Whitman, Reintgen - Radio Guided Surgery (Landes Bio Science Vademecum

LANDES LANDES
BIOSCIENCE V ade me c um BIOSCIENCE V ade me c um

Table of contents
1. Logistics and Organizational Aspects of
Radioguided Surgery Programs
8.
9.
Lymphoscintigraphy
Pathologic Evaluation of Sentinel Lymph

Radioguided
2. Intraoperative Gamma Radiation Detection Nodes in Breast Cancer
3.
and Radiation Safety
Training and Credentialing Physicians in

Radioguided Surgery
10.
11.
Nodes in Malignant Melanoma
The Technique of Minimally Invasive

Surgery
4. Melanoma Lymphatic Mapping: Scientific Radioguided Parathyroidectomy (MIRP)
Support for the Sentinel Lymph Node
12. Emerging Applications in Other Skin
Concept and Biological Significance of the
Sentinel Node Cancers
5. Sentinel Lymph Node Biopsy for 13. Radioguided Surgery and Vulvar
Carcinoma
Melanoma: Surgical Technique
14. Bone Lesion Localization Intraoperative
6. Technique for Lymphatic Mapping in
Breast Carcinoma Deaths
7. The Technique of Intraoperative Lymphatic

Mapping and Sentinel Lymphadenectomy
in Breast Cancer Using Blue Dye Alone
This is one of a new series of medical handbooks.

It includes subjects generally not covered in other handbook series, especially
many technology-driven topics that reflect the increasing influence of technology
in clinical medicine.
The name chosen for this comprehensive medical handbook series is Vademecum,
a Latin word that roughly means “to carry along”. In the Middle Ages, traveling
clerics carried pocket-sized books, excerpts of the carefully transcribed canons,
known as Vademecum. In the 19th century a medical publisher in Germany, Samuel
Karger, called a series of portable medical books Vademecum.
The Landes Bioscience Vademecum books are intended to be used both in the
training of physicians and the care of patients, by medical students, medical house
staff and practicing physicians. We hope you will find them a valuable resource.
Eric D. Whitman
All titles available at
www.landesbioscience.com
Douglas Reintgen
V A D E M E C U M
Radioguided Surgery
Eric D. Whitman, M.D.

Director, The Melanoma Center
Saint Louis, Missouri, U.S.A.
Douglas Reintgen, M.D.

Program Leader, Cutaneous Oncology
H. Lee Moffitt Cancer Center
and
Professor of Surgery
University of South Florida
Tampa, Florida, U.S.A.
L ANDES
B I O S C I E N C E
AUSTIN, TEXAS
U.S.A.
VADEMECUM
Radioguided Surgery
LANDES BIOSCIENCE
Austin
Copyright © 1999 Landes Bioscience

All rights reserved.
No part of this book may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any
information storage and retrieval system, without permission in writing from
the publisher.
Printed in the U.S.A.
Please address all inquiries to the Publishers:

Landes Bioscience, 810 S. Church Street, Georgetown, Texas, U.S.A. 78626
Phone: 512/ 863 7762; FAX: 512/ 863 0081
ISBN: 1-57059-569-0
Development of this book was supported in part by an unrestricted educa-

tional grant from U.S. Surgical Corporation.
Library of Congress Cataloging-in-Publication Data

Radioguided surgery / [edited by] Eric D. Whitman, Douglas Reintgen.
p. cm. -- (Vademecum)
Includes bibliographical references and index.
ISBN 1-57059-569-0
1. Breast--Cancer--Radiotherapy. 2. Breast--Cancer--Surgery. 3. Lymph
nodes--Diseases--Diagnosis. I. Whitman, Eric D. II. Reintgen, Douglas Scott.
III. Series.
[DNLM: 1. Surgical Procedures, Operative. 2. Radionuclide Imaging. WO
500 R129 1999]
RD667.5.R24 1999
617'.05--dc21
DNLM/DLC 98-53305
for Library of Congress CIP
While the authors, editors, sponsor and publisher believe that drug selection and dosage
and the specifications and usage of equipment and devices, as set forth in this book, are in
accord with current recommendations and practice at the time of publication, they make
no warranty, expressed or implied, with respect to material described in this book. In
view of the ongoing research, equipment development, changes in governmental regula-
tions and the rapid accumulation of information relating to the biomedical sciences, the
reader is urged to carefully review and evaluate the information provided herein.
Contents
1. Logistics and Organizational Aspects
of Radioguided Surgery Programs ...................................... 1
Eric D. Whitman
Introduction ....................................................................................... 1
Organizational Issues ......................................................................... 2
Initiating a Program ........................................................................... 2
Nuclear Medicine ............................................................................... 5
Pathology ............................................................................................ 7
Scheduling a Procedure ................................................................... 10
Purchasing a Probe ........................................................................... 13
Outcomes Management ................................................................... 14
Data Acquisition, Storage and Analysis ........................................... 15
Billing and Reimbursement ............................................................. 18
Conclusions ...................................................................................... 21
2. Intraoperative Gamma Radiation Detection
and Radiation Safety .......................................................... 23
David A. Hillier, Henry D. Royal
Introduction ..................................................................................... 23
Basic Radiation Terminology .......................................................... 23
Radiation Detection ......................................................................... 26
Gamma Probe Testing ...................................................................... 26
Proper Gamma Probe Use ............................................................... 29
Radiation Safety and Regulations ................................................... 33
3. Training and Credentialing Physicians
in Radioguided Surgery ..................................................... 39
Eric D. Whitman
Introduction ..................................................................................... 39
Training ............................................................................................ 39
Didactic Learning ............................................................................. 41
Experiential Learning ....................................................................... 42
Credentialing .................................................................................... 42
Summary .......................................................................................... 45
4. Melanoma Lymphatic Mapping: Scientific Support
for the Sentinel Lymph Node Concept
and Biological Significance of the Sentinel Node ............. 47
Merrick I. Ross, Jeffrey E. Gershenwald
Introduction ..................................................................................... 47
Scientific Support for the Sentinel Node Concept ......................... 48
Technical Advances .......................................................................... 50
Biologic Significance of the Sentinel Node ..................................... 52
Accurate Nodal Staging .................................................................... 55
Clinical Trials .................................................................................... 57
Concluding Comments .................................................................... 60
5. Sentinel Lymph Node Biopsy
for Melanoma: Surgical Technique .................................... 63
Fadi F. Haddad, Douglas Reintgen
Introduction ..................................................................................... 63
The Role of Nuclear Medicine—Surgical Perspective .................... 63
Description of the Technique of Intraoperative
Lymphatic Mapping ......................................................................... 65
Role of Pathology ............................................................................. 68
Updated Results of Lymphatic Mapping for Melanoma
from Our Institution ........................................................................ 70
Conclusion ....................................................................................... 70
6. Technique for Lymphatic Mapping
in Breast Carcinoma ........................................................... 72
Charles E. Cox
Introduction: Economic Imperatives .............................................. 72
Status of Axillary Node Dissection: Historical Overview .............. 73
Historical Perspective of the Development
of Lymphatic Mapping for Breast Cancer ....................................... 73
Technical Aspects of Lymphatic Mapping
for Breast Carcinoma ....................................................................... 74
Operative Technique for Lymphatic Mapping
of Breast Carcinoma ........................................................................ 76
Gamma Detection ............................................................................ 78
Summary .......................................................................................... 80
7. The Technique of Intraoperative Lymphatic Mapping
and Sentinel Lymphadenectomy in Breast Cancer
Using Blue Dye Alone ......................................................... 83
Philip I. Haigh, and Armando E. Giuliano
Introduction ..................................................................................... 83
Detailed Technique .......................................................................... 84
Complications .................................................................................. 88
Summary .......................................................................................... 89
8. Lymphoscintigraphy .......................................................... 90
Claudia G. Berman
Introduction ..................................................................................... 90
Dose Administration ........................................................................ 93
Imaging Protocol .............................................................................. 95
Radioguided Surgery ..................................................................... 101
Summary ........................................................................................ 105
9. Pathologic Evaluation of Sentinel Lymph Nodes
in Breast Cancer ................................................................ 106
Ni Ni K. Ku
Introduction ................................................................................... 106
Pathology Protocol ......................................................................... 107
Pathologic Findings ....................................................................... 108
Results ............................................................................................. 111
Conclusion ..................................................................................... 113
10. Pathologic Evaluation of Sentinel Lymph Nodes
in Malignant Melanoma ................................................... 115
Jane L. Messina
Introduction ................................................................................... 115
Review of the Literature ................................................................. 115
Pathologic Protocol for Handling Sentinel Nodes ....................... 116
Results ............................................................................................. 118
Discussion ...................................................................................... 121
11. The Technique of Minimally Invasive
Radioguided Parathyroidectomy (MIRP) ....................... 126
James Norman
Patient Selection and Anticipated Results ..................................... 127
Optimal Timing Between Sestamibi
and Operative Exploration .................................................... 127
Use of the Gamma Probe in the Operating Room ....................... 130
Summary ........................................................................................ 132
12. Emerging Applications in Other Skin Cancers ............... 133
C. Wayne Cruse
Introduction ................................................................................... 133
Squamous Cell Carcinoma ............................................................ 133
Merkel Cell Carcinoma .................................................................. 137
Technical Considerations ............................................................... 139
Summary ........................................................................................ 140
13. Radioguided Surgery and Vulvar Carcinoma ................. 142
Edward C. Grendys, Jr., James V. Fiorica
Introduction ................................................................................... 142
Epidemiology ................................................................................. 143
Anatomy and Lymphatic Drainage ............................................... 144
Surgical Management and Staging of Vulvar Carcinoma ............ 144
Intraoperative Lymphatic Mapping for Vulvar Carcinoma ......... 147
Technique of Vulvar Lymphoscintigraphy .................................... 147
Vulvar Melanoma ........................................................................... 148
Conclusion and the Future ............................................................ 148
14. Bone Lesion Localization ................................................. 153
Lary A. Robinson
Introduction ................................................................................... 153
Bone Metastases ............................................................................. 153
Radioisotope Bone Scan ................................................................ 154
Techniques of Bone Lesion Localization ....................................... 155
Conclusions .................................................................................... 161
Index .................................................................................. 163
Editors
Eric D. Whitman, M.D.
Director, The Melanoma Center
St. Louis, Missouri, U.S.A.
Chapters 1, 3
Douglas Reintgen, M.D.

