EXTENDED MEDIASTINAL NODE
DISSECTION FOR NON-SMALL CELL
The most important factor influencing survival of surgical
patients with non-small cell lung cancer (NSCLC) is
the presence or absence of lymph node involvement.
Considerable controversy still exists regarding surgical
treatment for patients with mediastinal lymph node
In 1960, Cahan  described the surgical procedure
for "radical lobectomy" in which one or two lobes were
excised in a block dissection of regional, i.e. hilar and
mediastinal lymph nodes. Our procedures have been
more extensive than Cahan's procedures. Routine
systematic dissection of the mediastinal lymph nodes
was performed in every case even if the preoperative
evaluation was N0 or N1, in accordance with the lymph
node map proposed by the Japan Lung Cancer Society
SURGICAL PROCEDURES FOR MEDICSTINAL
To improve the survival of N2 patients, we modified the
mode of lymph node dissection over the past 25 years
[4-9]. Before 1980, the mode of lymph node dissection
was less extensive compared with that of since 1981.
Since 1981, we have performed systematic and
extensive lymph node dissection in every curable case
of non-small cell lung cancer (NSCLC), even if the
preoperative evaluation was N0 or N1.
On the right side, the mediastinal pleura was
longitudinally incised along the trachea and esophagus
from the apex to the base of the right hemi-thorax. For
node dissection in the superior mediastinum, the azygos
vein was cut to mobilize the trachea, esophagus, and
superior vena cava. All of the accessible lymph nodes
in the superior mediastinum, i.e. the superior mediastinal
(#1), para-tracheal (#2), pretracheal (#3), retrotracheal
(#3p), and tracheobronchial angle (#4) nodes, were
removed with the surrounding fat pad. The node anterior
to the superior vena cava (#3a) was also routinely
removed, including thymic tissue. By these procedures
all of the lymph nodes and fat pad located around the
subclavian artery, trachea, right main bronchus,
ascending aortic arch, superior vena cava, and upper
thoracic esophagus could be completely removed.
For node dissection in the inferior mediastinum, the
incised posterior mediastinal pleura was reflected and
the pulmonary ligament was cut to expose the tracheal
bifurcation, both main stem bronchi, the pericardium,
and the lower thoracic esophagus. All of the lymph nodes
in this compartment, i.e. subcarinal (#7), paraesophageal
(#8), and pulmonary ligament (#9) and also contralateral
hilar (#10) and paraesophageal (#8) nodes were
dissected out with their surrounding fat pad.
On the left side, lymph node dissection in the inferior
mediastinum was performed similarly to that on the
right side. However, in the superior mediastinum, there
are great anatomical limitations on node dissection,
which is in marked contrast to the right side. This is
mainly due to obstruction of the surgical field by the
aortic arch. As a routine procedure in left thoracotomy,
the mediastinal pleura was incised just beneath the
aortic arch and the subaortic(#5), paraaortic (#6), #7
and #8 nodes are removed.
In addition to the routine dissection method, two kinds
of adjunctive procedures ((a) mobilization of aorta and
(b) median sternotomy) have been performed for more
extensive nodal dissection in the upper mediastinum.
Since 1981, the aortic arch and part of the descending
aorta are routinely mobilized by cutting Botallo's ligament
and a few intercostal arteries to allow more extensive
removal of the superior mediastinal nodes, i.e. #3, 3a,
3p and 4 nodes.
In addition to division of the Botallo's ligament, since
1986, the operative procedure was modified when left-
sided N2 lung cancer was diagnosed by preoperative
CT scanning or discovered at thoracotomy. For more
complete nodal dissection, median sternotomy was
performed after procedures in the left hemithorax had
been completed using the left postero-lateral approach.
By this procedure, #1, 2, 3, 3a, 4, and 7 nodes in the
ipsilateral side and also mediastinal and hilar lymph
nodes in the contralateral side could be completely
dissected [8,9]. By employing these dissection methods,
improved survival of left N2 patients was noted on
comparison with that undergoing routine nodal
Hata et al. (10) reported excellent results by routine
use of median sternotomy for left-side lung cancer in
* Department of Chest Surgery, Kanazawa Medical
University, Ishikawa, JAPAN.
Solunum 3, Özel Sayı 2:180-182, 2001
TUMOR DIAMETER AND INCIDENCE OF IYMPH
In Table I, incidences of lymph node metastasis in
relation to the size of the primary tumor are shown in
1255 patients with measurable tumor size. Overall,
the incidence of N0, N1, N2 and N3 diseases were
61.1%, 11.4%, 23.5%, and 4.1%, respectively. The
incidence of N2 disease increased as tumor size
increased. Fifteen percent of patients having 11 to
20mm tumor size have N2/N3 diseases. When the tumor
size increases to a range of 21 to 30mm, the incidence
increases to 25%. If the tumor measures more than
30mm, more than 30% demonstrate N2 disease.
