Intraoperative Neurophysiological Monitoring During Spinal Cord Stimulation Surgery: A Systematic Review

Objectives: This study aims to describe the state of literature regarding the use of intraoperative neurophysiological monitoring (IONM) during spinal cord stimulator surgery. Materials and Methods: A systematic review of the use of IONM during spinal cord stimulation (SCS) surgery was performed using the following three data bases: PubMed, Ovid MEDLINE, and Embase. Research techniques included systematic research following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocol by Cochrane, and backward searching. Qualitative analysis of included articles was performed using the methodologic index for nonrandomized studies assessment tool. Direction of effect, consistency across studies, and cost-effectiveness were narratively synthesized. Results: A total of 15 records were identi ﬁ ed through data base searching. All records used IONM methods under general anesthesia for guidance of epidural lead placement. IONM techniques used for determining lateralization in the found articles were compound muscle action potentials (CMAPs) ( n = 8), somatosensory evoked potentials (SSEPs) ( n = 3) or both ( n = 4). Motor evoked potentials were used in three trials for neuroprotection purposes. Two studies were comparative, and 12 were noncomparative. Conclusions: We found a good body of level II evidence that using IONM during SCS surgery is a valid alternative to awake surgery and may even be superior regarding pain management, cost-effectiveness, and postoperative neurologic de ﬁ cits. In direct comparison, the found evidence suggested using CMAP provided more consistently favorable results than using SSEP for midline placement of epidural leads under general anesthesia. Selection of IONM modality should be made on the basis of pathophysiology of disease, individual IONM experience, and the individual patient.


INTRODUCTION
Spinal cord stimulation (SCS) is a neuromodulatory intervention for treating medically refractory chronic neuropathic pain.It has become an effective treatment for various conditions, such as persistent spinal pain syndrome, complex regional pain syndrome, ischemic limb pain, angina pectoris, and peripheral neuropathy, in addition to some forms of visceral pain. 1,2To date, the exact mechanism of action for SCS remains not known.However, its basis originates from the gate control theory of Melzack and Wall 3 proposed in 1965.They described a pain gate located in the substantia gelatinosa coding nociceptive inputs.Depending on the activation of small and large neural fibers, the gate could be opened or closed and thus modify pain perception.Antidromic activation of Aβ afferents, blocking of spinothalamic tract transmission, supraspinal inhibition, and release of neurotransmitters and neuromodulators also are suspected to be involved. 4iven correct placement of the spinal cord stimulating electrode is crucial for postoperative pain relief, the surgery is commonly performed under conscious sedation. 5This allows patient interaction to assess the overlap between stimulator-induced paresthesia and the painful area.However, with local anesthesia, there is always the risk of loss of airway during surgery because of oversedation. 6[9][10] The technical advances of intraoperative neurophysiological monitoring (IONM) over the past two decades opened up possibilities of performing electrode implantation under general anesthesia, precluding patient interaction and eliminating the previously mentioned risk factors during surgery.
To date, there are two common techniques for IONM during spinal cord stimulator lead placement surgery for pain: evoked potential methods related to somatosensory pathways or evoked potential methods related to muscle responses.

Electromyography/Compound Muscle Action Potentials
Electromyography is a neurophysiological method to record spontaneous or triggered muscle activity.A compound muscle action potential (CMAP) is a triggered muscle response after stimulation of the corresponding nerve.Free-running electromyography (EMG) is generally used during SCS surgery.Shils and Arle found such an approach to be appropriate for signal detection. 11ith the EMG method, the observer may continuously assess CMAPs after stimulation through the epidural leads over the dorsal column (DC).The elicitation of CMAPs after stimulation of the DC can be explained through what is known as the centrally activated H-reflex. 11,12The H-reflex, also referred to as the Hoffmann reflex, is a mono-or oligosynaptic reflex activated by stimulation of Ia afferents running to the spinal cord. 13Ia afferents form synaptic connections to the a-motoneurons after entry through the dorsal horn.When stimulated, an orthodromic action potential travels through the Ia afferents.[16]

