Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin
Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report Daniel San-juan a,⇑, Juan Barges-Coll b, Juan Luis Gómez Amador b, Marite Palma Díaz c, Alfredo Vega Alarcón c, Enrique Escanio d, David J. Anschel e, Javier Avendaño Méndez Padilla b, Victor Alcocer Barradas b, Marco Antonio Alcantar Aguilar f, Maricarmen Fernández González-Aragón a a
Neurophysiology Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico Neurosurgery Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico Neuro-otology Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico d Neuro-ophthalmology Service, National Institute of Neurology and Neurosurgery, Mexico City, Mexico e Comprehensive Epilepsy Center of Long Island, St. Charles Hospital, Port Jefferson, NY, USA f Facultad de Medicina Mexicali, Universidad Autónoma de Baja California, Mexico b c
a r t i c l e
i n f o
Article history: Received 20 November 2013 Received in revised form 15 March 2014 Accepted 1 April 2014 Available online xxxx Keywords: Abducens nerve Intraoperative monitoring Skull base surgery
a b s t r a c t Background: To determine the reliability and usefulness of intraoperative monitoring of the abducens nerve during extended endonasal endoscopic skull base tumor resection. Methods: We performed abducens nerve intraoperative monitoring in 8 patients with giant clival lesions recording with needle electrodes sutured directly into the lateral rectus muscles of the eye to evaluate spontaneous electromyographic activity and triggered responses following stimulation of the abducens nerves. Results: A total of 16 abducens nerves were successfully recorded during endoscopic endonasal skull base surgeries. Neurotonic discharges were seen in two patients (12% [2/16] abducens nerves). Compound muscle action potentials of the abducens nerves were evoked with 0.1–4 mA and maintained without changes during the neurosurgical procedures. No patient had new neurological deficits or ophthalmological complications post-surgery. Conclusions: Intraoperative monitoring of the abducens nerve during the extended endonasal endoscopic approach to skull base tumors appears to be a safe method with the potential to prevent neural injury through the evaluation of neurotonic discharges and triggered responses. Ó 2014 Published by Elsevier Ltd.
1. Introduction Intraoperative electrophysiological methods to protect the ocular motor nerves from insults during skull base procedures are crucial to preventing postoperative neurological deficits (Sekiya et al., 1993). However, reports on intraoperative extraocular muscle monitoring have been sparse and many of these detail monitoring during craniotomies to remove skull base tumors (Kawaguchi et al., 1996; Fukaya et al., 1999, 2000; Ishihara et al., 2006). To our knowledge, these methods have not been tested during less invasive neurosurgical procedures such as endonasal endoscopic approach (EEA). ⇑ Corresponding author. Address: Av. Insurgentes Sur 3877, Col. La Fama, Tlalpan, México D.F. 14269, Mexico. Tel.: +52 5556063822x2527; fax: +52 5556064532. E-mail address: [email protected] (D. San-juan).
Two techniques, transeyelid and transconjunctival, have been developed to allow the insertion of needle or hookwire electrodes in the extraocular muscles with or without anesthetics (López, 2008; Nelson and Vasconez, 1995; Rivero et al., 1995). The application of the needle or hookwire electrodes in the intraorbital space can be performed quite safely as long as the electrode is inserted far enough from the globe within the intraorbital space, but may not always be placed exactly within the muscle (Eisner et al., 1995; Rivero et al., 1995; Grabb et al., 1997; Fukaya et al., 1999; López, 2008). This may contribute to the high sensitivity and low specificity of these techniques to prevent or predict postsurgical injuries (López, 2008; Rivero et al., 1995). Abducens nerve palsy is the most common isolated cranial nerve palsy (Brinar et al., 2007), and has been reported in 16– 50% of patients who undergo skull base operations (Sekiya et al., 1993; Kawaguchi et al., 1996). EEA of the skull base is a particular
http://dx.doi.org/10.1016/j.jelekin.2014.04.001 1050-6411/Ó 2014 Published by Elsevier Ltd.
