsburgh, PA
Donald Crammond, PhD UPMC, Neurosurgey Pittsburgh, PA
EP
Miguel Habeych, MD, MPH UPMC, Neurosurgey Pittsburgh, PA
TE D
Vincent F. Deeney, MD UPMC, Orthopedic Surgery Pittsburgh, PA
M AN U
Derrick Tint, BS University of Pittsburgh School of Medicine Pittsburgh, PA
SC
Lance Bodily, BS University of Pittsburgh School of Medicine Pittsburgh, PA
AC C
Jeffrey Blazer, PhD UPMC, Neurosurgey Pittsburgh, PA
Key Words: Scoliosis; Somatosensory evoked potential; Transcranial motor evoked potential Word Count: 4,355
ACCEPTED MANUSCRIPT
Somatosensory evoked potential monitoring during instrumented scoliosis
2
corrective procedures: Utility revisited.
3
Abstract:
4
Background: Intraoperative monitoring (IOM) using somatosensory evoked
5
potentials (SSEP's) plays an important role in reducing iatrogenic neurological
6
deficits during corrective pediatric idiopathic scoliosis procedures. However, for
7
unknown reasons recent reports have cited that the sensitivity of SSEPs to detect
8
neurological deficits has decreased, in some to be less than 50%. This current trend,
9
which is coincident with the addition of transcranial electrical motor evoked
10
potentials (tceMEP's), is surprising given that SSEPs are robust, reproducible
11
responses that were previously shown to have sensitivity and specificity of >90%.
12
Purpose: Our primary aim was to assess whether SSEPs alone can detect impending
13
neurological deficits with similar sensitivity and specificity as originally reported.
14
Our secondary aim was to estimate the potential predictive value of adding
15
tceMEP's to SSEP's monitoring in idiopathic scoliosis procedures.
16
Design: This was a retrospective review to analyze the efficacy of SSEP monitoring
17
in the group of pediatric instrumented scoliosis cases.
18
Patient sample: We retrospectively reviewed all consecutive cases of patients who
19
underwent idiopathic scoliosis surgery between 1999 and 2009 at Children's
20
Hospital of Pittsburgh. We identified 477 patients who had the surgery with SSEP
21
monitoring alone. Exclusion criteria included any patients with neuromuscular
AC C
EP
TE D
M AN U
SC
RI PT
1
1|Page
ACCEPTED MANUSCRIPT
disorders or unreliable SSEP monitoring. Patients who had incomplete
2
neurophysiology data or incomplete postoperative records were also excluded.
3
Outcome measures:Major outcomes measured were clinically significant
4
postoperative sensory or motor deficits, as well as significant intraoperative SSEP
5
changes.
6
Methods: Continuous interleaved upper and lower extremity SSEP's were obtained
7
throughout the duration of all procedures. We considered a persistent 50%
8
reduction in primary somatosensory cortical amplitude or a prolongation of
9
response latency by >10% from baseline to be significant. Persistent changes
10
represent significant deviation in SSEP amplitude or latency in more than two
11
consecutive averaged trials. Patients were classified into one of four categories with
12
respect to SSEP monitoring: true positive, false positive, true negative, and false
13
negative. The sensitivity, specificity, positive predictive value (PPV) and negative
14
predictive value (NPV) were then calculated accordingly.
15
Results: Our review of 477 idiopathic scoliosis surgeries monitored using SSEP's
16
alone revealed a new deficit rate of 0.63% with no cases of permanent injury.
17
Sensitivity=95.0%, specificity=99.8%, PPV=95%, NPV=99.8%. Using evidence-
18
based epidemiological measures, we calculated that the number needed to treat
19
(NNT) was 1,587 patients for one intervention to be performed that would have
20
been missed by SSEP monitoring alone. Additionally, the number needed to harm,
21
which represents the increase in false positives with the addition of tceMEP's, was
22
200.
