CME ARTICLE
MR Lymphangiography: How I Do It Lee M. Mitsumori, MD,1* Elizabeth S. McDonald, MD, PhD,2 Gregory J. Wilson, PhD,3 Peter C. Neligan, MD,4 Satoshi Minoshima, MD, PhD,3 and Jeffrey H. Maki, MD, PhD3 This article is accredited as a journal-based CME activity. If you wish to receive credit for this activity, please refer to the website: www. wileyhealthlearning.com/jmri
ACCREDITATION AND DESIGNATION STATEMENT Blackwell Futura Media Services designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM. Physicians should only claim credit commensurate with the extent of their participation in the activity. Blackwell Futura Media Services is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
EDUCATIONAL OBJECTIVES Upon completion of this educational activity, participants will be better able to: 1. Interpret an MR lymphangiography exam. 2. Recognize the pathophysiology and microsurgical treatment of lymphedema.
ACTIVITY DISCLOSURES No commercial support has been accepted related to the development or publication of this activity.
Faculty Disclosures: Editor-in-Chief: Mark E. Schweitzer, MD, has no relevant financial relationships to disclose. CME Editor: Scott B. Reeder, MD, PhD, discloses personal stock in Cellectar Biosciences and Neuwave Medical. CME Committee: Shreyas Vasanawala, MD, PhD, discloses research support from General Electric, and founder’s equity in Morpheus Medical. Scott K. Nagle, MD, PhD, discloses consulting fees from Vertex Pharmaceuticals for consulting in design of cystic fibrosis clinical trials involving imaging; and departmental research support from General Electric for evaluation of products and development. Mustafa R. Bashir, MD, discloses research support from Siemens Healthcare and Bayer Healthcare.
Tim Leiner, MD, PhD, discloses research support grant funding from Bracco, S.p.A., Philips Healthcare, and Bayer Healthcare. Bonnie Joe, MD, PhD, has no relevant financial relationships to disclose. Authors: Lee M. Mitsumori, MD; Elizabeth S. McDonald, MD, PhD; Greg Wilson, PhD; Peter C. Neligan, MD; and Satoshi Minoshima, MD, PhD have no relevant financial relationships to disclose. Jeffrey H. Maki, MD, PhD discloses grant support from Bracco Diagnostics for unrelated research on MR contrast agents. This manuscript underwent peer review in line with the standards of editorial integrity and publication ethics maintained by Journal of Magnetic Resonance Imaging. The peer reviewers have no relevant financial relationships. The peer review process for Journal of Magnetic Resonance Imaging is double-blinded. As such, the identities of the reviewers are not disclosed in line with the standard accepted practices of medical journal peer review. Conflicts of interest have been identified and resolved in accordance with Blackwell Futura Media Services’ Policy on Activity Disclosure and Conflict of Interest.
INSTRUCTIONS ON RECEIVING CREDIT For information on applicability and acceptance of CME credit for this activity, please consult your professional licensing board. This activity is designed to be completed within an hour; physicians should claim only those credits that reflect the time actually spent in the activity. To successfully earn credit, participants must complete the activity during the valid credit period. Follow these steps to earn credit: Log on to www.wileyhealthlearning.com Read the target audience, educational objectives, and activity disclosures. Read the article in print or online format. Reflect on the article. Access the CME Exam, and choose the best answer to each question. Complete the required evaluation component of the activity. This activity will be available for CME credit for twelve months following its publication date. At that time, it will be reviewed and potentially updated and extended for an additional period.
View this article online at wileyonlinelibrary.com. DOI: 10.1002/jmri.24887 Received Apr 15, 2014, Accepted for publication Nov 21, 2014. *Address reprint requests to L.M.M., Department of Radiology, Straub Clinic and Hospital, 888 South King St., Honolulu, HI 96813. E-mail: [email protected] From the 1Department of Radiology, Straub Clinic and Hospital, Honolulu, Hawaii, USA; 2Department of Radiology, University of Pennsylvania, Philadelphia, Pennsylvania, USA; 3Department of Radiology, University of Washington, Seattle, Washington, USA; and 4Department of Plastic Surgery, University of Washington, Seattle, Washington, USA
C 2015 Wiley Periodicals, Inc. V 1465
Journal of Magnetic Resonance Imaging
Lymphedema is a chronic progressive edematous disease that in the United States is most commonly related to malignancy and its treatment. Lymphaticovenular anastomosis is a recently introduced microsurgical treatment option for lymphedema that requires the identification and mapping of individual lymphatic channels. While nuclear medicine lymphoscintigraphy has been the primary imaging modality performed to evaluate suspected lymphedema, lymphoscintigraphy does not provide the spatial information necessary for presurgical planning. High-resolution dynamic 3D magnetic resonance imaging (MRI) can noninvasively image abnormal lymphatic channels to both diagnose lymphedema and depict the location and number of individual lymphatic channels for surgical planning. MR lymphangiography can be performed at 1.5T or 3.0T using multichannel phased array surface coils. The main components of the exam are a heavily T2-weighted 3D sequence to define the severity and extent of edema, a high-resolution dynamic 3D gradient echo imaging after intracutaneous contrast injection to visualize lymphatic channels, and a delayed 3D gradient echo sequence after intravenous contrast to define veins. This article reviews the pathophysiology and microsurgical treatment of lymphedema, presents the imaging protocol used at our institution, and describes exam interpretation and the image postprocessing performed for surgical planning. J. MAGN. RESON. IMAGING 2015;42:1465–1477.
