(Radiology. 1999;211:815-828.)
© RSNA, 1999
Cross-sectional Nodal Atlas: A Tool for the Definition of Clinical Target Volumes in Three-dimensional Radiation Therapy Planning1
Rafael Martinez-Monge, MD,
Patrick S. Fernandes, MD,
Nilendu Gupta, PhD and
Reinhard Gahbauer, MD
1 From the Division of Radiation Oncology, the Arthur G. James Cancer Hospital, Ohio State University, 300 W Tenth Ave, Columbus, OH 43210. Received July 15, 1998; revision requested August 27; revision received October 16; accepted November 23, 1999. Address reprint requests to R.M.
 |
Abstract
|
|---|
Virtual three-dimensional clinical target volume definition requires the identification of areas suspected of containing microscopic disease (frequently related to nodal stations) on a set of computed tomographic (CT) images, rather than the traditional approach based on anatomic landmarks. This atlas displays the clinically relevant nodal stations and their correlation with normal lymphatic pathways on a set of CT images.
Index terms: Computed tomography (CT), three-dimensional, 99.12917, 99.92 • Lymphatic system, 99.12917, 99.92 • Special reports • Treatment planning, 99.92
|
Introduction
|
|---|
When radiation is used with curative intent, the radiation volume usually encompasses the detectable tumor and the anatomic areas thought to be at risk for metastatic spread. The International Commission on Radiation Units and Measurements Report No. 50 (1) defines gross tumor volume (GTV) as the gross palpable or visible or demonstrable extent and location of the malignant growth. The same report defines clinical target volume (CTV) as a volume that contains a demonstrable GTV and/or is considered to contain (only) microscopic, subclinical extensions at a certain probability level. In clinical practice, the determination of the extent of the CTV is based on the knowledge of the patterns of spread for each specific disease presentation. Additional reliable information can be obtained from patterns-of-failure analysis and necropsy series.
For most tumors, the CTV will encompass one or more nodal stations, usually near the primary lesion. Traditionally, the location and boundaries of these nodal stations have been established in reference to anatomic landmarks during the standard simulation setup. Therefore, the radiation oncologist has been specifically trained to determine the boundaries of the different nodal stations on standard two-dimensional radiographs, especially in the anteroposterior and posteroanterior views. With the advent of three-dimensional (3D) virtual clinical target definition, the radiation oncologist faces the challenge of defining the CTV on cross-sectional CT or magnetic resonance images. Unfamiliarity with this new technique can make correlations with the known spatial references difficult to establish.
The present nodal atlas is intended to assist radiation oncologists who will use new 3D virtual clinical target definition and treatment planning programs.
|
CLASSIFICATION AND NOMENCLATURE
|
|---|
The anatomic patterns of lymphatic drainage for different organs to their first echelon (or efferent) nodal stations were taken from Rouviere's Anatomy of the Human Lymphatic System (2) and confirmed with other lymphatic anatomy textbooks (3,4). The main and accessory lymphatic routes for different organs that are relevant in radiation oncology are summarized in Tables 1 –8, with an explanation of the abbreviations appearing in the Key Box. When different subsites within an organ had unique drainage patterns, these were individualized in the Tables as well. The classification of and nomenclature for the different nodal areas usually followed the guidelines of Rouviere's system (2). In the classification of the mediastinal nodes, the widely used American Joint Committee on Cancer classification was chosen instead (5). The nodal areas represented are listed in the Key Box. When clinically relevant, some nodal stations were further divided into subgroups, which are noted as lowercase letters after the abbreviation codes provided in the Tables and the Figures. The nonparenthesized lowercase letters indicate differentiated subgroups, usually in the direction of the zz' axis. Parenthesized letters indicate subgroup subdivision, usually in the direction of the xx' axis (shown only for inguinal and external iliac nodes).
