Journal of Surgical Oncology 43:45-49 (1990)

Site, Size, and Tumour Involvement of Resected Extrapulmonary Lymph Nodes in Lung Cancer K. KAYSER, MD,PhD,

s. BACH, MD, H. BULZEBRUCK, Phtl, AND

I. VOGT-MOYKOPF,

MD,

G . PROBST, MD

From the Department of Pathology, Thoraxklinik, Heidelberg, Federal Republic of Germany

A prospective study was performed analyzing resected extrapulmonary lymph nodes of 544 operated lung carcinoma patients. Volume of lymph nodes was determined by weight. Lymph nodes were cut in serial sections 300 pm thick, and the volume of tumour metastasis in each resected lymph node was computed measuring the tumourous area in the lymph node sections. The following results were obtained: Percentage of resected lymph nodes varied with lymph node site and site of the primary lung cancer. Hilar lymph nodes were resected in 50% of the patients, lymph nodes of the main and stem bronchi in 57% and 63%, respectively. Tumour metastases were found in 10%-15% of the resected lymph nodes. The size of the lymph nodes varied to a large amount and was found to be independent of the lymph node site if no metastases could be detected. Lymph nodes measuring 10 ccm) in 62% of the cases (20/32). Tumourous involved lymph nodes of the main bronchus were found more frequently in cases of central localized lung cancer compared to carcinoma arising from the peripheral lung, where the opposite was seen in subaortal located lymph nodes. The percentage of lymph node involvement and size of tumour infiltrated lymph nodes was found to be independent of tumour cell type. Size of resected lymph nodes is not a reliable parameter for estimating existence of tumour infiltrations. KEY WORDS:mediastinal lymph nodes, lymph node size, lung resection

INTRODUCTION This article presents information about percentage and size of resected extrapulmonary lymph nodes in patients operated on for bronchus carcinoma. Selection criteria of patients for potentially curative resection are based upon the performance status, the morphological cell type, the local tumour extent, and the existence of tumour metastases [I-41. The size of the mediastinal lymph nodes in patients suffering from lung cancer has been suggested as the criteria for determining the presence or absence of metastatic disease [5-71. The technological progress of image processing, i.e., computed tomography, filtered chest X-rays, and nuclear resonance techniques, has influenced several authors to these techthe sensitivity and specificity niques for detecting mediastinal lymph node metastases 0 1990 Wiley-Liss, Inc.

prior to surgical treatment [8-101. Although it is generally accepted that “normal”-sized lymph nodes can be tumour infiltrated and that “enlarged” nodes need not necessarily be involved, the frequency of lung cancer metastases in normal-sized mediastinal lymph nodes is still a subject of debate affecting especially the sensitivity of image analyzing techniques. We now report our anatomic postsurgical findings of a prospective study on lung cancer and discuss the significance of metastasis to enlarged and normal-sized extrapulmonary lymph nodes.

Accepted for publication August 9, ,989. Address reprint requests to K . Kayser. M . D . , Ph.D., Dept. of Pdthology, Thoraxklinik, Amalienstr. 5 , D6900 Heidelberg, FRG.

46

Kayser et al.

TABLE I. Tuniour Cell Type and Sex of Operated Bronchus Carcinoma Patients Total

Females

Males

TABLE 11. Tumour Cell Type and Tumour Site of Operated Bronchus Carcinoma Patients Site

N

o/o

N

5%

N

8

Epidermoid Adeno Large cell Small cell Combined

238 125 41 23 27

52 27 10 4 7

27 39 9 3 6

32 46 4 7

265 164 56 26 33

49 30 10 5 6

Total

460

85

84

15

544

100

Cell type

11

MATERIALS AND METHODS Data of the prospective study were collected from January I , 1985, to December 31, 1987. Lymph nodes of 544 potentially curative operated lung carcinoma patients were classified according to region with a special node mapping scheme analogous to the scheme of the American Thoracic Society (ATS). Surrounding fatty tissue was carefully removed and each lymph node was weighed. The lymph nodes were cut in serial sections 300 pm thick using a cryostate and stained with hematoxylin and eosin (HE). All slides were carefully examined by light microscopy and tumourous area was measured if existent. The complete tumour volume was computed adding the tumourous areas multiplied by the thickness of the intersections (300 pm). Resection specimens (lobes and lungs of patients with primary lung carcinoma) were fixed by inflating air with moderate pressure (50 mmHg) and with buffered formalin (6%) into the major bronchi. Specimens were allowed to fixate an additional 24 hours. Specimens were cut into sagittal slices 6 mm thick starting from the hilar side. The primary tumour volume was computed measuring the tumourous area of each slide multiplied by its thickness and summing up the tumour volume of the slides. The histological cell type was determined according to the classification of the WHO [ 111 analysing histological slides of a complete tumour cross section stained with HE, alcian blue, periodic acid-Schiff, and Masson s stain. The procedure is described in detail elsewhere

[la. Data were fed into a computer (IBM 3090). Statistical analysis was performed using SAS programmes.

