| 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 Division of Cardiothoracic Surgery, Department of Surgery (Dr. Detterbeck); Department of Radiology (Dr. Falen); Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine (Dr. Rivera); Department of Radiation Oncology (Dr. Halle); and Division of Medical Oncology, Department of Internal Medicine (Dr. Socinski), University of North Carolina at Chapel Hill, Chapel Hill, NC.
Members of the Multidisciplinary Thoracic Oncology Program, University of North Carolina, Chapel Hill, NC.
Correspondence to: Frank C. Detterbeck, MD, FCCP, Division of Cardiothoracic Surgery, Medical School Wing C, Room 354 CB# 7065, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7065; e-mail: fdetter{at}med.unc.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
Key Words: lung cancer • positron emission tomography • staging
|
Introduction |
|---|
|
TOPAbstractIntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
The major issue in patients with a suspected or proven lung cancer is definition of the extent of disease (staging). The initial clinical evaluation of the patient provides a presumptive stage, although this may need to be confirmed by further tests. Although PET imaging is approved in the United States for staging of patients with lung cancer, there is confusion about the appropriate place of PET imaging among the array of available staging tests. Furthermore, there is controversy about whether PET imaging can obviate the need for invasive tests such as mediastinoscopy. Finally, it must be remembered that, in many instances, an invasive test is needed in order to provide confirmation of the diagnosis, completely aside from any staging considerations. This article reviews the data regarding the role of PET imaging in the extrathoracic and intrathoracic staging of patients with lung cancer.
|
Background |
|---|
|
TOPAbstractIntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
The studies comparing staging of the mediastinum by CT and PET with staging by CT alone, however, are flawed in that staging by CT alone is well known to be inaccurate. Among patients with discrete enlarged lymph nodes, 40% do not have malignant involvement, whereas an average of 15% of patients with normal nodes by CT scan do have N2,3 involvement.10 Therefore, optimal conventional staging should include mediastinoscopy or some other means of mediastinal node biopsy, at least for the subgroups of patients in whom CT is known to be inaccurate. Furthermore, extrathoracic metastases are present in some patients. Appropriate staging should include a clinical evaluation by a physician who is experienced in the signs and symptoms of distant metastases as well as imaging for distant metastases in subgroups of patients despite the absence of symptoms. PET should not be a substitute for a careful evaluation of the patient by a knowledgeable clinician. Staging by CT alone represents the absence of such clinical input.
The results of a prospective randomized study11 (known as the PET in Lung Cancer Staging [PLUS] trial) have been published in which conventional workup was compared with conventional workup plus PET in 188 patients who were thought by the primary care physician to be "potentially resectable." The major end point in this study11 was the rate of futile thoracotomy, which was defined as a thoracotomy without resection, a thoracotomy in patients who were found to have N2 or N3 involvement, or any recurrence or any death occurring within a year of surgery. Patients randomized to staging that included PET imaging had a significantly lower rate of futile thoracotomy (21% vs 41%, p = 0.003). The difference in outcomes between the two study arms is due both to a higher rate of recurrence within 1 year (18% vs 6% of operated patients), and the frequent lack of identification of N2,3 involvement prior to thoracotomy in the conventional staging cohort. The overall incidence of N2,3 disease was similar in the two study arms (23% and 26%), but in the conventional staging arm this involvement was established prior to resection in the minority (45%), whereas it was found prior to resection in 75% in the PET staging arm.
A closer look at the PLUS study11 suggests that it really compares staging with PET to very minimal staging. Conventional workup is vaguely defined as being consistent with the local routine. The patients underwent relatively few staging investigations, either by the primary care physician or the surgeon. Only 3% had a brain CT or MRI, and only 27% a bone scan, despite the fact that the patients had relatively extensive tumors (24% had clinical stage IIIA,B disease). Furthermore, the clinical evaluation appears to have been rather limited. Weight loss was present in 15% of patients, and 9% had a poor performance status. These characteristics are generally viewed as signs of distant metastases, yet the patients were considered to be potentially resectable. This is likely to be the explanation for the high rate of recurrence within 1 year in this study,11 particularly in the conventional staging group. Thus, it is not clear that the results of the PLUS study11 can be generalized to patients who undergo a careful clinical evaluation in conjunction with additional studies to confirm the stage when the reliability of the initial stage is known to be poor.
