Abstract
Purpose
To highlight the strategy of neck dissection for various subsites of oral squamous cell carcinoma.
Materials & Methodology
Retrospective study of 153 patients with 164 neck dissection was involved between 2010 and 2016. Predictor variables were patient demographics, biopsy reports, imaging assessment and outcome variables were type of neck dissection and reconstruction performed and histological assessment of regional metastasis to the neck in relation to various primary subsites was carried out.
Results
Out of 153 patients, 126 (82.3%) were males and 27 (17.6%) were females with male-to-female ratio being 4.6:1. The mean age among the patients was 49.9 which ranged from 20 to 80 years. Lymph node metastasis was found in 22.6% of T1 and T2 tumors and 77.4% of T3 and T4 tumors. Incidence for gingivobuccal sulcus accounted for 49.6% of primary sites, tongue and floor of the mouth for 15.2%, retromolar trigone for 11.7%, lower alveolus for 8.6%, upper alveolus for 5.9%, lower lip for 3.9%, buccal mucosa for 3.3% and hard palate for 2.6%. Histologically metastasis was seen in level Ib (46%), IIa (33.1%) followed by others. Level V involvement was seen only in 5.5%.
Conclusion
For all subsites for N0 neck, minimum level III clearance should be performed, and for positive neck in RMT region, level IV or level V clearance is warranted. Apart from subsite, other factors to be considered are tumor stage, tumor thickness (DOI) and morphological characteristics of the primary tumor. The role of lymph node metastasis, number, size, extracapsular spread, its proximity and fixity to greater vessels in the neck.
Keywords: Squamous cell carcinoma, Metastasis; neck dissection, Cervical lymph nodes, Distant metastasis
Introduction
The head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer [1] in the world with an annual estimated incidence of around 275,000 for oral carcinoma [2]. India has been cited frequently as the country with the highest incidence in the world with over 100,000 cases noted every year [3].
The most important factor in the prognosis of squamous cell carcinomas of the upper aerodigestive tract is the status of the cervical lymph nodes with cure rates dropping to half with the involvement of regional lymph nodes [4, 5].
The neck has the most extensive and variable lymphatic system of the whole body. Of the estimated 800 lymph nodes in the human body, 300 lymph nodes are situated in the neck [6]. Although many different pathways and anastomoses of the lymphatics in the neck exist, most primary tumors metastasize to the first and further echelons of lymph nodes in the neck in a predictable pattern. In general, anterior oral carcinomas initially spread to the submandibular (less frequently submental) or to the high jugular lymph nodes. In posterior oral cavity and oropharyngeal carcinomas most frequently, involved lymph nodes are high (and mid) jugular lymph nodes. Tumors approaching or crossing the midline can metastasize to both sides of the neck [7].
The anatomical distribution of metastases from the various intraoral sites can only really be determined by examining the lymph nodes in pathological specimens, and this in turn is capable of providing data on which to base a logical surgical approach to the neck [8].
Though neck node involvement depends on primary site, tumor biology, there is not much of data available from Indian subcontinent where incidence is high. Most of the available data are from other parts of world where the nature and behavior of the tumor expression are different. There are controversies regarding the type of neck dissection to be carried out due to various levels of neck node involvement from different subsites. The primary aim of this study is to evaluate the incidence of metastatic lymph nodes in relation to specific subsites in the oral cavity. The various subsites included in our study were buccal mucosa, lip, upper and lower alveolar ridge, retromolar trigone (RMT), tongue and floor of mouth and hard palate.
Materials and Methods
Retrospective study involved 153 patients out of 175 of biopsy-proven squamous cell carcinoma of the oral cavity treated at our Craniofacial Unit from January 2010 to August 2016. A review of the medical records of these patients was done which served as the source of data for the study with the consent of Institutional Review Board. The age of patients included in the study ranged from 20 to 80 years; all were biopsy proven with preoperative assessment of primary and neck node with CT/dentascan, MRI and ultrasound. Recordings of various different types of neck dissection like RND, MRND and SOHND were done. Mean follow-up was 18 months. Patients with recurrence of oral cancer and patients having undergone primary radiotherapy before surgery were excluded from the study and 22 patients for lack of medical records and follow-up.
