| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology and Center for Imaging Science, Samsung Medical
Center, Sungkyunkwan University School of Medicine, 50, Ilwon-Dong,
Kangnam-Ku, Seoul 135-710, Korea.
2 Division of Pulmonary and Critical Care Medicine, Department of Medicine,
Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul,
Korea.
3 Biostatistics Unit, Samsung Medical Center, Seoul, Korea.
Received December 6, 2004;
accepted after revision February 16, 2005.
Supported by grant R11-2002-103 from the Korea Science and Engineering
Foundation.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Eighty-four patients (50 men and 34 women; age range, 39-80 years; mean age, 61 years) with stage T1 lung cancer underwent a hemodynamic and a conventional morphologic CT study before curative surgical resection. Peak enhancement (maximum attenuation over the entire time course), net enhancement (peak enhancement minus preenhancement attenuation), maximum enhancement ratio (MER), time to peak enhancement, slope of enhancement on dynamic studies, nodule size, presence of tumor necrosis or thickening of bronchovascular bundles, and marginal characteristics on morphologic studies were analyzed and correlated with the presence of histologically determined mediastinal or hilar nodal metastasis.
RESULTS. Mediastinal or hilar nodal metastases were found at surgery in 26 (31%) of 84 patients: mediastinal nodes in 13 (15%) and hilar nodes in 19 (23%). Six (7%) had both mediastinal and hilar nodal metastasis. Peak enhancement, net enhancement, and MER were significantly associated (p = 0.001, 0.002, and 0.008, respectively) with the presence of mediastinal or hilar nodal metastasis. A peak attenuation of 110 H or greater and a net enhancement of 60 H or greater predicted nodal metastasis with accuracies of 73% (61/84 nodules) and 73% (61/84 nodules) and odd ratios of 4.98 and 5.94, respectively.
CONCLUSION. Stage T1 lung cancers showing peak enhancement of 110 H or greater or net enhancement of 60 H or greater on dynamic CT indicate a high likelihood of hilar or mediastinal nodal metastasis.
Keywords: chest imaging • CT • hemodynamic studies • lung • lung cancer • metastasis • oncology
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
3 cm in diameter) without nodal or
distant metastasis, designated as stage IA (T1 N0 M0) using the TNM staging
system, have a greater than 50% rate of 5-year postoperative survival
[1,
2]. Although initial studies
suggest a low prevalence of nodal metastases in stage T1 lung cancers, several
recent studies have reported relatively high frequencies of mediastinal lymph
node metastases [3,
4]. Seely et al.
[5], in a review of the CT
scans and surgical findings of 104 patients with T1 lesions by complete nodal
sampling using either mediastinoscopy or thoracotomy, found that 21% of
patients had nodal metastases. Although several studies have been published on the staging of T1 lung cancer, little information is available concerning CT morphologic or dynamic study factors predicting nodal or extrathoracic metastasis in T1 lung cancer. Moreover, contradictory results have been reported for T1 lung cancers in terms of the correlation between tumor size and metastasis [6-10]. Stage T1 lung cancers with a solid nature, a coarsely spiculated margin, and bronchovascular bundle thickening around the lesion have been reported to show more frequent local vessel invasion, lymph node metastasis, and extrathoracic metastasis [11]. However, it is not known with precision how these findings allow us to predict the presence of hilar or mediastinal lymph node metastasis.
Histologic evaluations of tumor microvessel densities and of expressions of vascular endothelial growth factor (VEGF) are important prognostic factors in non-small cell lung cancers [12, 13]. In other words, the likelihood of metastatic disease increases as the number of intratumoral microvessels increases in lung cancers. Furthermore, because mean peak attenuation values on dynamic CT reflect microvessel densities in lung cancer and are significantly higher in VEGF-positive lung cancers [14-16], the likelihood of metastasis is probably increased in lung cancers that show strong and early enhancement on dynamic CT performed with IV contrast agent.
The purpose of our study was to identify those CT findings of the various hemodynamic and morphologic characteristics of stage T1 lung cancers that enable the prediction of hilar or mediastinal nodal metastasis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Hemodynamic CT Studies
We performed hemodynamic chest CT using a 4-MDCT (LightSpeed QX/I, GE
Healthcare) or a 16-MDCT (LightSpeed Ultra or Ultra16 scanner, GE Healthcare).
