| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Section of Thoracic Surgery (Dr. Cerfolio), Department of Epidemiology (Dr. Bryant), University of Alabama School of Public Health; and Department of Medicine (Dr. Eloubeidi), Division of Gastroenterology and Hepatology, University of Alabama at Birmingham, Birmingham, AL.
Correspondence to: Robert J. Cerfolio, MD, Associate Professor of Surgery, Chief of Thoracic Surgery, Division of Cardiothoracic Surgery, University of Alabama at Birmingham, 1900 University Blvd, THT 712, Birmingham, AL 35294; e-mail: Robert.cerfolio{at}ccc.uab.edu
Abstract
Background: Despite normal mediastinal (N2) lymph nodes shown on positron emission tomography (PET) and CT, some physicians routinely perform mediastinoscopy and/or endoscopic ultrasound fine-needle aspiration (EUS-FNA) in patients with non-small cell lung cancer (NSCLC).
Methods: A prospective trial on patients with NSCLC who were clinically staged N2 negative by both integrated PET/CT and CT scan. All underwent mediastinoscopy and EUS-FNA and if N2 negative underwent thoracotomy with thoracic lymphadenectomy.
Results: There were 153 patients (107 men). Of these, 136 patients were clinically staged N0 and 17 patients were clinically staged N1. Of the 136 patients who were staged as N0, 5 patients (3.7%) had positive EUS-FNA results (three in the subcarinal node), and 4 patients (2.9%) had positive mediastinoscopy results (all in the #4R node; one was N3). Six of the remaining 127 patients (4.7%) had N2 disease after resection. Seventeen patients were clinically staged as N1 by integrated PET/CT. Four patients (23.5%) had positive EUS-FNA results (two in the subcarinal node), 3 patients (17.6%) had positive mediastinoscopy results (all in #4R node; two were N2 and one was N3), and none of the remaining 10 patients had N2 disease after resection. Patients with unsuspected N2 disease were twice as likely (relative risk, 2.1; 95% confidence interval, 1.24 to 2.51; p = 0.02) to have a maximum standardized uptake value (maxSUV) > 10 and poorly differentiated cancer (relative risk, 2.1; 95% confidence interval, 1.14 to 2.38; p = 0.03).
Conclusion: We do not recommend routine mediastinoscopy or EUS-FNA in patients who are clinically staged as N0 after both integrated PET/CT and CT. However, these procedures should both be considered in patients clinically staged as N1 after PET/CT, and/or in those with adenocarcinoma, upper-lobe tumors, or tumors with a maxSUV 
10.
Key Words: endoscopic ultrasound • mediastinoscopy • non-small cell lung cancer • N2 negative
Non-small cell lung cancer (NSCLC) claims the lives of more Americans each year than those taken by breast, colon, liver, prostate, and melanoma combined. It is the leading cause of cancer deaths worldwide and remains a pandemic. Yet, even after careful clinical staging followed by complete resection, 5-year survival rates for patients with pathologic stage Ia, Ib, II, and IIIa disease are much lower than for other solid-organ cancers, and are only 67%, 57%, 47%, and 23%, respectively.1 These poor surgical results for NSCLC may be due to inaccurate staging prior to resection (missed metastatic disease) or inaccurate staging at the time of surgery (a lack of mediastinal lymph node dissection). For this reason, surgeons often choose to perform mediastinoscopy in all patients prior to thoracotomy to ensure stage IIIa (from N2) disease is not missed. The presence of an N2 node is an indication for neoadjuvant therapy. Studies2345 have shown that if a patient receives neoadjuvant therapy prior to resection, there will be an increase in survival. We and others6 have shown the inaccuracies of clinical staging compared to pathologic staging. We have also shown the relative frequency of metastatic N2 disease in the posterior mediastinal lymph nodes. These nodes are not accessible via mediastinoscopy but are easily and accurately sampled by endoscopic ultrasound fine-needle aspiration (EUS-FNA). Thus, some7 argue that both mediastinoscopy and EUS-FNA should be performed routinely in all patients prior to resection. However, the practice of routine mediastinoscopy has been shown not to be cost-effective. The purpose of this study was to evaluate the incidence of unsuspected N2 disease by performing mediastinoscopy (which assesses the anterior mediastinal N2 nodes) as well as that of EUS-FNA (which samples the posterior mediastinal N2 nodes) in patients with NSCLC who are clinically staged as N2 negative. These two staging tests were both performed in patients prior to pulmonary resection who had been clinically staged as N2 negative after integrated positron emission tomography (PET)/CT and CT scan.