AC C
EP
TE D
M AN U
SC
RI PT
1
2|Page
ACCEPTED MANUSCRIPT
Conclusion: SSEP monitoring alone during idiopathic scoliosis continues to be a
2
highly reliable method for the detection and prevention of iatrogenic injury. Our
3
results confirm the high sensitivity and specificity of SSEP monitoring alone
4
published in earlier literature. As such, we suggest the continued use of SSEP alone
5
in idiopathic scoliosis surgeries. At this time we do not believe there is sufficient
6
data to support the addition of MEP monitoring, although further studies and
7
revised criteria for the use of MEP may provide added value for its use in the future.
SC
RI PT
1
M AN U
8 9 10
TE D
11 12
15 16 17
AC C
14
EP
13
18 19 3|Page
ACCEPTED MANUSCRIPT
1
3
Intro: For pediatric patients with severe scoliosis, instrumented spinal fusions are
RI PT
2
essential to halt progression and ameliorate symptoms. Iatrogenic injury during
5
scoliosis procedures can occur, most often related to derotation of the spinal column
6
causing direct injury to the spinal cord or impinging on the vasculature1. While the
7
incidence of iatrogenic spinal cord injury resulting in paraplegia is low, this is a
8
feared complication as the patients are generally young and otherwise healthy.
9
Intraoperative monitoring (IOM) using somatosensory evoked potentials (SSEP)
10
plays an important role in reducing iatrogenic neurological deficits during these
11
corrective scoliosis procedures. A large multi center study reported a reduction of
12
persistent and major persistent neurological deficits to decrease from 0.72%
13
without monitoring to 0.55% with SSEP monitoring2. The most recent reports cite
14
total neurological deficits from scoliosis surgery, the sum of transient and persistent
15
deficits, to be 0.99% and 1.84% for pediatric and adult patients respectively. Lower
16
incidence rates were reported in idiopathic cases having rates of 0.73% and 1.45%
17
for pediatric and adult patients respectively3.
18
Multiple research studies have reported that addition of transcranial motor evoked
19
potential (TcMEP) monitoring to SSEPs increases the sensitivity of IOM significantly,
20
during scoliosis surgery. Many of these studies conclude that TcMEP is a better
21
assay of the motor tracts, and suggest that it has greater efficacy in preventing
22
motor deficits4,5. Moreover, for reasons that are unclear, the reported sensitivity of
AC C
EP
TE D
M AN U
SC
4
4|Page
ACCEPTED MANUSCRIPT
SSEPs to detect a neurological deficit has decreased from close to 100% in large
2
studies 2,6 to less than 50%4,5,7 in more recent studies over the last decade. This new
3
development is surprising given that SSEPs provide robust responses that were
4
previously shown to be highly sensitive and specific1,2_ENREF_1. In addition,
5
TcMEPs lack well-defined and accepted alarm criteria to alert the surgeon about an
6
impending neurological deficit. Our primary aim is to assess whether SSEP alone
7
can detect impending neurological deficits with similar sensitivity and specificity as
8
previously reported. We also try to determine the potential need and value of
9
additional monitoring with TcMEPs, considering the current alarm criteria being
M AN U
SC
RI PT
1
used by the field.
11
Materials and Methods
12
Study Design
13
We retrospectively reviewed consecutive cases of patients who underwent scoliosis
14
surgery for idiopathic and congenital scoliosis between 1999 and 2009. We
15
identified 477 patients who had the surgery with SSEP monitoring. These
16
procedures were performed by the same surgeons (WTW, VD, JSG, SAM) at UPMC
17
Children’s Hospital.
18
Inclusion criteria included all patients with idiopathic and congenital scoliosis
19
undergoing instrumented spinal fusions for scoliosis correction. Exclusion criteria
20
included any patients with neuromuscular disorders (neurogenic scoliosis) or
21
abnormal baseline SSEP signals that did not allow monitoring. Patients who had
AC C
EP
TE D
10
5|Page
ACCEPTED MANUSCRIPT
incomplete neurophysiology data or incomplete postoperative records were also
2
excluded.
3
Neurophysiologic Monitoring:
4
Intraoperative SSEP monitoring is standard at our institution for all instrumented
5
spinal fusions of scoliosis patients based on its documented efficacy1. Baseline SSEP
6
values were obtained after the induction of anesthesia but before patient
7
positioning in all cases. Continuous upper and lower extremity interleaved SSEPs
8
were obtained throughout the procedure. Physician oversight and interpretation
9
was performed using a combined on-site and remote model implemented at UPMC.