Before the IV contrast injection, a series of images (13 images) was obtained
throughout the nodule covering 30 mm along the z-axis with 2.5-mm
collimation at 120 kVp, 90 mA, 0.8-sec gantry rotation time, and a table speed
of 3.75 mm/sec. Subsequently, two sets of CT parameters were used for dynamic
enhancement studies.
From March 2002 to September 2003 (31 patients), an additional nine series of images were obtained after the initial unenhanced scans at 20-sec intervals for 3 min after contrast medium injection (3 mL/sec, total of 120 mL of Iomeron 300 [iomeprol], Bracco) with a power injector (MCT Plus, Medrad) using the parameters of 120 kVp, 170 mA, 0.8-sec gantry rotation time, and a table speed of 3.75 mm/sec, over 8 sec (total 10 series of images obtained at 0, 20, 40, 60, 80, 100, 120, 140, 160, and 180 sec after contrast administration). After October 2003, different enhancement dynamic study schemes were used (53 patients). Because we needed to evaluate the delayed phase of the dynamic study for nodule characterization and we had some concern about radiation hazard, we switched the dynamic study protocol. Images were obtained at 30, 60, 90, and 120 sec, and at 4, 5, 9, 12, and 15 min after contrast medium injection (3 mL/sec, total of 120 mL of Iomeron 300 [iomeprol]) using 120 kVp, 90 mA, 0.8-sec gantry rotation time, and a table speed of 3.75 mm/sec, over 8 sec (total 10 series of images obtained at 0, 30, 60, 90, and 120 sec and 4, 5, 9, 12, and 15 min after contrast administration). All image data were reconstructed with thicknesses of 2.5 mm (13 images in each image series; total number of dynamic images = 13 x 10 = 130 images) using a standard algorithm. In all patients, regardless of the dynamic study parameters used, helical CT (125 mA, 120 kVp, 5-mm collimation, table speed of 15 mm/sec) scans were obtained during or after the dynamic studies from the lung apices to the level of the middle pole of both kidneys for tumor staging.
All dynamic and staging CT data were directly interfaced to our PACS (Centricity 1.0, GE Healthcare Integrated Imaging Solutions), which allowed all images to be displayed on monitors (four monitors, 1,536 x 2,048 image matrices, 8-bit viewable gray-scale, and 60-foot-lambert luminescence). Both mediastinal (window width, 400 H; window level, 20 H) and lung (window width, 1,500 H; window level, -700 H) window settings were used for viewing images.
Image Analysis
Dynamic study of nodules—After viewing all 130 dynamic CT
images as thumbnail images on PACS monitors, two radiologists (with 2 and 14
years of chest CT experience, respectively) together selected one image for
analysis from 13 images at a given time. Then the two radiologists
independently measured attenuation values. The selected image was the
transverse section with the largest diameter (scanned at the equator of the
nodule). We measured the attenuation value of the nodule at the same area on
the selected image for each cluster at each time. We examined a region of
interest (ROI) that covered about one half of the diameter of a nodule at the
equator. When we confronted calcified (n = 1; area at the selected
image, 3%), cavitary (n = 7; average area, 5.1%; range, 4-11%;
estimation of air-containing area using polygonal ROI method on our PACS), or
necrotic (n = 15; average area, 7.2%; range, 5-18%; estimation of
nonenhancing, relatively low-attenuation area using polygonal ROI method)
areas, we made the ROI as large as possible (up to one half the diameter) in
the areas away from these calcified, cavitary, or necrotic areas. The edges of
the nodule were avoided to prevent partial volume averaging. All measurements
in Hounsfield units were obtained from images at mediastinal window settings
to ensure that partial volume averaging was minimized. All measurements were
obtained at the time of the CT examination without knowledge of the histologic
diagnosis.
We analyzed the extent of tumor enhancement using peak enhancement and net enhancement. Peak enhancement attenuation was defined as a maximum attenuation value of the nodule over the entire time course of the dynamic study. Net enhancement attenuation was calculated by subtracting preenhancement attenuation from peak enhancement attenuation.
