| 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 Departments of Interdisciplinary Endoscopy (Dr. Fritscher-Ravens and Mr. Brandt), Nuclear Medicine (Dr. Bohuslavizki), Internal Medicine, Pulmonology (Drs. Bobrowski and Pforte), Radiology (Dr. Lund), and General Surgery (Dr. Knöfel), University Hospital Eppendorf, Hamburg, Germany.
Correspondence to: Annette Fritscher-Ravens, MD, Department of Gastroenterology, Royal London Hospital, Whitechapel, London E1 1BB, United Kingdom; e-mail: fri.rav{at}btopenworld.com
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsConclusionReferences |
|---|
Patients and methods: Following bronchoscopic evaluation, CT, PET, and EUS were performed to evaluate potential mediastinal involvement in 33 consecutive patients with bronchoscopic biopsy/cytology proven (n = 25) or radiologically suspected (n = 8) lung cancer prior to surgery. Surgical histology was used as "gold standard" to confirm the diagnosis of the primary tumor and the mediastinal status in all patients. Histology proved non-small cell lung cancer in 30 patients, neuroendocrine tumor in 1 patient, and benign disease in 2 patients.
Results: The mean age of the study group was 61.5 years (range, 41 to 80 years; 23 male patients). CT, PET, and EUS detected mediastinal lymph nodes (size, 0.4 to 1.6 cm) in 15, 14, and 27 patients (21 of which were suspected to be malignant on EUS), respectively. With respect to the correct prediction of mediastinal lymph node stage, the sensitivities of CT, PET, and EUS were 57%, 73%, and 94%. Specificities were 74%, 83%, and 71%. Accuracies were 67%, 79%, and 82%. Results of PET could be improved when combined with CT (sensitivity, 81%; specificity, 94%; accuracy, 88%). The specificity of EUS (71%) was improved to 100% by FNA cytology (EUS-guided FNA), which gave a tissue diagnosis including tumor type, without complications.
Conclusions: No single imaging method alone was conclusive in evaluating potential mediastinal involvement in apparently operable lung cancer and routine clinical conditions. A tissue diagnosis is extremely helpful. Because FNA can be performed at the same time as EUS, this combination emerged as the most useful technique in the evaluation of even very small mediastinal metastases of lung cancer. CT seems necessary additionally to evaluate the pretracheal region as well as the rest of the thorax, and PET may be valuable to detect distant metastases.
Key Words: CT • endoscopic ultrasound • operable lung cancer • positron emission tomography • transesophageal endosonography-guided fine needle aspiration
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsConclusionReferences |
|---|
|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsConclusionReferences |
|---|
Inclusion Criteria: Patients were only included in this prospective study if they were referred for surgery of the primary tumor with curative intent, including lymph node sampling of the mediastinum for intraoperative staging. All patients were assessed as fit for thoracic surgery. Prior to entering this study, an outpatient chest radiograph, CT scan, as well as bronchoscopy suggested that the primary tumor was resectable in terms of location and size. No distant metastases were identified on additional abdominal CT.
Patients were excluded from this study if they had proven N3 disease or extensive N2 disease, which would make curative surgery extremely unlikely. Clinical "extensive" or "bulky" N2 disease was defined by the presence of documented metastases (by CT) at multiple stations of the ipsilateral mediastinal lymph nodes.31 These patients were staged to demonstrate the presence of metastases using transbronchial needle biopsy (TBNA) if the lymph nodes seen were accessible to this technique or using EUS-FNA outside this study.