LYMPHATIC CIRCULATION AND LYMPHEDEMA
T
he normal lymphatic system provides drainage of macromolecules, metabolic waste products, and excess interstitial fluid that has not been resorbed by the venous circulation. The lymphatic system begins as blind-ending sinuses that progressively coalesce to form larger collecting lymphatic channels.1 Ultimately, the fluid within the peripheral lymphatic system is propelled to the thoracic duct where the lymph fluid is returned to the circulatory system at the left subclavian vein (Fig. 1a).2 Peripheral lymphedema is a chronic, progressive, edematous disease caused by abnormal lymphatic drainage. The disease is characterized by the accumulation of proteinrich fluid, inflammation, adipose tissue hypertrophy, and fibrosis in the interstitial space. More than 90 million people are affected by lymphedema worldwide, primarily the result of filarial infections (Wuscheria banrofti). In the U.S., peripheral lymphedema affects nearly 10 million people, where it is most commonly related to the treatment of breast and gynecological malignancy.1 Upper extremity lymphedema rates after surgical treatment for breast cancer vary between 2% and 65%, depending on the type of procedure performed, with the highest rates occurring following lymph node dissection or nodal radiation therapy.3 As with breast cancer, lower extremity lymphedema rates vary depending on the surgical treatment, occurring in 25–67% of patients after inguinal-femoral lymph node dissection and 2–41% after pelvic lymph node dissection.4 There are conservative and operative treatment options for peripheral lymphedema. Unfortunately, the efficacy of conservative treatment is highly variable between patients and is not curative.5 Operative approaches include debulking, lymph node transplantation, and microsurgical procedures.4,6 Since debulking procedures can in themselves worsen the severity of the lymphedema, and the efficacy of lymph node transplantation has not yet been established7,8; microsurgical treatment of peripheral lymphedema is increasingly being performed at different centers around the world.7,9–11 1466
MICROSURGICAL TREATMENT Since the lymphatic and venous systems flow in parallel to return fluid to the heart, the creation of a lymphaticovenular anastomosis (LVA) bypasses a lymphatic obstruction by shunting distal lymph flow into adjacent veins (Fig. 1b). Because of the small size (0.2–0.8 mm) of the subdermal lymphatic channels used for the procedure, microsurgical techniques are required.12,13 Using a surgical microscope, the surgeon localizes an individual subdermal lymphatic channel as well as an adjacent subdermal venule. Once a suitable lymphatic channel and venule are identified, both vessels are divided and a LVA is created (Fig. 2). While LVA can be performed at the time of the lymph node dissection,14 this strategy is currently not widely performed since not all patients develop lymphedema after lymph node dissection, and many oncologic surgeons lack the microsurgical expertise to perform the procedure. The main challenge with LVA procedures lies in finding the small subdermal lymphatic channels and venules needed for performing the anastomoses.10 While radionuclide lymphoscintigraphy has been considered the primary clinical imaging modality to diagnose lymphedema,1,4 its limited temporal and spatial resolution does not allow the identification and localization of individual lymphatic channels.15,16 Currently, indocyanine green (ICG) fluorescence lymphography and isosulfan blue dye are used intraoperatively for this purpose10,17 (Fig. 2). ICG fluorescence lymphangiography provides real-time mapping of subdermal lymphatics and can be used in the operating room; however, this technique has several disadvantages for the preoperative imaging of peripheral lymphedema. These include an inherent small imaging field-of-view,10,17 limited anatomic coverage,10,17 lack of spatial information, limited penetration depth,10,17 few anatomic landmarks to perform preoperative localization, and the inability to characterize the subcutaneous soft tissues. Thus, a noninvasive imaging technique is needed to preoperatively image an entire extremity, assess the soft tissues of the limb, determine whether suitable lymphatic channels are present, and define the location of the Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 1: a: Normal lymphatic drainage. The lymphatic system begins as blind-ending sinuses in the interstitial space that coalesce to form larger collecting channels. The collected lymph fluid then travels from the sinuses into deeper collecting lymphatic channels that drain into lymph nodes. Eventually the lymph fluid is propelled through the lymphatics to the subclavian vein and returned to the circulatory system. b: Lymphatic obstruction can be bypassed with a lymphatic-to-venular anastomosis (orange line) that redirects lymphatic flow into an adjacent vein. The venous system returns the redirected lymph flow back to the central circulation.
lymphatics relative to anatomic landmarks that the surgeon can use in the operating room.
MR LYMPHANGIOGRAPHY MR lymphangiography (MRL) is a modification of 3D volumetric contrast-enhanced MR angiography that provides a noninvasive technique for imaging the superficial lymphatics in patients with peripheral lymphedema. MRL exams offer the anatomic coverage to image an entire extremity with a high-resolution 3D dataset, providing sufficient temporal and spatial resolution to depict individual lymphatic channels for preoperative planning of LVA microsurgery.16,18 MRL is also capable of depicting dermal backflow, which December 2015
reflects proximal lymphatic obstruction. This is important for LVA planning since dermal backflow identifies areas of high intralymphatic pressure and excess lymphatic fluid.19 Compared to nuclear medicine lymphoscintigraphy, MRL is able to depict individual lymphatic channels as part of 3D datasets, does not involve ionizing radiation, and has shorter exam times.15,16 In addition, other MR exam components, such as 3D heavily T2-weighted and fat-suppressed sequences, can be included to evaluate the subcutaneous soft tissues to delineate the presence, severity, and extent of lymphedema, as well as associated soft-tissue changes such as adipose deposition and fibrosis.20 The primary difference between MRL and MR angiography is that the contrast for MR lymphangiography is injected intracutaneously. For MRL, a small volume (1.0 mL) of a water-soluble extracellular Gd-based MR contrast agent is injected intracutaneously in the interdigital web spaces of the hand or foot. The low molecular weight of the extracellular MR contrast agent used allows the contrast to be taken up by the lymphatic circulation through interendothelial openings and by vesicular transport.1,21 The safety of intracutaneous administration of extracellular Gdbased MR contrast agents was initially based on animal experiments that showed minimal tissue damage after extravasation 22 and no visible adverse effects after injection in dogs.23 Since the first few clinical reports,24,25 a number of studies have been performed that support the safe off-label use of the intracutaneous administration of extracellular Gdbased MR contrast for MRL. Different studies have reported no significant adverse effects following intracutaneous injection of gadopentate dimeglumine (Gd-DTPA, Magnevist),16,26 gadoterate meglumine (Gd-DOTA, Dotarem),24 gadoteridol (Gd-HPDO3A, Prohance),18,27,28 gadodiamide (Gd-DTPA-BMA, Omniscan),25,29 and gadobenate dimeglumine (Gd-BOPTA, Multihance).15,20,30 Four patient studies have included an evaluation of the safety of the intracutaneous injection of the contrast agent in their methods.21,24,25,29 In all studies, the exam was described as being well tolerated by all patients. There were no reported complications, although some patients in these studies described mild to moderate pain during the injection, and others noted transient swelling at the injection site. To reduce the pain of the injection, small-gauge needles are used, and a local anesthetic can be mixed with the contrast agent before the intracutaneous injection.21
MR Imaging An MRL exam consists of two main components: a 3D heavily T2-weighted sequence to depict the severity and distribution of lymphedema, and a high-resolution fat-suppressed 3D spoiled gradient echo (3D SPGR) sequence after the intracutaneous injection of Gd-based MR contrast to image lymphatic channels. It is important to acquire near1467
Journal of Magnetic Resonance Imaging
FIGURE 2: Procedural steps for microsurgical LVA. a: After ICG is injected intracutaneously, a fluorescence camera is used to image the dye taken up the subdermal lymphatics. In this case, two lymphatic channels are seen (arrows) coursing proximally from the injection sites in the hand. b: ICG fluorescence lymphography is used intraoperatively to trace the location of the subdermal lymphatics on the skin surface. c: A cutdown is then performed at selected locations along the path of a lymphatic channel. d: Isosulfan blue dye injected intracutaneously near the cutdown is taken up by lymphatics and allows visual identification of the underlying lymphatic channel. e: A lymphatic channel and suitable sized adjacent subcutaneous venule are ligated and prepared for an end-to-end anastomosis. f: Completed LVA. The lymphatic side of the LVA still contains the blue dye.
isotropic volumetric source datasets to enable the use of image postprocessing algorithms that generate reconstructed and reformatted images that can be viewed interactively dur-
ing exam interpretation. Image postprocessing also allows the creation of composite images for the surgeon to facilitate preoperative planning.16,31
TABLE 1. Typical MRL Protocol Sequence Parameters Used at Our Institution
3D T2w TSE
Single echo 3D T1w GRE (1.5T)
Dual echo 3D T1w GRE (3.0T)
Sequence
3D Multi-Shot TSE
3D T1-TFE
3D T1-mFFE
Orientation
Sagittal
Sagittal
Sagittal
Partial Fourier factor
0.8
0.85 x 0.675
0.85 x 0.85
Fat suppression
SPIR
SPIR 3
Dual-echo Dixon 3
360 x 221 x 147 mm3
Field-of-view
380 x 312 x 150 mm
485 x 162 x 100 mm
Voxel size
1.7 x 1.7 x 3.0mm3
1.3 x 1.3 x 1.0 mm3
1.2 x 1.2 x 1.6 mm3
TR
2500 ms
7.2 ms
6.2 ms
TE
350 ms
3.3 ms
1.5 ms / 2.8 ms
ETL
90
Flip angle
90o
30o
20o
SENSE factor
2 in AP direction
Scan time
3:47 per station
1:27 per dynamic
1:25 per dynamic
These sequence parameters are used with a Philips Achieva 1.5T and Ingenia 3T and can form the starting point for performing the exam on other vendor platforms. 1468
Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 3: A 3D heavily T2-weighted sequence is performed to assess the presence, extent, and severity of lymphedema. Images of the left arm of a 60-year-old female after left mastectomy who presents with chronic left upper extremity lymphedema. Fullvolume MIP image (a) shows the degree, extent, and location of peripheral lymphedema, which is most severe in the upper midarm. Transverse multiplanar reformats demonstrate the subcutaneous distribution of lymphedema (high T2 signal) in the upper (b) and lower (c) arm (corresponding to white reference lines in (a)). There is moderate lymphedema in the upper arm, primarily medial in distribution (b), whereas in the lower arm the lymphedema is mild and posteriorly located (c).