Program Leader, Cutaneous Oncology
Moffitt Cancer Center
Chapter 5
Contributors
Claudia G. Berman, M.D. Jeffrey E. Gershenwald, M.D.
Associate Professor of Radiology Assistant Professor of Surgery
University of South Florida Department of Surgical Oncology
Tampa, Florida, U.S.A. The University of Texas M. D. Anderson
Chapter 8 Cancer Center
Houston, Texas
Charles E. Cox, M.D. Chapter 4
University of South Florida Armando E. Giuliano, M.D.
Program Leader Chief of Surgical Oncology
Comprehensive Breast Cancer Center John Wayne Cancer Institute
Moffitt Cancer Center Santa Monica, California, U.S.A.
Tampa, Florida, U.S.A. Chapter 7
Chapter 6
Edward C. Grendys, Jr., M.D.
C. Wayne Cruse, M.D. Assistant Professor
Professor of Surgery Gynecologic Oncology
Division of Plastic Surgery H. Lee Moffitt Cancer Center
University of South Forida University of South Florida
Tampa, Florida, U.S.A. Tampa, Florida, U.S.A.
Chapter 12 Chapter 13
James V. Fiorica, M.D. Fadi F. Haddad, M.D.

Professor, Gynecologic Oncology Clinical Research Fellow
H. Lee Moffitt Cancer Center Cutaneous Oncology Program
University of South Florida Moffitt Cancer Center
Tampa, Florida, U.S.A. University of South Florida
Chapter 13 Tampa, Florida, U.S.A.
Chapter 5
Philip I. Haigh, M.D., FRCS(C) Lary A. Robinson, M.D.
Surgical Oncology Fellow Director, Division of Cardiovascular
John Wayne Cancer Institute and Thoracic Surgery
Santa Monica, California, U.S.A. Thoracic Oncology Program
Chapter 7 H. Lee Moffitt Cancer Center
David A. Hillier, M.D.- Ph.D. Tampa, Florida, U.S.A.
Instructor of Radiology, Washington Chapter 14
University School of Medicine
Mallinckrodt Institute of Radiology Merrick I. Ross, M.D., F.A.C.S.
St. Louis, Missouri, U.S.A. Associate Professor of Surgery
Chapter 2 Chief, Melanoma/Sarcoma Section
Department of Surgical Oncology
Ni Ni K. Ku, M.D. The University of Texas M. D. Anderson
Associate Professor of Pathology Cancer Center
H. Lee Moffitt Cancer Center Houston, Texas
University of South Florida Chapter 4
Chapter 9 Henry D. Royal, M.D.
Professor of Radiology, Washington
Jane L. Messina, M.D. University School of Medicine
Associate Professor of Pathology Associate Director,
University of South Florida Division of Nuclear Medicine,
Tampa, Florida, U.S.A. Mallinckrodt Institute of Radiology
Chapter 10 St. Louis, Missouri, U.S.A.
Chapter 2
James Norman, M.D.
Associate Professor of Surgery
and Internal Medicine
Director of Endocrine Surgery
Chapter 11
Preface
Radioguided surgical procedures, including sentinel node map-
ping and biopsy, are currently taking the American surgical com-
munity by storm. The explosive yet poorly measured growth in these
techniques is perhaps best indicated by the proliferation of clinical
and basic science research papers presented on the topic at meet-
ings such as the American College of Surgeons and the Society of
Surgical Oncology.
In preparing this handbook, we have endeavored to have experts
in their fields describe exactly what they do, how they do it, and
why. Our goal is to create an indispensable resource for the practic-
ing clinician: the ultimate arbiter of technical and logistical ques-
tions about radioguided surgical procedures. We have tried to avoid
overly academic or review-oriented descriptions of the results from
clinical centers of innovation. We prefer instead to focus on the fu-
ture practitioners of sentinel node biopsy and other techniques: the
practicing surgeon and other clinicians who will continue to pro-
vide the majority of clinical care for breast cancer and other dis-
eases and who therefore must be able to safely and effectively per-
form these procedures for their patients.
We hope that you find this handbook essential to your clinical
practice, and we invite your comments, suggestions, or questions.
Eric D. Whitman
Douglas Reintgen
Logistics and Organization 1
Logistics and Organizational Aspects 1

of Radioguided Surgery Programs
Eric D. Whitman
Introduction ............................................................................................................ 1
Organizational Issues ............................................................................................. 2
Initiating a Program ............................................................................................... 2
Nuclear Medicine .................................................................................................... 5
Patholo gy ................................................................................................................ 7
Scheduling a Procedure ......................................................................................... 10
Purchasing a Probe ............................................................................................... 13
Outcomes M anagement ........................................................................................ 14
Data Acquisition, Storage and Analysis ................................................................ 15
Billing and Reimbursement .................................................................................. 18
Conclusions ........................................................................................................... 21
INTRODUCTION
Initiating a radioguided surgical program involves much more than just learn-
ing the technical aspects of the procedure(s) and purchasing an intraoperative
gamma probe. Upon review of some of the major publications in the field,1-5 I
believe there are certain nonscientific issues that stand out as essential, yet be-
hind-the-scenes; the infrastructure of successful radioguided surgical (RGS) pro-
grams. For example, patient scheduling for procedures is immensely more com-
plicated, because it will routinely involve the coordination of multiple depart-
ments within the hospital, including the operating room, nuclear medicine, and
pathology. This coordination must function smoothly and precisely to enable the
procedure to be performed within the ideal window after radionuclide injection,
with timely pathologic evaluation (which likely involves specialized personnel).
The pathologic evaluation itself is critical; the data for sentinel node biopsy is
predicated upon the lowest possible false negative rate when compared to the pa-
thology of the remainder of the nodal chain. If you cannot ensure that you are
taking all possible steps to minimize the false negative rate, your institution should
not be performing sentinel node biopsy, since you are ultimately doing a disser-
vice to the patient because you are not providing the same quality information as
that published in the medical literature. In order to achieve this, a comprehensive
Radioguided Surgery, edited by Eric D. Whitman and Douglas Reintgen.