Methods
Informed consent was obtained from all patients. The physicians performing the different examinations were blinded to the results of the other examinations. The results of all imaging methods were compared with the results of surgery in all 33 patients (thoracotomy in 31 patients and thoracoscopy in 2 patients). The American Thoracic Society (ATS) mediastinal map for lymphadenopathy was used to describe the location of lymph nodes.32
CT: Chest and limited upper abdominal CT was performed with a helical scanner (Somatom Plus S or Somatom Plus 4; Siemens; Erlangen, Germany). Slice thickness was 5 mm with a pitch of 1.6. Images were reconstructed at 5-mm intervals. During CT, 100 mL of contrast medium (iopadimol; Solutrast 33; Byk Gulden; Konstanz, Germany) was administered IV at a rate of 2.0 mL/s with a power injector. Scan reading was performed by independent, experienced radiologists during routine examination time. The radiologists were blinded to the results of the clinical investigations, other imaging modalities, and histopathologic findings. Patients were staged by CT images according to the criteria of Webb et al.33
PET: PET scanning was performed following 6 h of fasting in order to minimize blood insulin levels and glucose utilization of the normal tissue. Whole-body images were acquired 60 min after IV injection of 370 megabecquerel (10 mCi) of FDG using an scanner (ECAT EXACT 47 [921]; Siemens/CTI; Knoxville, TN) with an axial field of view of 16.2 cm. For attenuation correction, a postinjection transmission scan was acquired using rotating 68Ge rod sources. Attenuation-corrected and uncorrected emission data were reconstructed by filtered back- projection using a Hanning filter with a cutoff frequency of 0.4 of the Nyquist frequency. Attenuation-corrected data were converted to standardized uptake values (SUVs) and documented with an upper threshold of an SUV of five. SUVs were computed by normalizing the measured tumor radioactivity concentration to the injected dose and the total body mass. Differences in plasma glucose levels were neglected. Scan reading was performed by two independent, experienced nuclear medicine physicians during routine examination time, blinded to the results of clinical investigations, other imaging modalities, and histopathologic findings. PET was considered "positive" when mediastinal metabolic activity in form of "focal enhancement" was revealed on a reconstructed image. The location of an abnormality was reported as being present in the upper, middle, and lower mediastinum and on the to right or left side. Patients were staged by PET images according to the Union Internationale Contre le Cancer (1987).34
EUS: Linear echoendoscopes, Pentax 34 UX (Pentax GmbH; Hamburg, Germany) or Olympus GIF UC 30P (Olympus Optical; Hamburg, Germany) were used to perform EUS on a routine list by one experienced endoscopist, who was blinded to the results of the other imaging modalities. The procedure was performed in the left lateral decubitus position with IV sedation, using 2.5 to 5 mg of midazolam and propofol in individual titrated doses. EUS morphology, presence and status of lymph nodes, and the status of vessels were assessed. Every mediastinal lymph node detected on EUS was regarded as "abnormal," since under physiologic conditions normal lymph nodes are rarely visible on EUS. Their echogenicity would be similar to the surrounding tissue.35 It is fairly common to see lymph nodes subcarinally, even in apparently healthy patients. They appear especially common in patients with lung disease and are mostly defined as "reactive" on biopsy. These elongated, oval, hyperechoic, or isoechoic nodes are not safe to be interpreted as benign, as micrometastases would not change the EUS appearance.36 If such an elongated hyper (iso)echoic subcarinal lymph node was the only one seen on EUS in this study, it was punctured to look for metastatic disease. Nodes were visible on ultrasound from a diameter of 3 to 5 mm onwards.
As EUS visualization of the mediastinum is limited by the presence of air-filled structures, it cannot image the anterior mediastinum (level 3,6) through the trachea. Although it once was thought that it would be difficult at EUS to access level 2r and 4r lymph nodes, recent series have used EUS-FNA to detect nodes and sample these levels6 37 38 (Fig 1 , top).
|
During the puncture process, the entire needle could be visualized approaching the target in the sector shaped sound field. Color Doppler ultrasound imaging was used to identify adjacent blood vessels, when undefined structures were interposed in the path of the needle. Guided into the target lesion, the needle was moved back and forth within the mass while applying suction with a 10-mL syringe. The aspirated material was smeared onto glass slides and air-dried before being sent to cytopathology. Two needle punctures were usually made to obtain adequate tissue specimens. After May Grünwald Giemsa staining had been performed, the smears were classified by two experienced and independent cytopathologists using a modified Papanicolaou-classification.40
Data Analysis
Following the international TNM system, mediastinal lymph node status was assessed as N0, N2, or N3 for each of the imaging modalities. When FNA was obtained, the result was classified accordingly. When no FNA was obtained (if no nodes were visible), the FNA result was defined to be N0. Every diagnostic modality, including FNA, was compared to the final result using the 2 x 2 table shown in Table 2
. Thus, correct-positive (rp) results were the number of patients who had identical N stages in malignant disease, correct-negative (rn) results were the number of patients with identical N stages in benign disease. Accordingly, the false-negative (fn) results were those with a lower N stage in imaging when compared to the final surgical result, and the false-positive (fp) results were those patients with a higher N stage reported for each imaging modality than in that found to be the final diagnosis. We have used the Friedman test41
to answer the question whether the observed variations between diagnostic modes should be attributed to random fluctuations and the Cochrane test42
for comparison of four 2 x 2 tables to describe the correct prediction of the lymph node status. The mediastinal lymph nodal status of the study group found at surgical histology was used as a "gold standard."