Somatosensory Evoked Potentials
Somatosensory evoked potentials (SSEPs) provide a measurement of sensory conduction from the peripheral nerves to the DC to medial lemniscus pathways.They can be recorded on every level between the stimulation site and cerebral cortex.SSEPs can be elicited by either cutaneous or dermatomal sensory nerve stimulation.SSEPs are much smaller in amplitude than a CMAP and thus can be difficult to measure because of cortical background noise or electromagnetic disturbance from the environment. 15,16In contrast to the single pulse recorded CMAP, SSEPs are averaged potentials.Collision studies also can be performed using SSEPs because the neural pathways involved in SSEP measurement are identical to those activated from SCS. 8,13 Collision of orthodromic action potentials from peripheral nerves and antidromic action potentials elicited from epidural electrodes would thus lead to reduction or elimination of the signal travelling to the cortex, similarly to negative interference of waves.Selection of anesthetics should be carefully considered given various sedative medications affect SSEPs. 16he efficacy of SCS has already been proven multiple times, 1,[17][18][19] and protocols exist for different methods of IONM during electrode placement surgery. 13However, to our knowledge, there is no review of the available literature about this topic.This study provides a summary of the state of the art regarding the use of intraoperative neurophysiological monitoring during SCS surgery.

MATERIALS AND METHODS
A literature search following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocol for systematic reviews by Cochrane 20 was conducted in three electronic data bases for medical-scientific literature, PubMed, Ovid MEDLINE, and Embase, between July 2022 and January 2023.The literature search was systematically conducted following a Population, Intervention, Comparison, and Outcomes search strategy presented in Table 1.
Studies were only included in the analysis if published in English or German language, if the full text was available, and if published after January 2010.

RESULTS
There were 1444 records identified among all data bases through searching (Fig. 1).After subtraction of duplicates found in multiple data bases and filtering from January 1, 2010 onward, 771 records remained.These documents were then title-and abstract-screened for eligibility, whereby 717 papers were excluded for not fulfilling the previously mentioned criteria.The remaining 54 publications were full-text reviewed, and backward searching was used to cover any missed records in the search.Another seven publications were found through backward searching, yielding 61 records in total; 46 of the 61 records were excluded because of unavailability of full text, no direct use of IONM for lead placement, or not appropriate outcome (eg, stimulator model comparison) or study design (eg, animal studies, protocols).
The remaining 15 identified records were assessed qualitatively using the methodologic index for nonrandomized studies (MINORS) 21 (Table 2).The IONM techniques used for placement of epidural stimulators in the identified studies were CMAP/EMG (n = 8), SSEP (n = 3), or both (n = 4).Two of the 15 records were comparative studies 9,24 ; the remaining 13 were noncomparative trials.All but the two comparative trials used IONM under general anesthesia without comparison with an awake control group.Number of patients, stimulation characteristics, and electrode types, in addition to number of columns per lead used, are listed in Table 2, as far as was identifiable in the publications.

EMG/CMAP
Eight records measuring CMAPs were found, two of which were the only comparative trials found in the systematic research. 9,24our of the eight records reported outcomes related to pain scores.Visual analog score (VAS) or numeric rating scale (NRS) were most used.Shils and Arle reported an average reduction in VAS of 52.11% at least one month postoperatively with paddle lead placement under general anesthesia. 11Average VAS before surgery was reported at 7.51 (± 1.93) and after surgery at 3.63 (± 2.47). 11alowski et al reported a similar reduction in NRS score of 52.2%, 24 weeks after implantation under general anesthesia. 24However, the comparison with the reduction in NRS in the awake group of 55% in this study did not show statistical significance. 24Hwang et al showed a reduction from baseline NRS of 29.23% in 40 of 46 patients. 25In our study, Schlaeppi et al 30 reported a mean NRS reduction in a cohort of 20 patients from 8.2 to 3.6 (56%) at the three-month follow-up.Two studies reported reductions in subjective pain relief or pain coverage. 7,10Mammis and Mogilner documented that 82% of patients reported pain relief at latest follow up. 10 Similar numbers were described by Air et al 7 at 84.2% of patients with adequate stimulation coverage and good pain relief.Pain-paresthesia overlap postoperatively was reported by Schlaeppi et al to be at 100% for 18 of 20 patients, with the remaining two reporting a 90% overlap.One study investigated device failure rate in awake vs asleep placement. 9Device failure was defined as any reoperation secondary to traumatic break of the SCS system, device malfunctioning, or stimulator removal due to lack of efficacy. 9This study showed an incidence of device failure in awake placement and under general anesthesia of 29.7% and 14.94%, respectively. 9ymmetry and amplitude of CMAP response were used in all seven studies to determine midline placement of the paddle leads over the DC.Shils and Arle 11 and Collison et al 23

