Sacrificing the buccal branch of the facial nerve during parotidectomy

Transcription

1 ORIGINAL ARTICLE Sacrificing the buccal branch of the facial nerve during parotidectomy Muthuswamy Dhiwakar, MS, FRCS,* Zubair A. Khan, MS Department of Otolaryngology Head and Neck Surgery, Kovai Medical Center and Hospital, Coimbatore, India. Accepted 5 May 2016 Published online 1 June 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. The need for and consequence of sacrificing the buccal branch of the facial nerve during parotidectomy is unknown. We sought to determine the indication, frequency, and functional outcome of buccal branch sacrifice. Methods. We conducted a prospective study of all cases of parotidectomy at a tertiary referral center. Results. Of 100 consecutive cases of parotidectomy, the buccal branch was sacrificed in 23 cases. This subgroup was more likely to have anterior or deep lesions (p <.001), retrograde facial nerve dissection (p 5.037), and immediate postoperative upper and lower facial weakness (p and.002, respectively). However, if the temporozygomatic and cervicomandibular branches were anatomically preserved, full facial (including buccal) function was restored. Conclusion. Deep or anterior lesions may warrant sacrifice of the buccal branch for adequate access and excision. However, this does not result in long-term impairment of facial function. VC 2016 Wiley Periodicals, Inc. Head Neck 38: , 2016 KEY WORDS: facial nerve, parotidectomy, neoplasm, neuropraxia, buccal branch INTRODUCTION Parotidectomy is a commonly performed procedure for benign and malignant disease involving the parotid salivary gland. Typically, the facial nerve is identified and branches are preserved to maintain normal facial function. However, one or more branches may need to be sacrificed if involved by disease. The upper (temporozygomatic) and lower (cervicomandibular) divisions are responsible for symmetric eye closure and lip movement, respectively. Their sacrifice, despite immediate repair, may permanently compromise function. 1 The buccal branch of the facial nerve is perceived to be responsible for buccal function, such as blowing the cheek, and may contribute to symmetric lip movement. 2 Nevertheless, its precise individual contribution to facial function is not clear. In this prospective study, we evaluated the necessity for and outcomes of buccal branch sacrifice. MATERIALS AND METHODS This study was done in accordance with local ethics committee approval. The following data were prospectively collected: patient demographics, extent of surgery, anatomic location of the lesion within the parotid gland, lesion characteristics on histopathology, and postoperative outcome. All patients had normal facial function (House Brackmann grade 1) 3 preoperatively. Excluded were *Corresponding author: M. Dhiwakar, Otolaryngology Head and Neck Surgery, Kovai Medical Center and Hospital, Avinashi Road, Coimbatore, India dhiwamahi@yahoo.com patients with preexisting facial weakness and those with sacrifice of facial nerve trunk or zygomatic or marginal mandibular branch during parotidectomy. The surgical technique followed the principles already described. 1 With particular reference to this study, the flap was raised only until the anterior limit of the lesion. Care was taken not to expose the distal branches of the facial nerve at this initial stage. The facial nerve trunk was identified as it exited the tympanomastoid groove, just superior to the posterior belly of digastric. The nerve was traced forward until its bifurcation into the upper and lower divisions (antegrade dissection). Further dissection depended on the location of pathology. For example, if a tumor involved the inferior part of the superficial lobe, only the lower branch was dissected forward. On the other hand, for a deep lobe tumor in a central location, both the upper and lower branches were dissected. Distal nerve dissection was kept limited and was done only if necessary in order to provide a safe anterior macroscopic margin. In those rare instances when the facial trunk could not be identified because of dense proximal pathology and distorted anatomy, retrograde dissection was done by first identifying a distal branch. Every effort was made to anatomically preserve all visible nerve branches while achieving en bloc removal without capsular breach. However, if the nerve was found to be adherent with no plane of separation from the neoplasm, it was sacrificed. No magnifying loupe or microscope was used. A nerve electrophysiological monitoring device was also not used. However, in cases undergoing buccal branch sacrifice, a hand-held nerve stimulator (Stimuplex A; B. Braun, Bethlehem, PA) was used at a setting of 1.0 ma to stimulate individual branches. Neuromuscular topography was HEAD & NECK DOI /HED DECEMBER

