- Email: [email protected]
An objective assessment of the advantages of retrograde parotidectomy
An objective assessment of the advantages of retrograde parotidectomy NEIL BHATTACHARYYA,
MD, FACS,
MARC E. RICHARDSON,
REEGT,
OBJECTIVE: This study was undertaken to determine whether the retrograde parotidectomy approach is more efficient than standard anterograde parotidectomy without compromise of surgical effectiveness. METHODS: A retrospective analysis of patients undergoing parotidectomy was conducted. Cases were divided into those undergoing retrograde facial nerve dissection and those undergoing standard anterograde facial nerve dissection. From the review of medical records, standard demographic information, surgical time, histopathology, estimated blood loss, and use of facial nerve monitoring were determined. Pathology was reviewed to determine the size of the overall resection specimen as well as the size of the lesion excised and margin status. Postoperative complications were also recorded. Statistical comparisons were conducted between these 2 approaches for these clinical variables such as surgical time, blood loss, tumor margin status and relative volume of tissue removed during parotidectomy. RESULTS: 45 patients undergoing parotidectomy met inclusion criteria. The average patient age was 50.8 years with a female preponderance (73%). There were 19 standard parotidectomies and 26 retrograde approaches. Compared to standard parotidectomy, retrograde parotidectomy consumed less operative time (3.2 versus 1.8 hours, respectively), decreased intraoperative blood loss (67.9 cc versus 40.3 cc, respectively), and resulted in less removal of normal parotid tissue (volume of normal parotid tissue removed in excess of tumor: 23.0 cc versus 6.0 cc, respectively). No significant difference in surgical margin status was noted between anterograde and retrograde parotidectomy (P ⴝ 0.452). CONCLUSIONS: In appropriately selected cases,
From the Division of Otolaryngology, Brigham and Women’s Hospital and the Department of Otology and Laryngology, Harvard Medical School (Dr Bhattacharyya)and the Department of Anesthesia, Brigham and Women’s Hospital (Drs Richardson and Gugino), Boston, MA. Reprint requests: Neil Bhattacharyya, MD, FACS, Division of Otolaryngology, 333 Longwood Ave., Boston, MA 02115; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2004.03.012
392
and LAVERNE D. GUGINO,
MD, PHD,
Boston, Massachusetts
compared with standard anterograde parotidectomy, retrograde parotidectomy is more efficient and spares normal parotid tissue without compromising surgical margins. Facial nerve monitoring provides a useful adjunct for retrograde dissection. (Otolaryngol Head Neck Surg 2004;131:392-6.)
T
umors of the parotid gland account for 80% of all salivary gland neoplasms.1 Parotid neoplasms may arise from salivary tissue of the parotid gland or from intraparotid lymph nodes, and rarely from neural structures within the parotid gland.2 The treatment of choice for most parotid lesions is surgical excision.3 Although some debate exists in the literature as to the need for preoperative imaging and/or fine needle aspiration biopsy prior to parotidectomy, most surgeons agree that standard superficial (supraneural) parotidectomy is effective treatment for most parotid lesions involving the superficial lobe. Most commonly, this approach involves identification of the main trunk of the facial nerve posteriorly in the gland. Several techniques and methods have been described to anatomically locate the facial nerve. These include identification of the nerve via its relationship with the tympanomastoid suture, the tragal pointer or the posterior belly of the digastric muscle. Each of these methods allows for an anterograde dissection of the facial nerve elevating the superficial lobe of the parotid gland along with the tumor from the nerve. However, depending on the size and location of the tumor, the extent of the dissection and the extent of salivary gland tissue removal may be more determined by the branching pattern and location of the facial nerve than by the size and extent of the tumor. This more extensive dissection is intuitively likely to consume more operative time, resects a larger portion of the superficial lobe possibly resulting in a larger contour defect, and may not be necessary to adequately resect the lesion. The retrograde facial nerve dissection approach for parotidectomy is technically more challenging and more time-consuming at the onset of surgery. However, for certain lesions, it may be more advantageous. First, for inferior or tail lesions of the parotid, typically only 1 or perhaps 2 branches of the facial nerve may require dissection, sparing exposure and potential trauma to the main trunk. Second, for lesions that are anteriorly based in the parotid gland, a smaller field of dissection may be
Otolaryngology– Head and Neck Surgery Volume 131 Number 4
BHATTACHARYYA et al 393
Table 1. Clinical parameters and outcome variables for anterograde versus retrograde parotidectomy Variables
Standard parotidectomy
Retrograde parotidectomy
52.8 68.4 89.5 0.6
49.4 76.9 80.8 0.9
0.386 0.734 0.681 0.516
3.2 65.8 34.0 7.7 26.3
1.8 48.2 13.9 4.8 9.1
⬍0.001 0.001 ⬍0.001 0.017 ⬍0.001
Clinical variable Age Gender (% female) Use of facial nerve monitoring (%) Mean number of lymph nodes removed Outcome variable* Surgical time (hours) Surgical blood loss (cc) Overall volume of pathology specimen (cc) Volume of lesion within specimen (cc) Volume of normal parotid tissue removed (cc)
P value
*Mean values are given for outcome variables.
