| 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, Gangnam-gu,
Seoul 135-710, Korea.
2 Department of Radiology and Medical Imaging Laboratory, Asan Medical Center,
University of Ulsan College of Medicine, Seoul, Korea.
Received October 24, 2006;
accepted after revision February 28, 2007.
Address correspondence to K. S Lee
(kyungs.lee{at}samsung.com).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. From July 2005 to May 2006, 127 patients (99 men, 28 women; mean age, 63 years) with histopathologically proven non-small cell lung cancer underwent both CT and MRI of the chest. Transverse MR images were obtained with T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood turbo spin-echo sequences on a 3-T MRI system. Two chest radiologists assessed CT images and then MR images. The morphologic features of lung cancer and the detectability of small nodules in lobes not containing primary tumors on MR images were compared with the findings on CT images, which were the reference standard.
RESULTS. The morphologic characteristics of primary cancer found on both T1- and T2-weighted images corresponded to those on CT images. The overall rates of detection of nodules in lobes not containing primary tumors were 57% (184 of 323 nodules) and 56% (180 of 323 nodules) on T1- and T2-weighted images, respectively (p = 0.64). In terms of detection of non-calcified nodules 5-10 mm in diameter, both T1- and T2-weighted images had a detection rate of 92% (48 of 52 nodules) (p =1.00).
CONCLUSION. Both T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood turbo spin-echo 3-T MR images depict clinically significant small (5-10 mm in diameter) noncalcified pulmonary nodules nearly as well as do CT scans.
Keywords: high field strength • lung • lung neoplasms • lung nodule • MRI
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Signal-to-noise ratio is directly related to static magnetic field strength. Because of improved signal-to-noise ratios, MR units with field strength greater than 1.5 T have been found to have upgraded capabilities in terms of temporal and spatial resolution [8-10]. High-field-strength magnets, however, are prone to problems, such as increased susceptibility effects due to magnetic field inhomogeneity, especially when large fields of view are used [3].
The turbo spin-echo (TSE) sequence has been proposed to reduce susceptibility artifacts and to increase signal-to-noise ratio. This sequence is based on rapid repetitive rephrasing with a train of multiple 180° refocusing pulses, particularly with short echo spacing. Therefore, TSE sequences appear to be suitable for lung MRI. A more refined T2-weighted TSE sequence adopts breath-hold ECG-gated and black blood technique to eliminate flow and motion artifacts in the lung [11]. In addition, STIR sequences for fat suppression can enhance the conspicuity of focal lung lesions [11, 12]. The HASTE sequence for breath-hold cardiac-gated T2-weighted imaging is another practical thoracic imaging sequence for lung parenchyma [13]. HASTE images, however, are usually blurred because of T2 decay in later echoes during data acquisition. To the best of our knowledge, these newly developed and refined pulse sequences, which are based on TSE applications, have not been used for thoracic imaging under high-field-strength conditions. The purpose of our study was to evaluate the diagnostic performance of 3-T MRI for the detection and characterization of primary NSCLC and of small nodules in lobes not containing primary tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In all 127 cases, the histopathologic cell type of the primary tumor was determined with percutaneous needle aspiration biopsy (n =15) or surgery (n = 112). The pathologic subtypes of lung cancer were as follows: adenocarcinoma in 60 (47%), squamous cell carcinoma in 41 (32%), NSCLC of unspecified subtype in 17 (13%), large cell neuroendocrine carcinoma in four (3%), sarcomatoid carcinoma (spindle cell carcinoma) in three (3%), and pleomorphic carcinoma in two (2%) of the cases.
CT Imaging
All chest CT scans were obtained with 4-MDCT (LightSpeed QX/i, GE
Healthcare) or 16-MDCT (LightSpeed Ultra or Ultra16, GE Healthcare) scanners.
After IV administration of contrast medium (2 mL/s; total, 80 mL of iomeprol,
Iomeron 300, Bracco) with a power injector (MCT Plus, Medrad), CT scans (125
mA, 120 kVp, beam width of 10-20 mm, beam pitch of 1.375-1.5) for tumor
staging were obtained from the lung apices to the level of the middle portion
of both kidneys. Image data were reconstructed with a bone algorithm and
5.0-mm section thickness. All image data were directly interfaced with a PACS
workstation (Path-speed, GE Healthcare), which displayed all image data on
monitors (four monitors, 1,536 x 2,048 matrix size, 8-bit viewable gray
scale, 60-foot-lambert luminescence). Both mediastinal (width, 400 H; level,
20 H) and lung (width, 1,500 H; level, -700 H) windows were viewed on these
monitors.
