The lymphatic system Modalities
Conventional radiography
The imaging test most often required in the evaluation of lymphatic system disease is the chest radiograph. Because the lungs are open to the atmosphere and frequently subject to infection, thoracic lymph nodes can be slightly larger than in other organs and are more Often calcified. For interpretation purposes, an upper size limit of 1 cm is generally acceptable. Although the total number of thoracic lymph nodes is about 100, only a few can be assessed by conventional radiography. Moderately large (2-3 cm at least), hilar, mediastinal, paracardiac and paraspinal lymphadenopathy can often be detected by postero-anterior and lateral projection radiography. A contour abnormality is most often the key finding in such cases. The following anatomic regions should be most particularly examined because they almost always contain lymph nodes in close proximity with pleural/lung reflections that are tangential to the x-ray beam. These are: the aorto-pulmonary window, the right tracheobronchial angle, the upper paratracheal regions bilaterally (Fig. 2 a), the subcarinal angle, the right cardiophrenic angle, the paraspinal reflection, the posterior junction line and the azygo-esophageal reflection line. The right bronchus intermedius, as well as the posterior aspects of the left main bronchus and right upper lobe bronchus are best assessed on lateral views where they present easily seen interfaces with aerated lung. In patients with well inflated lungs the retrostemal pre-aortic clear space helps evaluate the anterior mediastinum for masses or adenopathy. In other regions of the body, plain radiography is often ineffective in assessing lymphatic system disease because of lack of soft tissue contrast. Nonetheless, plain abdominal radiographs can be helpful in evaluating the liver and spleen for enlargement. Calcified or previously opacified lymph nodes can be effectively followed by plain radio-graphy.
|
a
|
Figure 1.
Normal lymph node.
a) CT scan. Note small sharply defined node in pretracheal retrocaval space.
b) MRI scan. T1-weighted series. Same node (arrow) appears low in signal intensity due to long T1 of normal lymphoid tissue.
|
|
b
|
The technique of lymphangiography, once very popular in the evaluation of various lymph node beds such as the lower extremities and abdomen to the level of the cisterna chyli is being abandoned in most centres in favour of computed tomography. At a few oncologic centres, lymphangiography is still preferred for the management of certain lymphomas such as Hodgkin's disease because opacified lymph nodes can be assessed repeatedly by simple radiographs over a period of several weeks and involvement of normal sized lymph nodes can be specifically detected by the pattern of opacification. In addition, unlike non-Hodgkin's lymphoma where peripheral nodes such as mesenteric and peripancreatic nodes are more often involved but cannot be visualized by lymphography, Hodgkin's disease involves axial lymph nodes such as para-aortic and iliac groups which are opacifiable. Using fine local dissection of superficial lymphatics after coloration of the interstitial fluids by a vital dye, lymphatics can be visualized and cannulated with very fine needles. Lipid based iodinadated contrast agents or even radioactive
agents can then be injected for evaluating lymph flow and intermediary lymph nodes. The difficulty of successful cannulation has limited the utilization of this technique to a few well skilled radiologists.
Computed tomography
The advent of computed tomography (CT) has revolutionized the imaging assessment of the lymphatic system. The high tissue density contrast between lymph nodes and the fat that often surrounds them is large enough to identify some of the normal and most of the enlarged lymph nodes in all body regions (Figs. 1, 2). The cross-sectional plane of imaging and the utilization of fast dynamic intravenous contrast-enhanced CT permits easy differentiation of blood vessels from lymph nodes in most cases provided sufficient fat surrounds the lymph node. There are, however, several limitations to CT. After initial optimism, it has become clear that CT cannot reliably detect the presence of small focal or diffuse disease in the liver or spleen
a
|
Figure 3.
a) CT with contrast enhancement shows no obvious lymph node.
b) T2-weighted MR scan shows obvious bright lymph node in right hilum.
