The extent and number of metastatic lymph nodes limit the efficacy of lymphadenectomy in patients with oesophageal carcinoma

Surgical Oncology 1994; 3: 187-l 92

The extent and number of metastatic lymph nodes limit the efficacy of lymphadenectomy in patients with oesophageal carcinoma K. YANO, T. OKAMURA, Y. YOSHIDA, T. EZAKI AND K. YASUMOTO The Second Department of Surgery, School of Medicine, University of Occupational and Environmental Health, Kitakyushu, Japan

The extent

and number of metastatic

carcinoma

involving

more than

efficacy of lymphadenectomy.

lymph nodes in 52 patients

one lymph

node were

analysed

with oesophageal to determine

the

The patients were divided into two groups according to

the number of metastatic lymph nodes, group A: l-4 nodes (34 cases) and group B: 5 or more nodes (18 cases). The survival of all patients in group A was statistically better than that of the patients in group B (P=O.O2). In cases where metastatic nodes were restricted

to one region. the survival of patients

in group A was significantly

better

than that of patients in group B (P=O.O08). By contrast, in cases where metastatic nodes extended to two or more regions, there was no difference between the groups. The present study suggests that lymphadenectomy is effective only for patients with l-4 metastatic nodes that are restricted to a single region. Lymphadenectomy is less efficacious in cases with five or more metastatic nodes, even when the nodes are restricted to one region, and is similarly ineffective in cases with metastatic nodes extending to two or more regions, regardless of the number of positive nodes. Surgical Oncology 1894; 3: 187-192. Keywords:

oesophageal carcinoma, metastatic lymph nodes.

INTRODUCTION

been conclusive. the

Although

aggressive

surgical

treatment

number of new adjuvant therapies mented

[I, 21, the prognosis

phageal carcinoma rates

as

between

reported

most

adjacent

was

tissue

and

of operation

tomy

concomitant

However,

world’s

explained

regional

with

been performed the effiacy

literature

Environmental

lymph

Extended

oesophageal

MATERIALS

of

involve nodes

l-1,

with

and number

of

special

of metastatic

Second

METHODS

metastasis

out of a total of 120 patients with primary

lymphadenec-

oesophageal

resection

resection in the Department

has

has not

Department

Yahatanishi-ku,

carcinoma,

Hospital of Occupational Japan,

of

University of Occupational and Iseigaoka,

AND

This study was based on 52 cases with lymph node

the

at the

to improve survival [8, 91.

Dr Koichi Yano, Health,

prognosis

by the finding

of lymphadenectomy

Surgery, School of Medicine,

to the extent

and

carcinoma

are

went Correspondence:

features

oesophageal

lymph nodes.

of patients with oeso-

carcinomas

[5-71.

with

reference

have been imple-

[3, 41. The poor prognosis

partly

patients

a

poor. Five year survival

the

oesophageal

time

therefore

in

8 and 20%

such patients that

remains

and

In the present study, we assessed

clinicopathological

from either

who

of Surgery II, University

1980 to 1992. The 52 patients total

oesophagectomy

gastrectomy

through

laparotomy.

Thirty-four

intra-abdomen), 187

surgical

and Environmental

right

and

18

underwent

(cervical, cases

Health, under-

or oesophago-

thoracotomy

cases

regional lymphadenectomies

Kita-

kyushu, Japan 807.

underwent

and three

intrathoracic,

underwent

two

K. Yano

188

regional

lymphadenectomies

thoracic,

intrathoracic

patients

received

operative cally

as having

oesophagus node.

curative

treatment

Dissected

resection

with

lymph

oesophageal

were

cervical

hiatus)

region,

celiac

according

to the

general

for

no pre-

Survival

curves

Esophageal

and axis,

test was

lymph

survival

Ilcx, Apple were

method.

used for curves.

