DOI: 10.1148/rg.241025713
(Radiographics. 2004;24:121-145.)
© RSNA, 2004
Clinical Breast Lymphoscintigraphy: Optimal Techniques for Performing Studies, Image Atlas, and Analysis of Images1
Borys R. Krynyckyi, MD,
Chun K. Kim, MD,
Martin R. Goyenechea, MD,
Peggy T. Chan, MD,
Zhuang-Yu Zhang, PhD and
Josef Machac, MD
1 From the Department of Radiology, Mount Sinai School of Medicine, New York, NY. Received August 14, 2002; revision requested November 18; final revision received August 21, 2003; accepted August 22. All authors have no financial relationships to disclose. Address correspondence to B.R.K., Department of Radiology, Box 1141, Mount Sinai Medical Center, One Gustave L. Levy Pl, New York, NY 10029-6574 (e-mail: syrob@msn.com).
 |
Abstract
|
|---|
Breast lymphoscintigraphy is increasingly performed before surgery to delineate the drainage to the sentinel node (SN) in the axilla. On the basis of the histologic status of harvested SNs, the disease status of the entire axilla can be predicted. This prediction allows a more limited dissection to be performed while maintaining staging accuracy comparable with that of classic axillary lymph node dissection. Lymphoscintigraphy assists surgeons in harvesting the SN during gamma probe–assisted axillary biopsy or dissection and provides a wide field of view survey, among other benefits. When certain injection protocols are used, lymphoscintigraphy can be performed in the afternoon before surgery the next morning, thus minimizing disruptions of tight surgical schedules. Image acquisition can be optimized and SN activity can be maximized by means of such factors as parameters for preparation of the radiotracer, injection techniques, energy settings for the gamma camera, breast displacement maneuvers, and techniques for marking and outlining the patient’s body. The ultimate goals are to delineate the true SN, maximize activity in the node for facilitated removal (even at next-day surgery), and deliver the information to the surgeon without delaying the surgical schedule.
© RSNA, 2004
Index Terms: Breast neoplasms, metastases, 07.33 • Breast neoplasms, radionuclide studies, 00.12161 • Breast neoplasms, staging, 00.32 • Lymphatic system, neoplasms, 997.8331 • Lymphatic system, radionuclide studies, 997.12961
|
LEARNING OBJECTIVES
|
|---|
After reading this article and taking the test, the reader will be able to:
- Discuss the theory behind select biopsy of sentinel lymph nodes in the setting of breast cancer and some of the issues associated with the procedure.
- Describe the technique of performing lymphoscintigraphy and methods of optimizing the procedure and the resulting images.
- Identify some of the patterns that can be seen at lymphoscintigraphy and their meanings.
|
Introduction
|
|---|
The goal of breast cancer treatment is to minimize the extent of tissue disruption while facilitating survival. Breast-conserving surgery (vs modified radical mastectomy) for appropriately small tumors with no reduction in survival has become well established (1,2). The staging and control of disease have been intimately associated with axillary lymph node dissection (ALND) as a method to detect metastatic spread of disease (potentially altering patient care) while providing regional control of disease in the axilla. After removal of diseased nodes, a prolongation in long-term survival has been suggested (3,4). In its classic form, ALND involves extensive dissection of most of the level I, II, and III nodes. It can lead to significant morbidity in the form of lymphedema and other conditions (5,6). With the trend toward earlier detection of less advanced disease, a completely negative axilla is being found more frequently, making the morbidity associated with ALND a less acceptable occurrence.
As with breast-conserving surgery, methods to lessen the negative effect of staging the axilla while still providing regional control have been developed. The sentinel node (SN) concept is a major advance toward this goal. As a staging technique to reduce the dissection field and postsurgical morbidity, sentinel lymph node biopsy has been validated for melanoma and now is being applied to breast carcinoma (7,8). The concept is based on the assumption of an orderly progression of tumor from the primary lesion through the lymphatic channels to the first lymph node encountered along those channels (the SN), which then becomes the site of the earliest distant metastases. A low skip rate, which refers to metastasis bypassing the first node to lodge in a more distant node downstream, has been observed in many studies. This allows the SN to provide an accurate prediction of the status of the entire axilla with respect to metastases without the need for extensive dissection of more distant nodes past the SN. By harvesting only the SN and any immediately adjacent nodes, a reduction in morbidity can be achieved while accurately predicting metastatic spread and the potential need for ALND for regional control when metastases are present (3,9).
By injecting a dye or radiocolloid around the region of the primary breast lesion, the lymphatic drainage of that region to the first node along the lymphatic pathways (the SN) can be deduced, either by dissection of the lymphatic pathways in the case of dye or with a gamma camera or handheld gamma probe sensitive to the emissions of the radiocolloid. The initial injection of radiocolloid around the primary breast lesion, its progression along the lymphatic channels, and phagocytosis by macrophages in the SNs is the basis of imaging with the planar gamma camera. Mapping out the lymphatic drainage prior to surgery to assist the surgeon in facilitating removal of the SN is the goal of the lymphoscintigraphy specialist.