|
LOCALIZATION OF NODAL STATIONS
|
|---|
The different nodal stations were outlined and labeled on five different sets of consecutive and equidistant CT images (head and neck, thorax, abdomen, male pelvis, and female pelvis) (Figs 1 –6). We elected to use CT images because they are the customary image support in most 3D virtual clinical target definition programs. The nodal stations on the cross-sectional images were localized by extrapolating information from cross-sectional anatomy atlases (6), lymphatic atlases (3), and vascular atlases (4). To allow easy correlation, the five sets of CT images were connected with a recognizable bone structure on a topogram (Fig 7).
|
REPRESENTATION OF THE ANATOMY OF THE LYMPHATIC SYSTEM IN THE NODAL ATLAS
|
|---|
The initial lymphatic system is composed of capillary lymphatics, which originate in the intima of the tissue and are immersed in the ground substance of the tissue space. These capillaries anastomose in networks to form the peripheral lymphatic plexuses. These plexuses are only represented in the atlas for certain organs that are frequently irradiated while intact (prostate, rectum, some head and neck subsites, etc) to facilitate the recognition of the organ or site and its spatial relationship with the surrounding nodal stations.
The lymphatic plexuses drain to the first echelon lymph nodal stations through precollecting and collecting ducts. Sometimes, there are intercalating lymph nodes in the path of the collecting ducts. In general, no intermediate paths between the organ of interest and the first echelon nodal station have been represented in the atlas to avoid unnecessary complexity. However, some clinically relevant intercalating nodes have been represented (superior and inferior rectal nodes and internal pudendal nodes).
The first nodal station reached by the lymphatic drainage of a given organ is called the first echelon nodal group. The first echelon lymph nodes connect to each other through postlymphonodal collecting ducts and finally drain to more central efferent lymph nodes or directly into the venous system through the main lymphatic trunks, depending on anatomic location. As a rule, the first echelon nodal stations for all the different organs of the head and neck, thorax, abdomen, and pelvis are represented. These are listed in the Tables 1–8. One exception to this rule has been the case of the small bowel and most of the large bowel. Due to the anatomic mobility of these organs, the efferent pre- and paraaortic nodal groups (fixed structures with reproducible location), rather than the first echelon nodal groups (juxtaintestinal and paracolic), are represented in the atlas. Patterns of anomalous nodal spread, such as retrograde spread, are not shown in this atlas.
|
ERRORS IN LOCALIZATION OF NODAL STATIONS
|
|---|
Lymphography is the technique of choice to visualize nodal groups. However, its decline in the field of diagnostic radiology has led to the decline of its use as a tool for radiation therapy planning, and it has been virtually abandoned for both purposes. However, currently used standard radiation ports still follow the boundaries determined during the lymphographic era. During standard simulation, the different nodal stations are not actually seen on the simulation radiographs; therefore, a margin of normal tissue is taken around the CTV to allow for any localization uncertainties. Three-dimensional virtual clinical target definition, like standard simulation, lacks visualization of nodal stations. However, because localization errors on cross-sectional images are minimized (as are the associated increases in CTV uncertainty and size of the planning taget volume), this problem is not as great. During 3D virtual clinical target definition, only a few of the nodal stations represented in the atlas are visible on a CT image. So, the exact location of a given station in an individual is difficult to determine. Furthermore, the internal structure of the lymphatic system (different normal variants among subjects) precludes any categorical statement (2–4). Therefore, we chose to outline wide areas rather than discrete locations for each nodal station to account for the differences in normal anatomic variability and nodal interconnection. Another difficult problem is the mobility of the nodal stations located proximally in the limbs (inguinal, axilla). The location of these nodal groups will vary greatly, depending on simulation positioning. The cross-sectional images provided in this atlas were taken during standard CT positioning and may differ from images taken for radiation planning in special positions. The topograms with bone landmarks may help in correlating the cross sections provided in this atlas with the reference points of the actual patients.
|
SYSTEMATIC VERSUS NONSYSTEMATIC DESCRIPTION
|
|---|
This atlas is strongly biased toward the radiation oncology standpoint. Some nodal areas have been arbitrarily discarded because they are rarely relevant in radiation oncology. We elected to ignore the distal and intermediate nodal stations of the extremities and the nodal stations of muscular groups. Tumors that spread to these nodal stations represent a very small percentage of the overall clinical practice in radiation oncology. We also chose not to represent lymph nodal stations pertaining to mobile structures (mesenteric nodes of the small bowel and most of the large bowel) because of spatial unpredictability. For these anatomic locations, only the efferent pre- and paraaortic nodal groups have been represented.