The

Inc.).

by

generalized

Wilcoxon

comparison

comparisons

the of the

between

on actual patient numbers

using the x2-test. A P value

intrathoracic

Computer,

calculated

statistical

Statistical

groups were performed

into deep,

CO.05 was considered

significant.

trunk to

intra-abdominal

common

hepatic),

of the

Japanese

rule

Kaplan-Meier

of the

divided

(superficial,

supraclavicular),

(epigastric Society

(Macintosh

histologi-

(from the area around the brachiocephalic the

computer

cell carcinoma

nodes

Cancer Staging. All data were entered into a micro-

intraThe 52

more than one metastatic

three groups as follows; para-oesophageal,

and

and were diagnosed

squamous

with

(cervical

and intra-abdomen).

et al.

Diseases

[IO].

RESULTS

The total

The clinicopathological

features of groups A and B

number of dissected lymph nodes over the 52 cases

are summarized

was 1771. The mean number of dissected lymph nodes per case was 34.1. The number of cases

of 44 men and 8 women ranging in age from 42 to 80 years. The mean age was 62.2 years. There were

according to the number of metastatic lymph nodes

no significant

is shown in Fig. 1. The maximum

sex, tumour

number of meta-

in Table 1. The 52 cases consisted

differences location,

between

tumour

groups

static lymph nodes in one case was 29. The average number of metastatic lymph nodes per case was

features. However, there were statistically

5.00. The 52 patients were divided

tumour

differences

into two groups

in T classification

significant

(the depth of primary

(P=

penetration)

in age,

size, or histological

0.008)

and

stage

regardless of the region of metastatic lymph nodes; group A includes cases with l-4 metastatic lymph

(P=O.O19). The extent of metastatic lymph nodes in groups A and B is shown in Table 2. In both groups, intrathoracic and intra-abdominal lymph nodes were

nodes, and group B includes cases with five or more

the most frequent

metastatic

cases in group A, 21 cases (61.8%) had metastatic

according to the number of metastatic lymph nodes,

nodes. There are 34 cases in group A,

and 18 cases in group B. All of the resected tumours were post-surgically staged by the TNM classification system of the American Joint Committee for

sites of metastasis.

nodes restricted to one region, while 13 (38.2%) had nodes in two or more regions. Of the 18 cases in group B, however, only 6 (33.3%) demonstrated

Figure 1. The number according

. I . 2’3‘4.5-6-7

8 Number of metastatic

9 lymph

Of the 34

II nodes

13

20

29

lymph

nodes.

metastatic cases.

to the number *Average

lymph

of cases of metastatic value of

nodes for all 52

Metastatic

lymph nodes of oesophageal

Table 1. Clinicopathological features of group A and group B

189

carcinoma

Group A n

P-value*

Group B VW

n

(%I

29

(85.3)

15

5

(14.7)

3

(83.3) (16.7)

NS

(11.1) (61.1) (27.7)

NS

(11.1) (22.2) (55.6) (11.1)

NS

Sex male

female Age <50 50-70 705

4

(11.8)

2

25

(73.5)

11

5

(14.7)

5

3

(8.8)

2

2

4

22

(5.9) (64.7)

10

7

(20.6)

2

<5

10

(29.4)

2

5-10

22

(64.7)

15

(11.1) (83.3)

2

(5.9)

1

(5.6)

well

11

(32.3)

2

moderately

14

(41.2)

12

9

(26.5)

4

(11.1) (66.7) (22.2)

TO

7

(20.6)

2

(11.1)

Tl

9

(26.5)

1

T2

10

(29.4)

2

(5.6) (11.1)

T3

8

(23.5)

13

(72.2)

118

17

(50.0)

3

(16.7)

Ill

17

(50.0)

15

(83.3)

Total

34

Tumour location cervical upper middle lower Tumour size (cm)

105 Histological

NS

differentiation

poorly

NS

Primary tumour

P = 0.008

Stage

P=O.O19

18

*P value based on x*-test. NS: not significant.

nodal metastases (66.6%)

had

regions

restricted

metastatic

(P= 0.036).