realized with this approach. Therefore, in appropriate instances, a retrograde approach may conserve normal parotid tissue, reduce operative time, and result in a more favorable cosmetic outcome. This study was conducted to objectively determine whether retrograde parotidectomy reduces operative time and conserves normal parotid tissue while still providing adequate surgical margin for tumor resection without additional facial nerve complications. METHODS Consecutive adult patients undergoing parotidectomy with a retrograde facial nerve dissection approach (cases) were identified from a prospectively maintained log of operative cases. A second group of patients undergoing standard parotidectomy with main trunk facial nerve identification served as the historical controls. The first author (N.B.) performed all of the surgical procedures. Medical records for patients in each group were systematically reviewed and clinical data were extracted and entered into a database. From the medical record, standard demographic information and the results from preoperative evaluation including fineneedle aspiration were identified. Operative notes were reviewed to determine use of facial nerve monitoring, type of facial nerve dissection, operative time, and estimated blood loss during surgery. From the pathology reports, the histopathology of the parotid lesion, margin status, three-dimensional size of the overall pathologic specimen, and size of the tumor within the specimen were determined. Postoperative complications such as facial nerve weakness, and wound complications such as sialocele formation, hematoma, and wound infection were also recorded. Exclusion criteria included planned total parotidectomy for known highgrade malignancy, parotid biopsy for salivary tissue for diagnostic purposes (i.e., rule out Sju¨ gren’s syndrome), and revision parotidectomy.
Comparisons were conducted between the cases (retrograde dissection group) and historical control group (standard anterograde dissection group) for the outcome measures of surgical time, surgical margin status, and volume of normal parotid tissue removed relative to the size of the primary tumor. Comparison was conducted between the 2 groups for surgical margin status as well. Statistical comparisons were conducted with the paired Student’s t-test or Pearson’s Chi-square, where appropriate with significance set at P ⬍ 0.05. RESULTS A total of 45 cases met the inclusion criteria after 6 cases were excluded from analysis (excluded cases: 2 simple parotid biopsies, 2 revision parotidectomies, and 2 cases of parotidectomy for chronic sialoadenitis). There were a total of 19 standard anterograde parotidectomies performed with identification of the main trunk of the facial nerve followed by anterograde facial nerve dissection constituting the control group. There were a total of 26 retrograde facial nerve dissection approaches. Descriptive clinical data are summarized in Table 1 for the 2 groups. For the standard parotidectomy approach group, there were 13 females with an average age of 52.3 years. For the retrograde parotidectomy approach group, there were 20 females with an average age of 49.4 years. Fine-needle aspiration cytology was obtained in 25 (78%) of the cases, and was nondiagnostic in 9 cases. The most common suggested neoplasm on fine-needle aspiration was pleomorphic adenoma totaling 15 cases. The findings on final histopathologic analysis are given in Table 2. Facial nerve monitoring was utilized for 17 of 19 of the standard parotidectomy cases and for 20 of 25 of the retrograde parotidectomy cases. As noted in Table 1, standard parotidectomy resulted in a significantly larger size of the overall pa-
Otolaryngology– Head and Neck Surgery October 2004
394 BHATTACHARYYA et al
Table 2. Distribution of final tissue histopathology results Histopathology Pleomorphic adenoma Mucoepidermoid carcinoma Lipoma Lymphoepithelioma Sialocele/sialolithiasis Lymphadenopathy Lymphoma Warthin’s tumor Adenocarcinoma Apocrine hidrocystoma Basal cell adenoma Sarcoidosis Squamous cell carcinoma Undifferentiated carcinoma Total
N
%
20 4 3 3 3 2 2 2 1 1 1 1 1 1 45
44.4 8.9 6.7 6.7 6.7 4.4 4.4 4.4 2.2 2.2 2.2 2.2 2.2 2.2 100.0
thology specimen submitted to pathology as compared with retrograde parotidectomy. In addition, the size of the lesion within the parotid gland tissue excised was also somewhat larger for patients undergoing standard parotidectomy. However, patients undergoing standard parotidectomy also had a much larger volume of normal tissue (that is, overall size of pathology specimen in cubic centimeters minus overall size of lesion within the specimen in cubic centimeters) removed as compared with patients undergoing retrograde parotidectomy. This difference represents normal salivary tissue removed in the course of extirpating the primary parotid lesion. Because the standard parotidectomy patients also had higher through tumor volumes, analysis of variance was conducted comparing volumes of normal parotid tissue removed between the 2 groups incorporating volume of the pathologic lesion as a covariate. This analysis demonstrated that even with lesion size as covariate (significance of covariate in the model P ⫽ 0.073), the retrograde parotidectomy approach resulted in smaller volumes of normal parotid tissue excised (P ⫽ 0.001). Therefore, retrograde parotidectomy exhibited statistically significant clinical advantages in terms of decreasing operative time by 1.4 hours, decreasing intraoperative blood loss by 27 cc, and significantly reducing the volume of normal parotid tissue excised. In the standard parotidectomy group, there was 1 case of temporary facial paralysis that resolved within 6 weeks postoperatively and no cases of permanent paralysis. Similarly, in the retrograde group, there were no cases of temporary or permanent paralysis of the facial nerve or of its noncervical (nonplatysmal) branches. One patient in the retrograde group had a wound complication with keloid formation. There were no cases in either group of postoperative
