Electrophysiologic analysis of injury to cranial nerve XI during neck dissection

Transcription

1 ORIGINAL ARTICLE Electrophysiologic analysis of injury to cranial nerve XI during neck dissection Bostjan Lanisnik, MD, 1 * Miha Zargi, MD, PhD, 2 Zoran Rodi, MD, PhD 3 1 Department of ENT Head and Neck Surgery, University Medical Center, Maribor, Slovenia, 2 University Clinic for ENT and Cervicofacial Surgery, University Medical Center, Ljubljana, Slovenia, 3 Institute for Clinical Neurophysiology, University Medical Center, Ljubljana, Slovenia. Accepted 6 January 2015 Published online 26 June 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. Despite preservation of the accessory nerve, a considerable number of patients report partial nerve damage after modified radical neck dissection (MRND) and selective neck dissection. Methods. Accessory nerve branches for the trapezius muscle were stimulated during neck dissection, and the M wave amplitude was measured during distinct surgical phases. Results. The accessory nerve was mapped in 20 patients. The M wave recordings indicated that major nerve damage occurred during dissection at levels IIa and IIb in the most proximal segment of the nerve. The M waves evoked from this nerve segment decreased significantly during surgery (analysis of variance; p 5.001). Conclusion. The most significant intraoperative injury to the accessory nerve during neck dissection occurs at anatomic nerve levels IIa and IIb. VC 2015 Wiley Periodicals, Inc. Head Neck 38: E372 E376, 2016 KEY WORDS: accessory nerve, nerve mapping, neck dissection, damage, level IIb INTRODUCTION The accessory nerve provides motor input to the sternocleidomastoid and trapezius muscles. Nerve damage during surgery can result in weakness and atrophy of the trapezius muscle, leading to shoulder syndrome. There is no clear agreement on the contribution of cervical nerves to the innervation of the trapezius muscle; however, some authors believe that the nerve has a purely proprioceptive role. 1,2 The anatomy of the accessory nerve in the anterior and posterior triangle of the neck has been comprehensively described in the literature. 3 Despite the routine preservation of this nerve during neck dissection, previous studies have demonstrated that functional outcomes after a modified radical neck dissection (MRND) were impaired compared with those after selective neck dissection. Furthermore, the functional results of selective neck dissection are impaired compared with presurgery levels. 4,5 This observed decline was attributed to the dissection at level V and resection of the cervical roots. 6 It has also been postulated that avoiding dissection of level IIb may reduce postoperative cranial nerve XI morbidity. 7 K oybasioglu et al 8 reported an improvement in nerve function after a modified neck dissection compared with lateral neck dissection because of increased tension of the sternocleidomastoid muscle. In this study, we mapped the accessory nerve during an MRND using electrophysiological techniques. The M *Corresponding author: B. Lanisnik, Department of ENT and Maxillofacial Surgery, University Medical Center Maribor, Ljubljanska 5, 2000 Maribor, Slovenia. bostjan.lanisnik@siol.net wave amplitude was recorded after electrical stimulation at different relevant anatomic locations along the nerve. PATIENTS AND METHODS Twenty patients were assessed during a neck dissection for head and neck cancer between January 2012 and January The purpose of this study was to determine the location of cranial nerve XI injury during neck dissection. The study was designed as a prospective cohort analysis of patients undergoing treatment for head and neck cancer without previous surgery or radiation therapy. All patients had N0 or N1 neck status and MRND of levels I to V on one or both sides with preservation of the cervical roots. All patients who required reconstruction that affected the shoulder girdle were excluded from this study. This study was approved by the Committee for Medical Ethics of the Republic of Slovenia and supported by the University Medical Center in Maribor, Slovenia. The MRND began in an anterior to posterior direction with dissection of the trunk of cranial nerve XI. The nerve was then dissected free between levels IIa and IIb. Next, the nerve was further dissected to level V. The cervical roots were preserved in all patients. The fatty tissue and lymph nodes were dissected between the cervical roots. All operations were performed using the same techniques. The surgical procedure was divided into 3 phases as follows: phase 1 corresponded to the dissection at levels I, IIa, and IIb; phase 2 corresponded to the dissection at levels III, IV, and V; and phase 3 consisted of the 60 minutes immediately after surgery completion. To determine the location of accessory nerve injury, we measured the M wave amplitude in the trapezius muscle. The M wave amplitude decreases after axonal injury E372 HEAD & NECK DOI /HED APRIL 2016