Since intracutaneously administered contrast is simultaneously absorbed by the venous circulation,5,16,21 at our institution we conclude the exam with an intravenous injection of Gd-based MR contrast to obtain a MR venogram 32 by repeating the high-resolution 3D SPGR sequence. We use the images from the MR venogram to facilitate the differentiation of superficial veins from enhancing lymphatic channels during exam interpretation. MRL can be performed at 1.5T27,29 or 3.0T.30,33 Patient positioning, coil placement, and scan orientation will depend on whether the exam is of an upper or lower extremity, as well as whether unilateral or bilateral imaging is performed. For the upper extremity, which is typically performed as a unilateral two-station exam, the patient is placed supine and head-first into the gantry. The arm of interest is at the patient’s side and surface coils are positioned to image the arm from the mid-hand to the shoulder. Optimally, a large z-axis phased array coil system December 2015
with automatic coil element selection is available for the full coverage. If not, the largest available phased array coil is used, and depending on arm length, some of the upper arm/shoulder may not be included in the imaging volume. Head-first patient positioning allows access to the patient’s hand for the intracutaneous contrast injection performed midway through the exam. For the lower extremities, MRL is performed either unilateral or bilateral depending on the history and clinical request. Patients are placed supine and feet-first on the scanner table, which permits access to the feet from the far side of the gantry for the intracutaneous contrast injection. Surface coils are positioned to cover the lower extremities from the mid-foot to the groin. As with upper extremity MRL, a large z-axis phased array coil system with automatic coil element selection, if available, is preferred, in which case the study is performed as a three-station exam. If not, the largest available phased array coil is used, and a two-station 1469
Journal of Magnetic Resonance Imaging
FIGURE 4: MRL appearance of enhancing lymphatic channels in a lymphedematous limb. Full volume coronal MIP reconstruction of a 74-year-old male with left leg swelling after femoral endarterectomy. On this unilateral lower extremity MRL, lymphatic channels appear irregular and beaded, vary in caliber, and are twisted and discontinuous (arrows). Subdermal veins are smooth and have a uniform caliber (arrowheads).
exam is performed.29 In this instance, the surface coil is first positioned to cover the lower legs from the mid-foot to the knees, and then repositioned to cover the thighs from the knees to the groin. Inhomogeneous fat suppression can result in areas of high signal that mask lymphatics and dermal backflow.16 Thus, for unilateral exams, the patient is offset laterally in order to place the extremity of interest as close to magnet isocenter as possible to improve shimming and the uniformity of fat suppression. The primary sequence components of our exam are: 1) a heavily T2-weighted 3D TSE with spectral fat suppression (SPIR), to define the severity and extent of edema, and 2) a dynamic fat-suppressed T1-weighted 3D SPGR (single echo 3D T1w GRE with spectral fat suppression or dual echo 3D T1w GRE with Dixon reconstruction) before and after the intracutaneous contrast injection to visualize enhancing lymphatic channels. Typical sequence parameters are presented in Table 1. Currently, we perform a minimum of seven dynamic phase 1470
acquisitions at 5-minute intervals (0–5–10–15–20–25–30 min).30 For the lower extremity exam using a single station coil, the distalmost station is imaged at the first five timepoints, after which the coil is moved to the upper leg for the last two timepoints. With larger coil arrays or in the upper extremity, the distalmost station of the lower extremity is imaged at the first five timepoints, with all of the stations imaged for the latter two timepoints. After these seven dynamic phases are acquired, the images are reviewed and additional phases are obtained as needed to completely image progressive lymphatic enhancement. A delayed postcontrast coronal fatsuppressed T1-weighted 3D SPGR is then performed at 120 seconds (upper extremities) to 180 seconds (lower extremities) after single dose intravenous injection of the same Gd-based MR contrast agent. This MR venogram is used to help differentiate lymphatics from veins. The 3D scans of unilateral upper or unilateral lower extremities are typically performed in the sagittal rather than coronal orientation for more efficient scanning and to avoid extraneous tissue outside the extremity of interest during the 3D processing. 3D scans of the bilateral lower extremities are performed in the coronal orientation for the larger scan coverage to include both lower extremities in the same scan field-of-view. Our average exam time for both upper and lower extremity MRL is 1.5–2.0 hours, similar to what has been reported in the literature.30 The same contrast injection protocol is used regardless of field strength. The intracutaneous injection of MR contrast consists of a mixture of 10 mL contrast (gadobenate dimeglumine, MultiHance, Bracco Diagnostics, Princeton, NJ), with 1 mL 1% lidocaine, and 1 mL sodium bicarbonate.30 The contrast mixture is drawn into a 5 or 10 cc syringe depending on whether a unilateral or bilateral exam is being performed. After the skin of the dorsal hand or foot is sterilely prepped, 1 mL of the contrast mixture is injected intracutaneously into each of the four interdigital web spaces of the hand or foot with a 26G needle. Subsequently, the injection sites are massaged for 60 seconds to facilitate lymphatic uptake.1,29 The intravenous contrast administration for the venogram can be performed manually or with a power injector, and consists of 0.1 mmol/kg of contrast (gadobenate dimeglumine) injected at a rate of 1.0 mL/sec followed by a 20-mL saline flush injected at the same rate. We selected gadobenate dimeglumine as the contrast agent for these exams because of its higher relaxivity, potential for protein binding, and thermal stability.34,35 Although its higher osmolality than other agents could in theory result in a greater degree of tissue injury,22,36 no complications have been described in previous studies that have used gadobenate dimeglumine for intracutaneous MRL injections.20,30 Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 5: Multiphase dynamic MRL and the time dependence of lymphatic enhancement. Images of the left arm of a 58-year-old female 10 years after left mastectomy with left upper extremity lymphedema being evaluated for LVA. Multiphase MRL full volume MIP image reconstructions prior to contrast administration, as well as 5 minutes, 10 minutes, 20 minutes, 25 minutes, and 30 minutes after intracutaneous contrast injection in the left hand. This series of MIP reconstructions shows progressive lymphatic enhancement with the greatest degree of enhancement at 30 minutes. Irregular radiating lymphatic channels are seen extending from the lateral forearm towards the medial left elbow in a characteristic irregular, beaded pattern (arrowheads). Note the small area of imperfect fat suppression causing an artifact in the lateral elbow labeled on the precontrast scan (*).