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
2
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
high risk technique requiring an accurate knowledge of its anatomy and physiology (Barges-Coll et al., 2010). In the present study, we adopted previous segmental classifications of the abducens nerve. From the endonasal endoscopic standpoint, it becomes much more relevant to consider interdural and gulfar segments, as well as proximal and distal cavernous segments separately since each of these segments has its own surgical relevance during endoscopic neurosurgery (Barges-Coll et al., 2010). Anatomical landmarks used to localize the abducens nerve intraoperatively are the vertebral basilar junction for the transclival approach, the lacerum segment of the carotid for the medial petrous apex approach, and second trigeminal branch nerve for Meckel’s cave approach (Barges-Coll et al., 2010). The present report describes a methodology to monitor abducens nerve function intraoperativetly by recording lateral rectus muscle spontaneous and triggered responses. The usefulness of this method is discussed in the context of preserving neural function during EEA. 2. Methods Neurophysiological monitoring of the activity of the external lateral rectus muscles of the eye was performed in eight giant lesions of the middle and lower clivus, performed through an EEA. All patients had a tumor that threatened the integrity of the abducens nerve. These tumors are routinely treated at our institution with a variety of extended EEA such as the transclival, suprapetrous (Meckel´s cave and anterolateral cavernous sinus), and medial petrous apex approaches (Kassam et al., 2005a,b). This study was approved by the local ethics and scientific committees. We explained to each patient or legally responsible guardian our method of intraoperative monitoring and informed consent was obtained in each case. All patients underwent a complete neuro-ophthalmic, neurological and general physical examination, as well as neuroimaging studies (Brain MRI 3.0 T; sequences T1, T2 and FLAIR pre-and postoperatively). 2.1. Anesthesia Anesthesia was induced in the patient by intravenous administration of propofol. Muscle relaxants [vecuronium or pancuronium) were only provided during intubation. A continuous infusion of propofol (2–4 mg/kg/h) was administered during surgery with bolus injections as needed. 2.2. Recording technique Disposable paired subdermal electrodes (Medtronic Xomed) insulated to within 5 mm of the tip with 2.5 mm spacing between electrodes were used to record the responses of external lateral rectus muscles with a Nicolet Endeavor CR (VIASYS Healthcare, USA) system. Electrodes were applied by a neuro-ophthalmologist after the patient was anesthetized but before the operation began. Both lateral rectus muscles were identified through the conjunctiva and needles were applied directly in the muscle after the opening of the conjunctiva. These recording electrodes were sutured through the muscle itself using Vicryl (Fig. 1A–C). The duration of time used to do so was 0.5–1 h. We used a bipolar montage per eye; an active negative electrode (G2) was placed adjacent to the active positive electrode (G1) in the lateral rectus muscle and one ground electrode was placed over the nasion (Fig. 1A–C). The bandpass was 30 Hz to 1.5 kHz, and no signal processing was used. EMG signal voltage ranged between 50 and 200 lV (amplifier gain applied; 5000 V/V) and the sweep speed at 200 ms per division. After the recording electrodes were secure,
Fig. 1. A–C. Technical position of the needle electrodes in the lateral rectus muscles. G: Ground. G1: Active positive. G2: Active negative.
an antibiotic ointment was applied to the patient’s eyes and the eyes were closed naturally. The electromyogram signal was recorded and sampled (3000 Hz) digitally for analysis. We used neurotonic discharges as an indicator of nerve irritation defined as distinctive discharges of a motor unit that appear as rapid, irregular bursts lasting several milliseconds or prolonged trains of one minute or more.
2.3. Stimulation technique After removal isolation of the vidian canal, the lateral sphenoid recess was flattened and exposure of the anterior cavernous sinus
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
3
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
Fig. 2. A–E. Patient 7. (A–B) Brain MRI with a giant Chordoma invading the lateral recess in both sides, along with the left posterior fossa. (C) Sagittal MRI showing the ventral extension of the tumor. (D) Destruction of the craniovertebral junction. (E) Pericraneal flap used to cover the large ventral defect. (F) Postoperative MRI (24 h).
was completed. Most of the patients had tumor invading that lateral recess (Fig. 2A–E), so aspiration of the tumor was done until the dura was exposed. The sixth nerve travels lateral to the paraclival carotid, just 10 mm above the lacerum segment of the carotid (Brinar et al., 2007; Barges-Coll et al., 2010). We could stimulate nerve over the dura in 5 of 8 cases at this segment. The other three cases were stimulated at the anterior cavernous sinus, just before entering the superior orbital fissure. A monopolar stimulator was
used to locate the abducens nerve during surgery. Shaft length was 100 mm, insulated to 2 mm from tip with a diameter of 0.75 mm (CardinalHealth). Cathodal stimulation was performed through this electrode, and an anodal electrode (disposable stainless-steel subdermal needle; CardinalHealth) was placed on the maxilla surface. Rectangular pulses of 0.5 ms duration were applied at a repetition rate of the 4.1 Hz. The maximum stimulus intensity was 4 mA (Sekiya et al., 1993; Kawaguchi et al., 1996).