hematoma, sialocele requiring drainage, or wound infection. Among the standard parotidectomy group, there were 13 negative margins, 5 focally close margins, and 1 positive margin, whereas among the retrograde group, there were 20 negative margins, 5 focally close margins, and 0 positive margins. There was no statistically significant difference in margin status between the standard parotidectomy and retrograde parotidectomy groups (P ⫽ 0.452, Chi-square). DISCUSSION The impetus for refocusing attention on the retrograde parotidectomy approach arose from our experience with significant cosmetic contour defects after standard superficial parotidectomy. Specifically, 2 patients who had small lesions of the parotid (1 with a small lymphoepithelioma and 1 with a small intraparotid lymph node with suspicious cells identified on fine-needle aspiration cytology) suffered from significant contour defects after parotidectomy. We therefore re-examined the need for the standard approach to the main trunk of the facial nerve and forward dissection toward the tumor. Therefore, we began to more purposefully attempt retrograde dissection for appropriate cases. In subsequent cases, we found that especially for inferior and more anteriorly located lesions a retrograde approach substantially reduced the extent of dissection, the extent of normal parotid gland sacrificed and the operative time. In the last half of the 20th century, surgeons have traditionally performed complete superficial parotidectomy, i.e., removal of the lateral lobe of the parotid gland, for benign lesions of the parotid gland involving the superficial lobe. Several authors have previously cited advantages of complete superficial parotidectomy including the versatility of this approach in the face of unclear or unknown pathology and uniformity of the approach and ease for standardization of teaching.4 In addition, maintaining a large cuff of normal parotid tissue around the neoplasm may decrease the chance of tumor spillage or capsular disruption. However, with mounting interest in minimally invasive surgery and a trend toward conservative resections, some debate has surfaced in the literature as to the need for complete superficial parotidectomy. Several authors have recommended more conservative subtotal or partial superficial parotidectomy as a treatment for benign lesions and even select low-grade malignancies.4,5 Smaller resections may have advantages including smaller contour defects after surgery, decreased operative time and potentially decreased rates of temporary and/or permanent facial nerve paralysis, especially of the main trunk. However, these advantages have not yet been system-
Otolaryngology– Head and Neck Surgery Volume 131 Number 4
atically demonstrated in the literature. Evidence does suggest that the incidence of Frey’s syndrome is decreased with smaller partial resections.6 With respect to tumor recurrence, several series have demonstrated that partial superficial parotidectomy likely results in similar recurrence rates ranging from 0% to 6% when compared with complete superficial parotidectomy.5,7 The use of the retrograde dissection approach for parotidectomy is a natural extension of these more conservative subtotal or partial superficial parotidectomies. Even if a partial parotidectomy is planned, if the facial nerve is localized and dissected from its main trunk with the anterograde approach, it is still likely that excess normal parotid tissue will require dissection before the tumor is encountered in the anterograde approach. Thus, especially for anterior, inferior, or tail lesions, a significant amount normal parotid tissue will be elevated and therefore sacrificed as part of the resection at the outset of the facial nerve dissection. This proximal dissection of the parotid gland is often not necessary to adequately resect many superficial lobe lesions. The surgical technique for retrograde parotidectomy has been described by others and our technique will only briefly be summarized here. After the induction of general anesthesia without long-term neuromuscular blockade, bipolar facial nerve monitoring electrodes are placed in the orbicularis oris and orbicularis oculi muscles on the ipsilateral side of the parotidectomy. A standard preauricular parotidectomy incision is marked out in the event that conversion to standard superficial parotidectomy with main trunk facial nerve dissection. However, only the inferior posterior aspect of this incision is initially carried out based on location of the parotid lesion. An anteriorly superiorly based skin flap was developed in the subcutaneous plane. At this point, the lesion is palpated and the suitability of the inferior approach with retrograde facial nerve dissection is