|
|
||||||||
Ann Thorac Surg 2004;78:238-244
© 2004 The Society of Thoracic Surgeons
a Department of Surgery, Toneyama National Hospital, Toneyama 5-1-1, Toyonaka 560-8552, Osaka, Japan
b Department of General Thoracic Surgery (E1), Osaka University Graduate School of Medicine, Yamadaoka 2-2, Suita 565-0871, Osaka, Japan
c Department of Surgery, National Kinki Central Hospital for Chest Diseases, Nagasonecho 1180, Sakai 591-8555, Osaka, Japan
d Department of Surgery, Osaka Prefectural Habikino Hospital, Habikino 3-7-1, Habikino 583-8588, Osaka, Japan
e Department of Mathematical Health Science, Osaka University Graduate School of Medicine, Yamadaoka 1-7, Suita 565-0871, Osaka, Japan
Accepted for publication February 6, 2004.
* Address reprint requests to Dr Inoue, Department of General Thoracic Surgery (E1), Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan
e-mail: masa{at}surg1.med.osaka-u.ac.jp
| Abstract |
|---|
|
|
|---|
METHODS: Retrospective survival analysis was performed using 128 patients who underwent curative pulmonary resection.
RESULTS: The overall 5-year survival rate was 45.3%. Univariate analysis showed the number of metastases, location (unilateral or bilateral), prethoracotomy carcinoembryonic antigen (CEA) level, hilar or mediastinal lymph-node metastasis, and Dukes' stage to be considerable prognostic factors. Among these, Dukes' A for the primary lesion and unilateral pulmonary metastasis were shown to be independent predictors of longer survival by multivariate analysis (p = 0.0093 and p = 0.0182, respectively). In patients treated with both pulmonary and hepatic metastastasectomies, a better prognosis was observed in those who received metachronous resection. Recurrence after a pulmonary metastasectomy frequently occurred in the thorax and the 3-year survival rate was 44.6% in patients who underwent a repeat thoracotomy.
CONCLUSIONS: Patients with unilateral metastasis and Dukes' A for the primary tumor benefit most from the resection of pulmonary metastasis from colorectal carcinoma. Further, the number of metastases, prethoracotomy CEA level, and hilar or mediastinal lymph-node involvement should be considered to determine the operative indication. Finally, periodic follow-up examinations for thoracic recurrence should be carefully performed as these patients may have a heightened risk of requiring a repeat thoracotomy.
| Introduction |
|---|
|
|
|---|
We previously reported that the resection of pulmonary metastasis from colorectal carcinoma is effective in patients with a normal CEA level and without lymph-node metastasis [8], although several points, such as the relationship between survival and the number of metastatic tumors and/or stage of the primary tumor, the recurrence pattern after pulmonary metastasectomy, and the effectiveness of repeat pulmonary metastasectomy, as well as others, are still unclear because of the small number of patient cases. In this report, we conducted a retrospective multicenter analysis of patients treated with pulmonary metastasectomy for colorectal carcinoma. From our results, we discuss the prognostic factors, patterns of hilar or mediastinal lymph-node involvement, the outcomes of patients with pulmonary metastasis found during treatment for the primary lesion, the surgical results of patients treated for both hepatic and pulmonary metastases, the recurrence sites after pulmonary metastasectomy, and the significance of repeat pulmonary metastasectomy in an attempt to define which patients can benefit from a metastasectomy associated with colorectal carcinoma.
| Patients and methods |
|---|
|
|
|---|
| Results |
|---|
|
|
|---|
|
|
|
In the last decade, we have performed lymph-node dissection, or at least sampling, and found mediastinal or hilar lymph-node metastases in 21 out of 89 patients (23.6%). The 5-year survival rate for patients without lymph-node involvement was 50.8% as compared with 19.3% for node-positive patients (p = 0.0047). Most positive nodes were located at the metastatic lobes or nearby sites in the mediastinum [9]. The level of node involvement, whether hilar or mediastinal, had no prognostic implications.