Table 1 Clinical and socio-demographic characteristics of eight patients who underwent surgical endoscopic approach and abducens nerve monitoring intraoperatively. Age (years)/ Sex
Pathological diagnosis
Other preoperative neurological findings
Preoperative neuro-ophthalmological findings
Surgical endonasal endoscopic approach
32/F
Meningioma dorsum sellae Schwannoma Trigeminal Nasopharingeal carcinoma Meningioma clival Chordoma Chordoma Chordoma Chordoma
None
Hemianopsia right eye
Transsellar, upper Transclival
6
V right nerve
None
8
V right nerve
None
Transpterygoid, medial petrous apex Transpterygoid
None
Left hemianopsia Mild bilateral VI nerve paresis VI right nerve paresis None Mild right VI nerve paresis
Transsellar, Transclival Mid Transclival Mid and lower Transclival Lower Transclival, Transodotoid Mild and lower transclival
12 12 13 8 6
33/F 54/F 38/F 82/M 28/M 37/F 41/F
IX, XII None
Follow-up (mo)
8
Abbreviations: M: Male, F: Female, mo: month. V: Trigeminal, VI: Abducens IX: Glossopharyngeal, XII: Hypoglossal.
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
4
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
neurological deficits (Fig. 3). Abducens nerve stimulation was performed an all cases over the dura; in one patient we could stimulate the nerve at the venous confluence at the posterior cavernous sinus. In all patients, compound muscle action potentials (CMAPs) were elicited. Latencies of CMAPs were 1.5(±0.4) milliseconds. Amplitudes of CMAPs were 120(±80) lV. Response thresholds were 0.1–4.0 mA. We did not see any changes in the amplitude of CMAPs during the operations, including the two patients who developed neurotonic discharges. No patient had new neurological deficits or ophthalmological complications post-surgery (Table 1), except subconjunctival hematomas which resolved at 2 months follow-up.
4. Discussion
Fig. 3. Patient no. 7. Female 37 years old with Chordoma. Free electromyographic recordings shows at 13:18 h neurotonic discharges in both lateral ocular muscles during the tumor resection. After 5 min these abnormalities disappeared when the surgeon finished the resection of the bulk of the tumor. At 15:29 h during the additional resection of the residual tumoral tissue, neurotonic discharges were seen only on the left lateral ocular muscle, the surgeon was alerted and stopped the maneuver and the EMG returned to the baseline pattern. Abbreviations: RLR:right lateral rectus muscle, LLR: left lateral rectus muscle.
