Society of Procedure Guideline for Lymphoscintigraphy and the Use of Intraoperative Gamma Probe for Sentinel Lymph Node Localization in Melanoma of Intermediate Thickness version 1.0, approved June 15, 2002

Authors: Naomi Alazraki, MD (Emory University School of Medicine, Veterans Affairs Medical Center, Atlanta, GA); Ed- win C. Glass, MD (Wadsworth Veterans Affairs Medical Center, Los Angeles, CA); Frank Castronovo, PhD (Harvard Medical School, Brigham and Women’s Hospital, Boston, MA); Renato A. Valdés Olmos, MD (Netherlands Cancer Insti- tute, Amsterdam, The Netherlands); and Donald Podoloff, MD (MD Anderson Cancer Center, Houston, TX).

I. Purpose to make a smaller incision directly over the node, based on the image and probe counts. The purpose of this guideline is to assist nuclear Lymphoscintigraphy images readily demon- medicine practitioners in recommending, perform- strate the unpredictability of lymphatic drainage ing, interpreting, and reporting the results of (1) patterns. Sentinel lymph node biopsy, after iden- lymphoscintigraphy for identifying sentinel lymph tification by lymphoscintigraphy and excision us- nodes for excisional biopsy in patients with ing the intraoperative gamma probe and/or blue melanoma and (2) the use of the intraoperative dye technique, is frequently performed in pa- gamma probe in the operating room. tients without either clinically apparent metas- tases or early intermediate-stage melanoma II. Background Information and Definitions (Clark level <4; Breslow thickness 0.76–4 mm) be- A. This guideline is written specifically for lym- cause of its significant diagnostic and prognostic phoscintigraphy in patients with primary in f o r m a t i o n . melanomas that originate in the skin. Staging of B. Definitions these tumors is based on tumor thickness (Bres- 1. Lymphoscintigraphy: Imaging pathways of low measurement) and level of skin invasion lymphatic flow and lymph nodes after injec- (Clark’s level), both of which are determined by tion of a that is absorbed the pathologist from a biopsy sample. Ample by the lymphatics. data correlate patient survival with Breslow and 2. Sentinel lymph node: The first lymph node in Clark measurements. a lymph node bed to receive lymphatic In the past, elective lymph node dissection drainage from a tumor. Often drainage to (ELND) of the lymphatic bed believed most more than 1 lymph node group and sentinel likely to drain the primary tumor site (based on node is identified. Sappey’s classic anatomic description of cuta- 3. Blue dye technique: Intraoperative injection neous lymphatic flow) was used as part of the (usually peritumoral) of isosulfan blue dye for staging procedure for melanoma. ELND has the purpose of staining lymphatic vessels and been a controversial staging procedure for pa- sentinel lymph nodes so that they can be iden- tients with intermediate (I and II) stage tified visually during surgery for excisional melanoma, because approximately 80% have tu- biopsy. mor-negative lymph nodes and therefore do not 4. Gamma-detecting intraoperative probe: need ELND, a procedure associated with signifi- Small, hand-held -detecting device cant morbidity and cost. The sentinel lymph that uses auditory signals and meter read-outs node excisional biopsy procedure, in contrast, is of counts detected. The intraoperative gamma simpler and not associated with significant mor- probe can be used effectively by the surgeon bidity, provides accurate information about lym- and nuclear medicine physician as a guide to phatic drainage patterns, and allows the surgeon find the radiolabeled sentinel lymph node(s). 136 · LYMPHOSCINTIGRAPHY