also used a plotting technique based on CMAP elicitation for individual contacts or columns of the lead to determine physiological midline.
A statistically significant reduction (p = 0.018) in operation time for placement under general anesthesia compared with the awake procedure was reported by one study. 24Reported operation time for general anesthesia was 88.9 (± 51.2) vs 125.2 (± 37.9) minutes during awake placement. 24The rate of extraneous paresthesia in the general anesthesia group (16.7% ± 23.1) was significantly lower (p < 0.001) than in the awake group (71.2% ± 30.3) in the same study. 24Collison et al 23 reported a correlation between cerebrospinal fluid thickness and postoperative energy consumption.They stated that this could greatly influence battery selection and prevent unnecessary expenses in device selection. 23eduction in baseline opioid use after device implantation was reported in 17 of 24 patients (70.83%) for one trial. 25Lead repositioning from apparent anatomical midline according to neurophysiological testing was reported by Shils and Arle 11 in 15.9% of the cases.Hwang et al 25 were the only group to use percutaneous implanted electrodes instead of paddle-type electrodes.One study used a different approach in CMAP elicitation using a double train stimulation paradigm to differentiate DC activation from CST  Table 2. MINORS Performed in Human.
Author Air et al 7 Balzer et al 8 Choi et al 22 Collison et al 23 Falowski et al 9 Falowski et al 24 Hwang et al 25 Mammis and Mogilner 10 Muncie et al 26 Penar and McSherry 27 Roth et al 6 Schoen et al 28

Shils and
Arle 11 Tamkus et al 29 Schlaeppi et al 30 activation. 30This stimulation technique was first described by Deletis et al. 12,30 SSEP Three of the 14 records used SSEPs for placement of stimulator leads under general anesthesia.Two of the three trials used an SSEP-Collision technique. 8,26Balzer et al 8 determined that a reduction in cortical SSEP amplitude >75% from baseline suggests lateralization if unilateral or midline positioning if bilateral.Muncie et al stated any decrease in cortical SSEP amplitude from baseline to be an orientation for midline placement. 26One group evaluated symmetry of response from SCS-elicited SSEPs compared with ulnar-nerve elicited SSEPs. 27The stimulated nerves for baseline SSEP recording in the three studies included ulnar nerve (n = 2), median nerve (n = 2), peroneal nerve (n = 1), and tibial nerve (n = 2).Postoperative pain relief was assessed in all three studies using SSEPs.Balzer et al reported excellent pain coverage in all patients with unilateral pain syndrome (n = 25) and sensory alterations in all patients with bilateral pain syndrome (n = 19) immediately after surgery. 8Muncie et al 26 reported five of six patients experiencing pain relief two years after surgery, ranging from 40% to 80%.In the single case report, Penar and McSherry 27 reported excellent pain coverage and relief postoperatively.