|
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsConclusionReferences |
|---|
CT demonstrated enlarged mediastinal lymph nodes in 15 of the 33 patients (45%), which were correctly staged (with respect to N stage) in 8 patients. The nodes detected were located in the pretracheal region (level 3) in two patients and paratracheally in five patients; three of these were located in level 4r and one each in levels 2l and 4l. Five nodes were seen in the aortopulmonary window (level 5), subcarinally (level 7) in six, and paraaortic in the lower mediastinum (level 8) in two (Table 3 ). In two patients, lymph nodes were localized in the pretracheal area, one of which was at a single site and the malignant mediastinal infiltration was consequently missed on EUS, which is unable to image the pretracheal area because of intervening air. Eighteen of 33 patients did not have enlarged nodes detected on CT. EUS found lymph nodes suspicious of malignancy in 8 of those 18 patients. EUS-FNA demonstrated metastatic involvement in four patients and benign "reactive" nodes in the other four patients. Surgical histology showed mediastinal nodal involvement (N2) in all of the four "positives" detected with EUS-FNA but not in other cases, indicating that CT had given fn results in 4 patients. In three patients, N2 nodes were seen on CT but final diagnosis was N0; in one patient with a contralateral node on CT, surgery revealed N2 stage. In addition to mediastinal mapping, CT demonstrated enlarged axillary lymph nodes in one patient. The sensitivity, specificity, accuracy, and positive and negative predictive values of CT in the diagnosis of mediastinal metastases were 57%, 74%, 67%, 62%, and 70% respectively (Table 4 ). The proportion of correctly predicted lymph node stages had an accuracy of 67%.
|
|
|
|
|
All 27 patients in whom lymph nodes were detected on EUS underwent FNA irrespective of their benign or malignant appearances. In 14 of a total of 16 patients with mediastinal metastases, EUS-FNA cytology demonstrated malignancy (88%). In the smears of two patients with subcarinal nodes, only inflammatory cells were seen. In these patients, surgical histology and TBNA (n = 1) proved N2 disease in pretracheal nodes (level 3, not visible by EUS) at a single site in one patient and an additional micrometastasis in a subcarinal node in the other patient. In seven other aspirates with benign cytology, the nodes were shown to be benign by surgery as well. Thus, sensitivity, specificity, and accuracy of EUS-FNA were 88%, 100%, and 91% when calculated with respect to the correct prediction of lymph node status including the anterior mediastinal area, which EUS cannot show. The clinical status of the patients was followed up for 4 days postprocedure, when the cytology results were discussed with the physician in charge. No complications were reported.
Results of CT in combination with PET were compared to EUS-FNA (Table 6 ). The accuracy, sensitivity, and specificity of CT plus PET, when one of the two imaging technique were predicting the N stage correctly, were similar for EUS-FNA. However, as lymph node status was discordant with both the techniques, a tissue diagnosis would be advantageous.