assessed. Working anteriorly and slightly below the inferior border, the marginal mandibular and/or platysmal branches of the facial nerve are identified just outside the free edge of the parotid gland and dissected in a retrograde manner elevating the tail and inferior portion of the parotid gland from the facial nerve. Once the lesion has been elevated such that the superior portion of the lesion is elevated from the interior branches of the facial nerve, the capsule of the parotid gland is sequentially sectioned, and the lesion is delivered inferiorly. Attachments to the sternocleidomastoid muscle are sectioned and the lesion with the surrounding cuff of normal parotid tissue is delivered. Often, branches of the great auricular nerve can be easily reflected posteriorly avoiding sectioning. Throughout the dissection and sectioning of parotid gland tissue, the facial nerve
BHATTACHARYYA et al 395
stimulating electrode is employed to help ensure the absence of any facial nerve branches or significant rami communicantes within the parotid tissue to be sectioned. In our experience, the use of the facial nerve monitor with a stimulating electrode has also accelerated identification of the distal branches of the facial nerve at the beginning of the dissection. As the individual anterior branches of the facial nerve are followed posteriorly, identification of communicating branches is also facilitated with the facial nerve stimulator and nerve monitor. The current data reinforce the previously held notion that partial parotidectomy with a retrograde approach conserves more normal parotid tissue. Importantly, the current data also indicate that with the retrograde approach surgical margin status is not compromised. Very often during parotidectomy surgery, 1 or more of the tumor margins is defined by the contact of the tumor with facial nerve branches.5 Therefore, identifying the individual facial nerve branches that may control the rate limiting step as the deep “tumor margin” with the retrograde approach may offer additional advantages with respect to operative time. For example, for tail lesions, identifying the platysmal or marginal mandibular branch of the facial nerve at the outset may determine the superior resection margin and obviate the need to dissect more superior branches or the trunk of the facial nerve. One disadvantage of the current study is that it was conducted retrospectively. Therefore, the possibility for selection bias exists and that smaller, more favorable lesions may have been unknowingly shunted towards a retrograde parotidectomy approach. Therefore, it seems prudent to recommend that the retrograde parotidectomy approach be considered for straightforward cases that are highly likely to represent benign disease in a favorable location. Also, when planning retrograde partial parotidectomy, fine-needle aspiration results may prove helpful in excluding cases that may not be ideal candidates for this approach. Such cases would include high-grade primary parotid malignancies, those cases that might require concurrent neck dissection, and potentially metastatic lesions.8 One potential drawback to the retrograde facial nerve dissection approach is that during the dissection, multiple communicating branches within the anterior parotid gland may be identified and unnecessarily followed. In these instances, the facial nerve monitor has proven helpful in separating minor communicating rami from true branches or divisions. Often times, true branches or divisions will stimulate on lower amperage settings than the communicating rami will. However, this still may be a technically challenging portion of the procedure requiring sound judgment and experience. Finally, surgeons
396 BHATTACHARYYA et al
should maintain a relatively low threshold for conversion to a standard anterograde parotidectomy approach when tumor or operative conditions dictate. Such conditions may arise when the tumor encases one or more branches of the facial nerve anteriorly or when it becomes evident that the posterior tumor margin will abut the main trunk of the facial nerve. REFERENCES 1. Eisele DW, Johns ME. Salivary gland neoplasms. In: Bailey BJ, editor. Head and neck surgery-Otolaryngology. Philadelphia: Lippincott Williams & Wilkins; 2001. pp. 1279-97. 2. Johns ME, Goldsmith MM. Incidence, diagnosis, and classification of salivary gland tumors. Oncology (Huntingt) 1989;3(Part 1):47-56; discussion 56-62.
Otolaryngology– Head and Neck Surgery October 2004
3. Fee WE, Tran LE. Evaluation of a patient with a parotid tumor. Arch Otolaryngol Head Neck Surg 2003;129:937-8. 4. Myssiorek D. Removal of the inferior half of the superficial lobe is sufficient to treat pleomorphic adenoma in the tail of the parotid gland. Arch Otolaryngol Head Neck Surg 1999;125: 1164-5. 5. Witt RL. Facial nerve function after partial superficial parotidectomy: an 11 year review (1987-1997). Otolaryngol Head Neck Surg 1999;121:210-3. 6. Wenmo C, Spandow O, Emgard P, et al. pleomorphic adenomas of the parotid gland: superficial parotidectomy or limited excision? J Laryngol Otol 1988;102:603-5. 7. Yoo GH, Eisele DW, Askin FB, et al. Warthin’s tumor: a 40 year experience at the Johns Hopkins Hospital. Laryngoscope 1994; 104:799-803. 8. Bhattacharyya N, Fried MP. Nodal metastasis in major salivary gland cancer: predictive factors and impact on survival. Arch Otolaryngol Head Neck Surg 2002;128:904-8.