EXAM INTERPRETATION AND 3D IMAGE POSTPROCESSING The overall objectives of MRL are two-fold: the first is to define the severity and extent of lymphedema, and the second to depict the presence, number, course, and location of enhancing lymphatic channels. We interactively review the source images on an independent 3D workstation to allow the real-time creation and viewing of 3D postprocessed images. Multiplanar reformations (MPR), maximum intenDecember 2015
sity projection (MIP) reconstructions, volume rendering (VR), and the 3D cursor are used interactively to characterize and localize a particular structure on different image reconstructions. Screen capturing composite postprocessed images also provides images that summarize the exam findings and depict the spatial and depth information needed by the surgeon for preoperative planning.31 To evaluate the presence and severity of peripheral lymphedema, multiplanar reformats of the heavily T21471
Journal of Magnetic Resonance Imaging
FIGURE 6: Use of 3D image postprocessing to determine the location and depth of individual lymphatic channels. MRL of the left upper extremity of a 48-year-old woman with lymphedema after left mastectomy and axillary lymph node dissection. a: Full volume MIP of the left arm. There is a dilated irregular enhancing lymphatic channel extending cranially from the hand injection site. The lymphatic channel wraps around the arm, beginning anteriorly, then coursing laterally to posteriorly along the arm. b: Axial multiplanar reformation at a proximal level of the forearm shows the posterior location of the lymphatic channel that is 8 mm deep to the skin at the 5 o’clock location. c: Axial MPR at the distal forearm shows the anterior-lateral location of the lymphatic channel that is 2 mm deep to the skin at the 2 o’clock location.
weighted 3D volumetric datasets are reviewed together with a full-volume MIP reconstruction of the scanned extremity (Fig. 3).20 We qualitatively describe the severity of edema as none, mild, moderate, or severe; and describe the location and extent of edema. We next examine the dynamic contrast enhanced 3D MRL, where lymphatic channels are seen as irregular, discontinuous, or twisted enhancing channels that progressively enhance and propagate proximally with time on the dynamic scans (Fig. 4).29,30 Full-volume MIPs are created for each dynamic phase to select the phase with the greatest degree of lymphatic channel enhancement (Fig. 5). Subsequent image review is then performed on the phase with the greatest degree of lymphatic enhancement. We have not found subtraction techniques32 useful, being hindered by patient movements that frequently occur during the pauses between the dynamic phases of the exam, creating misregistration artifacts that severely distort or obscure the small subdermal lymphatics. Interactive review of coronal, sagittal, and axial reformatted images with a 3D cursor allows localization and measurement of the depth of any 1472
lymphatic channel that can then be mapped on the full volume MIP reconstruction (Fig. 6). Different patterns of lymphatic drainage have been described 30 and are demonstrated in Fig. 7. Following intracutaneous injection of MR contrast, venous opacification is universally present.16,21 Thus, it is important to be able to distinguish between the enhancing subdermal lymphatics and subdermal veins. Features that help distinguish between the two relate in part to morphology. Subdermal veins are typically smooth, with uniform caliber (Fig. 4), and a smooth straight longitudinal course. In contrast, lymphatic channels vary in caliber, appear irregular and beaded (Fig. 4), and are often twisted and discontinuous.16,25,30 In addition to the irregular beaded appearance, another feature that can help differentiate subdermal lymphatics from adjacent veins relates to enhancement kinetics on dynamic MRL. On a dynamic MRL series, lymphatic enhancement generally increases and slowly progresses proximally with time, while venous enhancement (Fig. 8) decreases over time.29 Despite these features, we Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 7: Different MRL patterns of lymphatic drainage in lymphedematous limbs. a: Coronal MIP of a left arm—numerous reticular skin lymphatics and areas of dermal backflow (arrows). Incomplete fat suppression appears as scalloped regions of homogeneous high signal (*) that can obscure lymphatics and dermal backflow. b: Coronal oblique MIP of a left leg—radiating arranged or clustered, enhancing lymphatic channels that assemble at the medial knee and continue proximally into the thigh. c: Sagittal MIP of a left arm—discontinuous, minimally enhancing lymphatic channels (arrows). d: Coronal MIP of a left leg—dilated lymphatic channels with few branches.
found that it can still be difficult to determine whether an enhancing structure is a lymphatic channel or a vein. In these cases, directly comparing an image from the MRL with the same image from the delayed MR venogram can characterize the structure of interest as a lymphatic channel based on its presence or absence on the venogram (Fig. 9). It is important to note that enhanced lymphatic vessels may not be seen in normal limbs with MRL, which is believed to be due to faster lymphatic transport in a healthy limb.30 In addition to lymphatic channels, the presence, location, and size of areas of progressive skin enhancement reflecting dermal backflow should be described in the MRL report (Fig. 10).16,21 After interactive exam review, we include the following information in the final MRL report: 1) the presence, severity, location, and distribution of subcutaneous edema; 2) presence and pattern of abnormal lymphatic drainage; 3) size, number, course, and depth from the skin of enhanced lymphatic channels; and 4) the presence, size, and location of any areas of dermal backflow (Table 2). Referencing to a clock face has been helpful for describing the location of lymphatic channels and areas of dermal backflow. Depending on surgeon preference, measurements of limb circumference can also be included in the report to compare diseased versus unaffected limb sizes, and provide a quantitative metric to evaluate response to treatment. December 2015
FIGURE 8: Differentiating subdermal lymphatics from subdermal veins. Coronal MIP of a lymphedematous right arm at 5 minutes (a) and 30 minutes (b) after intracutaneous contrast administration showing evolution of enhancement of the subdermal lymphatic channels and veins. On dynamic MRL, the beaded and irregular lymphatic enhancement increases and progresses proximally with time (arrows) while venous enhancement (arrowheads) decreases.
1473
Journal of Magnetic Resonance Imaging
FIGURE 9: Differentiating subdermal lymphatics from veins. Viewing the same image projection on the MRL and MRV can confirm if a structure of interest represents an enhancing lymphatic or vein. Oblique MIP MRL of a right calf (a) shows several enhancing branching structures about the lateral malleous (arrows, arrowhead) that are difficult to classify as veins or lymphatics. On the MR venogram (b), the structures with arrows on the MRL are veins, as they enlarge and communicate with other veins on the MRV, whereas the structure with an arrowhead on the MRL is a lymphatic channel that does not change or communicate with veins on the MRV.
FUTURE DIRECTIONS The spatial and temporal resolution of MRL allow the depiction of individual subdermal lymphatic channels, thereby permitting the creation of reformatted images that facilitate preoperative selection of patients for microsurgical LVA treatment.16,18 MRL can also be used for postoperative assessment by demonstrating reductions in lymphedema, limb size, enhancing lymphatic channels, and dermal backflow.18 While the spatial resolution of early studies was not sufficient to visualize the postoperative anastomosis,18 small patient series at 3.0T suggest that MRL can assess the patency of surgically modified lymphatic channels.16,28 One of MR’s strengths is its versatility in the availability of multiple different contrast agents and pulse sequence techniques that can be employed to modify tissue contrast. Currently, there are nine gadolinium-based MR contrast agents that are clinically available for human use.37 While MRL has been primarily performed with extracellular fluid 1474
(ECF) agents, several of the contrast agents have particular properties that could in theory improve MRL. Gadobutrol is an ECF agent formulated with twice the molarity (1.0 mol/L) of most other agents.37 The higher concentration would allow for smaller intracutaneous injection volumes, conceivably improving patient comfort.25 Furthermore, since the optimal particle size for colloidbound radiotracers used for nuclear medicine lymphoscintigraphy is believed to be 50–70 nm,1 gadolinium chelates that associate with macroproteins (eg, gadobenate dimeglumine, gadoxetate, gadofosveset trisodium37) may in principle increase lymphatic and lymph node uptake of MR contrast, and also improve venous differentiation. A recent animal study investigated this property of gadofosveset for MRL at 3T by premixing gadofosveset with human serum albumin (physical size 7 nm) before intradermal injection in nude mice and monkeys. The authors reported that compared to intradermal injections of gadofosveset and gadopentetate (both with physical size
MR Lymphangiography: How I Do It Lee M. Mitsumori, MD,1* Elizabeth S. McDonald, MD, PhD,2 Gregory J. Wilson, PhD,3 Peter C. Neligan, MD,4 Satoshi Minoshima, MD, PhD,3 and Jeffrey H. Maki, MD, PhD3 This article is accredited as a journal-based CME activity. If you wish to receive credit for this activity, please refer to the website: www. wileyhealthlearning.com/jmri
ACCREDITATION AND DESIGNATION STATEMENT Blackwell Futura Media Services designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM. Physicians should only claim credit commensurate with the extent of their participation in the activity. Blackwell Futura Media Services is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
EDUCATIONAL OBJECTIVES Upon completion of this educational activity, participants will be better able to: 1. Interpret an MR lymphangiography exam. 2. Recognize the pathophysiology and microsurgical treatment of lymphedema.
