The Laryngoscope C 2014 The American Laryngological, V
Rhinological and Otological Society, Inc.
Great Auricular Nerve Preservation in Parotid Gland Surgery: Long-term Outcomes Alberto Grammatica, MD; Pietro Perotti, MD; Francesco Mancini, MD; Anna Bozzola, MD; Cesare Piazza, MD; Piero Nicolai, MD; Luca Oscar Redaelli de Zinis, MD Objectives/Hypothesis: To assess sensory outcomes and quality of life (QoL) in two groups of patients with and without great auricular nerve (GAN) preservation during parotidectomy. Study Design: Retrospective chart review. Methods: The posterior branch of the GAN was preserved in 42 patients (group A) and sacrificed in 13 (group B). Tactile, heat, and cold sensitivities were investigated by dividing GAN territory into seven areas. Comparisons between operated (OS) and nonoperated sides (NS) within each group, and between the OS of the two groups were made. The QoL questionnaire was administered. Results: In group A, normal tactile, heat, and cold sensitivities ranged from 16.7% to 66.7%, 11.9% to 73.8%, and 21.4% to 81%, respectively, in different OS areas. Significant differences between OS and NS were found, except for the preauricular superior area. In group B, normal tactile, heat, and cold sensitivities ranged from 0% to 61.5%, 0% to 53.8%, and 7.7% to 76.9%, respectively, in different OS areas. Significant differences between OS and NS were found except for the preauricular superior (tactile sensitivity), and preauricular superior and helix/concha areas (cold sensitivity). Comparing the OS tactile and thermic sensitivities between the two groups, only the lobule area showed differences. The preauricular inferior area was different only for heat. The QoL questionnaire showed different hypoesthesia extension between the two groups. All other items were comparable. Conclusions: Sensory deficits are commonly reported despite GAN preservation. Lobule and preauricular inferior areas showed differences in terms of tactile and thermic sensitivities, with better outcomes in group A. QoL seems tolerable despite GAN sacrifice. Key Words: Parotid gland, surgery, great auricular nerve, nerve sacrifice, nerve preservation, quality of life. Level of Evidence: 4 Laryngoscope, 00:000–000, 2014
INTRODUCTION Parotidectomy is a common, well-standardized, head and neck surgical procedure. Facial nerve injury, Frey’s syndrome, and aesthetic deformities are the most common complications following parotidectomy, whereas important, but less well-studied, sequelae include hypoanesthesia, paresthesia, and discomfort to the auricle and periauricular areas due to neurotmesis or neuroapraxia of the great auricular nerve (GAN). In fact, during parotidectomy, the GAN can be either mobilized and preserved or sacrificed to expose the posterior and inferior aspects of the gland. Expected consequences of nerve transection are numbness of the
Additional Supporting Information may be found in the online version of this article. From the Department of Otorhinolaryngology–Head and Neck Surgery, University of Brescia, Brescia, Italy. Editor’s Note: This Manuscript was accepted for publication October 14, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Alberto Grammatica, MD, Department of Otorhinolaryngology-Head and Neck Surgery, University of Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy. E-mail: [email protected] DOI: 10.1002/lary.25025
Laryngoscope 00: Month 2014
auricle, with possible, even more rare, neuroma formation1 and neurotic excoriation.2 As it is increasingly important to associate complete disease removal with good functional and aesthetic outcome, surgeons dealing with parotidectomy started to preserve the posterior branch of the GAN minimize postoperative discomfort. In 1989, Brown and Ord stated that GAN preservation ensured better sensory recovery after parotidectomy,3 although in the last decades several reports with conflicting results have emerged.4–7 It is generally accepted that, within the first 6 months after surgery, there is no subjective and objective difference in terms of sensory deficits (anesthesia, paresthesia, extent of sensory loss areas) comparing subjects with the GAN preserved to those in whom it was resected.4–6,8,9 The average area of anesthesia seems to decrease every 3 months during the first year, with half of patients without anesthesia after 12 months even when complete transection of the nerve has been accomplished.10 By contrast, long-term outcomes after GAN preservation during parotid surgery in terms of sensitivity and quality of life (QoL) have been reported less consistently. The aim of this article was therefore to study the long-term sensory outcomes and QoL in patients with and without GAN preservation during parotid surgery for benign diseases. Grammatica et al.: GAN Preservation in Parotid Gland Surgery
1
Statistical analysis was performed using the SPSS statistical package (SPSS Inc., Chicago, IL): the Mann-Whitney U test was used to compare two categorical variables on one continuous variable. The Kruskal-Wallis test was applied to compare more than two categorical variables on one continuous variable. A v2 test or Fisher exact test was employed to compare categorical variables. Results were expressed as medians and interquartile ranges (IQR).
RESULTS
Fig. 1. Division of the auricle and periauricular areas innervated by the great auricular nerve. (1) Preauricular superior area. (2) Preauricular inferior area. (3) Angle of the mandible. (4) Helix and concha. (5) Lobule. (6) Retroauricular posterior area. (7) Mastoidlateral neck. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
MATERIALS AND METHODS Formal review and approval of the local ethics committee was obtained for this study. All patients included underwent surgery for benign parotid gland disorders at our institution between 2010 and 2012. All procedures were performed by two experienced salivary gland surgeons (P.N., R.L.O.D.Z.). At least 1 year after surgery, we retrospectively evaluated this subset of patients, dividing them into two groups according to intraoperative preservation or sectioning of the GAN. Exclusion criteria included bilateral parotid neoplasms, previous chemotherapy, radiation and/or head and neck surgery, neck or periauricular traumas, diabetes, and peripheral or central neurologic and/or psychiatric diseases. To perform the study, we divided the skin innervated by the GAN into seven areas (modified from the study of Min et al.5): 1) preauricular superior; 2) preauricular inferior; 3) angle of the mandible; 4) helix and concha; 5) lobule; 6) retroauricular; 7) mastoid-lateral neck (Fig. 1). A single trained author (F.M.), blinded on the nerve status (preserved or not), tested tactile and thermal sensitivities by stimulating each area of the GAN territory, first on the healthy nonoperated side (NS) and subsequently on the operated side (OS). Sensitivity was graded according to three classes: class 0, no sensitivity; class 1, reduced sensitivity; class 2, normal sensitivity. Tactile sensitivity was tested using a monofilament esthesiometer (Gillis W. Long Hensenıs Disease Center, Carville, LA) by pressing the center of each area with the tip of the filament. Cold sensitivity was tested by holding the finger of a latex glove full of iced water against the center of each area. Heat sensitivity was evaluated using the tip of a steel spoon warmed in boiling water for 5 seconds and left for 15 seconds at room temperature until reaching 45 C. Each area was consecutively tested twice for tactile, cold, and heat sensitivities and then compared to the contralateral area. At the end of clinical evaluation, a QoL questionnaire (validated from Patel et al.)11 was administered (Fig. 2).
