Asian Cardiovasc Thorac Ann 2004;12:125-129
© 2004 Asia Publishing EXchange Ltd
Combined Procedures for Mediastinal Staging in Non-small Cell Lung Cancer
Masaki Tomita, MD,
Yasunori Matsuzaki, MD,
Masao Edagawa, MD,
Tetsuya Shimizu, MD,
Masaki Hara, MD,
Toshio Onitsuka, MD
Department of Surgery II, Miyazaki Medical College, Miyazaki, Japan
For reprint information contact: Masaki Tomita, MD Tel: 81 985 852 291 Fax: 81 985 855 563 Email: mtomita{at}post.miyazaki-med.ac.jp Department of Surgery II, Miyazaki Medical College, Kihara 5200, Kiyotake, Miyazaki, 889-1692, Japan.
 |
ABSTRACT
|
|---|
We investigated whether the combined use of computed tomography, thallium-201 single photon emission computed tomography and serum carcinoembryonic antigen level improves preoperative non-invasive mediastinal. 128 consecutive non-small cell lung cancer patients (85 adenocarcinomas, 31 squamous cell carcinomas and 12 others) who underwent a surgical resection were enrolled in this study. The results of the combined procedures were compared with the pathologic findings. Our results showed that the combined evaluation of mediastinal nodal involvement with the three procedures might increase underestimation, but decrease overestimation as compared to computed tomography alone. Thallium-201 single photon emission computed tomography for patients with enlarged nodes at computed tomography showed 81.3% and 100% of positive predictive value in overall and squamous cell carcinoma patients, respectively. The negative predictive value of thallium-201 single photon emission computed tomography for patients without enlarged nodes at computed tomography was highly accurate in adenocarcinoma (93.9%) as well as squamous cell carcinoma (94.4%). Combining computed tomography findings and serum carcinoembryonic antigen level had a poor predictive value. However, in patients with adenocarcinoma, a negative examination was highly accurate (95.2%). In conclusion, our results show a trend that combined use of the three procedures might improve non-invasive mediastinal staging in non-small cell lung cancer.
|
INTRODUCTION
|
|---|
The most important predictor of outcome in patients with non-small cell lung cancer is the presence or absence of mediastinal lymph node involvement (N2) with tumor.1,2 Therefore, preoperative diagnosis of N2 disease is very important in deciding a treatment strategy. Mediastinoscopic biopsy has been a mainstay in N2 disease diagnosis. However, mediastinoscopy is an invasive examination and requires operative intervention. A need exists for an accurate, non-invasive means of staging non-small cell lung cancer. Computed tomography (CT) is the imaging modality used most commonly for preoperative non-invasive mediastinal staging.3–8 However, CT criteria have been reported to yield a high false-positive result rate and some investigators reported that one third of patients with enlarged mediastinal lymph nodes on CT scan did not have nodal involvement.9,10 Therefore, the reliability of CT size criteria for node metastasis remains controversial. Imaging with whole-body positron emission tomography (PET) has been reported to be more accurate for the diagnosis of mediastinal node metastasis.11 Unfortunately, this technology is currently unavailable in many centers, including our institute. Clinical diagnosis of mediastinal nodal status (cN) is currently performed based on CT findings in many centers.
Other previous studies reported that thallium-201 single photon emission computed tomography (Tl-201 SPECT) is useful in evaluation of mediastinal node metastasis in non-small cell lung cancer.12–14 Elevated preoperative serum carcinoembryonic antigen (CEA) levels have been also reported as a clinical predictor of N2 disease in non-small cell lung cancer.15,16
In spite of these studies, very little attention has been given to the diagnostic value of the combined use of CT, Tl-201 SPECT and serum CEA levels in the evaluation of mediastinal node metastasis. What remains to be seen is whether the combined use of these examinations improves preoperative non-invasive mediastinal staging or not. The present study examines this question.