We also analyzed patient survival according to Dukes' stage as determined during surgery for a primary colorectal lesion in 95 patients (Dukes' A, 29; B, 8; C, 48; D, 10). The 5-year survival rates for patients with Dukes' A, B, C, and D were 68.7%, 38.1%, 31.9%, and 27.8%, respectively. Because the number of Dukes' B and D patient cases was small, we compared Dukes' A versus BD and found that the 5-year survival rate for those with Dukes' A was 68.7% as compared with 32.8% for those with BD (Fig 3).
|
|
|
Nine patients were diagnosed with colorectal carcinoma with pulmonary metastasis before treatment for the primary lesion (ie, Dukes' D, colorectal carcinoma). The primary site was the colon in 6 patients and the rectum in 3 patients. The tumor diameter ranged from 110 cm (average 3 cm). A lobectomy was performed in 5 patients and 4 patients underwent partial resection. The survival for these patients ranged from 1362 months (MST: 26 months), whereas 3 patients survived for more than 4 years. Recurrence after pulmonary resection occurred in the lung in 3 patients and the brain in 2 patients.
We also thoroughly surveyed the outcomes after pulmonary metastasectomy. Interestingly, intrathoracic recurrence, especially in the lung, was most frequent and occurred in 46 (53%) of the patients that showed evidence of recurrence. Recurrence in the liver, brain, bone, and peritoneum was detected in 12, 9, 9, and 8 patients, respectively. Among those suffering from intrathoracic recurrence, 20 patients underwent a repeat thoracotomy and their survival curve is shown in Figure 4. The 3- and 5-year survival rates after the second operation were 44.6% and 22.3%, respectively, with an MST of 35.4 months.
|
| Comment |
|---|
|
|
|---|
Sakamoto reported that no considerable differences in survival rate were found among patient groups for solitary, ipsilateral multiple, and bilateral metastases [13]. In this study, although only 12 patients (9.4%) suffering bilateral metastases underwent a metastasectomy, the 5-year survival rate was substantially worse than that for patients with unilateral metastasis. It is expected that the survival rate will decrease in association with an increase in the number of metastases in patients with bilateral metastases even if all metastatic nodules are macroscopically extirpated. The mean number of lesions resected in these patients with bilateral metastases was 4.2, therefore, we considered that such bilateral multiple metastatic patients had a minimal depressive effect on the survival curve of the bilateral metastases group. Furthermore, multivariate analysis revealed that tumor location (unilateral or bilateral) and not metastatic number was a most powerful and independent prognostic factor suggesting that patients with hemithoracic pulmonary lesions could be good candidates for resection. However, patients with bilateral metastases will gain no surgical benefit.
We found that patients with a solitary pulmonary metastasis had a considerably better survival rate than those with two or more by univariate analysis, although multivariate analysis did not show the metastatic number to be an independent prognostic factor. The number of pulmonary metastases has often been reported as an independent prognostic factor [1012, 16, 19, 20], whereas some studies have found no marked difference [7, 14]. CT imaging has recently been improved and is able to detect a tiny nodule allowing for preoperative knowledge of the exact number of pulmonary metastases. Thus, an accurate preoperative evaluation of nodule number, which may have an influence on prognosis, has become increasingly important. The size of the metastases was not a prognostic factor in our study and most other reports found no substantial relationship between tumor size and patient prognosis [12, 1416, 20], although Shirouzu [10] and Okumura [11] reported that patients with pulmonary metastases 3 cm or greater in diameter had worse survival than those with tumors less than 3 cm. In the analysis by Shirouzu, the results may have been influenced by the bias that 57% of the patients with metastases 3 cm or greater in diameter suffered from lymph-node metastases.
Sakamoto [13], Saito [18], Rena [21], Pfannschmidt [22], Higashiyama [23], and Headrick [24] have reported CEA to be a prognostic factor in patients with pulmonary metastasis from colorectal carcinoma. In this study, the outcome of patients with an elevated CEA level was inclined to be poor compared with those having a normal level (Table 2). However, when analyzing with a twofold cutoff value limit, the survival of the CEA high group was considerably worse than that of the CEA low group with a less than twofold cutoff (Table 2). Interestingly, recent reports by Saito [18] and Higashiyama [23] similarly showed that a prethoracotomy CEA greater than 10 ng/ml was a predictor of poor prognosis, whereas their cutoff values were set at 5 ng/ml. Namely, poor prognosis could be predicted in patients with a distinct elevation of CEA.