MD, FACS,
MARC E. RICHARDSON,
REEGT,
OBJECTIVE: This study was undertaken to determine whether the retrograde parotidectomy approach is more efficient than standard anterograde parotidectomy without compromise of surgical effectiveness. METHODS: A retrospective analysis of patients undergoing parotidectomy was conducted. Cases were divided into those undergoing retrograde facial nerve dissection and those undergoing standard anterograde facial nerve dissection. From the review of medical records, standard demographic information, surgical time, histopathology, estimated blood loss, and use of facial nerve monitoring were determined. Pathology was reviewed to determine the size of the overall resection specimen as well as the size of the lesion excised and margin status. Postoperative complications were also recorded. Statistical comparisons were conducted between these 2 approaches for these clinical variables such as surgical time, blood loss, tumor margin status and relative volume of tissue removed during parotidectomy. RESULTS: 45 patients undergoing parotidectomy met inclusion criteria. The average patient age was 50.8 years with a female preponderance (73%). There were 19 standard parotidectomies and 26 retrograde approaches. Compared to standard parotidectomy, retrograde parotidectomy consumed less operative time (3.2 versus 1.8 hours, respectively), decreased intraoperative blood loss (67.9 cc versus 40.3 cc, respectively), and resulted in less removal of normal parotid tissue (volume of normal parotid tissue removed in excess of tumor: 23.0 cc versus 6.0 cc, respectively). No significant difference in surgical margin status was noted between anterograde and retrograde parotidectomy (P ⴝ 0.452). CONCLUSIONS: In appropriately selected cases,
From the Division of Otolaryngology, Brigham and Women’s Hospital and the Department of Otology and Laryngology, Harvard Medical School (Dr Bhattacharyya)and the Department of Anesthesia, Brigham and Women’s Hospital (Drs Richardson and Gugino), Boston, MA. Reprint requests: Neil Bhattacharyya, MD, FACS, Division of Otolaryngology, 333 Longwood Ave., Boston, MA 02115; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2004.03.012
392
and LAVERNE D. GUGINO,
MD, PHD,
Boston, Massachusetts
compared with standard anterograde parotidectomy, retrograde parotidectomy is more efficient and spares normal parotid tissue without compromising surgical margins. Facial nerve monitoring provides a useful adjunct for retrograde dissection. (Otolaryngol Head Neck Surg 2004;131:392-6.)
T
umors of the parotid gland account for 80% of all salivary gland neoplasms.1 Parotid neoplasms may arise from salivary tissue of the parotid gland or from intraparotid lymph nodes, and rarely from neural structures within the parotid gland.2 The treatment of choice for most parotid lesions is surgical excision.3 Although some debate exists in the literature as to the need for preoperative imaging and/or fine needle aspiration biopsy prior to parotidectomy, most surgeons agree that standard superficial (supraneural) parotidectomy is effective treatment for most parotid lesions involving the superficial lobe. Most commonly, this approach involves identification of the main trunk of the facial nerve posteriorly in the gland. Several techniques and methods have been described to anatomically locate the facial nerve. These include identification of the nerve via its relationship with the tympanomastoid suture, the tragal pointer or the posterior belly of the digastric muscle. Each of these methods allows for an anterograde dissection of the facial nerve elevating the superficial lobe of the parotid gland along with the tumor from the nerve. However, depending on the size and location of the tumor, the extent of the dissection and the extent of salivary gland tissue removal may be more determined by the branching pattern and location of the facial nerve than by the size and extent of the tumor. This more extensive dissection is intuitively likely to consume more operative time, resects a larger portion of the superficial lobe possibly resulting in a larger contour defect, and may not be necessary to adequately resect the lesion. The retrograde facial nerve dissection approach for parotidectomy is technically more challenging and more time-consuming at the onset of surgery. However, for certain lesions, it may be more advantageous. First, for inferior or tail lesions of the parotid, typically only 1 or perhaps 2 branches of the facial nerve may require dissection, sparing exposure and potential trauma to the main trunk. Second, for lesions that are anteriorly based in the parotid gland, a smaller field of dissection may be
Otolaryngology– Head and Neck Surgery Volume 131 Number 4
BHATTACHARYYA et al 393
Table 1. Clinical parameters and outcome variables for anterograde versus retrograde parotidectomy Variables
Standard parotidectomy
Retrograde parotidectomy
52.8 68.4 89.5 0.6
49.4 76.9 80.8 0.9
0.386 0.734 0.681 0.516
3.2 65.8 34.0 7.7 26.3
1.8 48.2 13.9 4.8 9.1
⬍0.001 0.001 ⬍0.001 0.017 ⬍0.001
Clinical variable Age Gender (% female) Use of facial nerve monitoring (%) Mean number of lymph nodes removed Outcome variable* Surgical time (hours) Surgical blood loss (cc) Overall volume of pathology specimen (cc) Volume of lesion within specimen (cc) Volume of normal parotid tissue removed (cc)
P value
*Mean values are given for outcome variables.