Should all patients with a high likelihood of lung cancer undergo PET imaging? PET can certainly be useful in many instances. However, PET imaging should not be used as a substitute for a clinical evaluation of the patient by a physician. Furthermore, it must be remembered that a PET can also be a nuisance, eg, if it frequently draws attention to "noise" and creates a lot of commotion about things that are unimportant in the final analysis. This is especially true in settings where the incidence of actual worrisome events is very low. Therefore, it is best to consider specific patient groups as defined by the presumed cell type and stage in order to achieve a thoughtful integration of PET imaging into the workup of patients with a high likelihood of lung cancer. Before discussing how to incorporate PET imaging, though, it is worth reviewing the data behind the traditional approach to staging of patients with lung cancer.
Traditional Approach to Staging in Patients With Probable Lung Cancer
In many patients, a presumptive diagnosis of lung cancer can be made with a high degree of confidence on the basis of the patient’s age, risk factors, and the radiographic appearance of the tumor. In fact, it is often possible to predict whether the tumor is more likely to be small cell lung cancer (SCLC) or NSCLC (eg, rapid growth and marked mediastinal adenopathy without a peripheral lung lesion is typical of SCLC). Although tissue confirmation of malignancy and of the cell type must be obtained at some point, the diagnosis is generally secure enough that it is best to consider the presumptive stage and possible treatment options first. Often an approach can be identified that will establish both a cytologic diagnosis and provide confirmation of the presumptive stage as well (Table 1
). Furthermore, the treatment approach may influence how confirmation of the diagnosis is obtained (eg, a preoperative diagnosis is not needed if the treatment will be surgical resection). Therefore, the most efficient approach to confirming the diagnosis depends on the presumed cell type, the presumed stage, and the tentatively planned treatment.
|
Signs and symptoms of systemic metastases (eg, fatigue, weight loss, poor appetite, neurologic signs and symptoms, bone pain) must be verified because such a positive clinical evaluation carries a substantial false-positive (FP) rate.1213 Generally, this has been accomplished by a series of imaging tests such as a brain CT or MRI, a bone scan, and an abdominal CT. In some instances, however, a simple test such as a plain skeletal radiograph of a symptomatic area is definitive enough. When the symptoms and radiographic findings are typical of metastases at multiple sites, most physicians consider this adequate proof of systemic disease without further confirmation. If only a solitary metastatic site is present, tissue confirmation at this site is usually required because the suspected metastatic lesion is not cancer in approximately 10 to 15% of patients.121415
In patients with NSCLC and a negative clinical evaluation for the presence of distant metastases, there is controversy about the need to confirm this with additional tests. Most medical and radiation oncologists believe that such confirmation is required, whereas most surgeons do not. A more detailed look at the false-negative (FN) rate of the clinical evaluation suggests an explanation for the discrepancy in these viewpoints. In patients with stage I and II NSCLC, a negative clinical evaluation carries a FN rate of approximately 5%, as has been shown consistently in several studies121617 using different approaches to this issue. However, in patients with clinical stage III (cIII) NSCLC, the FN rate of a clinical evaluation is approximately 15 to 30%.121317 It must be stressed, however, that the clinical evaluation must be carefully done, and it has been demonstrated that there should be a low threshold to investigation of subtle symptoms.18 The additional investigation for possible asymptomatic distant disease has traditionally involved a head CT or MRI and a bone scan (and imaging of the liver and adrenals if this was not included with the chest CT).