|
PATIENTS AND METHODS
|
|---|
128 consecutive patients (85 adenocarcinomas, 31 squamous cell carcinomas and 12 others) with primary lung cancer who underwent surgical intervention at our institute recently were included in this study. The following patients were excluded: (1) patients who had not received complete resection which consisted of either a lobectomy or a pneumonectomy together with the regional lymph nodes dissection; (2) patients with a primary tumor directly invading the mediastinum. There were 89 men and 39 women. The ages ranged from 20 to 80 years, with an average of 64.6 years. All patients had undergone preoperative examinations, including CT scan, Tl-201 SPECT and serum CEA level, and received a surgical resection with mediastinal lymph node dissection. We evaluated the reliability of CT findings in the diagnosis of N2 disease on the basis of size criteria – that is, mediastinal lymph nodes 1.0 cm or larger in the shortest axis were diagnosed as metastatic nodes. Tl-201 SPECT images were obtained 15 minutes (early scans) and 3 hours (delayed scans) after intravenous administration of 222 MBq (6mCi) of Tl-201. All Tl-201 SPECT images were interpreted by nuclear radiologists. When there was at least one area of definitely increased radioactivity in the mediastinum, the images were considered positive. Serum CEA level was measured by means of the two-site immunoenzymometric assay (Tosoh Inc., Yamaguchi, Japan). The normal upper limit for this assay was 5 ng•mL–1. The time interval between these examinations and surgical resection was less than one month in all the patients. In our institute, we recently started preoperative mediastinoscopic examination, and 12 of overall 128 patients received preoperative mediastinoscopy. For the other remaining patients, preoperative mediastinoscopic examination was not performed, even in patients with clinical N2 (cN2) disease.
Surgically resected mediastinal lymph nodes were pathologically examined for metastasis in all patients. Pathologic N2 disease (pN2) was present when any mediastinal node was histologically involved. The results of combining CT, Tl-201 SPECT and serum CEA level in preoperative diagnosis for mediastinal nodal involvement and the discrepancy between clinical and pathological findings in N factor were analyzed. A true-positive result was documented when a patient had both cN2 and pN2 findings. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated by the following formulae:
The data of serum CEA level was analyzed using the Mann-Whitney U test. A McNemar test for correlated proportions (one-tailed p value) was performed to compare the effectiveness of combined examination versus CT alone in the detection of pN2 disease. Differences were considered significant when the p value was less than 0.05.
|
RESULTS
|
|---|
There were 31 among 128 overall patients (24.2%) with pN2 disease. Serum CEA level in pN2 patients was significantly higher than that in pN0-1 patients (16.1 ± 19.0 vs. 6.5 ± 8.69 ng•mL–1; p < 0.01).
The results of each single examination in evaluation of mediastinal node metastasis are shown in Table 1
. These results indicate that CT is superior among these 3 examinations. Table 1 also summarizes the results obtained by means of combined examination. Combined use of these 3 examinations showed higher false-negative rate and lower false-positive rate in comparison to CT alone. Overall, sensitivity, specificity, accuracy, PPV and NPV of the combination of the 3 examinations was 38.7%, 97.9%, 83.6%, 85.7% and 83.3% as compared to 71.0%, 90.7%, 85.9%, 71.0% and 90.7% with CT alone, respectively. These results indicate that specificity and PPV were higher but sensitivity and NPV were lower when combining CT, Tl-201 SPECT and serum CEA level rather than CT size criterion alone ( p < 0.05). The results of combining 2 of the 3 examinations also showed similar findings (Table 1). These results indicate that when combined examination is used, specificity and PPV for the diagnosis of pN2 disease is higher but sensitivity and NPV are lower compared to CT alone ( p < 0.05). In other words, the combined examination increases underestimation but decreases overestimation.
Since CT has been a mainstay in cN2 disease diagnosis, we investigated the additional diagnostic value of Tl-201 SPECT and CEA based on the CT findings. As shown in Table 2, the results of Tl-201 SPECT based on positive CT findings indicate that about 81.3% of patients with positive results at Tl-201 SPECT in addition to enlarged nodes at CT had pN2 disease ( p < 0.05).