ACTIVITY DISCLOSURES No commercial support has been accepted related to the development or publication of this activity.
Faculty Disclosures: Editor-in-Chief: Mark E. Schweitzer, MD, has no relevant financial relationships to disclose. CME Editor: Scott B. Reeder, MD, PhD, discloses personal stock in Cellectar Biosciences and Neuwave Medical. CME Committee: Shreyas Vasanawala, MD, PhD, discloses research support from General Electric, and founder’s equity in Morpheus Medical. Scott K. Nagle, MD, PhD, discloses consulting fees from Vertex Pharmaceuticals for consulting in design of cystic fibrosis clinical trials involving imaging; and departmental research support from General Electric for evaluation of products and development. Mustafa R. Bashir, MD, discloses research support from Siemens Healthcare and Bayer Healthcare.
Tim Leiner, MD, PhD, discloses research support grant funding from Bracco, S.p.A., Philips Healthcare, and Bayer Healthcare. Bonnie Joe, MD, PhD, has no relevant financial relationships to disclose. Authors: Lee M. Mitsumori, MD; Elizabeth S. McDonald, MD, PhD; Greg Wilson, PhD; Peter C. Neligan, MD; and Satoshi Minoshima, MD, PhD have no relevant financial relationships to disclose. Jeffrey H. Maki, MD, PhD discloses grant support from Bracco Diagnostics for unrelated research on MR contrast agents. This manuscript underwent peer review in line with the standards of editorial integrity and publication ethics maintained by Journal of Magnetic Resonance Imaging. The peer reviewers have no relevant financial relationships. The peer review process for Journal of Magnetic Resonance Imaging is double-blinded. As such, the identities of the reviewers are not disclosed in line with the standard accepted practices of medical journal peer review. Conflicts of interest have been identified and resolved in accordance with Blackwell Futura Media Services’ Policy on Activity Disclosure and Conflict of Interest.
INSTRUCTIONS ON RECEIVING CREDIT For information on applicability and acceptance of CME credit for this activity, please consult your professional licensing board. This activity is designed to be completed within an hour; physicians should claim only those credits that reflect the time actually spent in the activity. To successfully earn credit, participants must complete the activity during the valid credit period. Follow these steps to earn credit: Log on to www.wileyhealthlearning.com Read the target audience, educational objectives, and activity disclosures. Read the article in print or online format. Reflect on the article. Access the CME Exam, and choose the best answer to each question. Complete the required evaluation component of the activity. This activity will be available for CME credit for twelve months following its publication date. At that time, it will be reviewed and potentially updated and extended for an additional period.
View this article online at wileyonlinelibrary.com. DOI: 10.1002/jmri.24887 Received Apr 15, 2014, Accepted for publication Nov 21, 2014. *Address reprint requests to L.M.M., Department of Radiology, Straub Clinic and Hospital, 888 South King St., Honolulu, HI 96813. E-mail: [email protected] From the 1Department of Radiology, Straub Clinic and Hospital, Honolulu, Hawaii, USA; 2Department of Radiology, University of Pennsylvania, Philadelphia, Pennsylvania, USA; 3Department of Radiology, University of Washington, Seattle, Washington, USA; and 4Department of Plastic Surgery, University of Washington, Seattle, Washington, USA
C 2015 Wiley Periodicals, Inc. V 1465
Journal of Magnetic Resonance Imaging
Lymphedema is a chronic progressive edematous disease that in the United States is most commonly related to malignancy and its treatment. Lymphaticovenular anastomosis is a recently introduced microsurgical treatment option for lymphedema that requires the identification and mapping of individual lymphatic channels. While nuclear medicine lymphoscintigraphy has been the primary imaging modality performed to evaluate suspected lymphedema, lymphoscintigraphy does not provide the spatial information necessary for presurgical planning. High-resolution dynamic 3D magnetic resonance imaging (MRI) can noninvasively image abnormal lymphatic channels to both diagnose lymphedema and depict the location and number of individual lymphatic channels for surgical planning. MR lymphangiography can be performed at 1.5T or 3.0T using multichannel phased array surface coils. The main components of the exam are a heavily T2-weighted 3D sequence to define the severity and extent of edema, a high-resolution dynamic 3D gradient echo imaging after intracutaneous contrast injection to visualize lymphatic channels, and a delayed 3D gradient echo sequence after intravenous contrast to define veins. This article reviews the pathophysiology and microsurgical treatment of lymphedema, presents the imaging protocol used at our institution, and describes exam interpretation and the image postprocessing performed for surgical planning. J. MAGN. RESON. IMAGING 2015;42:1465–1477.
LYMPHATIC CIRCULATION AND LYMPHEDEMA
T
he normal lymphatic system provides drainage of macromolecules, metabolic waste products, and excess interstitial fluid that has not been resorbed by the venous circulation. The lymphatic system begins as blind-ending sinuses that progressively coalesce to form larger collecting lymphatic channels.1 Ultimately, the fluid within the peripheral lymphatic system is propelled to the thoracic duct where the lymph fluid is returned to the circulatory system at the left subclavian vein (Fig. 1a).2 Peripheral lymphedema is a chronic, progressive, edematous disease caused by abnormal lymphatic drainage. The disease is characterized by the accumulation of proteinrich fluid, inflammation, adipose tissue hypertrophy, and fibrosis in the interstitial space. More than 90 million people are affected by lymphedema worldwide, primarily the result of filarial infections (Wuscheria banrofti). In the U.S., peripheral lymphedema affects nearly 10 million people, where it is most commonly related to the treatment of breast and gynecological malignancy.1 Upper extremity lymphedema rates after surgical treatment for breast cancer vary between 2% and 65%, depending on the type of procedure performed, with the highest rates occurring following lymph node dissection or nodal radiation therapy.3 As with breast cancer, lower extremity lymphedema rates vary depending on the surgical treatment, occurring in 25–67% of patients after inguinal-femoral lymph node dissection and 2–41% after pelvic lymph node dissection.4 There are conservative and operative treatment options for peripheral lymphedema. Unfortunately, the efficacy of conservative treatment is highly variable between patients and is not curative.5 Operative approaches include debulking, lymph node transplantation, and microsurgical procedures.4,6 Since debulking procedures can in themselves worsen the severity of the lymphedema, and the efficacy of lymph node transplantation has not yet been established7,8; microsurgical treatment of peripheral lymphedema is increasingly being performed at different centers around the world.7,9–11 1466
MICROSURGICAL TREATMENT Since the lymphatic and venous systems flow in parallel to return fluid to the heart, the creation of a lymphaticovenular anastomosis (LVA) bypasses a lymphatic obstruction by shunting distal lymph flow into adjacent veins (Fig. 1b). Because of the small size (0.2–0.8 mm) of the subdermal lymphatic channels used for the procedure, microsurgical techniques are required.12,13 Using a surgical microscope, the surgeon localizes an individual subdermal lymphatic channel as well as an adjacent subdermal venule. Once a suitable lymphatic channel and venule are identified, both vessels are divided and a LVA is created (Fig. 2). While LVA can be performed at the time of the lymph node dissection,14 this strategy is currently not widely performed since not all patients develop lymphedema after lymph node dissection, and many oncologic surgeons lack the microsurgical expertise to perform the procedure. The main challenge with LVA procedures lies in finding the small subdermal lymphatic channels and venules needed for performing the anastomoses.10 While radionuclide lymphoscintigraphy has been considered the primary clinical imaging modality to diagnose lymphedema,1,4 its limited temporal and spatial resolution does not allow the identification and localization of individual lymphatic channels.15,16 Currently, indocyanine green (ICG) fluorescence lymphography and isosulfan blue dye are used intraoperatively for this purpose10,17 (Fig. 2). ICG fluorescence lymphangiography provides real-time mapping of subdermal lymphatics and can be used in the operating room; however, this technique has several disadvantages for the preoperative imaging of peripheral lymphedema. These include an inherent small imaging field-of-view,10,17 limited anatomic coverage,10,17 lack of spatial information, limited penetration depth,10,17 few anatomic landmarks to perform preoperative localization, and the inability to characterize the subcutaneous soft tissues. Thus, a noninvasive imaging technique is needed to preoperatively image an entire extremity, assess the soft tissues of the limb, determine whether suitable lymphatic channels are present, and define the location of the Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 1: a: Normal lymphatic drainage. The lymphatic system begins as blind-ending sinuses in the interstitial space that coalesce to form larger collecting channels. The collected lymph fluid then travels from the sinuses into deeper collecting lymphatic channels that drain into lymph nodes. Eventually the lymph fluid is propelled through the lymphatics to the subclavian vein and returned to the circulatory system. b: Lymphatic obstruction can be bypassed with a lymphatic-to-venular anastomosis (orange line) that redirects lymphatic flow into an adjacent vein. The venous system returns the redirected lymph flow back to the central circulation.