Enhancement dynamics were assessed using two indexes: the maximum enhancement ratio (MER) and the slope of enhancement (SLE). Time to peak enhancement (TTP, expressed in seconds) was also recorded. The MER was calculated using the equation MER = [(peak enhancement attenuation - preenhancement attenuation) / preenhancement attenuation]. The SLE (expressed in seconds-1) was calculated using the equation SLE = MER / time to peak attenuation [16-18].
Evaluation of marginal and internal characteristics of nodules—The same two radiologists who measured the attenuation values of nodules also assessed retrospectively the morphologic features on CT. They reached a decision on findings by consensus in terms of nodule size, presence of tumor necrosis or thickening of bronchovascular bundles, and marginal characteristics. Necrosis was regarded as present when a focal low-attenuation area was observed in a nodule, compared with the surrounding enhancing area on a dynamic study, and this was classified as less than 10%, 10-50%, or greater than 50% of nodule volume. Bronchovascular bundle thickening was considered to be present when normally tapering bronchovascular bundles directed toward a nodule were distinctly widened or irregularly thickened. Nodule margins were classified as smooth, lobulated, spiculated, or lobulated and spiculated.
Nodal stage at staging CT—Nodal station was evaluated according to the lymph node map definition for lung cancer staging proposed by Mountain and Dresler [1]. In the present study, nodes lying distal to the hilar region were classified as hilar nodes (N1 nodal state). Lymph node assessment was based on size: mediastinal nodes with a short-axis diameter of 10 mm or more were defined as abnormal. Hilar lymph nodes were considered to be positive for malignancy when their greatest diameter exceeded 10 mm [19]. Mediastinal and hilar nodes containing nodular or laminated calcification were regarded as benign irrespective of their size.
Surgical-Pathologic Correlation
Tumor resection and extensive mediastinal lymph node dissection were
performed by one of two experienced thoracic surgeons (with 17 and 12 years of
experience, respectively). In addition to lung resection surgery, surgeons
dissected all visible and palpable lymph nodes in the surgical field,
irrespective of size. Namely, all encountered lymph nodes were removed from
American Thoracic Society lymph node map areas
[1] of 10R (right hilar), 9
(pulmonary ligament), 8 (paraesophageal), 7 (subcarinal), 4R (right lower
paratracheal), 3 (prevascular and retrotracheal), and 2R (right paratracheal)
in tumors of the right lung, and from areas 10L (left hilar), 9, 8, 7, 6
(paraaortic), 5 (subaortic), and 4L (left lower paratracheal) of the left
lung.
A lung pathologist with 10 years of experience described tumors (i.e., histopathologic class, size, involvement of surrounding organ, necrosis, distance from the resection margin) and lymph nodes (location and number). Surgeons labeled dissected lymph nodes by numbering (describing nodal station) on the basis of the lymph node map definition for lung cancer staging [1]. The pathologist then evaluated the nodes as numbered in the surgical fields.
Statistical Analysis
Agreement in measured attenuation values of nodules between two observers
was analyzed by calculating the intraclass correlation coefficient. Because
two dynamic study protocols were used, we tested for the presence of
significant differences in the peak and net enhancements of the two protocols
using the Mann-Whitney test. The Mann-Whitney and Fisher's exact tests were
used to search for relationships between the presence of metastatic
lymphadenopathy and marginal characteristics, nodule size, presence of
bronchovascular bundles, necrosis, peak enhancement, net enhancement, SLE,
TTP, or MER as determined by CT studies. A p value of 0.05 or less
was considered to indicate statistical significance.
Logistic regression was used for multivariate analysis to investigate the independent predictor for nodal metastasis. To determine the cutoff values of peak and net enhancements corresponding to maximal accuracy for predicting hilar or mediastinal nodal metastasis, we used receiver operating characteristic (ROC) analysis. Diagnostic characteristics—that is, sensitivity, specificity, accuracy, positive predictive value, and negative predictive value—were calculated retrospectively at nearby cutoff values determined at ROC analysis, changing the values by 5 H in order to seek the best attenuation value to distinguish between metastatic and nonmetastatic lung cancer. We also calculated the sensitivity, specificity, accuracy, and positive and negative predictive values of nondynamic CT for staging nodal metastasis. Thereafter, we analyzed how and in which patients dynamic CT improved the prediction of mediastinal or hilar nodal metastasis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sixty-four patients (76%) had adenocarcinoma; 13 (15%), squamous cell carcinoma; three (4%), large cell carcinoma; two (2%), bronchioloalveolar carcinoma; and one each (1%), pleomorphic carcinoma and carcinoid.