CMAP/SSEP Combined
Four studies combined the two previously mentioned techniques to determine definitive placement of electrodes.Three of the four used the SSEP collision technique and CMAP response symmetry for determining physiologic midline. 6,22,29Schoen et al 28 did not further specify whether SSEP collision testing was used.Choi et al 22 and Tamkus et al 29 interpreted any SSEP amplitude reduction as a criterion for lateralization.Roth et al determined lateralized lead placement as a reduction in amplitude of ≥40%. 6erves stimulated for SSEP acquisition across the found studies were ulnar (n = 1), median (n = 2), and tibial nerve (n = 2).Choi et al 22 recorded motor evoked potentials (MEPs) in addition to CMAP and SSEP solely for neuroprotective purposes.They defined the warning criteria as a decrease in muscle MEP. 22Two alerts occurred during their series. 22No postoperative neurologic deficits were observed. 22Tamkus et al 29 similarly investigated occurrence of IONM alerts during paddle lead placement.They reported two alerts in their series in 111 patients, with no subsequent neurologic injury. 29wo groups quantified pain perception after surgical intervention. 6,22Choi et al 22 reported subjective relief in baseline pain of ≥50% in 17 of 25 patients one week after implantation.Roth et al 6 reported statistically significant changes (p < 0.05) in four of five used pain scores.One group considered the reoperation rate using combined CMAP and SSEP for stimulator placement 28 ; 13.6% of their patients underwent subsequent revision or removal of the stimulator. 28oth et al 6 reported a rate for CMAP and SSEP predicting paresthesia coverage of the painful areas with 82.7% and 69%, respectively.
Lead repositioning based on IONM feedback for optimal midline placement was reported by one study at a rate of 18.6%, 29 similar to that reported by Shils and Arle. 11he different IONM techniques are summarized and compared in Table 3.