The outcome variables were incidence and frequency of metastatic involvement of cervical lymph nodes at various node levels which were determined clinically and by histopathological examination of neck specimen, and this was correlated with specific subsites in the oral cavity. The subsites included were (1) buccal mucosa, (2) tongue and floor of mouth, (3) retromolar trigone, (4) gingivobuccal sulcus, (5) upper and (6) lower alveolus, (7) hard palate and (8) lip. The clinical TNM staging was done according to AJCC classification system, and the node levels were designated according to Robbins classification.
Results
There were 153 patients of biopsy-proven squamous cell carcinoma of the oral cavity, among which 126 (82.3%) were males and 27 (17.6%) were females with male-to-female ratio being 4.6:1. The mean age among the patients was 49.9 which ranged from 20 to 80 years.
Lymph node metastasis was found in 22.6% of T1 and T2 tumors and 77.4% of T3 and T4 tumors. Gingivobuccal sulcus (GBS) was the most frequently encountered site and accounted for 49.6% (75 cases) of the primary sites. Lateral border of tongue and floor of mouth was involved in 15.2% (23 cases), retromolar trigone (RMT) in 11.7% (18 cases), lower alveolus in 8.6% (13 cases), upper alveolus in 5.9% (nine cases), lower lip in 3.9% (six cases), buccal mucosa (BM) in 3.3% (five cases) and hard palate in 2.6% (four cases) of the primary sites (Table 1).
Table 1.
Subsite distribution among population
| Subsite | Number of patients | Percentage (%) |
|---|---|---|
| Gingivobuccal sulcus | 75 | 49.4 |
| Tongue and floor of mouth | 23 | 15.2 |
| Retromolar trigone | 18 | 11.5 |
| Lower alveolus | 13 | 8.6 |
| Upper alveolus | 9 | 5.7 |
| Hard palate | 4 | 2.6 |
| Lower lip | 6 | 3.7 |
| Buccal mucosa | 5 | 3.3 |
| Total | 153 |
Clinically, the lymph nodes were palpable in 130 (84.9%) patients and the total numbers of palpable nodes were 173, with a sensitivity of 64.5% and a specificity of 69.9%. The number of patients in whom nodes were not palpable clinically, i.e., N0, were 23 (15%). The distribution among various node levels in the order of decreasing frequency was level Ib (77.4%), level IIa (12.7%), level Ia (8.1%), level IIb (1.1%), level III (0.5%). Level IV and V lymph nodes were not palpable clinically in any of the cases. Clinically bilateral lymph nodes were palpable in ten patients out of which nine patients underwent ipsilateral neck dissection and one patient underwent bilateral neck dissection.
Among the 121 metastatic lymph nodes detected histopathologically from the neck dissection specimen, the frequency of distribution of various levels of lymph nodes was level Ib (46.2%), level IIa (33.1%), level III (8.2%), level Ia (4.9%), level IIb and level IV (3.3%), level V (0.8%) (Table 2).
Table 2.
Nodal distribution according to neck levels
| Node level | Clinically palpable node levels | Histopathological positive metastatic lymph nodes | ||
|---|---|---|---|---|
| Number | Percentage (%) | Number | Percentage (%) | |
| Level Ia | 14 | 8.1 | 6 | 4.9 |
| Level Ib | 134 | 77.4 | 56 | 46.2 |
| Level IIa | 22 | 12.7 | 40 | 33.1 |
| Level IIb | 2 | 1.1 | 4 | 3.3 |
| Level III | 1 | 0.5 | 10 | 8.2 |
| Level IV | 0 | 0 | 4 | 3.3 |
| Level V | 0 | 0 | 1 | 0.8 |
| 173 | 121 | |||
The frequency of the individual levels of lymph nodes for each specific subsite is as follows (Table 3).
Table 3.
Subsite clinical nodal distribution
| Level Ia | Level Ib | Level IIa | Level IIb | Level III | Level IV | Level V | |
|---|---|---|---|---|---|---|---|
| Gingivobuccal sulcus | 14.2% | 57.1% | 14.2% | – | 14.2% | – | – |
| Tongue and floor of the mouth | 3.8% | 73% | 19.2% | 3.8% | – | – | – |
| Retromolar trigone | – | 44% | 33.3% | – | 5.5% | 11.1% | 5.5% |
| Lower alveolus | 13.3% | 80% | 6.6% | – | – | – | – |
| Upper alveolus | – | 100% | – | – | – | – | – |
| Hard palate | – | 66.6% | 33.3% | – | – | – | – |
| Lower lip | 16.6% | 66.6% | 16.6% | – | – | – | – |
| Buccal mucosa | 10.2% | 73.8% | 14.7% | 1.1% | – | – | – |
Gingivobuccal Sulcus
Clinically detected nodes were most common in level Ib (57.1%). Levels Ia, IIa and III were involved in equal frequency (14.2%).