Mediastinal or hilar nodal metastases were present in 26 (31%) of the 84 patients. Of these 26 patients, 21 had adenocarcinoma; four, squamous cell carcinoma; and one, nonmucinous bronchoalveolar carcinoma. Mediastinal nodal metastases were present in 13 (15%) of the 84 patients: seven patients (8%) had isolated mediastinal nodal metastasis and six (7%) had both mediastinal and hilar nodal metastasis. Hilar nodal metastases were present in 19 (23%) of 84 patients: 13 (15%) had isolated hilar nodal metastasis and six had (7%) both hilar and mediastinal nodal metastasis. All mediastinal or hilar nodal metastases were ipsilateral (N1 or N2). On a per-nodal-station basis, a total of 428 mediastinal and hilar lymph nodes were sampled at surgery (mean number of nodal stations sampled per patient, 5.1) and of those, 35 metastatic lymph nodes were confirmed to be positive for malignancy at pathology.
The CT findings, including the hemodynamic and morphologic features of stage T1 lung cancer, are summarized in Tables 1, 2, 3. No statistically significant differences were found between patients with and without nodal metastasis in terms of marginal characteristics, the presence of bronchovascular thickening, size, necrosis, TTP, or SLE of a nodule (p = 0.104, 0.660, 0.178, 0.491, 0.348, and 0.179, respectively). Findings of peak enhancement, net enhancement, and MER (p = 0.001, 0.002, and 0.008, respectively) were significantly correlated with the presence of mediastinal or hilar nodal metastasis (Figs. 1A, 1B, 2A, 2B, and 2C).
|
|
|
|
|
|
|
|
When mediastinal nodal metastasis alone (n = 13), irrespective of hilar nodal metastasis, was considered, peak (p = 0.007) and net (p = 0.016) enhancement and MER (p = 0.046) were found to be significantly different for nodules with and without metastasis. When hilar nodal metastasis alone (n = 19), regardless of mediastinal nodal metastasis, was considered, peak (p = 0.026) and net (p = 0.034) enhancements were significantly higher in nodules with metastasis than in nodules without metastasis (Table 3).
On multivariate analysis, attenuation values at peak and net enhancements and MER were found to be independent positive predictors of nodal metastasis (p = 0.034, odds ratio = 1.032 for peak enhancement; p = 0.028, odds ratio = 1.035 for net enhancement; and p = 0.011, odds ratio = 4.868 for MER) irrespective of the CT findings concerning tumor margin, bronchovascular bundles, size, or tumor necrosis.
At ROC analysis, 112 H of peak enhancement and 61 H of net enhancement showed the highest accuracies for predicting the presence of hilar or mediastinal nodal metastasis. When we applied a cutoff value of 110 H of peak enhancement for the presence of metastatic lymph nodes, sensitivity was 54% (14/26 patients); specificity, 81% (47/58); accuracy, 73% (61/84); positive predictive value, 56% (14/25); and negative predictive value, 80% (47/59). When we applied a cutoff value of 60 H of net enhancement to the presence of metastatic lymph nodes, the sensitivity was 65% (17/26 patients); specificity, 76% (44/58); accuracy, 73% (61/84); positive predictive value, 55% (17/31); and negative predictive value, 83% (44/53) (Tables 4 and 5). When we applied a cutoff value of 110 H of peak enhancement only for the presence of mediastinal metastatic lymph nodes (N2), sensitivity was 62% (8/13 patients); specificity, 76% (54/71); accuracy, 74% (62/84); positive predictive value, 32% (8/25); and negative predictive value, 92% (54/59). Similarly when we applied a cutoff value of 60 H of net enhancement only to the presence of mediastinal metastatic lymph nodes (N2), the sensitivity was 77% (10/13 patients); specificity, 70% (50/71); accuracy, 71% (60/84); positive predictive value, 32% (10/31); and negative predictive value, 94% (50/53). On case-control (odds ratio) study, stage T1 lung cancers showing more than 110 H of peak enhancement were found to have a 4.98 times higher frequency of nodal metastasis than those showing less than 110 H of peak enhancement (95% confidence limits: 1.81, 13.72). T1 lung cancers showing more than 60 H of net enhancement had a 5.94 times higher frequency of nodal metastasis than those with less than 60 H of net enhancement (95% confidence limits: 2.17, 16.26).