|
|
Discussion
The current ATS guidelines for the staging of NSCLC suggest that CT should be considered the standard investigation for the evaluation of mediastinal involvement of lung cancer.43
Based on anatomic imaging, lymph nodes with a short axis of > 1 cm are considered to have malignant involvement, although it is known that some benign nodes may be larger, and that normal-sized nodes may contain malignancy.44
45
Thoracic CT has varying but limited sensitivity (43 to 81%), accuracy (59 to 85%) and, in some series, poor specificity (44 to 94%) in reported series (Table 1)
. It was hoped that alternative imaging techniques such as PET might give better results. Recent studies on PET reported a high sensitivity of 67 to 100%, with a specificity of 70 to 100% and an accuracy of 80 to 100% for the imaging of lymph nodes (Table 1)
. Unfortunately, nonspecific inflammatory reactions in lymph nodes may also accumulate FDG, and the benign diseases that could be associated with an elevated glucose uptake are not completely defined yet. The fp results with PET have been reported with tuberculosis, sarcoidosis, histoplasmosis, and other fungal infections.21
23
46
Several studies, comparing PET and CT (Table 1)
, demonstrate PET to be significantly superior to thoracic CT because it is able to demonstrate an increased tumor metabolism even in normal-sized nodes.13
Although it may usefully define distant metastases, a major drawback of PET is that the FDG uptake can only be localized to the hilar or mediastinal areas differentiating left-sided from right-sided activity.25
In routine usage at present, it has inherent limitations in defining the precise individual location of the involved groups of lymph nodes as defined in the ATS classification, which may have therapeutic implication in some of the patients. A more detailed anatomic classification of mediastinal lymph nodes may become possible by combining emission and transmission-corrected PET scans.21
EUS is an accurate, relatively noninvasive, highly sensitive imaging technique in the detection of lesions < 1 cm.47 Gress et al5 compared results of thoracic CT and EUS in predicting the involvement of mediastinal lymph nodes and reported a sensitivity of 49% and 84% for CT and EUS, respectively. In our study, suspected mediastinal nodes were seen on CT, PET, and on EUS with a sensitivity of 57%, 73%, and 94%, respectively. The fact, that only potentially operable patients with no or only rare ipsilateral lymph nodes seen on initial CT were included may explain the somewhat decreased sensitivity of CT and PET compared with some published series. Small lymph nodes are more difficult to detect on CT and may not take up glucose on PET, providing fn results. The higher resolution of EUS may be the reason that this technique was statistically significantly superior to CT or PET (Table 4) .
EUS with its additional option of guided FNA seems to have an added advantage in such a situation. EUS-FNA is less invasive with a negligible rate of complications, especially in the mediastinum.4 5 6 7 31 37 It can be carried out during the same session as an outpatient procedure without substantially increasing the costs. Moreover, it enables characterization of the tumor type into subtypes such as small cell, adenocarcinoma, or squamous cell cancer, which has an influence on the subsequent choice of therapy. The fp results are extremely rare in this type of cytology, so that there is no need to confirm a malignancy with another, more invasive method such as mediastinoscopy or thoracoscopy. In our study, EUS-FNA was performed in 27 patients with a sensitivity, specificity, and accuracy of 88%, 100%, and 91%, respectively. The sensitivity achieved in our patients was a little lower than previously reported data of EUS-FNA of mediastinal lymph nodes in lung cancer (90% vs 100%, respectively).4 5 6 7 31 37 The reason may be the small size of the lymph nodes punctured in this study (0.5 to 1.6 cm), as well as a relatively low frequency of punctures/node (n = 2). The specificity, however, was 100%; and to date, no fp result has been reported for mediastinal EUS-FNA.
In the present study of 33 patients with suspected lung cancer, PET was accurate in 79% and EUS was accurate in 82%. EUS was highly sensitive (94%) compared to CT and PET (57% and 73%, respectively), whereas specificity of PET (83%) was superior to that of EUS prior to FNA (71%). The higher sensitivity of EUS is probably due to its high imaging resolution, detecting any small lymph node with altered EUS morphology. The lower sensitivity of PET could be due its limitations in distinguishing nodes in close proximity to the primary tumor and its inability to identify very small nodes.14 18 20 In this series, however, PET detected two of the four N2 stages not seen on CT.
A combination of imaging methods might improve on the results of individual techniques by compensating for the individual drawbacks and supplementing the positive features. Thus, combination of CT plus PET was proposed as an alternative approach to avoid more invasive preoperative procedures to confirm the diagnosis. This combination, presumed to be positive when at least one modality was positive, was reported to be highly sensitive (93 to 96%), specific (up to 97%), and accurate (up to 96%).20 21 48 The specificity of CT plus PET was 94% in the present study and 100% using EUS-FNA.