The technique of lymphoscintigraphy offers several benefits to the surgeons: The wide field of view gamma camera images the entire chest, with the axillary and internal mammary node regions in one field of view. It can also facilitate detection of intramammary nodes in the breast itself. Lymphoscintigraphy can "map out" diffusion zones emanating from injection sites (interstitial diffusion of radiotracer away from the injection site) and can help identify SNs close to or partly hidden by these diffusion fields, especially when the sitting or standing position is used. This position is not readily available when the handheld gamma probe is used during surgery. The occasionally complex three-dimensional distribution of SNs is also delineated, which facilitates their removal, as well as the presence of secondary distant echelon nodes.
In addition, lymphoscintigraphy can alert the surgeons to "end-on" effects and dilated areas of prominent lymphatic channels, which might mistakenly be construed to represent SNs, by mapping out these findings and preparing the surgeons for the type of findings they might encounter with the handheld probe in this region. Surface contamination, which might be erroneously construed to represent SNs, is easily revealed with orthogonal views and triangulation methods. The activity will appear to reside at the surface in these situations. Surgeons just learning the intraoperative technique appreciate the information lymphoscintigraphy provides. Use of a large field of view camera in the initial survey along with the reference points provided by triangulation markings on the patient’s body facilitate surgical planning and assist the surgeon in use of the handheld probe. Failure to visualize an SN during lymphoscintigraphy alerts the surgical team that blue dye and traditional ALND might be needed.
Several of the technical issues in lymphoscintigraphy were previously presented (5,10,11). In this article, we present additional suggestions and further practical refinements that we have arrived at by performing daily clinical procedures. Our experience has also allowed us to observe the wide range of possible patterns of lymphatic drainage, which are discussed mainly in the figure legends. Specific topics discussed include the radiopharmaceutical preparation, injection techniques, additional injection site issues, pain control, techniques for preventing contamination, massaging the breast, image acquisition parameters, arm positioning and breast displacement maneuvers, triangulation techniques and body marking and outlining, internal mammary SNs, reasons to strive for a "hot" node, combined and hybrid injection techniques, blue dye and special staining of SNs, and factors affecting visualization and detection of SNs.
|
Radiopharmaceutical Preparation
|
|---|
In the United States, the only available radiotracer formulation as a lymphoscintigraphic agent is technetium-99m sulfur colloid, which can be filtered or unfiltered. The smaller filtered preparation (100 or 200 nm) (Acrodisc syringe filter; Pall, Ann Arbor, Mich) generally achieves better lymphatic penetration and is commonly used. Unfiltered preparations have a range of sizes that is difficult to categorize due to multiple preparation and measurement variables that affect the reported size range, but most particles (>75%) are probably 100–1,000 nm in diameter (8). It has been suggested that unfiltered preparations decrease diffusion from the lesion injection site (increased diffusion could potentially obscure closely adjacent nodes) but, in theory, at the expense of decreased delivery of activity to the SN for the same reasons. In addition, larger particles would limit the number of echelon nodes (nodes downstream of the first SN) seen along the lymphatic chain, although it has been suggested that harvesting echelon nodes along with the SN decreases the false-negative rate (12–14). A high specific activity preparation has been reported to further improve performance (15,16). Tc-99m nanocolloid, which is used outside the United States, has a much smaller average particle range, with almost all particles less than 100 nm in diameter (8).
|
Injection Techniques
|
|---|
This is perhaps the most controversial aspect of performing lymphoscintigraphy. Multiple injection techniques exist (Fig 1), with several proponents for each technique in the literature.
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Different sites of injection. Right: Drawing shows multiple potential injection sites in the right breast, with the injection patterns altered when close to the axilla. A, For uncomplicated lesions, injections can be performed at two to four perilesional locations. B, Lumpectomy sites can be injected at two to six locations, with a minimum of one injection on each side of the scar. C, When lesions are close to the axilla, it is reasonable to shift portions of the total injected dose away from the regions closest to the axilla to diminish the negative effects of injection site diffusion, which could hide adjacent SNs. Top left: Dermal injections are usually performed at the skin directly above the lesion, whereas areolar injections can be performed anywhere in the areola or at the aspect of the areolar-cutaneous junction closest to the lesion.
|
|
Perilesional Injections
Probably the most common technique used in the United States (8), perilesional (also called intraparenchymal or peritumoral) injections are administered generally around the lesion and can include injections deeper or superficial to the plane of the lesion (but the term usually signifies injections on the side of the lesion). Typically, two to four injections of an equally divided dose are performed. The number of injections can beincreased for very large lesions. Lumpectomy site injections call for at least two injections, one on each side of the scar, with multiple injections reported to increase the rate of node visualization (17) (Figs 1, 2). Final images are usually obtained 45 minutes to 1–2 hours or more after injection, as activity reaches the SN slowly as it diffuses through tissues to reach lymphatic vessels. Perilesional injections are more forgiving of position compared with intralesional injections, as discussed later. A greater mass of surrounding breast tissue undergoes injection, which has more lymphatic drainage compared with the tumor itself.
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. (a) Anterior midstudy image of the left breast, obtained after concurrent perilesional and intradermal injections, shows a tortuous lymphatic pathway leading to the SN from the injection sites (IS). (b) Anterior end-of-study image of the right side of the chest, obtained in another patient after injections into the upper (U) and lower (L) tissues flanking a lumpectomy scar and at the areolar-cutaneous junction (arrow), shows several nodes in the axilla (SNs).
|
|
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. (a) Anterior midstudy image of the left breast, obtained after concurrent perilesional and intradermal injections, shows a tortuous lymphatic pathway leading to the SN from the injection sites (IS). (b) Anterior end-of-study image of the right side of the chest, obtained in another patient after injections into the upper (U) and lower (L) tissues flanking a lumpectomy scar and at the areolar-cutaneous junction (arrow), shows several nodes in the axilla (SNs).
|
|
Total volumes of injected material generally range from just below 1 mL to a maximum of 8–16 mL (4). These large volumes were proposed as a method to open up lymphatic junctions and allow radiotracer to enter the channels. Others suggest that extreme volumes are not physiologic and even raise the question of tumor spread if such volumes are administered intralesionally (8,18,19). A more modest volume of 3–4 mL is probably adequate. Breast size also dictates the volume of injection, with smaller volumes recommended for small breasts and larger volumes for large breasts. A wide range of activity levels are used, typically 200 µCi (7.4 MBq) to 2–3 mCi (74–111 MBq), with a maximum of 10 mCi (370 MBq) in one study (20).
Problems with diffusion zone activity obscuring the SNs after injections close to the axilla can occur, especially if the primary lesion is located in the upper outer quadrant, close to the axillary nodes. The perilesional injection technique is one of the few that will delineate internal mammary nodes to any extent (Figs 3, 4). Deeper injections and small breast size have been associated with an increased rate of internal mammary node visualization. These two factors are most likely related to the proximity between the injection site and the deeper lymphatic channels draining to the internal mammary nodes (21).
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Importance of multiangle views in assessing overlap. Top: Anterior and lateral backlit end-of-study images of the left side of the chest show the injection sites (IS) and SNs. An internal mammary SN (IM) is also noted. Bottom: Images show an overlap effect for both the injection sites and the SNs, depending on the view.
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Sequence of consecutive anterior images. (a) Image obtained before "LymphoBoost" (LB) injection (injection at the areolar-cutaneous junction) shows a perilesional injection site (IS), internal mammary SNs (IM), and an axillary SN (arrow). The liver activity is due to capillary penetration or partial vascularization of the radiotracer dose during injections. (b) Image obtained after lymphoboost (LB) injection shows marked augmentation of axillary SN activity with additional nodes noted (arrow). (c, d) Images obtained with the arm up (c) or down (d) show a marked effect of arm position on the perceived locations of the SNs and the injection site with a potential corresponding shift in the surface markings. The position of the internal mammary SNs (arrow) remains fixed. (Reprinted, with permission, from reference 5.)
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Sequence of consecutive anterior images. (a) Image obtained before "LymphoBoost" (LB) injection (injection at the areolar-cutaneous junction) shows a perilesional injection site (IS), internal mammary SNs (IM), and an axillary SN (arrow). The liver activity is due to capillary penetration or partial vascularization of the radiotracer dose during injections. (b) Image obtained after lymphoboost (LB) injection shows marked augmentation of axillary SN activity with additional nodes noted (arrow). (c, d) Images obtained with the arm up (c) or down (d) show a marked effect of arm position on the perceived locations of the SNs and the injection site with a potential corresponding shift in the surface markings. The position of the internal mammary SNs (arrow) remains fixed. (Reprinted, with permission, from reference 5.)
|
|
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Sequence of consecutive anterior images. (a) Image obtained before "LymphoBoost" (LB) injection (injection at the areolar-cutaneous junction) shows a perilesional injection site (IS), internal mammary SNs (IM), and an axillary SN (arrow). The liver activity is due to capillary penetration or partial vascularization of the radiotracer dose during injections. (b) Image obtained after lymphoboost (LB) injection shows marked augmentation of axillary SN activity with additional nodes noted (arrow). (c, d) Images obtained with the arm up (c) or down (d) show a marked effect of arm position on the perceived locations of the SNs and the injection site with a potential corresponding shift in the surface markings. The position of the internal mammary SNs (arrow) remains fixed. (Reprinted, with permission, from reference 5.)
|
|
View larger version (129K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4d. Sequence of consecutive anterior images. (a) Image obtained before "LymphoBoost" (LB) injection (injection at the areolar-cutaneous junction) shows a perilesional injection site (IS), internal mammary SNs (IM), and an axillary SN (arrow). The liver activity is due to capillary penetration or partial vascularization of the radiotracer dose during injections. (b) Image obtained after lymphoboost (LB) injection shows marked augmentation of axillary SN activity with additional nodes noted (arrow). (c, d) Images obtained with the arm up (c) or down (d) show a marked effect of arm position on the perceived locations of the SNs and the injection site with a potential corresponding shift in the surface markings. The position of the internal mammary SNs (arrow) remains fixed. (Reprinted, with permission, from reference 5.)
|
|
The exact anatomic location of the injection is at times obscure in the literature, with terms such as subcutaneous or subdermal occasionally used, mainly in Europe. These terms vaguely refer to injections somewhere above the tumor but below the skin (subdermal), which could also be classified as perilesional in certain instances if the lesion is very close to the skin surface.
Cutaneous Injections
Cutaneous injections are generally delivered into the skin above the lesion, at a depth varying from just at or below the dermis (subdermal) to as shallow as possible (creating a wheal or blister). The question whether these surface injections delineate the same nodes as perilesional injections is controversial (Fig 5). This difference is indisputable where internal mammary nodes are concerned, which are not delineated by surface injections to any significant extent (0%–3%). In contrast, multiple authors have shown no significant difference for axillary node visualization or harvesting (outside the internal mammary node basins), claiming that the same axillary nodes are drained to, an "all roads lead to Rome" concept (22–24). Some authors have detected subtle differences, with discongruent nodes noted (not including internal mammary basin nodes) (20,25, 26).
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Potential connections between lymph nodes and lymphatic channels. (a) Drawing of a lymph node shows serial connections to germinal centers and alternate connections that mostly bypass the germinal centers, running through the node or on the surface without connecting (19). (b) Diagrams show a serial connection (which is most common) and an alternate parallel connection, which has reduced flow compared with that of serial connections. (c) Anterior midstudy image of the left side of the chest, obtained after perilesional and skin injections, shows a bright focus and a fainter focus (arrow), which appeared simultaneously. The latter is closer to the tumor in the breast and potentially represents a node fed through a side parallel connection or a node that is only partially visualized due to replacement by metastases. Because the fainter focus intensified over time and persisted on delayed views, ectasia was less likely. (d) Diagrams show possible connections between a tumor and lymph nodes. PL = perilesional injection site. 1, Perilesional injection demonstrates the SN and a more distant axillary echelon node. 2, Surface injection leads to augmentation of the nodes seen after the perilesional injection. EN = echelon node. 3, Surface injection leads to augmentation of the nodes seen after the perilesional injection, with additional foci appearing proximally, upstream from the originally delineated SNs. This appearance could represent pooling of activity in a dilated area, that is, a pseudo-SN (PSN) or a reverse echelon node (REN). A reverse echelon node is an upstream node that is distinct from the primary SN seen after the initial perilesional injection and that receives activity only from the surface injection. It is usually closer to the tumor and technically is not part of the tumor drainage, since it is not seen with the original perilesional injection but only after the surface injection. The more distant true SN is almost always augmented by surface injections, since it is downstream along the lymphatic chain. 4, Surface injection leads to activity bypassing the true SN because the lymphatic channels are not connected at that point. There is also demonstration of a distant echelon node, which might not contain tumor. The former finding is referred to as a missed SN (MSN) if no perilesional injection was performed; the latter finding is potentially referred to as a false SN (FSN) or a false-negative node (FNN) (5,25). (e) Anterior midstudy image of the left side of the chest, obtained after perilesional and areolar-cutaneous junction injections, shows activity tracking to the same sentinel and echelon nodes along different pathways. Such an "all roads lead to Rome" pattern is seen most of the time with combination injections. (Fig 5a reprinted, with permission, from reference 19; Fig 5d reprinted, with permission, from reference 5.)
|
|
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Potential connections between lymph nodes and lymphatic channels. (a) Drawing of a lymph node shows serial connections to germinal centers and alternate connections that mostly bypass the germinal centers, running through the node or on the surface without connecting (19). (b) Diagrams show a serial connection (which is most common) and an alternate parallel connection, which has reduced flow compared with that of serial connections. (c) Anterior midstudy image of the left side of the chest, obtained after perilesional and skin injections, shows a bright focus and a fainter focus (arrow), which appeared simultaneously. The latter is closer to the tumor in the breast and potentially represents a node fed through a side parallel connection or a node that is only partially visualized due to replacement by metastases. Because the fainter focus intensified over time and persisted on delayed views, ectasia was less likely. (d) Diagrams show possible connections between a tumor and lymph nodes. PL = perilesional injection site. 1, Perilesional injection demonstrates the SN and a more distant axillary echelon node. 2, Surface injection leads to augmentation of the nodes seen after the perilesional injection. EN = echelon node. 3, Surface injection leads to augmentation of the nodes seen after the perilesional injection, with additional foci appearing proximally, upstream from the originally delineated SNs. This appearance could represent pooling of activity in a dilated area, that is, a pseudo-SN (PSN) or a reverse echelon node (REN). A reverse echelon node is an upstream node that is distinct from the primary SN seen after the initial perilesional injection and that receives activity only from the surface injection. It is usually closer to the tumor and technically is not part of the tumor drainage, since it is not seen with the original perilesional injection but only after the surface injection. The more distant true SN is almost always augmented by surface injections, since it is downstream along the lymphatic chain. 4, Surface injection leads to activity bypassing the true SN because the lymphatic channels are not connected at that point. There is also demonstration of a distant echelon node, which might not contain tumor. The former finding is referred to as a missed SN (MSN) if no perilesional injection was performed; the latter finding is potentially referred to as a false SN (FSN) or a false-negative node (FNN) (5,25). (e) Anterior midstudy image of the left side of the chest, obtained after perilesional and areolar-cutaneous junction injections, shows activity tracking to the same sentinel and echelon nodes along different pathways. Such an "all roads lead to Rome" pattern is seen most of the time with combination injections. (Fig 5a reprinted, with permission, from reference 19; Fig 5d reprinted, with permission, from reference 5.)
|
|
View larger version (164K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Potential connections between lymph nodes and lymphatic channels. (a) Drawing of a lymph node shows serial connections to germinal centers and alternate connections that mostly bypass the germinal centers, running through the node or on the surface without connecting (19). (b) Diagrams show a serial connection (which is most common) and an alternate parallel connection, which has reduced flow compared with that of serial connections. (c) Anterior midstudy image of the left side of the chest, obtained after perilesional and skin injections, shows a bright focus and a fainter focus (arrow), which appeared simultaneously. The latter is closer to the tumor in the breast and potentially represents a node fed through a side parallel connection or a node that is only partially visualized due to replacement by metastases. Because the fainter focus intensified over time and persisted on delayed views, ectasia was less likely. (d) Diagrams show possible connections between a tumor and lymph nodes. PL = perilesional injection site. 1, Perilesional injection demonstrates the SN and a more distant axillary echelon node. 2, Surface injection leads to augmentation of the nodes seen after the perilesional injection. EN = echelon node. 3, Surface injection leads to augmentation of the nodes seen after the perilesional injection, with additional foci appearing proximally, upstream from the originally delineated SNs. This appearance could represent pooling of activity in a dilated area, that is, a pseudo-SN (PSN) or a reverse echelon node (REN). A reverse echelon node is an upstream node that is distinct from the primary SN seen after the initial perilesional injection and that receives activity only from the surface injection. It is usually closer to the tumor and technically is not part of the tumor drainage, since it is not seen with the original perilesional injection but only after the surface injection. The more distant true SN is almost always augmented by surface injections, since it is downstream along the lymphatic chain. 4, Surface injection leads to activity bypassing the true SN because the lymphatic channels are not connected at that point. There is also demonstration of a distant echelon node, which might not contain tumor. The former finding is referred to as a missed SN (MSN) if no perilesional injection was performed; the latter finding is potentially referred to as a false SN (FSN) or a false-negative node (FNN) (5,25). (e) Anterior midstudy image of the left side of the chest, obtained after perilesional and areolar-cutaneous junction injections, shows activity tracking to the same sentinel and echelon nodes along different pathways. Such an "all roads lead to Rome" pattern is seen most of the time with combination injections. (Fig 5a reprinted, with permission, from reference 19; Fig 5d reprinted, with permission, from reference 5.)
