C V 2011 The American Laryngological,

The Laryngoscope

Rhinological and Otological Society, Inc.

When to Manage Level V in Head and Neck Carcinoma?

Baris Naiboglu, MD; Ugur Karapinar, MD; Amit Agrawal, MD; David E. Schuller, MD; Enver Ozer, MD 
Objectives/Hypothesis: As superselective neck dissection strategy is gaining popularity to minimize postoperative morbidity and better life quality, we investigated the metastatic nodal status of level V neck lymph node group for head and neck squamous cell carcinoma in various primary sites. We have also aimed to display the impact of involvement of other nodal groups on level V. Study Design: Retrospective review of histopathologic examination of case series at a comprehensive cancer center. Methods: The study group was composed of 107 patients who underwent a type of neck dissection including level V among 243 patients. The impact of primary site and metastatic nodal status of other levels on metastasis to level V involvement were evaluated. Results: The most common primary tumor site was oropharynx (n 43), followed by oral cavity (n 32), larynx (n 16), carcinoma of unknown primary (n 10), and hypopharynx (n 6). General pathologic N positivity for all levels was 78.3% (76 of 97) when 10 carcinoma of unknown primary patients were excluded. Level V was involved in 13 of 107 (12.1%) patients. Level V was not involved in any patient when the other levels were not involved (0 of 21). Even when considering only N patients, the ratio of N positivity for level V is still <20% (13 of 86, 15.1%). Conclusions: Because level V was not involved in any patient when the other levels were not involved, it might be reasonable to preserve level V especially in clinically and intraoperatively N0 patients. Key Words: Head and neck carcinoma, lymphatic metastasis, level V, spinal accessory nerve. Level of Evidence: 4. Laryngoscope, 121:545547, 2011

INTRODUCTION
Level V encompasses all lymph nodes within the posterior triangle in which there exists three different lymphatic pathways: nodes located along the spinal accessory nerve (SAN), along the transverse cervical artery, and above the lateral two-thirds of the clavicle. Neck metastasis is the most important factor for prognosis of head and neck squamous cell carcinoma (HNSCC).1 The radical neck dissection described by Crile in 1906 is the starting point of surgical treatment of metastatic neck disease in the literature.2 It involves removal of all nodal groups besides nonlymphatic structures such as the sternocleidomasoid muscle, internal jugular vein, and spinal accessory nerve. The functional and cosmetic disability and severe shoulder pain after this type of resection has led to the evolution of modified radical dissections. Theoretically, neck dissections preserving the integrity of SAN should not result in
From the Department of OtolaryngologyHead and Neck Surgery (B.N.), Haydarpasa Numune Education and Research Hospital, Istanbul, Turkey, the Department of OtolaryngologyHead and Neck Surgery (U.K.), Denizli Military Hospital, Denizli, Turkey, Department of Otolaryngology Head and Neck Surgery (A.A., D.E.S., E.O.), Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, Comprehensive Cancer Center, the Ohio State University, Columbus, Ohio, U.S.A. Editors Note: This Manuscript was accepted for publication November 17, 2010. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Dr. Baris Naiboglu, Acibadem Yaprak  Sokak Altinturk Apartmani No 39/8 34668, Turkey. E-mail: drbnaib@ yahoo.com DOI: 10.1002/lary.21468

shoulder morbidity; however, this is not always the case. It was pointed out that patients undergoing modified radical neck dissection had significantly worse shoulder function than patients with selective neck dissection in which one or more nodal groups are preserved.3,4 Selective neck dissection has gained its indications as a result of this thought of minimizing postoperative morbidity while maintaining oncologic safety. It is becoming the treatment of choice even in clinically N necks. The most important issue for maintaining oncologic safety in selective neck dissection is dependent on knowing the specific pattern of spread of that carcinoma according to its primary location at head and neck. Therefore, we aimed to evaluate the metastatic nodal status of level V at both pathologically N and N0 necks and display the impact of involvement of level IIV nodes on level V.