As for hilar or mediastinal lymph-node involvement in patients with pulmonary metastasis from colorectal carcinoma, the previous analysis from a single institute resulted in a substantially better prognosis for patients without lymph-node metastasis [8]. Okumura, Saito, Pfannschmidt, and Headrick also reported the prognostic impact of lymph-node involvement [11, 18, 22, 24]. We confirmed those results in this study using a large number of patients and concluded that lymph-node metastasis was a prognostic factor (Table 2). However, multivariate analysis showed lymph-node involvement not to be an independent prognostic predictor when analyzed together with Dukes' stage (Table 3). We interestingly observed that most patients with lymph-node metastases were diagnosed with Dukes' BD for the primary lesion except for 2 patients. There may be some biological relationship between the invasiveness of the primary tumor and the intrathoracic lymph-node involvement. In addition, the metastatic site of nodes corresponded to the regional segmental or lobar nodes supporting the hypothesis of node metastasis from pulmonary metastatic lesions. Thus, we suppose that sufficient evaluation of intrathoracic lymph nodes, for instance, using CT, fluorodeoxyglucose-positron emission tomography, fiberscopic transbronchial needle aspiration, or mediastinoscopy is important in aiding the decision of the operative indication, as lymph-node involvement is not rare in patients with pulmonary metastasis from colorectal carcinoma. The results of this study showed that patients with proven node metastasis had no operative indication. We recommend lymph-node dissection or sampling to predict prognosis.
The invasiveness of colorectal carcinoma is generally classified using Dukes' stage in which a tumor limited to the muscularis propria is considered to be Dukes' A, that beyond the muscularis propria but without lymph-node metastasis is considered to be Dukes' B, that with lymph-node metastasis is considered to be Dukes' C, and a tumor remaining locally or with distant metastasis is considered to be Dukes' D [25]. Dukes' classification is reported to be a prognostic factor for patients undergoing hepatic metastasectomy for colorectal carcinoma [26], however, the prognostic impact of Dukes' stage for pulmonary metastasis has not been sufficiently assessed. Sauter found a better, though statistically insignificant, disease-free survival in Dukes' B compared with Dukes' C patients who underwent pulmonary resection of metastatic colorectal carcinoma [27]. We found the Dukes' stage to be an independent prognostic factor (Table 3). These results suggest that the degree of local invasiveness can influence the prognosis of patients with distant metastasis in colorectal carcinoma. We believe that Dukes' classification is useful to predict prognosis and should be considered for operative indication in patients with pulmonary metastasis.
Regarding the pulmonary and hepatic metastasectomy procedure, Gough reported 9 patients with a median survival of 27 months [28] and Lehnert noted that sequential resection was warranted in carefully selected patients with a 2-year survival rate of 70% [29]. Kobayashi also reported a 5-year survival rate of 31% in 47 patients and found that patients with a solitary pulmonary metastasis or a small number of hepatic metastases were good candidates for metastasectomy [30]. Further, a study by Robinson revealed that patients treated with metachronous resection survived longer as compared with synchronous resection [31]. Similarly, patients treated with metachronous resection showed a reasonably good outcome in this study. Thus, we consider that patients suffering from metachronous metastases to the liver and lung are potential operative candidates.