lymphatics relative to anatomic landmarks that the surgeon can use in the operating room.
MR LYMPHANGIOGRAPHY MR lymphangiography (MRL) is a modification of 3D volumetric contrast-enhanced MR angiography that provides a noninvasive technique for imaging the superficial lymphatics in patients with peripheral lymphedema. MRL exams offer the anatomic coverage to image an entire extremity with a high-resolution 3D dataset, providing sufficient temporal and spatial resolution to depict individual lymphatic channels for preoperative planning of LVA microsurgery.16,18 MRL is also capable of depicting dermal backflow, which December 2015
reflects proximal lymphatic obstruction. This is important for LVA planning since dermal backflow identifies areas of high intralymphatic pressure and excess lymphatic fluid.19 Compared to nuclear medicine lymphoscintigraphy, MRL is able to depict individual lymphatic channels as part of 3D datasets, does not involve ionizing radiation, and has shorter exam times.15,16 In addition, other MR exam components, such as 3D heavily T2-weighted and fat-suppressed sequences, can be included to evaluate the subcutaneous soft tissues to delineate the presence, severity, and extent of lymphedema, as well as associated soft-tissue changes such as adipose deposition and fibrosis.20 The primary difference between MRL and MR angiography is that the contrast for MR lymphangiography is injected intracutaneously. For MRL, a small volume (1.0 mL) of a water-soluble extracellular Gd-based MR contrast agent is injected intracutaneously in the interdigital web spaces of the hand or foot. The low molecular weight of the extracellular MR contrast agent used allows the contrast to be taken up by the lymphatic circulation through interendothelial openings and by vesicular transport.1,21 The safety of intracutaneous administration of extracellular Gdbased MR contrast agents was initially based on animal experiments that showed minimal tissue damage after extravasation 22 and no visible adverse effects after injection in dogs.23 Since the first few clinical reports,24,25 a number of studies have been performed that support the safe off-label use of the intracutaneous administration of extracellular Gdbased MR contrast for MRL. Different studies have reported no significant adverse effects following intracutaneous injection of gadopentate dimeglumine (Gd-DTPA, Magnevist),16,26 gadoterate meglumine (Gd-DOTA, Dotarem),24 gadoteridol (Gd-HPDO3A, Prohance),18,27,28 gadodiamide (Gd-DTPA-BMA, Omniscan),25,29 and gadobenate dimeglumine (Gd-BOPTA, Multihance).15,20,30 Four patient studies have included an evaluation of the safety of the intracutaneous injection of the contrast agent in their methods.21,24,25,29 In all studies, the exam was described as being well tolerated by all patients. There were no reported complications, although some patients in these studies described mild to moderate pain during the injection, and others noted transient swelling at the injection site. To reduce the pain of the injection, small-gauge needles are used, and a local anesthetic can be mixed with the contrast agent before the intracutaneous injection.21
MR Imaging An MRL exam consists of two main components: a 3D heavily T2-weighted sequence to depict the severity and distribution of lymphedema, and a high-resolution fat-suppressed 3D spoiled gradient echo (3D SPGR) sequence after the intracutaneous injection of Gd-based MR contrast to image lymphatic channels. It is important to acquire near1467
Journal of Magnetic Resonance Imaging
FIGURE 2: Procedural steps for microsurgical LVA. a: After ICG is injected intracutaneously, a fluorescence camera is used to image the dye taken up the subdermal lymphatics. In this case, two lymphatic channels are seen (arrows) coursing proximally from the injection sites in the hand. b: ICG fluorescence lymphography is used intraoperatively to trace the location of the subdermal lymphatics on the skin surface. c: A cutdown is then performed at selected locations along the path of a lymphatic channel. d: Isosulfan blue dye injected intracutaneously near the cutdown is taken up by lymphatics and allows visual identification of the underlying lymphatic channel. e: A lymphatic channel and suitable sized adjacent subcutaneous venule are ligated and prepared for an end-to-end anastomosis. f: Completed LVA. The lymphatic side of the LVA still contains the blue dye.
isotropic volumetric source datasets to enable the use of image postprocessing algorithms that generate reconstructed and reformatted images that can be viewed interactively dur-
ing exam interpretation. Image postprocessing also allows the creation of composite images for the surgeon to facilitate preoperative planning.16,31
TABLE 1. Typical MRL Protocol Sequence Parameters Used at Our Institution
3D T2w TSE
Single echo 3D T1w GRE (1.5T)
Dual echo 3D T1w GRE (3.0T)
Sequence
3D Multi-Shot TSE
3D T1-TFE
3D T1-mFFE
Orientation
Sagittal
Sagittal
Sagittal
Partial Fourier factor
0.8
0.85 x 0.675
0.85 x 0.85
Fat suppression
SPIR
SPIR 3
Dual-echo Dixon 3
360 x 221 x 147 mm3
Field-of-view
380 x 312 x 150 mm
485 x 162 x 100 mm
Voxel size
1.7 x 1.7 x 3.0mm3
1.3 x 1.3 x 1.0 mm3
1.2 x 1.2 x 1.6 mm3
TR
2500 ms
7.2 ms
6.2 ms
TE
350 ms
3.3 ms
1.5 ms / 2.8 ms
ETL
90
Flip angle
90o
30o
20o
SENSE factor
2 in AP direction
Scan time
3:47 per station
1:27 per dynamic
1:25 per dynamic
These sequence parameters are used with a Philips Achieva 1.5T and Ingenia 3T and can form the starting point for performing the exam on other vendor platforms. 1468
Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 3: A 3D heavily T2-weighted sequence is performed to assess the presence, extent, and severity of lymphedema. Images of the left arm of a 60-year-old female after left mastectomy who presents with chronic left upper extremity lymphedema. Fullvolume MIP image (a) shows the degree, extent, and location of peripheral lymphedema, which is most severe in the upper midarm. Transverse multiplanar reformats demonstrate the subcutaneous distribution of lymphedema (high T2 signal) in the upper (b) and lower (c) arm (corresponding to white reference lines in (a)). There is moderate lymphedema in the upper arm, primarily medial in distribution (b), whereas in the lower arm the lymphedema is mild and posteriorly located (c).