2 DHIWAKAR AND KHAN TABLE 1. Characteristics of the buccal intact versus buccal sacrificed groups. Characteristics Buccal intact no. of cases (N 5 77) Buccal sacrificed no. of cases (N 5 23) Difference, p value Location of tumor <.001* Superficial 55 1 Anterior 4 6 Deep Origin of buccal branch Upper division 6 2 NS Lower division NS Main bifurcation 10 5 NS Multiple 9 4 NS Distal buccal communication 11 4 NS Tumor capsule breach 1 0 NS Pathology Benign neoplasm NS Malignant neoplasm 9 6 NS Inflammatory/lymphovascular lesion 12 7 NS Mean maximum lesion dimension, cm 3.2 (SD 1.2) 3.8 (SD 1.8) NS Extent of parotidectomy NS Superficial 20 6 Near-total/total Margin positivity 1 2 NS Retrograde facial nerve dissection Concurrent neck dissection 10 4 NS Postoperative radiation Postoperative non-neural complication 19 7 NS Seroma 11 2 Salivary fistula 3 1 Abscess 4 2 Frey s syndrome 1 2 Abbreviation: NS, not significant. * Superficial vs anterior or deep. Dual or triple origin from upper/lower trunks/bifurcation. With another adjacent facial nerve branch. confirmed by assessing contraction of the partially exposed eyelid, buccal, and angle of mouth muscles, before the nerve was sacrificed. For the purpose of this study, lesions were classified into 3 types according to anatomic location within the parotid gland: (1) predominant involvement of any part of the superficial lobe (ie, parotid tissue lateral to facial nerve) was considered superficial (including lesions with minimal extension into the adjacent deep lobe); (2) anterior location with masseteric contact was considered anterior; and (3) predominant component involving the deep lobe and warranting dissection of both the upper (temporozygomatic) and lower (cervicomandibular) trunks was classified as deep. Involvement of the whole gland warranting total or near-total parotidectomy was also categorized as deep. Anatomy of facial nerve branching was carefully noted with particular reference to the buccal branch. Whether the latter originated from the upper temporozygomatic trunk, lower cervicomandibular trunk, or from bifurcation was recorded. Any distal communication of the buccal with other branches was also carefully studied and recorded. Postoperative facial nerve function with respect to the upper temporozygomatic and lower cervicomandibular trunks was recorded according to the House Brackmann grading system. 3 Symmetry of movement and appearance of nasolabial fold at rest was compared to the contralateral side. Oral commissure competence was also separately assessed by asking the patient to blow his/her cheek and checking symmetry by visual inspection and resistance to palpation. Patients were questioned if they had any difficulty chewing and food particle retention in the gingivobuccal sulcus. The study cohort was comprised of cases that underwent sacrifice of at least one buccal branch originating from the proximal segment of the upper or lower trunks or bifurcation of the main trunk. The control group was comprised of cases with anatomic preservation of all visible branches of the facial nerve. The 2 groups were compared with particular emphasis on intraoperative findings, nerve anatomy, and early (postoperative days 1 7) and late (second week to 6 months) functional outcomes. For categorical variables, the chisquare test with Yates correction was used. For continuous variables, the Mann Whitney U test was used. Twotailed p values <.05 were considered significant. RESULTS Of 110 consecutive cases of parotidectomy, 10 were excluded because of revision surgery or sacrifice of the temporozygomatic or cervicomandibular branch. One hundred cases of primary parotidectomy with anatomically preserved temporozygomatic and cervicomandibular 1822 HEAD & NECK DOI /HED DECEMBER 2016

3 BUCCAL NERVE SACRIFICE IN PAROTIDECTOMY FIGURE 1. (A) Mucoepidermoid carcinoma of the right parotid with no plane of separation with buccal nerve (arrowhead). (B) Near-total parotidectomy was performed; buccal branch from bifurcation sacrificed (white background), whereas smaller buccal branches from the lower division (arrow) and upper division (arrowhead) was preserved. [Color figure can be viewed at wileyonlinelibrary.com] branches were available for analysis and were included in this study. Of these, 23 (23%) underwent sacrifice of the buccal branch. The buccal preserved and buccal sacrificed groups were comparable with regard to age (44 vs 45 years) and proportion of men (57% vs 50%, respectively), although there was a preponderance of right-sided surgery in the former group (60% vs 44%). The buccal branch typically arose from the lower division. However, dual or triple origins of the buccal branches from the upper trunk, bifurcation, and lower trunk were apparent in 13 cases (13%; Table 1, Figure 1). Among the 4 cases in the study group with multiple origins of the buccal branch, only the thickest origin was sacrificed in 3 cases and dual origins were sacrificed in 