If no systemic metastases are found in patients with NSCLC, the status of the mediastinum becomes a crucial issue. Most physicians do not obtain further confirmation when there is extensive infiltration of the mediastinum to the point where discrete lymph nodes can no longer be discerned. Although there are no data that clearly define this, clinical practice suggests that this approach is justified. However, when there is enlargement of discrete lymph nodes, confirmation of the true status of the nodes is necessary, because there are extensive data that in approximately 40% of patients there is no mediastinal involvement despite the radiographic appearance.101920 Conversely, although a chest CT may suggest no mediastinal node involvement (all nodes are < 1 cm in short axis), approximately 15% of patients will be found to have nodal involvement.101920 The FN rate of a negative chest CT finding for the mediastinum is particularly high (20 to 25%) in the case of a central tumor, an adenocarcinoma, or if there is evidence of N1 node enlargement.1020 Therefore, the clinical stage of the mediastinum as defined by CT alone is notoriously inaccurate in a variety of settings, and further confirmation of the clinical stage is desirable. Traditionally this has been accomplished by invasive procedures such as mediastinoscopy, transbronchial needle aspiration, or esophageal ultrasound and needle aspiration.
|
Role of PET for Staging in Patients With Lung Cancer |
|---|
|
TOPAbstractIntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
It is unclear how PET imaging should best be integrated into the workup of patients with SCLC. A PET scan is likely to show systemic metastatic sites at least as well as CT or MRI scans of the bones, chest, and abdomen, although a brain CT or MRI is still likely to be necessary. While PET imaging currently remains a relatively expensive test, it is no more costly than multiple other scans. This would argue that PET imaging may be a reasonable alternative to multiple scans in most patients with SCLC. However, there are few data as yet to help with the decision regarding the role of PET imaging in the workup of patients with SCLC.2122
Probable NSCLC With Symptoms of Systemic Metastases
Some patients present with signs and symptoms of obvious advanced disease. For example, this may include patients with subcutaneous metastatic deposits, patients with symptoms and plain skeletal radiographs demonstrating metastases, or patients with palpable supraclavicular node enlargement or a large pleural effusion. Although these latter two examples only signify cIIIB disease, these patients have a poor prognosis and are treated the same as patients with clinical stage IV (cIV) NSCLC. There is little reason to obtain a PET scan in patients with such obvious advanced disease. A needle aspiration of one of the sites of involvement will yield a tissue diagnosis and prove the advanced nature of the disease without the expense of PET imaging.
Other patients present with more subtle signs of possible systemic disease, particularly constitutional symptoms such as fatigue, anorexia, or weight loss. Further investigation to find possible metastatic disease is warranted, given that the clinical evaluation for systemic metastases has a high FP rate.12 A PET scan is a reasonable alternative to a battery of other imaging studies. However, it must be realized that there are also many indeterminate findings on PET imaging that may require a battery of additional studies or interventions. Nevertheless, the data demonstrate that PET is more accurate than conventional imaging. In a study5 of 100 patients with lung cancer, PET imaging for distant metastases had a sensitivity of 91%, a specificity of 96%, a FP rate of 5%, and a FN rate of 7%. In contrast, conventional imaging was found to have a sensitivity of 80%, a specificity of 80%, a FP rate of 24%, and a FN rate of 17%.5 Furthermore, PET is more reliable when compared specifically with a radionucleotide bone scan. The sensitivity of PET imaging for detecting bone metastases is 90 to 92%, the specificity is 98%, the FN rate is 1 to 2%, and the FP rate is 8 to 10%,523 as opposed to an average sensitivity of 76%, specificity of 69%, FP rate of 63%, and FN rate of 9% for bone scanning in patients with lung cancer.12 Thus, the ability of PET to detect distant metastases is high, and the interpretation of a positive or negative PET result in an individual patient can be made with relative confidence. However, conventional imaging of the brain is still required, because the high glucose metabolism of the brain makes FDG-PET a poor imaging test in this organ.