Since intracutaneously administered contrast is simultaneously absorbed by the venous circulation,5,16,21 at our institution we conclude the exam with an intravenous injection of Gd-based MR contrast to obtain a MR venogram 32 by repeating the high-resolution 3D SPGR sequence. We use the images from the MR venogram to facilitate the differentiation of superficial veins from enhancing lymphatic channels during exam interpretation. MRL can be performed at 1.5T27,29 or 3.0T.30,33 Patient positioning, coil placement, and scan orientation will depend on whether the exam is of an upper or lower extremity, as well as whether unilateral or bilateral imaging is performed. For the upper extremity, which is typically performed as a unilateral two-station exam, the patient is placed supine and head-first into the gantry. The arm of interest is at the patient’s side and surface coils are positioned to image the arm from the mid-hand to the shoulder. Optimally, a large z-axis phased array coil system December 2015
with automatic coil element selection is available for the full coverage. If not, the largest available phased array coil is used, and depending on arm length, some of the upper arm/shoulder may not be included in the imaging volume. Head-first patient positioning allows access to the patient’s hand for the intracutaneous contrast injection performed midway through the exam. For the lower extremities, MRL is performed either unilateral or bilateral depending on the history and clinical request. Patients are placed supine and feet-first on the scanner table, which permits access to the feet from the far side of the gantry for the intracutaneous contrast injection. Surface coils are positioned to cover the lower extremities from the mid-foot to the groin. As with upper extremity MRL, a large z-axis phased array coil system with automatic coil element selection, if available, is preferred, in which case the study is performed as a three-station exam. If not, the largest available phased array coil is used, and a two-station 1469
Journal of Magnetic Resonance Imaging
FIGURE 4: MRL appearance of enhancing lymphatic channels in a lymphedematous limb. Full volume coronal MIP reconstruction of a 74-year-old male with left leg swelling after femoral endarterectomy. On this unilateral lower extremity MRL, lymphatic channels appear irregular and beaded, vary in caliber, and are twisted and discontinuous (arrows). Subdermal veins are smooth and have a uniform caliber (arrowheads).
exam is performed.29 In this instance, the surface coil is first positioned to cover the lower legs from the mid-foot to the knees, and then repositioned to cover the thighs from the knees to the groin. Inhomogeneous fat suppression can result in areas of high signal that mask lymphatics and dermal backflow.16 Thus, for unilateral exams, the patient is offset laterally in order to place the extremity of interest as close to magnet isocenter as possible to improve shimming and the uniformity of fat suppression. The primary sequence components of our exam are: 1) a heavily T2-weighted 3D TSE with spectral fat suppression (SPIR), to define the severity and extent of edema, and 2) a dynamic fat-suppressed T1-weighted 3D SPGR (single echo 3D T1w GRE with spectral fat suppression or dual echo 3D T1w GRE with Dixon reconstruction) before and after the intracutaneous contrast injection to visualize enhancing lymphatic channels. Typical sequence parameters are presented in Table 1. Currently, we perform a minimum of seven dynamic phase 1470
acquisitions at 5-minute intervals (0–5–10–15–20–25–30 min).30 For the lower extremity exam using a single station coil, the distalmost station is imaged at the first five timepoints, after which the coil is moved to the upper leg for the last two timepoints. With larger coil arrays or in the upper extremity, the distalmost station of the lower extremity is imaged at the first five timepoints, with all of the stations imaged for the latter two timepoints. After these seven dynamic phases are acquired, the images are reviewed and additional phases are obtained as needed to completely image progressive lymphatic enhancement. A delayed postcontrast coronal fatsuppressed T1-weighted 3D SPGR is then performed at 120 seconds (upper extremities) to 180 seconds (lower extremities) after single dose intravenous injection of the same Gd-based MR contrast agent. This MR venogram is used to help differentiate lymphatics from veins. The 3D scans of unilateral upper or unilateral lower extremities are typically performed in the sagittal rather than coronal orientation for more efficient scanning and to avoid extraneous tissue outside the extremity of interest during the 3D processing. 3D scans of the bilateral lower extremities are performed in the coronal orientation for the larger scan coverage to include both lower extremities in the same scan field-of-view. Our average exam time for both upper and lower extremity MRL is 1.5–2.0 hours, similar to what has been reported in the literature.30 The same contrast injection protocol is used regardless of field strength. The intracutaneous injection of MR contrast consists of a mixture of 10 mL contrast (gadobenate dimeglumine, MultiHance, Bracco Diagnostics, Princeton, NJ), with 1 mL 1% lidocaine, and 1 mL sodium bicarbonate.30 The contrast mixture is drawn into a 5 or 10 cc syringe depending on whether a unilateral or bilateral exam is being performed. After the skin of the dorsal hand or foot is sterilely prepped, 1 mL of the contrast mixture is injected intracutaneously into each of the four interdigital web spaces of the hand or foot with a 26G needle. Subsequently, the injection sites are massaged for 60 seconds to facilitate lymphatic uptake.1,29 The intravenous contrast administration for the venogram can be performed manually or with a power injector, and consists of 0.1 mmol/kg of contrast (gadobenate dimeglumine) injected at a rate of 1.0 mL/sec followed by a 20-mL saline flush injected at the same rate. We selected gadobenate dimeglumine as the contrast agent for these exams because of its higher relaxivity, potential for protein binding, and thermal stability.34,35 Although its higher osmolality than other agents could in theory result in a greater degree of tissue injury,22,36 no complications have been described in previous studies that have used gadobenate dimeglumine for intracutaneous MRL injections.20,30 Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 5: Multiphase dynamic MRL and the time dependence of lymphatic enhancement. Images of the left arm of a 58-year-old female 10 years after left mastectomy with left upper extremity lymphedema being evaluated for LVA. Multiphase MRL full volume MIP image reconstructions prior to contrast administration, as well as 5 minutes, 10 minutes, 20 minutes, 25 minutes, and 30 minutes after intracutaneous contrast injection in the left hand. This series of MIP reconstructions shows progressive lymphatic enhancement with the greatest degree of enhancement at 30 minutes. Irregular radiating lymphatic channels are seen extending from the lateral forearm towards the medial left elbow in a characteristic irregular, beaded pattern (arrowheads). Note the small area of imperfect fat suppression causing an artifact in the lateral elbow labeled on the precontrast scan (*).