realized with this approach. Therefore, in appropriate instances, a retrograde approach may conserve normal parotid tissue, reduce operative time, and result in a more favorable cosmetic outcome. This study was conducted to objectively determine whether retrograde parotidectomy reduces operative time and conserves normal parotid tissue while still providing adequate surgical margin for tumor resection without additional facial nerve complications. METHODS Consecutive adult patients undergoing parotidectomy with a retrograde facial nerve dissection approach (cases) were identified from a prospectively maintained log of operative cases. A second group of patients undergoing standard parotidectomy with main trunk facial nerve identification served as the historical controls. The first author (N.B.) performed all of the surgical procedures. Medical records for patients in each group were systematically reviewed and clinical data were extracted and entered into a database. From the medical record, standard demographic information and the results from preoperative evaluation including fineneedle aspiration were identified. Operative notes were reviewed to determine use of facial nerve monitoring, type of facial nerve dissection, operative time, and estimated blood loss during surgery. From the pathology reports, the histopathology of the parotid lesion, margin status, three-dimensional size of the overall pathologic specimen, and size of the tumor within the specimen were determined. Postoperative complications such as facial nerve weakness, and wound complications such as sialocele formation, hematoma, and wound infection were also recorded. Exclusion criteria included planned total parotidectomy for known highgrade malignancy, parotid biopsy for salivary tissue for diagnostic purposes (i.e., rule out Sju¨ gren’s syndrome), and revision parotidectomy.
Comparisons were conducted between the cases (retrograde dissection group) and historical control group (standard anterograde dissection group) for the outcome measures of surgical time, surgical margin status, and volume of normal parotid tissue removed relative to the size of the primary tumor. Comparison was conducted between the 2 groups for surgical margin status as well. Statistical comparisons were conducted with the paired Student’s t-test or Pearson’s Chi-square, where appropriate with significance set at P ⬍ 0.05. RESULTS A total of 45 cases met the inclusion criteria after 6 cases were excluded from analysis (excluded cases: 2 simple parotid biopsies, 2 revision parotidectomies, and 2 cases of parotidectomy for chronic sialoadenitis). There were a total of 19 standard anterograde parotidectomies performed with identification of the main trunk of the facial nerve followed by anterograde facial nerve dissection constituting the control group. There were a total of 26 retrograde facial nerve dissection approaches. Descriptive clinical data are summarized in Table 1 for the 2 groups. For the standard parotidectomy approach group, there were 13 females with an average age of 52.3 years. For the retrograde parotidectomy approach group, there were 20 females with an average age of 49.4 years. Fine-needle aspiration cytology was obtained in 25 (78%) of the cases, and was nondiagnostic in 9 cases. The most common suggested neoplasm on fine-needle aspiration was pleomorphic adenoma totaling 15 cases. The findings on final histopathologic analysis are given in Table 2. Facial nerve monitoring was utilized for 17 of 19 of the standard parotidectomy cases and for 20 of 25 of the retrograde parotidectomy cases. As noted in Table 1, standard parotidectomy resulted in a significantly larger size of the overall pa-
Otolaryngology– Head and Neck Surgery October 2004
394 BHATTACHARYYA et al
Table 2. Distribution of final tissue histopathology results Histopathology Pleomorphic adenoma Mucoepidermoid carcinoma Lipoma Lymphoepithelioma Sialocele/sialolithiasis Lymphadenopathy Lymphoma Warthin’s tumor Adenocarcinoma Apocrine hidrocystoma Basal cell adenoma Sarcoidosis Squamous cell carcinoma Undifferentiated carcinoma Total
N
%
20 4 3 3 3 2 2 2 1 1 1 1 1 1 45
44.4 8.9 6.7 6.7 6.7 4.4 4.4 4.4 2.2 2.2 2.2 2.2 2.2 2.2 100.0
thology specimen submitted to pathology as compared with retrograde parotidectomy. In addition, the size of the lesion within the parotid gland tissue excised was also somewhat larger for patients undergoing standard parotidectomy. However, patients undergoing standard parotidectomy also had a much larger volume of normal tissue (that is, overall size of pathology specimen in cubic centimeters minus overall size of lesion within the specimen in cubic centimeters) removed as compared with patients undergoing retrograde parotidectomy. This difference represents normal salivary tissue removed in the course of extirpating the primary parotid lesion. Because the standard parotidectomy patients also had higher through tumor volumes, analysis of variance was conducted comparing volumes of normal parotid tissue removed between the 2 groups incorporating volume of the pathologic lesion as a covariate. This analysis demonstrated that even with lesion size as covariate (significance of covariate in the model P ⫽ 0.073), the retrograde parotidectomy approach resulted in smaller volumes of normal parotid tissue excised (P ⫽ 0.001). Therefore, retrograde parotidectomy exhibited statistically significant clinical advantages in terms of decreasing operative time by 1.4 hours, decreasing intraoperative blood loss by 27 cc, and significantly reducing the volume of normal parotid tissue excised. In the standard parotidectomy group, there was 1 case of temporary facial paralysis that resolved within 6 weeks postoperatively and no cases of permanent paralysis. Similarly, in the retrograde group, there were no cases of temporary or permanent paralysis of the facial nerve or of its noncervical (nonplatysmal) branches. One patient in the retrograde group had a wound complication with keloid formation. There were no cases in either group of postoperative