PET imaging is particularly useful in patients with an atypical presentation, a solitary site of metastasis, or with lesions that are indeterminate on other scans. For example, patients may present with an enlarged adrenal gland, an indeterminate liver lesion, or a second pulmonary nodule. The FN rate of PET imaging in adrenal lesions is close to zero,52425 and is lower than that of any other test (such as MRI, noncontrast CT, or needle aspiration).12 The FP rate for PET imaging of adrenal lesions ranges from 0 to 8%.52425 PET imaging of the liver and lung has also been reported to have very low FN and FP rates,5 making this an excellent study to either rule out or rule in malignant involvement in these sites (provided the lesion is 
1 cm in diameter). However, because tissue confirmation of cancer is generally necessary, it may be more expeditious to perform a needle biopsy of a suspected metastatic site instead of a PET scan if the suspicion of malignant involvement is high.
cIII NSCLC
A large proportion of patients with probable NSCLC present with radiographic evidence of mediastinal involvement but without signs and symptoms of systemic disease. In these patients, the FN rate of the clinical evaluation appears to be high enough (15 to 30%) to warrant a search for systemic metastases despite the negative clinical evaluation.12 A PET scan is an alternative to the traditional approach using a bone scan, but it will not obviate the need for a brain CT or MRI. Although a PET scan may cost slightly more than a bone scan, a bone scan often results in indeterminate findings that prompt other imaging studies. Furthermore, PET imaging also has the advantage of potentially detecting unsuspected metastases at other sites. Thus, a rational argument can be made to consider a PET scan for extrathoracic staging in patients with cIII NSCLC.
Many studies34523262728 have reported finding distant metastases with PET in 10 to 40% of patients in general with lung cancer (often with limited verification of the PET results). Only one study29 has characterized the patients carefully with regard to clinical evaluation and clinical stage. In this prospective study,29 PET detected asymptomatic distant (nonbrain) metastases in 24% of 100 patients with cIII NSCLC. This rate of finding distant metastases in patients with cIII NSCLC is similar to the FN rate of the clinical evaluation in this patient population. In > 90% of these cases, the metastases were confirmed by cytology or follow-up, and no FP PET findings were noted. Of note, only 19% of the metastases involved the bone, suggesting an advantage for PET imaging over bone scan to look for asymptomatic distant disease. Similar results have been reported in two other studies3031 with respect to the clinical stage as defined by CT (but without data regarding the clinical evaluation): extrathoracic metastases were detected in 25% and 28% of patients with cIIIA,B NSCLC, respectively.3031
It does not appear that PET imaging should replace mediastinoscopy for confirmation of the status of the mediastinum in centers where mediastinoscopy is available. Pooling the results from a large number of series suggests that the average FP rate of PET imaging in the mediastinum is 16%,10 and a recent meta-analysis32 found a FP rate of 22%. In another review33 of data solely from patients with enlarged mediastinal nodes, the FP rate of PET in the mediastinum was an average of 13%. A positive PET scan result therefore warrants confirmation by mediastinoscopy at this time, and this was the conclusion reached in a recent evidence-based guideline as well.34 Alternative approaches for confirmation of enlarged and suspicious nodes include transbronchial needle aspiration, TTNA, and esophageal ultrasound with needle aspiration.
Whether a negative PET scan finding in the mediastinum obviates the need for mediastinoscopy is a matter of controversy. Several reviews6103536 have found average FN rates for PET in the mediastinum of 5 to 7% in general (involving patients with cIII as well as clinical stage I,II [cI,II] tumors). In a review33 of data from patients with enlarged mediastinal nodes, the FN rate of PET in the mediastinum was an average of 8%. By comparison, the average FN rate for mediastinoscopy is 9% (range, 0 to 16%).10 Although these are quite similar, some experts argue that it is important to find those patients with microscopic disease in the mediastinal lymph nodes, and that PET imaging will not detect this as well as mediastinoscopy.3435 Whether these results are interpreted as acceptable to avoid invasive biopsy in patients with a negative PET result, or as an argument to pursue this, is a matter of individual opinion. However, it is likely that in patients with enlarged mediastinal nodes, many clinicians would be uncomfortable accepting a negative PET result without a biopsy. This is confirmed by a recent meta-analysis,37 which estimated a 30% probability of malignant involvement of enlarged mediastinal nodes in the face of a negative PET scan.