We measured the latency and amplitude manually, considering the latency as the time between the stimulus and the peak EMG response and the amplitude the largest peak-to-peak EMG response. Also, we assumed a reduction P 50% of the baseline amplitude or absence of the compound muscle action potentials evoked as indication of nerve injury. We stimulated when the neurosurgeon needed to identify the abducens nerve, especially during tumor resection. 3. Results In the eight patients, a total of 16 abducens nerves were successfully monitored intraoperativetly. Patient clinical and sociodemographic characteristics are in Table 1. Two patients developed neurotonic discharges during the resection of the tumor which altered the procedure by the surgeon changing the path of dissection. In both cases the neurotonic discharges subsequently disappeared [mean 1 min] and there were no postoperative
In our study we successfully recorded spontaneous electromyography and triggered compound muscle action potentials from the lateral rectus muscles of the eye intraoperativetly to guide the EEA surgical approach to remove several types of intracranial tumors. No patient had new postoperative neurological deficits. The primary purpose of free-running EMG in the intraoperative setting is to either identify the nerve at a time when the anatomy is disrupted, such as during tumor resection procedures involving the spine, limbs, or posterior fossa; or to protect peripheral nerves at times when there is potential for peripheral axon traction or blunt trauma (Strommena and Crum, 2008). During the EEA the direct visual inspection of the extraocular nerves may be difficult and the use of either bipolar or monopolar stimulators is limited due to the narrow surgical field. Our technique could be useful to localize the extraocular nerves when normal anatomical relationships have been altered by tumors (Sekiya et al., 1993). Previous studies found difficulties using needle electrodes to record extraocular muscles because they were easily dislodged by slight operative manipulation such as scalp reflection or orbital roof removal (Sekiya et al., 1993, 2000). One potential risk of the needle electrodes is the accidental perforation of the globe or intraorbital bleeding due to inadvertent movement of the needle tips (Sekiya et al., 1993, 2000). Periorbital disc electrodes and hookwire electrodes are an alternative option and have the advantage of being less invasive; however, they are less accurate and can more easily be contaminated by facial or masseter muscles activities confusing the identification of the source of electrical activity (Fukaya et al., 1999; López, 2008; Sekiya et al., 1993). Neurotonic discharges are spontaneous potentials which occur in response to mechanical, thermal, or metabolic irritation of the nerve that innervates a muscle (Harper and Daube, 1998). Although neurotonic discharges are sensitive indicators of nerve irritation, their presence does not always indicate damage to axons and their absence does not guarantee lack of damage. For example sharp transection of a nerve may produce no neurotonic discharges (Nelson and Vasconez, 1995). Once the nerve is transected, an evoked response can still be recorded if the distal segment is stimulated. Similarly, if there is mechanical irritation to the distal segment, neurotonic discharges can be elicited. Failure to recognize this may mislead the neurophysiologist and surgeon as to the continuity of the proximal nerve segment. If a nerve transection is suspected, electrical stimulation of the proximal segment will not elicit an evoked potential. Additionally, damaged nerve has been shown to have a higher stimulation threshold (Holland, 2002). Neurotonic discharges tend to occur more often with cranial nerve manipulation than with peripheral nerve manipulation (Harner et al., 1987). Although a discharge alone is not predictive of neurologic deficit, the amount of discharges and, in particular, the longer neurotonic discharges do appear to correlate with postoperative functional deficit (Harner et al., 1987). Pharmacological
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
neuromuscular blockade will significantly attenuate motor activity and should be avoided as much as possible. Neurotonic discharges can, however, still be recorded with neuromuscular blocking agents producing up to 75% blockade which corresponds to a CMAPs amplitude decrement of less than 100% over four successive supramaximal repetitive nerve stimulations (Holland et al., 1998). Propofol anesthesia appears to have negligible influence on spontaneous electromyography recordings (Struys et al., 1998). A limitation of the presently described transconjunctival approach is the time consuming need of an ophthalmologist to suture the needle electrode to the extraocular muscle. However, theoretically and practically, this is the most precise and efficient way to record extraocular muscular activity, because the uninsulated tip of the electrodes are within the muscle itself (Sekiya et al., 2000). Blind application of needle or hookwire electrodes in the intraorbital space can be performed quite safely as long as the electrode is inserted far enough from the globe within the intraorbital space, but may