|
|
On a per-patient basis, the overall sensitivity of nondynamic staging CT for nodal metastasis was 27% (7/26 patients with positive nodes); specificity, 90% (52/58 patients with negative nodes); accuracy, 70% (59/84 patients); positive predictive value, 54% (7/13 patients); and negative predictive value, 73% (52/71 patients) (Table 6). On a per-nodal station basis, the overall sensitivity of nondynamic CT for detecting metastatic hilar or mediastinal nodes (in which lymph node metastasis was assessed on the basis of size criteria) was 17% (6/35 nodal groups); specificity, 98% (386/393); accuracy, 92% (392/428 nodal groups); positive predictive value, 46% (6/13 nodal groups); and negative predictive value, 93% (386/415 nodal groups).
|
Of 19 patients with false-negative nodal metastasis on nondynamic staging CT (T1 N0 stage), 16 were regarded to indicate a high probability of metastasis by applying the 60-H cutoff value of net enhancement and 14 by applying the 110-H cutoff value of peak enhancement to hemodynamic CT scans. Of six patients with false-positive nodal metastasis (four T1 N1, two T1 N2) on nondynamic CT, five were regarded to indicate a low probability of metastasis by applying the 60-H cutoff value of net enhancement, and all six were regarded to indicate a low probability of metastasis according to the 110-H peak enhancement cutoff value (Table 7).
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Given the increasing detection of microscopic metastasis in small stage T1 lung cancer lesions [5], a more reliable and useful indicator for predicting the possibility of lymph node metastasis is required.
A controversial relation between primary tumor size and disease stage at the time of presentation in stage T1 lung cancer has been reported [6, 9, 10]. Small T1 squamous cell carcinomas show much less chance of hilar or mediastinal nodal metastasis than do large squamous cell carcinomas. However, this is not the case with adenocarcinomas. Moreover, no good correlation exists between tumor size and the presence of hilar or mediastinal nodal metastasis in T1 adenocarcinomas [23, 24].
Aoki et al. [11] reported that a coarse spiculated margin and thickening of the bronchovascular bundles of nodules are seen at a significantly higher frequency in adenocarcinomas with lymph node metastasis than in those without metastasis. In our study, no statistically significant differences in the frequencies of a spiculated margin or in the presence of bronchovascular thickening were observed between patients with and those without nodal metastasis. Discrepancies between the two studies may be explained by differences in cell types of the cancers included. Our study included adenocarcinoma (76%) and other (24%) histologic types, including squamous, large cell, and pleomorphic carcinomas and carcinoid.
In our study, the extent of enhancement was found to be related to mediastinal or hilar nodal metastasis and to be independent of tumor size, margin, necrosis, and bronchovascular thickening. Peak and net enhancements and MER were associated with lymph node metastasis.
On a per-patient basis, the overall sensitivity of nondynamic CT for detecting nodal metastasis was 27% (7/26 patients with positive nodes); specificity, 90% (52/58 patients with negative nodes); and accuracy, 70% (59/84 patients). The lower sensitivity shown by our results compared with previous studies (at least > 41% in terms of detecting mediastinal nodal metastasis) [5, 25] is attributed to the different sample populations, the extent of lymph node dissection, and the rather strict size criterion (10 mm in any axis diameter rather than smaller diameter such as 7 mm) for hilar nodal metastasis. Dynamic CT helped predict nodal metastasis in 54% of patients, meeting the peak enhancement criterion of 110 H or more and in 65% of patients meeting the net enhancement criterion of 60 H or more. Patients with micrometastases, which were not identified by size criterion alone, were suggested to have metastases at dynamic CT. Actually, of 19 false-negative cases on nondynamic CT in this study, 74% (14/19 patients) showed more than 110 H of peak enhancement and 84% (16/19) showed more than 60 H of net enhancement. Consequently, the results of dynamic peak and net enhancement studies may be used to determine the aggressiveness of preoperative staging evaluation in patients with malignant nodules.
