Spinal accessory nerve monitoring with clinical outcome measures

Transcription

1 ORIGINAL WITT, GILLIS, ARTICLE PRATT Spinal accessory nerve monitoring with clinical outcome measures Robert L. Witt, MD; Theresa Gillis, MD; Robert Pratt, Jr., MA, D.ABNM Abstract We conducted a prospective study of 11 patients to (1) determine the feasibility of electrophysiologic monitoring of the spinal accessory nerve (SAN) during modified radical neck dissection, (2) determine whether a threshold increase in current is required to stimulate the SAN by comparing the amount of current on initial identification of the SAN and the amount of current after completion of the dissection prior to closure, and (3) determine whether clinical outcome measures of shoulder syndrome at 72 hours and 45 days postoperatively are affected by a threshold increase. We found that 3 of 11 patients (27.3%) experienced significant threshold increases (>0.4 ma) on completion of the dissection. Of 8 patients who completed a shoulder syndrome evaluation, 3 (37.5%) had scapular winging, mild to moderate pain, and less than 90% of shoulder abduction. Two of 3 patients with shoulder syndrome had a threshold increase on electrophysiologic monitoring. We conclude that electrophysiologic monitoring of the SAN is feasible. It did not identify a threshold increase in more than 70% of patients. Electrophysiologic integrity of the SAN did not completely correlate with clinical outcome measures for shoulder syndrome. Introduction A significant number of articles has been published on electrophysiologic monitoring of (1) the facial nerve during otologic and parotid surgery and (2) the recurrent laryngeal nerve during thyroid and parathyroid surgery. However, information is limited on electrophysiologic monitoring of the spinal accessory nerve (SAN) during modified radical neck dissection. From the Department of Surgery, Christiana Care Health Systems, Wilmington, Del., and the Department of Otolaryngology Head and Neck Surgery, Jefferson Medical College, Philadelphia (Dr. Witt); the Department of Medicine, Christiana Care Health Systems, Helen F. Graham Cancer Center, Newark, Del. (Dr. Gillis); and Surgical Monitoring Associates, Bala Cynwyd, Pa. (Mr. Pratt). Reprint requests: Robert L. Witt, MD, 2401 Pennsylvania Ave., #112, Wilmington, DE Phone: (302) ; fax: (302) ; RobertLWitt@aol.com Originally presented at the Triological Society Southern Section Meeting; Jan. 14, 2006; Naples, Fla. The major morbidity associated with modified radical neck dissection and selective neck dissection is shoulder syndrome, which is caused by surgical trauma to the SAN. Shoulder syndrome includes pain, weakness, and deformity of the shoulder-girdle mechanism. Clinical parameters of the shoulder-girdle mechanism that can be measured are shoulder shrug weakness, limitations in shoulder-abduction and/or shoulder-flexion active range of motion, scapular winging at rest, and pain, typically in a characteristic location across the upper border of the trapezius muscle. Trapezius weakness is evidenced by lateral winging, which is differentiated from the medial winging seen in weakness of the serratus anterior muscle. Objective information on the results of SAN-sparing neck dissections (modified and selective neck dissections) is important because shoulder syndrome secondary to SAN trauma is the most common morbidity of neck dissection. Debate continues regarding the role of neck dissection, particularly following chemoradiation treatment. We conducted a study to (1) determine the feasibility of electrophysiologic monitoring of the SAN during modified radical neck dissection, (2) determine whether a threshold increase in current is required to stimulate the SAN by comparing the amount of current on initial identification of the SAN and the amount of current after completion of the dissection prior to closure, and (3) determine whether clinical outcome measures of shoulder syndrome are affected by a threshold increase. Patients and methods We conducted a prospective study of 11 consecutively presenting patients aged 39 to 77 years (mean: 62) who underwent modified (zones 1 through 5 with preservation of the SAN but sacrifice of the sternocleidomastoid muscle and jugular vein) radical neck dissection performed by a single surgeon (R.L.W.) at a single institution. Selective and radical neck dissections were not included in this series. Electrophysiologic recording of the amount of current on initial identification of the SAN was compared with the amount of current recorded at the completion of the procedure. Clinical correlation measured parameters of shoulder syndrome (shrug, flexion, abduction, winging, 540 ENT-Ear, Nose & Throat Journal August 2006