MRI
All MRI studies were performed with a 3-T unit (Achieva, Philips Medical
Systems) equipped with a gradient system capable of a maximum gradient
amplitude of 80 mT/m, rise time of 0.2 millisecond, and slew rate of 200
T/m/s. A sensitivity-encoding cardiac coil with a six-coil element was used to
obtain axial images of the thorax.
A breath-hold T2-weighted triple inversion black blood TSE sequence was used with cardiac gating and the following parameters: TR/TEeff, 1,200-1,800/60 (2 R-R intervals); echo-train length, 21; field of view, 400 mm; acquisition matrix size, 256 reconstructed to 512; slice thickness, 5 mm with 1-mm interslice gap. A double-inversion blood-nulling preparation pulse was applied at the R-wave trigger to suppress blood signals with an inversion delay time of 697.6 milliseconds and was followed by a third inversion pulse to generate spectral presaturation attenuated by inversion recovery contrast enhancement for fat suppression and a greater sensitivity for tissue fluid with an inversion delay of 90 milliseconds. Two slice acquisitions per breath-hold were continued for 40-44 slices to image all of both lungs. The scanning time per breath-hold was 16 seconds, giving rise to a total acquisition time of 5.20-5.52 minutes, depending on total number of slices. A sensitivity-encoding factor of 2 was applied in the phase-encoding direction to increase acquisition speed.
A T1-weighted 3D multishot turbo field-echo sequence was used during breath-holds. The sequence was optimized with a turbo field-echo factor of 50, 24 turbo field-echo shots, and a radial turbo direction with the following acquisition parameters: TR/TE, 3.0/1.49; flip angle, 10°; number of excitations, 2; field of view, 400 mm; acquisition matrix size, 224 reconstructed to 512. Fat suppression was achieved with the spectral presaturation attenuated by inversion recovery method with an inversion delay of 90 milliseconds. Four consecutive slabs without a gap were obtained in the foot-to-head direction to image all of both lungs in a single breath-hold per slab. The slab thickness was 66 mm, and each slab was divided into 11 partitions to produce 11 images with a 6-mm section thickness. Imaging time per breath-hold was 13.5 seconds for one slab, giving rise to a total acquisition time of 54 seconds for four slabs covering the whole thorax.
Image Analysis
Two chest radiologists with 7 and 3 years of experience in chest CT
interpretation and each with 2 years of experience in thoracic MRI
interpretation, who reached the decisions on findings by consensus, performed
all measurements retrospectively in a step-by-step manner
[7]. First, the two chest
radiologists assessed the CT images, which were used as reference standards.
The reviewers evaluated primary lung cancer nodules for size, location, and
morphologic features. The morphologic features estimated were presence or
absence of a spiculated or lobulated margin, a satellite lesion, or a cavity
within a nodule. The reviewers also assessed the presence of pulmonary nodules
in lobes not containing primary tumors. If nodules were present, the
radiologists recorded size, number, location, and presence or absence of
calcification. The sizes of the pulmonary nodules were classified into three
groups: 3-5 mm, 5-10 mm, and more than 10 mm in longest diameter.
Second, the two chest radiologists evaluated MR images obtained with T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood TSE sequences using the same procedure as for the CT images. To avoid recall bias, the assessments of CT scans and MR images were scheduled more than 2 months apart. Third, corresponding CT and MRI data sets were reviewed simultaneously by the two radiologists for side-by-side comparison of primary lung cancer nodules in terms of size and morphologic characteristics. This assessment was followed by similar comparisons of the visibility, number, and size of nodules in lobes not containing primary tumors. Six small nodules in lobes not containing primary tumors were diagnosed histopathologically after wedge lung resection. The other small nodules in lobes not containing primary tumors were observed by follow-up with serial CT studies.