not always be placed exactly within the muscle (Eisner et al., 1995; Rivero et al., 1995; Grabb et al., 1997; Fukaya et al., 1999; López, 2008). However, hookwire electrodes have the advantages of being easier to apply, less time consuming, and an ophthalmologist is not needed to apply them (López, 2008). The amplitude and latencies of CMAPs recorded in our patients had similar values to previous studies (Fukaya et al., 1999; Kawaguchi et al., 1996). We used monopolar stimulation because it may be superior to bipolar stimulation for mapping the course of a nerve that has been obscured or distorted by a tumor (Fukaya et al., 1999; Kawaguchi et al., 1996). Loss of CMAP responses during manipulation of a lesion or higher stimulus threshold after manipulation suggests that a cranial nerve deficit will be present post-operatively. While a normal CMAP response at a low stimulus threshold (
Contents lists available at ScienceDirect
Journal of Electromyography and Kinesiology journal homepage: www.elsevier.com/locate/jelekin
Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report Daniel San-juan a,⇑, Juan Barges-Coll b, Juan Luis Gómez Amador b, Marite Palma Díaz c, Alfredo Vega Alarcón c, Enrique Escanio d, David J. Anschel e, Javier Avendaño Méndez Padilla b, Victor Alcocer Barradas b, Marco Antonio Alcantar Aguilar f, Maricarmen Fernández González-Aragón a a
Neurophysiology Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico Neurosurgery Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico Neuro-otology Department, National Institute of Neurology and Neurosurgery, Mexico City, Mexico d Neuro-ophthalmology Service, National Institute of Neurology and Neurosurgery, Mexico City, Mexico e Comprehensive Epilepsy Center of Long Island, St. Charles Hospital, Port Jefferson, NY, USA f Facultad de Medicina Mexicali, Universidad Autónoma de Baja California, Mexico b c
a r t i c l e
i n f o
Article history: Received 20 November 2013 Received in revised form 15 March 2014 Accepted 1 April 2014 Available online xxxx Keywords: Abducens nerve Intraoperative monitoring Skull base surgery
a b s t r a c t Background: To determine the reliability and usefulness of intraoperative monitoring of the abducens nerve during extended endonasal endoscopic skull base tumor resection. Methods: We performed abducens nerve intraoperative monitoring in 8 patients with giant clival lesions recording with needle electrodes sutured directly into the lateral rectus muscles of the eye to evaluate spontaneous electromyographic activity and triggered responses following stimulation of the abducens nerves. Results: A total of 16 abducens nerves were successfully recorded during endoscopic endonasal skull base surgeries. Neurotonic discharges were seen in two patients (12% [2/16] abducens nerves). Compound muscle action potentials of the abducens nerves were evoked with 0.1–4 mA and maintained without changes during the neurosurgical procedures. No patient had new neurological deficits or ophthalmological complications post-surgery. Conclusions: Intraoperative monitoring of the abducens nerve during the extended endonasal endoscopic approach to skull base tumors appears to be a safe method with the potential to prevent neural injury through the evaluation of neurotonic discharges and triggered responses. Ó 2014 Published by Elsevier Ltd.
1. Introduction Intraoperative electrophysiological methods to protect the ocular motor nerves from insults during skull base procedures are crucial to preventing postoperative neurological deficits (Sekiya et al., 1993). However, reports on intraoperative extraocular muscle monitoring have been sparse and many of these detail monitoring during craniotomies to remove skull base tumors (Kawaguchi et al., 1996; Fukaya et al., 1999, 2000; Ishihara et al., 2006). To our knowledge, these methods have not been tested during less invasive neurosurgical procedures such as endonasal endoscopic approach (EEA). ⇑ Corresponding author. Address: Av. Insurgentes Sur 3877, Col. La Fama, Tlalpan, México D.F. 14269, Mexico. Tel.: +52 5556063822x2527; fax: +52 5556064532. E-mail address: [email protected] (D. San-juan).
Two techniques, transeyelid and transconjunctival, have been developed to allow the insertion of needle or hookwire electrodes in the extraocular muscles with or without anesthetics (López, 2008; Nelson and Vasconez, 1995; Rivero et al., 1995). The application of the needle or hookwire electrodes in the intraorbital space can be performed quite safely as long as the electrode is inserted far enough from the globe within the intraorbital space, but may not always be placed exactly within the muscle (Eisner et al., 1995; Rivero et al., 1995; Grabb et al., 1997; Fukaya et al., 1999; López, 2008). This may contribute to the high sensitivity and low specificity of these techniques to prevent or predict postsurgical injuries (López, 2008; Rivero et al., 1995). Abducens nerve palsy is the most common isolated cranial nerve palsy (Brinar et al., 2007), and has been reported in 16– 50% of patients who undergo skull base operations (Sekiya et al., 1993; Kawaguchi et al., 1996). EEA of the skull base is a particular
http://dx.doi.org/10.1016/j.jelekin.2014.04.001 1050-6411/Ó 2014 Published by Elsevier Ltd.
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
2
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
high risk technique requiring an accurate knowledge of its anatomy and physiology (Barges-Coll et al., 2010). In the present study, we adopted previous segmental classifications of the abducens nerve. From the endonasal endoscopic standpoint, it becomes much more relevant to consider interdural and gulfar segments, as well as proximal and distal cavernous segments separately since each of these segments has its own surgical relevance during endoscopic neurosurgery (Barges-Coll et al., 2010). Anatomical landmarks used to localize the abducens nerve intraoperatively are the vertebral basilar junction for the transclival approach, the lacerum segment of the carotid for the medial petrous apex approach, and second trigeminal branch nerve for Meckel’s cave approach (Barges-Coll et al., 2010). The present report describes a methodology to monitor abducens nerve function intraoperativetly by recording lateral rectus muscle spontaneous and triggered responses. The usefulness of this method is discussed in the context of preserving neural function during EEA. 2. Methods Neurophysiological monitoring of the activity of the external lateral rectus muscles of the eye was performed in eight giant lesions of the middle and lower clivus, performed through an EEA. All patients had a tumor that threatened the integrity of the abducens nerve. These tumors are routinely treated at our institution with a variety of extended EEA such as the transclival, suprapetrous (Meckel´s cave and anterolateral cavernous sinus), and medial petrous apex approaches (Kassam et al., 2005a,b). This study was approved by the local ethics and scientific committees. We explained to each patient or legally responsible guardian our method of intraoperative monitoring and informed consent was obtained in each case. All patients underwent a complete neuro-ophthalmic, neurological and general physical examination, as well as neuroimaging studies (Brain MRI 3.0 T; sequences T1, T2 and FLAIR pre-and postoperatively). 2.1. Anesthesia Anesthesia was induced in the patient by intravenous administration of propofol. Muscle relaxants [vecuronium or pancuronium) were only provided during intubation. A continuous infusion of propofol (2–4 mg/kg/h) was administered during surgery with bolus injections as needed. 2.2. Recording technique Disposable paired subdermal electrodes (Medtronic Xomed) insulated to within 5 mm of the tip with 2.5 mm spacing between electrodes were used to record the responses of external lateral rectus muscles with a Nicolet Endeavor CR (VIASYS Healthcare, USA) system. Electrodes were applied by a neuro-ophthalmologist after the patient was anesthetized but before the operation began. Both lateral rectus muscles were identified through the conjunctiva and needles were applied directly in the muscle after the opening of the conjunctiva. These recording electrodes were sutured through the muscle itself using Vicryl (Fig. 1A–C). The duration of time used to do so was 0.5–1 h. We used a bipolar montage per eye; an active negative electrode (G2) was placed adjacent to the active positive electrode (G1) in the lateral rectus muscle and one ground electrode was placed over the nasion (Fig. 1A–C). The bandpass was 30 Hz to 1.5 kHz, and no signal processing was used. EMG signal voltage ranged between 50 and 200 lV (amplifier gain applied; 5000 V/V) and the sweep speed at 200 ms per division. After the recording electrodes were secure,
Fig. 1. A–C. Technical position of the needle electrodes in the lateral rectus muscles. G: Ground. G1: Active positive. G2: Active negative.
an antibiotic ointment was applied to the patient’s eyes and the eyes were closed naturally. The electromyogram signal was recorded and sampled (3000 Hz) digitally for analysis. We used neurotonic discharges as an indicator of nerve irritation defined as distinctive discharges of a motor unit that appear as rapid, irregular bursts lasting several milliseconds or prolonged trains of one minute or more.
2.3. Stimulation technique After removal isolation of the vidian canal, the lateral sphenoid recess was flattened and exposure of the anterior cavernous sinus
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
3
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
Fig. 2. A–E. Patient 7. (A–B) Brain MRI with a giant Chordoma invading the lateral recess in both sides, along with the left posterior fossa. (C) Sagittal MRI showing the ventral extension of the tumor. (D) Destruction of the craniovertebral junction. (E) Pericraneal flap used to cover the large ventral defect. (F) Postoperative MRI (24 h).
was completed. Most of the patients had tumor invading that lateral recess (Fig. 2A–E), so aspiration of the tumor was done until the dura was exposed. The sixth nerve travels lateral to the paraclival carotid, just 10 mm above the lacerum segment of the carotid (Brinar et al., 2007; Barges-Coll et al., 2010). We could stimulate nerve over the dura in 5 of 8 cases at this segment. The other three cases were stimulated at the anterior cavernous sinus, just before entering the superior orbital fissure. A monopolar stimulator was
used to locate the abducens nerve during surgery. Shaft length was 100 mm, insulated to 2 mm from tip with a diameter of 0.75 mm (CardinalHealth). Cathodal stimulation was performed through this electrode, and an anodal electrode (disposable stainless-steel subdermal needle; CardinalHealth) was placed on the maxilla surface. Rectangular pulses of 0.5 ms duration were applied at a repetition rate of the 4.1 Hz. The maximum stimulus intensity was 4 mA (Sekiya et al., 1993; Kawaguchi et al., 1996).
Table 1 Clinical and socio-demographic characteristics of eight patients who underwent surgical endoscopic approach and abducens nerve monitoring intraoperatively. Age (years)/ Sex
Pathological diagnosis
Other preoperative neurological findings
Preoperative neuro-ophthalmological findings
Surgical endonasal endoscopic approach
32/F
Meningioma dorsum sellae Schwannoma Trigeminal Nasopharingeal carcinoma Meningioma clival Chordoma Chordoma Chordoma Chordoma
None
Hemianopsia right eye
Transsellar, upper Transclival
6
V right nerve
None
8
V right nerve
None
Transpterygoid, medial petrous apex Transpterygoid
None
Left hemianopsia Mild bilateral VI nerve paresis VI right nerve paresis None Mild right VI nerve paresis
Transsellar, Transclival Mid Transclival Mid and lower Transclival Lower Transclival, Transodotoid Mild and lower transclival
12 12 13 8 6
33/F 54/F 38/F 82/M 28/M 37/F 41/F
IX, XII None
Follow-up (mo)
8
Abbreviations: M: Male, F: Female, mo: month. V: Trigeminal, VI: Abducens IX: Glossopharyngeal, XII: Hypoglossal.
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
4
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
neurological deficits (Fig. 3). Abducens nerve stimulation was performed an all cases over the dura; in one patient we could stimulate the nerve at the venous confluence at the posterior cavernous sinus. In all patients, compound muscle action potentials (CMAPs) were elicited. Latencies of CMAPs were 1.5(±0.4) milliseconds. Amplitudes of CMAPs were 120(±80) lV. Response thresholds were 0.1–4.0 mA. We did not see any changes in the amplitude of CMAPs during the operations, including the two patients who developed neurotonic discharges. No patient had new neurological deficits or ophthalmological complications post-surgery (Table 1), except subconjunctival hematomas which resolved at 2 months follow-up.
4. Discussion
Fig. 3. Patient no. 7. Female 37 years old with Chordoma. Free electromyographic recordings shows at 13:18 h neurotonic discharges in both lateral ocular muscles during the tumor resection. After 5 min these abnormalities disappeared when the surgeon finished the resection of the bulk of the tumor. At 15:29 h during the additional resection of the residual tumoral tissue, neurotonic discharges were seen only on the left lateral ocular muscle, the surgeon was alerted and stopped the maneuver and the EMG returned to the baseline pattern. Abbreviations: RLR:right lateral rectus muscle, LLR: left lateral rectus muscle.