The thorax was the most frequent site of recurrence after pulmonary metastasectomy. Mori also reported similar results which showed intrathoracic recurrence in 58% of 35 patients including 79% in the lung and 21% in the lymph nodes [14]. From these results and the feasible outcome after repeated metastasectomy, we concluded that patients undergoing pulmonary metastasectomy from colorectal carcinoma should be carefully followed up, especially in the chest as they may have a heightened need for repeat thoracotomy. Our recommendation for follow-up is a chest roentgenogram at least every 3 months and a CT scan every year. The serum CEA level may also be useful to predict recurrence, because 5 patients who underwent repeated metastasectomy showed CEA elevation. However, no applicable prognostic predictor to select candidates for repeat metastasectomy was found among the variables tested in this study. Therefore, we would consider repeat metastasectomy for good-risk patients whose lesions are resectable.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y. Tanaka, Y. Maniwa, W. Nishio, M. Yoshimura, and Y. Okita The optimal timing to resect pulmonary metastasis Eur. J. Cardiothorac. Surg., June 1, 2008; 33(6): 1135 - 1138. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Soyka, P. Veit-Haibach, K. Strobel, S. Breitenstein, A. Tschopp, K. A. Mende, M. P. Lago, and T. F. Hany Staging Pathways in Recurrent Colorectal Carcinoma: Is Contrast-Enhanced 18F-FDG PET/CT the Diagnostic Tool of Choice? J. Nucl. Med., March 1, 2008; 49(3): 354 - 361. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. M. Pawlik, R. D. Schulick, and M. A. Choti Expanding Criteria for Resectability of Colorectal Liver Metastases Oncologist, January 1, 2008; 13(1): 51 - 64. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Nakajima, T. Murakawa, T. Fukami, A. Sano, M. Sugiura, and S. Takamoto Is Finger Palpation at Operation Indispensable for Pulmonary Metastasectomy in Colorectal Cancer? Ann. Thorac. Surg., November 1, 2007; 84(5): 1680 - 1684. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Pfannschmidt, H. Dienemann, and H. Hoffmann Surgical Resection of Pulmonary Metastases From Colorectal Cancer: A Systematic Review of Published Series Ann. Thorac. Surg., July 1, 2007; 84(1): 324 - 338. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Treasure, M. Utley, and I. Hunt When professional opinion is not enough BMJ, April 21, 2007; 334(7598): 831 - 832. [Full Text] [PDF] |
||||
![]() |
G. Veronesi, F. Petrella, F. Leo, P. Solli, P. Maissoneuve, D. Galetta, R. Gasparri, G. Pelosi, T. De Pas, and L. Spaggiari Prognostic role of lymph node involvement in lung metastasectomy J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 967 - 972. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Tanaka, H. Shimada, M. Ueda, K. Matsuo, I. Endo, and S. Togo Long-Term Characteristics of 5-Year Survivors After Liver Resection for Colorectal Metastases Ann. Surg. Oncol., April 1, 2007; 14(4): 1336 - 1346. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Furak, I. Trojan, T. Szoke, L. Tiszlavicz, J. Eller, and G. Lazar Visceral pleural infiltration as a negative prognostic factor in lung metastasis Interactive CardioVascular and Thoracic Surgery, April 1, 2007; 6(2): 196 - 199. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Takahashi, K. Nagai, N. Saito, M. Konishi, T. Nakagohri, N. Gotohda, M. Nishimura, J. Yoshida, and T. Kinoshita Multiple Resections for Hepatic and Pulmonary Metastases of Colorectal Carcinoma Jpn. J. Clin. Oncol., March 1, 2007; 37(3): 186 - 192. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Yedibela, P. Klein, K. Feuchter, M. Hoffmann, T. Meyer, T. Papadopoulos, J. Gohl, and W. Hohenberger Surgical Management of Pulmonary Metastases from Colorectal Cancer in 153 Patients Ann. Surg. Oncol., November 1, 2006; 13(11): 1538 - 1544. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Poston, R. Adam, and J.-N. Vauthey Downstaging or Downsizing: Time for a New Staging System in Advanced Colorectal Cancer? J. Clin. Oncol., June 20, 2006; 24(18): 2702 - 2706. [Full Text] [PDF] |
||||
![]() |
C. Casali, A. Stefani, E. Storelli, and U. Morandi Prognostic factors and survival after resection of lung metastases from epithelial tumours Interactive CardioVascular and Thoracic Surgery, June 1, 2006; 5(3): 317 - 321. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. C. Hanna, T. A. Ponsky, G. D. Trachiotis, and S. M. Knoll Colon Cancer Metastatic to the Lung and the Thyroid Gland Arch Surg, January 1, 2006; 141(1): 93 - 96. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Inoue Reply Ann. Thorac. Surg., August 1, 2005; 80(2): 791 - 791. [Full Text] [PDF] |
||||
![]() |
C.S. Pramesh, R. C. Mistry, and V. V. Upasani Is Identification of Prognostic Factors for Lung Metastasectomy Sufficient to Establish Selection Criteria? Ann. Thorac. Surg., August 1, 2005; 80(2): 790 - 790. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |