Navigation

Lung Cancer Pictures

A present, mediastinal lymph node metastases from primary lung cancer are nonoperatively diagnosed mainly by computed tomography (CT). However, as the imaging ability of CT is not sufficient to show lymph nodes clearly and identify lymph node metastases by their sizes, the accuracy of diagnoses on clinical N stage by CT has been as low as 59.8 percent. (1)

On the other hand, transesophageal endoscopic ultrasound examination (EUS) has been significantly improved in the last ten years and has been mainly used in the field of gastroenterology. Although there have only been a few reports on the diagnostic ability of EUS for lymph node metastases from primary lung cancer, it is worth paying attention to the diagnostic capability of EUS. (2)

Among patients with primary lung cancer who were admitted to the National Cancer Center Hospital from July 1987 to April 1988 for surgical treatment, 132 patients underwent EUS. Pathologic examination was made on 101 of the 132 patients. In this article, we studied the diagnostic capability of EUS on these 101 patients. The mediastinal lymph node dissection and the pathologic examination were not made on the other 31 patients for various reasons: no operation because of obvious N3 nodal involvement or distant metastasis, receiving only pulmonary resection as symptomatic therapy for their pneumonia and/or hemoptysis, and receiving only exploratory thoracotomy because of dissemination.

PATIENTS

The patients were 79 men and 22 women. Their ages ranged from 32 to 81 years (average, 63.3 years). Adenocarcinoma was observed in 55 patients; squamous cell carcinoma, 33 patients; small cell carcinoma, five patients; large cell carcinoma, four patients; adenosquamous cell carcinoma, two patients; and simultaneous adenocarcinoma and squamous cell carcinoma, two patients. The involved areas by tumors were as follows: right lung in 50 patients (upper lobe, 29; middle lobe, five; lower lobe, 16) and left lung in 51 patients tipper lobe, 30; lower lobe, 20; upper and lower lobes, one).

METHODS

A GF-UM2 radial scanner with 7.5-MHz (Olympus Co Ltd) was used in a balloon method with deaerated water. Using endoscopic ultrasound observation equipment of EU-M2 (Olympus Co Ltd), imaging pictures were recorded on videotapes and pictures were taken with a camera (Polaroid).

The premedication was performed in the same way as upper gastrointestinal endoscopy, and the patients assumed the left lateral decubitus position during EUS.

The diagnostic standard for positive metastases was based on the thickness, clearness of contours, and existence of low echo images of fusion or lobulation of lymph nodes.

As an imaged lymph node and a resected one were occasionally very difficult to identify, the metastasis was determined in an individual site.

Based on the classification of Naruke et all, the sites of mediastinal lymph nodes were divided into the following eight groups, including the supraclavicular lymph node aDd a part of the hilar lymph node. However, the examination on the site of the paraaortic lymph node (No. 6) was not made as the identification of lymph nodes was difficult in that region.

1. Supraclavicular lymph node: Located at the upper part of the horizontal line at the top of the upper rim of the subclavian artery or vein.

2. Highest mediastinal lymph node (No. 1): The site is defined by the horizontal line at the top of the upper rim of the subclavian artery or vein and the horizontal line at the left point of the trachea where the tipper rim of the brachiocephalic artery ascends to the right, crossing in front of the trachea.

3. Paratracheal and pretracheal lymph node (No. 2 and 3): between No. 1 and No. 4

4. Right tracheobronchial lymph node (right No. 4): Located on the obtuse angle level between the trachea and right main bronchus, with and inside the azygos vein.

5. Left tracheobronchial lymph node and Botallo's lymph node (left No. 4 and 5): Located between the aortic arch, the left pulmonary artery, and the obtuse angle between the trachea and left main bronchus.

6. Parasubcarinal lymph node (No. 7, 8, and 10): subcarinal lymph node (No. 7), paraesophageal lymph node (No. 8), and hilar lymph node (No. 10) adjacent to No. 7 were counted as one region, as they are difficult to distinguish, especially when the lymph node is swollen.