2 INJURY TO CRANIAL NERVE XI DURING NECK DISSECTION because of conduction blocks and/or a desynchronization in conduction velocity of individual axons. Physiologically, the M wave amplitude decreases with the distance between the stimulation point and muscle because of action potential desynchronization. The electrophysiological recordings were performed using NIM 2.0 (Medtronic, Jacksonville, FL) with superficial and subcutaneous electrodes (inter-electrode distance of 0.5 cm) placed over the trapezius muscle close to each other. The active (G1) superficial electrode was placed over the muscle belly, and the reference (G2) electrode was placed over the acromion. The ground electrode was placed in the subcutaneous tissue of the jugulum. A bipolar subcutaneous electrode was inserted into the trapezius muscle at the level of the G1 superficial electrode. The superficial electrodes detected the response of a greater number of motor units compared with the bipolar subcutaneous electrodes. Appropriate anesthetic techniques for intraoperative nerve monitoring were used in combination with intravenous anesthetics and opiates (propofol and remifentanil). A bolus of a fast-acting muscle relaxant was administered to facilitate orotracheal intubation. The accessory nerve was stimulated using a handheld hook electrode. Stimulus intensity was gradually increased until the M wave amplitude remained stable. Next, the intensity was increased an additional 20% to obtain a supramaximal electrical stimulus (from ma). The stimulation procedure was repeated at each stimulation point at 5-minute intervals and 2 more measurements were done 30 and 60 minutes after the dissection was completed (see Figure 1). The M wave amplitude was automatically measured during the recording. The stimulation points were used to divide the nerve into 3 anatomic segments: the trunk of cranial nerve XI that passes through level IIa/IIb (between points A and B), the trapezius branch that passes through level V (between points B and C), and the most distal segment below the trapezius (distal to point C). The M wave amplitudes were normalized to the highest amplitude measured for each patient and separately for each stimulation point. The dissection time was not the same in all patients; thus, the raw timeline data could not be compared. Therefore, the data were processed to correspond to defined surgical phases. The highest M wave amplitude for each of the 3 stimulation points was measured at the beginning of each phase for all patients. RESULTS Measurements were completed in all 20 patients. The results from the bipolar subcutaneous and superficial electrodes correlated for all measurements, therefore, only the average relative M wave amplitude measured in the trapezius muscle using the superficial electrodes is presented in Figure 2. The dissection time was not similar for all patients (Figure 2 presents raw timeline data); therefore, these data were processed to correspond to the same 3 distinct surgical phases for each patient. The raw timeline data show a marked decrease in M wave amplitude after stimulation at point A but not at points B and C. The M wave amplitudes across the different surgical phases are presented in Figure 3. The mean relative amplitudes at the beginning of phase 1 after stimulation at points A, B, and C were 0.96, 0.95, and 0.95, respectively. At the FIGURE 1. The anatomic course of the accessory nerve with 3 marked stimulation points (A, B, and C) is shown. The trapezius muscle root is marked with black arrows, and the cervical roots are retraced with suture loops. beginning of phase 2, the mean relative amplitude after stimulation at point A was significantly lower after dissection of levels I, IIa, and IIb ( , analysis of variance; p 5.001) compared to the response at the start of surgery. After phase 2, the M wave amplitude after stimulation at point A did not return to its presurgical value, and the amplitude during phase 3 remained significantly lower compared to the amplitude at the beginning of surgery. The relative M wave amplitude measured in the trapezius muscle after stimulation of the accessory nerve at points B and C did not significantly change during surgery. The results are presented in Table 1. Figure 4 shows representative screenshots of M wave amplitude measurements after stimulation at point A during the 3 surgical phases of 2 patients. The M wave amplitude decreased after stimulation at point A during all 3 surgical phases. This result was observed with the superficial and bipolar subcutaneous electrodes. The form of the M wave did not change at any time during surgery. This result suggests that the most significant injury to the nerve occurred at the most proximal segment of the nerve between stimulation points A and B (level IIa/IIb). In addition, the damage to the nerve occurred during phase 1, and the result reflects only mild damage to the lower segments of cranial nerve XI. DISCUSSION Shoulder morbidity after modified neck dissection is well documented. The contribution of cranial nerve XI to HEAD & NECK DOI /HED APRIL 2016 E373