Laryngoscope 00: Month 2014
2
Ninety patients underwent parotid gland surgery for benign diseases in the time interval considered. Fifty-five (61%) met the inclusion criteria for sensory testing and questionnaire compilation and were therefore enrolled in the study. Patients were divided in two groups: group A (N 5 42), in which the posterior branch of the GAN was preserved, and group B (N 5 13), in which the GAN was sacrificed. There were 28 (51%) females and 27 (49%) males. In group A there were 24 (57%) females and 18 (43%) males, whereas in group B there were 4 (31%) females and 9 (69%) males. The difference in gender distribution was not statistically significant (P 5.09). Age ranged from 24 to 79 years (median, 54 years; IQR, 42–61 years) in the entire cohort of patients. The ages of patients in group A varied between 24 and 79 years (median, 53 years; IQR, 42–62 years), whereas the ages of patients in group B ranged between 24 and 67 years (median, 57 years; IQR, 42–58 years), with no statistically significant difference (P 5.9). Definitive histopathology revealed 37 pleomorphic adenomas, 15 Warthin’s tumor, one myoepithelial neoplasm, one cyst, and one lipoma. All surgical interventions performed were defined as superficial parotidectomy, except for one case of pleomorphic adenoma of the deep lobe, which required subtotal parotidectomy. The period of time between surgery and sensory testing plus QoL questionnaire administration varied between 12 and 46 months (median, 24 months; IQR, 19–34 months). Median time between the two events was 27 months (IQR, 20–36 months) for Group A and 21 months (IQR, 19–32 months) for group B (P 5.2). In group A, normal tactile sensitivity of the OS was observed in 16.7% to 66.7% of patients depending on the area tested. Comparison of the degree of tactile sensitivity among OS and NS patients showed a significant difference for all the areas tested except area 1. Normal heat sensitivity varied from 11.9% to 73.8% of patients. All the areas tested showed significant difference between OS and NS. Normal cold sensitivity was preserved in 21.4% to 81% of patients. In this case, there was significant difference between the two sides except for area 1 (see Supporting Table I in the online version of this article). Overall sensitivity in this group of patients showed a trend toward improvement over time, although statistical analysis did not evidence significant differences among patients tested early (at 1 year after surgery) versus those evaluated later. In group B, normal tactile sensitivity in the OS was observed in 0% to 61.5% of patients depending on the area tested. Comparison of the degree of tactile Grammatica et al.: GAN Preservation in Parotid Gland Surgery
Fig. 2. Quality-of-life questionnaire validated by Patel et al.11
sensitivity among OS and NS patients showed significant differences for all the areas tested except area 1. Normal heat sensitivity varied from 0% to 53.8% of patients. All areas examined showed significant differences between OS and NS. Normal cold sensitivity was preserved in 7.7% to 76.9% of patients. There was significant difference between the two sides for cold sensitivity except for areas 1 and 4 (see Supporting Table II in the online version of this article). No trend of overall sensitivity recovery over time was evidenced in this group. Laryngoscope 00: Month 2014
Comparing the degree of sensitivity of OS of the two groups, the only significant difference was observed for area 5 for all types of sensitivity tested and for area 2 when considering heat sensitivity only (see Supporting Table III in the online version of this article). The median number of abnormal sensations investigated by the QoL questionnaire was 2 (IQR, 1–3) for group A and 2 (IQR, 2–3) for group B (P 5.4) (Table I). Analyzing as an independent variable the symptom “pain” by grouping under such a category all the Grammatica et al.: GAN Preservation in Parotid Gland Surgery
3
TABLE I. Type of Sensation Perceived in the Auricle and Periauricular Areas According to the Patel et al.11 Questionnaire. Type of Sensation
Group A
Group B
Stinging
11.9% (5)
23.1% (3)
0.4
Abnormal sensation Burning
57.1% (24) 4.8% (2)
61.5% (8) 15.4% (2)
1 0.2
Lack of feeling
59.5% (25)
76.9% (10)
0.3
Pain Lack of sensitivity
31% (13) 57.1% (24)
15.4% (2) 61.5% (8)
0.5 1
19% (8)
30.8% (4)
0.5
35.7% (15)
38.5% (5)
0.1
Discomfort Hypersensitivity
P Value
abnormal sensations (stinging, burning, hypersensitivity, and discomfort), we found these were more frequent in group B (76.9% of patients) than in group A (64.3%), although without any statistically significant difference (P 5.5). No statistically significant difference in the number of abnormal sensations was seen considering the time elapsed after surgery. By analyzing the postoperative symptoms that fall under the broad category of “pain” we observed how these tend to decrease over time in both groups, with a median time in group A not reporting pain of 34.2 months (IQR, 21.2–37.6 months) versus 23.5 months (IQR, 19.7–33.1 months) reporting pain (P 5.1), and a median time in group B not reporting pain of 31.8 months (IQR, 25.5–31.8 months) versus patients presenting pain of 20.1 months (IQR, 17–23.7 months), although not statically significant (P 5.2). The frequency, length, anxiety, extension, interference with daily routine, and worry about abnormal sensations were compared between the two groups. The only subjective sensation with a statistically significant difference was the mean hypoesthesic area extension, which turned out to be 2 (the size of a penny 5 20.3 mm) in group A and 4 (the size of a half dollar 5 30.6 mm) in group B (Table II).