Table I: Tumor size and frequency of lymph node
metastases in NSCLC
METASTATIC SPREAD OF N2 DISEASE
All dissected lymph nodes were sent for pathological
examination. In this study, there were some patients
with multiple metastatic lymph nodes within a single
level, but such nodes were defined as one metastatic
focus in this study. Figure 1 shows the distribution of
mediastinal lymph node metastases in patients, who
underwent complete dissection of the mediastinal lymph
node. Open circles represent single-level metastases,
and solid circles multi-level metastases. Upper lobe
lesions involved more metastatic levels than lower level
lesions. Common metastatic levels were #7, 4, 3, 2,
In patients with multi-level metastasis most of the
metastases were noted in the regional lymph nodes,
i.e. upper lobe lesion mostly involved the upper
mediastinal nodes and lower lobe lesion mostly involved
those in the lower mediastinum. However, in cases of
right lower lobe tumors and left lower lobe tumors,
metastases to the superior mediastinum (nonregional
mediastinum) were observed more frequently than
metastases of upper lobe tumors to the inferior
Figure I: Metastatic spread of N2 disease in relation to
the location of the primary tumor.
RLL: right lower lobe. RML: right middlle lobe. RUL: right upper lobe. LUL: left
upper lobe. LLL: Left lower lobe
RATIONALE FOR EXTENSIVE IYMPH NODE
As indicated above, nodal spread of metastasis in
the mediastinum is frequently noted and extensively
scattered. Therefore, during surgery, extensive
mediastinal node dissection should be performed.
In addition, further rationale for extensive mediastinal
node dissection are as follows: Firstly, even T1
lesion may induce high frequency of N2 disease as
shown in Table I. Secondly, the accuracy of CT
scan in demonstrating N2 disease is not very high.
We did not use mediastinoscopy as a routine
preoperative diagnostic procedure, and preoperative
staging were mainly done by CT scan. The diagnostic
rate of N2 disease were sensitivity; 67%, specificity;
81% and accuracy; 77%. Thirdly, there is no method
of detecting latent (microscopic) metastasis by gross
intraoperative findings. Fourthly, there are many
preoperatively unproven (latent) N2 disease detected
after postsurgical examination of the resected
specimens. Among the 659 patients who were
evaluated as N0 by preoperative evaluation, 89
patients (13.5%) had pN2 disease verified by
postoperative pathological examination.
Furthermore, among the 269 patients with pN2
disease, 40% were diagnosed as cN0 (n=89) or cN1
RESULTS OF SURGERY
Since 1988, we have done radical extensive lymph
node dissection in 1517 cases as a routine
procedure. The mortality mortality within 30 days
after operation were noted in 19 patients, constituting
mortality rate of 1.2%. As a postoperative
complication, 5 patients suffered from postoperative
chylothorax with no mortality.
There were 218 stage IIIA-N2 patients underwent
resection with 2 (0.9%) operative mortality. Their
operative radicality was 152 (70%) complete
resection and 66 with incomplete resection. The 5-
year survival rate of patients who underwent
complete resection was 30%, whereas that of the
incompltely resected cases were 5%. In total, the
5-year survival rate of the resected N2 patients was
23%. The survival rate of the completely resected
pN2 patients who were evaluated as N0-1
preoperatively showed the 5-year survival rate of
36%, showing significantly better survival than that
of the cN2 patients which was 27%. There was a
significant difference between the two groups.
When prognostic factors affecting the survivals of
patients with N2 disease were analysed, favourable
factors were complete resection, one-level
metastasis, preoperatively unproven N2 (cN0-1),
T1-2N2M0, intranodal microscooic metastasis,
without #7 nodal involvement and primary tumor
less than 20mm. On the other hand, unfavourable
factors were incomplete resection, multi-level
metastases, radiologically evident N2 disease, T3-
4N2 disease, extranodal expansion, number, #7
node metastases, or tumor more than 50 mm.
1. Cahan WG : Radical lobectomy. J Thorac Cardiovasc
Surg 1960; 39:555-570.
The Japan Lung cancer Society. General rule for
clinical and pathological records of lung cancer.
4th ed. Tokyo, Kanehara Publishing Comp 1995,
Naruke T, Suemasu K, Ishikawa S: Lymph node
mapping and curability at various levels of metastasis
in ?resected lung cancer. J Thorac Cardiovasc
Surg 1978; 76:832-839.
Watanabe Y, Shimizu J, Oda M, et al.:Aggressive
surgical intervention in N2 non-small cell lung
cancer of the lung. Ann Thorac Surg 1991;51: 253 -261.
Watanabe Y, Shimizu J, Tsubota M, et al: Mediastinal
spread of metastatic lymph node in?bronchogenic
carcinoma. Chest 1990; 97:1059-1065.
Watanabe Y, Hayashi Y, Shimizu J, et al: Mediastinal
nodal involvement and the prognosis of non-small
cell lung cancer. Chest 1991;100:422-428.
Watanabe Y: Review: Results of surgery for N2
non-small cell lung cancer. Ann Thorac Cardiovasc
Surg 1996; 2: 85-94.
Watanabe Y, Ichihashi T, Iwa T: Median sternotomy
as an approach for pulpmonary surgery. Thorac
Cardiovas Surgeon 1988; 36: 227-231.
Watanabe Y, Shimizu J, Oda M et al: Improved
survival in left N2 lung cancer by modification of
operative procedure. Thorac Cardiovasc Surgeon
10. Hata E, Hayakawa K, Miyamoto H, et al.: Rationale
for extended lymphadenectomy for lung cancer.
Theor Surg 1990;5:19-25.
Extended mediastinal node dissection for non-small cell lung cancer