Tongue and Floor of Mouth
The most frequently involved node level clinically was level Ib (73.1%), followed by IIa (19.2), Ia (3.8%) and IIb (3.8%). Levels III, IV and V were not detected clinically. The frequency of lymph node levels identified upon histopathological analysis was as follows: level Ib (33.3%) > level IIa (33.3%), level III (14.2%), level IV (9.5%), levels Ia and IIb (4.7%). Level V was not reported in any of the cases. Bilateral involvement at level IIa was detected histopathologically in one case.
Retromolar Trigone
Most frequently detected node level was level Ib (44.4%) followed by level IIa (33.3%) and level IV (11.1%). Level III and Level V were involved in one specimen each (5.5%).
Lower Alveolus
Clinically, 80% were in level Ib, 13.3% in level Ia and 6.6% in level IIa.
Histologically frequently involved node level was level Ib (43.7%), followed by level IIa (37.5%), levels Ib, IIb and III (6.2%) in the decreasing order of frequency, respectively.
Upper Alveolus
Clinically, level Ib was palpable in eight (100%) cases.
Final histopathology report showed two cases positive for metastatic lymph nodes at level Ib. No other lymph node groups were detected.
Hard Palate
Clinically, the frequency of lymph node involvement was 66.6% (two out of three) for level Ib and 33.3% (one out of three) for level IIa.
Histologically confirmed metastatic lymph nodes were in levels Ib and IIa (50% each).
Lower Lip
On clinical evaluation, 66.6% (four out of six) of lymph nodes were detected at level Ib and 16.6% (one out of six) each at levels Ia and IIa.
Metastatic deposits were noted in only one case at level IIa.
Buccal Mucosa
The most frequently involved node level clinically was level Ib (73.8%), followed by IIa (14.7%), Ia (10.2%) and IIb (1.1%) in the decreasing order of frequency, respectively.
The frequency of lymph node levels identified upon histopathological analysis was as follows: level Ib (50%), level IIa (30.3%), level III (8.9%), level Ia (7.1%), level IIb (3.5%).
Discussion
Overall incidence of neck node metastasis depends upon various factors, and the need to address the neck at what level is controversial and most debated. The purpose of this present paper was to assess the different levels of nodes involvement for subsites of oral cavity. Since the evolution of neck dissection first described by George Crile in the early twentieth century, modification for status of radical neck dissection was done by Martin who advocated level I–V clearance with spinal accessory nerve, sternocleidomastoid muscle and IJV; later, Suarez in 1963 popularized modified radical neck dissection with the preservation of important nonlymphatic structures. The role of adjuvant and neoadjuvant radiotherapy and chemotherapy has also come up in the management of neck nodes to prevent locoregional failure. There has been controversy regarding the type of neck dissection, level of clearance and morbidity. In spite of various factors influencing this decision, we should also consider the attitude and knowledge of the patient toward the disease and compliance for regular follow-up before deciding upon the type of neck clearance.
OSCC is a locoregional disease known notoriously for regional and distant metastasis [1]. The most important factor in the prognosis of OSCC is the status of the cervical lymph nodes. The cure rates drop to half with the involvement of regional lymph nodes [4, 5]. Not only the incidence of locoregional recurrence, but also the risk of distant metastases increases as the tumor burden in the neck increases. Other factors apart from presence or absence of lymph node metastases that influence prognosis are size, number, level and tumor thickness and extracapsular spread [7].
The neck has the most extensive and variable lymphatic system of the whole body. Of the estimated 800 lymph nodes in the human body, about 300 lymph nodes are situated exclusively in the neck. Thus, one-fifth and one-sixth of all nodes in the body are located in either side of the neck [6].
Pretreatment assessment of the status of the cervical lymph nodes is an essential tool for optimal treatment planning in oral cancer patients [7]. Palpable nodes are suspected as metastatic based on their size, consistency and location. But this method has not been found to be very reliable for the assessment of the neck. Modern imaging techniques such as computed tomography (CT), magnetic resonance imaging (MRI) and ultrasound-guided FNAC have developed in the recent years as more reliable methods. But the exact determination of metastatic deposits in the lymph nodes is possible only by histopathological examination of the nodes obtained from neck dissection.