M AN U
SC
RI PT
1
In all cases, a board-certified Neurophysiologist (American Board of
11
Neurophysiological Monitoring) was on-site, in person available for interpretation
12
while physician (neurologists) oversight, supervision, and interpretation were
13
performed in-person on-site or remotely. The oversight physician provided
14
supervision to 4-5 cases simultaneously on average with a maximum of 8 cases.
EP
15
TE D
10
Upper Extremity SSEPs:
17
Median or ulnar nerve stimulation was performed bilaterally in an alternating
18
fashion at the wrist with sub dermal needle electrode pairs. P4/Fz and P3/Fz scalp
19
electrodes montages were used (per the international 10–20 system) for recording
20
cortically generated responses. A cervical electrode was localized at the C7 spinous
21
process or mastoid (M) and referenced to Fz was used for subcortical recordings.
AC C
16
6|Page
ACCEPTED MANUSCRIPT
Stimulation frequency was 2.45 Hz with duration of 0.2 milliseconds. Band pass
2
filters were set at 10 to 300 Hz with a gain of 20k for cortical recordings and 30 to
3
1,000 Hz for subcortical recordings with a gain of 50K. Averages were computed for
4
either 128 or 256 trials, depending on the signal to noise ratio.
5
RI PT
1
Lower Extremity SSEPs.
7
For the lower extremities, bilateral alternating tibial nerve stimulation was used.
8
Peroneal nerve stimulation was used when high quality tibial nerve responses could
9
not be elicited. Tibial nerve stimulation was performed at the medial malleolus of
10
the ankle with sub dermal needle electrode pairs with a proximally placed cathode
11
and an anode placed approximately 1 cm distally. The peroneal nerve was
12
stimulated using pairs of sub dermal needles located at the head of the fibula and
13
medially in the popliteal fossa. Recordings were obtained from the scalp and
14
cervical region with subdermal electrodes. Pz/Fz and P4/P3 scalp electrodes
15
montages were used (per the international 10–20 system) for recording cortical
16
potentials. A cervical electrode was localized at the C7 spinous process or at the
17
level of the mastoid and referenced to Fz was used for subcortical potentials.
18
Stimulation frequency was 2.45 Hz with duration of 0.2 milliseconds. Band pass
19
filters were set at 30 to 300 Hz for cortical recordings and 30 to 1,000 Hz for
20
cervical recordings. Averages were computed for either 128 or 256 trials depending
21
on the signal quality. IOM for the procedures in this study did not involve
AC C
EP
TE D
M AN U
SC
6
7|Page
ACCEPTED MANUSCRIPT
transcranial motor evoked potentials (TcMEPs) or stimulus triggered EMG
2
recording in response to pedicel screw stimulation.
3
Anesthesia:
4
General anesthesia with inhalation agents (1 MAC or less) and adjunct intravenous
5
agents were used for all patients. Muscle relaxants were used during induction and
6
maintained throughout the procedure.
SC
RI PT
1
M AN U
7 8
Alarm Criteria:
9
The initial recordings made after induction of anesthesia and before positioning were used as baseline potentials. Continuous SSEP signal averages were collected.
11
We considered a persistent, 50% reduction in primary somatosensory cortical
12
amplitude or a prolongation of response latency by >10% from baseline to be
13
significant. “Persistent changes” represent changes in amplitude or latency of the
14
SSEPs in more than two averaged trials. The primary reason for using more than 2
15
trials is to eliminate technical issues like noise as the reason for the change. These
16
criteria are generally agreed on in the literature as being of optimal sensitivity and
17
specificity for detecting iatrogenic spinal cord injury8-10. Henceforth, these
18
threshold signal changes will be referred to as “significant.”