Our results of the combination of CT and PET showed superior discriminatory characteristics to each of the imaging techniques alone. The reasons for studying the combination of CT plus PET are to avoid the risks of complications associated with the invasive confirmatory procedures such as mediastinoscopy (Fig 1 , middle) or thoracoscopy, TBNA through fiberoptic bronchoscope (Fig 1) , transthoracic CT, or sonographically guided FNA, and to reduce the costs of these additional procedures. Although anatomic and metabolic imaging modalities are highly effective in demonstrating mediastinal lymph nodes, our results as well as the experience of others reveal that none of these techniques are reliable in predicting the nature or the stage of the lesions when performed alone. Hence, tissue diagnosis is necessary to confirm metastases in the lymph nodes.30
Several methods for achieving tissue diagnosis of mediastinal lymph nodes are available. TBNA during bronchoscopy can offer a unique opportunity to identify mediastinal lymph nodes detected on preprocedure CT scan avoiding a separate staging procedure. Because the target nodes are not visible through the bronchoscope, the needle has to be advanced "blind" through the airway wall using landmarks identified on CT. TBNA is limited to needle aspiration of lymph nodes in areas adjacent to the trachea or bronchi (levels 2, 4, 7, 10; Fig 1 , bottom). It is less successful if the lymph nodes are small.48 While the overall sensitivity runs at approximately 70% in the best series, and may improve to 95% in accessible lymph nodes, which are > 1.5 cm, the sensitivity in nodes smaller 1.0 to 1.5 cm is substantially reduced.49 50 51 In a recent prospectively performed study of 66 patients, Wiersema et al52 found that EUS-FNA was superior to TBNA in the diagnosis of malignant mediastinal lymph nodes in NSCLC (92% vs 73%). TBNA guided with CT fluoroscopy might improve TBNA results, but the success is still limited by the size of the lymph nodes of 1 cm, and real-time imaging is not available during puncture.48 TBNA might be improved by using endobronchial ultrasound (EBUS), since this technique can image even small paratracheal and bronchial lymph nodes of 3 mm. However, as the ultrasound image is radial, real-time EBUS-guided biopsy is not yet possible. This new technique can be very helpful in orientating for a TBNA to follow.53 54 55
In relation to these techniques, one major drawback of EUS and hence that of EUS-FNA is its inability to image the anterior mediastinum (level 6; Fig 1 , top) and the rest of the thorax beyond the mediastinum. CT is consequently required for the evaluation of pretracheal nodes and the rest of the thorax. Generally, multiple groups of lymph nodes are involved when there are metastases in the mediastinum and whenever pretracheal nodes were seen on CT; other groups of nodes may also be present which may be detectable on EUS. However, this may not always be the case, as seen in the present study, where one patient had a single pretracheal metastasis. A combination of CT and EUS with guided FNA appears a logical combination that might significantly reduce the costs as well as the morbidity related to other invasive preoperative diagnostic procedures for tissue diagnosis. In rare cases when isolated pretracheal lymph nodes are present, mediastinoscopy may be required.
To our knowledge, this report is the first study comparing all three modalities in the diagnosis of mediastinal metastases of potentially operable lung cancer patients. We evaluated these three modalities on routine lists using standard equipment. This has the inherent advantage that some of these results might be applied to general hospitals, rather than those few with access to the latest "high-tech" or experimental equipment. Although expertise in EUS-FNA may not yet be available in all general hospitals, the application of EUS-FNA technology in the chest is relatively simple and does not require great skill. The endoscope is in a straight position, and tip displacement is unusual. Puncture is easier using EUS in the chest than EUS-FNA at other sites. Most learners of EUS-FNA start their biopsy experience in the mediastinum. Although uptake has been slow in some countries, EUS is becoming increasingly available. In Germany for example, > 300 linear EUS units have been sold and there are a quite a few nonacademic gastroenterologists who are very experienced with EUS. Probably more than one half of the larger pulmonary hospitals in Germany have purchased EUS equipment and/or have access to expertise in EUS-FNA, and we have trained several pulmonologists in our unit.