|
|
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d. Potential connections between lymph nodes and lymphatic channels. (a) Drawing of a lymph node shows serial connections to germinal centers and alternate connections that mostly bypass the germinal centers, running through the node or on the surface without connecting (19). (b) Diagrams show a serial connection (which is most common) and an alternate parallel connection, which has reduced flow compared with that of serial connections. (c) Anterior midstudy image of the left side of the chest, obtained after perilesional and skin injections, shows a bright focus and a fainter focus (arrow), which appeared simultaneously. The latter is closer to the tumor in the breast and potentially represents a node fed through a side parallel connection or a node that is only partially visualized due to replacement by metastases. Because the fainter focus intensified over time and persisted on delayed views, ectasia was less likely. (d) Diagrams show possible connections between a tumor and lymph nodes. PL = perilesional injection site. 1, Perilesional injection demonstrates the SN and a more distant axillary echelon node. 2, Surface injection leads to augmentation of the nodes seen after the perilesional injection. EN = echelon node. 3, Surface injection leads to augmentation of the nodes seen after the perilesional injection, with additional foci appearing proximally, upstream from the originally delineated SNs. This appearance could represent pooling of activity in a dilated area, that is, a pseudo-SN (PSN) or a reverse echelon node (REN). A reverse echelon node is an upstream node that is distinct from the primary SN seen after the initial perilesional injection and that receives activity only from the surface injection. It is usually closer to the tumor and technically is not part of the tumor drainage, since it is not seen with the original perilesional injection but only after the surface injection. The more distant true SN is almost always augmented by surface injections, since it is downstream along the lymphatic chain. 4, Surface injection leads to activity bypassing the true SN because the lymphatic channels are not connected at that point. There is also demonstration of a distant echelon node, which might not contain tumor. The former finding is referred to as a missed SN (MSN) if no perilesional injection was performed; the latter finding is potentially referred to as a false SN (FSN) or a false-negative node (FNN) (5,25). (e) Anterior midstudy image of the left side of the chest, obtained after perilesional and areolar-cutaneous junction injections, shows activity tracking to the same sentinel and echelon nodes along different pathways. Such an "all roads lead to Rome" pattern is seen most of the time with combination injections. (Fig 5a reprinted, with permission, from reference 19; Fig 5d reprinted, with permission, from reference 5.)
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5e. Potential connections between lymph nodes and lymphatic channels. (a) Drawing of a lymph node shows serial connections to germinal centers and alternate connections that mostly bypass the germinal centers, running through the node or on the surface without connecting (19). (b) Diagrams show a serial connection (which is most common) and an alternate parallel connection, which has reduced flow compared with that of serial connections. (c) Anterior midstudy image of the left side of the chest, obtained after perilesional and skin injections, shows a bright focus and a fainter focus (arrow), which appeared simultaneously. The latter is closer to the tumor in the breast and potentially represents a node fed through a side parallel connection or a node that is only partially visualized due to replacement by metastases. Because the fainter focus intensified over time and persisted on delayed views, ectasia was less likely. (d) Diagrams show possible connections between a tumor and lymph nodes. PL = perilesional injection site. 1, Perilesional injection demonstrates the SN and a more distant axillary echelon node. 2, Surface injection leads to augmentation of the nodes seen after the perilesional injection. EN = echelon node. 3, Surface injection leads to augmentation of the nodes seen after the perilesional injection, with additional foci appearing proximally, upstream from the originally delineated SNs. This appearance could represent pooling of activity in a dilated area, that is, a pseudo-SN (PSN) or a reverse echelon node (REN). A reverse echelon node is an upstream node that is distinct from the primary SN seen after the initial perilesional injection and that receives activity only from the surface injection. It is usually closer to the tumor and technically is not part of the tumor drainage, since it is not seen with the original perilesional injection but only after the surface injection. The more distant true SN is almost always augmented by surface injections, since it is downstream along the lymphatic chain. 4, Surface injection leads to activity bypassing the true SN because the lymphatic channels are not connected at that point. There is also demonstration of a distant echelon node, which might not contain tumor. The former finding is referred to as a missed SN (MSN) if no perilesional injection was performed; the latter finding is potentially referred to as a false SN (FSN) or a false-negative node (FNN) (5,25). (e) Anterior midstudy image of the left side of the chest, obtained after perilesional and areolar-cutaneous junction injections, shows activity tracking to the same sentinel and echelon nodes along different pathways. Such an "all roads lead to Rome" pattern is seen most of the time with combination injections. (Fig 5a reprinted, with permission, from reference 19; Fig 5d reprinted, with permission, from reference 5.)
|
|
Intradermal injections above the lesion result in significantly faster and brighter visualization of SNs than perilesional injections (8,10,11,24). Final imaging can usually be performed after only 30–45 minutes, as the radiotracer reaches SNs much faster due to the rich lymphatic drainage of the skin to the axilla compared with the drainage of the deeper parenchyma of perilesional injections. The average volume of injection is also much less, typically 0.2–1 mL. The amount of activity used is usually also smaller than for perilesional injections, typically ranging from 150 µCi (5.6 MBq) to 1–2 mCi (37–74 MBq). Problems with diffusion from the injection site interfering with the detection of SNs also exist, but this is probably less of a problem than for perilesional injections.
Areolar Region Injections
On the basis of anatomic considerations, that is, the presence of a highly dense subareolar lymphatic plexus that drains the areola to the axilla, use of injections in this region was examined by several authors (25,27–34). Variations in technique exist, with some authors achieving an infiltration of the whole areolar region with radiotracer or using a quadrant-based approach depending on lesion location, whereas others performed injections at a fixed site or variable sites. Different volumes have been used and injected at different depths.
As with skin injections, problems of discongruence between areolar and perilesional injections exist concerning internal mammary node detection, the rate of which is also very low. Issues of potential discongruence similarly exist with areolar injections as for skin injections above the tumor in their ability to delineate the same axillary nodes as perilesional injections. In many studies, these issues were not directly addressed by using radionuclide-only injections, as blue dye was used as a comparative agent instead. However, the rate of complete discongruence, which is defined as a completely missed SN from an areolar-only injection, appears to be very low based on the reported comparative results.