3 LANISNIK ET AL. FIGURE 2. The time course for the mean relative M wave amplitude along with the SE is presented. T_0 indicates the beginning of the measurements after accessory nerve trunk dissection. The measurements were recorded every 5 minutes. T_30 and T_60 indicate the resting phase 30 and 60 minutes after surgery completion, respectively. shoulder morbidity has been addressed in several studies, and data suggest that a level V and/or level IIb dissection increases shoulder morbidity and is responsible for nerve function deterioration. 9 The purpose of the present study was to determine which nerve level and dissection phase produced the most damage to the accessory nerve. Our analysis showed that the largest decrease in nerve function, measured as a decrease in the relative M wave amplitude, occurred during phase 1 after stimulation at point A. This phase corresponds to dissection at levels IIa and IIb, and stimulation at point A corresponds to the function of the most proximal point of the nerve. During FIGURE 3. The relative mean M wave amplitude along with the SE of the trapezius muscle during 3 consecutive phases (1, 2, and 3) after stimulation of the accessory nerve at 3 anatomic levels (A, B, and C) is shown. The relative M wave amplitude of the trapezius muscle decreased after stimulation at point A during the end of surgical phase 1. These data indicate injury to the nerve during this phase. The relative amplitude after stimulation at point B or C was not significantly changed. E374 HEAD & NECK DOI /HED APRIL 2016

4 INJURY TO CRANIAL NERVE XI DURING NECK DISSECTION TABLE 1. The mean relative M wave amplitude in the trapezius muscle after stimulating different locations of the accessory nerve (points A, B, and C) during surgery is shown (analysis of variance; no. of patients 5 20). Relative M wave amplitude Relative M wave amplitude change (%) Stimulation location Phase 1 Phase 2 Phase 3 Phase 1! phase 2 A B C (27), p (5), ns (2), ns Phase 2! phase (8), ns (2), ns (2), ns Abbreviation: ns, not statistically significant. The figures in bold indicate statistical significance. phase 2, the dissection occurred mainly within levels III, IV, and V, and no further deterioration of nerve function was observed. This result was evident based on the lack of significant changes to the relative amplitude in phase 3; furthermore, the amplitude during phase 3 was found to be slightly improved, suggesting that nerve damage may be reversible. Other segments of the accessory nerve below point B showed relatively mild fluctuations in M wave amplitude that were not statistically significant. Our results clearly show that the most significant damage occurred between points A and B during dissection at levels IIa and IIb. This nerve segment is dissected from its bed and subjected to traction during this part of the surgery. The nerve damage did not dissipate until after the surgery; however, this does not necessarily mean that the nerve damage is irreversible. Intraoperative electrophysiological studies of the accessory nerve are scarce. Lee et al10 used intraoperative nerve monitoring in 25 patients during selective or MRND and showed that none of the patients developed significant shoulder symptoms. Witt and Rejto11 studied the shift in electrophysiological thresholds and concluded that selective neck dissection patients had a smaller threshold shift compared with MRND patients, but electrophysiological integrity of the cranial nerve IX did not completely correlate with clinical outcome measures. Soo et al2 measured the amplitude and shape of the M waves during surgery to clarify the contribution of each cervical branch to trapezius muscle innervation. Our results show FIGURE 4. The M wave amplitude was measured with superficial (SUPERF) and subcutaneous (SUBCUT) electrodes. The response recorded by these 2 methods is largely similar; however, the difference in amplitudes indicates that a smaller number of motor units are detected with subcutaneous electrodes. As shown by the data from 2 representative patients, the amplitude but not the form of the M wave (stimulation at point A) was altered during surgery. HEAD & NECK DOI /HED APRIL 2016 E375