1 case. Two or more buccal branches arose from the lower division in 3 cases in each group. In all 3 cases in the study group, only the thicker more proximal branch was sacrificed. Fifteen cases exhibited distal communicating loops or branches between the buccal and marginal mandibular nerves (see Figure 2). Of these, 5 had additional visible communication with the zygomatic branch. No patient underwent primary neurorrhaphy or cable nerve grafting of the sacrificed buccal branch. The buccal sacrificed group had a disproportionately higher incidence of anterior or deep lobe involvement (p <.001; Figure 1) and higher incidence of retrograde nerve dissection (p 5.037). In the buccal preserved group, no patient had postoperative weakness of zygomaticotemporal function. Nine patients (12%) had weakness of the cervicomandibular function, of which all except 1 patient exhibited full recovery at 6 months. In comparison, 2 patients (9%) in the buccal sacrificed group had early weakness of zygomaticotemporal function (p 5.051) that fully recovered. Similarly, 10 patients (43%) had immediate weakness of cervicomandibular function (p 5.002), all of which fully recovered at 6 months. The buccal sacrificed group also had more cases of early grade 3 or worse function (p 5.007; Table 2). Cases with cervicomandibular weakness in both groups also had coexistent oral commissure incompetence. However, except for the 1 case in the buccal preserved group with late persistent House Brackmann grade 2 cervicomandibular function, all cases in both groups exhibited normal oral commissure competence, along with symmetric movement and appearance at rest by 6 months. No patient reported difficulty in chewing or food particle retention on either side. DISCUSSION To our knowledge, the current report is the first to examine the need for and outcomes of buccal branch sacrifice during parotidectomy. Our data shows that 50% of cases with deep or anterior lesions warrant sacrifice of the buccal branch for adequate access and en bloc excision. This is associated with early lower facial nerve weakness in a significant proportion of cases. Nevertheless, if the other facial branches are anatomically FIGURE 2. Left parotidectomy: distal loop (double arrowhead) demonstrated between the lower division (arrow) and the buccal branch (arrowhead). [Color figure can be viewed at wileyonlinelibrary.com] HEAD & NECK DOI /HED DECEMBER

4 DHIWAKAR AND KHAN TABLE 2. Postoperative facial nerve function. Facial nerve branch Buccal intact no. of cases (N 5 77) Buccal sacrifice no. of cases (N 5 23) Early* Late Early* Late Temporal/zygomatic Grade 1: 77 Grade 1: 77 Grade 1: 21 Grade 1: 23 Grade 2: 1 Grade 2: 0 Grade 4: 1 Grade 4: 0 Marginal mandibular Grade 1: 68 Grade 1: 76 Grade 1: 13 Grade 1: 23 Grade 2: 7 Grade 2: 1 Grade 2: 5 Grade 2: 0 Grade 3: 2 Grade 3: 0 Grade 3: 3 Grade 3: 0 Grade 4: 2 Grade 4: 0 * Postoperative day 1 7. Second week to 6 months. Grade 5 House Brackmann grade. preserved, complete recovery can be anticipated with no residual dysfunction within any facial subunit. Early weakness, particularly of the lower face, occurred in 19% of the cases overall. This is likely a neuropraxic outcome induced by surgical trauma to the marginal mandibular nerve, and not due to any putative loss of motor fibers because of buccal nerve sacrifice. Evidence of lower facial weakness even in the buccal preserved group, and full recovery of function in the buccal sacrificed group support the neuropraxia hypothesis. Previous case series in which the buccal branch was not sacrificed have similarly documented early neuropraxia in 18% to 40% of cases with full recovery in the majority. 1,2,4,5 In the current report, not only did the buccal sacrificed group have more cases of early weakness, but also a higher proportion with severe neuropraxia. This is a result of the increased difficulty of dissection imposed by the less favorably located deep and anterior lesions. 1 The need for retrograde dissection, which is known to increase the risk of neuropraxia, 6 was also significantly more, further illustrating the challenging anatomy and pathology encountered in this group. The overall number of interconnections observed in the current study between the buccal and adjacent branches is significantly less when compared with previously published cadaveric studies. In contrast to surgical series, cadaveric analysis allows for special neural stains and exhaustive nerve dissection down to microscopic branches. This facilitates exponentially more anatomy to be revealed. For example, cadaveric studies have demonstrated buccal branches from both zygomaticotemporal and cervicomandibular divisions in at least 50% of cases, 7,8 in contrast to only 13% in the current study. More importantly, in a study of 350 cadaver heads by Davis et al, 9 distal communication between the buccal and marginal mandibular branches was found in every specimen. Others have similarly documented distal communication between the buccal and adjacent facial branches as a universal phenomenon. 