DISCUSSION GAN is the largest superficial branch of the cervical plexus, providing sensory innervation to the skin overlying the parotid gland, external ear, and posterior auricular region. The cervical plexus gives rise to multiple
superficial cutaneous branches in addition to the GAN, including the lesser occipital, transverse cutaneous, and supraclavicular nerves, generally originating from the posterior aspect of the sternocleidomastoid muscle (SCM) into the so called Erb’s point. The GAN arises from the anterior branch of the second and third cervical nerves or directly from the latter,8 passes around the posterior margin of the SCM, and runs on the muscle’s lateral surface beneath the platysma, in close relationship with the external jugular vein (EJV). During parotidectomy, GAN identification can be facilitated by McKinney’s point located 6.5 cm below the external auditory meatus.3,4,12 Moreover, in 50% of cases, the EJV lies anterior and parallel to the nerve. The GAN then proceeds superiorly dividing into anterior and posterior branches at the inferior margin of the parotid gland. According to Ozturk et al.’s analysis, its most frequent branching point is located at the superior third of the SCM (type 1, 54%), followed by the middle third (type 2, 4%), and inferior third (type 3, 27%). The absence of branching was also observed (type 4, 15%).13 The anterior branch supplies the skin of the face and that overlying the parotid. The posterior branch may be single or multiple (up to 3) (Fig. 3) and innervates the skin over the mastoid process and the posterior aspect of the auricle, with a filament piercing the lobule and innervating it and the concha.6,14 GAN sparing is not crucial for the final outcome of parotid surgery, but its functional deficit can cause discomfort, with reported cases of traumatic lesions to the preauricular skin and lobule due to the failure in recognizing pain and other tactile sensations.11,15 Sparing the posterior branch of the GAN requires an additional 5 to 10 minutes of surgical time, does not significantly affect the overall duration of surgery,8 and should be accomplished whenever it does not modify the surgical approach and/or affect its oncologic radicality. In the English literature, several reports on GAN preservation have been published, with most focused on short-term results,5,7,8,16 whereas only limited data are available on long-term (more than 1 year) outcomes.9,17 Moreover, some controversial issues have been reported concerning the utility of posterior branch sparing in terms of better postoperative sensory outcomes. When considering 1-year follow-up, several studies have demonstrated that posterior branch GAN preservation allows reduced sensory loss and quicker recovery of
TABLE II. Feelings Experienced by Patients According to the Patel et al. Questionnaire. Type of Sensation
Range
Group A, Median (IQR)
Group B, Median (IQR)
P Value
Frequency Length
0–6 0–6
3.5 (1–6) 6 (1–6)
4 (4–6) 6 (0–6)
.2 .7
Anxiety
0–6
1.5 (0–2)
2 (0–4)
.6
Extension Interference to daily routine
0–4 0–6
2 (1–3) 0 (0–1)
4 (3–4) 0 (0–1)
.003 .9
Worry about these abnormal feelings
0–6
1 (0–2)
0 (0–1)
.4
IQR 5 interquartile range.
Laryngoscope 00: Month 2014
4
Grammatica et al.: GAN Preservation in Parotid Gland Surgery
Fig. 3. Intraoperative view of right parotid gland. Note the great auricular nerve main trunk dividing into three posterior branches lying on the posterolateral aspect of the gland (the anterior branch had already been divided). [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
sensation, reaching normal levels at 6 to 12 months postoperatively.4–7 Biglioli et al. advocated that sensation recovery, especially for the angle of the mandible, may depend on the presence of a GAN accessory anterior branch that divides before entering the parotid or on collateral innervations by the mandibular branch of the trigeminal nerve.8 Moreover, Ryan and Free indicated that when the posterior branch is preserved, collateral innervation by the auriculotemporal, lesser occipital, and transverse cutaneous nerves, as well as the patient’s adaptation, may play an adjunctive role.17 Despite these observations, in daily practice, when evaluating patients treated by parotid gland surgery, it is common that one of the main discomforts reported is sensory loss at the level of the periparotid skin areas and auricle, regardless of the intraoperative management of the GAN and time elapsed after surgery. In the present study, we found that, even after more than 1 year following surgery, OS tactile sensitivity in both groups was persistently different compared to the NS. In group A, all areas suffered some type of sensory deficit. The difference between OS and NS approached statistical significance for area 1 (preauricular superior) and was significant for the other six areas, with a more pronounced sensory loss for area 2 (preauricular inferior), area 3 (angle of the mandible), area 6 (retroauricular), and area 7 (mastoid/lateral neck). The same is true for group B, in which area 1 showed a borderline nearsignificant difference, whereas the other six showed a statistically significant difference (more pronounced for areas 2, 3, 5, and 6), with worse outcomes in terms of sensory loss. Only area 5 (lobule) was significantly different considering the OS of the two groups in terms of tactile sensitivity (69% of patients in group B reporting reduced sensitivity vs. 45% of group A; P 5.04). This is related to the fact that the GAN posterior branch, usually spared during gland mobilization, directly supplies the lobule. These data differ from the retrospective study of Christensen and Jacobsen comparing long-term outcomes in patients with and without GAN preservation, where no difference in sensation of the cheek and manLaryngoscope 00: Month 2014
dible was found, but with less sensory dysfunction of the earlobe.15 It is interesting how, in both groups, various degrees of tactile and thermal anesthesia are reported, with the retroauricular (area 6) and angle of the mandible (area 3) being the most represented in group A, and the angle of the mandible (area 3), helix/concha (area 4), lobule (area 5), and mastoid/lateral neck (area 7) mainly involved in group B. In the study by Ryan and Free,17 after a follow-up of 4 to 5 years, 26% of patients with GAN preservation showed anesthesia of the lobule and 26% at the level of the mandibular angle. These data are worse than those observed in our study, because in group A, no anesthesia was present in the lobule, whereas only 7% reported anesthesia at the level of the angle of mandible. On the other hand, Biglioli et al. reported that in patients in whom the GAN was resected, after 8 years of follow-up, 90% of patients reported anesthesia at the angle of the mandible.8 These results are substantially worse than our results, where 23% of patients in group B reported anesthesia in this specific area, followed