International Journal of Academic Medicine

: 2016  |  Volume : 2  |  Issue : 2  |  Page : 132--144

Neuromonitoring protocol for spinal cord stimulator cases with case descriptions

Steven Falowski, Andres Dianna 
 Department of Neurosurgery, St. Luke's University Health Network, Bethlehem, PA 18015, USA

Correspondence Address:
Steven Falowski
Department of Neurosurgery, St. Luke's University Health Network, 801 Ostrum Street, Suite 302, Bethlehem, PA 18015


Spinal cord stimulation (SCS) relies on the ability to create an overlap of paresthesia on the painful regions. Electrode implantation has historically been performed with awake intraoperative testing to allow the patient to report on the device-induced paresthesia. More recently, the use of neuromonitoring has come into favor and can be used for SCS placement, while the patient remains fully anesthetized throughout the surgery. This is a critical evaluation of the neuromonitoring technique and protocol with an in-depth description of neuromonitoring for SCS placement using electro-myography (EMG) responses in both cervical and thoracic electrode placement. There is an explanation for the interpretation of the EMG responses, as well as case reports of two patients. Neuromonitoring is used to determine myotomal coverage, as a marker that corresponds with dermatomal coverage. This article demonstrates some of the critical steps for both the surgeon and neuromonitoring group to implement this technique, as well as the clinical results of paresthesia coverage in patients. This protocol can be utilized in implementing neuromonitoring into a practice for those implanting SCS systems. The following core competencies are addressed in this article: Medical knowledge, patient care, practice-based learning and improvement, system-based practice, interpersonal and communication skills. This article addresses the gap in knowledge base to implement an approach to improve patient care and outcome.

How to cite this article:
Falowski S, Dianna A. Neuromonitoring protocol for spinal cord stimulator cases with case descriptions.Int J Acad Med 2016;2:132-144

How to cite this URL:
Falowski S, Dianna A. Neuromonitoring protocol for spinal cord stimulator cases with case descriptions. Int J Acad Med [serial online] 2016 [cited 2022 Aug 15 ];2:132-144
Available from:

Full Text


Spinal cord stimulation (SCS) is an adjustable, nondestructive, neuromodulation procedure utilizing electrodes on the spinal cord for the management of neuropathic pain. The most common indications worldwide include postlaminectomy syndrome or failed back surgery syndrome, complex regional pain syndrome (CRPS), ischemic limb pain, and angina. Indications have been extended to include the treatment of intractable pain due to other causes including visceral/abdominal pain, cervical neuritis pain, spinal cord injury pain, postherpetic neuralgia, and neurogenic thoracic outlet syndrome. The procedures are most commonly performed by neurosurgeons or anesthesiologists specializing in pain management [1],[2],[3] but other specialties, such as rehabilitation medicine and orthopedic surgery, have also demonstrated interest in the procedure.

The success of the therapy relies on the ability to create an overlap of the painful regions with paresthesia.[4] Extensive work has been done previously in describing the mapping of the spinal structure and the relationship between the spinal level of stimulation and the somatotopy of paresthesia.[5] Electrode implantation can be performed with awake intraoperative testing using local anesthetic and intravenous sedation or under general anesthesia which precludes the ability for the physician to interact with the patient.

The awake operation is generally chosen to allow the patient to report on the device-induced paresthesia and also the assessment of the discomfort thresholds. It is commonly accepted that intraoperative testing in the awake patient is likely to lead to optimal placement of the electrode as well as ensure safety in case of neurologic compromise. However, more recent publications have examined the utility of neuromonitoring with the placement of the patient asleep.[6],[7],[8] The studies referenced have shown equal efficacy with possible lower complication and revision rates.

Neuromonitoring is used to determine myotomal coverage, as a marker that corresponds roughly with dermatomal coverage. This is performed while the patient remains fully anesthetized throughout the surgery, in what amounts to a specialized thoracic or cervical laminectomy procedure. Asleep placement may be coming into favor. With the exception of the few studies looking at asleep placement that give very brief explanations there have been no critical descriptions of the neuromonitoring technique and protocol, as well as a lack of description for the interpretation of the data. This can prove extremely helpful in implemented neuromonitoring into a practice. We seek to give an in depth description of neuromonitoring for SCS placement using electromyography (EMG) responses.