Volumes of injection and activity amounts are typically similar to those for skin injections above the tumor, although some centers have used substantially greater volumes of 4–5 mL over multiple or single subareolar injection sites (27,29, 34). Activity reaches the SN at a similar or even faster rate than for dermal injections above the tumor. In addition, more of the total dose reaches the SN compared with skin injections above the tumor.
Intralesional Injections
Intralesional injections have been performed outside the United States but rarely within the United States. Intralesional injections have been considered by several authors as the most accurate, as they simulate drainage from the tumor itself (35–37). However, there can be poor lymphatic drainage from the lesion because of poor lymphatic permeation of solid tumor tissues (8). In addition, questions have been raised about spread of disease from intratumoral needle tracks and the effects the injected material has on dislodging tumor cells (8).
The main technical problem with this technique is the difficulty of finding the center of the lesion if the lesion is not palpable, is palpable but is small or deep, or is multifocal, without use of special techniques such as ultrasonographic (US) or mammographic guidance. Even if the lesion is palpable, the center is still not easily determined. Whether the increased placement accuracy of the radiotracer (compared with that of other perilesional techniques) truly makes a difference clinically has not been answered adequately. As with other perilesional-type injections, problems with diffusion zone activity obscuring SNs from injections close to the axilla occur, but these problems can be minimized if smaller volumes of injection are used, which typically range from 0.2 to 1 mL.
|
Additional Injection Site Issues
|
|---|
Wire localization poses a challenge in identifying the site to inject, as the location where the wire exits the skin is obviously not the true position of the lesion deep in the breast. Review of mammograms and the surface markings made during US for lesions originally localized with these techniques can help define the general area to inject.
Fluid-filled lumpectomy sites pose their own challenge as to where to inject. Injection into a serosanguineous cavity, which can be very large, will rarely lead to SN visualization. Injection into the tissues surrounding the cavity (or into the skin if surface injections are used) is desirable, but the needle tip position cannot always be reliably guided by plunger withdrawal maneuvers to check for serosanguineous fluid. US guidance can help with the injections. Nevertheless, some investigators have reported reduced rates of SN visualization in these cases, whereas others have not found any differences (17,38,39). Injection into the wall at multiple sites has been suggested to improve performance (17) (Figs 1, 2).
Finally, for any deep injection where the tumor is close to the chest wall, it is important to avoid injection into the pectoral muscle or a breast prosthesis for obvious reasons.
|
Pain Control
|
|---|
Experience has shown a need for routine control of pain during injections. Shallow intradermal injections are more painful than intraparenchymal ones. A topical anesthetic has worked well for us. EMLA cream alone (lidocaine 2.5% and prilocaine 2.5%; Astra Pharmaceuticals, Wayne, Pa) has proved to be effective in controlling local surface pain during injections for all but areolar-cutaneous junction injections (11,16). The cream needs to be applied to the skin for at least 20–30 minutes to be fully effective. For even more sensitive shallow areolar injections, besides use of a topical anesthetic, lidocaine can be added to the injection syringes for additional pain control (as discussed later).
|
Techniques for Preventing Contamination
|
|---|
Good injection technique will prevent contamination and reduce the chance of inappropriately labeling such artifacts as nodes. Methods of avoiding contamination include applying negative pressure to the syringe during withdrawal and immediately dabbing the injection site with a gauze pad. This gauze pad is immediately discarded for a fresh one. Bandaging the site after all injections are completed and keeping the patient’s hands away from the sites will also help (11). Dynamic acquisitions are useful. Contamination will not intensify with time, whereas true SNs will. Adding an air bubble into the syringe on top of the dose will help ensure delivery of the entire dose. Care must be exercised to avoid spraying portions of the dose.
|
Massaging the Breast
|
|---|
Massaging the injection site after injecting the radiotracer has been suggested to improve the distribution of radiotracer and increase SN visualization (8,40). Mild massaging for a short period is probably beneficial. Vigorous and prolonged massaging has been criticized by some as not physiologic and potentially harmful by spreading micrometastases and can lead to contamination if the proper technique is not used (41,42).
|
Acquisition Parameters
|
|---|
We recommend a large field of view gamma camera equipped with a high-resolution collimator. A 256 x 256 matrix can be used to obtain images with high resolution. To eliminate some of the negative effects of scatter from the injection site, the energy window for Tc-99m can be offset upward or dual energy channels can be used (Fig 6) (5,11). The dual-channel method allows a facility to view images in either a combined channel mode or separately to determine the benefit of this technique with any particular camera system. Dynamic image acquisitions for all phases of the study (60-second frames obtained for a 6–15-minute session), if time permits, are suggested for several reasons. Different summations of frames to produce optimal images are then possible. Images presented to the surgeons can be summed to include frames with and without the transmission source outlines when nodes are occasionally faint. Contamination is easier to detect in a dynamically obtained data set. In addition, first-line SNs can be easily differentiated from more distant echelon nodes.