by the helix/concha in 15% of cases. To the best of our knowledge, this is the first study addressing long-term thermal sensitivity. Regarding heat stimulation, as for tactile, the area and degree of thermal sensitivity are directly related to GAN preservation. The preauricular inferior (area 2) and lobule (area 5) showed significant differences between the two groups, whereas the angle of the mandible (area 3) resulted in more severe sensory loss despite nerve preservation. Regarding cold stimulation, only lobule (area 5) showed a difference between the two groups. Considering thermal stimulation, Biglioli et al. found that roughly one-third of patients with the GAN preserved had some type of cold sensitivity deficit, with no anesthesia reported, whereas for heat stimulation, about 90% had anesthesia located in the ear lobe and mandibular regions.8 Although thermal sensitivity is difficult to study due to its more subjective perception and lower number of receptors compared to tactile sensitivity, we found that, overall, despite GAN posterior branch preservation, some grade of hypoesthesia is usually present. In our QoL survey, no significant difference in terms of number of sensations was seen between the two groups, and the only feeling that was different concerned the extent of the area affected; in particular, this was twice as prominent in group B. None of the other items showed any statistically significant difference, especially for pain and other similar disturbing feelings. This means that nerve sectioning does not severely impact either the everyday life of patients or the type of abnormal sensations perceived; this may be partly due to the patient’s adaptation during the postoperative course, as previously hypothesized by Ryan and Free.17 Our results are comparable to those reported by Patel et al., where 57% of patients with the GAN sacrifice reported at least one abnormal sensation, although the mean number decreased in time, from 2.3 in the first year to 0.2 after 5 years. Moreover, among symptomatic patients, 77% reported little or no concern by their Grammatica et al.: GAN Preservation in Parotid Gland Surgery
5
abnormal feelings, and 90% reported no interference with daily activities. These data suggest that, in spite of nerve sectioning, symptoms tend to improve after 1 year from surgery and progressively lessen in severity and impact on QoL.11 One potential flaw of our study is represented by its nonrandomized design that caused a slight imbalance between the number of patients in groups A and B, in favor of GAN preservation, which is, at the moment, our preferred surgical philosophy. It is still speculative and partly unclear how, in patients with the GAN preserved and despite a lengthy period after surgery, some do not recover normal nerve function or have better outcomes than those described in the present report. This may be explained, at least in part, by the scar tissue formed after surgery, possibly encasing the preserved posterior branch, and preventing axonal regeneration due to the diminished vascularization of these tissues.
CONCLUSION Lack of sensitivity and other sensory dysfunctions are commonly reported by patients of both groups without significant differences. The angle of the mandible, lobule, and preauricular areas showed differences in terms of tactile and thermal sensitivities with slightly better outcomes in the GAN-preserved group. However, no long-term difference in terms of QoL between patients with or without GAN sacrifice after parotid surgery was observed. After at least 1 year from surgery, the impact of GAN sacrifice in terms of QoL appears to be tolerable. Comparison between the two groups showed a statistically significant difference only in extension of the discomforted area (reduced in group A), with no difference of the other subjective items. Even though the importance of GAN preservation during superficial parotidectomy should not be overemphasized during preoperative counseling and not pursued when-
Laryngoscope 00: Month 2014
6
ever potentially affecting oncologic radicality, sparing its branches may be beneficial to minimize postoperative discomfort, anesthesia, and dysesthesia, with no increase in postoperative pain.
BIBLIOGRAPHY 1. Moss CE, Johnston CJ, Whear NM. Amputation neuroma of the great auricular nerve after operations of the parotid gland. Br J Oral Maxillofacial Surg 2000;38:537–538. 2. Fardy MJ. Neurotic excoriations complicating superficial parotidectomy—a case report. Br J Oral Maxillofac Surg 1993;31:41–42. 3. Brown JS, Ord RA. Preserving the great auricular nerve in parotid surgery. Br J Oral Maxillofac Surg 1989;27:459–466. 4. Hui Y, Wong DSY, Wong LY, Ho WK, Wei WI. A prospective controlled double-blind trial of great auricular nerve preservation at parotidectomy. Am J Surg 2003;185:574–579. 5. Min JH, Lee SH, Lee YS, et al. It is necessary to preserve the posterior branch of the great auricular nerve in parotidectomy? Otolaryngol Head Neck Surg 2007;137:636–641. 6. Porter MJ, Wood SJ. Preservation of the great auricular nerve during parotidectomy. Clin Otolaryngol 1997;22:251–253. 7. Vieira MB, Maia AF, Ribeiro JC. Randomized prospective study of the validity of the great auricular nerve preservation in parotidectomy. Arch Otolaryngol Head Neck Surg 2002;128:1191–1195. 8. Biglioli F, D’Orto O, Bozzetti A, Brusati R. Function of the great auricular nerve following surgery for benign parotid disorders. J Craniomaxillofac Surg 2002;30:308–317. 9. Suen DTK, Chow TL, Lam CYW, Wong S, Lam SH. Sensation recovery improved by great auricular nerve preservation in parotidectomy: a prospective double-blind study. ANZ J Surg 2007;77:374–376. 10. Ryan WR, Fee WE Jr. Great auricular nerve morbidity after nerve sacrifice during parotidectomy. Arch Otolaryngol Head Neck Surg 2006;132: 642–649. 11. Patel N, Har-El G, Rosenfeld R. Quality of life after great auricular nerve sacrifice during parotidectomy. Arch Otolaryngol Head Neck Surg 2001; 127:884–888. 12. McKinney P, Katrana DJ. Prevention of injury to the great auricular nerve during rhytidectomy. Plast Reconstr Surg 1980;66:675–679. 13. Ozturk CN, Ozturk C, Huettern F, Drake RL, Zins JE. A failsafe method to avoid injury to the great auricular nerve. Aesthetic Surg J 2014;34: 16–21. 14. Zohar Y, Siegal A, Siegal G, Halpern M, Levy B, Gal R. The great auricular nerve; does it penetrate the parotid gland? An anatomical and microscopical study. J Craniomaxillofac Surg 2002;30:318–321. 15. Christensen NR, Jacobsen SD. Parotidectomy. Preserving the posterior branch of the great auricular nerve. J Laryngol Otol 1997;111:556–559. 16. Yokoshima K, Nakamizo M, Fukumoto A, et al. Significance of preserving the posterior branch of the great auricular nerve in parotidectomy. J Nippon Med Sch 2004;5:323–327. 17. Ryan WR, Fee WE. Long-term great auricular nerve morbidity after sacrifice during parotidectomy. Laryngoscope 2009;119:1140–1146.