 Description of Neuromonitoring Protocol

Muscle coverage for EMG responses is the primary concern in utilizing neuromonitoring for the placement of SCS electrodes when done under general anesthesia. In addition, it allows for monitoring of the spinal cord in case of injury during placement of an electrode. Monitoring for safety of the cord will include somatosensory evoked potentials (SSEP) as a baseline, but may also include transcranial motor evoked potentials (TceMEP).

General anesthesia is performed with total intravenous anesthesia regimen that includes propofol and benzodiazepines. Intubation is performed and the patient is positioned in the prone position on chest rolls. Neuromuscular blocking agents are not utilized after the intubation of the patient, as this will interfere with motor signals. Sterile, 1.3 cm, 27 gauge subdermal needle electrodes are placed in pairs for bilateral coverage. Symmetrical placement of the monitoring leads is imperative given that the basis for the neurophysiologic mapping resides in response amplitude comparisons. EMG responses are determined with the addition of surgeon interpretation of the placement of the electrode, as well as fluoroscopy for confirmation. This interpretation will then be utilized to determine the physiological midline, laterality, and orientation of the electrode, and myotomal coverage as a marker for anticipated dermatomal paresthesia.

Thoracic spinal cord stimulator placement

For thoracic cord placement, the surgeon will usually perform a laminectomy at the lower thoracic levels. Therefore, in addition to lumbar and sacral nerve levels, coverage for local thoracic nerve roots is desirable. Monitoring electrodes are placed in the periumbilical rectus abdominis muscles for mid-lower thoracic electrode placement to achieve sensitivity in the T8/12 spinal nerve root distributions. Our accepted protocol incorporates the following for prone thoracic laminectomy for SCS placement [Table 1].{Table 1}

Ulnar and posterior tibial nerve (PTN) SSEP, lower extremity transcranial electrically-stimulated motor evoked potentials with compound muscle action potential response (TceMEP-CMAP), spontaneous EMG (SEMG), triggered EMG (TEMG), and train-of-four (TOF).

Muscle channels are as follows:

Bilateral upper-lower abdominals (rectus abdominis segments recorded above, representing T5–9 nerve root levels, and below, representing T10–12 nerve root levels, the umbilicus in active reference montage), iliopsoas ([ILIO], representing L1–2), adductor-quadriceps ([ADD-QUAD], representing L2, 3, 4, in active reference montage), tibialis anterior (TA, representing L5), medial gastrocnemius (MG, representing S1–2), and abductor hallucis-extensor hallucis brevis (AH-EHB, representing S1, 2, 3, in active-reference montage). In certain situations, it may be advantageous to utilize active-reference linking of the tibialis anterior and medial gastrocnemius to allow for additional lead placement in the patient's buttocks (gluteus maximus [GMax L5, S1, S2]) or posterior thigh (“Hamstrings” [Ham, L4, L5, S1, S2] (semitendinosus, semimembranosus, biceps femoris short and long head)) to better cover specific regions of a patient's pain pattern for readout of the stimulus artifact portion of the response preferentially appearing in these leads over other muscles with the same nerve root level. Using six pairs of muscles for EMG and motor recording necessitates that upper extremity motor evoked potentials and PTN SSEP peripheral (popliteal fossa) channels are eliminated when using a 16 channel system. Moreover, it will be necessary to create a separate SEMG window for the SSEP/EMG mode that contains one less pair of muscles than the SEMG window in the CMAP or EMG modes, due to channel constraints, so that readout from channels used in SSEP collection are not displayed in the window intended to view muscle activity.