© 1999 Landes Bioscience
2 Radioguided Surgery
and consistent pathology evaluation algorithm employing the techniques described

1 elsewhere in this handbook must be implemented.
Many of the recommendations in this chapter cannot unfortunately be based
on published guidelines; rather they have been synthesized by me during my ex-
perience as the organizer of radioguided surgical programs at two philosophically
and geographically distinct hospitals, the first a large urban university medical
center and the second a medium sized suburban community hospital. This chap-
ter, therefore, focuses on the logistics behind a successfully implemented radio-
guided surgical program.
ORGANIZATIONAL ISSUES
The most important organizational issues that I have encountered in my in-

volvement with RGS programs at two very different institutions are the need to
develop a consistent set of standardized procedures for all components of the pro-
gram, and to have agreement between all involved parties (i.e., pathology, nuclear
medicine, surgery, administration, etc.) on how these policies and procedures will
be implemented prior to scheduling the first sentinel node case. These organiza-
tional issues can be divided into Program Initiation, Nuclear Medicine, Pathol-
ogy, and Scheduling components.
INITIATING A PROGRAM
In most institutions, surgeons have been the driving force behind initiating
these procedures, usually after being trained at one of the specialized courses. As
discussed in another chapter in this handbook, many courses do not appear to
adequately prepare attendees for the logistical hurdles they must negotiate upon
their return to their home institutions.
This begins with surgeon training and credentialing. The hospital, with input
from interested physicians, should decide before any RGS is performed who would
be allowed to perform what procedures, and in which situations. This ideal situa-
tion is often not possible for a variety of reasons. Alternative solutions are dis-
cussed in chapter 3 (“Training and Credentialing”). The important point is that
some regulation of RGS should initially occur to prevent inappropriate usage of
the advanced technology in ways that might ultimately harm patients. These regu-
lations should include training requirements, credentialing, and outcome mea-
surements and analysis, to ensure that the patients are receiving a level of care
consistent with published results from more experienced referral centers.
The involved physicians must also decide which RGS procedures will be per-
formed and whether these procedures are best performed under a research proto-
col with Institutional Review Board (IRB) approval. The decision to seek IRB ap-
proval for some or all of the RGS operations is to an extent dependent upon the
comfort and training level of the individual surgeon, the clinical research interests
of the involved physicians, and the specific requirements of the institution. Most
institutions today are not obtaining IRB approval for sentinel node biopsy for 1
melanoma, unless there are specific research interests being pursued. Although
there is an ongoing clinical study evaluating SLN biopsy for melanoma coordi-
nated through the John Wayne Cancer Institute,6 there have been other publica-
tions and programs that strongly state that SLN biopsy for melanoma is the stan-
dard of care.7,8 Unlike breast cancer, standard surgical treatment of melanoma
does not include lymph node dissection, despite multiple attempts to justify “pro-
phylactic” lymph node dissection in recent years.9 However, the recent approval
of interferon alfa-2b (Intron A®, Schering Plough Inc., Madison, New Jersey) to
treat patients at high risk for recurrent melanoma, especially those with lymph
node positive disease, has put a premium on the early detection of nodal me-
tastases in patients with at least intermediate thickness melanoma. The highest
level of sensitivity to detect nodal disease is currently available only through SLN
sampling, provided the pathologic examination meets the criteria discussed else-
where in this chapter and others. Previously, there was no treatment that was proven
effective for patients at high risk of melanoma recurrence, but a long-term
multicenter randomized prospective trial (ECOG 1684) established Intron A as
the only approved adjuvant therapy for patients at risk.10 Thus, SLN biopsy for
melanoma detects nodal dissemination at an early (clinically occult) time point,
while enabling the separation of those patients whose high-risk primary lesions
have apparently not yet metastasized. Further, SLN biopsy eliminates the logisti-
cal need for prophylactic lymph node dissections, converting all lymph node dis-
sections in melanoma patients to “therapeutic.” For these reasons, and the relative
ease with which most of the melanoma SLN are located, IRB approval is not gen-
erally pursued by most physicians performing SLN biopsy for patients with this
cancer.
The situation is much different for SLN biopsy for breast cancer, a disease far
more prevalent in the United States than melanoma. Due to its more common
occurrence, it is more likely to be treated by nonspecialist physicians in a commu-
nity setting than melanoma. Conversely, many melanoma patients in the United
States and elsewhere are referred relatively early in their treatment to specialists in
the field, who will also be more experienced with SLN biopsy techniques. Con-
sidering the differences in disease prevalence and typical referral patterns in this
country, most potential RGS procedures for general surgeons are likely to be for
patients with newly diagnosed breast cancer. This scenario presents several prob-
lems. First, the technical aspects of SLN biopsy for breast cancer are generally
considered to be more difficult than the same procedure for melanoma. The rea-
sons for this difference are covered elsewhere in the handbook. Second, there are
fewer published data regarding SLN biopsy for breast cancer than for melanoma,
and the data published reflect the extended learning curve and greater potential
for failure associated with SLN biopsy for breast cancer. Third, breast cancer care
in the United States occurs within a politically charged environment, amplifying
any mistakes while simultaneously encouraging rapid adoption of any technique
offering potential advantages.
I strongly believe that all breast cancer SLN procedures should at least initially
1 be done under an IRB approved protocol at this time (early 1999). An ideal single-
armed protocol should specify that all patients would undergo SLN biopsy fol-
lowed by standard axillary node dissection, regardless of the pathology of the sen-
tinel node. The purposes, in my mind, of performing all SLN biopsies for breast
cancer under IRB approval are:
1) to ensure that the sentinel node biopsies are performed with consistent
technique by surgeons trained in these procedures,
2) to provide consistent nuclear medicine and pathologic examination
protocols and algorithms for all breast cancer SLN biopsies,
3) to comprehensively monitor clinical outcomes from the procedures, and
analyze these outcomes to prove statistically significant equivalence be-
tween your institution’s results and those published in the literature from
major referral centers.
Thus, the ideal IRB approved protocol will enable the investigator to state at its
predetermined accrual endpoint that SLN biopsy for breast cancer at that institu-
tion, performed by the surgical co-investigators and processed by the nuclear
medicine physicians and pathologists according to the criteria and algorithms
specified to the IRB, is statistically equivalent to SLN biopsy techniques and re-
sults published in the literature. Further, based on the high quality outcomes ob-
tained, that particular institution and clinicians are now able to offer SLN biopsy
without routine axillary node dissection, as the procedure has evolved at more
experienced centers.
The American College of Surgeons has obtained National Cancer Institute fund-
ing for clinical trials in surgical oncology (ACSCOG program).6 One of the first
trials will be a multicenter examination of sentinel node biopsy for breast cancer.
As with any large trial, there will be pros and cons to participating, and most
hospitals/physicians will not. This is not meant to make any statement regarding
that trial; that is simply how these situations typically evolve. Nonetheless, the
positive media reports of sentinel node biopsy for breast cancer will only acceler-
ate, placing significant pressure on all physicians treating this disease to offer SLN
biopsy sooner rather than later. In this setting, I have written a complete, generic,
non-institution specific, protocol for breast cancer SLN biopsy, meeting the crite-
ria and schematic examples set forth above. This protocol and consent form can
be obtained free of charge by e-mail from me at [email protected]. This
protocol is designed so that it may be submitted as is to any IRB after the name of
the principal and any associate investigators have been typed onto the front page.
This protocol is not designed to be a replacement for the ACSCOG trial; it is merely
a way for all institutions and physicians to safely and consistently begin to per-
form SLN biopsy for breast cancer.
Performance of other RGS procedures, particularly minimally invasive radio-
guided parathyroidectomy, without IRB approved protocol monitoring is less well
defined, as these procedures to date have been less commonly utilized. However,
the decision to perform any RGS procedure with IRB approval must only be made
with the recognition that part of what IRB approval provides is medicolegal pro-
tection, by ensuring that all participants sign a detailed informed consent and
understand the investigational nature of their therapy. Performing any RGS pro- 1
cedure without an investigational protocol exposes the clinician to this medicole-
gal risk, and demands full disclosure to the patient of the procedural components,
published literature, personal clinician experience, and risk-benefit comparison.
NUCLEAR MEDICINE
Lymphoscintigraphy for SLN procedures is performed using standard tech-