Grammatica et al.: GAN Preservation in Parotid Gland Surgery
Rhinological and Otological Society, Inc.
Great Auricular Nerve Preservation in Parotid Gland Surgery: Long-term Outcomes Alberto Grammatica, MD; Pietro Perotti, MD; Francesco Mancini, MD; Anna Bozzola, MD; Cesare Piazza, MD; Piero Nicolai, MD; Luca Oscar Redaelli de Zinis, MD Objectives/Hypothesis: To assess sensory outcomes and quality of life (QoL) in two groups of patients with and without great auricular nerve (GAN) preservation during parotidectomy. Study Design: Retrospective chart review. Methods: The posterior branch of the GAN was preserved in 42 patients (group A) and sacrificed in 13 (group B). Tactile, heat, and cold sensitivities were investigated by dividing GAN territory into seven areas. Comparisons between operated (OS) and nonoperated sides (NS) within each group, and between the OS of the two groups were made. The QoL questionnaire was administered. Results: In group A, normal tactile, heat, and cold sensitivities ranged from 16.7% to 66.7%, 11.9% to 73.8%, and 21.4% to 81%, respectively, in different OS areas. Significant differences between OS and NS were found, except for the preauricular superior area. In group B, normal tactile, heat, and cold sensitivities ranged from 0% to 61.5%, 0% to 53.8%, and 7.7% to 76.9%, respectively, in different OS areas. Significant differences between OS and NS were found except for the preauricular superior (tactile sensitivity), and preauricular superior and helix/concha areas (cold sensitivity). Comparing the OS tactile and thermic sensitivities between the two groups, only the lobule area showed differences. The preauricular inferior area was different only for heat. The QoL questionnaire showed different hypoesthesia extension between the two groups. All other items were comparable. Conclusions: Sensory deficits are commonly reported despite GAN preservation. Lobule and preauricular inferior areas showed differences in terms of tactile and thermic sensitivities, with better outcomes in group A. QoL seems tolerable despite GAN sacrifice. Key Words: Parotid gland, surgery, great auricular nerve, nerve sacrifice, nerve preservation, quality of life. Level of Evidence: 4 Laryngoscope, 00:000–000, 2014
INTRODUCTION Parotidectomy is a common, well-standardized, head and neck surgical procedure. Facial nerve injury, Frey’s syndrome, and aesthetic deformities are the most common complications following parotidectomy, whereas important, but less well-studied, sequelae include hypoanesthesia, paresthesia, and discomfort to the auricle and periauricular areas due to neurotmesis or neuroapraxia of the great auricular nerve (GAN). In fact, during parotidectomy, the GAN can be either mobilized and preserved or sacrificed to expose the posterior and inferior aspects of the gland. Expected consequences of nerve transection are numbness of the
Additional Supporting Information may be found in the online version of this article. From the Department of Otorhinolaryngology–Head and Neck Surgery, University of Brescia, Brescia, Italy. Editor’s Note: This Manuscript was accepted for publication October 14, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Alberto Grammatica, MD, Department of Otorhinolaryngology-Head and Neck Surgery, University of Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy. E-mail: [email protected] DOI: 10.1002/lary.25025
Laryngoscope 00: Month 2014
auricle, with possible, even more rare, neuroma formation1 and neurotic excoriation.2 As it is increasingly important to associate complete disease removal with good functional and aesthetic outcome, surgeons dealing with parotidectomy started to preserve the posterior branch of the GAN minimize postoperative discomfort. In 1989, Brown and Ord stated that GAN preservation ensured better sensory recovery after parotidectomy,3 although in the last decades several reports with conflicting results have emerged.4–7 It is generally accepted that, within the first 6 months after surgery, there is no subjective and objective difference in terms of sensory deficits (anesthesia, paresthesia, extent of sensory loss areas) comparing subjects with the GAN preserved to those in whom it was resected.4–6,8,9 The average area of anesthesia seems to decrease every 3 months during the first year, with half of patients without anesthesia after 12 months even when complete transection of the nerve has been accomplished.10 By contrast, long-term outcomes after GAN preservation during parotid surgery in terms of sensitivity and quality of life (QoL) have been reported less consistently. The aim of this article was therefore to study the long-term sensory outcomes and QoL in patients with and without GAN preservation during parotid surgery for benign diseases. Grammatica et al.: GAN Preservation in Parotid Gland Surgery
1
Statistical analysis was performed using the SPSS statistical package (SPSS Inc., Chicago, IL): the Mann-Whitney U test was used to compare two categorical variables on one continuous variable. The Kruskal-Wallis test was applied to compare more than two categorical variables on one continuous variable. A v2 test or Fisher exact test was employed to compare categorical variables. Results were expressed as medians and interquartile ranges (IQR).
RESULTS
Fig. 1. Division of the auricle and periauricular areas innervated by the great auricular nerve. (1) Preauricular superior area. (2) Preauricular inferior area. (3) Angle of the mandible. (4) Helix and concha. (5) Lobule. (6) Retroauricular posterior area. (7) Mastoidlateral neck. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
MATERIALS AND METHODS Formal review and approval of the local ethics committee was obtained for this study. All patients included underwent surgery for benign parotid gland disorders at our institution between 2010 and 2012. All procedures were performed by two experienced salivary gland surgeons (P.N., R.L.O.D.Z.). At least 1 year after surgery, we retrospectively evaluated this subset of patients, dividing them into two groups according to intraoperative preservation or sectioning of the GAN. Exclusion criteria included bilateral parotid neoplasms, previous chemotherapy, radiation and/or head and neck surgery, neck or periauricular traumas, diabetes, and peripheral or central neurologic and/or psychiatric diseases. To perform the study, we divided the skin innervated by the GAN into seven areas (modified from the study of Min et al.5): 1) preauricular superior; 2) preauricular inferior; 3) angle of the mandible; 4) helix and concha; 5) lobule; 6) retroauricular; 7) mastoid-lateral neck (Fig. 1). A single trained author (F.M.), blinded on the nerve status (preserved or not), tested tactile and thermal sensitivities by stimulating each area of the GAN territory, first on the healthy nonoperated side (NS) and subsequently on the operated side (OS). Sensitivity was graded according to three classes: class 0, no sensitivity; class 1, reduced sensitivity; class 2, normal sensitivity. Tactile sensitivity was tested using a monofilament esthesiometer (Gillis W. Long Hensenıs Disease Center, Carville, LA) by pressing the center of each area with the tip of the filament. Cold sensitivity was tested by holding the finger of a latex glove full of iced water against the center of each area. Heat sensitivity was evaluated using the tip of a steel spoon warmed in boiling water for 5 seconds and left for 15 seconds at room temperature until reaching 45 C. Each area was consecutively tested twice for tactile, cold, and heat sensitivities and then compared to the contralateral area. At the end of clinical evaluation, a QoL questionnaire (validated from Patel et al.)11 was administered (Fig. 2).