|
ATLAS LIMITATIONS
|
|---|
This atlas cannot be used as a tool to diagnose or predict nodal involvement. It only provides a guide to the general anatomic pathways of nodal drainage of the normal organs shown. It cannot help forecast individual spread patterns. When used for 3D virtual clinical target definition, this nodal atlas can help in accurately defining the location of the different nodal stations that are to be included in the CTV. Deciding which nodal stations should be included in the CTV depends on the level of spread probability that is clinically assumable for that specific disease manifestation and the normal tissue complication probability for that intended dose level. These two issues are a matter of clinical judgment that is beyond the general purpose of this atlas.
|
POTENTIAL ADVANTAGES
|
|---|
A nodal atlas allows systematic definition of CTVs in those clinical situations in which the irradiation of nodal areas is clinically relevant, which may facilitate reliable intercommunication among institutions. The current tendency, however, is to define CTVs on a nonanatomic basis, probably due to the lack of a common anatomic language. Some ongoing radiation therapy protocols have elected to define the corresponding CTVs as a volume around the GTV. The CTV is then defined mathematically as a margin around the GTV. Although reproducible among institutions, this definition carries the important risk of including in the CTV some areas with a minimal probability of tumor involvement or areas that may be at higher risk for treatment toxicity. A nodal atlas allows a better spatial understanding of the valuable information contained in the surgical pathology reports. For those patients undergoing postoperative radiation therapy, the decision-making process essentially depends on the surgical findings. Improving the ability to locate high-risk areas (as defined by the surgical information) on a cross-sectional image must necessarily improve the quality in treatment planning and delivery. We are currently evaluating the potential for redesigning radiation plans on the basis of nodal information from surgical series and 3D reconstruction with the aid of this atlas.
Finally, the atlas-based 3D definition of CTVs has the potential to improve the therapeutic ratio. Nodal groups with a very low probability of metastatic involvement can be excluded from the CTV, thus expanding beam arrangement possibilities and allowing increased treatment intensity. We are also currently evaluating the volumetric implications (normal tissue dose-volume histograms) of using 3D virtual simulation with either a nodal atlas or standard bone landmarks.
The present atlas should aid the radiation oncologist in accurately locating on cross-sectional images the different nodal stations that will correspond to the chosen clinical target volume.
|
Acknowledgments
|
|---|
The authors thank John Croyle for his help in the preparation of the graphic matrix of the illustrations and David Carpenter for editorial assistance.
|
Footnotes
|
|---|
Abbreviations: CTV = clinical target volume
GTV = gross tumor volume
3D = three-dimensional
Author contributions: Guarantor of integrity of entire study, R.M., R.G.; study design, R.M.; definition of intellectual content, R.M.; literature research, R.M., P.S.F.; data acquisition and analysis, R.M., N.G.; manuscript preparation, R.M.; manuscript review, R.G.
|
References
|
|---|
-
International Commission on Radiation Units and Measurements. Prescribing, recording and reporting photon beam therapy ICRU Report 50. Washington, DC: International Commission on Radiation Units and Measurements, 1993.
-
Rouviere H. Anatomy of the human lymphatic system Ann Arbor, Mich: Edwards, 1938.
-
Donini I, Battezatti M. The lymphatic system Padua, Italy and London, England: Piccin Medical Books, 1972.
-
Uflacker R. Atlas of vascular anatomy Baltimore, Md: Williams & Wilkins, 1997.
-
American Joint Committee on Cancer. AJCC cancer staging manual 5th ed. Philadelphia, Pa: Lippincott-Raven, 1997.
-
Ellis H, Logan B, Dixon A. Human cross sectional anatomy Oxford, England: Butterworth-Heinemann, 1991.
TOP
Abstract
Introduction