We choose to display the abdominal, ILIO, ADD-QUAD, tibialis anterior, and medial gastrocnemius channels only in that mode, leaving out the AH-EHB channels, which are least relevant during the portions of the surgery where SSEP data collection is most relevant. In CMAP or EMG mode, however, the muscles take the precedence over the SSEP channels, so that all the six muscle pairs can be simultaneously displayed in the SEMG window [Figure 1] and [Figure 2].{Figure 1}{Figure 2}

A typical thoracic stimulator placement proceeds as follows:

Setup for typical lower thoracic/lumbar case with prone patient on chest rolls Anesthesia: With TceMEP: TIVA (preferred) or combination infusion + inhalant up to 0.5 MAC (No N2O), with short-acting relaxant for intubation. Without TceMEP: Inhalant up to 1.0 MAC (limited to 50% ET N2O), with short-acting relaxant for intubation Obtain initial data with baselines Track train of four to make sure you will have maximum sensitivity for muscle responses during testing of stimulator electrode Prior to electrode placement, switch to EMG test/window, with both SEMG and TEMG windows displayed in vertical split to observe left/right symmetry at each muscle/nerve level. Recommended parameters are SEMG sweep 200 ms/div and gain (sensitivity) 200 µv/div, TEMG sweep 10 ms/div and gain (sensitivity) 100 µv/div with automatic capture and store active to allow for posttesting review of responses if needed (The viewing epoch of 200 ms/div for SEMG is chosen to complement the ability to see multiple stimulus artifact and/or CMAP responses in a single window while also allowing for the resolution of some potentially higher frequency CMAP trains and/or neurotonic discharge. The 10 ms/div sweep for the TEMG window allows for more critical evaluation of any individual CMAP response and higher frequency fibrillations, fasciculation's, or neurotonic discharge) Begin testing after impedances of the electrode are measured. Determine electrode area that is being used to stimulate, such as “middle middle” or “top right” which is dependent on the type of lead placed. This can vary from single column leads to multi-column paddle electrodes. A bipole configuration with a single cathode and anode is most commonly used. Surgeon will potentially make adjustments to electrode positioning to optimize left/right symmetry and/or coverage of the patient's area(s) of pain. During testing, you will see stimulus artifact at low levels of stimulation, then stimulus artifact accompanied by time-locked CMAP responses at higher levels. Typical stimulation parameters are frequencies of 10 Hz (range from 4 to 20), 200 µs pulse width (varies from 100 to 500), and level increasing from 0 to approximately 10–12 mA. When viewed in the 200 ms/div sweep SEMG window, both the stimulation artifact and CMAP will appear as spikes; in the 10 ms/div sweep TEMG window, the stimulation artifact spike and CMAP will appear distinctly different in morphology. Comparing strong amplitude symmetry is easier in SEMG due to all spike appearance, while comparing lower amplitude CMAP symmetry is easier in TEMG, because you can evaluate the shape of the CMAP left versus right, as well as its amplitude Once testing is complete and the lead is in final position, the implantable pulse generator (IPG) is connected. The impedances and IPG/system function is tested Continue running SSEP during closing.


SSEP: Standard stimulation lead placement for ulnar (medial volar wrist with anode distal to cathode) and PTN (medial posterior ankle with anode distal to cathode) SSEP. Standard scalp recording electrode placement (10–20 system Fpz, Cz', C3', C4', Cv5, and noncephalic signal ground) and bilateral Erb's point (just superior to mid-clavicle) leads EMG: Rectus abdominis leads, active-reference upper to lower (forming square pattern around umbilicus). Bipolar in ILIO, TA, and MG; active-reference ADD-QUAD and AH-EHB TOF: Taken at left or right PTN stimulation to ipsilateral AH-EHB recording TceMEP-CMAP: Stimulation leads at C3 and C4, biased slightly medially to optimize lower extremity responses. Recording: From all EMG channels plus contralateral AH-EHB. Upper extremity motor evoked potential response leads may be placed in the patient but not plugged into the system unless needed to confirm/refute change in motor evoked potential from lower extremities linked to a surgical event.

Cervical spinal cord stimulator placement

Monitoring for cervical SCS placement can be done in a similar fashion. The main difference is the selection of muscles used for EMG and TceMEP-CMAP recordings [Table 2]. There is a higher risk of spinal cord compression in the region of the cervical enlargement, so it is recommended that TceMEP-CMAP recordings should always be obtained [Figure 3].{Table 2}{Figure 3}

Although the stimulation parameters are the same as thoracic stimulator placement, the thresholds for responses are usually lower in the cervical spine. Therefore, lower pulse widths and amplitudes should be attempted initially. Evaluation and interpretation of data is otherwise similar to thoracic stimulator placement.