Laryngoscope 00: Month 2014
2
Ninety patients underwent parotid gland surgery for benign diseases in the time interval considered. Fifty-five (61%) met the inclusion criteria for sensory testing and questionnaire compilation and were therefore enrolled in the study. Patients were divided in two groups: group A (N 5 42), in which the posterior branch of the GAN was preserved, and group B (N 5 13), in which the GAN was sacrificed. There were 28 (51%) females and 27 (49%) males. In group A there were 24 (57%) females and 18 (43%) males, whereas in group B there were 4 (31%) females and 9 (69%) males. The difference in gender distribution was not statistically significant (P 5.09). Age ranged from 24 to 79 years (median, 54 years; IQR, 42–61 years) in the entire cohort of patients. The ages of patients in group A varied between 24 and 79 years (median, 53 years; IQR, 42–62 years), whereas the ages of patients in group B ranged between 24 and 67 years (median, 57 years; IQR, 42–58 years), with no statistically significant difference (P 5.9). Definitive histopathology revealed 37 pleomorphic adenomas, 15 Warthin’s tumor, one myoepithelial neoplasm, one cyst, and one lipoma. All surgical interventions performed were defined as superficial parotidectomy, except for one case of pleomorphic adenoma of the deep lobe, which required subtotal parotidectomy. The period of time between surgery and sensory testing plus QoL questionnaire administration varied between 12 and 46 months (median, 24 months; IQR, 19–34 months). Median time between the two events was 27 months (IQR, 20–36 months) for Group A and 21 months (IQR, 19–32 months) for group B (P 5.2). In group A, normal tactile sensitivity of the OS was observed in 16.7% to 66.7% of patients depending on the area tested. Comparison of the degree of tactile sensitivity among OS and NS patients showed a significant difference for all the areas tested except area 1. Normal heat sensitivity varied from 11.9% to 73.8% of patients. All the areas tested showed significant difference between OS and NS. Normal cold sensitivity was preserved in 21.4% to 81% of patients. In this case, there was significant difference between the two sides except for area 1 (see Supporting Table I in the online version of this article). Overall sensitivity in this group of patients showed a trend toward improvement over time, although statistical analysis did not evidence significant differences among patients tested early (at 1 year after surgery) versus those evaluated later. In group B, normal tactile sensitivity in the OS was observed in 0% to 61.5% of patients depending on the area tested. Comparison of the degree of tactile Grammatica et al.: GAN Preservation in Parotid Gland Surgery
Fig. 2. Quality-of-life questionnaire validated by Patel et al.11
sensitivity among OS and NS patients showed significant differences for all the areas tested except area 1. Normal heat sensitivity varied from 0% to 53.8% of patients. All areas examined showed significant differences between OS and NS. Normal cold sensitivity was preserved in 7.7% to 76.9% of patients. There was significant difference between the two sides for cold sensitivity except for areas 1 and 4 (see Supporting Table II in the online version of this article). No trend of overall sensitivity recovery over time was evidenced in this group. Laryngoscope 00: Month 2014
Comparing the degree of sensitivity of OS of the two groups, the only significant difference was observed for area 5 for all types of sensitivity tested and for area 2 when considering heat sensitivity only (see Supporting Table III in the online version of this article). The median number of abnormal sensations investigated by the QoL questionnaire was 2 (IQR, 1–3) for group A and 2 (IQR, 2–3) for group B (P 5.4) (Table I). Analyzing as an independent variable the symptom “pain” by grouping under such a category all the Grammatica et al.: GAN Preservation in Parotid Gland Surgery
3
TABLE I. Type of Sensation Perceived in the Auricle and Periauricular Areas According to the Patel et al.11 Questionnaire. Type of Sensation
Group A
Group B
Stinging
11.9% (5)
23.1% (3)
0.4
Abnormal sensation Burning
57.1% (24) 4.8% (2)
61.5% (8) 15.4% (2)
1 0.2
Lack of feeling
59.5% (25)
76.9% (10)
0.3
Pain Lack of sensitivity
31% (13) 57.1% (24)
15.4% (2) 61.5% (8)
0.5 1
19% (8)
30.8% (4)
0.5
35.7% (15)
38.5% (5)
0.1
Discomfort Hypersensitivity
P Value
abnormal sensations (stinging, burning, hypersensitivity, and discomfort), we found these were more frequent in group B (76.9% of patients) than in group A (64.3%), although without any statistically significant difference (P 5.5). No statistically significant difference in the number of abnormal sensations was seen considering the time elapsed after surgery. By analyzing the postoperative symptoms that fall under the broad category of “pain” we observed how these tend to decrease over time in both groups, with a median time in group A not reporting pain of 34.2 months (IQR, 21.2–37.6 months) versus 23.5 months (IQR, 19.7–33.1 months) reporting pain (P 5.1), and a median time in group B not reporting pain of 31.8 months (IQR, 25.5–31.8 months) versus patients presenting pain of 20.1 months (IQR, 17–23.7 months), although not statically significant (P 5.2). The frequency, length, anxiety, extension, interference with daily routine, and worry about abnormal sensations were compared between the two groups. The only subjective sensation with a statistically significant difference was the mean hypoesthesic area extension, which turned out to be 2 (the size of a penny 5 20.3 mm) in group A and 4 (the size of a half dollar 5 30.6 mm) in group B (Table II).