MATERIALS AND METHODS

Between January 1, 2005 and December 12, 2007, 243 patients had neck dissections due to various types of previously untreated HNSCC excluding sinonasal, nasopharynx, thyroid, and salivary gland carcinoma at the Department of OtolaryngologyHead and Neck Surgery at the James Cancer Hospital and Solove Research Institute of the Ohio State University Medical Center. Level V had not been routinely removed due to early cancer stage, clinically N0 neck, primary tumor site, or a combination of these factors. Therefore, the evaluation was carried out only over 107 patients. A retrospective review of pathologic reports of the specimens were done. The distinction of nodal groups had been done according to the separation of specimen by surgeon intraoperatively. In the majority of the cases, level V lymph nodes were identified according to the

Laryngoscope 121: March 2011

Naiboglu et al.: Managing Level V in HNSCC

545

TABLE I. Distribution of N1 Patients According to Primary Tumor Location and Nodal Levels.
Primary Site No. of Patients Patients With Lymph Node Metastasis: Levels IIV n (%) Patients With Lymph Node Metastasis: Level V n (%)

Oropharynx Oral cavity Larynx Carcinoma of unknown primary Hypopharynx Total

43 32 16 10 6 107

39 (90.7) 18 (56.3) 15 (93.7) 10 (100) 4 (66.7) 76 (71.0)

6 (13.9) 2 (6.2) 3 (18.7) 1 (10) 1 (16.6) 13 (12.1)

sternocleidomastoid muscle relation in the specimen. Level V nodes were submitted separately at the remaining few patients. All of the operations were performed by three senior surgeons (D.E.S., A.A., E.O.). Metastatic nodal status of all levels have been determined. Distibution of prevalence of lymph node positivity for level V according to the primary site and relation of level V positivity with other levels were evaluated.

RESULTS
The most common primary tumor site was oropharynx (OP) (n 43), followed by oral cavity (OC) (n 32), larynx (Lx) (n 16), carcinoma of unknown primary (CUP) (n 10), and hypopharynx (HP) (n 6). When regarding all nodal levels, pathologic N rates were as follows: Lx (15 of 16), OP (39 of 43), HP (4 of 6), and OC (18 of 32). General N positivity was 78.3% (76 of 97) when 10 patients with CUP were excluded. Level 5 was involved in 13 of 107 (12.1%) patients. The distribution according to the primary sites is as follows: Lx (3 of 16, 18.7%) followed by HP (1 of 6, 16.6%), OP (6 of 43, 13.9%), CUP (1 of 10, 10%), and OC (2 of 32, 6.2%) (Table I). One of the other levels was also positive in all level V-involved patients. In other words, metatasis was not isolated to level V in any of the cases. Level V was not involved in any patient when the other levels were not involved (0 of 21). Of those 13 patients, 10 showed multiple metastasis at other levels (levels IIV). Metastatic lymph node with a size of 6 cm or larger was determined in two patients. Even when considering only N patients, the ratio of N positivity for level V is still <20% (13 of 86, 15.1%). The distribution according to primary site is as follows: Lx 20% (3 of 15), OP 15.4% (6 of 39), HP 25% (1 of 4), OC 1.1% (2 of 18) and CUP 10% (1 of 10) (Table II).

integrity. Iatrogenic nerve damage is always possible. Intraoperative SAN division has been reported as an unexpected complication in 1.68% of modified radical neck dissections in a large case series.5 Dissections extending into level V, whether radical, modified radical, or part of a more selective procedure, are shown to cause a significantly higher incidence of postoperative shoulder dysfunction than dissections not extending to level V.6,7 Therefore, it seems reasonable not to dissect level V if there is not a risk of metastasis to that region. There are several reports about metastasis to level V in the literature. It was reported as to be between 4.8% and 7.4%.811 We found an overall prevalence of 12.1%. Our result is slightly higher than those in the literature. This is probably due to the selection bias while deciding to include level V in that neck dissection. Our patient population consisted mainly of clinically N patients. The impact of this high rate of preoperative lymph node positivity on pathologic neck and level V positivity are obvious in our results. Involvement of level V is expected to be even lower in a randomized group of patients, which would include more N0 and fewer N neck. Overall, the most probable primary tumor site with level V lymph node metastasis was Lx (18.7%) followed by HP (16.6%), OP (13.9%), CUP (10%), and OC (6.2%). Among those 13 patients with level V lymph node positivity, none of them was isolated to level V. Metatasis was observed only when one or more of the other levels had been involved. Level V positivity was observed in 13 patients among 86 N necks. This rate is still <20%. One exception is hypopharyngeal carcinoma