7. Pulmonary ligament lymph node (No. 9): Located on the posterior wall and lower part of the inferior pulmonary vein

8. Retrotracheal lymph node (No. 3p): Located on the posterior portion of the trachea

The imaging ability of EUS was examined by counting the number of sites in which lymph nodes were imaged and the number of imaged lymph nodes by EUS in 509 sites in 101 patients on whom histopathologic examination had been made. Study was carried out in line with results obtained by the images recorded and the histopathologic results in terms of the following: (1) verification of sensitivity, specificity, positive predictive accuracy, negative predictive accuracy and overall accuracy of EUS images; (2) analysis on total images recorded by EUS, excluding images of the right superior mediastinal lymph node, which was in the blind angle of EUS; and (3) histologically confirmed metastatic stages.

At the same time, CT images taken for the 101 patients were reviewed to confirm its imaging capability on metastases to lymph nodes in sizes larger than I em in the shortest diameter and the result of the review was compared with that attained by EUS. The CT scanners used in this study were TCT60A30 and/or TCT900S (Toshiba Co Ltd, Tokyo). The scanning time for the former machine was 4.5 s. Twelve milliliters of 65 percent diatrizoate meglumine (Angiographin) was injected at 1 to 2 ml/s in one shot, and scanning of one slice at one time was started 25 s after the injection. For the latter machine, the scanning time was 1 s. One hundred milliliters of 65 percent diatrizoate meglumine (Angiographin) was injected at 0.8 to 1 ml/s, and scanning of six to seven slices at one time was started 40 s after the injection.

To evaluate sensitivity, specificity, positive and negative predictive accuracy, as well as overall accuracy of EUS in percentage, at first each image was diagnosed and classified into four such categories as "true positive," "true negative," "false positive," and "false negative, " followed by rating in percentage calculated for respective categories in such manner as follows: Sensitivity: number of "true positive" divided by total number of "true positive" and "false negative." Specificity: number of "true negative" divided by total number of "true negative" and "false positive." Positive predictive accuracy: number of "true positive" divided by total number of "true positive" and "false positive." Negative predictive accuracy: number of "true negative" divided by total number of true negative" and "false negative." Overall accuracy: total number of true positive" and true negative" divided by total number of true positive," "true negative," "false positive," and "false negative."

The metastases of lymph nodes were histopathologically classified into the following three stages: (I) mild metastasis: micrometastasis barely observed on the sinusoid; (II) moderate metastasis: metastasis between I and III; and (III) severe metastasis: metastasis invaded more than half of the lymph node. The diagnostic capability of EUS in determining the metastatic stages of the lymph node was evaluated by comparing the results of EUS with the above-mentioned histopathologic classification.

RESULTS

The Imaging Ability of Lymph Nodes of EUS

Although the right side of the supraclavicular lymph node, right side of No. 1, right No. 2 and 3, and right No. 4 were generally impossible to be observed because of the air echoing through the trachea and bronchus, other regions were clearly and continuously observed by EUS.

The number of sites in which lymph nodes were imaged and the number of lymph nodes imaged by EUS and CT were compared in Table 1 [omitted] . Both EUS and CT satisfactorily imaged lymph nodes in left No. 4 and 5 and No. 7, 8, and 10, especially. EUS-imaged lymph nodes in almost all the cases. The lymph nodes in No. 9 were rarely imaged by CT, but 22 of 56 cases were imaged by EUS. In the right superior mediastinum (the right side of No. 1, right No. 2 and 3, and right No. 4), it was difficult to image lymph nodes by EUS except significantly enlarged ones. Except for the region of the right superior mediastinum, the numbers of imaged lymph nodes per one region were 1.6 by CT and 2.7 by EUS. There were some cases in which lymph nodes were imaged as one by CT but revealed as plural by EUS. Thus, the more detailed properties of each lymph node (configuration of marginal line, metastatic lobulation, etc, and internal echo) were observable by EUS compared with CT .

Four hundred eight-two lymph nodes were imaged by EUS, of which 238 (49.4 percent) were less than 5 mm in the shortest diameter, 201 (41.7 percent) were between 5 and 10 mm. Among them, 45 (60.0 percent) of 75 metastatic lymph nodes diagnosed as positive by EUS were sized between 5 and 10 mm (shortest diameter).

