Intraoperative Neurophysiological Monitoring During

Intraoperative Neurophysiological Monitoring During Sacrectomy Procedures Faisal R. Jahangiri, M.D., CNIM, DABNM, FASNM1; Sami Al Eissa, M.D.2; Anila F. Jahangiri, Ph.D.3; Amro Al-Habib, M.D., FRCSC, MPH4 1

Division of Neurology, Department of Medicine Division of Orthopedics, Department of Surgery 3 King Abdullah International Medical Research Center King Fahad National Guard Hospital King Abdulaziz Medical City Riyadh, Saudi Arabia 4 Neurosurgery Division, Department of Surgery College of Medicine King Saud University Riyadh, Saudi Arabia 2

Neurodiagn J. 53:312–322, 2013 © ASET, Missouri

Intraoperative Neurophysiological Monitoring During Sacrectomy Procedures Faisal R. Jahangiri, M.D., CNIM, DABNM, FASNM1; Sami Al Eissa, M.D.2; Anila F. Jahangiri, Ph.D.3; Amro Al-Habib, M.D., FRCSC, MPH4 1

Division of Neurology, Department of Medicine Division of Orthopedics, Department of Surgery 3 King Abdullah International Medical Research Center King Fahad National Guard Hospital King Abdulaziz Medical City Riyadh, Saudi Arabia 4 Neurosurgery Division, Department of Surgery College of Medicine King Saud University Riyadh, Saudi Arabia 2

ABSTRACT. Previously intraoperative neurophysiological monitoring (IONM) has not been used along with a computer based navigation system for en bloc resection of a sacral Ewing sarcoma. In order to improve the post-operative neurological outcome of the patient we decided to include IONM in our procedure. A partial or complete resection of a sacral tumor may result in the loss of neurological functions due to close proximity of vascular, neural, and visceral structures. A prolonged two-stage surgical procedure may be a high risk procedure for position related brachial plexus injury. An 18-year-old male presented with left lower extremity weakness, which worsened with gait. His MRI was consistent with a sacral mass causing compression on the left S1 and S2 roots. A surgical resection was planned with anterior and posterior approaches. IONM helped guide the surgical team to prevent damaging the sacral roots on the normal side (right) and position related upper extremity brachial plexus injuries. Author’s Email: [email protected] Received: June 9, 2013. Accepted for publication: July 24, 2013.

312

IONM DURING SACRECTOMY PROCEDURES

313

Our technique involving IONM can be used safely for accurate en bloc removal of a sacral tumor with a safe margin while protecting the neural function and minimizing recurrence. This case report demonstrates that intraoperative neurophysiological monitoring was useful in identifying and reversing impending nerve injury during sacrectomy surgery. Significant changes were seen in ulnar and posterior tibial somatosensory evoked potentials (SSEPs). We recommend that IONM should be considered for safe margin en bloc sacral tumor resection and prevention of injury to the sacral root and brachial plexus. KEY WORDS. Bulbocavernosus reflex (BCR), computer assisted navigation system, Ewing’s sarcoma, intraoperative neurophysiological monitoring (IONM), sacrectomy, somatosensory evoked potential (SSEP), spontaneous electromyography (s-EMG), triggered electromyography (t-EMG).

OBJECTIVE Intraoperative neurophysiological monitoring (IONM) is a real-time assessment of the functional integrity of the peripheral or central nervous system during a high risk surgical procedure. One of the main objectives of utilizing IONM during these cases is to reduce any position related brachial plexus injuries due to prolonged compression and ischemia (Jahangiri et al. 2011).

INTRODUCTION The benefits of intraoperative neurophysiological monitoring (IONM) have been most commonly studied and published for spine, brain, vascular, and ear, nose, and throat (ENT) surgeries. IONM is considered a standard of care for scoliosis surgeries (Nuwer et al. 1995). During lengthy surgical procedures, IONM can provide protection of neural tissues such as brain, spinal cord, and peripheral nerves. Utilizing a multi-modality approach of neurophysiological monitoring; somatosensory evoked potentials (SSEP), bulbocavernosus reflex (BCR), and spontaneous and triggered electromyography (s-EMG and t-EMG); gives a broad coverage of the nervous system. It is beneficial during non-spine cases (Eager et al. 2011). Each modality has its own benefits and limitation as well as unique specificity and sensitivity. The modalities used were SSEP, BCR, and s-EMG and t-EMG. In order to maximally benefit from IONM, teamwork between surgical, anesthesia, and neurophysiological monitoring teams is needed (Michler et al. 2013). Position related brachial plexus injury during surgery is a significant cause of postoperative neurological deficit. Upper extremity ulnar and median nerves SSEPs are

