The Relationship between the Localization, Size, Stage and Histopathology of the Primary Laryngeal Tumor with Neck Metastasis

Objective: In this study we aimed the investigate the laryngectomy (partial and total) and neck dissection (functional, radical) in patients undergoing larynx identified as a result of the operation of primary tumour localization, size, stage and histopathology of the relationship between neck metastasis. Materials and Methods: The files of 118 patients who has had operation of laryngectomy (partly, total) and neck dissection (functional, radical) because of larynx cancer in the Ear Nose and Throat (ENT) Clinic in the last ten years (1997-2007) have been evaluated retrospectively. Results: 116 of them were male and 2 of them were female. Age average was 56+/-22. 189 neck dissections have been applied to 118 patients. While in supraglottic tumours, 55.2% cervical metastasis and 28.9% occult metastasis have been determined, in transglottic tumours 35.1% cervical metastasis and 18.9% occult metastasis have been determined. It has been seen that while the cervical metastasis ratio of tumours bigger than 4 cm is 55.8%, the tumours smaller than 4 cm is 27.27% (p=0.002). The cervical metastasis rate has been determined as 26.3% in T1 cases, 25.8% in T2 cases, 40.6% in T3 cases and 58.3% in T4 cases. In tumours with advanced T stage, the rate of cervi-cal metastasis has been seen more than the ones with early T stage (p=0.027). Cervical metastasis has been determined as 27.3% in good differential (G1) tumours, 53.4% in medium differential (G2) tumours, and 74.1% in bad differential (G3) tumours (p=0.005). Conclusion: Considered in the light of clinical parameters specified in the cervical lymph node metastasis, N 0 patients with laryngeal cancer, tumour, according to the anatomical location as unilateral or bilateral elective neck dissection should be performed and his-topathological specimens taken in the event of a positive detection of metastases concluded that postoperative radiotherapy should be performed. Özet Amaç: Bu çalışmada larenjektomi (parsiyel, total) ve boyun disek-siyonu (fonksiyonel, radikal) yapılan hastalarda, operasyon sonucu larenkste tespit edilen pirimer tümörün lokalizasyonu, büyüklüğü, evresi ve histopatolojisi ile boyun metastazı arasındaki ilişkiyi araş-tırmayı amaçladık. (p=0,027). İyi diferansiye (G1) tümörlerde %27,3, orta diferansiye (G2) tümörlerde %53,4, kötü diferansiye (G3) tümörlerde %71,4 oranında servikal metastaz tespit edildi (p=0,005). Sonuç: Belirtilen klinik parametreler ışığında servikal lenf nodu me-tastazı düşünülen, N 0 larenks kanserli olgularda; tümörün anatomik olarak yerleşim yerine göre unilateral veya bilateral elektif boyun diseksiyonu yapılmalı ve alınan spesmenlerde histopatolojik (+) me-tastaz tespit edilmesi durumunda postoperatif radyoterapi mutlaka uygulanmalıdır kanaatine varılmıştır.


Introduction
The most important factors which define the prognosis in laryngeal cancers are the place of primer lesion, the size of lesion, whether or not there is a cervical metastasis, and cellular differentiation [1].The risk of metastasis to cervical lymphatics has a crucial importance in laryngeal carcinoma.The presence of lymph node decreases the probability of local control and survival [2].In this study, we aimed at making a survey of the relationship between the localization, the size, the phase, and the histopathology of the primer tumour, determined in larynx by operation, and neck metastasis in the patients who was operated (functional and radical) neck dissection and (partial, total) laryngectomy in our clinic in the last decade.

Materials and Methods
The files of 118 successive patients treated with (functional and radical) neck dissection and (partial, total) laryngectomy operations because of laryngeal cancer in our clinic in the years between 1997-2007 were retrospectively analysed.Patients whom neck dissection was not operated were excluded from the study.Ages and sexes of the patients, localization of the tumour, whether or not there was a lymph node in the neck before the operation, operations performed, operation findings and histopathological results of the postoperative specimen were retrospectively analysed.The results of preoperative neck examination were compared with the histopathological examination results of the postoperative neck dissection specimen.As a result of preoperative clinic examination and postoperative pathological evaluation, necks with positive tumour metastasis were evaluated as N (+); and, the necks, accepted as N 0 (no palpable lymph node) in preoperative clinic examination and getting N (+) (tumour metastasis positive and lymph node presence) in postoperative pathological evaluation were evaluated as N false (-) (occult metastasis).Patients who were determined to have node by preoperative palpation and determined not to have metastasis in pathological examination of postoperative specimen were accepted as N false (+) (false +).Lymphadenopathy (LAP) bigger than 1.5 m in the neck was accepted as clinically N (+) in physical examination.Neck metastases were evaluated with regard to the level of lymph node.We had no case of laryngeal cancer having isolated subglottic location.T1, T2, T3, and T4 phases, differentiation degrees, and sizes of supraglottic, glottic and transglottic tumours and the presence of cervical metastasis in these tumours were examined in each case separately.The relationship between localization, size, phase, and histopathological differentiation degree of the tumour and neck metastasis was evaluated.

