Navigate
By: Justin W. Silverstein, DHSc, FASNM, FASET1,2
1Neuro Protective Solutions, New York, NY
2Department of Neurology, Northwell Health/Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, New York, NY
In the early 2000s, Luis Pimenta revolutionized lumbar spine surgery by introducing the transpsoas approach for lumbar interbody arthrodesis. This approach offered distinct advantages over traditional methods, such as avoiding major vessels and peritoneal contents (common in anterior approaches) and preserving paraspinal musculature and thecal sac integrity (typical in posterior approaches). Additionally, the minimally invasive nature of the procedure leads to faster recovery, and shorter hospital stays. This approach is also capable of accommodating larger footprint cages compared to conventional transforaminal and posterior lumbar interbody fusion techniques.
Despite these significant benefits, the transpsoas approach presents unique risks that were previously uncommon for spine surgeons. These include the potential for injury to the lumbar plexus or fully formed peripheral nerves, resulting in postoperative neuropathies or plexopathies—distinct from radiculopathies. Given that the lumbar plexus resides within the substance of the psoas muscle, this was the first time a surgical approach necessitated a requisite neuromonitoring protocol.
In 2006, Ozgur and colleagues1 introduced the extreme lateral lumbar interbody fusion (LLIF) technique in their pioneering publication. This groundbreaking work not only detailed the surgical procedure but also established a crucial neuromonitoring paradigm. Initially, only spontaneous electromyography (S-EMG) and triggered electromyography (T-EMG) were employed during these procedures. S-EMG served to continuously assess nerve function and alert the surgeon to potential impending neurological compromise. Conversely, T-EMG was utilized to precisely locate and approximate lumbar plexus elements within the psoas muscle during the approach to the disc space.
Despite the recommended use of neuromonitoring and mapping techniques, patients were still experiencing postoperative femoral nerve deficits or lumbar plexopathies, with the literature reporting an incidence rate of 36%.2 This led to the hypothesis that while approach-related injuries can occur, the primary mechanism of injury was likely attributed to the tensile force on the nerve exerted by the retractor blades rendering S-EMG essentially ineffective in detecting these injuries. S-EMG has a notably high false negative rate as a neuromonitoring modality. Moreover, S-EMG lacks the ability to detect slow stretch/compressive injuries to nerves that may result in conduction block, a capability present with evoked potentials.3-5
To access the disc space, an incision is made, followed by a meticulous finger dissection until the upper surface of the psoas muscle is reached. At this juncture, the psoas is methodically opened in a cylindrical fashion through sequential dilation. These dilators serve a dual purpose, acting also as directional monopolar stimulation probes. During each dilation sequence, the dilator is electrified, eliciting T-EMG responses.
The intensity level of the stimulus required to activate the T-EMG response serves as a metric for gauging the proximity of the dilator to neural elements. Lower values indicate closer proximity to the nerve, while higher values signify greater distance. The dilators are typically designed with shielding on one side, enabling directional mapping. The surgeon rotates the dilator to effectively determine the spatial relationship between neural elements within the psoas muscle and the dilator. Following this, a tubular retractor is inserted and expanded. The surgical field is then carefully examined using a monopolar stimulation probe to verify the absence of any lumbar plexus elements within the designated working corridor for the procedure.
While T-EMG is adept at guiding the surgeon to the disc space in LLIF procedures, it does not provide insight into the functional continuity of the nerve. In cases where nerve injury is attributed to retraction, the stimulation from a probe within the surgical field is likely occurring at or distal to the site of nerve injury.
In 2010, Juan Uribe initiated a series of publications addressing the management of postoperative deficits following transpsoas approaches to the spine.6,7 He also shared insights into his own complications during lectures at various symposiums. At the annual meeting for the Society of Lateral Access Surgery in 2010, I had the privilege of attending one of his lectures. Witnessing the specific femoral nerve injuries he presented, I promptly jotted down “saphenous nerve SSEP” on a piece of paper. I discreetly passed it to a surgeon colleague I was sitting next to—an early adopter of the transpsoas approach and a subsequent research collaborator—and expressed my belief that this could be a valuable method for monitoring femoral nerve function.
Upon returning from the conference, I was determined to master the acquisition of saphenous nerve somatosensory evoked potentials (SSEPs). I delved into my anatomy books, tracing the nerve’s course and identifying the optimal targeting approach. With Institutional Review Board (IRB) approval secured, I began implementing saphenous nerve SSEPs in all lumbar spine surgeries within our practice.