|
|
 b
|
thus limiting its potential in accurate staging of lymphomas. In the absence of surrounding fat, CT fails to detect many lymph nodes (Fig. 3). Furthermore, CT cannot differentiate malignant from reactive lymph nodes on the basis of density or appearance and its accuracy in cancer staging is thus limited. Size being the only criterion CT can rely upon for assessing normal and abnormal nodes in early metastatic involvement, microscopic or minimal macroscopic disease cannot be detected. CT is useful in monitoring the response of lymph node masses to therapy by monitoring size changes, however, it cannot differentiate residual active disease from fibrotic residual lymph nodes (see below). Despite its limitations, CT remains the best method for imaging the lymphatic system because of its relative ease of use and interpretation. In addition, because it can help direct precisely needle biopsy to lesions as small as 1 cm in regions of difficult access, CT has become a good management tool by helping provide direct histologic confirmation in a minimally invasive fashion. This latter role is bound to grow further in the future as familiarity with such techniques develops but also because an increasing number of molecular genetic markers that provide very specific and sensitive diagnoses with minute amounts of tissue material are being
a
|
Figure 4.
a) T1-weighted and
b) T2-weighted MR scans show a large metastatic lymph node in the right hilum with typical high signal intensity on T2-weighted scan. Note also a smaller node in the precarinal region that remains dark on the T2-weighted scan. This low signal would suggest that this node is normal or fibrotic. However, at pathologic examination, microscopic disease was found. MRl is not reliable in excluding the presence of minimal disease in metastatic nodes.
|
b |
a
|
Figure 5.
a) CT scan shows two small calcified lymph nodes.
b) MRl scan suggests that a larger node with central fat is present and does not clearly detect calcification.
(Courtesy of Dr. Richard Webb, UCSF) |
b
|
discovered at a rapid pace.
Several more specific contrast agents for CT of the reticulo-endothelial system have also been investigated. Lipid based agents such as EOE-13 have been tested in animals and humans. These contrast agents increase the density of the liver and spleen as they are picked up by the reticulo-endothelial cells. This type of agent has not, however, gained wide acceptance. Attempts have also been made at developing methods to visualize lymph nodes of the mesentery by using orally administered oily based iodinated agents with very limited success.
Magnetic resonance imaging
MRI is an imaging technique with greater tissue contrast capabilities than CT. As with CT, lymph nodes can be distinguished from surrounding fat by their different relaxation characteristics. Lymph nodes have much longer T1-relaxation times than fat and thus appear as low intensity structures on T1-weighted images. However, the T2 relaxation times are similar and lymph nodes are not always distinguishable on T2-weighted sequences and are best visualized on T1-weighted sequences (Fig. 1). With the advent of MRI, it was hoped that specific differences in tissue signals will be found between malignant and non-malignant tissues, because tissue parameters such as the T1- and T2-relaxation times vary as a complex function of water content, protein composition and structure. Even though statistically significant differences in both the T1 and T2 relaxation times of malignant and non-malignant nodes do exist, the overlap between signal characteristics is such that diagnostic utility is limited in the individual patient (Fig. 4). MRI can to some extent differentiate fibrotic lymph nodes which exhibit low signal intensity on T2-weighted series from inflammatory or neoplastic nodes which appear much brighter and this has found some limited use in the differentiation of retro-peritoneal fibrosis from tumor, post-radiation fibrosis from recurrent tumor in lymphomas, rectal and uterine cervical cancers. Microscopic disease is not detectable and calcification may not always be detected with MRI (Fig. 5). Because of these limitations, as well as the greater cost and longer examination times of MRI, CT remains the preferred method of investigation for lymph node pathology except in patients with allergy to iodinated contrast agents. In the evaluation of the liver and spleen, MRI is equally accurate to CT and in selected instances, can show diffuse or small multifocal involvement of the liver and spleen
|
a
|
b
|
Figure 6.
Untreated Hodgkin's disease in anterior mediastinum (arrows).
a) T1-weighted scan
b) T2-weighted scan. Note high but heterogeneous signal intensity on T2-weighted scan, typical of active lymphoma.
|
|
a
|
b
|
|
c
|
d
|
Figure 7.
Evolution of treated Hodgkin's disease. A series ofT2-weighted MRl scans at time of diagnosis (a); 6 weeks (b); 3 months (e) and l year (d) after initiation of therapy. Note the decreasing signal intensity typical of fibrosis and the parallel change in size typical of good therapeutic response.
|
that cannot be detected with
CT. In evaluating liver and spleen, it is felt that
MRI and
CT play complimentary roles and are often used jointly in patients with known malignancies in whom accurate evaluation of the liver and spleen is necessary.
MRI is very effective in the evaluation of the bone marrow. Fatty marrow is easily differentiated from red marrow and the normal conversion patterns of red fatty marrow with aging have been well described.