View larger version (73K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. (a) Anterior images obtained with dual energy channels. The image obtained with the lower channel shows little useful information, but the image obtained with the higher channel shows a partly hidden SN. (b) Lateral images of another patient obtained with dual energy channels. The image obtained with the lower channel shows shallow angle scatter from the arm, but the image obtained with the higher channel shows an SN (arrow), which is no longer obscured by the scatter.
|
|
View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. (a) Anterior images obtained with dual energy channels. The image obtained with the lower channel shows little useful information, but the image obtained with the higher channel shows a partly hidden SN. (b) Lateral images of another patient obtained with dual energy channels. The image obtained with the lower channel shows shallow angle scatter from the arm, but the image obtained with the higher channel shows an SN (arrow), which is no longer obscured by the scatter.
|
|
Injections in the skin above the tumor or at the areola can create a rapidly changing pattern, with prominent transient lymphatic channels, bolus streaming, and areas of ectasia that manifest as transient accumulation of activity in focally dilated lymphatic channels, producing focal "hot spots." These can be potentially confusing to interpret for centers just starting to perform lymphoscintigraphy. In these cases, dynamic and delayed imaging can help identify true SNs, which become brighter over time (Figs 7–9) (5,11). Finally, viewing in cine mode, as is also done with gastrointestinal bleeding studies, makes use of the mind’s ability to better detect subtle differences in patterns when the data are presented dynamically (43).
View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Early (a) and late (b) anterior images of the left side of the chest show an SN (arrow) and the injection site (IS). There is also an "inverted J"-type pattern, which represents a lymphatic channel that courses above the SN before turning inferiorly, deep to the SN. Potential causes of this pattern include pathways in the skin plexus that follow a large, cranially bulging breast before connecting to the deeper plexus of axillary lymphatics (dashed curve in b). Other possible causes include upward "tenting" of lymphatics due to a raised arm position, thus producing an inflection point, or alternate pathways secondary to complete replacement of nodes by the tumor. (Reprinted, with permission, from reference 5.)
|
|
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Early (a) and late (b) anterior images of the left side of the chest show an SN (arrow) and the injection site (IS). There is also an "inverted J"-type pattern, which represents a lymphatic channel that courses above the SN before turning inferiorly, deep to the SN. Potential causes of this pattern include pathways in the skin plexus that follow a large, cranially bulging breast before connecting to the deeper plexus of axillary lymphatics (dashed curve in b). Other possible causes include upward "tenting" of lymphatics due to a raised arm position, thus producing an inflection point, or alternate pathways secondary to complete replacement of nodes by the tumor. (Reprinted, with permission, from reference 5.)
|
|
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Anterior midstudy image of the right side of the chest shows the injection site (IS) and an SN. There is also a focus of activity in the breast (arrow), which was excised. Histologic analysis demonstrated that the focus was a dilated lymphatic channel. No node was found. Such foci are known as pseudo-SNs. They usually disappear on delayed images and do not intensify over time.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. (a) Anterior early image of the left side of the chest shows a focus of activity (arrow). This is not an SN but a lymphatic channel seen "end on," which can be erroneously interpreted as an SN because of the end-on effect. (b-d) Multiangle sequence of images presented from anterior (b) to lateral (d) shows the end-on effect. The lymphatic channel appears as an arch leading from the injection site to an SN, which is mostly hidden on the anterior images (a and b) by the injection site. D = dermal injection site, P = perilesional injection site. (Reprinted, with permission, from reference 5.)
|
|
View larger version (75K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. (a) Anterior early image of the left side of the chest shows a focus of activity (arrow). This is not an SN but a lymphatic channel seen "end on," which can be erroneously interpreted as an SN because of the end-on effect. (b-d) Multiangle sequence of images presented from anterior (b) to lateral (d) shows the end-on effect. The lymphatic channel appears as an arch leading from the injection site to an SN, which is mostly hidden on the anterior images (a and b) by the injection site. D = dermal injection site, P = perilesional injection site. (Reprinted, with permission, from reference 5.)
|
|
View larger version (67K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9c. (a) Anterior early image of the left side of the chest shows a focus of activity (arrow). This is not an SN but a lymphatic channel seen "end on," which can be erroneously interpreted as an SN because of the end-on effect. (b-d) Multiangle sequence of images presented from anterior (b) to lateral (d) shows the end-on effect. The lymphatic channel appears as an arch leading from the injection site to an SN, which is mostly hidden on the anterior images (a and b) by the injection site. D = dermal injection site, P = perilesional injection site. (Reprinted, with permission, from reference 5.)
|
|
Top
Abstract
LEARNING OBJECTIVES