Initially, my efforts focused on targeting the infrapatellar branch of the saphenous nerve. Regrettably, I encountered challenges in obtaining consistent signals from this stimulation site. I shifted my approach to the mid-thigh region, utilizing 13mm needle electrodes to pinpoint the groove between the sartorius and vastus medialis muscles (adductor canal) [Figure 1]. Based on my research indicating the nerve to be at a depth of approximately 5cm at this location, I applied a high pulse width of 1ms, resulting in stable, reproducible recordings with 40mA of stimulation.
Additionally, I explored stimulating the saphenous nerve at its most distal site, just above the anterior ankle, using sticker electrodes. While this approach produced excellent recordings, I harbored reservations about whether I was truly capturing saphenous nerve SSEPs or if there was potential volume conduction to the peroneal nerves. Recognizing that the mid-thigh approach offered the most isolated method for obtaining saphenous nerve SSEPs, I opted to concentrate on this method and establish my case series.
During the course of my research, I came across a remarkable paper by Davis et al. (2011)8 that provided an insightful review of lumbar plexus anatomy in cadavers, specifically in relation to the placement of retractor blades utilized in LLIF procedures. Their work elegantly demonstrated the trajectory of the fully formed femoral nerve as it traverses the disc space at the L4-L5 level. They were pioneers in asserting that encountering nerve roots in these procedures is unlikely, emphasizing the importance for surgeons to be prepared to encounter the trunk of the fully formed femoral nerve. The excerpt below, extracted from the Davis et al. paper, effectively encapsulates the state of neuromonitoring for transpsoas approaches circa 2011.
“Compounding factors are lack of visualization of these structures and reliance on intraoperative neuromonitoring techniques that may be only marginally effective. Current neuromonitoring methods utilize triggered electromyography with use of an insulated probe to identify neural structures. This modality may help to avoid direct spearing of the neural motor structures but does little thereafter. It does not reliably assess the integrity of nerves transversing the surgical site because stimulation of retractor blades typically occurs distal to the site of traction or compression. Monitoring of somatosensory evoked potentials (SSEP) is an additional modality that tests the sensory tracts but has limited usefulness because detection of nerve root injury is difficult with this modality and because the posterior tibial and peroneal nerves that are monitored are not the neural structures that are at risk during the transpsoas lateral approach.”
The timeline of efforts and collaborative work undertaken to address postoperative deficits in these procedures is intriguing. However, it is worth noting that the methods employed at the time primarily revolved around carefully timing surgical retraction and subsequent post-operative management of complications.
In 2011, we encountered our initial instance of isolated approach side saphenous nerve SSEP deterioration during surgery. We promptly alerted the surgeon, but at that time, meaningful interventions were unknown. While there was an initial attempt to elevate blood pressure, no other effective countermeasures were implemented. Unfortunately, the response remained attenuated throughout the procedure, and upon awakening, the patient exhibited sensorimotor deficits in the femoral nerve distribution. Regrettably, this injury proved to be a lasting femoral nerve impairment persisting for years following the procedure. Nonetheless, it served as a pivotal case, providing us with valuable insights on how to intervene and which countermeasures should be employed when an alert is encountered. These measures encompassed the release or removal of surgical retraction, expediting the procedure, and augmenting blood pressure at the first sign of a hyperacute femoral nerve conduction failure.
I had the privilege of presenting this case, along with the novel technique, for the first time ever at the 2012 Society for Lateral Access Surgery annual conference in San Diego, CA. Subsequently, I disseminated our data at numerous regional, national, and international symposia over the ensuing years. In 2013, following a presentation on the topic at the Western Society of END Technologists, I had the opportunity to meet a neurologist from UC Irvine who expressed interest in the technique. I provided her with guidance on optimizing stimulation at the mid-thigh, leading to subsequent publications from her group that validated our own research (which I will elaborate on later).
Furthermore, in 2013, our research was selected for presentation during the best paper session at the International Society for the Advancement of Spinal Surgery (ISASS) annual conference in Vancouver, BC, Canada. The technique garnered significant interest at ISASS, and we were honored to be interviewed by ISASS for a sneak peek of our presentation prior to the event.
After years of refining our techniques and gathering data, we achieved a significant milestone in 2014 with the publication of our seminal paper titled “Saphenous Nerve Somatosensory Evoked Potentials: A Novel Technique to Monitor the Femoral Nerve During Transpsoas Lumbar Lateral Interbody Fusion Surgery” in Spine. In this study involving 46 patients, we successfully obtained saphenous nerve SSEP data in 89% of the cohort, with 100% sensitivity and specificity in identifying postoperative femoral nerve injury. Instances where approach side saphenous nerve SSEP deterioration occurred, and did not recover, were associated with patients waking up with a range of femoral nerve injuries, from sensory impairments to sensorimotor deficits.