DISCUSSION
The neuromodulation appropriateness consensus committee (NACC) recommended the definition of successful SCS implantation to be ≥50% pain relief from baseline, as originally stated by Kumar et al. 19,31 Falowski et al 24 presented the only two comparative studies to date with a head-to-head comparison of the procedure performed awake with that performed under general anesthesia using the CMAP technique. 9They showed that pain relief in the general anesthesia group was almost identical to that in the awake group at latest follow-up, with both groups reaching the >50% pain relief on group average defining successful implantation as stated by the NACC. 24,31Despite being a nonrandomized trial with possible group heterogeneity, this study shows that performing SCS implantation with neurophysiological guidance under general anesthesia is at least equally as effective as the awake procedure.The studies by Mammis and Mogilner 10 and Shils and Arle 11 , although nonrandomized and noncomparative, show consistency with the findings of Falowski et al, with similar results for surgery under general anesthesia using CMAP.In addition, Falowski et al 24 presented other considerable outcomes in the same study.Operation time in the general anesthesia group with an average of 94.4 minutes was reported to be significantly shorter (p = 0.01) than in the awake group, with an average of 130.6 minutes.This could seriously affect the cost-effectiveness of the procedure, favoring general anesthesia over the awake procedure.Furthermore, painparesthesia coverage and extraneous sensory alterations were reported to be significantly lower in the general anesthesia group (p = 0.05 and p < 0.001, respectively). 24This addresses other important aspects of pain management, with considerable impact on quality of life other than pain relief itself.
When evaluating the SSEP technique for epidural simulator placement, less strong evidence was found.Balzer et al 8 performed the largest retrospective review to date using the SSEP-Collision technique in 44 patients.They reported excellent pain relief for all patients. 8Muncie et al 26 and Penar and McSherry 27 provided consistency with similar results.However, cohort sizes in the latter were considerably smaller, with six patients and one patient, respectively, causing us to consider the results with caution.Furthermore, none of the three papers using SSEP for placement of stimulator leads used an objective assessment for pain at baseline and after intervention, relying solely on subjective statements.
In a head-to-head comparison of the two techniques for SCS implantation under general anesthesia, we found more consistent data favoring the use of CMAP than that of SSEP.Roth et al 6 reported that CMAP had a higher prediction rate for lead lateralization than did SSEP, with 82.7% over 69%.None of the seven studies using SSEP alone or in combination with CMAP stated a greater efficacy of SSEP than that of the EMG-based technique, showing consistent favoring of the CMAP.Furthermore, the applicability of the two techniques needs to be considered.Although SSEP techniques require trained personnel for measurement and interpretation, CMAP techniques are more robust and may be performed guided by the surgeon.This makes the CMAP modality more accessible for surgeons than measuring SSEP.
The two techniques also may find different fields of application based on the underlying pathology of the disease.For example, in patients with peripheral sensory nerve disorders, such as diabetic polyneuropathy, no or pathological SSEP would be obtainable, preventing reliable assistance for midline lead placement.In contrast, CMAP cannot be elicited and used as the placement criteria in diseases affecting the lower motoneuron, such as spinal muscle atrophy.Epidemiologically, peripheral sensory nerve disorders have a higher prevalence than do denervating diseases, implying a wider applicability of CMAP than of SSEP for epidural lead placement.Nevertheless, we deem the potential of SSEP for determining physiologic midline under general anesthesia useful.
The use of MEPs was described as an adjunct for neuroprotective purposes and did not serve as a tool for midline determination.However, using IONM to provide instant feedback on several neural structures and their integrity during a procedure is another beneficial aspect of intraoperative neurophysiological monitoring.As described by Balzer et al 8 and Tamkus et al, 29 IONM can successfully aid in detection of imminent neurologic injury and prevent postoperative neurologic deficits during surgery.Although the rate of such complications is reported to be low, 29,31,32 consequences are dramatic.IONM delivers a safe approach in preventing such incidents even further.
When comparing the benefits of SCS implantation with IONM under general anesthesia with those of awake conditions, several factors must be considered.Anesthetic agents can have a major impact on the reliability of the mentioned IONM techniques and should be carefully considered. 16For example, CMAP measurement is only possible if muscle relaxants are avoided or if their effect has completely worn off after intubation.MEPs and SSEP latency or amplitude also can be affected by multiple halogenated agents used for anesthesia. 16This proves, as stated above, that IONM needs a well-trained team for correct application and reliable interpretation.
Another aspect when performing the procedure awake is the risk of airway loss due to oversedation, 6,7,10,11,24 especially given general practice is to place the patient in a prone position.A history of spinal surgeries with need for extensive scar tissue dissection, intolerance of certain local anesthetics, medical comorbidities, or anxiety can be other factors precluding the procedure under awake conditions. 7,11,25,26hils and Arle, 11 in addition to Tamkus et al, 29 showed that the lead reposition rate from anatomic midline due to IONM feedback was between 16% and 19%.The lead reposition rate in the study by Schlaeppi et al 30 using the double train stimulation paradigm was 45%.This implies that the anatomic and physiologic midline of the spinal cord can vary and are not necessarily congruent.Therefore, using IONM aids optimizing lead placement and thus helps maximize pain relief and reduce risk of extraneous paresthesia.This gives IONM another significant advantage in performing the surgery under general anesthesia rather than awake for optimal outcome.
With the elaborated criteria listed above, we state that the use of intraoperative neurophysiological monitoring, especially CMAP, for epidural lead placement under general anesthesia is a valid alternative to the awake procedure.We found solid evidence proving the method to be at least as beneficial as, if not superior to, the awake procedure regarding postoperative pain relief, extraneous paresthesia, neuroprotection, and cost-effectiveness.
However, this review is not free of limitations.The main risk lies in the level of evidence in the covered literature.No randomized controlled trials were found in the systematic research, and only two records compared their intervention with a control group.In addition, outcomes were often vaguely described as good or excellent pain relief.These valuations are considered very subjective and lack objective parameters such as pain questionnaires for cross-comparison.It also is possible that the search strategy did not sufficiently retrieve relevant records or gray literature.Backward searching of found articles was performed to cover any missed studies and reduce the risk of insufficient research output.No additional records contributing to the results of this review were identified.Another limitation is the intercomparability of the included studies.Group heterogeneity in the form of patient populations, pain conditions, surgical technique, and stimulation parameters limited the ability to pool end points.This systematic review should serve as an overview of the strengths, weaknesses, and possibilities of IONM during epidural lead placement.