314


used as reliable indicator of brachial plexus function (Chung et al. 2009) during prolonged and even short duration surgical procedures. Upper extremity SSEP is typically used during spine and sometimes during brain surgeries for preventing position related brachial plexus injuries. SSEP is not being used during other surgeries where there is still a high risk of post-operative neuropathy. We also utilized lower extremity posterior tibial nerve SSEPs to prevent any injury to the spinal cord and lower extremities. We are presenting the benefit of utilization of IONM during sacral tumor resection for preventing perioperative position related brachial plexus injury. The functional integrity of spinal cord roots S2-S3-S4 including sensory and motor pathways can be monitored intraoperatively by BCR (Deletis and Vodušek 1997). The BCR is an oligosynaptic reflex consisting of sensory (afferent) and motor (efferent) pathways. The sensory pathways of the BCR are a part of pudendal nerve. The motor component supplies external anal sphincter muscles and bulbocavernosus muscles. Therefore, including BCR monitoring during removal of tumors involving the sacral roots helps in monitoring the sensory pathways, motor pathways, and spinal gray matter involving sacral roots (Vodušek et al. 1993). Anesthetic inhalational agents and muscle relaxants suppress BCR because of multiple synapses involved as well as neuromuscular junctions.

CASE REPORT Clinical Scenario and Images An 18-year-old male patient presented with minimal weakness of left ankle/toes plantar-flexion, which got worse with walking. The magnetic resonance imaging (MRI) showed a sacral mass involving S1 and S2 vertebras; with more extension to the left side (Figure 1). This sacral mass (tumor) appeared to be compressing the left S1 and S2 nerve roots. It was extending into the left paravertebral space and through the neural foramen into the extradural intraspinal canal causing severe compression and displacement of the thecal sac to the right side. It produced a mass effect on the adjacent parts of the sigmoid but no evidence of invasion. The left sacroiliac (SI) joint was also involved by the tumor. Computed tomography (CT) confirmed the involvement of the left SI joint with the lytic tumor. There were no metastatic lesions identified on body imaging. CT-guided biopsy confirmed the diagnosis of Ewing sarcoma. Pre-operative session of chemotherapy was given to shrink the tumor. Surgical Procedure Anesthesia was administered using total intravenous anesthesia (TIVA) with propofol and remifentanil infusion (Anschel et al. 2008). The TIVA protocol without inhalational agents and any muscle relaxant is recommended for optimal recording


315

FIG. 1. Preoperative MRI. There is a large mass lesion, hypointense in T1 and hyperintense in T2, originating mainly from the left side of S1 extending into the sacral ala and left side of S2 with extension into the left paravertebral space and through the neural foramen into the extradural intraspinal canal causing severe compression and displacement of the thecal sac to the right side. It exerts a mass effect on the adjacent parts of the sigmoid. The left sacroiliac joint is entangled by the tumor (black arrows).

of both SSEP and EMG responses (Hilibrand et al. 2004). Neuromuscular blockade was only used for intubation. For the remaining duration of the procedure, a train of 4/4 twitches was maintained (Figure 2). Surgery was performed in two stages. The first stage was anterior to free the anterior tumor surface from major vessels and abdominal organs. Using intraoperative navigation, anterior cortical cutting was performed around the tumor mass in the sacrum and iliac bone with safe margins. Stage two included a laminectomy from L5 to S3 with pedicle screw fixation from L3 to S1 and right iliac screw placement. Ligation was performed of the left L5, S1, S2, and S3 roots. Navigator oriented posterior cortex cutting around the mass to meet the anterior cutting was done. Following soft tissue release, total tumor resection was performed with a safe margin (Figure 3). It included subtotal sacrectomy, left SI joint, and adjacent part of the iliac bone. Reconstruction was done using a fibular bone graft between the two iliac bones and a titanium cage from the lower L3 edge to the sacrum (Figure 4). Intraoperative Neurophysiological Monitoring (IONM) After patient intubation, electrodes were placed for somatosensory evoked potentials (SSEP), electromyography (EMG), bulbocavernosus reflex (BCR), and train of