AC C
EP
TE D
10
19 20
Record Review:
8|Page
ACCEPTED MANUSCRIPT
The preoperative and postoperative charts of all patients were examined and
2
compared to assess the presence of a new neurological injury. All motor, sensory,
3
bowel, or bladder symptoms of new onset were considered to be iatrogenic in
4
nature. This assessment was performed blinded to the neurophysiology data. A
5
new motor, sensory, bowel or bladder deficit is defined as a change in the
6
neurological status of the patient as noted in the chart in the progress note,
7
consultant note, and or discharge notes when compared to their preoperative
8
neurological condition. The neurophysiology data was examined in conjunction
9
with neurophysiology technologist notes to assess all cases in which a significant
M AN U
SC
RI PT
1
change in waveforms occurred. This assessment was performed blinded to the
11
medical charts documenting postoperative neurological status. For those cases in
12
which a significant change was observed, examination of the technician notes/data
13
was performed to correlate the change, if possible, with the corresponding
14
procedural cause—instrumentation, hypotension, positioning, correction of the
15
deformity—as well as the intervention performed, if any, to correct the changes in
16
signals and whether or not the signals returned. The medical charts were reviewed
17
to assess the neurological status of the patient post-operatively, upon discharge, and
18
at follow up. Transient deficits were defined as those that were observed post-
19
operatively but had resolved at discharge or subsequent visits with the duration for
20
recovery noted. Permanent deficits were defined as those that were not resolved at
21
discharge or subsequent follow up. The most recent long-term follow-up
22
appointment was also reviewed, with an average follow-up of 4.4 years.
AC C
EP
TE D
10
23 9|Page
ACCEPTED MANUSCRIPT
Data Analysis:
2
To obtain diagnostic test statistics significant changes were classified as true
3
positive (TP), false positive (FP), true negative (TN), or false negative (FN) with the
4
following criteria:
5
TP: Significant SSEP signal changes accompanied by a new postoperative neurologic
6
deficit; or, a case where significant SSEP signal deterioration occurred as the result
7
of a recognized intraoperative cause, event or complication; or, a case where a
8
significant SSEP signal deterioration improved to baseline value after a specific
9
intraoperative intervention.
M AN U
SC
RI PT
1
TN: Normal intraoperative SSEP signals in the absence of new postoperative deficits.
11
FP: persistent significant SSEP signal deterioration, which did not improve with
12
intraoperative interventions and the patient, woke up neurologically intact.
13
FN: Normal intraoperative SSEP signals with a new postoperative neurologic deficit.
14
Results:
15
Demographic information:
16
Age:
17
The average age of the patients overall was 14 years old, with a range from 1-28
18
years old. 41 patients were between the age of 1-10 years old, 404 patients were
19
between the ages of 11-18 years old, 31 patients were between the ages of 19-25
20
years old, and one patient was 28 years old.
AC C
EP
TE D
10
10 | P a g e
ACCEPTED MANUSCRIPT
Among those with significant IONM changes, the average age was 14.5 years old. Of
2
patients with significant IONM changes, 19 were between the ages of 11-18 and one
3
was between the ages of 19-25 years old.
4
Of the 3 patients with new post-operative deficits the average age was 18 years old
5
with ages of 16, 22, and 16.
6
Sex:
7
There were 352 females and 125 males in this series, for 74% and 26% of the series
8
respectively.
9
Among the 20 patients with significant IONM changes, 14 were female and 6 were
M AN U
SC
RI PT
1
males, 70% and 30% respectively.
11
Among the 3 patients with postoperative deficits 1 was male and 2 were female,
12
33% and 66% respectively.
13
Operative Date:
14
Overall 120 cases were performed between 6/14/99-7/8/03 (group 1), 120 cases
15
between 7/10/03-3/17/05 (group 2), 119 cases between 3/17/05-7/10/07 (group
16
3) and 118 cases between 7/12/07-3/12/09 (group 4).
EP
AC C
17
TE D
10
18
Among the 20 cases with significant changes, 9 correspond to group one, 3 to group
19
2, 6 to group 3, and 2 to group 4.
20
11 | P a g e
ACCEPTED MANUSCRIPT
1
Of the 3 patients with post-operative deficits, 2 correspond to group 1, 0 to group 2,
2
1 to group 3, and 0 to group 4.
RI PT
3 Post operative deficits and SSEP changes (table 1):
5
The incidence of significant SSEP change was 20/477 (4.2%). The incidence of new
6
postoperative deficits was 3/477 (0.63%) (table 1). The incidence of new
7
postoperative deficit in patients who had changes in SSEPs was 2/477 (0.42%).