Statistical Analysis
The figures attained after the study were statistically evaluated in a computer working under Microsoft® Windows XP operating system, using SPSS package software (SPSS 13.0 for Windows).Performing the Chi-Square test, ordinal figures, the relationship between them and the relationship between them and the shared figure were determined.

Results
Of 118 patients with laryngeal cancer included in the study, 116 were male, and 2 were female.The youngest patient was 34, and the oldest was 78.
As for the surgical treatment operated on the neck, 118 patients were operated 189 neck dissections.173 of these were functional neck dissections, and 16 were radical neck dissections.Also, 71 patients were operated bilateral neck dissections.
The relationship between the localization of the tumour and cervical metastasis has been shown in Table 1, and the relationship between the anatomic region the tumour has occupied in laryngeal parts has been shown in Table 2.It has been determined in our study that the supralaryngeal tumours are much more metastatic to the neck.When the contralateral cervical metastasis was compared with the anatomic localization the primary tumour occupied, it was determined that 6 (13.3%) of 45 tumour cases which were reaching the other side crossing the midline, showing ipsilateral localization, 3 (25%) of 12 piriform sinus located tumour cases, 6 (18.7%) of 32 tumour cases with epiglottis involvement, 1 (5.3%) of 19 tumour cases showing anterior commissural involvement, and 1 (50%) of 2 tumour cases with tongue base involvement did contralateral cervical metastasis.
The relationship between the histopathological differentiation degree of the tumour and cervical metastasis has been shown in Table 3.This relationship was found very meaningful in our study (p=0.005;p<0.01).
According to Tumor Node Metastasis of American Joint Committee on Cancer (AJCC TNM) classification, the relationship between the T phase of the tumour and cervical metastasis has been shown in Table 4.While cervical metastasis was found in 13 (26%) and occult metastasis was found in 9 (18%) of 50 advanced-phase (T 1 -T 2 ) cases, cervical metastasis was found in 34 (50%) and occult metastasis was found in 16 (23.5%) of advanced-phase (T 3 -T 4 ) 68 cases.In our study, there was a meaningful relationship between the T phase of the tumour and the risk of cervical metastasis (p=0.027;p<0.05).However, the relationship between the T phase of the tumour and the risk of occult metastasis was not found to be meaningful (p=0.639;p>0.05).The relationship between the size of the tumour and cervical metastasis is shown in Table 5.The reason for the use of 4 cm as limit value in the size of the tumour was that the diameter of the primary lesion was found between 2-4 cm in the most of the cases and that value of 4 cm was median as of the distribution of the cases.As a result of our study, there was a very meaningful relationship between the size of the tumour and cervical metastasis (p=0.002;p<0.01).
The relationship between the localization of the tumour and the level of cervical metastasis it did isshown in Table 6.
In 6 (12.7%) of our 47 cases showing cervical metastasis, perinodal invasion was seen.In our study, it has been determined that perinodal invasion is not connected with the localization, the T phase, and the differentiation degree of the tumour.