niques and protocols, well described in chapter 8. Most radiologists are at mini-
mum familiar with, if not experienced in, this procedure and the technical aspects
of lymphoscintigraphy itself should be expected to go well. There are, however,
several logistical issues to be considered from the perspective of coordinating
nuclear medicine activities with other clinical specialties within a RGS program.
First, it is important to discuss the purposes of lymphoscintigraphy with the
nuclear medicine physicians before initiating the RGS program. Unlike a stan-
dard lymphoscintigram, a study performed on the day of SLN surgery must meet
several constraints, mostly to facilitate scheduling by the surgeon and operating
room. As discussed in a subsequent section of this chapter, the schedule a patient
follows on the day of SLN surgery requires multiple departments within the hos-
pital to function within fixed time constraints. This type of scheduling is atypical
for most nuclear medicine studies. We have had to reinforce to the nuclear medi-
cine physicians, technologists, and administrative staff our scheduling priorities.
From the surgical standpoint, the essential elements of the sentinel lymph node
lymphoscintigraphy are:
1) identify where the sentinel node(s) is/are (basin only, see below);
2) scan all basins at risk to confirm which do not have SLN within them;
and,
3) deliver the patient to the operating room holding area with copies of
the relevant scans so as to not delay the start of the scheduled operation.
We do not ask that the radiologists mark or tattoo the skin over the SLN, un-
like many other programs. We have found that this requires extra time in the nuclear
medicine suite and also unfortunately may be misleading to the surgeon. The po-
sition of the patient during lymphoscintigraphy is likely to be different than the
position during the operation, causing the skin marking to move relative to the
underlying SLN once the patient is fully positioned, prepped and draped. I have
personally made (and have seen others make) misplaced incisions because of well
intentioned but ultimately inaccurate skin markings. Also, marking the skin loca-
tion of the SLN within a standard, routinely dissected lymphatic basin such as the
axillary or inguinal locations is superfluous; all surgeons should be experienced
enough to locate an axillary sentinel node without the assistance of a skin mark.
With the patient completely positioned for the procedure on the operating room
table, I make my own skin marks, using the intraoperative gamma probe to inter-
rogate the skin overlying the nodal basin identified by the lymphoscintigraphy,
before prepping and draping the area. This is particularly important in areas other
1 than the axilla or groin, such as the head and neck, flank, and scapula, where both
lymph node location and patient positioning are more variable.
Breast SLN lymphoscintigraphy is slightly different, as the melanoma SLN tend
to “light up” more quickly and with more intensity, enabling melanoma patients
to complete their scans sooner. Breast cancers have a more predictable drainage
pattern, so that typically only one view is necessary, of the ipsilateral chest and
axilla. The breast SLN tend not to concentrate as much radioactivity, which may
make them more difficult to locate on lymphoscintigraphy, particularly in pa-
tients with upper outer quadrant lesions, where proximity to the high concentra-
tion of radionuclide at the primary site may obscure the SLN on the scan.
Early in our experience, we had significant problems scheduling breast SLN
lymphoscintigraphy and surgery, due to difficulties obtaining satisfactory (using
standard, non-SLN criteria) images which led to large delays in transporting the
patient to the operating room. We have discussed these issues at length with our
nuclear medicine physician colleagues and have arrived at compromises designed
to enable high quality lymphoscintigraphy to coexist with the scheduling con-
straints of the surgeon and the operating room. Following injection of the radio-
nuclide at the primary site in the breast, the patient is scanned once, at about 75
minutes. The scanned area encompasses the ipsilateral chest, showing the axilla,
supraclavicular, and internal mammary nodal chains. The patient and a copy of
the scan, regardless of the findings, are then sent to the operating room. We have
found (unpublished data) that nodes are unlikely to show up on scans if they do
not appear by this time point. Further, we utilize both the blue dye and the radio-
nuclide to localize the breast SLN and have found these techniques complemen-
tary, particularly in cases where no lymph nodes are visualized by lympho-
scintigraphy. In our experience (unpublished data), SLN can still be localized fol-
lowing a “negative” lymphoscintigram, either by the blue dye alone and/or with
the gamma probe in cases where the primary injection site obscures the view of
the axilla by the scanner. During the time between radionuclide injection and
scans, our breast SLN patients are transported to admitting or preoperative hold-
ing areas, as necessary, to complete their pre-surgical evaluation and registration,
more effectively utilizing their time in the hospital preoperatively.
For minimally invasive parathyroidectomy, we have asked our radiologists to
follow the protocol described by the group at the University of South Florida,
available on the internet at http://endocrineweb.com. We have emphasized the
timing issues, as described by Dr. Norman,11 where the maximal separation of
counts in the parathyroid and thyroid glands should be between 2 and 4 hours. It
is, therefore, imperative that the patient’s procedure begin somewhere in this time
frame.
Overall, we have found the nuclear medicine physicians very responsive to our
needs, regardless of the institution involved. However, it has been very rewarding
for us and our patients to establish the operative and scheduling constraints of
lymphoscintigraphy and RGS with them, so that the scans are performed in a
consistent manner that maximizes the information provided to the surgeon while
minimizing potential disruption of surgical schedule and acquisition of unneces-

sary data. 1
PATHOLOGY
The use of sentinel lymph node biopsy for cancer staging is based on the hy-
pothesis, subsequently supported by data from multiple centers in several differ-
ent cancers, that the SLN is the most informative node for that primary cancer;
that if the cancer has regionally metastasized, it will be to that node first, and
finally, if the SLN is not pathologically positive, no other lymph node in the pa-
tient can have any evidence of metastasis. The pathologic examination of the
SLN may be the key component testing this hypothesis, to ensure that adequate
evaluation of the sentinel nodes are performed, while proceeding appropriately
and cost effectively in this era of cost containment in medical care. This area has
proved to be a struggle in many centers, as there is no absolute proof that earlier
detection of cancer metastases, particularly at the microscopic level, contributes
to patient prognosis in melanoma or breast cancer. Concerns have also been raised
as to whether a pathologist/institution can be reimbursed for the extra time, re-
agents, and testing necessary to perform serial sections and immunohistochemi-
cal stains on some or all of the SLN harvested for each patient. Our SLN program
was initially encumbered by these same concerns, but upon review of the litera-
ture it was clear that all of the leading centers of innovation utilized an aggressive
pathologic evaluation scheme, involving both serial section analysis and immu-
nohistochemical staining. If we did not implement a similar pathologic al-
gorithm, would we be providing the same level of prognostic information to the
patient or referring physician?
I concluded, based on discussions with physicians at the pioneering centers in
RGS, that it would be impossible for us to have confidence in our SLN pathology
results without performing serial section and immunohistochemical analysis, since
we would likely miss micrometastatic disease, increasing our regional recurrence
risk. This conclusion has been reinforced by several recent publications. In the
first instance, all melanoma patients with recurrence in the nodal basin after nega-
tive SLN biopsy had the tissue blocks reexamined with serial section and immu-
nohistochemical analysis. An overwhelming percentage of the cases would have
been pathologically positive if the more comprehensive/aggressive approach had
been taken initially.12 Another report confirmed the utility of a more comprehen-
sive pathologic examination, concluding that about 40% of patients with
micrometastatic disease would be “missed” without serial sections and immuno-
histochemical staining (see chapter 10). Finally, researchers at John Wayne Cancer
Institute examined the validity of the more comprehensive pathologic algorithm
to examine SLN in breast cancer by also subjecting the non-SLN to this aggressive
pathologic examination protocol. In 1087 non-SLN, only 1 node had evidence of
micrometastatic disease not found on routine bivalved H&E examination.13 This
underscores both the need for serial sections and immunohistochemical staining
1 Sentinel Node Pathologic Evaluation

Breast Cancer
Suburban Surgical Associates
Missouri Baptist Hospital
Sentinel Node(s) Obtained BJC Health Systems
St. Louis, Missouri
Place in saline and

transport fresh to May, 1998
Pathology Eric Whitman, MD
Charles Short, MD
Serial sections cut ("Bread

Loaf", every 2-3 mm)
Scrape prep or touch prep Results conveyed to

of bisected node surgeon in operating room
H & E exam of central slides

(i.e. as if bisected) only
no H & E of all
Positive? sections
yes
Exam no cytokeratin immunohistochemistry

Positive?
complete of all sections
yes
Positive Exam
complete Positive?
Result
yes no
Positive Exam Exam