Materials and Methods
This is a prospective study between September 2004 and September 2005. All consecutive patients with clinically staged I or II (N2 negative) NSCLC who presented to one general thoracic surgeon were eligible. A clinical stage was determined using the TNM classification8 after a preoperative F-18 fluorodeoxyglucose (FDG)-integrated PET/CT and a CT using 5-mm collimated cuts. Both tests were mandatory for entry into this study and had to be performed within 4 weeks of nodal biopsy and/or resection. Both procedures were performed on separate days on an outpatient basis. Mediastinoscopy was performed under general anesthesia, and EUS-FNA was performed using IV sedation. In addition, the entry criteria included pathologic confirmation of NSCLC either before surgery or via positive mediastinoscopy or EUS-FNA results, or after thoracotomy and pulmonary resection. Patients < 19 years of age, those who received neoadjuvant radiochemotherapy, those who refused to be in the study, or those could not undergo mediastinoscopy or EUS-FNA were excluded. Patients who were found to have benign disease after resection were also eliminated. Patients with clinically suggested N2, N3, or M1 disease were also excluded from this study even if those test results turned out to be false-positive. A power analysis found that approximately 140 patients would be needed to answer the study question. This study and the prospective database used to collect and store the data were both approved by the University of Alabama at Birmingham Institutional Review Board. Patient consent was obtained for entry in this study and into the database.
Imaging
Integrated PET/CT scans were performed (GE Discovery LS PET-CT Scanner; GE; Milwaukee, WI). Patients were asked to fast for 4 h and then subsequently received 555 MBq (15 mCi) of FDG IV followed by PET after 1 h. The scans were performed from the skull base to mid-thigh level. A CT examination was used for attenuation correction of PET images. The scanning time for emission PET was 5 min per bed position. Iterative reconstruction with CT attenuation correction was performed. The most recent CT scan of the chest was available for visual correlation. Maximum standardized uptake value (maxSUV) was determined by drawing regions of interest (ROIs) on the attenuation-corrected FDG-PET images around the primary tumor. It was then calculated by the software contained within the PET or PET/CT scanner by the following formula9:
 |
Mediastinoscopy and EUS-FNA
Patients were carefully staged pathologically. Both mediastinoscopy and EUS-FNA were performed in all patients even if one test result yielded N2 disease. During mediastinoscopy, lymph nodes in the anterior paratracheal area, nodal stations 2R, 4R, 2L, 4L, and the superior portion of 7 were evaluated. All stations were investigated; if lymph nodes were identified, biopsies were performed irrespective of their appearance. EUS-FNA was performed as previously described,10 and biopsies of the posterior mediastinal lymph nodes (subcarinal [7], periesophageal [8], and inferior pulmonary ligament [9]) were performed in all patients. All nodal stations were evaluated and sampled unless no nodal tissue was present at that station. If all lymph node stations appeared normal by endoscopic ultrasound (EUS), at least one needle aspirate of the subcarinal lymph node station was performed in all patients. If there was no evidence of N2 or N3 disease after mediastinoscopy or EUS-FNA, patients underwent thoracotomy, pulmonary resection, and complete thoracic lymphadenectomy. Pathologic review was performed via standard techniques, and immunohistochemical staining was employed if deemed necessary by the pathologist. The pathologic stage was assessed using the international staging system.
Statistical Analysis
Data were stored using Excel (Microsoft; Seattle WA) and imported into SAS version 8.02 for analysis (SAS Institute; Cary NC). 
2 analysis or Fisher exact test were used among discrete variables to produce a relative risk ratio and examine for potentially significant differences among those with and without N2 disease. For continuous variables such as the median maxSUV and the nodule size, the Student t test or the Mann-Whitney U test were used to compare means for nonnormally distributed variables. The Student t test was used to compare means for normally distributed variables.
Results
Patient characteristics are shown in Table 1 . There were no procedure-related complications or deaths after mediastinoscopy or EUS-FNA. Three patients refused entry into the study and were eliminated from further analysis. The results for patients clinically staged as N0 and N1 after integrated PET/CT are shown in Figure 1 . Thirteen patients had unsuspected benign pulmonary nodules that were resected during this study, and they were eliminated from further analysis. As shown, the incidence of N2 disease discovered via mediastinoscopy in a patient clinically staged as N0 was only 2.9%. It was only 3.6% via EUS-FNA. The prevalence of N2 found via mediastinoscopy in a patient clinically staged as N1 by integrated PET/CT was 17.6%. It was 23.5% via EUS-FNA. Many variables were evaluated, and some characteristics of the 22 patients who had unsuspected N2 disease are shown in Table 2 . Table 3 depicts the results of the two staging procedures based on the lobar location of the primary tumor. It illustrates the incidence of a positive test result if mediastinoscopy and/or EUS-FNA was performed selective for tumors in certain lobar locations.