III. Common Indications 1. Radiotracer injections a. Approximately 0.1mL containing at least A. Sentinel node localization and excision using ra- 3.7 MBq (100 µCi) Tc-99m sulfur colloid, dionuclide methods are performed in the care of filtered (0.22µ millipore filter), is adminis- patients with: tered as 4–8 peritumoral intradermal injec- 1. Intermediate stage primary melanoma (Bres- tions (fewer than 4 may be performed if ap- low 0.76 mm–4.0 mm). propriate) within 1 cm from the melanoma 2. No clinical evidence of nodal involvement. or the excisional biopsy site at which the 3. No clinical evidence of distant tumor spread. melanoma was located. B. Exclusions may include patients with: b. Injections should surround the lesion or 1. Extensive previous surgery in the region of biopsy site to best sample lymphatic the primary tumor site or targeted lymph drainage in all directions. An exception node bed. may be in primary cutaneous melanoma in 2. Patients with known metastases. the head or neck, in which lymphatic drainage is often caudad. Injections infe- IV. Procedure rior to those lesions may be omitted to avoid masking a sentinel node in close A. Patient Preparation proximity to the primary lesion. There are no dietary or medication restrictions c. The injection should be performed by an au- for the procedure. Patients should follow preop- thorized user physician or his/her designee. erative restrictions if the procedure is performed on the same day as scheduled surgery. d. The high pressure of the intradermal bleb B. Information Pertinent to Performing the Proce- can result in leakage of some of the ra- dure dioactivity on needle removal. Care should The patient usually has had a previous biopsy be taken to avoid radioactive contamina- that showed cutaneous malignancy and the tion, which can be confused with lymph pathological classification of the tumor (Breslow, node “uptake.” thickness of the tumor, and/or Clark, skin-layer e. Gentle finger massage should be consid- penetration of the tumor, which are prognostic). ered at each injection site to promote up- Recent extensive surgical excision of the primary take of the tracer into lymphatic channels lesion can be the cause of a rare unsuccessful and lymphatic flow. study because of surgical disruption of lym- f. The injection site should be covered with a phatic vessel integrity. bandaid or cotton ball to prevent leakage C. Precautions of tracer through the needle puncture site. If surgery is to be performed using the intraop- 2. Dosimetry erative gamma probe to assist in finding the sen- In the dosimetry tables included here, the lo- tinel node, the tracer must be injected approxi- cal radiation dose has been ignored and the mately 0.5–3 h before surgery. If surgery is effective dose has been calculated assuming further delayed (by 6 or more h), another image that 20% of the administered activity has been before surgery is advisable to define further mi- absorbed systemically (Kaplan, 1985). The gration of tracer to additional nodes, if any. reasons for ignoring the local radiation dose D. are that: No radiopharmaceuticals have been specifically a. Deterministic effects (e.g., local skin necro- approved by the U.S. Food and Drug Adminis- sis) are not a concern for Tc-99m–labeled tration for lymphoscintigraphy in the United radiopharmaceuticals. States. Tc-99m sulfur colloid is used in the b. The estimated local radiation dose varies United States in a filtered (usually 0.22-mm fil- greatly, depending on the assumptions tration) or unfiltered form. Smaller particles are used (Castronovo 1994; ICRP, 60). generally recommended as more frequent visu- c. The local radiation dose contributes little to alization of lymphatic channels is achieved with the effective dose. A 50-rem dose averaged the filtered formulation. over 10 cm2 skin is comparable with an ef- There is worldwide variation in radiopharma- fective dose of 1.7 mrems (Baum, 2001). ceuticals used for lymphoscintigraphy. Tc-99m d. Melanoma is very uncommon in children antimony trisulfide colloid (particle size, 0.015–0.3 but lymphoscintigraphy is occasionally µm) and Tc-99m nanocolloid (particle size, performed in children to determine the 0.05–0.8 µm) are used in Australia and Europe. cause of extremity swelling. SOCIETY OF NUCLEAR MEDICINE PROCEDURE GUIDELINES MANUAL JUNE 2002 · 137

Radiation Dosimetry for Adults Radiopharmaceutical Administered Activity Organ Receiving the Effective Dose2 La r g e s t Radiation Dose1 mSv/MBq MB q mGy/MBq (rem/mCi) (m C i ) (r a d / m C i ) Tc-99m small 15 – 35 0.015 0.0019 or large colloids1 Intradermal Spleen (0.5-1.0) (0.057) (0.0071) 1 ICRP 53, pp. 180 and 182. 2 ICRP 80. Note: values in this table are only 20% of the values found in ICRP 80 because of the assumption that only 20% of the administered activity is absorbed systemically.