Result complete complete
Positive Negative
Result Result
Fig. 1.1. Sentinel node pathologic evaluation.
of the SLN and the apparent safety of not performing as comprehensive an exami-
nation of the non-SLN.
During implementation of a single-armed IRB approved study of SLN biopsy
for breast cancer at our institution, we included an algorithm for comprehensive
pathologic examination of the SLN, shown in Figure 1.1. This algorithm was de-
veloped by myself and Dr. Charles Short, director of pathology at Missouri Bap-
tist Medical Center in St. Louis, Missouri. The algorithm is designed to provide
up to four separate data points for each SLN. Each SLN is separately examined
and reported upon at each level, but the rules for “stopping” further examination 1
apply if any SLN is positive, to eliminate unnecessary examinations and expense.
The first data point is an intraoperative examination, using either touch prep or
scrape prep techniques of the bivalved central section of each SLN, as described in
chapter 9. The second data point is H&E examination of the bivalved or central
sections of the SLNs, as would otherwise be performed for all non-SLN biopsies.
All SLN will undergo both of these examinations, regardless of the results of the
intraoperative (immediate) testing. However, should any node be positive by the
second (permanent H&E slide) test, no further pathologic examinations are war-
ranted. This avoids the hours and expense of performing serial section analysis
and immunohistochemical staining on all SLN in every case. We also do not rely
solely on the immediate, intraoperative evaluation, since we are concerned about
false positive results by this method, particularly since not all pathologists will at
least initially be as experienced as Drs. Ku (chapter 9) or Messina (chapter 10) in
these techniques. Should the immediate examination be positive for metastatic
disease, but the central H&E section is negative, we will proceed on to serial sec-
tion analysis, and if necessary, immunohistochemical staining to either confirm
or refute the immediate findings from the touch or scrape prep. For these reasons
also we are not yet using the immediate intraoperative evaluation to change our
surgical plan until we have confirmed the validity of these techniques in our
pathologist’s hands. Therefore, all melanoma patients will have SLN biopsy per-
formed initially without intraoperatively changing to a complete node dissection
unless the node is grossly involved with malignancy. For breast patients, we per-
form the SLN biopsy as part of an IRB approved protocol as above, so that all
patients currently receive complete axillary node dissection immediately after the
SLN biopsy. We anticipate acquiring enough data during this IRB approved breast
SLN protocol to be able to fully implement immediate or intraoperative examina-
tion using the techniques described elsewhere when we begin breast SLN biopsies
off-protocol.
The third data point is the result of H&E examination of serial sections, pre-
pared by slicing the SLN at 2-3 mm intervals and preparing slides for examination
at each level. It is anticipated that 20-30% of negative SLN by central section ex-
amination will be positive by this level of testing. Finally, if all previous tests have
proved negative (or if the only positive test result was the immediate evaluation),
the serial sections are submitted for immunohistochemical analysis, using the
antibodies described elsewhere in this handbook by Drs. Messina and Ku, respec-
tively. Each data point of the pathologic examination process is recorded sepa-
rately, as shown on our breast SLN data sheet (Fig. 1.2), which is also included in
the document set of our IRB protocol.
Implementation of this standardized pathologic algorithm, developed jointly
by the surgical and pathologic teams at our institution, has been very successful
clinically and administratively, involving both departments in the development of
a new way of approaching lymph node biopsy and examination. I strongly
recommend a similar approach at institutions new to SLN biopsy, to enable from
Fig. 1.2. Breast sentinel lymph node mapping study data entry sheet.
the beginning a pathologic protocol that provides results consistent with those
published by the leading centers of SLN innovation.
SCHEDULING A PROCEDURE
To schedule a RGS procedure, it is necessary to coordinate the availability and

timing of several different procedures and departments within the hospital. We
have found the following guidelines to work best:
1) We never obtain a radionuclide localization study before the day of sur-

gery, except initially for patients with primary hyperparathyroidism as we and 1
our referring physicians became more comfortable with that procedure. We found
that physicians preferred to obtain a Sestamibi scan before referring the patient
for surgery, as this is a new procedure and they liked to know before the actual
referral whether a MIRP was possible. Following the initial cases, preoperative
Sestamibi scans are more commonly done only on the day of surgery.
For melanoma and breast cancer cases, there is no reason to subject the patient to
the added discomfort, time, risk, and cost of lymphoscintigraphy at any time except
immediately before surgery. For these two diseases, the decision to perform SLN
mapping is not based on the results of the lymphoscintigram, but on the patient’s
diagnosis, clinical stage, and risk of nodal metastases. Regardless of the results of a
lymphoscintigram performed at a time remote from the RGS procedure, another
radionuclide injection will need to be accompanied by at least one set of scans, so
that the radiologist and hospital can be reimbursed for the injection. Therefore,
even if the nodal chain has been “localized” by a lymphoscintigram performed
days or weeks in advance, a second scan will still be necessary on the day of sur-
gery. For melanoma cases, the lymphoscintigraphy will without fail show a senti-
nel node or nodes somewhere; the decision on patient positioning or sometimes
even level of anesthesia will depend upon the findings of the scan and for this
reason we ask our anesthesiologists and operating room staff to be flexible in their
preparation for melanoma SLN cases, as we never know exactly how we will ap-
proach the case until after the scans are complete. In our experience, breast can-
cers will occasionally not show a definite SLN on preoperative lymphoscintigraphy,
even though one can often be found at the time of surgery, even in the absence of
a “positive” lymphoscintigram. This points out the complementary utility of the
lymphazurin blue dye and the radionuclide, as discussed elsewhere.
2) Schedule all phases of the RGS procedure through one knowledgeable per-
son to try to ensure maximal coordination of the various departments. Our RGS
scheduling is completely coordinated by a single RN experienced in surgical on-
cology, who understands the issues and constraints of both the surgery and the
surgeon. A typical day of surgery schedule for our RGS patients is as follows:
0700 Report to admitting, NPO after midnight. Registration pa-
perwork completed; blood tests and admitting radiographs
(if not already completed) are done.
0815 Patient transported to nuclear medicine
0830 Radionuclide injection
0845-1000 Scans performed
1015 Patient transported to operating room holding area with scans
1030 Patient evaluated by anesthesia, surgeon reviews scans to fi-
nalize choice of anesthetic requirements and patient
positioning.
1100 Scheduled case start time
We provide the patients and their families with an outline of their day of sur-
gery schedule, so they understand the logistics of the day. We also prepare them
for what may be a long day by emphasizing in our communications with them the
1 complexity and uncertainty in coordinating several different departments of the
hospital, and that they should be prepared to spend most of the day at the hospi-
tal, even though the surgery itself is outpatient.
3) Prepare mentally for the inevitable scheduling breakdowns. In addition to
the patients and their families, the operating surgeons must accept the difficulties
in getting all of these procedures and tests to be performed in all patients without
delays. Some of my colleagues have been frustrated by the hospital’s seeming in-
ability to get a lymphoscintigraphy done in time so that the SLN procedure can be
successfully fit into an already tight operating schedule. The following “rules” are
helpful:
-Do not attempt to perform any RGS procedures before 11 AM (noon for
breast cancers) until your institution has experienced 15-20 cases.
- Do not schedule outpatient office hours after your first few SLN proce-
dures, until you and your office have a better appreciation for how long
the surgery and its preceding schedule will take.
- Minimize the number of non-SLN cases you schedule after a SLN proce-
dure, to avoid delays to other elective surgeries.
4) Be creative and prepared to modify your scheduling system based on what
works best at your institution. We are constantly tinkering with the logistics of
our SLN scheduling. For example, we have tried to actually schedule patient trans-
port aides, to improve the efficiency of the transport process. Another modifica-
tion of our schedule has been to send the patient down to nuclear medicine for
radionuclide injection before completing the hospital registration paperwork, etc.,
then bringing the patient back to the admitting area during the gap between in-
jection and scanning. The results of these modifications are mixed; cases still seem
to start at 1100 at the earliest.
Finally, it is important to remember that for melanoma SLN cases, the surgical
approach and position may change based on the location and number of sentinel
node basins identified by the preoperative lymphoscintigraphy. This is also pos-
sible to a lesser extent with breast SLN cases (supraclavicular and internal mam-
mary sentinel nodes are possible). This is most likely to occur with head and neck
melanomas or torso melanomas. For torso melanomas, we always request a bean
bag cushion on the operating table, as the axillary nodal basins are often involved,
and it is possible to put the patient in the decubitus position and allow access to
both the SLN basin and the melanoma primary. Operating room personnel and
anesthesiologists at this institution now expect that the final choice of anesthetic
technique and patient position will not occur until after the scans have been re-
viewed by the surgeon.
Fig. 1.3. Example of an intraoperative gamma probe, specifically

designed and optimally engineered for radioguided surgical procedures,
the Navigator gamma guidance system (US Surgical Corporation,
Norwalk, CT).
PURCHASING A PROBE
Before any RGS procedures can take place, it will obviously be necessary for
your institution to purchase a gamma detector for intraoperative use. There are
currently several different devices available on the marketplace, and it is not the
intention of this author or textbook to recommend one over the other. Currently
available gamma detectors are different in several ways, but market-driven changes
and upgrades make it impossible to discuss the current models in a timely fashion
within a handbook like this. I recommend evaluation of each probe by the in-
volved surgeons, with attention to the three key components of any RGS device:
1) The ability to provide directional, or vectoring, information, to guide
surgical dissection and removal of radiolabeled tissue, by use of colli-
mation to narrow the “field of view” of the probe,
2) Elimination of extraneous or interfering radiation readings, from ei-
ther the primary site or background radiation in the room or building,
with adequate shielding of the radiation detector probe, and
3) Discrimination between radioactive and non-radioactive tissue, includ-
ing filtering of radiation soft tissue scatter, through device incorpora-
tion of energy threshold adjustments.
Fig. 1.4. Sentinel lymph node data sheet for melanoma procedures.
OUTCOMES MANAGEMENT
As your institution progresses in its development of a RGS program, it is im-