hematoma, sialocele requiring drainage, or wound infection. Among the standard parotidectomy group, there were 13 negative margins, 5 focally close margins, and 1 positive margin, whereas among the retrograde group, there were 20 negative margins, 5 focally close margins, and 0 positive margins. There was no statistically significant difference in margin status between the standard parotidectomy and retrograde parotidectomy groups (P ⫽ 0.452, Chi-square). DISCUSSION The impetus for refocusing attention on the retrograde parotidectomy approach arose from our experience with significant cosmetic contour defects after standard superficial parotidectomy. Specifically, 2 patients who had small lesions of the parotid (1 with a small lymphoepithelioma and 1 with a small intraparotid lymph node with suspicious cells identified on fine-needle aspiration cytology) suffered from significant contour defects after parotidectomy. We therefore re-examined the need for the standard approach to the main trunk of the facial nerve and forward dissection toward the tumor. Therefore, we began to more purposefully attempt retrograde dissection for appropriate cases. In subsequent cases, we found that especially for inferior and more anteriorly located lesions a retrograde approach substantially reduced the extent of dissection, the extent of normal parotid gland sacrificed and the operative time. In the last half of the 20th century, surgeons have traditionally performed complete superficial parotidectomy, i.e., removal of the lateral lobe of the parotid gland, for benign lesions of the parotid gland involving the superficial lobe. Several authors have previously cited advantages of complete superficial parotidectomy including the versatility of this approach in the face of unclear or unknown pathology and uniformity of the approach and ease for standardization of teaching.4 In addition, maintaining a large cuff of normal parotid tissue around the neoplasm may decrease the chance of tumor spillage or capsular disruption. However, with mounting interest in minimally invasive surgery and a trend toward conservative resections, some debate has surfaced in the literature as to the need for complete superficial parotidectomy. Several authors have recommended more conservative subtotal or partial superficial parotidectomy as a treatment for benign lesions and even select low-grade malignancies.4,5 Smaller resections may have advantages including smaller contour defects after surgery, decreased operative time and potentially decreased rates of temporary and/or permanent facial nerve paralysis, especially of the main trunk. However, these advantages have not yet been system-
Otolaryngology– Head and Neck Surgery Volume 131 Number 4
atically demonstrated in the literature. Evidence does suggest that the incidence of Frey’s syndrome is decreased with smaller partial resections.6 With respect to tumor recurrence, several series have demonstrated that partial superficial parotidectomy likely results in similar recurrence rates ranging from 0% to 6% when compared with complete superficial parotidectomy.5,7 The use of the retrograde dissection approach for parotidectomy is a natural extension of these more conservative subtotal or partial superficial parotidectomies. Even if a partial parotidectomy is planned, if the facial nerve is localized and dissected from its main trunk with the anterograde approach, it is still likely that excess normal parotid tissue will require dissection before the tumor is encountered in the anterograde approach. Thus, especially for anterior, inferior, or tail lesions, a significant amount normal parotid tissue will be elevated and therefore sacrificed as part of the resection at the outset of the facial nerve dissection. This proximal dissection of the parotid gland is often not necessary to adequately resect many superficial lobe lesions. The surgical technique for retrograde parotidectomy has been described by others and our technique will only briefly be summarized here. After the induction of general anesthesia without long-term neuromuscular blockade, bipolar facial nerve monitoring electrodes are placed in the orbicularis oris and orbicularis oculi muscles on the ipsilateral side of the parotidectomy. A standard preauricular parotidectomy incision is marked out in the event that conversion to standard superficial parotidectomy with main trunk facial nerve dissection. However, only the inferior posterior aspect of this incision is initially carried out based on location of the parotid lesion. An anteriorly superiorly based skin flap was developed in the subcutaneous plane. At this point, the lesion is palpated and the suitability of the inferior approach with retrograde facial nerve dissection is