Typical setup to include

SSEP: Standard stimulation lead placement for median (M) N, ulnar (U), and PTN. Standard scalp recording electrode placement (10–20 system Fpz, Cz', C3', C4', with Cv5 lead relocated to M1 or M2 due to exposure constraints, and noncephalic signal ground) and bilateral Erb's point leads SEMG/TEMG: Selection of muscles is somewhat dependent on patient's pain distribution. For general coverage of cervical levels C3–C8 and thoracic T1: Bipolar montage for middle trapezius (TRAP, nerve root levels C3–4), deltoid (Delt C5), biceps brachii (Bic C6), triceps (Tri C7), and abductor pollicis brevis (APB C8) and abductor digiti minimi manus (ADM T1) in active-reference montage. If pain pattern skips shoulders and is most prevalent in hands: Bipolar Delt (C5), Bic (C6), Tri (C7), APB (C8), and ADM (T1). In addition, it may be advantageous in certain situations to use an active-reference montage for the biceps-triceps and include leads in the patient's brachioradialis (C6), and/or flexor carpi ulnaris (C7, C8) for pain distribution primarily below the elbow TOF: Taken at left or right PTN stimulation site to ipsilateral AH-EHB recording TceMEP-CMAP: Stimulation leads at C3 and C4. Recording from all EMG channels, ipsilateral AH-EHB, plus contralateral APB-ADM (or APB), and AH-EHB.

 Physician Interpretation of Data

Neuromonitoring can be used to determine differences in anatomic midline and physiologic midline, as well as determine myotomal coverage as a marker for dermatonal coverage. The physician will make adjustments in positioning of the lead based on the feedback from neuromonitoring to optimize symmetry and/or coverage. Stimulation artifact will first be elicited at lower levels of stimulation and will then be followed by time-locked CMAP responses as the levels increase. Stimulation artifact can be a good marker for determining coverage, but should only be loosely interpreted until there are time-locked true motor responses. At times, we will titrate down to 5 Hz from the 10 Hz recommendation, so that the slower pulse rate makes it easier to see the stimulus in the muscle. Although the stimulation parameters are similar between cervical and thoracic, responses are usually elicited earlier and at lower thresholds in the cervical spine. It is best to start at lower pulse widths and amplitudes in the cervical evaluation. This is likely from the tighter canal and smaller CSF space in the cervical spine compared to thoracic spine.

Determination of physiologic midline is imperative in obtaining accurate paresthesia coverage. Stimulation artifact will initially be seen in thoracic nerve roots. Regardless of placement, it is expected that bilateral stimulation will be seen in nerve roots secondary to current spread and the ease of exciting the nerve roots. It is therefore suggested to only loosely use this data for symmetry. This same holds true for the activation of the abdominal muscles and illiopsoas muscles. Although true muscle responses can be expected, absolute symmetry and interpretation may be difficult. In the case of abdominal muscles, there must be true symmetric placement of recording electrodes, and there is also spread from nerve root activation. In the case of illiopsoas muscles, the recording electrodes cannot be placed directly into the muscle secondary to body habitus, in addition to the muscle being quite large and diffuse, which can lead to difficult interpretation of symmetry. It is the author's recommendations to use this data, but only in conjunction with motor responses from the ADD-QUAD and distal when determining midline. Finally, an additional marker of symmetry and physiologic midline is to determine the timing of firing on both sides. It may be visualized that the amplitude on both sides is symmetric, but a slight preference off midline can be determined by monitoring which side and muscle groups fired first. This equates to an exact determination of placement and precise stimulation.