2 SPINAL ACCESSORY NERVE MONITORING WITH CLINICAL OUTCOME MEASURES and pain) at 72 hours and 45 days. Patients underwent a complete shoulder-girdle evaluation by a physiatrist (T.G.) who specializes in the rehabilitation of head and neck cancer patients. Patients received no formal physical therapy instruction between the initial and follow-up evaluations. All were instructed to limit their reach to no higher than shoulder height and to avoid carrying more than 5 lbs with the affected arm. Postoperative discharge information included instructions for gentle neck range-of-motion exercises, pectoral stretches performed in a supine or reclined position with the arm abducted less than 90, and scapular retraction exercises. Functional evaluations were scheduled to be performed by the physiatrist within the first 72 hours postoperatively and at approximately 45 days postoperatively. The physiatrist was blinded to the results of the electrophysiologic monitoring threshold data. Pain. The physiatric evaluation included a subjective rating of pain on a scale of 0 (no pain) to 10 (worst pain imaginable). Patients were specifically asked to describe the pain along the superior border of the scapula in the affected shoulder. Scapular winging. Patients were examined while they were in resting and active scapular positions and during shoulder range-of-motion activities. The presence of scapular winging (i.e., lateral displacement and caudal rotation of the glenoid of the scapula) was assessed with the affected arm at rest and with the patient seated. Scapular winging was rated on a scale of 0 to 3, with 0 indicating no significant displacement or no asymmetry with the contralateral scapula, 1 indicating trace or minimal displacement, 2 indicating moderate displacement, and 3 indicating severe displacement. Shoulder shrug. Strength in the upper trapezius was assessed by shoulder shrug and rated on a scale of 0 to 5, with 0 indicating no active muscle contraction, 1 indicating trace contraction, 2 indicating weak contraction manifested by an inability to raise the scapula against gravity, 3 indicating contraction sufficient to raise the scapula against gravity only, 4 indicating an ability to raise the scapula against some resistance by the examiner, and 5 indicating an ability to raise the scapula against substantial resistance by the examiner. Flexion and abduction. Shoulder active range of motion was measured with the patient in a seated position. Flexion was measured with a goniometer from 0 to 180 (full flexion). Abduction was also measured from 0 to 180, and compensatory efforts such as contralateral flexion and/or rotation of the trunk were eliminated. For the purposes of this study, the active range-of-motion measures were simplified into two outcomes: less than 90 and 90 or more. Procedures. Electrophysiologic monitoring was carried out by a clinical neurophysiologist (R.P.) who had met the professional requirements established by the American Board of Neurophysiological Monitoring (ABNM). The ABNM s minimum educational prerequisite is a master s degree in a health science. Three years of experience qualifies an individual to seek ABNM certification, which requires passing both a written and oral examination. Direct electrical stimulation of neural and nonneural tissue was administered intraoperatively during neck dissection by a 12-channel EpochXP computer system (Axon Systems; Hauppauge, N.Y.) with a Toshiba Satellite laptop (not all 12 channels were used for this procedure). Both electrically triggered and spontaneous electromyography (EMG) were evaluated. Stimulation was provided by a concentric bipolar stimulator with the following parameters stimulation rate: 2.1 Hz; duration: 100 microseconds (μsec) for neural tissue and 200 μsec for nonneural tissue. Intensity was recorded in milliamperes (ma) to account for impedance variability, and single pulses were delivered. Recording parameters were as follows bandpass filtering: 30 to 500 Hz; time window: 20 milliseconds (msec); scale: 200 microvolts (μv) per division. Both spontaneous EMG and neuromuscular responses to electrical stimulation were recorded from left and right trapezius muscles, which are innervated by the