assessed. Working anteriorly and slightly below the inferior border, the marginal mandibular and/or platysmal branches of the facial nerve are identified just outside the free edge of the parotid gland and dissected in a retrograde manner elevating the tail and inferior portion of the parotid gland from the facial nerve. Once the lesion has been elevated such that the superior portion of the lesion is elevated from the interior branches of the facial nerve, the capsule of the parotid gland is sequentially sectioned, and the lesion is delivered inferiorly. Attachments to the sternocleidomastoid muscle are sectioned and the lesion with the surrounding cuff of normal parotid tissue is delivered. Often, branches of the great auricular nerve can be easily reflected posteriorly avoiding sectioning. Throughout the dissection and sectioning of parotid gland tissue, the facial nerve
BHATTACHARYYA et al 395
stimulating electrode is employed to help ensure the absence of any facial nerve branches or significant rami communicantes within the parotid tissue to be sectioned. In our experience, the use of the facial nerve monitor with a stimulating electrode has also accelerated identification of the distal branches of the facial nerve at the beginning of the dissection. As the individual anterior branches of the facial nerve are followed posteriorly, identification of communicating branches is also facilitated with the facial nerve stimulator and nerve monitor. The current data reinforce the previously held notion that partial parotidectomy with a retrograde approach conserves more normal parotid tissue. Importantly, the current data also indicate that with the retrograde approach surgical margin status is not compromised. Very often during parotidectomy surgery, 1 or more of the tumor margins is defined by the contact of the tumor with facial nerve branches.5 Therefore, identifying the individual facial nerve branches that may control the rate limiting step as the deep “tumor margin” with the retrograde approach may offer additional advantages with respect to operative time. For example, for tail lesions, identifying the platysmal or marginal mandibular branch of the facial nerve at the outset may determine the superior resection margin and obviate the need to dissect more superior branches or the trunk of the facial nerve. One disadvantage of the current study is that it was conducted retrospectively. Therefore, the possibility for selection bias exists and that smaller, more favorable lesions may have been unknowingly shunted towards a retrograde parotidectomy approach. Therefore, it seems prudent to recommend that the retrograde parotidectomy approach be considered for straightforward cases that are highly likely to represent benign disease in a favorable location. Also, when planning retrograde partial parotidectomy, fine-needle aspiration results may prove helpful in excluding cases that may not be ideal candidates for this approach. Such cases would include high-grade primary parotid malignancies, those cases that might require concurrent neck dissection, and potentially metastatic lesions.8 One potential drawback to the retrograde facial nerve dissection approach is that during the dissection, multiple communicating branches within the anterior parotid gland may be identified and unnecessarily followed. In these instances, the facial nerve monitor has proven helpful in separating minor communicating rami from true branches or divisions. Often times, true branches or divisions will stimulate on lower amperage settings than the communicating rami will. However, this still may be a technically challenging portion of the procedure requiring sound judgment and experience. Finally, surgeons
396 BHATTACHARYYA et al
should maintain a relatively low threshold for conversion to a standard anterograde parotidectomy approach when tumor or operative conditions dictate. Such conditions may arise when the tumor encases one or more branches of the facial nerve anteriorly or when it becomes evident that the posterior tumor margin will abut the main trunk of the facial nerve. REFERENCES 1. Eisele DW, Johns ME. Salivary gland neoplasms. In: Bailey BJ, editor. Head and neck surgery-Otolaryngology. Philadelphia: Lippincott Williams & Wilkins; 2001. pp. 1279-97. 2. Johns ME, Goldsmith MM. Incidence, diagnosis, and classification of salivary gland tumors. Oncology (Huntingt) 1989;3(Part 1):47-56; discussion 56-62.
Otolaryngology– Head and Neck Surgery October 2004
3. Fee WE, Tran LE. Evaluation of a patient with a parotid tumor. Arch Otolaryngol Head Neck Surg 2003;129:937-8. 4. Myssiorek D. Removal of the inferior half of the superficial lobe is sufficient to treat pleomorphic adenoma in the tail of the parotid gland. Arch Otolaryngol Head Neck Surg 1999;125: 1164-5. 5. Witt RL. Facial nerve function after partial superficial parotidectomy: an 11 year review (1987-1997). Otolaryngol Head Neck Surg 1999;121:210-3. 6. Wenmo C, Spandow O, Emgard P, et al. pleomorphic adenomas of the parotid gland: superficial parotidectomy or limited excision? J Laryngol Otol 1988;102:603-5. 7. Yoo GH, Eisele DW, Askin FB, et al. Warthin’s tumor: a 40 year experience at the Johns Hopkins Hospital. Laryngoscope 1994; 104:799-803. 8. Bhattacharyya N, Fried MP. Nodal metastasis in major salivary gland cancer: predictive factors and impact on survival. Arch Otolaryngol Head Neck Surg 2002;128:904-8.