| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Thoracic Surgery, Scientific Institute H San Raffaele, Milan, Italy
Accepted for publication October 28, 2008.
* Address correspondence to Dr Melloni, Unità Operativa di Chirurgia Toracica Ospedale San Raffaele, Via Olgettina 60, Milan, 20132, Italy (Email: giulio.melloni{at}hsr.it).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods: Fifty-one patients staged N2 or N3 after positron emission tomography imaging underwent transbronchial needle aspiration.
Results: A malignant adenopathy was identified in 26 patients (51%) that were excluded from operation and referred for neoadjuvant chemotherapy or chemoradiotherapy according to the mediastinal status (N2 or N3), as determined on the positron emission tomography image. In the remaining 25 patients (49%), samples were considered adequate negative in 12 patients, inadequate in 11, or inconclusive in 2. These patients underwent mediastinoscopy. Mediastinoscopy showed N2 disease in 19 cases, and the patients received neoadjuvant chemotherapy. In the remaining 6 cases no mediastinal involvement was identified and patients underwent operation. Postoperatively, 5 patients were staged N0 and 1 was staged N2. For transbronchial needle aspiration, sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 76%, 100%, 100%, 33%, and 79%, respectively.
Conclusions: Transbronchial needle aspiration avoided a mediastinoscopy in approximately half of lung cancer patients referred for operation with positive mediastinal positron emission tomography, sparing the associated costs and risks of more invasive surgical procedures. The minimally invasive mediastinal staging algorithm that we proposed seems to be efficacious and easily applicable in clinical practice.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Mediastinoscopy is often used to exclude or confirm mediastinal lymph node metastases in patients who are suitable for surgical resection. Mediastinoscopy is still considered the gold standard for preoperative staging of the mediastinum in patients with NSCLC [2]. The procedure provides excellent access anatomically to the right and left paratracheal and subcarinal lymph nodes, which are the key to the staging of most NSCLCs [3]. However, with a sensitivity of about 80%, mediastinoscopy is also not always a perfect procedure, and its results may be surgeon-dependent [2, 3].
Transbronchial needle aspiration (TBNA) is a minimally invasive method for sampling mediastinal lymph nodes adjacent to the tracheobronchial wall. Although numerous studies have demonstrated that TBNA has a high level of sensibility, is safe, and is cost-effective compared with mediastinoscopy [4–6], it is still underused both by pulmonologists and thoracic surgeons. Surveys have demonstrated that only 11% to 30% of pulmonologists use TBNA [5]. Moreover, very few surgical groups are routinely using TBNA as a staging modality in patients with potentially resectable NSCLC. The main reason is the belief that this bronchoscopic procedure is not particularly useful because it does not have an effect on the management of lung cancer [5].
Skepticism concerning TBNA is clearly in contrast with the data from the literature, however. Numerous studies have found the sensitivity of TBNA to be similar to mediastinoscopy [2]. Other analyses have demonstrated that TBNA avoids further invasive testing, such as mediastinoscopy, in a large percentage of patients with lung cancer [6]. A recent article [7], moreover, has shown that combination of TBNA and FDG-PET has the potential to allow adequate mediastinal staging of NSCLC.
The objectives of this study were to analyze our experience with the use of TBNA as a minimally invasive procedure alternative to mediastinoscopy in the presurgical staging of NSCLC patients with positive mediastinal FDG-PET and to propose a staging algorithm that combines performance characteristics of the three methods of FDG-PET, TBNA and mediastinoscopy.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
According to the standard of clinical practice of our department (Fig 1), all patients with suspected or newly diagnosed NSCLC, functionally and technically suitable for surgical resection, and clinically staged N2/N3, M0 (stage IIIa and IIIb) after FDG-PET underwent minimally invasive mediastinal investigation with TBNA to obtain cytologic confirmation from the right paratracheal (2R, 4R), left paratracheal (2L, 4L), and subcarinal (7) stations.
|
If patients had positive TBNA proven N2 or N3 disease, neoadjuvant chemotherapy or chemotherapy and radiotherapy were scheduled, respectively. If patients had negative, inadequate, or inconclusive TBNA specimens, mediastinoscopy was performed, and the patients underwent pulmonary resection, neoadjuvant chemotherapy or chemotherapy or radiotherapy, or both if a N0, N2 or N3 disease were shown, respectively (Fig 1).