E. Image Acquisition node(s) in frontal and lateral views, and/or in 1. Acquisition parameters the patient’s position as it would be on the op- Sequential or continuous imaging begins im- erating room table. Skin marking can be per- mediately after completion of injections and formed using the probe in nuclear medicine, if continues for 30–60 minutes. Continuous available, or using a radioactive marker imaging may include dynamic imaging at 30 source. For trunk lesions, images must in- seconds per frame for 2–30 minutes and/or clude both axillary and both inguinal regions. sequential static images every 5 minutes for Some unusual locations for sentinel nodes in- up to 1 hour or until the sentinel lymph node clude periscapular, costal, and umbilical re- is identified. For trunk lesions, images of both gions. The lateral view is essential, especially axillary and inguinal regions should be ob- for distinguishing periscapular from axillary tained. Lateral or oblique views are often nodes. For head and neck lesions, images helpful to uncover multiple nodes that overlie should include the entire head and neck in an- one another on a single projection. The pa- terior, posterior and/or oblique projections tient’s body contour may be defined on the (whichever places the primary tumor and image in relation to the injection site, lymph sentinel node(s) closest to the camera face), drainage, and sentinel node(s) by transmis- and the lateral view. Sentinel lymph nodes sion imaging or by manual tracing with an ex- and lymph drainage are particularly variable ternal source. Other anatomic structures and unpredictable in the head and neck. In marked on the images using external sources, addition to nuclear medicine personnel, the particularly in head and neck images, may be surgeon should understand imaging tech- useful. Transmission images can be obtained niques and projections. using a Co-57 flood source placed so that the 3. In-transit nodes patient is between the camera and the source For extremity lesions, the knee or elbow re- for about 10–15 seconds during dynamic and gions should be included in the field of view static image acquisitions. on dynamic and static images to detect “in- 2. Skin marking transit” (also called “intercalated”) lymph Skin is marked over the imaged sentinel nodes, which are sentinel lymph nodes.

Radiation Dosimetry in Children (5 year old) Radiopharmaceutical Administered Organ Receiving the Effective Dose2 Ac t i v i t y La r g e s t Radiation Dose1 mSv per MBq mB q mGy per MBq (rem per mCi) (m C i ) (rad per mCi) Tc-99m small or 15–35 0.050 0.0036 large colloids1 Intradermal Spleen (0.5-1.0) (0.185) (0.013) 1 ICRP 53, pp. 180 and 182. 2 ICRP 80. Note: values in this table are only 20% of the values found in ICRP 80 because of the assumption that only 20% of the administered activity is absorbed systemically. 138 · LYMPHOSCINTIGRAPHY