316


FIG. 2. Train of Four (TOF) data presenting all four twitches present. Left: Four present TOF responses. Right: TOF in histogram view.

four (TOF). For SSEPs, stimulation was set up for the posterior tibial nerve in lower extremities (at medial malleolus) and the ulnar nerve in upper extremities (at wrist) bilaterally. Stimulation intensity was set to 25 mA, duration of 0.3 msec, and repetition rate of 3.7 hertz (Hz) for ulnar nerves. The stimulation intensity was set to 60 mA, duration of 0.3 msec and a repetition rate of 3.7 Hz for posterior tibial nerves. For SSEP recording, posterior tibial nerves were stimulated at the ankles and the ulnar nerves were stimulated at the wrists bilaterally. Recording electrodes for spontaneous electromyography (s-EMG) were placed in the tibialis anterior (TA), medial gastrocnemius (MG), abductor hallucis (AH), extensor hallucis brevis (EHB) in the lower extremities, and external anal sphincter muscles for monitoring sacral roots. For BCR stimulation, surface pad electrodes were placed on the dorsal surface of the penis. The recording subdermal EMG electrodes were placed in the external anal sphincters bilaterally. We set the alarm criteria for SSEP as more than 10% increase in latency or more than 50% decrease in amplitude of the waveforms. To preserve the anal sphincter function, EMG and BCR (Rodi and Vodušek 2001) were monitored bilaterally throughout the surgical procedure. Stimulation setup for BCR responses was setup as double-pulse electrical stimulation with inter-stimulus interval (ISI) of 3.0 msec and a stimulation rate of 2.3 Hz (Deletis and Shils 2002). For BCR, the alarm criteria set was a significant or a complete loss of responses or a significant change in the waveform morphology. The surgeon was continuously informed of the sphincter s-EMG and BCR responses especially during the resection of left S1, S2, and S3 nerve roots resection. Right S1, S2, and S3 nerve root responses remained stable during the left side resection.


FIG. 3. Sacral tumor after resection measuring about 6 cm x 4 cm x 5.8 cm.

317

318


FIG. 4. Post-operative MRI showing reconstruction using a fibular bone graft between the two iliac bones and a titanium cage from the lower L3 edge to the sacrum.

Surgery This sacrectomy was a two stage procedure, stage one was an anterior approach and stage two was a posterior approach. During this surgery, the navigation system was used in meeting the anterior image cuts done in the first stage to the posterior images created in the second stage, in order to resect part of sacrum with safe margins around it. Stage one (which lasted nine hours) was an anterior approach. This approach was done in order to ligate the external iliac artery bilaterally allowing an anterior release and anterior cortex cutting around the mass. After this, an anterior cut was made in the sacrum to the right side of the tumor with a safe margin using a navigation system. IONM was done continuously during stage one. IONM data remained stable throughout stage one. The second stage (which lasted 14 hours) was a posterior approach. This stage included a lower lumbar spine instrumentation with right iliac screw placement, bilateral laminectomy of L5, S1-S2, and S3 to expose the dura, ligation of the left L5, S1, S2, and S3 roots, navigator oriented posterior cortex cutting around the mass to meet the anterior cutting, release of the sacral piece, and taking out the mass. A fibular bone graft was inserted and fixed between the two iliac bones.


319

Changes Changes in neurophysiological data were noted after 15 hours into the surgery during the posterior resection. Loss of ulnar nerve SSEP in the left upper extremity was noted. A follow-up recording confirmed the loss in left ulnar Erb’s (brachial plexus), subcortical, and cortical responses (Figure 5). Surgeon and the anesthesia team were immediately informed. The anesthesiologist repositioned the shoulders twice but there was no improvement in the SSEP signals. The surgeon decided to stop the tumor resection and adjusted the shoulder pads. Left ulnar SSEP responses returned close to the baselines within 30 minutes. Surgical resection of the sacral tumor was resumed. A drop in left lower posterior tibial nerve SSEP followed the division of the left L5, S1, and S2 roots resection (Figure 6) along with loss of left BCR responses. During the remaining part of the tumor resection upper extremity ulnar SSEP, right lower posterior tibial nerve SSEP, and right BCR remained stable.