perative that appropriate outcomes be analyzed, to document the success rate of
your efforts. In the near future, it is likely that both third party payers and patients
will expect to see some objective evaluation of the volume and results of all surgi-
cal procedures, particularly newer ones or those perceived as requiring special
skills and/or training. This evaluation is easiest to initiate at the beginning of a
program, rather than in midstream.
Fig. 1.5. Minimally invasive radioguided parathyroidectomy (MIRP) patient information registry.
DATA ACQUISITION, STORAGE AND ANALYSIS
The following guidelines and suggestions are offered based only on our spe-
cific clinical experiences, and on my personal experience as a database program-
mer over the past 20 years. The data acquired by each institution is different and is
governed by the needs and interests of the principal “investigators” or primary
clinicians. Additionally, any national protocols such as the coming ACSCOG trial(s)
tend to acquire more data to fit all the criteria and interests of the organizers.
Based on my experience, I have found it best to attempt to streamline and reduce
the amount of information gathered for data storage. Much of the data accumu-
1 lated for routine studies ends up serving no useful purpose; it is never cited in
publications, reports, or presentations, but was included initially because it was
thought to be important and then never removed.14 Data should be acquired and
stored so that when we review our data registry we have a better understanding of
how our RGS practice is evolving, the number and type of procedures we are
performing, and what the results are.
Our primary data entry sheets are designed to fit on one page only and are
disease specific. Our data sheets for breast SLN, melanoma SLN, and MIRP pro-
cedures are shown in Figures 1.2, 1.4, and 1.5, respectively. First, demographic
information is stored to document what kind of patients had this procedure done.
Second, diagnostic data is recorded, enough to classify patients by the extent of
their disease and types of treatment given. Third, preoperative studies should be
recorded, with their results. For RGS procedures, this is typically the nuclear medi-
cine scan and possibly any preoperative staging studies.
Fourth, the operative data is recorded. The first three types of information are
ideally recorded preoperatively, by either the physician or a specialized nurse
clinician, but the operative data is most commonly recorded by the circulating
nurse in the operating room. This process is facilitated by a simplified data re-
cording sheet, with large and obvious data entry points, and minimal duties for
the “recording nurse” other than actually writing down specific information. In
that regard, I have eliminated much of the information that many other surgeons
report recording. Our only goals in recording information about sentinel nodes
are to confirm the identify of the biopsied node as a sentinel node, using the
criteria discussed elsewhere, and to subsequently prove that there are no further
SLN in that basin. Therefore, for each SLN, the only intraoperative data recorded
are the identifier (which much be identical to the identifier sent with the speci-
men to pathology, usually either a number or a letter), the basin, the counts per
second(cps) ex vivo, and whether or not it was blue-stained. After all SLN are
removed, I record the cps of the basin after excision to document the absence of
other SLN. Recording of non-SLN cps is usually superfluous, as the current de-
vices available allow clear distinction of the radioactivity levels between sentinel
and non-sentinel nodes. After excision, there are boxes for the pathologic findings
for each level of pathologic examination of each SLN, as described earlier in this
chapter. For breast SLN patients, the pathologic findings of the completion lymph
node dissection are recorded on this same data sheet; node dissection results for
melanoma patients are recorded elsewhere in our data registry (form not shown).
During MIRP procedures, our data is more focused on the localization infor-
mation provided by the gamma detection device. We record the settings of the
device, the type of probe used, and the cps in all four quadrants of the thyroid
gland (right upper, right lower, left upper, and left lower) initially, after incision,
and after thyroid gland dissection. Finally, the cps of the excised gland is recorded,
and the ratio of the parathyroid tissue cps to the central thyroid gland (docu-
mented to be important by Dr. Norman’s group).11
Our information is stored in computerized, password protected, data regis-

tries running on Filemaker Pro v4.0 for Windows/Macintosh (Claris Corp., 1
Cupertino, California). These databases (there is a separate one for each disease
process) automatically calculate the outcome parameters described below, where
appropriate, permitting continuous evaluation of our RGS program.
To adequately evaluate the success of a RGS program, it is best to establish a
common vocabulary that describes the outcome parameters of interest. These
parameters can then also be used to critically appraise the relevant literature, pro-
viding a common basis for comparison between published results and local re-
sults. My evaluation of the SLN literature has led to my use of the following terms
to describe SLN outcomes: success rate, yield, positivity, sensitivity, false nega-
tive rate, exclusivity, and local failure rate.
Success Rate: The ability to locate SLN in all basins identified by preoperative
lymphoscintigraphy. Note that in the literature, this is not the biopsy of all nodes
identified on nuclear scans; it is the biopsy of at least one node from each basin
found to contain at least one SLN by lymphoscintigraphy. The expected rate for
melanoma should approach 100%. The rate for breast cancer, with its documented
longer learning curve, should be above 90% within the first 20-30 cases.
Yield: Number of sentinel nodes biopsied per patient (sometimes broken down
per basin). Generally, for both melanoma and breast cancer, a mean of 1.5-2.0
SLN are biopsied per patient, with an upper range of 5-6 nodes.
Positivity: Percentage of patients who will have lymph node metastases identi-
fied by SLN pathologic examination. This number varies with patient diagnosis
and known primary lesion prognostic factors, principally tumor size for breast
cancer and Breslow thickness and ulceration for melanoma. This rate will also be
affected by the type of pathologic evaluation performed, as discussed earlier in
this chapter and in other chapters as noted.
Sensitivity/False Negative Rate: The ability of SLN biopsy to identify all pa-
tients with lymphatic metastases. These numbers are related by the equation:
100 = Sensitivity + False Negative Rate
As previously stated, the goal of SLN biopsy is to maximize the sensitivity while
minimizing the false negative rate. This goal justifies the performance of the ex-
tended, comprehensive, pathologic evaluation algorithm. From a cancer progno-
sis and treatment perspective, probably the most important outcome variable is
the false negative rate. If this variable result is unacceptably high (based on litera-
ture values, greater than 5%),1 then patients in that treatment cohort will have too
great a chance of developing regional or distant recurrence after being treated as
“node-negative”. Also, once you and your institution move beyond the routine
performance of complete node dissection regardless of SLN pathology, sensitivity
and false negative rate are impossible to calculate and can only be referenced his-
torically. Therefore, as any institution begins to perform SLN, particularly for breast
cancer where the standard of care remains complete axillary node dissection, these
two variables must be calculated during the IRB protocol phase. The statistical
endpoint of the IRB protocol that I have written for breast cancer, described ear-
1 lier, is patient accrual high enough to document statistical equivalence of the
cohort’s false negative rate for SLN biopsy to published results, or 5%.
Exclusivity: This is the incidence of patients whose only nodal metastases are
in SLN. The early publications of melanoma SLN biopsy all identified an interest-
ing phenomenon, that patients with sentinel node metastases rarely had non-SLN
metastases.15 This is consistent with the concept of the SLN as the primary drain-
ing lymph node of the primary cancer site with the corollary that the other, non-
sentinel lymph nodes in the basin had a significantly lesser role in “draining” that
site of the body, and were therefore at substantially lower risk for metastatic in-
volvement. More recent publications have placed the exclusivity rate of SLN in
melanoma patients above 90%. Interestingly, the exclusivity rate for breast cancer
patients is lower, at 65-70%.16 The reasons for this difference is unclear, but may
be due to intrinsic differences in the lymphatic drainage from skin versus breast
tissue.
Local Failure Rate: Incidence of nodal metastases occurring after negative SLN
biopsy of that lymphatic basin. This is one of the potential poor outcomes of a
negative SLN biopsy, particularly one done without following the comprehensive
pathologic evaluation schemes described in this handbook. A recent publication
suggests that the local failure rate for melanoma patients following negative SLN
biopsy is at least partially affected by the type of pathologic examination per-
formed.12 The local failure rate for breast cancer patients following negative
SLN biopsy without routine completion axillary node dissection is completely
unknown.
The outcome measurements for MIRP are substantially different. Since this is
a newer procedure, employing a novel technical approach to a standard opera-
tion, I believe the outcome measurements currently should concentrate on local-
ization parameters, criteria for parathyroid identification and success rates. Ac-
cordingly, we record information about the relative differences between cps mea-
sured over the diseased gland and elsewhere in the neck. We also record the type
of instrumentation used and its settings. Finally, we record the cps ex vivo, opera-
tive time, and the pathologic and clinical outcome. This information is intended
to facilitate future procedures through eliminating operative steps or frozen sec-
tion analysis, or providing other technical “tricks” to make the procedure go more
smoothly.
BILLING AND REIMBURSEMENT
This is simultaneously the most difficult, most challenging and most contro-
versial and problematic section to write or edit in this handbook. It may also be
perceived as the most valuable section by the general practitioner, who after all
would like to reasonably and appropriately be compensated for the performance
of a specialized operative procedure that requires special training and perhaps
national credentialing. During talks that I give on RGS, discussions of billing and
reimbursement are the most intently listened-to sections of my presentation.