CONCLUSIONS
To our knowledge, this is the first systematic review of the use of IONM for spinal cord stimulator placement surgery.We deliver a solid body of level II evidence that using IONM during SCS surgery is a valid alternative to awake surgery.We found evidence for superior pain relief, fewer postoperative neurologic deficits, and less extraneous paresthesia, in addition to shorter operation times improving cost-effectiveness for neurophysiologically guided lead placement in the asleep patient.We state a B degree of recommendation for performing SCS lead placement under general anesthesia with the use of IONM.However, more research in the form of large randomized controlled trials with objective outcome parameters is needed to further evaluate and eventually set IONM as the new standard of SCS lead placement under general anesthesia over the awake procedure.In direct comparison, the found evidence suggests using CMAPs provides more consistently favorable results than does the use of SSEPs for midline placement of epidural leads under general anesthesia.The selection of IONM modality should be based on the pathophysiology of disease, personal IONM experience, and the individual patient.

COMMENTS
This systematic review regarding the use of IONM during spinal cord stimulator surgery gives a profound overview about the actual state of the art and future directions.Despite the valid recommendations of this study, future research in the form of large randomized controlled trials with objective outcome parameters is needed.
Frank Patrick Schwarm, MD Giessen, Germany *** At times, we encounter patients who prefer not to be awakened midprocedure during neuromodulator implantation.In addition, patients may not be cognitively intact when roused from sedation, compromising their ability to respond clearly to commands.Addressing this issue, IONM during spinal cord stimulation implantation offers a viable solution that benefits both patient comfort and procedural outcomes.Using IONM allows us to keep patients under general anesthesia throughout the procedure.This approach does not compromise the outcome; rather, it could potentially enhance it.Furthermore, avoiding the necessity for patients to be awake during testing can significantly expedite the implantation process, thus boosting patient satisfaction.It is essential to note that the successful implementation of IONM necessitates additional trained personnel to monitor and interpret the data.This requirement may present a hurdle for some practitioners.However, it is worth noting that neurosurgery routinely uses this technique to mitigate nerve injury during spinal surgeries, indicating its practicality and value.In conclusion, the integration of IONM during spinal cord stimulation implantation offers a promising avenue to improve patient comfort, procedure efficacy, and overall satisfaction.Although it requires more resources, the potential benefits may well justify the investment.
Billy Huh, MD, PhD Houston, TX, USA *** Surgical spinal cord stimulator lead placement is sometimes required when a percutaneous lead placement is difficult or not possible due to anatomical restrictions.Such paddle lead placements require a spine surgery which necessitates general anesthesia and monitoring the neurological functions during such surgery will enhance the safety profile for this surgery.

Figure 1 .
Figure 1.Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram.Number of records identified using the search strategy for the three data bases PubMed, Ovid MEDLINE, and Embase with consecutive review and eventual exclusion process.

Table 1 .
Population, Intervention, Comparison, and Outcomes Search Strategy.Pain relief is summarized as all measurements for postoperative pain coverage such as pain questionnaires (Oswestry disability index, VAS, McGill pain questionnaire, pain catastrophizing scale) or subjective pain relief.

Table 3 .
Item 8, A control group having the gold standard intervention.Item 9, Contemporary groups.Item 10, Baseline equivalence of groups.Item 11, Prospective calculation of the sample size.Item 12, Statistical analyses adapted to the study design.Overview of Study Characteristics, End Points, and Stimulation Parameters for the Final Fifteen Records.
Item 1, A stated aim of the study.Item 2, Inclusion of consecutive patients.Item 3, Prospective collection of data.Item 4, End point appropriate to the study aim.Item 5, Unbiased evaluation of end points.Item 6, Follow-up period appropriate to the major end point.Item 7, Loss of follow-up ≤5%.