In summary, PET imaging appears to be useful as an alternative to conventional imaging to rule out unsuspected systemic disease in patients with cIII NSCLC. Of course, information about the level of PET activity in the mediastinum will be gained as well. However, positive PET uptake in the mediastinum still warrants confirmation by mediastinoscopy (or other means). Although there is controversy about whether to accept a negative mediastinal PET scan in general, most clinicians would pursue mediastinoscopy in patients with enlarged mediastinal nodes by CT despite a negative PET result. However, in communities in which invasive staging of mediastinal nodes is not available, a strong argument for PET staging of the mediastinum can be made.
cI,II NSCLC
A search for distant metastases is traditionally thought to be unwarranted in patients with cI,II NSCLC if the clinical evaluation is negative, as several avenues of investigation consistently have found that the incidence of detectable metastases is < 5%.12 It could be argued that these data are a reflection of the ability to detect metastases, and that PET imaging may have an increased detection rate. Two studies3038 have reported that PET detected extrathoracic metastases (subsequently confirmed) in 1% and 9% of patients who had cI,II tumors by radiographic assessment alone, without providing data about the clinical evaluation. PET detected distant metastases in 5.2% of 287 patients (77% clinical stage I [cI], 7% clinical stage II [cII], 16% cIIIA) with a negative clinical evaluation in another prospective trial (all metastases subsequently confirmed).39 Only one study29 has addressed the rate of detection of distant metastases specifically in patients with cI,II NSCLC by CT and by clinical evaluation. In this prospective study,29 PET imaging detected asymptomatic distant (nonbrain) metastases in 8% of 39 patients with cI tumors and 18% of 28 patients with cII tumors.29 This suggests that PET may be warranted for the detection of distant metastases in patients with cII tumors, whereas the detection rate in patients with cI tumors is probably not high enough to justify this test when all of the data are considered together. Of note, studies1740 evaluating the rate of detection of distant metastases by traditional tests after a negative clinical evaluation have reported results for both cI and cII combined, although they have involved mostly cI disease.
Patients with cI,II tumors have a radiographically negative mediastinum by definition. However, there is ample evidence that the FN rate of CT for mediastinal involvement is approximately 20 to 25% in patients with central tumors, with an adenocarcinoma or with N1 node enlargement.10 The rate of detection of mediastinal node involvement by PET imaging for such subsets of patients has never been reported. Thus, it is unclear whether PET can or should replace mediastinoscopy for mediastinal staging in patients with central tumors, adenocarcinoma, or clinical N1 disease.
If a PET scan has been obtained, the data suggest that a positive PET result in the mediastinum should be confirmed by a biopsy. In a review,33 the FP rate of PET in the mediastinum for patients without mediastinal node enlargement was 14%, and other reviews1031 have found FP rates of 16% and 22% for patients in general, without regard to whether the nodes were enlarged or not. However, it is controversial whether or not to confirm a negative PET result in the mediastinum. The FN rate of PET in the mediastinum is 8% in patients with cI,II tumors,33 and several reviews6103236 have found average FN rates for PET in the mediastinum of 5 to 7% in general. Some physicians would argue that this FN rate is low enough to be acceptable, while others would argue that one should not rely on a negative PET study in clinical settings in which the chance of nodal involvement by CT criteria is high (central tumors, adenocarcinoma, or N1 node enlargement) because FN rates for PET have not been reported for such subsets of patients without mediastinal node enlargement.3234 A recent meta-analysis37 estimated approximately a 5% probability of malignant involvement of normal-sized mediastinal nodes in the face of a negative PET scan, despite a pretest probability of malignant involvement of 25%.
In patients with peripheral cI tumors and a negative clinical evaluation, there appears to be little role for PET imaging for staging. The incidence of either systemic metastases or mediastinal node metastases in this population is quite low (< 10%).1012 Therefore, neither additional scans for distant disease nor mediastinoscopy are considered warranted in the traditional workup of these patients. The chance of detection of distant metastasis by PET is < 10% in patients with cI NSCLC,293038 making it difficult to justify PET imaging in these patients when one considers the problems of FP and equivocal results, and the resultant further testing and delays. With regard to the detection of possible mediastinal node involvement in patients with peripheral cI tumors, the fact that thoracotomy and node dissection discloses < 10% with positive mediastinal nodes argues strongly against the use of PET for further mediastinal staging in these patients as well. A 10% rate of positive FDG uptake in the mediastinum among 84 patients with cI tumors was reported in the only study38 that specifically addressed this population. However, 60% of these positive PET results in the mediastinum turned out to be FPs,38 underscoring the drawbacks of pursuing such imaging if the incidence of disease is low. A lack of benefit to PET imaging is confirmed by a randomized study41 consisting primarily (92%) of patients with cI NSCLC, in which there was no difference in the resection rate or incidence of early recurrences among 179 patients (mean follow-up, 10 months).