It should come as no surprise to experienced surgeons that there is no existing 1
CPT code that even comes close to describing the procedure performed during
sentinel node biopsy.17,18 The pathologists and nuclear medicine radiologists have
all performed their portions of the SLN procedure before, even if not in the exact
same fashion, and therefore billing for those segments of an SLN biopsy is straight-
forward. For MIRP operations, Sestamibi scanning and the pathologic examina-
tion of the parathyroid are standard procedures that again are covered by avail-
able CPT codes.
No discussion of billing and reimbursement in this era would be complete
without a disclaimer: Given the uncertain nature of billing for any new procedure,
the following codes are given as examples and are based on (limited) experience
and (limited) published advice from the AMA. These are based on our current think-
ing in clinical practice and are not intended to be suggestions to other physicians or
office managers. It is possible that by the time this handbook is published, the situ-
ation may have changed completely and these coding examples may no longer be
appropriate or viable. Any physicians who employ these codes would be doing so at
their own risk and agree to hold the author, editors, and publisher blameless.
Table 1.1. Billing for RGS procedures

Procedure CPT Code Description Amount*
SLN Biopsy, Melanoma

38999 Unlisted lymphatic Varies**
procedure
78195-26 Intraoperative
lymphatic mapping
SLN Biopsy, Breast CA
On Phase I Protocol 38745 Axillary lymphadenec-
tomy, complete
lymphatic mapping
After Protocol complete 38999 Unlisted lymphatic See axillary
procedure SLN**
lymphatic mapping
MIRP
60500 Parathyroidectomy
parathyroid mapping
* Amount billed is usual and customary rate, except as otherwise noted

** For CPT code 38999, we bill the following amounts:
$1,000 SLN biopsy, inguinal node (up to two nodes)
$1,200 SLN biopsy, axillary node (up to two nodes)
$1,400 SLN biopsy, head and neck node (up to two nodes)
$100 Additional amount for each pair of SLN after first two in basin
75% of sum SLN biopsy, more than one basin
(e.g., bilateral axillary SLN biopsy = (1,200 + 1,200) x 0.75 = $ 1,800)
We believe that in the absence of an established all-encompassing code for

1 sentinel node biopsy, surgeons should code for the surgical removal of the SLN(s)
and (separately) for the professional component of interpretation of the readout
from the gamma detector device to localize the lymph nodes at risk. The second
portion can be billed using an existing CPT code of 78195 with the modifier “-26”.
The CPT code 78195 is for nuclear medicine scanning to localize lymphatic drain-
age. We add the text, “intraoperative lymphatic mapping,” to distinguish what we
have done from the work of the radiologists who will likely also be using that
same code, without the modifier, for the lymphoscintigraphy itself. The modifier
is added to emphasize the interpretation aspect of the portion of the procedure
we are billing for. The existing CPT codes for lymph node biopsy describe biopsy
of nodes from a limited number of locations (CPT codes 38500-38530). For a few
basins, they differ for superficial or deep nodal location, but in general describe
only the act of biopsying an otherwise unspecified lymph node or nodes. We be-
lieve that these codes are inadequate to describe what occurs during SLN biopsy
and do not appropriately reimburse the surgeon for the technically demanding
procedure that required additional training and the purchase of a novel device.
We therefore code all melanoma SLN procedures using 38999, unlisted lymphatic
procedure. We enclose a two page explanatory letter with each bill submitted, de-
scribing what SLNs are, why they are biopsied, how they are biopsied, and what
Table 1.2. Reimbursement for RGS procedures*

Payor** % Reimbursed*** Notes
Insurance Co #1 50%
Insurance Co #2 100% HMO
Insurance Co #3 15-25%
Insurance Co #5 85%
Insurance Co #6 100% small private insurer
Insurance Co #8 Under appeal “Not part of benefit package”
Insurance Co #9 Under appeal “Included as part of wide

excision”
* This experience is based on billing as described in Table 1.1, after about 50 melanoma
cases in 1998
** Payors not identified
*** This is the percentage of the billed amount for CPT 38999 paid
(All payors immediately reimburse for CPT 78195-26 at usual rates)
our fee schedule is for all SLN biopsies. We also enclose a copy of the operative
report, with the portion describing sentinel node mapping and biopsy highlighted. 1
This provides the payor with what we hope is adequate information about the
procedure in general to enable them to reimburse the surgeon. Further, it shows
them that we believe this to be a procedure that will be performed multiple times
in the future, so that we have established a consistent fee schedule, accounting for
different basins, their relative complexity from a surgical perspective, and the num-
ber of nodes biopsied. This fee schedule is shown in Table 1.1. Currently, we are
performing all breast SLN procedures under IRB protocol with routine axillary
node dissection. As such, we code only for the deep axillary node dissection (CPT
38745) for the procedure. In the future, we will likely adopt the same billing pro-
tocol for breast SLN as used for melanoma cases. We bill MIRP procedures using
the standard parathyroidectomy code (CPT 60500).
The second portion of the bill for all SLN procedures is 78195-26. The com-
plete axillary node dissections that we are currently performing after all breast
SLN biopsies are also billed with this code, in addition to CPT 38745, as described
above. The MIRP procedures have the corresponding code of 78070-26 for the
intraoperative mapping of the parathyroid adenoma location.
The first test of a billing strategy is the reimbursement rate. To date, all insur-
ance companies that we have billed have immediately paid on 78195-26. This
amount is generally in the $50-$100 range. As might be expected the SLN biopsy
billing using CPT 38999 is more problematic and has produced a variety of re-
sponses. The results of these policies five months after implementation are shown,
without corporate or governmental identifiers, in Table 1.2. To date, this billing
strategy seems appropriate and has been received relatively well by third party
payers, but we anticipate that as SLN becomes more prevalent, the AMA will cre-
ate new CPT codes for these procedures.
CONCLUSIONS
This chapter has described the implementation of a successful RGS program,