SAN; the spontaneous EMG activity of both sides was compared throughout the procedure to rule out artifacts secondary to random fluctuations in muscle tension. Subdermal-needle recording electrodes were placed in the trapezius muscle (figure, A) after induction of anesthesia, and a ground electrode was placed on the ipsilateral arm. Also, the status of the motor endplate for EMG recording was evaluated by train-of-four stimulation of the left posterior tibial nerve, with recording from the left abductor hallucis muscle (figure, B). In all cases, the train-of-four stimulation before nerve exposure demonstrated the proper function of the motor endplate for valid recording of EMG activity. Threshold of response to electrical stimulation was determined by the ascending-descending method. With the bipolar stimulator in contact with the SAN, the intensity of stimulation was slowly increased from 0 ma until a demonstrable response waveform was elicited. Then the intensity was slowly decreased until the waveform disappeared. The lowest stimulation intensity in ma at which a visible response could be obtained was considered the threshold. This technique was used before and after tumor resection. Results Three of the 11 (27.3%) patients had significant threshold increases (>0.4 ma) on completion of the dissection (table Volume 85, Number 8 541

3 WITT, GILLIS, PRATT A B Figure. A: Subdermal-needle recording electrodes are placed in the trapezius muscle. B: The left posterior tibial nerve is stimulated, with recording from the left abductor hallucis muscle. 1). The remaining 8 patients (72.7%) had increases of less than 0.1 ma or no increase at all. Eight patients completed both shoulder syndrome evaluations, one at 72 hours and the other at approximately 45 days. All 8 had shoulder flexion greater than 90 at the 45-day follow-up (table 2). Three of the 8 (37.5%) had scapular winging, mild to moderate pain, and less than 90 of shoulder abduction at the 45-day follow-up (tables 1 and 2). Two of 3 patients with shoulder syndrome had a threshold increase on electrophysiologic monitoring, while the other did not (table 1). Of the 3 patients who did not undergo a shoulder syndrome evaluation by the physiatrist, none described subjective shoulder dysfunction. These 3 patients were evaluated at 45 days in the office of the surgeon and found to have a strong shoulder shrug, flexion and abduction greater than 90, and no scapular winging; 2 experienced no pain, and 1 reported minimal pain (tables 1 and 2). None of these 3 patients had a threshold shift on electrophysiologic monitoring. In all, the rate of shoulder syndrome in the entire cohort was 27.3% (3/11). Discussion The trapezius muscle has upper, middle, and lower parts. The rotatory action of the scapula is made possible by the upper and middle parts contributing to upward rotation of the scapula and abduction of the arm at the glenohumeral joint and elevation of the arm above the shoulder level. 1 The trapezius muscle is responsible for the primary rotatory and supportive actions of the shoulder-girdle mechanism. 2 Resection of the SAN can result in shoulder syndrome as a result of denervation of the trapezius muscle. Shoulder syndrome results in downward displacement of the scapula and downward and forward drop of the shoulder. Pain is appreciated at the upper margin of the scapula, often described as a deep ache. Arm abduction can be limited to less than 90. The functional evaluation components chosen for this study included pain, scapular derangement at rest, strength of shoulder shrug, and active range of motion in shoulder flexion and abduction. These outcomes were chosen for ease and objectivity of measurement, as well as for functional implications. Pain is of obvious concern to the patient and clinician. The pain of shoulder syndrome along the upper scapular border has been attributed to the overstretch and fatigue of the levator scapula and rhomboid muscles that occur when denervation of the trapezius leads to loss of scapular support and stability. 3-6 Scapular position at rest was chosen as an indication of the degree of trapezius weakness. Active range of motion in an upright position was measured to indicate adequacy of trapezius strength for functional tasks. Flexion greater than 90 can be achieved in some patients with trapezius palsy because of the stabilizing forces of the serratus anterior muscle. Flexion measurements were obtained in an attempt to clarify that distinction. In all but 1 patient, initial flexion measurements did exceed 90. Active abduction of the shoulder was measured in all patients to indicate upper and middle trapezius strength. True abduction beyond 90 cannot be achieved in the upright position with complete trapezius paralysis or severe denervation. 6 Scapular retraction, chiefly measuring middle trapezius strength, was not chosen as a measure because it was difficult to objectively quantify this movement or ability. Shoulder shrug strength quantifies chiefly upper trapezius function, although levator scapulae and rhomboideus muscle groups also contribute some effort to this movement. Nodal recurrence rates in appropriately selected patients are not higher in SAN-preserving neck dissections. Patients undergoing SAN-sparing neck dissection have less disability than those who undergo radical neck dis- 542 ENT-Ear, Nose & Throat Journal August 2006

4 SPINAL ACCESSORY NERVE MONITORING WITH CLINICAL OUTCOME MEASURES Table 1. Clinical outcome measures of scapular winging, pain, and threshold shift Scapular winging Pain Threshold Pt.* Initial Follow-up Initial Follow-up shift (ma) * Patients 1 through 8 were evaluated by the physiatrist (T.G.), and patients 9,10, and 11 were evaluated by the surgeon (R.L.W.). Scapular winging was rated on a scale of 0 (no significant displacement or no asymmetry with the contralateral scapula) to 3 (severe displacement); pain was rated on a scale of 0 (none) to 10 (worst pain imaginable). section. 7 In a prospective study of 109 patients, Leipzig et al found that patients whose SAN, jugular vein, and sternocleidomastoid muscle were spared had a clinical dysfunction rate of 30%, compared with 50% when only the SAN was spared and 60% with radical neck dissection. 8 Not all patients who undergo radical neck dissection develop shoulder syndrome, and yet patients whose SAN is spared have a significant rate of shoulder syndrome. Motor contributions from C2, C3, and C4 to the trapezius have therefore been studied, but they have been found to be insignificant. 9 The occurrence of shoulder syndrome as a consequence of SAN-sparing neck dissection has been attributed chiefly to either an inadvertent transection of the SAN or strong traction during the course of surgery. 8 SAN-sparing neck dissections have been reported to result in some cases of temporary but reversible dysfunction that can be improved with physical therapy. However, improvement generally is not observed beyond 6 months. 10 Radiation therapy has been shown to be an additional factor in a patient s permanent disability. 11 Our study did not control for radiation. In 2002, Friedenberg et al reported that postoperative electrodiagnostic findings with needle EMG did not predict ultimate functional outcome in SAN neuropathy. 12 In 1985, Sobol et al reported that at 16 weeks postsurgery, moderate to severe EMG abnormalities were found in 65% of patients who had undergone modified radical Table 2. Clinical outcome measures of shrug, flexion, and abduction Surgical Shrug Flexion Abduction Pt.* Age Handedness side Initial Follow-up Initial Follow-up Initial Follow-up 1 65 Right Right Right Left <90 < Right Right Right Left <90 < Right Left < Right B 2 2 <90 90 <90 < Left Left Right Left < Right Left Right Right Left Right * Patients 1 through 8 were evaluated by the physiatrist (T.G.), and patients 9,10, and 11 were evaluated by the surgeon (R.L.W.). Shrug strength was rated on a scale of 0 (no movement) to 5 (normal movement). Volume 85, Number 8 543

5 WITT, GILLIS, PRATT neck dissection, compared with 22% of those who had undergone supraomohyoid neck dissection. 13 This same study found improvement in all parameters at 1 year after nerve-sparing surgery. In effect, dissection of the SAN may not predictably alter the objective shoulder disability of a particular patient. 8 This might be attributable to the role of unaffected shoulder-girdle muscles. In 2002, Midwinter and Willatt reported on the use of a nerve monitor during neck dissection. 