Bronchoscopic TBNA procedures were performed by 4 of the authors (G. M., A. C., P. C., M. C.) using a cytologic 22-gauge needle. The location of the needle insertion was determined by correlating the findings of the preprocedure CT and FDG-PET scans with landmarks in the airway. On removal of the needle, the aspirate was smeared onto microscope slides. The prepared and stained slides were immediately assessed microscopically by a cytopathologist to determine if the material present was adequate for a preliminary diagnosis. If the preliminary interpretation was negative or inadequate, additional passes were made with the needle into the same or different lymph node at the discretion of the physician. A final diagnosis was made after all processed slides were evaluated by the cytopathologist.
To avoid false-positive results, the aspiration was always done before possible endobronchial biopsies. In case of bilateral involvement of the mediastinum, the aspiration was preferentially performed at the level of the contralateral (N3) adenopathy if technically feasible. Mediastinoscopy was performed under general anesthesia by four experienced thoracic surgeons using a video mediastinoscope. Stations 2R, 4R, 2L, 4L, and 7 (highest part of the station) were sampled, if visible lymph nodes were present. The location of the lymph nodes sampled both by TBNA and mediastinoscopy and those dissected during the procedure was classified according to Mountain/Dresler regional nodal stations for lung cancer staging [8]. In all patients who underwent operation, pulmonary resection was associated with systematic lymph node dissection.
Statistical Analysis
Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of TBNA were calculated using the standard definitions. These variables were calculated after exclusion of inadequate and inconclusive samples. All positive results were considered true-positive for statistical analysis because false-positive results were considered unlikely and no surgical verification was performed. For negative TBNA samples, surgical exploration (mediastinoscopy or thoracotomy) was used as the reference standard.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
In 26 patients a malignant lymphadenopathy was demonstrated, with an overall diagnostic yield of 51%. NSCLC was diagnosed in 6 patients, adenocarcinoma in 11, squamous cell carcinoma in 8, and large cell carcinoma in 1. These 26 patients with positive TBNA were excluded from operation and referred for neoadjuvant chemotherapy or chemotherapy and radiotherapy according to the mediastinal nodal status (N2 or N3 disease) as determined on the FDG-PET scan. In the remaining 25 patients (49%), samples from TBNA were considered adequate negative in 12 patients (23%), inadequate in 11 (21%), and inconclusive in 2 (5%).
All 25 patients without a specific diagnosis underwent mediastinoscopy. Mediastinoscopy showed the presence of N2 disease in 19 (37.3%), and they received neoadjuvant chemotherapy. In the remaining 6 (11.8%), mediastinoscopy revealed no metastatic mediastinal lymph nodal involvement, and patients underwent lung resection with lymphadenectomy. Postoperatively, 5 patients were staged N0 and 1 N2 (presence of a single metastasis at the level of the posterior part of the subcarinal 7 station). For TBNA, the sensitivity of assessing mediastinal lymph node metastases was 76% (26 of 34); specificity was 100% (4 of 4); positive predictive value was 100% (26 of 26); negative predictive value was 33% (4 of 12); and diagnostic accuracy was 79% (30 of 38).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In our Department of Thoracic Surgery, most patients with NSCLC suitable for operation usually undergo noninvasive staging with FDG-PET scan. Until few years ago in our clinical practice, mediastinoscopy was used to confirm or exclude the presence of metastases in patients with positive PET mediastinal lymph nodes. Recently, however, we have changed this staging strategy, as reported in the algorithm (Fig 1).
Many reasons led to the choice of an initial minimally invasive staging strategy using FDG-PET scan in combination with TBNA in patients with potentially resectable NSCLC.
First, FDG-PET scan is always more frequently used in clinical practice owing to its accuracy [1, 10] and widespread diffusion. In fact, PET technology is now available in many departments of nuclear medicine, reducing the waiting list and making it possible to quickly obtain a PET appointment.
Second, TBNA and mediastinoscopy have the same sensibility: 78% [2].