5 LANISNIK ET AL. that the greatest injury to the cranial nerve XI occurs at level IIa/IIb and are also supported by other clinical studies. The impact of a level IIa/IIb dissection on shoulder morbidity was studied by Parikh et al. 12 In a double-blind randomized study, these authors compared electromyography and nerve conduction studies preoperative and postoperatively and concluded that a clinically meaningful difference can be detected if level IIb is included in the dissection. Their data suggest that patients with a level IIb dissection have a greater mean decrease in the trapezius M-response amplitude; thus, inclusion of this level causes additional morbidity. The authors also concluded that their sample size was small and that further studies were needed to address the influence of level IIb dissection on postoperative trapezius muscle morbidity. 12 An electromyography study of the trapezius muscle after selective neck dissection analyzed the influence of preservation and resection of the C2 and C4 rami of the cervical plexus. The authors concluded that partial or complete denervation of the trapezius muscle was caused by axonal injury of the cranial nerve XI because traction and excision of the C2 to C4 rami further increased the dysfunction of the upper trapezius muscle. 7 Here, we showed that dissection at level IIb and freeing the nerve have an impact on nerve function. This injury at level IIb may be associated with anatomic variants and surgical approach to identifying the cranial nerve XI and manipulation during dissection Further studies are needed to correlate these findings with clinical outcome measures. Other authors have found significant damage to the nerve during dissection of level V, whereas we could not demonstrate a decrease in the M wave amplitude in the distal segments of the nerve that are anatomically positioned within level V. 6 Nevertheless, we propose that the injury to cranial nerve XI during dissection is multifactorial, which is also supported by clinical studies showing that electrophysiological results and clinical outcome measures did not correlate completely. 11,12 CONCLUSION This study was not intended to address the appropriateness of a level IIb or level V dissection from an oncologic standpoint; rather, this study was designed to determine which level and phase of surgery yielded the greatest degree of injury to cranial nerve XI. A sample size of 20 patients is relatively small; however, we showed that major damage to the nerve occurs during level IIa/IIb dissection. This result was confirmed by a decrease in the M wave amplitude when the nerve was stimulated at its most proximal point. In addition, the injury occurred during the dissection of level II. These results need to be confirmed in further independent studies using a larger sample population. REFERENCES 1. Brodal A. Neurological anatomy in relation to clinical medicine. Third ed. New York, NY: Oxford University Press; Soo KC, Strong EW, Spiro RH, Shah JP, Nori S, Green RF. Innervation of the trapezius muscle by the intra-operative measurement of motor action potentials. Head Neck 1993;15: Lloyd S. Accessory nerve: anatomy and surgical identification. J Laryngol Otol 2007;121: Cheng PT, Hao SP, Lin YH, Yeh AR. Objective comparison of shoulder dysfunction after three neck dissection techniques. Ann Otol Rhinol Laryngol 2000;109(8 Pt 1): Chepeha DB, Taylor RJ, Chepeha JC et al. Functional assessment using Constant s Shoulder Scale after modified radical and selective neck dissection. Head Neck 2002;24: Cappiello J, Piazza C, Giudice M, De Maria G, Nicolai P. Shoulder disability after different selective neck dissections (levels II IV versus levels II V): a comparative study. Laryngoscope 2005;115: Tsuji T, Tanuma A, Onitsuka T, et al. Electromyographic findings after different selective neck dissections. Laryngoscope 2007;117: K oybasioglu A, Tokcaer AB, Uslu S, Ileri F, Beder L, Ozbilen S. Accessory nerve function after modified radical and lateral neck dissections. Laryngoscope 2000;110: Bradley PJ, Ferlito A, Silver CE, et al. Neck treatment and shoulder morbidity: still a challenge. Head Neck 2011;33: Lee CH, Huang NC, Chen HC, Chen MK. Minimizing shoulder syndrome with intra-operative spinal accessory nerve monitoring for neck dissection. Acta Otorhinolaryngol Ital 2013;33: Witt RL, Rejto L. Spinal accessory nerve monitoring in selective and modified neck dissection. Laryngoscope 2007;117: Parikh S, Tedman BM, Scott B, Lowe D, Rogers SN. A double blind randomised trial of IIb or not IIb neck dissections on electromyography, clinical examination, and questionnaire-based outcomes: a feasibility study. Br J Oral Maxillofac Surg 2012;50: Lanisnik B, Zargi M, Rodi Z. Identification of three anatomical patterns of the spinal accessory nerve in the neck by neurophysiological mapping. Radiol Oncol 2014;48: Tatla T, Kanagalingam J, Majithia A, Clarke PM. Upper neck spinal accessory nerve identification during neck dissection. J Laryngol Otol 2005;119: Lee SH, Lee JK, Jin SM, et al. Anatomical variations of the spinal accessory nerve and its relevance to level IIb lymph nodes. Otolaryngol Head Neck Surg 2009;141: E376 HEAD & NECK DOI /HED APRIL 2016