In patients with negative results at CT, NPV of Tl-201 SPECT was 92.2% indicating that about 92% of Tl-201 SPECT negative patients with negative results at CT did not have pN2 disease. However this did not have statistical significance ( p = 0.22). The PPV of serum CEA level in addition to CT positive results did not show a favorable result (Table 2
). However, in patients with negative results at CT, serum CEA level showed 94.8% of NPV ( p < 0.01).
In patients with adenocarcinoma (n = 85), as shown in Table 2, Tl-201 SPECT showed relatively high false-positive rate in patients with positive results at CT, thus, PPV was 62.5%. Serum CEA level did not show a favorable result in patients with positive results at CT also. On the other hand, Tl-201 SPECT showed 93.9% of NPV in patients with negative results at CT but this did not reach statistical significance ( p = 0.22). Serum CEA level showed 95.2% of NPV in patients with negative results at CT ( p < 0.01).
In patients with squamous cell carcinoma (n = 31), the result of CT alone showed higher false-positive rate than that in adenocarcinoma (Table 2). When combined with Tl-201 SPECT, Tl-201 SPECT showed 100% of PPV in patients with positive results at CT but this did not have statistical significance ( p = 0.13). The PPV of serum CEA level in patients with enlarged nodes at CT was 71.4%. In squamous cell carcinoma patients with negative result at CT, Tl-201 SPECT showed 94.4% of NPV but this also did not have statistical significance ( p = 0.55). Serum CEA level did not show a high NPV in squamous cell carcinoma patients with negative result at CT (55.6%).
|
DISCUSSION
|
|---|
Several previous studies suggested that induction chemotherapy or chemoradiation followed by surgical resection resulted in pathologic down-staging and better long-term survival in patients with N2 disease.17 Therefore, accurate preoperative staging of the mediastinum is important in deciding lung cancer treatment strategy.
CT in the assessment of mediastinal lymph node involvement remains unsettled. The most accurate method of prethoracotomy assessment of N2 disease is a mediastinoscopy. However, many centers currently use CT of the mediastinum as a screening examination, opting for mediastinoscopy in those patients with positive mediastinal nodes at CT. The predictive value and accuracy of CT in determining mediastinal node involvement is lower in the presence of distal obstructive pneumonitis or old granulomatous disease.9,10 Furthermore, Suzuki et al18 reported that 17.9% of patients with negative mediastinal nodes by means of CT (cN0-1) had pN2 disease. Therefore, the reliability of CT size criteria for mediastinal node involvement remains controversial.
In addition to CT, Tl-201 SPECT has been reported to be useful in the evaluation of mediastinal node involvement in lung cancer.12–14 In our results, however, the use of Tl-201 SPECT alone in the evaluation of mediastinal node involvement was not favorable. Furthermore, the location of abnormal Tl-201 accumulation was only roughly estimated and could not be evaluated precisely according to the lymph node mapping system. Therefore, we believe that accurate location of the abnormal mediastinal lymph nodes is not always possible with Tl-201 SPECT alone.
Elevated preoperative serum CEA level has been also reported as a clinical predictor of N2 disease in non-small cell lung cancer.15,16 In our series, mean serum CEA level in pN2 patients was significantly higher than that in pN0-1 patients. However, use of the serum CEA level alone in evaluation of mediastinal node metastasis was not favorable. One of the reasons might be that the serum CEA level is known to vary depending on other factors such as smoking.19
Previous studies have been performed to evaluate the accuracy of CT or Tl-201 SPECT in the evaluation of mediastinal node involvement.3–10,12–14 However, most of these studies demonstrated a diagnostic value of single examination separately, and very little attention has been given to the combined examination with CT, Tl-201 SPECT and serum CEA level. Our results show that the combined examination of mediastinal nodal involvement with CT, Tl-201 SPECT and serum CEA level might increase underestimation, but decrease overestimation. From these results, the combined examination does not always improve the diagnostic value for N2 disease. Furthermore, because of the increase in underestimation, the combined examination might not be suitable for a screening examination of N2 disease. However, the combined examination showed higher PPV (about 80%).