DISCUSSION
Radical neck dissection was the standard treatment for neck metastasis at the beginning of 20th century. Preservation of nonlymphatic structures has evolved first and followed by sparing of some nodal groups throughout the century for the sake of minimizing postoperative morbidity. There is a trend toward selective neck dissections even in clinically positive necks. Level V is the subject of focus for sparing due to the presence of SAN in the posterior triangle. Shoulder syndrome may arise postoperatively because of traction or devascularization even in the presence of gross anatomic nerve Laryngoscope 121: March 2011

TABLE II. Distribution of Level V Positivity Among N1 Patients According to Primary Tumor Location.
No. of N (Levels IIV) Patients Patients With Lymph Node Metastasis: Level V n (%)

Primary Site

Oropharynx Oral cavity Larynx Carcinoma of unknown primary Hypopharynx Total

39 18 15 10 4 86

6 (15.4) 2 (11.1) 3 (20) 1 (10) 1 (25) 13 (15.1)

Naiboglu et al.: Managing Level V in HNSCC

546

but the sample size is too small (four patients) to make a conclusion. The highest rates are observed in laryngopharyngeal carcinomas and the least at oral cavity carcinoma. Lim reported a prevalence of level V positivity of 5% in his series of 93 N oral cavity and oropharyngreal carcinoma patients. He concluded that level V lymph node group may be preserved in patients with clinically N oral cavity and oropharyngeal carcinoma patients below nodal stage T2a.12 Shah advised to dissect level V only when multiple nodal levels had been involved.9 In a study of 168 neck dissections, level V was not involved without positive nodes in other levels.13 In a large series consisting of 344 neck dissections for oropharyngeal and hypopharyngeal carcinomas, isolated skip metastases outside the levels II, III, or IV occurred in only one patient (0.3%). Otherwise, level I or V involvement was always associated with nodal metastases at other levels.14 The situation is not different for laryngeal carcinoma. Level V was rarely involved but always in conjunction with neck node metastases in levels II, III, or IV in a series of 262 radical neck dissections.15 The reason behind this is that the lymphatic drainage from a particular anatomic site is predictable and follows a sequential route. The first station lymph node of any primary site flows sequentially into the next lymph node station in the chain. In a previous study by us, we found that the major primary site with level I lymph node metastasis was OC followed by CUP, OP, and the least was laryngopharynx.16 Thus, this situation is entirely reversed for level V according to the result of the current study. Brennan et al. studied the incidence of recurrent nodal disease in level V in patients who had previously had a staging or therapeutic dissection of the neck, with or without postoperative radiotherapy. They reported that 41 patients (26%) developed recurrent metastasis out of 177 neck dissections in the neck and only four of them (3%) were in level V. In these four patients, level III or IV lymph nodes were shown histologically to have extracapsular spread at the time of the original dissection. None of the metastases at level V were from oral or oropharyngeal primary tumors.17 According to Davidson, obvious involvement of level IV must exist to dissect level V.10 However, level IV was not involved at 4 of 13 patients in this study. When considering the results of this study together with the reports of aforementioned authors, it might be reasonable not to dissect level V if the patient is clinically N0 and there is no scepticism about the involvement of level V intraoperatively at cases of which surgery is the first or sole modality. For clinically N necks, the risk for laryngopharyngeal carcinoma is considerably higher than oral cavity carcinomas.

