| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Surgery, University of Manitoba, St. Boniface General Hospital,
409 Tache Ave., Winnipeg, Manitoba R2H 2A6, Canada.
2 Department of Human Anatomy and Cell Sciences, University of Manitoba, Basic
Medical Sciences Bldg., Winnipeg, Manitoba R3E 3J7, Canada.
3 Schering AG, Contrast Media Research, Muellerstr. 178, 13342 Berlin,
Germany.
Received February 4, 2002; accepted after revision May 30, 2002.
Supported by Schering and the St. Boniface General Hospital Research
Foundation.
OBJECTIVE. The goals of our study were to determine the technical feasibility of dynamic MR lymphangiography for detecting sentinel lymph nodes using Gadomer-17 contrast material to compare these results to the clinically standardized blue dye contrast agent, and to show MR imaging—guided needle localization marking of the sentinel nodes.
MATERIALS AND METHODS. Six anesthetized swine underwent MR imaging before and for 1 hr after interstitial injection of Gadomer-17 in the posterior tongue and intradermally in the stifle (knee). Using MR imaging guidance, we percutaneously placed a histology needle into the sentinel node or nodes. A standardized intraoperative sentinel node detection procedure was then performed using isosulfan blue contrast agent. Sentinel nodes identified on the MR images were compared with and matched to those identified by biopsy needle localization and the isosulfan blue contrast agent.
RESULTS. Interstitial and intradermal injection of Gadomer-17 resulted in a lasting, increased MR signal in the regional lymph nodes. Seven sentinel nodes were identified in both the neck and the groin, with one pig showing two sentinel nodes in the neck and another showing two sentinel nodes in the groin. In all cases, the isosulfan blue contrast agent agreed with the location of the sentinel node determined by Gadomer-17 and marked with the needle.
CONCLUSION. In this large animal model, dynamic MR angiography of sentinel lymph nodes is technically feasible using a contrast agent (Gadomer-17) and can be easily combined with MR imaging—guided needle localization.
Sentinel lymph node biopsy [1, 2] is becoming an accepted tool for treating regional lymphatics in patients with malignant melanoma and is undergoing evaluation in breast cancer and other tumors [1]. Despite the theoretic advantages of the technique and the encouraging results of nonrandomized trials, the definitive answer about substituting sentinel lymph node biopsy for elective lymph node dissection is not known [1, 3, 4].
The sentinel lymph node is the first lymph node to receive lymphatic drainage from a tumor. Current sentinel lymph node techniques identify the sentinel lymph node or nodes at the time of intraoperative dissection after interstitial injection of radiocolloid that may be combined with interstitial isosulfan blue contrast agent. This one-time evaluation does not distinguish multiple sentinel nodes from second-echelon or other regional lymph nodes [5]. Dynamic lymphoscintigraphy [6, 7] is becoming more commonly applied and provides temporal resolution to address the issue of first-appearance criterion [5]; however, this imaging modality has a limited spatial resolution. None of the current techniques reliably identifies all possible sentinel nodes [1, 3, 8].
MR imaging has proven to be a useful technique for the detection and depiction of lymph nodes in the head and neck and in urologic and pelvic cancers. Published techniques have used IV injection of contrast agents including the gadopentetate dimeglumine polyglucose macro-complex [9, 10] and dextran-coated ultrasmall superparamagnetic iron oxide (Combidex; Advanced Magnetics, Cambridge, MA) [11, 12] or the interstitial injection of contrast agents such as gadopentetate dimeglumine polymers [13], Gadofluorine-8 (Schering, Berlin, Germany) [14], Gadomer-17 (Schering) [15], and super-paramagnetic iron oxide [16,17,18] to determine the metastatic status of lymph nodes. None of these studies attempted to specifically identify the sentinel lymph nodes or to compare the methods with the currently accepted sentinel lymph node techniques. One of us previously showed the use of Combidex for interstitial sentinel node lymphangiography [19].
This study was designed to establish the use of interstitial and intradermal injection of an investigational MR contrast agent, Gadomer-17, for the identification and localization of sentinel lymph nodes in a swine model and to compare Gadomer-17 with with the isosulfan blue contrast agent.
Materials and Methods
Our study was performed in accordance with the Canadian Council on Animal Care guidelines and was approved by the University of Manitoba Protocol Management and Review Committee. Six female swine (40-50 kg) were used in the study.
The pigs were sedated, intubated, and anesthetized using isoflurane and had a lateral or median auricular vein catheter and a urinary catheter placed. A slow saline infusion was maintained for emergency vascular access for sedation of the pigs, with diazepam if required. Complete vital signs were taken every 20 min, and three lead ECG and end tidal CO2 parameters were measured continuously. Pigs were positioned prone in the MR imaging magnet and held in position by limb restraint and foam cushions.