Discussion
The presence of cervical lymph node is an important factor in terms of the prognosis in head and neck cancers [3][4][5].
It has been stated that while 5-year survival is between 29-54% in patients who have laryngeal cancer with nodal metastasis, it changes between 64-94.5% in patients who do not have local metastasis [6,7].Size and number of the involved lymph node, and whether there is extracapsular spread or not affects survival [8,9].It has been stated that, if the size and number of the involved lymph node increase, if there is lymph node involvement in more than one level, and if extracapsular spread exists, there is a significant decrease in survival [3,8,9].Therefore, it is important to know the factors which affect the neck metastases and to determine the treatment to be operated on the neck [10].
In our study, cervical metastasis was discovered in 21 (55.2%) of 38 supraglottic tumours.The neck was evaluated as N0 in physical examination in 10 cases, although metastasis (occult metastasis) was found in the neck dissection specimen.No cervical metastasis was found in any of the 6 glottic cancer cases.In 26 (35.1%) of 74 glottic cancer patients, cervical metastasis was found, and occult metastasis was found in 14 (18.9%) of these 74 patients.We had no  any isolated subglottic located case.It has been determined in our study that the results obtained are in accord with the literature [3,5].In our study, in keeping with the literature again, the rates of cervical metastasis have been determined as 48.4% for epiglottis, 53.8% for ariepiglottic plica, 41% for ventricular band, 44.4% for ventricular, 31.8% for cord vocal, 50% for thyroid cartilage, 68.7% for piriform sinus, 100% for tongue base involvement and 54.5% for preepiglottic area involvement [11].
In our study, there was contralateral cervical metastasis in 17 patients.Of these patients, 8 patients had midline located tumour, and 6 had the tumour crossing the other side reaching the midline.In addition, 3 patients had piriform sinus located tumour while 16 of 17 contralateral metastases had also ipsilateral lymph node metastasis and these results are considered to be accordant with the literature [11].
The increase in grade of the primary tumour increases the risk of neck metastasis [5,13,14].It has been emphasized that the rate of cervical metastasis is higher in tumours with poor histopathological differentiation (grade=G 3 ) than those with good differentiation (grade=G 1 ) [13,15].In our study, out of 118 patients cervical metastasis was found in 27.3% (18 of 66 cases) of cases with good differentiation (G 1 ), in 55% (25 of 45 cases) of cases with moderate differentiation (G 2 ), and in 71.4% (5 of 7 cases) of cases with poor differentiation (G 3 ) (p<0.01).
The relationship between the size of the tumour and cervical metastasis is disputable.Kirchner and friends [16], in a study of 50 cases, have found metastasis at the rate of 55% in transglottic cancers, bigger than 4 cm, and at the rate of 25% in tumours, smaller than 4 cm.It has been stated in this study that cervical lymph node metastases depend on the localization of the primary tumour rather than its size.In our study, cervical metastasis was found in 29 (55.8%) of 52 laryngeal tumours, bigger than 4 cm and in 18 (27.27%) of 66 laryngeal tumours, smaller than 4 cm (p<0.01).
It has been emphasized that cervical lymph node metastasis is at I and II neck level at the most in many supraglottic and glottic located laryngeal carcinoma cases [17].In a study consisting of 40 patients with clinic N0, the main localizations of all occult metastases been determined to be ipsilateral II and/or III.Occult metastasis could not be found in the lymph nodes on the levels of I, IV, and V [18].Following the directions which can be early lymph node metastasis in laryngeal carcinomas that are clinically N0, it has been emphasized that lateral neck dissection which has been operated on is an acceptable treatment method [19,20].It has been stated that involvement of the area I increases in the case of the involvement of the areas II-IV as well as the probability of isolated involvement is rarely seen [19,20].In our study, being accordant with the literature, of 47 cases which were laryngeal cancer and being cervical metastatic, 15 cases (31.9%) were metastatic to level II, 4 cases (8.5%) to level III, 5 cases (10.6%) to level IV, 11 cases (23.4%) to level II and III, 2 cases (4.25%) to level III and IV, 4 cases (8.5%) to level I, II, and III, and 1 case to all the levels.Also, it was determined that 80% of the cases which had involvement with area IV had cartilage invasion.
When the patients with extracapsular spread were compared with the patients of the same phase who had cervical metastasis limited to lymph node, a meaningful decrease in survival was determined [21].Extracapsular spread was seen in 6 cases in our study.
It has been stated that the probability of pathological (+) lymph node presence in the neck specimen obtained by operation changes between 13.6% and 37% in the laryngeal cancer patients whose necks are clinically non-palpable to lymph nodes [19,20].Probably, it will never be possible to be able to determine micrometastases in neck lymph nodes correctly with imaging methods.The risk of cervical metastasis, concerning the characteristic features of the primary tumour (localization, size, grade, infiltration degree), is at stake [22].Therefore, the treatment of the neck should be handled spryly, even if the neck is clinically negative [23].
The debates increase especially on the neck examination and patient group accepted as N 0 .In this case, there are three options: elective radiotherapy, elective neck dissection and neck follow-up until encountering a metastatic disease [28,29].Some of the authorities advocate elective neck dissection because of the fallibility of palpation and the high risk of lymph node metastasis of supraglottic tumours [30,31].Also, some authorities have adopted and recommended elective neck dissection as standard treatment because it decreases recurrence rates in clinically negative necks [32][33][34].Nowadays, detailed histopathological evaluation of the lymph nodes removed during the neck dissection is the most reliable method available for the diagnosis of the correct locations of lymph nodes [35].
Consequently, the risk of cervical metastasis to be associated with the characteristic features of the primary tumour (localization, size, T stage, histopathological differentiation degree), was determined.We believe that, in N0 laryngeal cancers, it is necessary to do unilateral or bilateral elective neck dissection according to anatomic localization of the tumour, in cases where positive cervical lymph node is thought to exist considering clinical parameters (tumour size, T stage and histopathological differentiation degree), and to do postoperative radiotherapy in case that histopathological (+) metastasis is found in taken specimens.