Third, although EUS-FNA is considered an accurate method for minimally invasive staging of NSCLC [2], we use TBNA because we believe it has some advantages. TBNA is more likely to be able to make a difference in managing NSCLC patients suitable for operation than EUS-FNA. EUS-FNA has excellent access to stations 9, 8, 7, 6, and 5; however, it cannot sample the right and left paratracheal lymph nodes (stations 2 and 4). These nodes are those more frequently involved in NSCLC. Moreover, the presence of metastatic paratracheal lymph nodes has a big effect on patient management. In fact, the nearly all patients with nodal metastases at levels 2 and 4 are considered for neoadjuvant chemotherapy. On the contrary, many patients with metastases confined to levels 9, 8, 6, or 5 may be considered for primary operation, followed by adjuvant chemotherapy, without the need to preoperatively obtain tissue confirmation of neoplastic nodal involvement. In addition, EUS-FNA requires sophisticated, expensive equipment and specific training. As a result, the current application of EUS-FNA is limited to selected centers and most probably will not be available widespread. In contrast, TBNA is a method that has the potential for widespread use because no sophisticated or costly equipment is necessary [6].
Fourth, compared with mediastinoscopy, TBNA is less invasive and associated with less morbidity [2, 4–6].
We also chose to use this noninvasive staging system because in the last 5 years we have considerably improved our experience with TBNA, performing more than 100 procedures per year, and consequently increasing our diagnostic yield. TBNA, in fact, is the most operator-dependent of all bronchoscopic procedures, with a diagnostic yield strictly related to the operator's experience [5].
Finally, we chose to use TBNA in all NSCLC patients functionally and technically suitable for surgical resection with positive mediastinal FDG-PET also considering the limitations of mediastinoscopy. Some studies have demonstrated that approximately half of the false-negative results of this surgical procedure are related to the performance of incomplete mediastinoscopies [3].
A further consideration on the rationale of our presurgical staging algorithm of NSCLC patients is that an increasing number of studies have demonstrated that only stage IIIa NSCLC patients with mediastinal downstaging after neoadjuvant therapy will benefit from a subsequent surgical therapy [11]. As a consequence, it is likely that an increasing number of patients will require a repeat mediastinoscopy, which is a technically demanding procedure. It is evident that the use of our type of semi-invasive staging would avoid the risk performing a technically difficult repeat mediastinoscopy in a high percentage of patients.
The major limitation of TBNA in NSCLC staging is that it is impossible to systematically study the mediastinum. Contrary to mediastinoscopy, which allows sampling of multiple nodal stations with a secondary detailed assessment of the extent of mediastinal involvement, TBNA does not usually permit adequate samples to be obtained from all the accessible mediastinal lymph node. In most cases aspiration efforts during TBNA may be limited to the sampling of only the best target, which corresponds to the more accessible enlarged lymph node. This limitation is more evident when CT scan alone is used to guide TBNA. However, this limitation may be minimized when FDG-PET images are used in combination with CT for TBNA guidance.
To illustrate this concept, Figure 3 shows a typical case. In this patient, an operator who performed TBNA on the basis of the CT scan would probably sample only the more accessible enlarged lymph node, in this case the right paratracheal lymph node (2R station), thus understaging the disease. Whereas, the operator using FDG-PET as means of guidance for the procedure, only sampling the more accessible nodal station (2R station), will be able to correctly stage the disease as N3 according to the PET findings that show presence of an increased uptake of 18F-FDG at the level of the contralateral left paratracheal nodal station.
|
The overall diagnostic yield of TBNA (51%) in this study was relatively low compared with some published studies [2]. However, we should consider that our patient population was highly select. In fact, all patients were potential surgical candidates and almost all of them had a minimal mediastinal lymph node involvement. The target lymph nodes in most cases were only slightly enlarged (1 to 1.5 cm). Moreover, many patients had only one positive mediastinal nodal station at FDG-PET (clinically minimal disease). On the other hand, we know from the published data that the diagnostic yield of TBNA depends critically on the lymph node size [4, 5, 12], and this could partly explain why the diagnostic yield varies so widely, ranging from 15% to more than 85% [13]. However, even if in our experience the diagnostic yield was relatively low, it is notable that TBNA avoided a mediastinoscopy in approximately half of this very select patient population.