Imaging was performed on a 0.2-T system (Open Viva; Siemens, Erlangen, Germany) and included both fast low-angle shot gradient-echo sequences (TR/TE, 22/8; flip angle, 30°; slice thickness, 4 mm; gap, 1 mm) and spin-echo sequences (200/8.1; flip angle, 80°; slice thickness, 4 mm; gap, 1 mm) in axial and coronal planes of the head and neck (posterior canthus to the suprasternal notch) and the hind limb (stifle [knee] to the mid abdomen). A 21-cm multipurpose and 35-cm flexible body receive coils were used for the head and neck and stifle, respectively. After obtaining baseline MR images, we injected 0.6 mL of Gadomer-17 (500 mmol of gadolinium per liter of Gadomer-17, approximately 7 mmol of gadolinium per kilogram of body weight using a 1.0-mL syringe and a 1.5-inch 23-gauge needle) interstitially into either the left or right lateral posterior tongue immediately adjacent to the mandibular third premolar at a depth of 1 cm and intradermally at the right or left stifle over the tibial tuberosity. Each pig was imaged at 0, 5, 10, 20, 40, and 60 min, and the sentinel node or nodes were identified as the first node or nodes to show enhancement. After the 60-min imaging study and using MR fluoroscopy (gradient-recalled acquisition in steady-state sequence, 30.6/13.1; flip angle, 90°), we percutaneously placed a 10 cm x 20-gauge 20-cm MR-compatible histology needle (E-Z-EM, Westbury, NY) into the sentinel node or nodes and left it as a marker.
Immediately after the percutaneous needle localization, the pigs were taken to surgery, and a standardized intraoperative sentinel node detection procedure was performed at the same Gadomer-17 contrast injection locations—in the tongue and hind limb—using 1% isosulfan blue contrast agent. In the neck, a midline skin incision was made from the level of the mentalis to the suprasternal notch to permit rapid visual detection of the contrast agent. Two milliliters of isosulfan blue contrast agent were injected into the tongue in the same locations as the Gadomer-17 contrast agent. The draining lymph vessels were identified and their course followed by immediate dissection, with careful attention not to dislodge the biopsy needle. For the lower limb, 2 mL of isosulfan blue contrast agent was injected intradermally at the stifle, and skin dissection (patella to femoral vein) was completed 2 min later to prevent disruption of any superficial lymphatic vessels. Lymph node or nodes along the dyed draining vessel were noted for blue coloration. Sentinel nodes identified on the MR images were compared with and matched to those identified by biopsy needle localization and the contrast agent.
Results
Interstitial and intradermal injection of Gadomer-17 in all six pigs resulted in increased signal in regional lymph nodes after transportation via lymphatic vessels into the lymph nodes in the neck and groin, respectively (Fig. 1A,1B,1C,1D). Sentinel lymph nodes were identified as the first node or nodes to show enhancement in the neck and groin. Enhancement occurred less than 5 min after injection. A single sentinel lymph node located in a grouping of multiple nodes was easily identified and marked. In some pigs, second-echelon nodes were identified beginning 30 min after injection. The sentinel lymph nodes reached maximal enhancement approximately 20 min after injection and maintained this level of enhancement for a minimum of 60 min (Fig. 2). In total, seven sentinel lymph nodes were identified in the neck and seven in the groin, with one pig showing two sentinel nodes in the neck and another showing two sentinel nodes in the groin. All 14 nodes were successfully located using MR fluoroscopy and percutaneous placement of the histology needles (Figs. 3A and 3B). In all cases, the isosulfan blue contrast agent sentinel lymph node method agreed with the location of the sentinel lymph node determined by Gadomer-17 and marked with the needle (Fig. 3C).
|
|
|
|
|
|
|
|
Discussion
Our study has shown that Gadomer-17 may be a valuable contrast agent for the detection of sentinel nodes. Gadomer-17, which is undergoing clinical investigation, is a highly branched globular macromolecule (dendrimer) contrast agent with 24 gadolinium complexes at its surface with an apparent molecular weight of 30-35 kd [20]. It has a high T1 relaxivity of 17.3 and 18.7 L/mmol-1 · sec-1 in water and plasma, respectively. Gadomer-17 has a biphasic blood half-life of approximately 9 and 168 min, respectively, with complete clearance mainly by glomerular filtration [20]. Interstitial injection of low doses (1-10 µmol/kg of body weight) of Gadomer-17 in the hind limbs of dogs showed a strong signal increase in iliac and inguinal lymph nodes within 15 min of injection [21]. The detection of lymph node metastases in VX2 tumor—bearing rabbits also has been possible because no early uptake occurs within intranodal metastatic deposits (Misselwitz et al., unpublished data). It has not been determined whether this differentiation is possible in a clinical setting. The minimal metastatic deposit size that can be identified has also not been determined.