FIG. 5. Ulnar nerve somatosensory evoked potential (SSEP) data showing positioning effect. There is a loss/decrease in amplitude bilaterally in peripheral, subcortical, and cortical responses. Trace nine (9) is the baseline response, while traces 142 (right) and 145 (left) are the new responses with changes bilaterally (black arrows).

320


FIG. 6. Posterior tibial nerve somatosensory evoked potential (SSEP) stack data showing significant decrease in amplitude of left side responses at 23:11 (black arrow). There is a loss/decrease in amplitude in left peripheral, subcortical, and cortical responses followed by cutting of left L5, S2, and S2 nerve roots that were involved in the sacral tumor. Right side responses remained stable.

Post-operative The patient’s recovery from surgery was uneventful. Post-surgical MRI showed tumor-free margins of resection. Neurologically, left ankle and toes plantar-flexion was 0/5. There was also loss of sphincter control for three days that might have been caused by neuropraxia due to cauda equina traction during the surgery. One week later, post-operative sphincter function returned to normal with no deficits. Patient had full control of bowel and bladder functions. Patient’s upper extremity sensory and motor functions were intact. Chemotherapy was started soon after; radiotherapy was delayed for about six weeks. One year post-operatively patient was free of tumor and had no neurological deficits in upper extremities. Patient’s bowel and bladder functions/control were also intact.

DISCUSSION Brachial plexopathy may result from improper positioning of the shoulder and/or arms during prolonged surgical procedures in the prone position. Position related


321

nerve injury could be secondary to traction on the nerve or direct pressure causing focal ischemia. When the arms are tucked by tape it exerts extra pressure on brachial plexus and may result in post-operative ulnar neuropathy. It is very important to set up for an Erb’s point recording in addition to subcortical and cortical responses for upper SSEPs. This will help is differentiating and troubleshooting the causes of loss of SSEP responses. Immediate recognition of the changes in signal and readjusting/ repositioning of the shoulders/arms improve the SSEP signals and recovery is possible. If there is delayed or no intervention or no real-time intraoperative neurophysiological monitoring this may result in permanent post-operative neurological deficit. Cheney et al. (1999) describes six hundred and seventy (670) anesthesia related nerve injury claims. Most frequent sites of nerve injuries in 48% of these non-monitored cases were brachial plexus and ulnar nerve (Cheney et al 1999). In another study by Parks (1973), the most common type of nerve injury was brachial plexopathy in 38% of the injuries, with a rate of 0.14% in 50,000 surgical procedures. The importance of upper extremity SSEP monitoring for position related nerve injury during sacral tumor resection is presented in this case report. To avoid any long term or permanent post-operative neurological impairment an early intervention is demonstrated as well. While upper limb SSEP monitoring is usually done for spine and brain surgeries, it should be considered for other surgeries such as anterior/ posterior sacral tumor resections as well where there is a risk of brachial plexus injury due to positioning and pressure during prolonged surgical procedures. BCR responses were continuously recorded and were not very robust during this procedure. During the second (posterior) part of the surgery there was a decrease in the amplitude of left lower posterior tibial nerve SSEP following the division of the left L5, S1, and S2 roots resection. There was also an immediate loss of left BCR responses following left L5, S1, and S2 roots resection. Surgeon was informed of these changes after nerve roots resection. The right lower posterior tibial nerve SSEP and right BCR remained stable throughout the surgical procedure.

CONCLUSION Utilization of multi-modality intraoperative neurophysiological monitoring during non-spinal surgeries in addition to the navigation system can prevent or decrease the risk of devastating post-operative neurologic deficits. These deficits are usually a result of brachial plexus injuries due to patient arm positioning during prolong surgical procedures. IONM, specifically upper SSEPs, should be considered for nonspinal procedures that involves prone positioning for a long period of time and where there is a risk of brachial plexus injuries.