MRI is considered accurate in evaluating the presence or absence of
diffuse or
focal bone marrow disease. In many cases, the MR examination may be positive for metastatic disease to the bone marrow before
radionuclide bone scanning. In primary bone marrow diseases such as myeloma,
lymphoma and leukemia, MR imaging of the bone marrow is the most accurate method to stage and follow bone marrow involvement.
MRI can also be useful in the management of patients with known lymphoma in whom partial regression of tumor masses is often observed. These residual masses most often represent the fibrotic residual of a sterilized tumor mass. It can, however, also represent residual active disease. With CT, it is impossible to differentiate fibrosis from residual active disease. With MRI, significant differences in the signal intensity pattern of fibrosis and residual tumor have been demonstrated on T2-weighted series (Figs. 6 and 7). MRI is thus used at several centres as a method of monitoring the residual masses in patients with known lymphomas. If the signal intensity of these masses is low on T2-weighted series, they are presumed to represent residual fibrosis. However, if the signal intensity remains high beyond six months following initial therapy they are considered to represent residual disease and further diagnostic steps are undertaken. More importantly, reappearance of high signal intensity foci in a previously fibrotic appearing mass is a reliable sign of tumor recurrence.
To enhance the detection and assessment of lymphatic system pathology, multiple contrast agents are being investigated for use with MRI. MRI has the distinct advantage of greater sensitivity to smaller amounts of contrast agent as compared to computed tomography. Because of this increased sensitivity, efforts have been directed to the development of superparamagnetic iron oxide particles which are cleared from the blood stream by the reticulo-endothelial system. The presence of a superparamagnetic particle reduces signal intensity considerably due to a marked shortening of the T2 relaxation time. In the liver and spleen, the reticulo-endothelial system effectively clears such agents and leads to a marked decrease in the signal intensity of normal liver and spleen, thus enhancing the detection of underlying pathology. In addition, since the clearance from the blood stream is not immediate, these agents can be used to measure perfusion in various organs. If made small enough, these superparamagnetic iron oxide particles can leave the vascular space before trapping by reticular endothelial cells and can penetrate the interstitial extra-cellular space. Since foreign particles in the interstitial space are primarily cleared through the lymphatics, these particles eventually accumulate in the lymph nodes. Early clinical trials have shown that normal lymph nodes will readily accumulate these particulate contrast agents whereas lymph nodes involved by metastastic disease do not. It is hoped that these newer agents will enable more accurate assessment of loco-regional lymph nodes, thus facilitating the staging of various types of cancers.
Because of the superb sensitivity of radionuclide imaging, albeit at lower spatial resolution, the lymphoreticular system has been extensively imaged with multiple radioactively labelled agents. Using reticulate agents such as Technetium 99m sulfur colloid particles, the liver and spleen can easily be demonstrated. Using smaller particles, bone marrow uptake can be improved, however, bone marrow imaging with radionuclide techniques has not gained wide acceptance because of the inconsistency of uptake. Over the years, many efforts have been expended in the development of more specific agents including labelled monoclonal antibodies and cells such as leukocytes in the hope of using immunological mechanisms for agent localization. These efforts have generally failed. A notable exception is the use of Gallium 67 scanning which appears to more reliably detect areas of disease activity in both inflammatory and neoplastic processes. For example, in the management of lymphoma, Gallium 67 scanning is reliable at assessing the disease after therapy. Over the past few years, the use of single photon emission computed tomography (SPECT) has improved the spatial resolution of radionuclide methods. Progress in radiochemistry and labelling agents may lead to an increased use of SPECT for assessing diseases of the lymphatic system. More recently, the development of metabolic agents such as FDG-glucose with positron emission tomography (PET) offer the hope of directly observing the metabolic activity of diseased tissues. Already, preliminary studies show that positive FDG-glucose imaging in lymph nodes correlates highly with the presence of metastatic disease. This concept is being used in lung, colorectal and breast cancer as a new method of detecting lymph node involvement. Greater experience will be needed to assess the diagnostic accuracy of such approaches. With the advent of total body high resolution efficient PET scanners, it is possible that cancer staging may be more accurately performed with PET in the future. The specificity of the metabolic activity associated with increased uptake of FDG-glucose, however, needs to be ascertained before widespread clinical use.
Elias Zerhouni