Advances in technology have led to paddle electrodes with complex arrays. These arrays can span multiple vertebral levels, as well as vary in width across the canal. The complexity of contacts is also variable with present paddle electrodes having up to 32 contacts, as well as having up to five columns. It is because of this complexity that proper placement is of importance for use of an entire electrode array. However, the increased length and width of these paddle arrays has led to the added complexity of ensuring the array is lying flat and flush in the canal. Neuromonitoring can be utilized to ensure that the lead is not canted or tilted. This is done by determining the amplitudes at which responses are first generated. This should be consistent across the lead. If there is large discrepancies across or within columns, it may be a sign that the lead is tilted or canted since the distance from the contact to the spinal cord is varying.

Determination of myotomal coverage as a marker for dermatonal coverage can be used to ensure proper paresthesia coverage as well. Mapping of the spinal structure and the relationship between the spinal level of stimulation and the somatotopy of paresthesia has been previously described.[5] Anatomy and somatotopy may differ slightly among patients and therefore the lead may be moved either cranial/caudal or medial/lateral in an attempt to activate the specific fibers and generate coverage on the neuromonitoring. This use of the monitoring is most important in patients with very specific pain areas, such as those patients with CRPS in a specific area of the limb.

In some instances, there may be concern that the proposed protocol may not give adequate or reliable data for lead placement. In these rare situations, there is the option of trying SSEP collision testing in addition to the EMG protocol to look for correlation. One can also titrate down to 5 Hz from the 10 Hz recommendation for frequency as previously stated so that the slower pulse rate makes it easier to see the stimulus in the muscle. Raising the frequency may lead to increased firing, but can be at the cost of saturating the signal and losing the ability to determine symmetry. However, this is an option in circumstances where signals are weak or determined unreliable. Another approach to broaden the signal strength may include raising the pulse width for improved penetration of the fibers/cord or by activating more contacts on the lead simultaneously. Simple bipole configurations will usually work well, but one can also consider a double-guarded cathode configuration (+ - +) or transverse stimulation. Finally, one can always use the stimulation artifact as a loose estimate of coverage.

 Patient Case Series

Patient 1

This is a 51-year-old male with pain in his posterior right thigh/calf/foot greater than low back pain. His trial X-ray demonstrated single percutaneous midline lead placement at T9 with programming of the top contacts to generate his paresthesia coverage.

Intraoperative placement included a Penta Electrode (St. Jude Medical) with rechargeable Protege IPG [Figure 4]. Lead was placed at the level of inferior T8 and full T9 with decision to have a midline placement slightly favoring the right side.{Figure 4}

Intraoperative programming

First lead placement tested middle-middle (contacts 8+ 9−) that demonstrated stronger left sided signals, and therefore the lead was repositioned [Figure 5].{Figure 5}

After repositioning of the lead to move slightly right, the testing included middle-middle (contacts 8+ 9−), middle-right (12+ 13−), middle left (4+ 5−) [Figure 6].{Figure 6}

Immediate postoperative programming

Program (1) contact 8+ 9− at PW 200 µs and Rate 30 Hz. Comfort at 2.0 mA.

Generated bilateral lower back greater on right side with posterior greater than anterior thigh and calf to ankle with mild sensation in foot.

Program (2) contact 4+ 5− PW 200 µs and Rate 30 H. Comfort at 2.0 mA.

Generated bilateral lower back greater on left side with posterior greater than anterior thigh and calf to foot.

Two week postoperative programming

Single program with contact 9+ 11− at PW 150 µs and rate at 50 Hz. Comfort at 2.0 V.

Generated mainly right posterior lower extremity to calf with coverage in right lower back.


The patient had a pain pattern that included low back and right lower extremity. Bilateral coverage would need to be obtained for the treatment of his lower back pain, as well as taking into account the desire of the patient to favor right lower extremity coverage. Initial placement in the operating room demonstrated a slight left placement with greater amplitudes in the left iliopsoas and left ADD-QUAD. Secondary to this the lead was shifted slightly to the right side. At that point, testing demonstrated that middle contacts of the Penta electrode had a very slight increase in amplitude in the right ADD-QUAD compared to the left ADD-QUAD demonstrating a midline placement with very slight preference to the right side, which is ideal in this patient. Further testing included columns right and left of the midline to ensure that isolated coverage could be obtained in each lower extremity.