The Diagnostic Ability of EUS to Detect Lymph Node Metastases

Among 509 sites studied, 69 were histopathologically positive for metastasis, but in the results of EUS examination, 75 lymph nodes in 48 sites were positive. The metastases in each site were accurately diagnosed except the right superior mediastinum. The lymph nodes on the right superior mediastinum were difficult to be imaged, but once they were imaged, the accurate diagnoses were obtained. The comparison of the diagnostic results obtained by EUS and CT were summarized in Table 3 [omitted]. The results obtained by EUS in 509 sites were as follows: sensitivity, 53.6 percent; specificity, 97.5 percent; positive predictive accuracy, 77.1 percent; negative predictive accuracy, 93.1 percent; and overall accuracy, 91.6 percent. The number of sites, excluding the right superior mediastinum, was 225, and their results were 80.6 percent, 95.0 percent, 72.5 percent, 96.7 percent, and 92.9 percent, respectively. There were many sites with false negative results by CT (47 sites), and the sensitivity of the whole site was 31.9 percent. Thus, the results obtained by EUS were more dominant than those of CT. The cases in which no lymph nodes or small-sized lymph nodes with negative signs of metastasis were imaged by CT, but were proved as positive by EUS, are shown in Figures 5, 6, and 7.

Relation between the Histopathologic Findings and the Results of EUS on the Metastatic Lymph Nodes

In 509 sites, 69 sites were pathologically proved to have lymph node metastases. Among them, 30 of 42 sites with severe metastases, 5 of 17 with moderate metastases, and 2 of 10 with mild metastases were proved true positive by EUS. And in the sites excluding the right superior mediastinal lymph node, the metastases were pathologically identified in 37 sites, and of these, EUS revealed the metastases in 22 of 23 sites with severe metastases, 5 of 8 with moderate metastases, and 2 of 6 with mild metastases. The diagnoses determined for severe lymph node metastases by EUS were comparatively accurate, while the diagnoses by EUS were inaccurate for the cases with mild metastases.

DISCUSSION

The five-year survival rates of patients with N2 lung cancer undergoing surgical treatment have recently been reported to be approximately 23 percent (4) and 30 percent. (5) These survival rates showed improvement to some extent, but the improvement was not sufficient enough. Therefore, multidisciplinary treatments, such as neoadjuvant therapy, etc, should be used more in the future. Taking this into consideration, accurate pretreatment staging for patients with lung cancer is extremely important as a fundamental procedure although it is very difficult to accomplish.

At present, preoperative diagnoses for mediastinal lymph node metastases of primary lung cancer are mainly made by CT scanning. Since CT has no blind angle, all of the mediastinum could be examined. However, its ability to image lymph nodes is not sufficient. Due to the movement and the partial volume effect of the pulmonary artery and vein, the aortic arch and the left atrium, aortopulmonary window (left No. 4 and 5), subcarinal and paraesophageal lymph nodes are not sufficiently imaged by CT. (6-8)

According to a report on lung cancer, the sites of No. 3p, 4, 5, 7, 8, and 10 had been observable by EUS. (2) The site of para-aortic lymph nodes could not be diagnosed, as it had been barely possible to identify the node with the 7.5-MHz radial scanning system of EUS. However, we identified sites from the left supraclavicular to the left superior mediastinum (No. 1 and 2), from left No. 4 to 9 were observed in succession by EUS in all the cases, while the right superior mediastinum was not observed since it was in a blind angle to the EUS due to the positions of the esophagus and trachea.

Mitani et al (9) reported that CT scanning at the interval of 1 cm made it possible to image lymph nodes in only 25 percent of targeted sites. With the detailed study on normal subjects, Genereux and Howie (10) reported that about 40 percent of lymph nodes had not been imaged in No. 7 and left No. 4. In this study, lymph nodes in the left No. 4 and 5 and No. 7, 8, and 10 were imaged by EUS in almost all the cases, and lymph nodes in No. 9 were imaged in 40 percent of the subjected cases. When imaging abilities of EUS and CT were compared for sites excluding the right superior mediastinum, which could not be observed by EUS, it was verified that EUS showed significantly more superior ability than CT, with results in imaging of 62.3 percent against 40.1 percent attained by CT, and the numbers of lymph nodes imaged in one site were 2.7 (EUS) and 1.6 (CT).