|
|
|
|
The overall 5-year survival rate of patients with NSLCC is only 13%. However, survival may be improved with the identification of N2 disease prior to thoracotomy and pulmonary resection. Since the early reports by Roth et al23 and Rosell et al,45 the standard of care for patients with N2 disease has been neoadjuvant chemoradiotherapy. Resection is usually reserved for those who are downstaged (ie, have resolution of their N2 disease). Although these small studies have never been confirmed by larger multi-institutional trials and the benefits of neoadjuvant therapy for microscopic N2 disease have never been shown, this treatment modality remains our current standard of care. Thus, it is disheartening that, even after performing integrated PET/CT in all patients prior to resection, we still found a 14% incidence of unsuspected N2 disease in our prospective study in 2005.6 This incidence is similar to that reported by De Leyn et al11 in 1996. Because of this relatively high incidence, Haddad et al7 studied the routine use of mediastinoscopy in 149 patients clinically staged as N0 and only found a 2.7% positive rate for N2 disease. Similarly, we report a 2.9% incidence. Based on this low incidence, Haddad and associates7 concluded that mediastinoscopy was not cost-effective in this group of patients. Because we showed that unsuspected N2 disease was more often located in the posterior mediastinal lymph nodes6 that are not accessible via mediastinoscopy, we performed this prospective study and added EUS-FNA to the algorithm.
Our prospective study found that that only 2.9% of patients clinically staged as N0 after integrated PET/CT and CT had positive mediastinoscopy results and only 3.7% had positive EUS-FNA results. Thus, we cannot recommend the routine use of these two tests in patients clinically staged as N0. This high negative predictive value of PET for patients judged as N0 has been shown before.12 However, in patients with clinical N1 disease suggested by integrated PET/CT, we found a relatively high incidence (total of 17.6% after mediastinoscopy and 23.5% after EUS-FNA) of unsuspected N2 disease. Thus, both of these tests seem appropriate for these patients.
There is a distinct predilection for the location of N2 disease based on the lobar location of the primary tumor in patients with NSCLC. We recently reported a retrospective review of almost 1,000 patients13 and found almost identical results to earlier reports from Naruke et al14 and Kotoulas et al.15 All three of these studies found that tumors in the right upper lobe were most likely to metastasize to the 4R and 2R stations; tumors in the right middle lobe to the 7 station; tumors in the right lower lobe to the 4R and 7 stations; tumors in the left upper lobe to the 5 and 6 stations; and finally tumors in the left lower lobe most commonly to the 5 and 7 stations. If we had known this data beforehand and changed the design of this study and limited mediastinoscopy to only those patients with clinically staged N0 disease with a right upper lobe lesion, we would have found a 10.4% positive rate for mediastinoscopy. Similarly, had we only performed EUS-FNA in patients staged N0 with tumors only in the right lower lobe tumor, EUS-FNA results would have been positive 15.4% of the time. This type of a targeted approach may increase the cost-effectiveness of performing mediastinoscopy and EUS-FNA even when patients are clinically staged N0.
Knowing the lobar location of the primary may be a clue as to what nodal station to sample. This clue can be coupled with factors that predict who is more likely to have N2 disease as opposed to where it may occur. These factors include the maxSUV of the primary as determined by integrated PET/CT, tumor histology, degree of differentiation, size,16 and maybe its location in the lobe (central as opposed to peripheral). Perhaps consideration of all of these factors may help guide the most appropriate test and help identify which patients are more likely to have unsuspected N2 disease, and in what nodal station they are most likely to occur.
In conclusion, we do not recommend the routine use of mediastinoscopy or EUS-FNA in all patients with NSCLC who are clinically staged N0 after integrated PET/CT. However, we do recommend both mediastinoscopy and EUS-FNA in patients clinically staged as N1 after integrated PET/CT. Importantly, a more targeted search for unsuspected N2 disease may make these biopsy procedures more cost-effective. We recommend a prospective multi-institutional study that uses factors known prior to resection that may predict the likelihood of having N2 disease, such as the maxSUV of the tumor, the size, histology, the degree of tumor differentiation, and the location of the tumor in the lobe. This information coupled with knowing the lobar location of the primary may help direct which tests to order and in which patients. This should increase the positive predictive value and thus the cost-effectiveness of these minimally invasive but expensive nodal biopsy procedures. This approach may help increase the relatively poor survival of patients who have unsuspected N2 disease.
Footnotes
Abbreviations: EUS = endoscopic ultrasound; EUS- FNA = endoscopic ultrasound fine-needle aspiration; FDG = F-18 fluorodeoxyglucose; maxSUV = maximum standardized uptake value; NSCLC = non-small cell lung cancer; PET = positron emission tomography; ROI = region of interest
This article was accepted for poster presentation at the Western Thoracic Surgical Association Meeting, June 2006.
The authors have no financial conflicts of interest to disclose.
Received for publication May 18, 2006. Accepted for publication June 20, 2006.
References
This article has been cited by other articles:
|
|
 |
|
  M. S. Sawhney, Y. Bakman, A. M. Holmstrom, D. B. Nelson, F. A. Lederle, and R. F. Kelly Impact of Preoperative Endoscopic Ultrasound on Non-small Cell Lung Cancer Staging Chest, September 1, 2007; 132(3): 916 - 921. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