Immediate postoperative programming matched the intraoperative testing demonstrating the utility of using the intraoperative monitoring to also guide programming following the procedure. A second program was also given to deliver left-sided coverage secondary to the patients desire to have equal stimulation in her lower extremities following the procedure. This was consistent with the middle-left programming seen during intraoperative testing.

Subsequent programming at 2 weeks postoperative utilized a program to direct midline coverage to the right side to generate lower back coverage, as well as isolated right lower extremity coverage. This programming demonstrates the stability of the lead in the postoperative period and also confirms the intraoperative monitoring findings.

Patient 2

This is a 51-year-old female with pain in her lower back and bilateral lower extremities greater on the left side. Her trial X-ray demonstrated dual percutaneous midline leads placed at T8 with programming of the middle contacts to generate her paresthesia coverage. Intraoperative placement included a tripole electrode (medtronic) with rechargeable restore sensor IPG [Figure 7]. Lead was placed at the level of inferior T8 and T9 with decision to have a midline placement slightly favoring the left side.{Figure 7}

Intraoperative programming

Initial placement of the lead delivered the desired results with midline coverage that slightly favored the left side at the top of the lead secondary to a slight tilt off midline. The testing included middle-middle (contacts 6+ 7−), middle-right (12+ 13−), top-middle (6+ 5−) [Figure 8].{Figure 8}

Immediate postoperative programming

One program was given to the patient that included a double-guarded cathode at contact 6 + 7 − 8+ at PW 270 µs and rate 40 Hz. Comfort at 4.0 mA.

Generated bilateral back and hip to the back of both legs slightly greater left to right.

Two week postoperative programming

The same programming was maintained with identical paresthesia coverage as immediate postoperative.


The patient had a pain pattern that included low back and bilateral lower extremity with greater pain on the left side. Secondary to this midline placement of the lead would need to be obtained. Testing of the initial placement in the operating room demonstrated that middle-middle contacts of the tripole electrodes had identical amplitudes in all muscle groups with a slight left-sided preference in the illiopsoas muscles confirming a midline placement. As previously described, the iliopsoas muscle group is not a reliable marker of comparing amplitudes. However, secondary to this, as well as the intraoperative X-ray demonstrating a very slight tilt of the lead [Figure 9], the top-middle of the lead was tested confirming that there was slight left bias of the lead with Left ADD-QUAD firing prior to the right side. Further testing included the column right of the midline to ensure that isolated coverage could be obtained in the right lower extremity. The decision was made to leave the lead as initially placed.{Figure 9}

Immediate postoperative programming matched the intraoperative testing demonstrating the utility of using the intraoperative monitoring to also guide programming following the procedure. The decision was made to use a double-guarded cathode configuration to isolate the signal in the middle portion of the lead and away from the top of the lead to ensure bilateral coverage without favoring the left side. This is an example of intraoperative monitoring guiding the postoperative programming.

Subsequent programming at 2 weeks postoperative utilized the same program without change. This programming demonstrated stability of the lead in the postoperative period and also confirms the intraoperative monitoring findings.


The distribution of the stimulation-induced paresthesia is imperative to obtain efficacy for pain control. Electrode implantation can be performed either awake (local anesthetic and intravenous sedation) or asleep (under general anesthesia). An advantage to testing an awake patient yields immediate feedback regarding the stimulation-induced paresthesia. When the procedure is performed with the patient asleep, there is the use of fluoroscopic imaging with or without evoked motor responses to assure proper electrode positioning.

Patients awakening from anesthesia may be very uncooperative and disoriented, while a patient who is too sedated would not be able to answer questions during the testing. Either could result in complete failure of the procedure or an accidental neural injury. Although this is accepted as the more common implantation method, there is a trend toward implantation under general anesthesia. Although minimally invasive techniques and other methods of awake placement have advantages, implantation under general anesthesia may be desirable. Patients undergoing lead revision frequently have extensive epidural scarring, requiring multilevel laminectomies to place the electrode appropriately which can also be similar in those patients with extensive thoracic degenerative disease. An awake patient may be predisposed to movement which can lead to migration of the electrode with undesired stimulation and ultimately treatment failure. It can also lead to decreased patient satisfaction. The implanting surgeon may prefer to implant in an asleep patient to address these factors.