DISCUSSION GAN is the largest superficial branch of the cervical plexus, providing sensory innervation to the skin overlying the parotid gland, external ear, and posterior auricular region. The cervical plexus gives rise to multiple
superficial cutaneous branches in addition to the GAN, including the lesser occipital, transverse cutaneous, and supraclavicular nerves, generally originating from the posterior aspect of the sternocleidomastoid muscle (SCM) into the so called Erb’s point. The GAN arises from the anterior branch of the second and third cervical nerves or directly from the latter,8 passes around the posterior margin of the SCM, and runs on the muscle’s lateral surface beneath the platysma, in close relationship with the external jugular vein (EJV). During parotidectomy, GAN identification can be facilitated by McKinney’s point located 6.5 cm below the external auditory meatus.3,4,12 Moreover, in 50% of cases, the EJV lies anterior and parallel to the nerve. The GAN then proceeds superiorly dividing into anterior and posterior branches at the inferior margin of the parotid gland. According to Ozturk et al.’s analysis, its most frequent branching point is located at the superior third of the SCM (type 1, 54%), followed by the middle third (type 2, 4%), and inferior third (type 3, 27%). The absence of branching was also observed (type 4, 15%).13 The anterior branch supplies the skin of the face and that overlying the parotid. The posterior branch may be single or multiple (up to 3) (Fig. 3) and innervates the skin over the mastoid process and the posterior aspect of the auricle, with a filament piercing the lobule and innervating it and the concha.6,14 GAN sparing is not crucial for the final outcome of parotid surgery, but its functional deficit can cause discomfort, with reported cases of traumatic lesions to the preauricular skin and lobule due to the failure in recognizing pain and other tactile sensations.11,15 Sparing the posterior branch of the GAN requires an additional 5 to 10 minutes of surgical time, does not significantly affect the overall duration of surgery,8 and should be accomplished whenever it does not modify the surgical approach and/or affect its oncologic radicality. In the English literature, several reports on GAN preservation have been published, with most focused on short-term results,5,7,8,16 whereas only limited data are available on long-term (more than 1 year) outcomes.9,17 Moreover, some controversial issues have been reported concerning the utility of posterior branch sparing in terms of better postoperative sensory outcomes. When considering 1-year follow-up, several studies have demonstrated that posterior branch GAN preservation allows reduced sensory loss and quicker recovery of
TABLE II. Feelings Experienced by Patients According to the Patel et al. Questionnaire. Type of Sensation
Range
Group A, Median (IQR)
Group B, Median (IQR)
P Value
Frequency Length
0–6 0–6
3.5 (1–6) 6 (1–6)
4 (4–6) 6 (0–6)
.2 .7
Anxiety
0–6
1.5 (0–2)
2 (0–4)
.6
Extension Interference to daily routine
0–4 0–6
2 (1–3) 0 (0–1)
4 (3–4) 0 (0–1)
.003 .9
Worry about these abnormal feelings
0–6
1 (0–2)
0 (0–1)
.4
IQR 5 interquartile range.
Laryngoscope 00: Month 2014
4
Grammatica et al.: GAN Preservation in Parotid Gland Surgery
Fig. 3. Intraoperative view of right parotid gland. Note the great auricular nerve main trunk dividing into three posterior branches lying on the posterolateral aspect of the gland (the anterior branch had already been divided). [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
sensation, reaching normal levels at 6 to 12 months postoperatively.4–7 Biglioli et al. advocated that sensation recovery, especially for the angle of the mandible, may depend on the presence of a GAN accessory anterior branch that divides before entering the parotid or on collateral innervations by the mandibular branch of the trigeminal nerve.8 Moreover, Ryan and Free indicated that when the posterior branch is preserved, collateral innervation by the auriculotemporal, lesser occipital, and transverse cutaneous nerves, as well as the patient’s adaptation, may play an adjunctive role.17 Despite these observations, in daily practice, when evaluating patients treated by parotid gland surgery, it is common that one of the main discomforts reported is sensory loss at the level of the periparotid skin areas and auricle, regardless of the intraoperative management of the GAN and time elapsed after surgery. In the present study, we found that, even after more than 1 year following surgery, OS tactile sensitivity in both groups was persistently different compared to the NS. In group A, all areas suffered some type of sensory deficit. The difference between OS and NS approached statistical significance for area 1 (preauricular superior) and was significant for the other six areas, with a more pronounced sensory loss for area 2 (preauricular inferior), area 3 (angle of the mandible), area 6 (retroauricular), and area 7 (mastoid/lateral neck). The same is true for group B, in which area 1 showed a borderline nearsignificant difference, whereas the other six showed a statistically significant difference (more pronounced for areas 2, 3, 5, and 6), with worse outcomes in terms of sensory loss. Only area 5 (lobule) was significantly different considering the OS of the two groups in terms of tactile sensitivity (69% of patients in group B reporting reduced sensitivity vs. 45% of group A; P 5.04). This is related to the fact that the GAN posterior branch, usually spared during gland mobilization, directly supplies the lobule. These data differ from the retrospective study of Christensen and Jacobsen comparing long-term outcomes in patients with and without GAN preservation, where no difference in sensation of the cheek and manLaryngoscope 00: Month 2014
dible was found, but with less sensory dysfunction of the earlobe.15 It is interesting how, in both groups, various degrees of tactile and thermal anesthesia are reported, with the retroauricular (area 6) and angle of the mandible (area 3) being the most represented in group A, and the angle of the mandible (area 3), helix/concha (area 4), lobule (area 5), and mastoid/lateral neck (area 7) mainly involved in group B. In the study by Ryan and Free,17 after a follow-up of 4 to 5 years, 26% of patients with GAN preservation showed anesthesia of the lobule and 26% at the level of the mandibular angle. These data are worse than those observed in our study, because in group A, no anesthesia was present in the lobule, whereas only 7% reported anesthesia at the level of the angle of mandible. On the other hand, Biglioli et al. reported that in patients in whom the GAN was resected, after 8 years of follow-up, 90% of patients reported anesthesia at the angle of the mandible.8 These results are substantially worse than our results, where 23% of patients in group B reported anesthesia in this specific area, followed by the helix/concha in 