We measured the latency and amplitude manually, considering the latency as the time between the stimulus and the peak EMG response and the amplitude the largest peak-to-peak EMG response. Also, we assumed a reduction P 50% of the baseline amplitude or absence of the compound muscle action potentials evoked as indication of nerve injury. We stimulated when the neurosurgeon needed to identify the abducens nerve, especially during tumor resection. 3. Results In the eight patients, a total of 16 abducens nerves were successfully monitored intraoperativetly. Patient clinical and sociodemographic characteristics are in Table 1. Two patients developed neurotonic discharges during the resection of the tumor which altered the procedure by the surgeon changing the path of dissection. In both cases the neurotonic discharges subsequently disappeared [mean 1 min] and there were no postoperative
In our study we successfully recorded spontaneous electromyography and triggered compound muscle action potentials from the lateral rectus muscles of the eye intraoperativetly to guide the EEA surgical approach to remove several types of intracranial tumors. No patient had new postoperative neurological deficits. The primary purpose of free-running EMG in the intraoperative setting is to either identify the nerve at a time when the anatomy is disrupted, such as during tumor resection procedures involving the spine, limbs, or posterior fossa; or to protect peripheral nerves at times when there is potential for peripheral axon traction or blunt trauma (Strommena and Crum, 2008). During the EEA the direct visual inspection of the extraocular nerves may be difficult and the use of either bipolar or monopolar stimulators is limited due to the narrow surgical field. Our technique could be useful to localize the extraocular nerves when normal anatomical relationships have been altered by tumors (Sekiya et al., 1993). Previous studies found difficulties using needle electrodes to record extraocular muscles because they were easily dislodged by slight operative manipulation such as scalp reflection or orbital roof removal (Sekiya et al., 1993, 2000). One potential risk of the needle electrodes is the accidental perforation of the globe or intraorbital bleeding due to inadvertent movement of the needle tips (Sekiya et al., 1993, 2000). Periorbital disc electrodes and hookwire electrodes are an alternative option and have the advantage of being less invasive; however, they are less accurate and can more easily be contaminated by facial or masseter muscles activities confusing the identification of the source of electrical activity (Fukaya et al., 1999; López, 2008; Sekiya et al., 1993). Neurotonic discharges are spontaneous potentials which occur in response to mechanical, thermal, or metabolic irritation of the nerve that innervates a muscle (Harper and Daube, 1998). Although neurotonic discharges are sensitive indicators of nerve irritation, their presence does not always indicate damage to axons and their absence does not guarantee lack of damage. For example sharp transection of a nerve may produce no neurotonic discharges (Nelson and Vasconez, 1995). Once the nerve is transected, an evoked response can still be recorded if the distal segment is stimulated. Similarly, if there is mechanical irritation to the distal segment, neurotonic discharges can be elicited. Failure to recognize this may mislead the neurophysiologist and surgeon as to the continuity of the proximal nerve segment. If a nerve transection is suspected, electrical stimulation of the proximal segment will not elicit an evoked potential. Additionally, damaged nerve has been shown to have a higher stimulation threshold (Holland, 2002). Neurotonic discharges tend to occur more often with cranial nerve manipulation than with peripheral nerve manipulation (Harner et al., 1987). Although a discharge alone is not predictive of neurologic deficit, the amount of discharges and, in particular, the longer neurotonic discharges do appear to correlate with postoperative functional deficit (Harner et al., 1987). Pharmacological
Please cite this article in press as: San-juan D et al. Intraoperative monitoring of the abducens nerve in extended endonasal endoscopic approach: A pilot study technical report. J Electromyogr Kinesiol (2014), http://dx.doi.org/10.1016/j.jelekin.2014.04.001
D. San-juan et al. / Journal of Electromyography and Kinesiology xxx (2014) xxx–xxx
neuromuscular blockade will significantly attenuate motor activity and should be avoided as much as possible. Neurotonic discharges can, however, still be recorded with neuromuscular blocking agents producing up to 75% blockade which corresponds to a CMAPs amplitude decrement of less than 100% over four successive supramaximal repetitive nerve stimulations (Holland et al., 1998). Propofol anesthesia appears to have negligible influence on spontaneous electromyography recordings (Struys et al., 1998). A limitation of the presently described transconjunctival approach is the time consuming need of an ophthalmologist to suture the needle electrode to the extraocular muscle. However, theoretically and practically, this is the most precise and efficient way to record extraocular muscular activity, because the uninsulated tip of the electrodes are within the muscle itself (Sekiya et al., 2000). Blind application of needle or hookwire electrodes in the intraorbital space can be performed quite safely as long as the electrode is inserted far enough from the globe within the intraorbital space, but may not always be placed exactly within the muscle (Eisner et al., 1995; Rivero et al., 1995; Grabb et al., 1997; Fukaya et al., 1999; López, 2008). However, hookwire electrodes have the advantages of being easier to apply, less time consuming, and an ophthalmologist is not needed to apply them (López, 2008). The amplitude and latencies of CMAPs recorded in our patients had similar values to previous studies (Fukaya et al., 1999; Kawaguchi et al., 1996). We used monopolar stimulation because it may be superior to bipolar stimulation for mapping the course of a nerve that has been obscured or distorted by a tumor (Fukaya et al., 1999; Kawaguchi et al., 1996). Loss of CMAP responses during manipulation of a lesion or higher stimulus threshold after manipulation suggests that a cranial nerve deficit will be present post-operatively. While a normal CMAP response at a low stimulus threshold (