8
Although patient 3 did have changes in SSEPs, they were not clearly correlated with
9
the deficit. This resulted in 1/477 (0.21%) of patients with a post-operative
10
neurological deficit without changes in SSEPs. The incidence of serious and
11
permanent neurological injury was 0/477 (0.00%).
12
True Positives:
13
19/477 cases were determined to be true positives as they were observed to have
14
significant changes in intraoperative SSEP monitoring associated with a clear
15
surgical cause. This allowed an intervention to be made that resulted in the return
16
of the signals to baseline or resulted in a deficit. As the discussion above indicates,
17
the lack of correlation between the deficit and SSEP changes in patient 3 compels us
18
to classify this patient as both a false positive and false negative rather than a true
19
positive.
20
Clinical context of SSEP changes (table 3):
AC C
EP
TE D
M AN U
SC
4
12 | P a g e
ACCEPTED MANUSCRIPT
As outlined in table 3 below 10/19 of these changes were related to instrumentation,
2
5/19 were related to hypotensive episodes, 3/19 were related to poor positioning,
3
and 1/19 was related to a tight blood pressure cuff.
4
Rod placement (n=5): Patients 2, 4, 7, 8, 9 had SSEP changes associated with rod
5
placement and subsequent derotation. In patients 2,4,7,8, and 9 adjustment of the
6
rods led to return of the SSEP responses and were not associated with a
7
postoperative deficit.
8
Wire Tightening (n=3): Patients 10, 14*, and 16 had SSEP changes associated with
9
tightening of sub laminar wires. In patients 10 and 16 loosening of the wires was
M AN U
SC
RI PT
1
associated with return of the SSEP responses and no postoperative deficit. In
11
patient 14 the SSEP responses improved to a degree but remained poor relative to
12
baseline for an extended period of time before returning to baseline at the end of the
13
procedure. The patient subsequently woke up with left sided weakness of the upper
14
and lower extremities.
15
Hook Placement (n=1): Patient 18 had deterioration of SSEP responses associated
16
with placement of hooks. Removal and replacement resulted in the responses
17
returning to baseline. No new postoperative deficit was observed.
18
Removal of old instrumentation (n=1): Patient 6 was undergoing a PSF revision and
19
had a deterioration of SSEP signals with removal of his old instrumentation.
20
However, the signal soon returned to baseline and no deficit was observed.
AC C
EP
TE D
10
13 | P a g e
ACCEPTED MANUSCRIPT
Poor Patient Positioning (n=3): Patients 17*, 19, and 20 had SSEP changes that were
2
correlated with poor positioning. 19 and 20 had their SSEP signals return fully to
3
baseline with adjustment of positioning and did not have postoperative deficits.
4
While patient 17 had some improvement of SSEP signals with adjustment of
5
positioning, they remained poor relative to his preoperative baseline throughout the
6
procedure.
7
Hypotension (n=5): Patients 1, 11, 12, 13, 15 all experienced hypotensive (less than
8
or equal to 60mmHg MAP) episodes intraoperatively that were associated with
9
changes in SSEP signals. All of these SSEP changes resolved after increasing the
M AN U
SC
RI PT
1
MAP and none were associated with a postoperative deficit.
11
Tight Pressure cuff (n=1): Patient 5 had a poor upper right extremity response that
12
was thought to be associated with a tight blood pressure cuff. Once removed the
13
signals returned to baseline and the patient did not have a deficit postoperatively.
14
Clinical context of patients with SSEP changes and deficits (table 2):
15
Patient 14 was a 23yr old woman with kyphoscoliosis who, during the course of
16
instrumentation, had complete loss of bilateral peroneal nerve SSEPs. These
17
responses improved when the instrumentation wires were loosened and returned
18
to baselines at the end of the procedure. The patient awoke with a motor deficit,
19
which at the time of discharge had improved to 2/5 strength in the left leg and 3/5
20
in the right leg. Her deficit had fully resolved, with 5/5 strength in upper and lower
21
extremities at 3 month follow up.