The use of EMG/SSEP during implantation was devised in 1998[9] with single-column leads. Advances in technology and electrode design has led to further descriptions and clinical application in placement of surgical leads via a laminotomy.[10] In 2011, a retrospective review looking at electrode implantation performed either awake or under general anesthesia was evaluated.[6] Conclusions reported that implantation performed under general anesthesia was associated with fewer failure rates and fewer re-operations. Results showed that the incidence of device failure following spinal cord implantation performed under general anesthesia using neurophysiologically-guided electrode placement was significantly lower than that following implantation in the awake patient. It was concluded that reduction in patients movements during surgery made it less likely to dislodge electrodes and their connections. There was no difference in the incidence of other major complications, including infections and poorly-placed electrodes that required repositioning.

Although with its limitations, it demonstrated that placement under general anesthesia with neurophysiologic mapping was a viable option that did not compromise therapeutic efficacy or result in additional complications and was at least as effective as patient feedback. An additional retrospective review in 2011 also concluded that the use of intraoperative electrophysiology for the placement of SCS paddle leads under general anesthesia is a safe and efficacious alternative to awake surgery.[7] There results showed that immediately postoperatively, 75 of 78 patients reported that the paresthesia coverage was as good as (or better than) that of the SCS trial. A total of 64 of the 78 implanted patients reported continued pain relief with stimulator use in long-term follow-up.

Shils and Arle in 2012[8] retrospectively analyzed 172 electrodes implanted between September 2008 and July 2011. All patients had their SCS placed under general anesthesia, and EMG was recorded from upper or lower limb muscle groups related to the placement of the stimulator electrode. Their conclusions stated that asleep placement with neurophysiologically-guided electrode placement is safe and appears to be at least as accurate and efficacious as using the awake SCS placement technique based on a 50% improvement in the visual analogue scale. Another interesting thing in the study was that based on the electromyographic recording technique, the electrodes were repositioned intraoperatively in 15.9% of patients and the stimulation used to elicit the CMAP s came from antidromic activation of the dorsal columns and not from the corticospinal tract. The use of intraoperative neuromonitoring may lead to increased repositioning's of the lead secondary to increased accuracy of generating appropriate paresthesia coverage on the optimal portion of the lead.

It is important to note that there are different protocols that could be implemented in utilizing electrophysiological monitoring such as SSEP collision testing.[11] Balzer et al. concluded that SSEP can be used safely and successfully for predicting the lateralization of cervical spinal cord stimulator placement. Roth et al. prospectively analyzed 73 patients implanted utilizing EMG and SSEP collision testing [12] to verify lead placement. Patient pain and function were assessed as well as stimulation parameters at 6 months. Statistically significant improvements were observed for both pain and function. EMG and SSEP testing appropriately lateralized leads in 89.0% and 69.0% of the patients, respectively. EMG predicted the active contacts being utilized at follow-up with 82.7% sensitivity. This demonstrates both the accuracy with placement utilizing neuromonitoring, as well as predicting ideal contacts to aid in programming.

The protocol described in this paper utilizes EMG responses which the authors feel may be more reliable and demonstrate increased specificity. It becomes imperative that with the protocol described here that the electrodes are placed symmetrically on the body in order to interpret laterality. Both SSEP collision testing and EMG electrophysiological monitoring may help to optimize electrode positioning and improve pain control outcomes. A prospective randomized analysis will be helpful looking at a comparison of awake versus asleep placement. The limitation in this protocol description lies in the lack of a prospective approach in obtaining data against the accepted approach with awake placement. Although asleep placement may be desirable to some, it comes at the cost of a learning curve and the necessary steps for implementation into a practice.