Statistical Analysis
The Friedman test was used to analyze statistical differences between
longest measured diameters of primary lung cancer nodules on MDCT images and
on T1-weighted 3D TSE and T2-weighted triple inversion black blood TSE 3-T MR
images. Differences between the two 3-T MRI sequences in terms of diagnostic
efficacy for morphologic characterization of primary lung cancer nodules and
for detection of the presence of pulmonary nodules in lobes not containing
primary tumors were tested for significance with the McNemar test and
Bonferroni correction for multiple comparisons
[14]. Significance was
accepted at p < 0.05 for all tests.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The morphologic characteristics of primary lung cancer nodules found on CT scans were as follows: spiculated and lobulated margins in 97% (109 of 112), accompanying satellite lesions in 13% (15 of 112), and a cavity within the primary lung cancer nodule in 17% (19 of 112) of the patients. These morphologic features were depicted on T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood TSE images with the following accuracy rates: spiculated and lobulated margins, 97% (109 of 112) and 97% (109 of 112); satellite lesions, 93% (104 of 112) and 96% (107 of 112); and a cavity within the primary cancer nodule, 95% (106/112) and 96% (108/112). No significant differences were found between the two MRI sequences in depiction of the morphologic characteristics of primary lung cancer nodules.
Detection Rate of Pulmonary Nodules in Lobes Not Containing Primary Tumors
Three hundred twenty-three reference standard pulmonary nodules in lobes
not containing primary tumors were found with CT in 89 (70%) of the 127 NSCLC
patients. Of 323 nodules, 137 (42%) contained calcification, but the other 186
nodules did not. The overall detection rates for small nodules in lobes not
containing primary tumors were 57% (184 of 323) on T1-weighted 3D turbo
field-echo images and 56% (180 of 323) on T2-weighted triple inversion black
blood TSE MR images (p = 0.64). Twenty-three (11%) of 207 small
nodules detected on T1-weighted 3D turbo field-echo and 32 (15%) of 212
nodules detected on T2-weighted triple inversion black blood TSE images were
false-positive findings. The positive predictive values for these nodules are
shown in Table 2.
|
The rate of detection of noncalcified nodules on T1-weighted turbo field-echo images was 53% (99 of 186); the rate for calcified nodules was 62% (85 of 137). The rate of detection of noncalcified nodules on T2-weighted triple inversion black blood TSE MR images was 57% (106 of 186); the rate for calcified nodules was 54% (74 of 137) (Table 2) (Fig. 1A, 1B, 1C).
|
|
|
|
In contrast, three biopsy-proven small metastatic nodules in lobes not containing primary tumors manifested higher signal intensity and greater diameter on T2-weighted triple inversion black blood TSE images than on T1-weighted 3D turbo field-echo images (Fig. 2A, 2B, 2C). Three biopsy-proven small benign pulmonary nodules (anthracofibrotic nodules) had lower signal intensity and smaller diameter on T2-weighted triple inversion black blood TSE images (Fig. 3A, 3B, 3C). For nodules larger than 10 mm in diameter, two calcified nodules were missed on T2-weighted triple inversion black blood TSE images, and a nodule of ground-glass opacity was missed on T1-weighted turbo field-echo images. Overall, detection rates according to nodule diameter were similar for T1-weighted turbo field-echo and T2-weighted triple inversion black blood TSE MRI (Table 3).
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In clinical practice, because they have little likelihood of being malignant, pulmonary nodules smaller than 5 mm in diameter found incidentally or identified in patients with lung cancer or with extrathoracic malignant tumors are not of great concern. Kim et al. [17] reported that pulmonary nodules smaller than 10 mm in diameter in lobes not containing primary tumors in patients with surgically resectable lung cancer have a low risk of metastasis (4%, six of 138 nodules). Therefore, it appears that clinically significant small (5-10 mm in diameter) pulmonary nodules are likely to be well depicted on T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood TSE 3-T MR images.