10 Detailed anatomic examinations confirm the existence of interconnections between the buccal nerve and a neural plexus formed from branches of the zygomatic and maxillary nerves that innervate muscles around the eye and upper nose, such as orbicularis oculi, procerus, and corrugator supercilii. 10 Similarly, interconnections have been consistently demonstrated between the buccal and marginal mandibular nerves to form a plexus that innervates the buccinator and orbicularis oris. 11 In addition, there exists a group of muscles that stretch from the frontal process of the maxilla and zygomatic arch to the lateral part of the upper lip, such as zygomaticus and quadratuslabiisuperioris. These are innervated by a vertical plexus of interconnections between the ophthalmic and maxillary divisions of the trigeminal nerve and zygomatic, buccal, and marginal mandibular divisions of the facial nerve. 11,12 In fact, Yang et al 13 did a detailed study of the buccal nerve using special stains to demonstrate connections as far below as the mental nerve to supply the lower lip. This was also confirmed by others. 7 The distal buccal intercommunication rate recorded in the current report is only 15%, which compares unfavorably with the 100% recorded in the cadaveric series. Our low rate is attributable to our desire to minimize unnecessary distal nerve dissection. It is also plausible that the absence of an intraoperative loupe or microscopic magnification use in the current study might have led to smaller intercommunications being missed. Thus, although an individual facial branch is wellknown to supply a group of named muscles, there is rich cross-innervation distally with other branches of the facial nerve and trigeminal nerve. Mid and lower facial muscle groups are not reliant solely on the buccal nerve for motor function. In fact, given the functional outcomes observed in this report, the contribution of the buccal branch to mid and lower facial function is probably negligible. This allows for safe sacrifice of the proximal buccal branch without producing lasting functional sequelae. For deep or anteriorly located lesions, we believe that buccal nerve sacrifice facilitates access to the margins of the lesion and en bloc tumor excision, preventing capsular breach, and may reduce surgical trauma to other more important branches, such as the zygomatic and marginal mandibular. The data presented here offers the surgeon the flexibility to sacrifice the buccal nerve if necessary, particularly when encountering deep or anterior lesions that are surgically challenging to remove. CONCLUSION Deep or anterior lesions may warrant sacrifice of the buccal branch for adequate access and excision. However, this 1824 HEAD & NECK DOI /HED DECEMBER 2016

5 BUCCAL NERVE SACRIFICE IN PAROTIDECTOMY does not result in long-term impairment of any facial function, provided other branches are anatomically preserved. REFERENCES 1. Ikoma R, Ishitoya J, Sakuma Y, et al. Temporary facial nerve dysfunction after parotidectomy correlates with tumor location. Auris Nasus Larynx 2014;41: Mahmmood VH. Buccal branch as a guide for superficial parotidectomy. J Craniofac Surg 2012;23:e447 e House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;93: Nouraei SA, Ismail Y, Ferguson MS, et al. Analysis of complications following surgical treatment of benign parotid disease. ANZ J Surg 2008;78: Roh JL, Park CI. Function-preserving parotid surgery for benign tumors involving the deep parotid lobe. J Surg Oncol 2008;98: Cannon CR, Replogle WH, Schenk MP. Facial nerve in parotidectomy: a topographical analysis. Laryngoscope 2004;114: Tzafetta K, Terzis JK. Essays on the facial nerve: Part 1. Microanatomy. Plast Reconstr Surg 2010;125: Erbil KM, Uz A, Hayran M, Mas N, Senan S, Tuncel M. The relationship of the parotid duct to the buccal and zygomatic branches of the facial nerve: an anatomical study with parameters of clinical interest. Folia Morphol (Warsz) 2007;66: Davis RA, Anson BJ, Budinger JM, Kurth LR. Surgical anatomy of the facial nerve and parotid gland based upon a study of 350 cervicofacial halves. Surg Gynecol Obstet 1956;102: Caminer DM, Newman MI, Boyd JB. Angular nerve: new insights on innervation of the corrugator supercilii and procerus muscles. J Plast Reconstr Aesthet Surg 2006;59: Hwang K, Jin S, Park JH, Chung IH. Innervation of the procerus muscle. J Craniofac Surg 2006;17: Baker DC, Conley J. Avoiding facial nerve injuries in rhytidectomy. Anatomical variations and pitfalls. Plast Reconstr Surg 1979;64: Yang HM, Won SY, Lee JG, Han SH, Kim HJ, Hu KS. Sihler-stain study of buccal nerve distribution and its clinical implications. Oral Surg Oral Med Oral Pathol Oral Radiol 2012;113: HEAD & NECK DOI /HED DECEMBER