This paper demonstrates a single center's standard protocol for the placement of SCS electrodes. The surgeon uses this monitoring data with EMG responses to guide electrode placement. The results of several studies [6],[7],[9],[10],[12] show that when the procedure is performed in this fashion under general anesthesia, it is a viable option. In our experience, this combination has proved effective for reliable electrode placement after a percutaneous trial. In addition, this technique for placement has been used in an “open trial” situation when a percutaneous trial cannot be performed whether from scar tissue or degenerative disease hindering lead placement. Under these situations, a trial electrode, either paddle or cylindrical, can be placed under EMG guidance with general anesthesia via laminectomy. In the situation of patients who have undergone a successful percutaneous trial, a multi-array electrode can subsequently be placed with the trial as a guidance for placement. The orientation and laterality of the lead can be confirmed and adjusted based on the neurophysiologic surveillance and guidance while utilizing the same level of the trial. Its strongest utility is in determining physiologic midline as opposed to anatomic midline, as well as determining myotomal response from which the corresponding dermatomal paresthesia coverage may be inferred, and finally guides positioning of the columns of a paddle to better serve the patient with fiber selection. Considering that implantation under general anesthesia may be associated with a lower device failure rates than awake implantation, and does not compromise therapeutic efficacy it has become a viable option. This protocol can be utilized in implementing neuromonitoring into a practice for those implanting SCS systems.


This protocol can be utilized in implementing neuromonitoring for those implanting spinal cord stimulators. It allows for a method of asleep placement which uses myotomal coverage as a marker that corresponds with dermatomal coverage amounting to a specialized thoracic or cervical laminectomy placement of electrodes. This may lead to improved patient satisfaction, outcomes, decreased complications, and ease of placement for the implanting physician.


The work was carried out at St. Luke's University Health Network without funding.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Devulder J, De Colvenaer L, Rolly G, Caemaert J, Calliauw L, Martens F. Spinal cord stimulation in chronic pain therapy. Clin J Pain 1990;6:51-6.
2Devulder J, Vermeulen H, De Colvenaer L, Rolly G, Calliauw L, Caemaert J. Spinal cord stimulation in chronic pain: Evaluation of results, complications, and technical considerations in sixty-nine patients. Clin J Pain 1991;7:21-8.
3Racz GB, McCarron RF, Talboys P. Percutaneous dorsal column stimulator for chronic pain control. Spine (Phila Pa 1976) 1989;14:1-4.
4Falowski S, Celii A, Sharan A. Spinal cord stimulation: An update. Neurotherapeutics 2008;5:86-99.
5Barolat G, Massaro F, He J, Zeme S, Ketcik B. Mapping of sensory responses to epidural stimulation of the intraspinal neural structures in man. J Neurosurg 1993;78:233-9.
6Falowski SM, Celii A, Sestokas AK, Schwartz DM, Matsumoto C, Sharan A. Awake vs. asleep placement of spinal cord stimulators: A cohort analysis of complications associated with placement. Neuromodulation 2011;14:130-4.
7Mammis A, Mogilner AY. The use of intraoperative electrophysiology for the placement of spinal cord stimulator paddle leads under general anesthesia. Neurosurgery 2012;70 2 Suppl Operative: 230-6.
8Shils JL, Arle JE. Intraoperative neurophysiologic methods for spinal cord stimulator placement under general anesthesia. Neuromodulation 2012;15:560-71
9Feler C. Apparatus and method for positioning spinal cord stimulation leads. United States Patent 6027456; 2000.
10Kumar K, Lind G, Winter J, Gupta S, Bishop S, Linderoth B. Spinal Cord Stimulation: Placement of Surigcal Leads via Laminotomy-Technques and Benefits. Neurmodulaton. ed. Krames ES, Peckham PH, Rezai AR. Oxford, UK: Elsevier; 2009. p. 1005-12.
11Balzer JR, Tomycz ND, Crammond DJ, Habeych M, Thirumala PD, Urgo L, et al. Localization of cervical and cervicomedullary stimulation leads for pain treatment using median nerve somatosensory evoked potential collision testing. J Neurosurg 2011;114:200-5.
12Roth SG, Lange S, Haller J, De La Cruz P, Kumar V, Wilock M, et al. A prospective study of the intra- and postoperative efficacy of intraoperative neuromonitoring in spinal cord stimulation. Stereotact Funct Neurosurg 2015;93:348-54.