|
Conclusion: Who Needs a PET? |
|---|
|
TOPAbstractIntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
|
In patients with discrete mediastinal node enlargement by CT, mediastinoscopy (or other invasive biopsy) is likely to be indicated regardless of the PET findings in the mediastinum. A positive mediastinal PET result should be confirmed by a biopsy because the FP rate is approximately 15 to 20%. An exception to this policy can be justified in patients with a positive PET scan result and diffuse mediastinal infiltration by CT—in whom the CT appearance alone is generally accepted as reliable. A negative PET scan result, however, is also likely to prompt mediastinoscopy in patients with enlarged mediastinal nodes, especially when tissue confirmation of the diagnosis is still necessary.
In patients with no clinical evidence of systemic metastases and no evidence of mediastinal node involvement by CT, the role of PET imaging of the mediastinum is controversial. For certain subgroups of these patients (those with a central tumor, an adenocarcinoma, or N1 nodal involvement), the reliability of the negative CT result is notoriously poor (FN rate of 20 to 25%). The role of PET imaging of the mediastinum has not been defined specifically for these subsets of patients, and mediastinoscopy remains the most reliable approach. A positive PET result would warrant confirmation by mediastinoscopy. Whether a negative PET result obviates the need for mediastinoscopy is controversial, with some physicians arguing that it should and others that it should not, and no data to resolve this.
In patients with a peripheral cI lung cancer, there are reasonable data to justify not pursuing a PET scan for staging. The evidence indicates that the incidence of finding either unsuspected distant metastases or unsuspected mediastinal involvement on PET imaging is quite low.
|
Footnotes |
|---|
Received for publication April 14, 2003. Accepted for publication September 4, 2003.
|
References |
|---|
|
TOPAbstractIntroductionBackgroundRole of PET for...Conclusion: Who Needs a...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. E. Stinchcombe, M. A. Socinski, L. M. Gangarosa, and A. H. Khandani Lung Cancer Presenting With a Solitary Colon Metastasis Detected on Positron Emission Tomography Scan J. Clin. Oncol., October 20, 2006; 24(30): 4939 - 4940. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Bunyaviroch and R. E. Coleman PET Evaluation of Lung Cancer J. Nucl. Med., March 1, 2006; 47(3): 451 - 469. [Full Text] [PDF]
|
|
|
|
|
|
C. S. Lathan, B. A. Neville, and C. C. Earle The Effect of Race on Invasive Staging and Surgery in Non-Small-Cell Lung Cancer J. Clin. Oncol., January 20, 2006; 24(3): 413 - 418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. C. Detterbeck Relearning the Lessons of the Past: Are We Making Progress? Ann. Thorac. Surg., November 1, 2005; 80(5): 1979 - 1980. [Full Text] [PDF]
|
|
|
|
|
|
L. K. Shankar and D. C. Sullivan Functional Imaging in Lung Cancer J. Clin. Oncol., May 10, 2005; 23(14): 3203 - 3211. [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]
|
|
|
|
|
|
B. E. Hillner, R. Tunuguntla, and M. Fratkin Clinical Decisions Associated With Positron Emission Tomography in a Prospective Cohort of Patients With Suspected or Known Cancer at One United States Center J. Clin. Oncol., October 15, 2004; 22(20): 4147 - 4156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. C. Detterbeck, S. Falen, M. P. Rivera, J. S. Halle, and M. A. Socinski Seeking a Home for a PET, Part 1: Defining the Appropriate Place for Positron Emission Tomography Imaging in the Diagnosis of Pulmonary Nodules or Masses Chest, June 1, 2004; 125(6): 2294 - 2299. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