F. Intervention the surgeon expeditiously. A brief written report Use of the gamma probe in the operating room: with annotated images should be sent to the op- Using the images and skin markings as guides, erating room with the patient, or verbal commu- the probe (placed centrally over the regions of nication with the surgeon and annotated images highest counts) can be used to select the opti- may be sent with the patient to the operating mum location for incision. The probe is placed in room. If nuclear medicine physicians or technol- a sterile sleeve or glove for intraoperative use in ogists goes to the operating room, they should be the surgical field. The surgeon uses the probe to capable of verbally explaining the imaging re- guide dissection to the hot node(s) and places the sults and skin markings. probe in the surgical bed after node excision to The final report should contain the following confirm removal of the hot node(s). In working information in addition to routinely reported with the probe, it is important to direct the probe items, such as the radiopharmaceutical, dose ad- away from activity at the injection sites. Counts ministered, method of injection, and the imaging are recorded per unit time with the probe in the protocol: operative field, over the node before excision (in 1. Location(s) of sentinel lymph node(s), includ- vivo) and after excision (ex vivo). A background ing in-transit nodes, if present; tissue count is also recorded with the probe 2. Presence of lymph channels, if visualized on pointing away from the injection site, nodal ac- images; tivity, or other physiologic accumulations (i.e., 3. Number of secondary lymph nodes visual- bladder, liver). ized on images; and G. Processing 4. Probe recorded counts (per second) of sen- To optimize visualization of draining lymph tinel node(s) in vivo, ex vivo, and in the surgi- nodes, lowering the upper threshold of the com- cal bed after node excision. puter display or utilizing other contrast enhance- J. Quality Assurance ment methods can enhance the low count areas When external markers are used to assist in skin of the image. marking of sentinel nodes, care should be taken H. Interpretation Criteria to avoid skin contamination. Appropriate cali- 1. Image criteria bration of the probe must be performed accord- a. Dynamic images aid in identifying the sen- ing to manufacturer instructions. tinel lymph node as the first to receive K. Sources of Error drainage from the tumor site. The sentinel 1. Skin contamination. lymph node is not necessarily the hottest 2. Operative positioning different from posi- node, although that is often the case. tioning when overlying skin is marked. b. Separate lymphatic channels that drain to 3. Failure to begin imaging immediately after in- discrete, different lymph nodes identify jection. each of those as distinct sentinel lymph 4. Failure to use image enhancement techniques, nodes, even though they may be located in if warranted to visualize low count areas. the same anatomic region. When drainage 5. Failure to communicate results expeditiously to more than 1 anatomic region is seen (e.g., to the surgeon. axilla and inguinal), each of those regions 6. Mistaking a sentinel node for a secondary must have at least 1 sentinel lymph node. node because it has fewer counts than an- 2. Intraoperative probe criteria other visualized lymph node. a. A sentinel lymph node usually has at least 10 times the background counts, taken at a V. Radiation Safety Considerations location remote from the injection site. b. Various probe criteria have been employed Radiation safety issues arise for operating room per- for identification of the sentinel node (e.g., sonnel, pathology personnel, and nuclear medicine counts per second recorded for the pre- personnel. Nurses, physicians, and technologists in- sumed sentinel node have been compared volved with patients who undergo lymphoscintigra- with nonsentinel nodes in vivo, ex vivo, or phy and intraoperative probe sentinel node localiza- with background counts in vivo). tion studies are instructed to wear radiation badges I. Reporting according to each institution’s policy. Actual expo- Because surgery usually will be performed be- sures from the activity levels used for these proce- fore a typed report is available, it is important dures are sufficiently low that badging is not essen- that the results of this study be communicated to tial for individuals who are not involved in other SOCIETY OF NUCLEAR MEDICINE PROCEDURE GUIDELINES MANUAL JUNE 2002 · 139