12

The survival of patients in group B was significantly

more

lower than that of the patients in group A (P=O.O2).

to one region while

nodes

The extent

in groups A and B according

in two

or

of metastatic to tumour

nodes

location

is

The

survival

lymph

curves

shown in Table 3. The tumours in the middle portion

and B are shown

showed

with

a broader distribution

stases compared There

were

of lymph node meta-

with tumours in the other portions.

no statistical

differences

between

the

two groups in this respect. are shown three

in Fig. 2. One patient

patients

diseases

were

in group excluded

restricted

B who

in group A and died

from

other

from the survival analysis.

patients

with

curves

of

metastatic

to one region in groups A

in Fig. 3. The survival of patients

nodes

in group A was significantly

higher than that in group B (P=O.O06). extending

The survival curves of patients in groups A and B

of

nodes restricted

patients

with

metastatic

The survival lymph

are shown in Fig. 4. There was no significant ence

between

survival.

nodes

to two or more regions in groups A and B the

two

groups

with

differ-

respect

to

190

K. Yano et al.

Table 2. Extent of metastatic

lymph

Location

nodes

of metastatic

lymph

nodes

in groups

A and B Group

A

Group n (%)

n (%) One region

B

n (%)

21 (61.8)*

n (%)

6 (33.3)*

cervical

4 (11.8)

2 (11.1)

intrathoracic

8 (23.5)

4 (22.2)

intra-abdominal

9 (26.5)

0 (0.0)

Two regions cervical

12 (35.3)

8 (44.4)

+ intrathoracic

intrathoracic

+ intra-abdominal

Three regions cervical

0 (0.0)

2 (11.1)

12 (35.3)

6 (33.3)

1 (2.9)

+ intrathoracic

4 (22.2)

+ intra-abdominal

1 (2.9)

Total

4 (22.2)

34 (100)

18 (100)

*P= 0.036.

Table 3. Comparison Tumour

of the extent Metastatic

of lymphatic

lymph

metastases

between

groups

A and B according

node location

Number

to tumour

location

of cases

location Total

Group A

Group

n

n

n

3

2

lntrathoracic

1

2

Cervical

+ intrathoracic

1

1

Cervical

+ intrathoracic

0

1

1

0

Cervical

B

5 Cervical 6

Upper

+ intra-abdominal

32

Middle Cervical lntrathoracic Intra-abdominal Cervical + intrathoracic lntrathoracic Cervical

+ intra-abdominal

+ intrathoracic

+ intra-abdominal

Lower lntrathoracic Intra-abdominal lntrathoracic + intra-abdominal

long term

DISCUSSION There number

have

been

studies

of lymph node metastases

oesophageal

carcinoma

nodal involvement the depth positive

a few

[I l-l

with

the

positive

nodes

in patients with

positive

examining

31. The frequency

has been closely correlated

of tumour

invasion,

of with

and the number

nodes has been examined

results [II].

patients

of

with respect to

16.3%

nodes and

one

or

5

0

1

2

Five year two

in the

abdomen

in the thorax

14.3%,

survival

positive

alone,

alone,

respectively,

rates of

nodes, were

while

with

or

with

14.5%,

nearly

all

patients with more than three positive nodes or with positive nodes in both the thorax and abdomen due to recurrence

within

three

died

years of operation

Metastatic

lymph nodes of oesophageal

Figure 2. Survival curves in patients with 1-4 metastatic nodes (group A) and five or more metastatic nodes (group 6). There was a significant difference between the two groups (P=O.O2).

191

carcinoma

1

I

I

I

12

24

36

48

I 60

Months after operation

P - 0.006

Figure 3. Survival curves of groups A and 6 in patients with metastatic lymph nodes restricted to one region. There was a statistically significant difference between the two groups (P=O.O06).

[12]. It has also been reported

0

12

24

that patients with six

[13]. The above studies examined

the metastatic

and did not perform

only the correla-

nodes with

prognosis

survival analysis based on the

number of positive nodes according metastatic

to the extent of

lymph nodes. Similar to previous studies,

the present study found that in many cases with five or

more

penetrated the

metastatic

nodes,

the adventitia

metastatic

nodes

regions. We have now determined patients with l-4

metastatic

the

tumour

of the oesophagus,

extended

60

tally greater than that of patients with five or more metastatic

of positive

46

Months after operation

or more positive nodes have a very poor prognosis tion of numbers

36

to two

or

whom

the positive nodes extended

static

nodes.

The

lymphadenectomy resection

and

lymph nodes is statisti-

are restricted

regions is poor regardless

static

that the survival of

nodes

to one region.