It has been found that most primary tumors follow an almost constant and predictable route via local lymphatics to the first and second echelon lymph nodes. The first and second echelon lymph nodes are at the highest risk to harbor occult metastasis in early-stage oral carcinomas. Therefore, knowledge about the first and second echelons is very important for all clinicians dealing with these patients. It has been shown by various studies that level I serves as the first echelon of lymph nodes for oral carcinomas [7].
The cervical lymphatics are organized into four functional drainage pathways—the main lymphatic pathway, the posterior pathway, the anterior pathway and the superficial lateral pathway [9]. The oral cavity drains lymph via the main lymphatic pathway to the submental (level Ia) and submandibular nodes (level Ib). The submental nodes direct the lymphatic fluid toward the submandibular nodes which then subsequently direct it to the common entry point to the jugular chain at level IIa. From here, the lymph descends down the jugular chain to levels III and IV to finally enter into the major lymphatic ducts. Also, the structures near the midline have a tendency toward bilateral or contralateral drainage [9].
In our study, we found that there were 25 patients (16.3%) with clinically N0 neck who had histopathological-proven metastases in four cases majority being at level Ib (3/4) and IIa (1/4).
One hundred thirty (84%) patients were classified as clinically N+, and incidence of histopathological distribution of metastasis was level Ib, level IIa, level III, levels IV and IIb, level V. This is in concordance with the previous studies of Shah et al. [10].
In the present study, we have attempted to determine the incidence of lymphatic metastasis at each neck level according to the various subsites in the oral cavity.
Gingivobuccal Sulcus
Also called as “Indian cancer” because of its high incidence (60–75%). In this present study, out of 153 cases, the incidence was 75 cases (49.6%). Eighty-six percent of patients presented with T4 and stage 4 lesions with evidence of bone involvement. In our study, level Ib was found to be the involved node level (50%) in the histopathologically examined neck specimens of gingivobuccal sulcus primaries. This was closely followed by level IIa (30.3%), level III (8.3%), and level Ia (7.1%). Level IIb was involved in 3.5% of cases. In no instance were levels IV and V detected in this group of patients.
Similar results have been previously documented by Lindberg & Shah [10–12]. This clearly shows that most metastases in gingiva buccal sulcus primaries occur to levels Ib, IIa and III. Our findings confirm those of previous studies and demonstrate that there is a predictable pattern of lymph node metastasis for gingivobuccal sulcus primaries.
These findings warrant a minimum level III clearance of neck nodes in N0 cases of gingivobuccal sulcus primaries.
Tongue and Floor of Mouth
Tongue deserves special consideration because of its unique structure, site, function, lymphatic drainage and challenges posed for approach to resection and reconstruction. The incidence of tongue squamous cell carcinoma in India is 14/1,00,000/year in males to 2.4/1,00,000/year in females.
In our series, the incidence was 15.2% (23 cases) out of which seven cases (30.4%) were T1/T2 lesions and 16 cases (69.5%) were T3/T4 lesions. Level Ib and IIa lymph node metastasis was recorded at both levels 33.3%, respectively. Level III involvement was present in 19.5% of cases, and clinically these were posteriorly located lesions. There was no significant involvement of level IV (9.5%) and level V which had no incidence of metastatic deposit. Nine patients with clinically midline involvement underwent MRND on ipsilateral and SOHND on contralateral side. There was evidence of level IIa involvement in one patient justifying the bilateral neck dissection.
From these findings, we can infer that minimum level IV clearance is required for T3/T4 N0/N+ tongue and floor of mouth primaries. Also, the possibility of bilateral and/or contralateral spread should be considered for lesions approaching midline and crossing lingual septum which necessitates bilateral neck dissection. Shah et al. [11] have documented similar results though they have considered tongue and floor of mouth as separate subsites. They have also demonstrated posterior triangle (level V) metastases in their series of floor of mouth cases which we have not encountered. Nithya et al. in their series have also reported level II to be most frequently involved for oral tongue primaries [13].
Retromolar Trigone
Tumors of retromolar trigone (RMT) subsite are aggressive and pose poor prognosis in terms of disease-free survival (DFS) due to its proximity to mandible, maxilla and masticator space. The incidence of RMT tumors ranges from 4 to 15%. There were 18 (11.7%) cases of retromolar trigone in our study. Clinically, 100% level Ib lymph nodes were palpable in this series of cases of RMT primaries. On histopathological examination, metastasis was evident most commonly in level Ib (44%) and IIa (33%), respectively. There was one case of level V (5.5%) (posterior triangle) metastasis. This was the only case of level V metastasis among all the subsites considered.