Numerous studies have shown that TBNA is less expensive than mediastinoscopy [4, 6]. An assessment of the cost effectiveness of mediastinal staging with TBNA is beyond the scope of this analysis. In our opinion, however, the main advantages of TBNA as an alternative to mediastinoscopy are not only its low cost but also its low invasiveness and the better utilization of clinical resources secondary to the use of this endoscopic procedure. Obviously, if two methods have the same accuracy, the choice will be the less invasive one. Moreover, it is obvious that performing TBNA in an endoscopic room results in a better utilization of the operating room, where an operation may be planned in a more rational way. Indeed, all surgeons are well aware of the difficulties in scheduling an operation in the operating room for suspected N2 disease without knowing in advance whether the procedure will finish after the mediastinoscopy or continue on with a thoracotomy.
Recently, endobronchial ultrasonography (EBUS) has been used for TBNA guidance (EBUS-TBNA). Preliminary studies have demonstrated that EBUS-TBNA may lead to improvements in the results of N-staging of NSCLC, increasing the percentage of adequate cytologic samples in patients with small lymph nodes or lymph nodes located in difficult anatomic positions [14]. Some studies [15–18] have demonstrated that EBUS-TBNA has a higher sensitivity (range, 69% to 98.7%) and negative predicted value (range, 88% to 97%) than conventional TBNA. However, the role of EBUS-TBNA in the mediastinal staging of NSCLC remains to be determined because the literature examining this procedure is relatively new and these excellent results have to be confirmed in further analyses.
To date, only three prospective, randomized studies comparing traditional TBNA and TBNA guided by EBUS have been conducted, but the reported results were conflicting. Shannon and colleagues [19] found no significant difference between EBUS guidance and traditional TBNA. In the study of Herth and colleagues [20], EBUS guidance increased the diagnostic yield only of subcarinal lymph nodes. In the study of Wallance and colleagues [15], EBUS-TBNA had a higher sensitivity than traditional TBNA (69% vs 36%, p = .003). Moreover, the effect of EBUS-TBNA on the need for further invasive staging procedures in most of the published analyses was similar to what we noted in our experience.
In this study a mediastinoscopy was eluded in 51% of the patients who underwent conventional TBNA. Vincent and colleagues [16] and Bauwens and colleagues [17] reported that surgical staging procedures were avoided, respectively, in the 43% and 56% of the patients who underwent EBUS-TBNA. According to the most recent guidelines [2, 9], negative results for both conventional TBNA and EBUS-TBNA always need to be surgically confirmed. From this point of view, therefore, the performance characteristics of conventional TBNA and EBUS-TBNA seem to be very similar. In addition, experience with EBUS-TBNA is still limited to only a few centers, and the cost of this innovative technology is high considering that last-generation echobronchoscopes cost more than $150,000.
Conventional TBNA, therefore, remains a simple, easily applicable, and inexpensive method for preoperative staging of NSCLC patients with positive mediastinal PET [6].
In conclusion, even if the number of patients evaluated in this study does not allow definitive conclusions, the algorithm that we propose seems to be convincing from a theoretic point of view and is easily applicable in clinical practice. TBNA avoided mediastinoscopy in approximately half of NSCLC patients suitable for operation with positive FDG-PET mediastinal lymph nodes, leading to a reduction of the costs and a better use of clinic resources. The main advantage of this type of staging is its low invasiveness. Considering that the FDG-PET is nowadays commonly used in the presurgical staging of NSCLC patients, the only requirement for using our algorithm is the appropriate skills to perform TBNA. More experience and further studies are required to confirm these preliminary results and to verify the possibility of using EBUS-TBNA as a substitute for traditional TBNA. Nevertheless, we believe that these preliminary data may help many thoracic surgeons not only to learn and practice TBNA but also to reevaluate their mediastinal staging approach.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  F. Detterbeck Invited Commentary Ann. Thorac. Surg., February 1, 2009; 87(2): 378 - 378. [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 |