It is crucial to highlight that our control SSEPs (posterior tibial nerve and deep peroneal nerve) consistently remained stable even when saphenous SSEPs exhibited deviations. Additionally, there were no instances of notable S-EMG activity observed.
In 2015, Bederman and colleagues9 from UC Irvine introduced their initial abstract and subsequent manuscript10 employing saphenous nerve SSEPs in LLIF procedures. Adopting the mid-thigh stimulation site, they obtained reliable saphenous nerve SSEPs in 52 out of 62 patients (84%). Their independent validation of our research yielded identical results—100% sensitivity and specificity in detecting postoperative femoral nerve injury. Notably, 7 patients experienced intraoperative saphenous nerve SSEP deterioration, with only 1 returning to baseline with intervention. This particular patient remained neurologically intact upon awakening, while the remaining 6 patients exhibited varying degrees of femoral nerve sensorimotor deficits.
From 2013 to 2020, our ongoing saphenous nerve SSEP research was consistently selected for podium presentation at every ISASS annual conference, except for 2019 when we did not submit. Notably, we were honored with inclusion in the best paper session for the second time at the 2020 ISASS annual conference in San Juan, PR.
In 2013, with the addition of Jon Block to our research team, we integrated motor evoked potentials targeting quadriceps muscles as a specific adjunct to our neuromonitoring paradigm, further enhancing the safeguarding of the femoral nerve during LLIF procedures. This pivotal addition led to the publication of our influential paper titled “Motor Evoked Potentials for Femoral Nerve Protection in Transpsoas Lateral Access Surgery of the Spine” in 2015.11
The following year, in 2016, we presented our data on appropriate interventions for approach side femoral nerve neuromonitoring deteriorations.12 Our findings demonstrated that if the retractors were released or removed within 10 minutes of an alert, the signals would recover within 1 minute. Conversely, if the intervention was delayed (>10 minutes), the mean signal recovery time extended to 7 minutes. Moreover, when no intervention was initiated, no signal recovery was observed, resulting in patients awakening with femoral nerve deficits. This pattern persisted over the ensuing decade, aligning with the results obtained by the UC Irvine group, to our knowledge the only other group publishing on Saphenous nerve SSEPs and providing outcome data.
In 2017, Bederman et al.13 once again published their follow-up case series, involving 38 patients who underwent LLIF procedures with saphenous nerve SSEP monitoring. They reliably obtained mid-thigh saphenous nerve SSEPs in 89% of their cohort. Notably, they encountered two alerts, both of which were promptly addressed through the release of retraction. This timely intervention resulted in the recovery of degraded signals, ultimately leading to the absence of postoperative femoral nerve deficits across the entire cohort.
In 2021, Sánchez Roldán et al.14 conducted a study focused on optimizing saphenous nerve SSEPs recordings. They performed a comparison study between the proximal (mid-thigh) stimulation site and a distal site (positioned approximately halfway between the knee and ankle) placing electrodes between the tibia and medial gastrocnemius. They achieved successful mid-thigh stimulation in 77% of their patients and observed that the distal stimulation site yielded either similar or superior amplitudes. For both sites, they utilized a 0.5ms pulse width and applied intensities ranging from 40 to 50mA. Specifically, 13mm needles were employed at the mid-thigh, while surface electrodes were used at the distal site.
While their study was primarily focused on technique without providing outcome data for comparison, it holds significant importance as it presents an alternative stimulation site for a technically challenging SSEP procedure. Notably, our own approach has evolved over the years, with a shift away from 13mm needles at the proximal site. For approximately the past decade, we have transitioned to using 22mm needles and initiating stimulation at 20mA, adjusting intensity levels as needed.
In 2022, we published our latest research utilizing saphenous nerve SSEPs and MEPs with quadriceps recordings in The Spine Journal.15 This comprehensive study involved an examination of 172 surgeries, revealing saphenous nerve SSEP and/or MEP with quadriceps recordings deterioration in 19 patients. We successfully reversed these degradations in 17 patients; however, regrettably, we encountered persistent signal loss in the remaining 2 patients. Both of these patients experienced significant femoral nerve palsies upon awakening. Encouragingly, all 17 patients with signal recovery awoke neurologically intact, mirroring the outcomes of the remainder of patients who had no alerts throughout the procedure.
This study adds substantial weight to the effectiveness of employing these modalities in monitoring femoral nerve function during LLIF procedures. Additionally, our findings indicated that retraction time did not significantly influence signal loss. As long as the evoked potentials remained intact, the retractor could be kept open for the necessary duration to complete the procedure. However, it is crucial to promptly implement countermeasures at the initial indication of electrical conduction failure from the evoked potentials.
To the best of our knowledge, Khajavi and Niznik16 stand as the only other group to assess both saphenous SSEPs and MEPs with quadriceps recordings simultaneously. They presented their data alongside ours at the ISASS annual conference in 2015.
While this editorial primarily focuses on saphenous nerve SSEPs, it’s essential to acknowledge the researchers who have embraced MEPs with quadriceps recordings. For example, Chaudary et al. (2015)17 and Riley et al. (2019)18 have published outcome data that mirrors our own over the past decade and a half. When saphenous nerve SSEPs and/or MEPs with quadriceps recordings are present at the conclusion of the procedure, patients emerge from surgery without any deficits. Conversely, if these recordings deteriorate and fail to recover, patients wake up with a femoral nerve deficit. It is worth noting that this correlation is unique to LLIF procedures, given that the injury affects the fully formed peripheral nerve, at the point where there is no longer any alternative overlapping innervation from nerve roots.
While these modalities may hold potential benefits in detecting upper lumbar nerve root injuries, we currently lack the data for direct comparison. However, in the context of LLIF procedures, the sensitivity and specificity of these modalities are both an impressive 100%, an uncommon achievement for a neuromonitoring technique. Our consistent outcomes have received validation from multiple independent research groups.
It is crucial to highlight that in every publication addressing this subject, S-EMG consistently failed to serve as an indicator for impending evoked potential deterioration or postoperative deficits.
As of the time of writing, no outcome studies have been conducted using the distal stimulation site. Therefore, our recommendation is to consider monitoring the saphenous nerve SSEP at both sites, particularly if difficulties arise in obtaining signals from the proximal site. It’s important to note that improper parameters for distal site stimulation may lead to the activation of other nerves, potentially obscuring any deterioration in femoral nerve function.
While the proximal stimulation site has its limitations, including potential movement (which can be mitigated by adjusting the repetition rate) and the depth of the nerve itself, it’s worth noting that in LLIF procedures, the patient is typically secured enough to minimize any disruptive effects of movement during surgery. Some helpful techniques for obtaining signals from the proximal site include positioning the patient in a frog-legged position to enhance visibility of the groove between the sartorius and vastus medialis muscles, inserting the electrodes at a 90-degree angle, using 22mm needle electrodes, and initiating stimulation at 20mA with a 1ms pulse width (Figure 2).
Over the past decade, saphenous nerve SSEPs have become firmly established in the field of surgical neurophysiology and spine surgery. Before 2012, the topic of saphenous nerve SSEPs was never broached. Fast forward to 2023, and it’s impossible to attend a neurodiagnostic, neuromonitoring, or spine surgery conference without encountering discussions about them. In fact, we’ve reached a juncture where companies specializing in spinal instrumentation are incorporating saphenous nerve SSEPs into their device applications. Furthermore, saphenous nerve SSEPs are a prerequisite modality for candidates seeking eligibility to sit for the CNIM complex spine micro-credential board examination. This signifies a notable evolution in the field.
When utilized appropriately, the incorporation of saphenous nerve SSEPs and motor evoked potentials with quadriceps recordings in transpsoas approaches to the spine has the potential to virtually eradicate post-operative femoral nerve deficits.
References:
[1]Ozgur BM, Aryan HE, Pimenta L, Taylor WR. Extreme Lateral Interbody Fusion (XLIF): a novel surgical technique for anterior lumbarinterbody fusion. Spine J 2006;6(4):435–43. https://doi.org/10.1016/j.spinee.2005.08.012.
[2]Hijji FY, Narain AS, Bohl DD, et al. Lateral lumbar interbody fusion: a systematic review of complication rates. Spine J 2017;17(10):1412–9. https://doi.org/10.1016/j.spinee.2017.04.022.
[3]Raynor BL, Padberg AM, Lenke LG, Bridwell KH, Riew KD, Buchowski JM, Luhmann SJ. Failure of intraoperative monitoring to detect postoperative neurologic deficits: a 25-year experience in 12,375 spinal surgeries. Spine (Phila Pa 1976). 2016;41:1387–93
[4]Silverstein J, Mermelstein L, DeWal H, Basra S. Saphenous nerve somatosensory evoked potentials. Spine 2014;39(15):1254–60. https://doi.org/10.1097/brs.0000000000000357.
[5]Duncan JW , Bailey RA , Baena R . Intraoperative decrease in amplitude of somatosensory-evoked potentials of the lower extremities with interbody fusion cage placement during lumbar fusion surgery . Spine (Phila Pa 1976) 2012 ; 37 : E1290 – 5 .
[6]Uribe JS, Arredondo N, Dakwar E, Vale FL. Defining the safe working zones using the minimally invasive lateral retroperitoneal transpsoas approach: an anatomical study [published correction appears in J Neurosurg Spine. 2010 Dec;13(6):798]. J Neurosurg Spine. 2010;13(2):260-266. doi:10.3171/2010.3.SPINE09766
[7]Ahmadian A, Deukmedjian AR, Abel N, Dakwar E, Uribe JS. Analysis of lumbar plexopathies and nerve injury after lateral retroperitoneal transpsoas approach: diagnostic standardization. J Neurosurg Spine 2013;18(3):289–97.
[8]Davis TT, Bae HW, Mok MAJJ, Rasouli A, Delamarter RB. Lumbar plexus anatomy within the psoas muscle: implications for the transpsoas lateral approach to the l4-l5 disc. J Bone Joint Surg Am 2011;93(16):1482–7. https://doi.org/10.2106/jbjs.j.00962.
[9]Bederman SS, Jain N, Yanni D, Guillen H, Lilit M. Saphenous Nerve Somatosensory Evoked Potentials Monitoring during Lateral Interbody Fusion. Global Spine Journal. 2015;5(1_suppl):s-0035-1554571-s-0035-1554571. doi:10.1055/s-0035-1554571
[10]Jain NS, Bederman SS, Phan K, Yanni DS, Guillen H, Mnatsakanyan L. Saphenous nerve somatosensory evoked potentials monitoring during lateral interbody fusion. Spine J 2015;15(10). https://doi.org/10.1016/j.spinee.2015.07.393.
[11]Block J, Silverstein JW, Ball HT, et al. Motor evoked potentials for femoral nerve protection in transpsoas lateral access surgery of the spine. The Neurodiagn J 2015;55(1):36–45. https://doi.org/10.1080/21646821.2015.1012456.
[12]Silverstein, JW, Block, J, Madhok, R, Goldstein, M, Basra, S, Mermelstein, L, et al. Alerts for femoral nerve monitoring during transpsoas lateral lumbar interbody fusion procedures. Abstract presented at: The Annual Meeting for ISASS; 2016; Las Vegas, NV.
[13]Bederman SS, Jain N, Yanni D, Guillen H, Lilit M. Saphenous nerve somatosensory evoked potentials monitoring during lateral interbody fusion. Global Spine J 2015;5. https://doi.org/10.1055/s-0035-1554571
[14]Sánchez Roldán MÁ, Mora Granizo F, Oflidis V, et al. Optimizing the methodology for saphenous nerve somatosensory evoked potentials for monitoring upper lumbar roots and femoral nerve during lumbar spine surgery: technical note. J Clin Monit Comput. 2022;36(4):1079-1085. doi:10.1007/s10877-021-00737-6
[15]Silverstein JW, Block J, Smith ML, et al. Femoral nerve neuromonitoring for lateral lumbar interbody fusion surgery. Spine J. 2022;22(2):296-304. doi:10.1016/j.spinee.2021.07.017
[16]Khajavi, K, Niznik, G. Femoral nerve monitoring during lateral transpsoas approach for lumbar interbody fusion. Abstract presented at: The Annual Meeting for the International Society for the Advancement of Spine Surgery; 2015; San Diego, CA.
[17]Chaudhary K, Speights K, McGuire K, White AP. Trans-cranial motor evoked potential detection of femoral nerve injury in transpsoas lateral lumbar interbody fusion. J Clin Monit Comput 2015;29(5):549–54. https://doi.org/10.1007/s10877-015-9713-8.
[18]Riley MR, Doan AT, Vogel RW, Aguirre AO, Pieri KS, Scheid EH. Use of motor evoked potentials during lateral lumbar interbody fusion reduces postoperative deficits. Spine J 2018;18(10):1763–78.https://doi.org/10.1016/j.spinee.2018.02.024.