PET has been known to be more sensitive than CT in the detection of small-sized metastatic nodes in the mediastinum. However, PET is still limited in the detection of microscopic metastatic nodes [26, 27]. Therefore, dynamic studies may be comparable to PET in the prediction of mediastinal nodal metastasis. Further prospective comparison studies between PET and enhanced dynamic CT are required to document the efficacy of the two diagnostic methods.
The results of our study may affect the diagnostic procedures for stage T1
lung cancer. The usefulness of performing mediastinoscopy in patients with T1
lung cancer has been the subject of much debate for some time. Some reports
have provided evidence that it should be done because of the 10-15% chance of
a clinically unsuspected mediastinal node being positive
[28,
29], whereas many surgeons do
not favor routine mediastinoscopy for peripheral T1 lesions because they think
that the procedure is overly invasive, with significant morbidity and
occasional mortality. However, when the results of dynamic studies are
considered, patients can be stratified in terms of likely benefit from
mediastinoscopy. Patients showing high peak (
110 H) or net ( 60 H)
enhancement on dynamic studies may undergo mediastinoscopy, and patients with
less enhancement may undergo direct surgery without mediastinoscopy because
these patients have little chance of having positive hilar or mediastinal
nodal metastasis, or they have more chance of microscopic metastasis on
pathologic examination.
Our results may also be helpful when planning appropriate surgical treatment, especially in less invasive surgical interventions. In 1995, the Lung Cancer Study Group reported the results of a randomized controlled trial of a comparison between limited resection and lobectomy for clinical stage T1 N0 M0 non-small cell lung cancer [30]. This trial showed the inferiority of limited resection in terms of local relapse and prognosis because patients with pathologic stage N1 or N2 disease may have been included. However, for accurately selected patients (e.g., patients with little enhancement on dynamic studies), limited surgical resection may have some advantages over a standard operation without impairing pulmonary function, especially when stage T1 lung cancer patients have marginal pulmonary functional reserve [31].
Our study has its limitations. First, our dynamic images were not obtained using a unified imaging protocol. We switched the protocol from an early-phase-oriented dynamic study to both early- and delayed-phase-oriented protocols to observe how the delayed phase of dynamic studies might be useful at distinguishing malignant and benign nodules. Therefore, peak attenuation values and times to peak enhancement in the two protocols, especially the latter (delayed-phase-oriented) protocol in which images were obtained at wider time intervals, may have not reflected real values. However, no statistically significant difference was observed in peak and net enhancement values of the two protocols. Second, our study clearly has a selection bias. Patients with cervical mediastinoscopy-positive lymph nodes or extrathoracic metastases were excluded. We also excluded patients with a lung cancer of less than 5 mm in diameter or who underwent incomplete lymph node dissection. Third, substantial overlap exists in the pattern of tumor enhancement in patients with and without nodal metastasis. The statistically significant results in our study are applicable for the entire group of patients. The results of our study may be less satisfactory for the individual patient.
In conclusion, peak enhancement, net enhancement, and MER of stage T1 lung cancers on dynamic CT are found to be significantly associated with the presence of mediastinal or hilar nodal metastasis. T1 lung cancers showing a peak enhancement of 110 H or more or a net enhancement of 60 H or more on dynamic CT are found to have a higher likelihood of hilar or mediastinal lymph node metastasis.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. A Yi, K. S. Lee, B.-T. Kim, S. S. Shim, M. J. Chung, Y. M. Sung, and S. Y. Jeong Efficacy of Helical Dynamic CT Versus Integrated PET/CT for Detection of Mediastinal Nodal Metastasis in Non-Small Cell Lung Cancer Am. J. Roentgenol., February 1, 2007; 188(2): 318 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. J. Jeong, C. A. Yi, and K. S. Lee Solitary Pulmonary Nodules: Detection, Characterization, and Guidance for Further Diagnostic Workup and Treatment Am. J. Roentgenol., January 1, 2007; 188(1): 57 - 68. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B.-T. Kim, K. S. Lee, S. S. Shim, J. Y. Choi, O J. Kwon, H. Kim, Y. M. Shim, J. Kim, and S. Kim Stage T1 Non-Small Cell Lung Cancer: Preoperative Mediastinal Nodal Staging with Integrated FDG PET/CT--A Prospective Study Radiology, November 1, 2006; 241(2): 501 - 509. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