15% of cases. To the best of our knowledge, this is the first study addressing long-term thermal sensitivity. Regarding heat stimulation, as for tactile, the area and degree of thermal sensitivity are directly related to GAN preservation. The preauricular inferior (area 2) and lobule (area 5) showed significant differences between the two groups, whereas the angle of the mandible (area 3) resulted in more severe sensory loss despite nerve preservation. Regarding cold stimulation, only lobule (area 5) showed a difference between the two groups. Considering thermal stimulation, Biglioli et al. found that roughly one-third of patients with the GAN preserved had some type of cold sensitivity deficit, with no anesthesia reported, whereas for heat stimulation, about 90% had anesthesia located in the ear lobe and mandibular regions.8 Although thermal sensitivity is difficult to study due to its more subjective perception and lower number of receptors compared to tactile sensitivity, we found that, overall, despite GAN posterior branch preservation, some grade of hypoesthesia is usually present. In our QoL survey, no significant difference in terms of number of sensations was seen between the two groups, and the only feeling that was different concerned the extent of the area affected; in particular, this was twice as prominent in group B. None of the other items showed any statistically significant difference, especially for pain and other similar disturbing feelings. This means that nerve sectioning does not severely impact either the everyday life of patients or the type of abnormal sensations perceived; this may be partly due to the patient’s adaptation during the postoperative course, as previously hypothesized by Ryan and Free.17 Our results are comparable to those reported by Patel et al., where 57% of patients with the GAN sacrifice reported at least one abnormal sensation, although the mean number decreased in time, from 2.3 in the first year to 0.2 after 5 years. Moreover, among symptomatic patients, 77% reported little or no concern by their Grammatica et al.: GAN Preservation in Parotid Gland Surgery
5
abnormal feelings, and 90% reported no interference with daily activities. These data suggest that, in spite of nerve sectioning, symptoms tend to improve after 1 year from surgery and progressively lessen in severity and impact on QoL.11 One potential flaw of our study is represented by its nonrandomized design that caused a slight imbalance between the number of patients in groups A and B, in favor of GAN preservation, which is, at the moment, our preferred surgical philosophy. It is still speculative and partly unclear how, in patients with the GAN preserved and despite a lengthy period after surgery, some do not recover normal nerve function or have better outcomes than those described in the present report. This may be explained, at least in part, by the scar tissue formed after surgery, possibly encasing the preserved posterior branch, and preventing axonal regeneration due to the diminished vascularization of these tissues.
CONCLUSION Lack of sensitivity and other sensory dysfunctions are commonly reported by patients of both groups without significant differences. The angle of the mandible, lobule, and preauricular areas showed differences in terms of tactile and thermal sensitivities with slightly better outcomes in the GAN-preserved group. However, no long-term difference in terms of QoL between patients with or without GAN sacrifice after parotid surgery was observed. After at least 1 year from surgery, the impact of GAN sacrifice in terms of QoL appears to be tolerable. Comparison between the two groups showed a statistically significant difference only in extension of the discomforted area (reduced in group A), with no difference of the other subjective items. Even though the importance of GAN preservation during superficial parotidectomy should not be overemphasized during preoperative counseling and not pursued when-
Laryngoscope 00: Month 2014
6
ever potentially affecting oncologic radicality, sparing its branches may be beneficial to minimize postoperative discomfort, anesthesia, and dysesthesia, with no increase in postoperative pain.
BIBLIOGRAPHY 1. Moss CE, Johnston CJ, Whear NM. Amputation neuroma of the great auricular nerve after operations of the parotid gland. Br J Oral Maxillofacial Surg 2000;38:537–538. 2. Fardy MJ. Neurotic excoriations complicating superficial parotidectomy—a case report. Br J Oral Maxillofac Surg 1993;31:41–42. 3. Brown JS, Ord RA. Preserving the great auricular nerve in parotid surgery. Br J Oral Maxillofac Surg 1989;27:459–466. 4. Hui Y, Wong DSY, Wong LY, Ho WK, Wei WI. A prospective controlled double-blind trial of great auricular nerve preservation at parotidectomy. Am J Surg 2003;185:574–579. 5. Min JH, Lee SH, Lee YS, et al. It is necessary to preserve the posterior branch of the great auricular nerve in parotidectomy? Otolaryngol Head Neck Surg 2007;137:636–641. 6. Porter MJ, Wood SJ. Preservation of the great auricular nerve during parotidectomy. Clin Otolaryngol 1997;22:251–253. 7. Vieira MB, Maia AF, Ribeiro JC. Randomized prospective study of the validity of the great auricular nerve preservation in parotidectomy. Arch Otolaryngol Head Neck Surg 2002;128:1191–1195. 8. Biglioli F, D’Orto O, Bozzetti A, Brusati R. Function of the great auricular nerve following surgery for benign parotid disorders. J Craniomaxillofac Surg 2002;30:308–317. 9. Suen DTK, Chow TL, Lam CYW, Wong S, Lam SH. Sensation recovery improved by great auricular nerve preservation in parotidectomy: a prospective double-blind study. ANZ J Surg 2007;77:374–376. 10. Ryan WR, Fee WE Jr. Great auricular nerve morbidity after nerve sacrifice during parotidectomy. Arch Otolaryngol Head Neck Surg 2006;132: 642–649. 11. Patel N, Har-El G, Rosenfeld R. Quality of life after great auricular nerve sacrifice during parotidectomy. Arch Otolaryngol Head Neck Surg 2001; 127:884–888. 12. McKinney P, Katrana DJ. Prevention of injury to the great auricular nerve during rhytidectomy. Plast Reconstr Surg 1980;66:675–679. 13. Ozturk CN, Ozturk C, Huettern F, Drake RL, Zins JE. A failsafe method to avoid injury to the great auricular nerve. Aesthetic Surg J 2014;34: 16–21. 14. Zohar Y, Siegal A, Siegal G, Halpern M, Levy B, Gal R. The great auricular nerve; does it penetrate the parotid gland? An anatomical and microscopical study. J Craniomaxillofac Surg 2002;30:318–321. 15. Christensen NR, Jacobsen SD. Parotidectomy. Preserving the posterior branch of the great auricular nerve. J Laryngol Otol 1997;111:556–559. 16. Yokoshima K, Nakamizo M, Fukumoto A, et al. Significance of preserving the posterior branch of the great auricular nerve in parotidectomy. J Nippon Med Sch 2004;5:323–327. 17. Ryan WR, Fee WE. Long-term great auricular nerve morbidity after sacrifice during parotidectomy. Laryngoscope 2009;119:1140–1146.
Grammatica et al.: GAN Preservation in Parotid Gland Surgery