The diagnoses of lymph node metastases by CT are made in accordance with the sizes of lymph nodes. However, various diagnostic standards for positive metastases have been reported as follows: all the imaged lymph nodes, (11) 1 cm or more in diameter, (12,13) 1.5 em or more, (7,8,14,15) and 2 cm or more. (6) This suggests that the diagnoses of metastases by the sizes of lymph nodes are apt to be inaccurate. As the rates of false positive diagnoses of swelling lymph nodes due to inflammation are from 15 to 33 percent, the necessity of histopathologic examination for confirmation has been reported quite often. (6-8,11-14) Martini et al (7) and Patterson et al (16) have also pointed out that as far as the sizes of lymph nodes are used as standards for diagnosing metastases, the accuracy of diagnosis is limited. Some articles have discussed the findings of ultrasound examination of lymph nodes, (17,18) and we also have reported our experiences with ultrasound examination in the cervical region for prescalene nodes., Table 5 [omitted] shows the findings of ultrasound examination of lymph nodes discussed in that study it shows the findings of internal echoes of the lymph nodes were not specific to the metastasis. However, the metastasized lymph nodes could be differentiated from nonmetastasized nodes by their clear contours and thickened images with fusion or lobulation. Thus, we decided to exclude the sizes of lymph nodes from the standard factors for diagnosing metastases. Excellent results were obtained in sensitivity, specificity, positive predictive accuracy, negative predictive accuracy, and overall accuracy. Among the lymph nodes examined by EUS, 91 percent were less than 10 mm in diameter, and 60.5 percent of lymph nodes diagnosed as positive for metastasis were also less than 10 mm in diameter. Thus, diagnosis of metastasis by size of lymph nodes is difficult, and the sensitivity rates obtained by CT have shown inaccuracy of 31.9 percent. Therefore, EUS imaging is an appropriate diagnostic procedure for lymph node metastases not in line with the sizes of lymph nodes, but established standards, because serious metastases to lymph nodes small in size could be detected and diagnosed by EUS imaging, although EUS imaging has difficulty in identifying micrometastases accurately to the extent of other imaging means.

According to the results of this study, EUS is useful for diagnosing mediastinal lymph node metastases originating from primary lung cancer. The features EUS confers are: (1) all sites except the right superior mediastinum can be observed; (2) compared with CT, the imaging ability of lymph nodes within the observable extent centering around the middle and inferior mediastinum is significant; (3) furthermore, metastases could be diagnosed from some factors except size, as each lymph node is clearly seen in an imaged area.

There have been many reports (7,15,19-21) in which mediastinoscopy and CT were used together in considering the limited diagnoses obtained by CT only However, EUS is not only useful in covering sites invisible to mediastinoscopy, but it is also able to make accurate prognoses nonsurgically and noninvasively Therefore, more accurate diagnoses of mediastinal lymph node metastases can be obtained by using EUS and CT together, which covers the EUS's blind angle of the right superior mediastinum.

CONCLUSION

Endoscopic ultrasound examination showed significant images of lymph nodes except in the right superior mediastinum. Using EUS, metastases to mediastinal lymph nodes were diagnosed with some standards except the size. The results were as follows: sensitivity, 80.6 percent; specificity, 95.0 percent; positive predictive accuracy, 72.5 percent; negative predictive accuracy, 96.7 percent; and overall accuracy, 92.9 percent. Endoscopic ultrasound examination was able to accurately diagnose highly metastasized lymph nodes even if they were small, but it was inaccurate in diagnosing micrometastases.

REFERENCES

1 Kondo D, Naruke T, Kondo H, Goya T, Tsuchiya R, Suemasu K. Evaluation on significance of neck ultrasound examination in primary lung cancer: detection of non-palpable metastatic scalene nodes. J Jpn Assoc Chest Surg 1988; 2:117-24

2 Soga H, Takeuchi R, Kurita H, Komatsubara S, Shimizu N, Teramoto S. Ultrasonic evaluation of mediastinal lymph node metastasis, especially to subcarinal node, in lung cancer by trans-esophageal radial scan. Rinsho Geka 1987; 42:1405-10

3 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; 76:832-39

4 Sawamura K, Lee Y, Suemasu K, Miyazawa N, Yamaguchi Y, Nakada T, et al. Analysis of N2 lung cancer undergoing surgical resection with complete mediastinal lymph node dissection. Haigan 1988; 28:1-9

5 Martini N, Flehinger BJ. Surgical treatment of lung carcinoma. Surg Clin North Am 1987; 67:1037-49

6 Ferguson MK, MacMahon H, Little AG, Golomb HM, Hoffman PC, Skinner DB. Regional accuracy of computed tomography of the mediastinum in staging of lung cancer. J Thorac Cardiovasc Surg 1986; 91:498-504