AC C
EP
TE D
10
14 | P a g e
ACCEPTED MANUSCRIPT
Patient 17 was 14yr old boy who had changes in the right median nerve cortical and
2
subcortical responses. This was believed to be secondary to positioning of the
3
patient. He had right upper extremity weakenss (3/5)in the wrist flexor, extensor,
4
finger flexor, extensor muscles. He improved to a 5/5 power in the right upper
5
extremity before discharge.
6
False Positive and False negative:
7
Patient 3 was a 17yr old girl who showed loss of her left tibial nerve SSEPs during
8
rod placement. A subsequent test of peroneal nerve SSEP indicated a presence of a
9
good cortical response. A wake up test was performed intraoperatively and the
M AN U
SC
RI PT
1
patient moved all four extremities. The tibial response recovered and was at
11
baseline values at closure. The patient however woke up with a post-operative
12
deficit that was observed to be in the distribution of the musculocutaneous nerve.
13
True Negatives:
14
We had a total of 457/477 (95.8 %) patients who had no changes in intraoperative
15
monitoring and no neurological deficit.
16
Statistical analysis:
17
All the 477 patients were classified into one of the four categories (TP, FP, TN, FN)
18
as described in the data analysis and a (2x2) table was thus created. The sensitivity,
19
specificity, positive predictive value and negative predictive value then calculated
20
accordingly (table 4).
AC C
EP
TE D
10
15 | P a g e
ACCEPTED MANUSCRIPT
This results in the following test statistics for SSEP monitoring: sensitivity=
2
19/20=95.0%, specificity= 457/458=99.8%, PPV= 19/20=95%,
3
NPV=457/458=99.8%.
4
True positive rate 19/477= 3.98%; False positive rate 1/477= 0.21%; True negative
5
rate 457/477= 95.8%; False negative rate 1/477= 0.21%.
6
Discussion
7
The results of our instrumented scoliosis procedures monitored with SSEP alone
8
reveal a total injury rate of 0.63% and a permanent deficit rate of 0.00%. SSEP
9
monitoring alone was observed to have 95.0% sensitivity and 99.8% specificity.
10
Our results for iatrogenic injury are slightly lower than, but consistent with, the
11
most recent rates of total injury and persistent/serious injury rate for idiopathic
12
scoliosis procedures in the literature1-3. Of the deficits that occurred during this
13
series all resolved within days to weeks. Significant changes in SSEP responses
14
were observed in 20 cases (4.2%), with 19/20 (95.0%) being true positives. This is
15
similar to other reports in the literature2,3_ENREF_2. As table 3 outlines, 10/20
16
changes were due to instrumentation, 5/20 were the result of hypotension during
17
surgery, 3/20 were the result of poor patient positioning, and one was the result of a
18
tight blood pressure cuff with distal extremity ischemia.
19
The majority of SSEP changes in this series were related to insertion, adjustment, or
20
removal of the instrumentation used during correction of the scoliotic deformity.
21
Placement of instrumentation involves corrective rods, wires, screws, and hooks, as
22
well as manual tightening of these instruments to achieve some correction of the
AC C
EP
TE D
M AN U
SC
RI PT
1
16 | P a g e
ACCEPTED MANUSCRIPT
spinal column and to prevent further deformation1,2. Placement of instrumentation
2
and subsequent derotation can result in impingement on the vasculature of the
3
spinal cord or direct compression of the neural structures, resulting in ischemia1,2.
4
Animal studies have confirmed that a significant loss of SSEP responses results from
5
“electrical failure” of the neural structures. This change precedes “ion pump failure”
6
and ensuing infarction. Timely intervention during the critical window between
7
electrical and ion pump failure can prevent permanent neurological deficits11. In
8
9/10 of the cases involving changes related to instrumentation, immediate
9
interventions resulted in return of the SSEP signals to baseline with no
M AN U
SC
RI PT
1
postoperative deficit in our study. However, in 1 case the patient’s responses were
11
slow to return and the patient experienced a transient deficit that resolved in
12
approximately 1 month. Collectively this data would suggest that SSEP monitoring
13
aided in avoiding several postoperative deficits and may have helped in reducing
14
the severity of the deficits that did occur, as has been suggested in prior studies12.
15
Low MAP (