radiation procedures. The decision to badge person- Nucl Med 1999;26:84–90. nel whose only radiation involvement is with sen- Castronovo FP, McKusick AK, Strauss HW. Dosimetric tinel node studies lies with the local institution. The consequences of radiopharmaceutical infiltrations. following information is useful in making this de- Invest Radiol 1994;29:59–64. t e r m i n a t i o n . Glass EC, Essner R, Morton DL. Kinetics of three lym- A. The administered radioactivity is <18.5 MBq (500 phoscintigraphic agents in patients with cutaneous µCi), of which approximately 1%, 185 kBq (5 melanoma. J Nucl Med 1998;39:1185–1190. µCi), might migrate to a sentinel lymph node. Hillier DA, Royal HD. Intraoperative gamma radiation B. The radiation dose to the hands of the surgeon detection and radiation safety. In: Whitman ED, has been estimated to be 5–94 µSv (0.5–9.0 mrem) Reintgen D, eds. Radioguided Surgery. Austin, TX: per patient. Relative to the radiation doses hu- Landes Bioscience; 1998:23–38. mans receive in 1 year from cosmic and natural International Commission on Radiological Protection. background sources (about 3 mSv [300 mrem] ef- Radiation Dose to Patients from Radiopharmaceuticals. fective whole body dose), a surgeon could per- ICRP Publication 53. New York, NY: Pergamon form roughly 30–60 melanoma sentinel node Press; 1987. surgeries in a year and not receive as much fin- International Commission on Radiological Protection. ger radiation exposure as that received by the Recommendations of the International Commis- whole body from the natural environmental. sion on Radiological Protection. ICRP Publication C. Radiation doses to pathology personnel who 60. Washington, DC; 1991:1–3. handle the radioactive sentinel node and pri- International Commission on Radiological Protection. mary tumor specimen (including the injection Radiation Doses to Patients from Radiopharmaceu- site) for a limited period would be no greater ticals. ICRP Publication 80. Washington, DC; 1990. than that received by the surgeon. Therefore, the Kaplan WD, Piez CW, Gelman RS, et al. Clinical com- histological specimen can be processed without parison of two radiocolloids for internal mammary delay, and patient care is not compromised. lymphoscintigraphy. J Nucl Med 1985;26:1382–1385. D. from the operating room Logic JR, Balch CM. Defining lymphatic drainage pat- (sponges, etc.) and pathology should be collected terns with cutaneous lymphoscintigraphy in according to institutional radiation safety proce- melanoma patients. In: Human Melanoma: Surgical dures. This waste will also be a biohazard and Treatment, Pathology and Progress. Philadelphia, PA: should be handled accordingly. J.B. Lippincott Company; 1984. Miner TJ, Shriver CD, Flicek PR, et al. Guidelines for the VI. Issues Requiring Further Clarification safe use of radioactive materials during localization and resection of the sentinel lymph node. Ann Surg A. Optimal radiopharmaceutical. Oncol 1999;6:75–82. B. Importance and reproducibility of identifying Mudun A, Murray DR, Herda SC, et al. Early stage the first lymph node to appear on imaging. melanoma: lymphoscintigraphy, reproducibility C. Outcomes for patients using strategies based on of sentinel node detection, and effectiveness of the sentinel node studies. intraoperative gamma probe. R a d i o l o g y 1 9 9 6 ; 1 9 9 : 1 7 1 – 1 7 5 . VII. Concise Bibliography Murray DR, Carlson GW, Greenlee R, et al. Surgical management of malignant melanoma using dy- Alazraki N, Eshima PD, Eshima LA, et al. Lym- namic lymphoscintigraphy and gamma probe- phoscintigraphy, the sentinel node concept, and the guided sentinel lymph node biopsy: the Emory ex- intraoperative gamma probe in melanoma, breast perience. Am Surg 2000;66:763–767. cancer, and other potential cancers. Semin Nucl Med Statius Muller MG, van Leeuwen PA, Borgstein PJ, et 1997;27:55–67. al. The sentinel node procedure in cutaneous Baum JW. Analysis of potential radiobiological effects melanoma: an overview of 6 years’ experience. Eur related to a unified skin dose limit. Health Phys J Nucl Med 1999;26:S20–S25. 2001;80:537–543. Uren RF, Howman-Giles RB, Shaw HM, et al. Lym- Bennett LR, Lago G. Cutaneous lymphoscintigraphy in phoscintigraphy in high-risk melanoma of the malignant melanoma. Semin Nucl Med. 1 9 8 3 ; trunk: predicting draining node groups, defining 13:61–69. lymphatic channels and locating the sentinel node. Bongers V, Borel Rinkes IH, Barneveld PC, et al. To- J Nucl Med 1993;34:1435–1440. wards quality assurance of the sentinel node proce- Waddington WA, Keshtgar MR, Taylor I, et al. Radia- dure in malignant melanoma patients: a single in- tion safety of the sentinel lymph node technique in stitution evaluation and a European survey. Eur J breast cancer. Eur J Nucl Med 2000; 27:377–391. 140 · LYMPHOSCINTIGRAPHY

VIII. Disclaimer be quite different than the spectrum of patients seen in a more general practice setting. The appropriate- The Society of Nuclear Medicine has written and ness of a procedure will depend in part on the preva- approved guidelines to promote the cost-effective lence of disease in the patient population. In addi- use of high quality nuclear medicine procedures. tion, the resources available to care for patients may These generic recommendations cannot be applied vary greatly from one medical facility to another. For to all patients in all practice settings. The guidelines these reasons, guidelines cannot be rigidly applied. should not be deemed inclusive of all proper proce- Advances in medicine occur at a rapid rate. The dures or exclusive of other procedures reasonably date of a guideline should always be considered in directed to obtaining the same results. The spectrum determining its current applicability. of patients seen in a specialized practice setting may