RESULTS A synopsis of the material is given in Table I. A total of 544 patients (460 males and 84 females) could be included in the study. Epidermoid carcinoma contributed to the material in 49% followed by adenocarcinoma and large cell anaplastic carcinoma (30% and 10% respectively). Site of the tumours classified according to cell type is given in Table 11. The majority of the tumours

Central

Peripheral

Total

Cell t w e

N

oiu

N

%

N

Epidermoid Adeno Large cell Small cell Combined

193 80 33 13 21

57 23

35 42

10 4 6

72 84 23 I3 12

6 6

265 164 56 26 33

49 30 10 5 6

Total

340

63

204

37

544

100

11

%

TABLE 111. Number of Resected and Tumour Infiltrated Lymph Nodes According to Lymph Node Mapping (N = 544 Operations)" No. of lymph nodes Resected"

Tumour infiltrated

Site

N

o/o

N

O/o

112 3 4 5 6 7 8 9

31 1 379 I83 272 286 162 205 I74 282

57 70 34 50 53 30 38 32 52

41 38 24 42 52 31 II 16 50

13

10

Total

2,254

10

13 15 18 19 5 9 18

305

*I12 = peribronchial; 3 = main bronchus; 4 = tracheobronchial; 5 = subcarinal; 6 = paratracheal; 7 = subaortal; 8 = pulmonary ligament; 9 = paraoesophageal; 10 = non-classified. "According to No. of resections.

(N = 341, 62%) developed at a central site, i.e., from a main or stem bronchus, whereas 203 carcinomas (38%) were located in the peripheral lung. The number of resected lymph nodes in relation to the lymph node mapping is given in Table 111. Lymph nodes of the main and stem bronchi were resected in 70% of the patients separately from the excised lung specimens, and the hilar lymph nodes in 50%. All in all 2,254 lymph nodes were excised in addition to the surgical specimens, i.e., an average of 4.1 lymph nodes per lobe (lung). The percentage of the excised lymph nodes was found to be independent of the side of the primary lung carcinoma (i.e., right or left lung). Lymph nodes located at 3-5 were excised more frequently in central localized tumours compared to peripheral localized tumours (Table IV). Tumour infiltrations in lymph nodes at 3 were detected twice as often in centrally located tumours compared to peripherally located tumours (15% vs. 7%) and somewhat higher in lymph nodes at 5-7 (Table IV). The mean tumour volume (confidence limits, P >

47

Lymph Nodes in Lung Cancer TABLE V. Volume of the Primary Lung Carcinoma and of Tumour Infiltrations in Extrapulmonary Lymph Nodes (Confidence Limits P 2 0.95)

TABLE IV. Number of Resected and Tumour Infiltrated Lvmoh Nodes Accordine to Tumour Site* Tumour site CENTRAL (N

=

PERIPHERAL (N

340)

Tumour Resecteda infiltrated

N

%

N

%

N

I12 3 4 5 6 7 8 9 10

180 254 120 181 187 98 128 114 169

53 75 35 53 55 29 38 34 50

28 39 16 31 36 21 6 8 34

16 15 13 17 19 21 5 7 20

1.431

219

204)

Tumour Resectedd infiltrated

Site

Total

=

%

N

%

131 64 125 61 63 31 91 45 99 49 64 31 7 7 3 8 60 29 113 55

13 9 8 II 16

10 7 13 12 16 16 7 13 12

823

96

10

5 8 16

*I12 = peribronchial; 3 = main bronchus; 4 = tracheobronchial; 5 = subcarinal; 6 = paratracheal; 7 = subaortal; 8 = pulmonary ligament; 9 = paraoesophageal; 10 = non-classified. "According to No. of resections.

0.95) of the primary lung carcinoma and of the resected pulmonary lymph nodes in relation to the cell type is given in Table V. No statistically significant differences in the volume of the primary lung carcinoma according to cell type could be found. This is in accordance with data already published [ l ] . The tumour mass in the additionally excised lymph nodes was found to be larger in cases of small cell anaplastic carcinoma and large cell anaplastic carcinoma compared to epidermoid carcinoma and adenocarcinoma. These findings are in agreement with survival data showing an improved survival of patients suffering from epidermoid/adenocarcinoma compared to patients operated on for large cell and small cell anaplastic carcinoma [ l ] . The percentage of excised lymph nodes with and without tumour infiltrations according to the size of the lymph nodes is given in Table VI showing that about 10% of small and medium sized lymph nodes (