14 They concluded that the monitor helped identify and preserve the SAN. It was not the objective of our study to use the nerve-integrity-monitor probe to identify the SAN. In fact, we used the nerve-monitor probe only after initial identification of the SAN. We found that it is feasible to monitor the SAN during SAN-sparing neck dissection. Three of 11 patients (27.3%) had a current (ma) threshold increase with a modified radical neck dissection. Patients with selective neck dissection were not included in this series, but they may have an equal or lower percentage of threshold shift in a neck dissection in which the SAN is dissected less extensively. Although 2 of 3 patients with a threshold shift did develop shoulder syndrome in our study, the other did not. A threshold shift could not predict shoulder syndrome. A larger series may determine whether a correlation between electrophysiologic threshold increase and shoulder syndrome exists. In head and neck surgery, great care and meticulous precision are taken during facial nerve dissection in parotid surgery and during recurrent laryngeal nerve dissection in thyroid dissection. Nerve integrity monitoring for routine parotid and thyroid surgery goes beyond the standard of care for these procedures and has not been proven to reduce nerve injury, 15,16 yet many surgeons, nonetheless, monitor the facial and recurrent nerve but not the SAN. The fairly routine use of nerve integrity monitoring for the facial and recurrent nerve would suggest a higher priority for these cranial nerves than the rarely reported use of monitoring of the SAN. There may still be a judgment among some surgeons that shoulder syndrome is a minimal price to pay for cancer survival. Quality-of-life studies would suggest otherwise. 17 In an age of chemoradiation, the role of staged neck dissection is evolving. The future role of neck dissection will be determined primarily by prospective studies of survival. A second factor is the morbidity of neck dissection. A heightened attention to SAN dissection with the use of precise technique, magnification, and perhaps nerve integrity monitoring may lead to reduced morbidity. In conclusion, electrophysiologic monitoring of the SAN is feasible. In this series, the anatomic and electrophysiologic integrity of the SAN was achieved in all cases. A threshold increase was not identified in more than 70% of patients. Electrophysiologic integrity of the SAN did not completely correlate with clinical outcome measures for shoulder syndrome. Nine of 10 patients (90.0%) who did not experience an electrophysiologic threshold increase did not develop shoulder syndrome. References 1. Mackenzie J. The morphology of the sternomastoid and trapezius muscles. J Anat 1955;89: Soo KC, Strong EW, Spiro RH, et al. Innervation of the trapezius muscle by the intra-operative measurement of motor action potentials. Head Neck 1993;15: Ewing MR, Martin H. Disability following radical neck dissection: An assessment based on the postoperative evaluation of 100 patients. Cancer 1952;5: Nahum AM, Mullally W, Marmor L. A syndrome resulting from radical neck dissection. Arch Otolaryngol 1961;74: Ballantyne AJ, Guinn GA. Reduction of shoulder disability after neck dissection. Am J Surg 1966;112: Saunders WH, Johnson EW. Rehabilitation of the shoulder after radical neck dissection. Ann Otol Rhinol Laryngol 1975;84: Short SO, Kaplan JN, Laramore GE, Cummings CW. Shoulder pain and function after neck dissection with or without preservation of the spinal accessory nerve. Am J Surg 1984;148: Leipzig B, Suen JY, English JL, et al. Functional evaluation of the spinal accessory nerve after neck dissection. Am J Surg 1983; 146: Nori S, Soo KC, Green RF, et al. Utilization of intraoperative electroneurography to understand the innervation of the trapezius muscle. Muscle Nerve 1997;20: Remmler D, Byers R, Scheetz J, et al. A prospective study of shoulder disability resulting from radical and modified neck dissections. Head Neck Surg 1986;8: Schuller DE, Reiches NA, Hamaker RC, et al. Analysis of disability resulting from treatment including radical neck dissection or modified neck dissection. Head Neck Surg 1983;6: Friedenberg SM, Zimprich T, Harper CM. The natural history of long thoracic and spinal accessory neuropathies. Muscle Nerve 2002;25: Sobol S, Jensen C, Sawyer W II, et al. Objective comparison of physical dysfunction after neck dissection. Am J Surg 1985;150: Midwinter K, Willatt D. Accessory nerve monitoring and stimulation during neck surgery. J Laryngol Otol 2002;116: Witt RL. Facial nerve monitoring in parotid surgery: The standard of care? Otolaryngol Head Neck Surg 1998;119: Witt RL. Recurrent laryngeal nerve electrophysiologic monitoring in thyroid surgery: The standard of care? J Voice 2005;19: Kuntz AL, Weymuller EA Jr. Impact of neck dissection on quality of life. Larynogoscope 1999;109: ENT-Ear, Nose & Throat Journal August 2006