322


ACKNOWLEDGEMENTS The authors would like to thank Dr. Mohannud Mjurkesh, Assistant Consultant, Division of Orthopedics, Department of Surgery; Mr. Abdullah Al Bishi, Supervisor Clinical Neurophysiology Laboratory, Division of Neurology, Department of Medicine; and Muhammad Imtiaz, Neuronavigation Unit, Surgical Nursing, King Fahad National Guard Hospital, King Abdulaziz Medical City, Riyadh, Saudi Arabia for their valuable contributions and help with preparing this article.

REFERENCES Anschel DJ, Aherne A, Soto RG, Carrion W, Hoegerl C, Nori P, Seidman PA. Successful intraoperative spinal cord monitoring during scoliosis surgery using a total intravenous anesthetic regimen including dexmedetomidine. J Clin Neurophysiol 2008; 25(1):56–61. doi: 10.1097/ WNP.0b013e318163cca6. Cheney FW, Domino KB, Caplan RA, Posner KL. Nerve injury associated with anesthesia: a closed claims analysis. Anesthesiology 1999; 90(4):1062–69. Chung I, Glow JA, Dimopoulos V, Walid MS, Smisson HF, Johnston KW, Robinson JS, Grigorian AA. Upper-limb somatosensory evoked potential monitoring in lumbosacral spine surgery: a prognostic marker for position-related ulnar nerve injury. Spine J 2009 9(4):287–95. Deletis V, Shils J (Editors). Neurophysiology in Neurosurgery. New York, NY: Academic Press; 2002; p. 211–12. Deletis V, Vodušek DB. Intraoperative recording of the bulbocavernosus reflex. Neurosurgery 1997; 40(1):88–93. Eager M, Shimer A, Jahangiri FR, Shen F, Arlet V. Intraoperative neurophysiological monitoring (IONM): lessons learned from 32 case events in 2069 spine cases. Am J Electroneurodiagnostic Technol 2011; 51(4):247–63. Fourney DR, Gokaslan ZL. Current management of sacral chordoma. Neurosurg Focus 2003 Aug 15; 15(2):E9. Hilibrand AS, Schwartz DM, Sethuraman V, Vaccaro AR, Albert TJ. Comparison of transcranial electric motor and somatosensory evoked potential monitoring during cervical spine surgery. J Bone Joint Surg Am 2004; 86-A(6):1248–53. Jahangiri FR, Holmberg A, Vega-Bermudez F, Arlet V. Preventing position-related brachial plexus injury with intraoperative somatosensory evoked potentials and transcranial electrical motor evoked potentials during anterior cervical spine surgery. Am J Electroneurodiagnostic Technol 2011; 51(3):198–205. Jahangiri FR. Surgical Neurophysiology: A Reference Guide to Introperative Neurophysiological Monitoring (IONM): Second Edition. Charleston, South Carolina: CreateSpace Publisher; 2012. Michler RP, Unsgård G, Rossvoll I. [Neurophysiological monitoring during surgery]. Tidsskr Nor Laegeforen. 2013 Feb 5;133(3):306–11. [Article in Norwegian] Nuwer MR, Dawson EG, Carlson LG, Kanim LE, Sherman JE. Somatosensory evoked potential spinal cord monitoring reduces neurologic deficits after scoliosis surgery: Results of a large multicenter survey. Electroencephalogr Clin Neurophysiol 1995; 96:6–11. Parks BJ. Postoperative peripheral neuropathies. Surgery 1973; 74(3):348–57. Rao G, Chang GJ, Suk I, Gokaslan Z, Rhines LD. Midsacral amputation for en bloc resection of chordoma. Neurosurgery 2010; 66 (3 Suppl Operative):41–44. Rodi Z, Vodušek DB. Intraoperative monitoring of the bulbocavernosus reflex: the method and its problem. Clin Neurophysiol 2001; 112:879–83. Schwartz DM, Drummond DS, Hahn M, Ecker ML, Dormans JP. Prevention of positional brachial plexopathy during surgical correction of scoliosis. J Spinal Disord 2000; 13(2):178–82. Vodušek DB, Deletis V, Kiprovski K. Intraoperative bulbocavernosus reflex monitoring: decreasing the risk of postoperative sacral dysfunction. Neurourol Urodyn 1993; 12:425.