Contrastingly, Shah et al. have not reported any level V metastasis in RMT primaries but other findings are similar to their study [11]. Woolgar has found level II as the most commonly involved followed by level I. Level III was involved in one case. There was no evidence of level IV or V involvement in their study [14].
In our study, the presence of level IV and V lymph nodes could be due to tumor biology and factors like tumor thickness and endophytic growth pattern. These findings in our study warrant level V clearance in cases of primary tumors arising from RMT.
Lower Alveolus
The tumor spread to lower alveolus has been well established. The controversy regarding the extent, resection of the mandible, i.e., rim sparing, segmental or hemimandibulectomy, continues with a greater role of reconstruction options in this subsite. This study Dentascan was the imaging modality to evaluate the extent of bone infiltration.
In the present study, 13 cases (8.6%) of lower alveolus were observed. As found in majority of the oral cavity subsites, lower alveolus also has propensity for metastasis to level Ib (43.7%) and IIa (37.5%). Levels IIb (6.2%) and III (6.2%) were detected histopathologically though lymph nodes at these levels were not palpable clinically in our patients. Levels IV and V were not detected. Thus, minimum level III clearance is required for these subsites. Shah et al. 11 have showed that gum primaries (alveolus) had metastasized to level I and II majorly and subsequently to levels III and IV. Level V spread, though in a few cases, was reported in their study in contrast to our study.
Upper Alveolus and Hard Palate
The incidence of this subsite is uncommon in our series as it was 13 patients (8.5%). Majority of the patients were reported with advanced disease with tendency for extension into infratemporal fossa (ITF). To downstage the disease, role of neoadjuvant chemotherapy is gaining popularity. In our series, three (1.96%) patients had level Ib and one (0.6%) patient at level IIa had histopathologically involved metastasis.
Zhang et al. have reported in their study that the positive lymph node detection was from levels I to III in cases of maxillary squamous cell carcinomas and metastases rate was higher for maxillary alveolus than that of the hard palate. The level of nodal involvement supports our study, but the metastasis rate was higher for hard palate (1/4 patients, i.e., 25%) than maxillary alveolus (2/9 patients, i.e., 22.2%) [15].
Lower Lip
Out of the six cases of lower lip which was analyzed, only one case was reported positive for neck metastasis. This was found at level IIa. This could be due to site of the lesion and bypassing of level Ia in this case suggestive of skip metastasis to level IIa.
Califano et al. in their study observed level I metastasis in 100% cases, level II and III in 28% cases and level IV in 4% cases of clinically N positive necks [16].
David Sharpe et al. in his study had ten patients of lower lip carcinoma. He found that upper internal jugular nodes (level II) were involved [8]. In three cases, level I lymph nodes were bypassed completely. We encountered similar findings where only level IIa was involved.
Buccal Mucosa
Buccal mucosa primary not extending into the sulcus was seen in five cases (3.3%). Level Ib was the most frequently involved level followed by level IIb. Narendra et al. in their study showed involvement of level I to III. Level V involvement was seen in very few of their cases (3–4%) [17]. We did not find any spread to III, IV or V levels. The implication of this finding suggests SOHND with level III clearance for pure buccal mucosa lesions without the extension into sulcus or alveolar bone.
Conclusion
These findings indicate that for N0 neck minimum level III clearance should be done during neck dissection for all oral carcinoma primaries and for N positive necks at least level IV or level V clearance is warranted. But the level of clearance during neck dissection does not solely depend on the subsite involved. We also need to take into consideration other tumor factors such as size, T stage, tumor thickness and morphological characteristics of the primary tumor. The role of lymph node metastasis, number, size, extracapsular spread, its proximity and fixity to greater vessels in the neck. Though the outcome of the study determines the type of neck dissection, similarly the audit of disease-free survival (DFS) and regional recurrence in the neck forms one of the shortcomings of this study.
Funding
Authors would like to declare there is no funding from institution or commercial company with regard to this study.
Compliance with Ethical Standards
Conflict of interest
None.
Ethical Clearance
The manuscript and study were cleared for publication from institutional review board (IRB No: 2017/P/OS/62).