In other words, about 80% of patients with positive results for 2 or 3 of these examinations have pN2 disease. It has been reported that preoperative induction therapy might improve the surgical outcome of pN2 patients.17–19 Therefore, we consider that it might be acceptable to perform preoperative induction therapy without performing mediastinoscopy for patients positive for combined examinations. On the other hand, mediastinoscopy is indicated in patients with enlarged mediastinal nodes at CT with negative results of Tl-201 SPECT and/or serum CEA level to compensate for the diagnostic limit of Tl-201 SPECT and serum CEA level.
In the present study, Tl-201 SPECT showed 81.3% of PPV in patients with a positive CT result. Although our results of Tl-201 SPECT alone did not show a favorable result in evaluation of mediastinal node involvement, Tl-201 SPECT might be useful for patients with enlarged nodes at CT. It is generally accepted that squamous cell carcinomas have a high proportion of enlarged tumor-free lymph nodes and the false-positive rate was higher by means of CT size criteria. Our results are similar. However, when combining with Tl-201 SPECT, PPV was significantly improved. Furthermore, NPV of Tl-201 SPECT based on the CT findings in overall, adenocarcinoma and squamous cell carcinoma patients was 92.2%, 93.9% and 94.4%, respectively. Taken together, our results show a trend that patients without pN2 disease might be predictable by combining CT and Tl-201 SPECT findings. However, some results did not have statistical significance. A couple of explanations can be considered. First, our results of CT alone were somewhat favorable compared to those of previous studies.9,10 Therefore, the additional effect of Tl-201 SPECT did not reach statistical significance. Secondly, the number of patients in each group is small. Further studies are required.
PPV of serum CEA level based on the CT findings did not indicate a favorable result. On the other hand, in patients with negative CT result, NPV of serum CEA level in overall, adenocarcinoma and squamous cell carcinoma patients was favorable – although some data did not have statistical significance. Therefore, combining CT findings and serum CEA level has a poor predictive value. However, a negative examination is highly accurate in patients with adenocarcinoma but not squamous cell carcinoma. Takamochiet al20 also recently reported that cases of pN0 disease can be predicted more correctly by combining preoperative serum CEA level and CT findings in pulmonary adenocarcinoma.
In conclusion, our results show a trend that combined use of CT, Tl-201 SPECT and serum CEA level might improve non-invasive mediastinal staging in non-small cell lung cancer.
View larger version (189K):
[in this window]
[in a new window]
|
Figure 1. Image of 54-year old woman with adenocarcinoma of the right upper lobe. (A) Contrast material-enhanced CT scan shows an enlarged pretracheal node. (B) Delayed Tl-201 SPECT shows increased abnormal accumulation corresponding to the mediastinal lesion. Her serum CEA level was 25.4 ng•mL–1.
|
|
|
REFERENCES
|
|---|
- Paulson DL, Reisch JS. Long-term survival after resection for bronchogenic carcinoma. Ann Surg
1976;184:324–32.[Medline]
- Martini N, Flehinger BJ. The role of surgery in N2 lung cancer. Surg Clin North Am
1987;67:1037–48.[Medline]
- Daly BD, Mueller JD, Faling LJ, Diehl JT, Bankoff MS, Karp DD, et al. N2 lung cancer: outcome in patients with false-negative computed tomographic scans of the chest. J Thorac Cardiovasc Surg
1993;105:904–10.[Abstract]
- Dillemans B, Deneffe G, Verschakelen J, Decramer M. Value of computed tomography and mediastinoscopy in preoperative evaluation of mediastinal nodes in non-small cell lung cancer. A study of 569 patients. Eur J Cardiothorac Surg