EXAM INTERPRETATION AND 3D IMAGE POSTPROCESSING The overall objectives of MRL are two-fold: the first is to define the severity and extent of lymphedema, and the second to depict the presence, number, course, and location of enhancing lymphatic channels. We interactively review the source images on an independent 3D workstation to allow the real-time creation and viewing of 3D postprocessed images. Multiplanar reformations (MPR), maximum intenDecember 2015
sity projection (MIP) reconstructions, volume rendering (VR), and the 3D cursor are used interactively to characterize and localize a particular structure on different image reconstructions. Screen capturing composite postprocessed images also provides images that summarize the exam findings and depict the spatial and depth information needed by the surgeon for preoperative planning.31 To evaluate the presence and severity of peripheral lymphedema, multiplanar reformats of the heavily T21471
Journal of Magnetic Resonance Imaging
FIGURE 6: Use of 3D image postprocessing to determine the location and depth of individual lymphatic channels. MRL of the left upper extremity of a 48-year-old woman with lymphedema after left mastectomy and axillary lymph node dissection. a: Full volume MIP of the left arm. There is a dilated irregular enhancing lymphatic channel extending cranially from the hand injection site. The lymphatic channel wraps around the arm, beginning anteriorly, then coursing laterally to posteriorly along the arm. b: Axial multiplanar reformation at a proximal level of the forearm shows the posterior location of the lymphatic channel that is 8 mm deep to the skin at the 5 o’clock location. c: Axial MPR at the distal forearm shows the anterior-lateral location of the lymphatic channel that is 2 mm deep to the skin at the 2 o’clock location.
weighted 3D volumetric datasets are reviewed together with a full-volume MIP reconstruction of the scanned extremity (Fig. 3).20 We qualitatively describe the severity of edema as none, mild, moderate, or severe; and describe the location and extent of edema. We next examine the dynamic contrast enhanced 3D MRL, where lymphatic channels are seen as irregular, discontinuous, or twisted enhancing channels that progressively enhance and propagate proximally with time on the dynamic scans (Fig. 4).29,30 Full-volume MIPs are created for each dynamic phase to select the phase with the greatest degree of lymphatic channel enhancement (Fig. 5). Subsequent image review is then performed on the phase with the greatest degree of lymphatic enhancement. We have not found subtraction techniques32 useful, being hindered by patient movements that frequently occur during the pauses between the dynamic phases of the exam, creating misregistration artifacts that severely distort or obscure the small subdermal lymphatics. Interactive review of coronal, sagittal, and axial reformatted images with a 3D cursor allows localization and measurement of the depth of any 1472
lymphatic channel that can then be mapped on the full volume MIP reconstruction (Fig. 6). Different patterns of lymphatic drainage have been described 30 and are demonstrated in Fig. 7. Following intracutaneous injection of MR contrast, venous opacification is universally present.16,21 Thus, it is important to be able to distinguish between the enhancing subdermal lymphatics and subdermal veins. Features that help distinguish between the two relate in part to morphology. Subdermal veins are typically smooth, with uniform caliber (Fig. 4), and a smooth straight longitudinal course. In contrast, lymphatic channels vary in caliber, appear irregular and beaded (Fig. 4), and are often twisted and discontinuous.16,25,30 In addition to the irregular beaded appearance, another feature that can help differentiate subdermal lymphatics from adjacent veins relates to enhancement kinetics on dynamic MRL. On a dynamic MRL series, lymphatic enhancement generally increases and slowly progresses proximally with time, while venous enhancement (Fig. 8) decreases over time.29 Despite these features, we Volume 42, No. 6
Mitsumori et al.: MR Lymphangiography
FIGURE 7: Different MRL patterns of lymphatic drainage in lymphedematous limbs. a: Coronal MIP of a left arm—numerous reticular skin lymphatics and areas of dermal backflow (arrows). Incomplete fat suppression appears as scalloped regions of homogeneous high signal (*) that can obscure lymphatics and dermal backflow. b: Coronal oblique MIP of a left leg—radiating arranged or clustered, enhancing lymphatic channels that assemble at the medial knee and continue proximally into the thigh. c: Sagittal MIP of a left arm—discontinuous, minimally enhancing lymphatic channels (arrows). d: Coronal MIP of a left leg—dilated lymphatic channels with few branches.
found that it can still be difficult to determine whether an enhancing structure is a lymphatic channel or a vein. In these cases, directly comparing an image from the MRL with the same image from the delayed MR venogram can characterize the structure of interest as a lymphatic channel based on its presence or absence on the venogram (Fig. 9). It is important to note that enhanced lymphatic vessels may not be seen in normal limbs with MRL, which is believed to be due to faster lymphatic transport in a healthy limb.30 In addition to lymphatic channels, the presence, location, and size of areas of progressive skin enhancement reflecting dermal backflow should be described in the MRL report (Fig. 10).16,21 After interactive exam review, we include the following information in the final MRL report: 1) the presence, severity, location, and distribution of subcutaneous edema; 2) presence and pattern of abnormal lymphatic drainage; 3) size, number, course, and depth from the skin of enhanced lymphatic channels; and 4) the presence, size, and location of any areas of dermal backflow (Table 2). Referencing to a clock face has been helpful for describing the location of lymphatic channels and areas of dermal backflow. Depending on surgeon preference, measurements of limb circumference can also be included in the report to compare diseased versus unaffected limb sizes, and provide a quantitative metric to evaluate response to treatment. December 2015
FIGURE 8: Differentiating subdermal lymphatics from subdermal veins. Coronal MIP of a lymphedematous right arm at 5 minutes (a) and 30 minutes (b) after intracutaneous contrast administration showing evolution of enhancement of the subdermal lymphatic channels and veins. On dynamic MRL, the beaded and irregular lymphatic enhancement increases and progresses proximally with time (arrows) while venous enhancement (arrowheads) decreases.
1473
Journal of Magnetic Resonance Imaging
FIGURE 9: Differentiating subdermal lymphatics from veins. Viewing the same image projection on the MRL and MRV can confirm if a structure of interest represents an enhancing lymphatic or vein. Oblique MIP MRL of a right calf (a) shows several enhancing branching structures about the lateral malleous (arrows, arrowhead) that are difficult to classify as veins or lymphatics. On the MR venogram (b), the structures with arrows on the MRL are veins, as they enlarge and communicate with other veins on the MRV, whereas the structure with an arrowhead on the MRL is a lymphatic channel that does not change or communicate with veins on the MRV.
FUTURE DIRECTIONS The spatial and temporal resolution of MRL allow the depiction of individual subdermal lymphatic channels, thereby permitting the creation of reformatted images that facilitate preoperative selection of patients for microsurgical LVA treatment.16,18 MRL can also be used for postoperative assessment by demonstrating reductions in lymphedema, limb size, enhancing lymphatic channels, and dermal backflow.18 While the spatial resolution of early studies was not sufficient to visualize the postoperative anastomosis,18 small patient series at 3.0T suggest that MRL can assess the patency of surgically modified lymphatic channels.16,28 One of MR’s strengths is its versatility in the availability of multiple different contrast agents and pulse sequence techniques that can be employed to modify tissue contrast. Currently, there are nine gadolinium-based MR contrast agents that are clinically available for human use.37 While MRL has been primarily performed with extracellular fluid 1474
(ECF) agents, several of the contrast agents have particular properties that could in theory improve MRL. Gadobutrol is an ECF agent formulated with twice the molarity (1.0 mol/L) of most other agents.37 The higher concentration would allow for smaller intracutaneous injection volumes, conceivably improving patient comfort.25 Furthermore, since the optimal particle size for colloidbound radiotracers used for nuclear medicine lymphoscintigraphy is believed to be 50–70 nm,1 gadolinium chelates that associate with macroproteins (eg, gadobenate dimeglumine, gadoxetate, gadofosveset trisodium37) may in principle increase lymphatic and lymph node uptake of MR contrast, and also improve venous differentiation. A recent animal study investigated this property of gadofosveset for MRL at 3T by premixing gadofosveset with human serum albumin (physical size 7 nm) before intradermal injection in nude mice and monkeys. The authors reported that compared to intradermal injections of gadofosveset and gadopentetate (both with physical size