We have also found that the prognosis of patients in

had more

lymph nodes, but only in cases in which

present

nodes

Lymphadenectomy

may

metastatic

or more

with

of meta-

suggests

to a single

not be effective

when those nodes are restricted

lymph

that

oesophageal

only in cases with l-4 restricted

with

five

study

concomitant

is warranted

lymph

to two or more

of the number

metaregion.

in cases

nodes,

even

to one region and

K. Yano et al.

Group 8 (n 2 9)

Group

A (II:

13)

Figure

4. Survival

and B in patients 0’

lymph

t

I

I

12

24

I

Months after

46

to two or more regions. Therefore,

order to effect

a longer

metastatic

60

difference

after

nodes in

resection

of groups

A

metastatic

extending

to two or

No significant

was observed

in survival

the two groups.

for carcinoma

of the oesophagus.

Br J

Surg 1987; 74: 165-8.

post-

7. Galandiuk Cancer

the latter two

in

nodes

more regions. between

should be considered survival

I

operation

with

therapy

of

36

in cases

adjuvant

number

I

extending operative

any

I

curves with

S, Hermann

of esophagus:

RE, Gassman

JJ, Cosgrove

the Cleveland

clinic experience.

DM.

Ann Surg 1986; 203: 101-8.

groups.

8. Akiyama

H,

Udagawa

Tsurumaru

HY, Suzuki

carcinoma

M,

Watanabe

G,

M. Development

of the esophagus.

Ono

Y,

of surgery

Am J Surg 1984;

for 147:

9-16. 9. Mansour

REFERENCES

KA,

surgery

Downey

without

RS.

preoperative

Esophageal adjuvant

carcinoma:

chemotherapy.

Ann Thorac Surg 1989; 48: 201-5. 1. Naunheim

KS, Petruska

chemotherapy

and

carcinoma.

J

P, Roy TS, era/.

radiotherapy

Thorac

for

Cardiovasc

Surg

Preoperative

10. Japanese

oesophageal

lines

1992;

103:

Kanehara

SJ, Stewart

result

JR, Sawyers

with

neoadjuvant

oesophageal

carcinoma.

JL,

et a/. Preliminary

therapy Ann

and

resection

Thorac Surg 1993;

11

for 56:

282-7. 3. Wang

LS, Hsiung

M, Chein KU. Gastric

carcinoma

of oesophagus:

substitution an analysis

for

4. Skinner

DB.

esophagus

En

bloc

and cardia.

resection

for

metastasis

5. Akiyama

abstract).

of

H,

Tsurumaru

of surgical

oesophagus.

Kawamura for

T,

carcinoma

Ono

Y.

of the

van

M, Obertop

Andel

JG, van

H, Mud Houten

thoracic

T, Ando qualitative

edn.

on

Tokyo:

of

N, Kawahara

H, Abe

analysis

the

carcinoma

on

of

the

0.

lymph

esophagus

J Jpn Surg Sot 1986; 86: 424-34. of the patients

esophagus,

of metastatic (English

S, Shimazu esophageal

tion in three tionship

Ann Surg 1981; 194: 438-46.

Schattenkerk WMH,

M,

treatment

8th

Guide

Studies

with

with

special

lymph

nodes

abstract).

carcinoma

of

reference

to the

and

meta-

their

J Jpn Surg Sot 1988;

68: 413-21. 13. Natsugoe

Principles

Diseases.

Pathological

Esophagus,

N. Prognosis

number

neoplasms

and

static portions

85: 59-71.

boom

node

Esophageal and

Co., 1992.

the thoracic

J Thorac Cardiovasc Surg 1983;

the

H, Kakegawa

Quantitative

12. Omori

for

Clinical

of

of

368 cases. Ann Thorac Surg 1992; 53: 289-94.

6. Eeftinck

Fujita

(English

resectable

the

Carcinoma

887-95. 2. Hoff

Society

for

HJ, EijkenH. Survival

Gastroenterol

after

areas. With special

between

metastatic

H, Baba M, et al. Recurrence cancer

lymph

lymph reference

recurrence nodes

and

(English

Surg 1991; 24: 2888-93.

node

of

dissec-

to the relanumber

abstract).

of Jpn

J