In our study population, a large number of pulmonary nodules in lobes not containing primary tumors were calcified (42%, 137 of 323 nodules). Heavily calcified lesions have been classically characterized as having low signal intensity on both T1- and T2-weighted MRI, because calcium salts do not contain mobile protons and thus have no signal intensity on MR proton images [18, 19]. However, calcification per se shortens T2-weighted but not T1-weighted relaxation time. Thus, calcified small nodules may appear as unidentifiable lesions with highly exaggerated signal void effects under high-field-strength conditions. Moreover, calcified nodules can be visualized as lesions of variable signal intensity, even as high-signal-intensity lesions, on T1-weighted images. This factor depends on the particle size and precise composition of calcium salt aggregates [19, 20]. Our results for calcified nodules in lobes not containing primary tumors corroborate the aforementioned theories. On T2-weighted triple inversion black blood TSE images, the rate of detection of noncalcified pulmonary nodules in lobes not containing primary tumors was increased compared with the rate for calcified nodules. However, on T1-weighted 3D turbo field-echo images, the rate of detection of noncalcified nodules was decreased compared with the rate for calcified nodules (Table 2).
Other results of our study concern the efficacy of the T2-weighted triple inversion black blood TSE sequence in detection of metastatic pulmonary nodules. All three biopsy-proven metastatic pulmonary nodules in lobes not containing primary tumors had higher signal intensity and appeared larger in diameter on T2-weighted triple inversion black blood TSE images than on T1-weighted 3D turbo field-echo images. On the other hand, an anthracofibrotic nodule had lower signal intensity and appeared smaller in diameter on T2-weighted triple inversion black blood TSE images. These findings corroborated the fact that the fibrotic components of lesions have low MRI signal intensity and that closely packed tumor cells produce isointensity to high signal intensity on T2-weighted images [16, 21]. Therefore, size comparison between T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood TSE sequences may help characterize small pulmonary nodules in lobes not containing primary tumors. Another advantage of the T2-weighted triple inversion black blood TSE sequence in the detection of small pulmonary nodule stems from its suppression of blood signal intensity. This technique causes vascular structures to become invisible in the peripheral aspect of the lung and thus improves detection of small pulmonary nodules.
There were limitations to our study. First, we included a relatively small number of nodules, especially nodules larger than 5-10 mm in diameter. Second, when we determined the optimal sequence for fast T2-weighted images of the thorax, we did not prospectively compare HASTE and T2-weighted triple inversion black blood TSE sequences. As a result, we are unsure which sequence is better for depicting small lung nodules. We plan to rectify this limitation with a prospective comparative study of the efficacy of detection of lung and mediastinal lesions with fast T2-weighted image sequences—namely, HASTE, T2-weighted triple inversion black blood TSE, and T2-weighted turbo STIR sequences—on a 3-T MRI system. Third, we lacked true volumetric acquisitions for the MRI sequences for comparison with the volumetric nature of MDCT. Fourth, we reconstructed both mediastinal and lung window images using a bone algorithm. Mediastinal window images reconstructed with a bone algorithm may have rings of high signal intensity at the interface of soft tissue and air. Thus small nodules may have falsely appeared bright and calcified.
In conclusion, 3-T MRI with T1-weighted 3D turbo field-echo and T2-weighted triple inversion black blood TSE sequences well shows primary lung cancer lesions with good anatomic resolution and lesion contrast in NSCLC patients. These sequences allow detection of clinically significant pulmonary nodules—that is, noncalcified nodules larger than 5 mm in diameter in lobes not containing primary tumors. Moreover, under 3-T conditions, MR images obtained with the T2-weighted triple inversion black blood TSE sequence are expected to more accurately depict small metastatic nodules in lobes not containing primary tumors than are images obtained with the T1-weighted 3D turbo field-echo sequence.
|
TopAbstractIntroductionMaterials 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. M. Shin, K. S. Lee, B.-T. Kim, H. Kim, O J. Kwon, J. Y. Choi, and M. J. Chung Non-Small Cell Lung Cancer Staging: Efficacy Comparison of Integrated PET/CT versus 3.0-T Whole-Body MR Imaging Radiology, August 1, 2008; 248(2): 632 - 642. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A Yi, K. S. Lee, J. Han, M. P. Chung, M. J. Chung, and K. M. Shin 3-T MRI for Differentiating Inflammation- and Fibrosis-Predominant Lesions of Usual and Nonspecific Interstitial Pneumonia: Comparison Study with Pathologic Correlation Am. J. Roentgenol., April 1, 2008; 190(4): 878 - 885. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
