Despite the apparent
advantages of MR imaging- no ionizing radiation, the
ability to image in any plane, superior contrast
resolution, and the lack of need for intravenous
contrast -- it remains a secondary imaging modality for
the evaluation of bronchogenic carcinoma [14]. However,
as protocols evolve and scan times decrease, MR is being
investigated for lung cancer staging.
Like CT, MR is not specific
in differentiating hyperplastic from malignant
lymphadenopathy and primarily relies upon size criteria
to suggest the presence of nodal metastases [23,64].
Other findings that suggest a malignant node include
high signal intensity with eccentric cortical thickening
or obliteration of the fatty hilum on T2-weighted FSE
images [190]. Benign lymph nodes typically demonstrate
low signal intensity or a normal fatty hilus [190].
False positive exams can occur with inflammatory or
infectious adenopathy [190]. In general, standard MR
imaging has not been shown to be superior to CT for
mediastinal lymph node staging [29,65], nor for the
evaluation of hilar adenopathy [37]. A meta-analysis
found a sensitivity of 87% and a specificity of 88% on a
per patient basis for identification of pathologic
adenopathy in non-small cell lung cancer patients [208].
Diffusion weighted imaging, may beneficial for lymph
node classification [207]. Diffuse weighted imaging
(DWI) is based on the difference in the diffusion of
water molecules in biologic tissue [207]. The
restriction of the diffusion of water molecules in
malignant tumors results in a decreased apparent
diffusion coefficient (ADC) compared to normal tissues
[207]. A meta-analysis of DWI in the characterization of
nodal metastases in lung cancer patients found a pooled
per patient sensitivity of 68%, and a specificity of 92%
[207].
Attempts have been made to improve the accuracy of MR
imaging through the use of superparamagnetic biodegradable
iron-oxide particles [143]. These agents are phagocytized
by the normal reticuloendothelial system and shorten T2
relaxation times (which results in decreased signal
intensity). Therefore, normal nodes will lose signal
compared to a baseline exam, while lack of signal change
is indicative of metastatic involvement [143]. Although
promising, these agents have not gained significantly in
popularity. STIR MR imaging has also been evaluated for
the detection of lymph node metastases [187,203]. Abnormal
signal has been defined as a signal intensity greater than
muscle (or it can also be quantified in relationship to a
phantom) [187]. Reported sensitivities are 77-100%,
specificity 71-97%, PPV 83%, NPV 87%, and an accuracy of
85-96% [187]. False positive examinations can occur in
association with inflammation, active sarcoid, and motion
[187].
Similarly, MRI has not been
found to be superior to CT in determining the
T-classification of the primary lesion [65]. However, as
a result of it's multiplanar capabilities, MRI is
probably better in the assessment of superior sulcus
tumors, the aorto-pulmonary window, and in assessing
mediastinal, chest wall or diaphragmatic invasion
[14,23,29,49,65,66], and good spatial resolution appears
to be the main factor for detection of parietal invasion
[47]. Reported sensitivity for determination of chest
wall invasion is 88-90%, and specificity is 86-100%
[14,47]. Despite it's improved delineation of chest wall
invasion, MR imaging is still less accurate for the
detection of rib destruction when compared with CT [67].
Additionally, with helical CT and the ability to perform
more detailed reconstruction images, the differences
between CT and MRI in the evaluation of chest wall
invasion may be diminishing [53,54,99].
Whole body MR imaging utilizing real time gradient echo
imaging and a sliding table platform has become feasible
[184,185]. Compare to PET/CT, whole body MR imaging is
better at identifying brain and liver metastases, while
PET/CT is better at identification og lymph node and soft
tissue lesions [184].
Given the lack of
significant improvement in the staging of bronchogenic
carcinoma except in selected instances as discussed
above, it would appear that the routine use of MRI for
the staging of bronchogenic carcinoma is not
indicated.
MRI in chest wall and
mediastinal invasion
Findings on MRI which
suggest the presence of chest wall invasion include
[47,49]:
1) Signal intensity
identical to that of the primary tumor within the chest
wall on T1-weighted images
2) Intraparietal hyperintense
signal on T2-weighted images
3) Focal high-signal
abnormality extending into the chest wall from the
adjacent lung tumor on T2 weighted images
4) Chest wall thickening on
T2 weighted images compared to the uninvolved side
5) Diffuse increased signal
within the chest wall soft tissues on T2 weighted
images
6) Intraparietal enhancement
following the administration of gadolinium
7) For superior sulcus
tumors, obliteration of the adipose tissue rim above the
lung apex [14]
Examples of MRI in patients
with Bronchogenic Carcinoma
Example 1: The images
below are from a patient with a superior sulcus tumor
(T1-weighted image left, T2-weighted image right).
Findings indicating chest wall invasion include signal
intensity identical to that of the primary tumor
within the chest wall and obliteration of the adipose
tissue rim above the lung apex (yellow arrows). The
subclavian vein is indicated by the blue arrow. The
ability to reconstruct images in multiple planes with
MR can aid in accurately defining tumor anatomy. (Case
courtesy of Julie Takasugi, MD. Veterans
Administration Medical Center, Department of
Radiology, Seattle WA.)
NOTE:
To load a higher resolution view, simply click directly on the image
below.
Example 2: This patient
with a superior sulcus tumor had evidence of rib
destruction on CT (black arrows). The sagittal
T1-weighted MR image also demonstrates the rib
destruction (blue arrow). Overall, however, MR is felt
to be less sensitive than CT for the detection of rib
destruction. (Case courtesy of Julie Takasugi, MD.
Veterans Administration Medical Center, Department of
Radiology, Seattle WA.)
NOTE:
To
load a higher resolution view, simply click
directly on either image below.
Example 3: The image on
the right is an axial T1-weighted image demonstrating
a mass invading the left pulmonary artery. The mass
abuts the descending thoracic aorta for an arch of
approximately 90 degrees. Two peripheral metastatic
foci can also be identified on the image. The image on
the left is an oblique coronal T1-weighted exam with
and without gadolinium showing the mass and its
heterogeneous enhancement. (Case courtesy Scott Flamm,
Cardiovascular MR Imaging Section, Department of
Radiology, Cleveland Clinic Foundation).
NOTE:
To
load a higher resolution view, simply click
directly on either image below.
Example 4: This is another
example of the excellent anatomic resolution that can
be obtained with MR. The superior sulcus tumor in this
case is easily identified by its low signal (black
arrows) compared to the higher signal intensity fat on
this T1-weighted coronal image. (Case courtesy of
Julie Takasugi, MD. Veterans Administration Medical
Center, Department of Radiology, Seattle WA.)