Similar to EUS in many countries, other potentially valuable imaging or tissue sampling techniques, such as EBUS and/or TBNA, remain relatively underutilized.49 56 57 It has been shown that only 12% of North American bronchoscopists routinely use TBNA and 29% use it occasionally. Reasons for this include operator inexperience, low yields, and concern regarding great vessel puncture, etc.56 57 58
Considering all available less-invasive methods available for staging lung cancer, one might create an algorithm along the following lines (cost-efficacy studies excluded). Initially, all patients with lung masses on chest radiography should be sent for thoracic CT scan prior to bronchoscopy. If on CT large lymph nodes are seen in levels 2r, 4r, 6, or 7, TBNA should be considered during diagnostic bronchoscopy (Fig 1 , bottom). If lymph nodes are seen in level 5 or in level 7 (and sometimes in level 2 and 4), or are small, EUS-FNA is warranted (Fig 1 , top). Optionally, PET may be performed after CT. If the result is negative, patients should be considered for surgery; if PET is positive for lymph nodes, TBNA or EUS could be performed depending to the location of the nodes seen. The results of this study make it reasonable to suggest that patients who are selected for potentially curative surgery may benefit from EUS and EUS-FNA. Irrespective of prior CT or PET scan findings, N2 or even N3 involvement of small nodes may be found at EUS-FNA unexpectedly, which might have a substantial impact on the further management.
|
Conclusion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsConclusionReferences |
|---|
|
Footnotes |
|---|
Received for publication February 5, 2002. Accepted for publication July 8, 2002.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsConclusionReferences |
|---|
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]
|
|
|
|
 |
|
 C. Schimmer, K. Neukam, and O. Elert Staging of non-small cell lung cancer: clinical value of positron emission tomography and mediastinoscopy Interactive CardioVascular and Thoracic Surgery, August 1, 2006; 5(4): 418 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Eloubeidi, A. Tamhane, V. K. Chen, and R. J. Cerfolio Endoscopic Ultrasound-Guided Fine-Needle Aspiration in Patients With Non-Small Cell Lung Cancer and Prior Negative Mediastinoscopy Ann. Thorac. Surg., October 1, 2005; 80(4): 1231 - 1239. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Vilmann and S. S. Larsen Endoscopic ultrasound-guided biopsy in the chest: little to lose, much to gain Eur. Respir. J., March 1, 2005; 25(3): 400 - 401. [Full Text] [PDF]
|
|
|
|
|
|
R. C. Rintoul, K. M. Skwarski, J. T. Murchison, W. A. Wallace, W. S. Walker, and I. D. Penman Endobronchial and endoscopic ultrasound-guided real-time fine-needle aspiration for mediastinal staging Eur. Respir. J., March 1, 2005; 25(3): 416 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. LeBlanc, B. M. Devereaux, T. F. Imperiale, K. Kesler, J. M. DeWitt, O. Cummings, D. Ciaccia, S. Sherman, P. Mathur, D. Conces, et al. Endoscopic Ultrasound in Non-Small Cell Lung Cancer and Negative Mediastinum on Computed Tomography Am. J. Respir. Crit. Care Med., January 15, 2005; 171(2): 177 - 182. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Eloubeidi, R. J. Cerfolio, V. K. Chen, R. Desmond, S. Syed, and B. Ojha Endoscopic Ultrasound-Guided Fine Needle Aspiration of Mediastinal Lymph Node in Patients With Suspected Lung Cancer After Positron Emission Tomography and Computed Tomography Scans Ann. Thorac. Surg., January 1, 2005; 79(1): 263 - 268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. C. Detterbeck, J. F. Vansteenkiste, D. E. Morris, C. A. Dooms, A. H. Khandani, and M. A. Socinski Seeking a Home for a PET, Part 3: Emerging Applications of Positron Emission Tomography Imaging in the Management of Patients With Lung Cancer Chest, November 1, 2004; 126(5): 1656 - 1666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S M Wildi, M A Judson, M Fraig, W E Fickling, N Schmulewitz, S Varadarajulu, S S Roberts, P Prasad, R H Hawes, M B Wallace, et al. Is endosonography guided fine needle aspiration (EUS-FNA) for sarcoidosis as good as we think? Thorax, September 1, 2004; 59(9): 794 - 799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C M Richardson and M D Peake Endoscopic (oesophageal) ultrasound guided fine needle aspiration (EUS-FNA) Thorax, July 1, 2004; 59(7): 546 - 547. [Full Text]
|
|