7 Martini N, Heelan R, Westcott J, Bains MS, McCormack P, Caravelli J, et al. Comparative merits of conventional, computed tomographic, and magnetic resonance imaging in assessing mediastinal involvement in surgically confirmed lung carcinoma. J Thorac Cardiovasc Surg 1985; 90:639-48

8 Faling LJ, Pugatch RD, Jung-Legg Y, Daly BDT Jr, Hong WK, Robbins AH, et al. Computed tomographic scanning of the mediastinum in the staging of bronchogenic carcinoma. Am Rev Respir Dis 1981; 124:690-95

9 Mitani N, Shimotakahara T, Sannou K, Nishijima H, Arimura T, Baba K, et al. CT diagnosis of mediastinal lymph node metastasis of lung cancer: comparison between removed lymph nodes and CT in four areas. Haigan 1986; 26:643-49

10 Geneureux GP, Howie JL. Normal mediastinal lymph node size and number: CT and anatomic study AJR 1984; 142:1095-1100

11 Rea HH, Shevland JE, House AJS. Accuracy of computed tomographic scanning in assessment of the mediastinum in bronchial carcinoma. J Thorac Cardiovasc Surg 1981; 81:825-29

12 Backer CL, Shields TW, Lockhart CG, Vogelzang R, LoCicero J III. Selective preoperative evaluation for possible N2 disease in carcinoma of the lung. J Thorac Cardiovasc Surg 1987; 93:337-43

13 Richey HM, Matthews JI, Helsel RA, Cable H. Thoracic CT scanning in the staging of bronchogenic carcinoma. Chest 1984; 85:218-21

14 Daly BDT, Faling LJ, Bite G, Gale ME, Bankoff MS, Jung-Legg Y, et al. Mediastinal lymph node evaluation by computed tomography in lung cancer. J Thorac Cardiovasc Surg 1987; 94:664-72

15 Schnyder PA, Gamsu G. CT of the pretracheal retrocaval space. AJR 1981; 136:303-08

16 Patterson GA, Ginsberg RJ, Poon PY, Cooper JD, Goldberg M, Jones D, et al. A prospective evaluation of magnetic resonance imaging, computed tomography, and mediastinoscopy in the preoperative assessment of mediastinal node status in bronchogenic carcinoma. J Thorac Cardiovasc Surg 1987; 94:679-84

17 Kumegawa H. Study of the endoscopic ultrasonography for esophageal carcinoma. Jpn J Gastroenterol Surg 1985; 18:1774-83

18 Murata Y, Muroi M, Yoshida M, Okushima N, Sugiyama A, Ide H, et al. Endoscopic ultrasonography in the diagnosis of esophageal carcinoma. J Jpn Bronchoesophagol Soc 1986; 37:378-85

19 Osborne DR, Korobkin M, Ravin CE, Putman CE, Wolfe WG, Sealy WC, et al. Comparison of plain radiography, conventional tomography and computed tomography in detecting intrathoracic lymph node metastasis from lung carcinoma. Radiology 1982; 142:157-61

20 Cougblin M, Deslauriers J, Beaulieu M, Fournier B, Piraux M, Rouleau J, et al. Role of mediastinoscopy in pretreatment staging of patients with primary lung cancer. Ann Thorac Surg 1985; 40:556-60

21 Baron RL, Levitt RG, Sagel SS, White MJ, Roper CL, Marbarger JR Computed tomography in the preoperative evaluation of bronchogenic carcinoma. Radiology 1982; 145:727-32

Table 5-Findings of Ultrasound Examination on Cervical Lymph Nodes (Nine Sites, 17 Lymph Nodes in Seven Pathologically Examined Patients)

                                          Positive      Negative
                                         Metastasis    Metastasis
                                            (15)          (2)
Contours
   Clear                                     9             0
   Partly clear                              6             2
   Unclear                                   0             2
Internal echoes
   Homogenous, strongly hypoechoic           3             0
   Fine granular, diffuse, strongly          7             2
     hypoechoic
   Fine granular, irregular, strongly        1             0
     hypoechoic
   Roughly granular, irregular,              1             0
     strongly hypoechoic
   Roughly granular, irregular,              3             0
     strongly hypoechoic
Shapes
   Flat                                      0             0
   Thickened, lobulated, fused              15             0
   Round, weak                               0             2

COPYRIGHT 1990 American College of Chest Physicians
COPYRIGHT 2004 Gale Group