The primary weakness of this study is that it is a retrospective study. It was not possible to give a rate of level V metastatic involvement among the entire group of HNSCC patients because not all the selective neck dissections include level V due to primary tumor location, early cancer stage, clinically N0 neck, or a combination of these factors.

CONCLUSIONS
Metastasis to level V is not expected if the other levels are not involved. Even if the other levels are involved, risk of occult metastasis to level V is not above 20%. The results of this study may help surgeons to make the decision of level V preservation in selected cases to avoid morbidities related to SAN and additional level V dissection in patients who will be treated with surgery as first or sole modality.

BIBLIOGRAPHY
1. Spiro RH, Alfonso AE, Farr HW, Strong EW. Cervical node metastasis from epidermoid carcinoma of the oral cavity and oropharynx. Am J Surg 1974;128:562567. 2. Crile GW. Excision of cancer of the head and neck. JAMA 1906;47: 17801786. 3. Chepea DB, Taylor RJ, Chepea JC, et al. Functional assessment using Constants shoulder scale after modified radical and selective neck dissection. Head Neck 2002;24:432436. 4. Cappiello J, Piazza C, Giudice M, et al. Shoulder disability after different selective neck dissections (level IIIV versus levels IIV): a comparative study. Laryngoscope 2005;115:259263. 5. Prim MP, De Diego JI, Verdaguer JM, et al. Neurological complications following functional neck dissection. Eur Arch Otorhinolaryngol 2006;263: 473476. 6. Terrell JE, Welsh DE, Bradford CR, et al. Pain, quality of life, and spinal accessory nerve status after neck dissection. Laryngoscope 2000;110: 620626. 7. Laverick S, Lowe D, Brown JS, Vaughan ED, Rogers SN. The impact of neck dissection on health-related quality of life. Arch Otolaryngol Head Neck Surg 2004;130:149154. 8. Schuller DE, Platz CE, Krause CJ. Spinal accessory lymph nodes: a prospective study of metastatic involvement. Laryngoscope 1978;88: 439449. 9. Shah JP. Patterns of cervical lymph node metastasis from squamous carcinomas of the upper aerodigestive tract. Am J Surg 1990;160:405409. 10. Davidson BJ, Kulkarny V, Delacure MD, Shah JP. Posterior triangle metastasis of squamous cell carcinoma of the upper aerodigestive tract. Am J Surg 1993;166:395398. 11. McDuffie CM, Amirghahari N, Caldito G, et al. Predictive factors for posterior triangle metastasis in HNSCC. Laryngoscope 2005;115:21142117. 12. Lim YC, Koo BS, Lee JS, Choi EC. Level V lymph node dissection in oral and oropharyngeal carcinoma patients with clinically node-positive neck: is it absolutely necessary? Laryngoscope 2006;116:12321235. 13. Cole I, Hughes L. The relationship of cervical lymph node metastases to primary sites of carcinoma of the upper aerodigestive tract: a pathological study. Aust N Z J Surg 1997;67:860865. 14. Candela FC, Kothari K, Shah JP. Patterns of cervical node metastases from squamous carcinoma of the oropharynx and hypopharynx. Head Neck 1990;12:197203. 15. Candela FC, Shah J, Jaques DP, Shah JP. Patterns of cervical node metastases from squamous carcinoma of the larynx. Arch Otolaryngol Head Neck Surg 1990;116:432435. 16. Ozer E, Karapinar U, Ryoo C, Agrawal A, Schuller DE. When to address level I in neck dissections. Otolaryngol Head Neck Surg 2010;142: 355358. 17. Brennan PA, Hoffman GR, Mackenzie N, et al. Recurrent nodal metastases in the posterior triangle: implications for treatment of the atypical tumour. Br J Oral Maxillofac Surg 2006;44:8386.

Laryngoscope 121: March 2011

Naiboglu et al.: Managing Level V in HNSCC

547