In our experiment, injection methods were chosen to mimic current clinical practice for sentinel lymph node studies—namely, interstitial injection, to mimic intratumor injection for lesions such as squamous cell carcinoma of the head and neck [22] or breast tumors [1]—whereas intradermal injection mimicked peritumor injection for malignant melanomas [2]. The differences in the maximal T1 signal between the interstitial and intradermal routes, allowing differences in baseline T1, may relate to a higher absorption of Gadomer-17 into the bloodstream from muscle with a resulting overall lower absorption of Gadomer-17 into the lymphatics. The enhanced sentinel lymph node could be easily distinguished from the surrounding tissues after either injection method at the respective maximal T1 signal strength. It has been shown in the hind limbs of dogs that the minimal effective dose for enhancement of lymph nodes by Gadomer-17 is 2.5 µmol of gadolinium per kilogram of body weight [21].
Previous experimental studies using ultra-small superparamagnetic iron oxide and other contrast agents for nonspecific lymphangiography have not addressed temporal changes or specific node ordering [13,14,15,16,17,18], which is critical to the establishment of true sentinel node status [23]. Our previous study with Combidex [19] showed a significant signal loss artifact in sentinel lymph nodes but also in lower level nodes that had taken up the contrast agent over time. The intraoperative observation of brown staining of the node or nodes by ultrasmall superparamagnetic iron oxide without MR lymphangiography is not a sufficient identifier of a sentinel node because this contrast agent tracks beyond the first node and into the second-echelon nodes, as do contrast agents such as isosulfan blue, radiocolloids and other contrast agents. Furthermore, significant blood staining in the operative field or intense staining from isosulfan blue dye mapping could obscure this coloration.
Reliable and accurate identification of all sentinel nodes is necessary before lymphatic mapping and sentinel node biopsy can replace blind elective lymph node dissection. Although lymphatic mapping with radiocolloids directs and minimizes the amount of dissection necessary to identify sentinel lymph nodes, close proximity of the tumor and injection to the nodal basin, such as seen in tumors of the head and neck, can result in interfering background radiation [24, 25]. This effect can be seen in patients with melanoma in whom the primary skin lesion overlies the nodal basin. The spatial resolution and anatomic detail provided by MR imaging combined with MR imaging—guided needle localization of the sentinel lymph node could facilitate this process.
Dynamic MR sentinel lymph node mapping combined with MR imaging—guided fine-needle aspiration biopsy could establish a method for in vivo evaluation of sentinel lymph nodes and could preclude a need for open biopsy, especially when combined with the advancements in molecular techniques in pathology [26]. If a need still exists for open biopsy, the MR imaging—guided needle localization or placement of a marker wire would facilitate this procedure, particularly in complex areas such as the head and neck. As shown in our previous study [19], the strong negative signal artifacts from Combidex would have prevented monitoring of internodal needle positioning during percutaneous MR imaging—guided procedures. Gadomer-17 provides positive signal enhancement and permits reliable visualization of the needle during placement.
Recent experiments in our laboratory have shown that when obtaining cells for cytology from sentinel lymph nodes using Gadomer-17, MR imaging—guided fine-needle aspiration biopsy is feasible (Nason R, Torchia M, unpublished data) (Fig. 4). Although a positive finding of metastatic cells at fine-needle aspiration biopsy would be diagnostic, a false-negative biopsy result could occur if the internodal metastasis was not sampled or if only micrometastases were present. It would be beneficial if, during MR lymphangiography, a semipermanent local contrast-marking of the sentinel lymph node could simultaneously be achieved to provide a positive identification of the same sentinel lymph node during neck dissection performed at a later time. MR imaging—guided fine-needle aspiration and semipermanent lymph node marking are currently undergoing investigation in our laboratory.
|
Gadomer-17 is a useful contrast agent with interstitial MR lymphangiography for the detection of sentinel lymph nodes. The contrast enhancement of the sentinel lymph nodes allows reliable MR imaging—guided needle localization and may allow MR imaging—guided fine-needle aspiration biopsy. Based on the current experiment, the magnet time required for MR imaging—guided lymphangiography with needle localization in an open configuration magnet would be approximately 25 min, with MR imaging—guided fine-needle aspiration biopsy estimated to require an additional 5-10 min.
Acknowledgments
We thank Richard Nason and James Thliveris of the University of Manitoba for their technical support and encouragement.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Kobayashi, S. Kawamoto, Y. Sakai, P. L. Choyke, R. A. Star, M. W. Brechbiel, N. Sato, Y. Tagaya, J. C. Morris, and T. A. Waldmann Lymphatic Drainage Imaging of Breast Cancer in Mice by Micro-Magnetic Resonance Lymphangiography Using a Nano-Size Paramagnetic Contrast Agent J Natl Cancer Inst, May 5, 2004; 96(9): 703 - 708. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