1994;8:37–42.[Abstract]
- Backer CL, Shields TW, Lockhart CG, Vogelzang R, LoCicero J 3rd. Selective preoperative evaluation for possible N2 disease in carcinoma of the lung. J Thorac Cardiovasc Surg
1987;93:337–43.[Abstract]
- Daly BD Jr, Faling LJ, Pugatch RD, Jung-Legg Y, Gale ME, Bite G, et al. Computed tomography. An effective technique for mediastinal staging in lung cancer. J Thorac Cardiovasc Surg
1984;88:486–94.[Abstract]
- Breyer RH, Karstaedt N, Mills SA, Johnston FR, Choplin RH, Wolfman NT, et al. Computed tomography for evaluation of mediastinal lymph nodes in lung cancer: correlation with surgical staging. Ann Thorac Surg
1984;38:215–20.[Abstract]
- Glazer GM, Orringer MB, Gross BH, Quint LE. The mediastinum in non-small cell lung cancer: CT-surgical correlation. AJR Am J Roentgenol
1984;142:1101–05.[Abstract/Free Full Text]
- Daly BD Jr, Faling LJ, Bite G, Gale ME, Bankoff MS, Jung-Legg Y, et al. Mediastinal lymph node evaluation by computed tomography in lung cancer. An analysis of 345 patients grouped by TNM staging, tumor size, and tumor location. J Thorac Cardiovasc Surg
1987;94:664–72.[Abstract]
- Dales RE, Stark RM, Raman S. Computed tomography to stage lung cancer. Approaching a controversy using meta-analysis. Am Rev Respir Dis
1990;141:1096–1101.[Medline]
- Valk PE, Pounds TR, Hopkins DM, Haseman MK, Hofer GA, Greiss HB et al. Staging non-small cell lung cancer by whole-body positron emission tomographic imaging. Ann Thorac Surg
1995;60:1573–82.[Abstract/Free Full Text]
- Yokoi K, Okuyama A, Mori K, Tominaga K, Miyazawa N, Takizawa I, et al. Mediastinal lymph node metastasis from lung cancer: evaluation with Tl-201 SPECT – comparison with CT. Radiology
1994;192:813–7.[Abstract/Free Full Text]
- Yamamoto Y, Nishiyama Y, Satoh K, Ohkawa M, Kameyama K, Hayashi E, et al. Comparative evaluation of Tc-99m MIBI and Tl-201 chloride SPECT in non-small-cell lung cancer mediastinal lymph node metastases. Clin Nucl Med
2000;25:29–32.[Medline]
- Yamamoto Y, Nishiyama Y, Fukunaga K, Satoh K, Ohkawa M. Tl-201 chloride and Tc-99m MIBI accumulation in lung adenocarcinoma in patients with and without distant metastases. Clin Nucl Med
2001;26:751–6.[Medline]
- Takamochi K, Nagai K, Suzuki K, Yoshida J, Ohde Y, Nishiwaki Y. Clinical predictors of N2 disease in non-small cell lung cancer. Chest
2000;117:1577–82.[Abstract/Free Full Text]
- Suzuki K, Nagai K, Yoshida J, Nishimura M, Takahashi K, Nishiwaki Y. Clinical predictors of N2 disease in the setting of a negative computed tomographic scan in patients with lung cancer. J Thorac Cardiovasc Surg
1999;117:593–8.[Abstract/Free Full Text]
- Faber LP, Kittle CF, Warren WH, Bonomi PD, Taylor SG 4th, Reddy S, et al. Preoperative chemotherapy and irradiation for stage III non-small cell lung cancer. Ann Thorac Surg
1989;47:669–77.[Abstract]
- Suzuki K, Nagai K, Yoshida J, Nishimura M, Takahashi K, Nishiwaki Y. Clinical predictors of N2 disease in the setting of a negative computed tomographic scan in patients with lung cancer. J Thorac Cardiovasc Surg
1999;117:593–8.[Abstract/Free Full Text]
- Fukuda I, Yamakado M, Kiyose H. Influence of smoking on serum carcinoembryonic antigen levels in subjects who underwent multiphasic health testing and services. J Med Syst
1998;22:89–93.[Medline]
- Takamochi K, Nagai K, Yoshida J, Suzuki K, Ohde Y, Nishimura M, et al. Pathologic N0 status in pulmonary adenocarcinoma is predictable by combining serum carcinoembryonic antigen level and computed tomographic findings. J Thorac Cardiovasc Surg
2001;122:325–30.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