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Li X, Shen Y, Di B and Song Q (2012) Metastasis of head and neck squamous cell carcinoma. Intech. ISBN: 978-953-51-0024-9
- 2.Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics. J Cancer Res Clin Oncol. 2005;55(2):74–108. doi: 10.3322/canjclin.55.2.74. [DOI] [PubMed] [Google Scholar]
- 3.Warnakulasuriya S. Global epidemiology of oral and oropharyngeal cancer. Oral Oncol. 2009;45(4–5):309–316. doi: 10.1016/j.oraloncology.2008.06.002. [DOI] [PubMed] [Google Scholar]
- 4.Shah J. Patterns of cervical lymph node metastasis from squamous carcinomas of the upper aerodigestive tract. Am J Surg. 1990;160(4):405–409. doi: 10.1016/S0002-9610(05)80554-9. [DOI] [PubMed] [Google Scholar]
- 5.Snow G, Annyasa A, Slooten E, Bartelink H, Hart A. Prognostic factors of neck node metastasis. Clin Otolaryngol. 1982;7(3):185–192. doi: 10.1111/j.1365-2273.1982.tb01581.x. [DOI] [PubMed] [Google Scholar]
- 6.Som P. Lymph nodes of the neck. Radiology. 1987;165(3):593–600. doi: 10.1148/radiology.165.3.3317494. [DOI] [PubMed] [Google Scholar]
- 7.Van den Brekel M, Leemans C, Snow G. Assessment and management of lymph node metastases in the neck in head and neck cancer patients. Crit Rev Oncol Hematol. 1996;22(3):175–182. doi: 10.1016/1040-8428(96)00191-6. [DOI] [PubMed] [Google Scholar]
- 8.Sharpe D. The pattern of lymph node metastases in intra-oral squamous cell carcinoma. Br J Plast Surg. 1981;34(1):97–101. doi: 10.1016/0007-1226(81)90110-7. [DOI] [PubMed] [Google Scholar]
- 9.Wang Y, Ow T, Myers J. Pathways for cervical metastasis in malignant neoplasms of the head and neck region. Clin Anat. 2011;25(1):54–71. doi: 10.1002/ca.21249. [DOI] [PubMed] [Google Scholar]
- 10.Shah J, Candela F, Poddar A. The patterns of cervical lymph node metastases from squamous carcinoma of the oral cavity. Cancer. 1990;66(1):109–113. doi: 10.1002/1097-0142(19900701)66:1<109::AID-CNCR2820660120>3.0.CO;2-A. [DOI] [PubMed] [Google Scholar]
- 11.Lindberg R. Distribution of cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts. Cancer. 1972;29(6):1446–1449. doi: 10.1002/1097-0142(197206)29:6<1446::AID-CNCR2820290604>3.0.CO;2-C. [DOI] [PubMed] [Google Scholar]
- 12.Pandey M, Shukla M, Nithya C. Pattern of lymphatic spread from carcinoma of the buccal mucosa and its implication for less than radical surgery. J Oral Maxillofac Surg. 2011;69(2):340–345. doi: 10.1016/j.joms.2010.02.031. [DOI] [PubMed] [Google Scholar]
- 13.Nithya C, Pandey M, Naik B, Ahamed I. Patterns of cervical metastasis from carcinoma of the oral tongue. World J Surg Oncol. 2003;1(1):10. doi: 10.1186/1477-7819-1-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Woolgar J. Detailed topography of cervical lymph-node metastases from oral squamous cell carcinoma. Int J Oral Maxillofac Surg. 1997;26(1):3–9. doi: 10.1016/S0901-5027(97)80837-5. [DOI] [PubMed] [Google Scholar]
- 15.Zhang W, Peng X. Cervical metastases of oral maxillary squamous cell carcinoma: a systematic review and meta-analysis. Head Neck. 2016;38(S1):E2335–E2342. doi: 10.1002/hed.24274. [DOI] [PubMed] [Google Scholar]
- 16.Califano L, Zupi A, Massari P, Giardino C. Lymph-node metastasis in squamous cell carcinoma of the lip. Int J Oral Maxillofac Surg. 1994;23(6):351–355. doi: 10.1016/S0901-5027(05)80053-0. [DOI] [PubMed] [Google Scholar]
- 17.Narendra H, Tankshali R. Prevalence and pattern of nodal metastasis in pT4 gingivobuccal cancers and its implications for treatment. Indian J Cancer. 2010;47(3):328. doi: 10.4103/0019-509X.64720. [DOI] [PubMed] [Google Scholar]