from initial conception and organization, through outcome analysis and billing/
reimbursement. The goal of this chapter was to “fill in the cracks” in any efforts to
establish a sentinel node and radioguided surgical program. Any additional ques-
tions may be forwarded to me at [email protected].
ACKNOWLEDGMENTS
The author would like to acknowledge the assistance of Ms. Anita Boatman,
RN and Ms. Patricia Eichholz, both indispensable to any smoothly running surgi-
cal practice.
REFERENCES
1 1. Krag D, Weaver D, Ashikaga T et al. The sentinel node in breast cancer: A
multicenter validation study. N Engl J Med 1998; 339:941-946.
2. Reintgen D, Cruse CW, Wells K et al. The orderly progression of melanoma nodal
metastases. Ann Surg 1994; 220:759-767.
3. Albertini JJ, Lyman GH, Cox C et al. Lymphatic mapping and sentinel node bi-
opsy in the breast cancer patient. JAMA 1996; 276:1818-1822.
4. Veronesi U, Paganelli G, Galimberti V et al. Sentinel-node biopsy to avoid axil-
lary dissection in breast cancer with clinically negative lymph-nodes. Lancet 1997;
349:1864-1867.
5. Giuliano AE, Jones RC, Brennan M, Statman R. Sentinel lymphadenectomy in
breast cancer. J Clin Oncol 1997; 15:2345-2350.
6. Reintgen D, Haddad F, Pendas S et al. Lymphatic mapping and sentinel lymph
node biopsy. In: Care of the Surgical Patient. Scientific American, 1998:1-17.
7. Emilia JCD, Lawrence Jr,W. Sentinel lymph node biopsy in malignant melanoma:
the standard of care? J Surg Oncol 1997; 65:153-154.
8. Reintgen D, Balch CM, Kirkwood J, Ross MI. Recent advances in the care of the
patient with malignant melanoma. Ann Surg 1997; 225:1-14.
9. Balch CM, Soong S, Bartolucci AA et al. Efficacy of an elective regional lymph
node dissection of 1 to 4 mm think melanomas for patients 60 years of age and
younger. Ann Surg 1996; 224(3):255-266.
10. Kirkwood JM, Strawderman MH, Ernstoff MS, Smith TJ, Borden EC, Blum RH.
Interferon Alfa-2b Adjuvant therapy of high-risk resected cutaneous melanoma:
The Eastern Cooperative Oncology Group Trial ECOG 1684. J Clin Oncol 1996;
14:7-17.
11. Norman J, Chheda H. Minimally invasive parathyroidectomy facilitated by in-
traoperative nuclear mapping. Surg 1997; 122:998-1004.
12. Gershenwald JE, Colome MI, Lee JE et al. Patterns of recurrence following a
negative sentinel lymph node biopsy in 243 patients with stage I or II melanoma.
J Clin Oncol 1998; 16:2253-2260.
13. Turner RR, Ollila DW, Krasne DL, Giuliano AE. Histopathologic validation of
the sentinel lymph node hypothesis for breast carcinoma. Ann Surg 1997;
226:271-278.
14. Whitman ED, Frisse ME, Kahn MG. The impact of data sharing on data quality.
Proceedings of the American Medical Informatics Association 1995; 19:952.(Ab-
stract)
15. Joseph E, Brobeil A, Glass F et al. Results of complete lymph node dissection in
83 melanoma patients with positive sentinel nodes. Ann Surg Oncol 1998;
5:119-125.
16. McMasters KM, Giuliano AE, Ross MI et al. Sentinel-lymph-node biopsy for
breast cancer - not yet the standard of care. N Engl J Med 1998; 339:990-994.
17. Coding consultation: Lymph nodes and lymphatic channels. CPT Assistant 1998;
8:10.
18. Physicians’ current procedural terminology: CPT ‘98. Fourth edition, American
Medical Association, Chicago, IL, 1997.
Radiation Safety 23
Intraoperative Gamma Radiation

Detection and Radiation Safety 2
David A. Hillier, Henry D. Royal
Introduction .......................................................................................................... 23
Basic Radiation Terminolo gy ............................................................................... 23
Radiation Detection ............................................................................................. 26
Gamma Probe Testing .......................................................................................... 26
Proper Gamma Probe Use .................................................................................... 29
Radiation Safety and Regulations ........................................................................ 33
INTRODUCTION
Effective intraoperative use of a gamma probe requires familiarity with basic

principles of radiation, radiation detection and radiation safety. Basic radiation
terminology, radiation detection, gamma probe testing, proper gamma probe use,
radiation safety and regulatory issues will be presented in this chapter.
BASIC RADIATION TERMINOLOGY
The terms used to describe radiation are confusing and complex. To make
matters worse, two different sets of terms are commonly used to describe radia-
tion related quantities. The United States has persisted in its use of older terms
whereas much of the rest of the world has adopted newer terms call Standard
International (SI) units. Despite this difficulty, mastery of a basic radiation vo-
cabulary is necessary in order to understand some important fundamental con-
cepts regarding radiation and in order to effectively communicate with other
medical personnel who routinely work with radiation.
Radiation is a general term describing the outward propagation of energy in
any one of a wide variety of forms (Fig. 2.1). A broad distinction is made between
“ionizing” radiation, which has sufficiently high energy to strip electrons from
atoms, and the less energetic “nonionizing” radiation.
Ionizing radiation is subdivided into non-particulate and particulate radia-
tion. Non-particulate ionizing radiation (photons) consists of x-rays and gamma
rays. Higher energy x-rays and gamma rays penetrate the body, are detectable
outside the body and therefore are useful for imaging. X-rays and gamma rays are,
by definition, distinguished by the way in which they are formed and are
Transition from high to low energy state

(the excess energy is given off as radiation)
2 Nonionizing Ionizing
60 - cycle
Radiofrequency radiation Nonparticulate Particulate
Microwaves
Infra-red light
Visible light X-rays
Ultraviolet light Gamma rays Charged Uncharged
Alpha particles Neutrons

Beta particles Neutrinos
Protons
Deuterons
Fission fragments
Fig. 2.1. Types of radiation.
indistinguishable once they have been emitted. X-rays result from transitions in
the energy state of an electron. Gamma rays and the particulate forms of ionizing
radiation are created when the nucleus undergoes a transition from a high energy
state to a lower energy state. Gamma rays generally are of higher energy than x-
rays, but there is considerable overlap. The magnitude of the energy content of a
photon is expressed in electron volts (eV). Visible light photons (one form of
nonionizing radiation) have energies of a few eV. X-rays and gamma rays used in
diagnostic imaging have energies in the range of tens to hundreds of kilo-elec-
tron-volts (1000 eV = 1 kilo-electron volt or 1 keV).
Charged particulate radiation (e.g., alpha-rays and beta-rays) does not pen-
etrate the body well and causes much greater biological harm than similar amounts
of x-rays and gamma-rays. Charged particles can be used for therapy but not for
imaging. Uncharged particles (e.g., neutrons and neutrinos) do penetrate the body,
but are not easily detectable and neutrons have unfavorable dosimetry.
The words “radiation” and “radioactive” are often confused. An atom that is
unstable spontaneously gives off radiation and is therefore radioactive. In con-
trast, an x-ray machine is not radioactive since it cannot spontaneously give off
radiation (without an external power source). Patients who have had a chest ra-
diograph do not spontaneously emit radiation. In contrast, patients who have
been injected with a radioactive material will continuously emit radiation for a
period of time.
The amount of radiation that patients emit from diagnostic nuclear medicine
studies is quite small and not a significant problem (see the section on radiation
safety below). The amount of radiation emitted decreases with time due to physi-
cal decay of the radionuclide (defined by a physical half-life) and elimination from
the body (defined by a biological half-life). The physical half-life (t1/2p) depends
on the radionuclide (radioactive atom or radioisotope). The most commonly used
Radiation Safety 25
radionuclide, Technetium-99m (Tc-99m), has a t1/2p of 6 hours. This means that

the amount of radiation decreases by a factor of two every 6 hours from physical
decay alone. The biological half-life (t1/2b) depends upon the chemical compound
to which the radionuclide is bound. The combination of the radionuclide and
2
carrier molecule is termed a “radiopharmaceutical”. In the United States, the most
commonly used radiopharmaceutical for sentinel node lymphoscintigraphy is fil-
tered Tc-99m sulfur colloid. For this radiopharmaceutical the rate at which the
radiation decreases is primarily determined by physical decay since its t1/2b is very
long.1,2
The amount (activity) of a radionuclide is expressed in terms of the number
of atoms that are decaying (disintegrations) per unit time (second). This term is
used because the number of counts detected by a radiation meter is generally
directly related to the amount of radionuclide. For reasons discussed below, it is
difficult to simply relate the amount of a radionuclide to the dose that a worker or
patient might receive. In the United States, the unit for the measurement of the
amount of activity is the curie (3.7 x 1010 disintegrations/second). The curie ap-
proximately equals the number of atoms disintegrating per second in a gram of
radium. The SI unit for the amount (activity) of a radionuclide is the becquerel
(1.0 disintegration/second). Curie amounts of radionuclides are not used in diag-
nostic studies. For most diagnostic studies, millicurie (mCi; 1000 mCi = 1 curie)
amounts of a radionuclide are injected. For lymphoscintigraphy, microcurie (µCi;
1000 µCi = 1 mCi = 37 megabecquerels [MBq]) amounts are transported to the
lymph nodes.
As a first approximation, the potential harmful effects of radiation are related
to the amount of energy that was deposited from the radiation per gram of tissue.
The unit used for absorbed dose is the rad (1 erg/ gram of tissue). The corre-
sponding SI unit is the gray (1 gray = 100 rads). Another important determinant
of the possible effects of radiation is the type of radiation. Some types of radiation
are more damaging than others for the same amount of energy deposited because
the energy is deposited more densely. For example, alpha rays can cause up to 20
times as much damage as gamma rays. The unit that takes into account the bio-
logical effectiveness of the radiation is called the rem. The SI unit for the rem is
the sievert (1 sievert [Sv] = 100 rems). One rad of alpha rays might equal 20 rems
but one rad of gamma rays equals one rem. A rem also equals 1000 millirems
(mrem). Other determinants of the possible effects of radiation include the por-
tion of the body exposed and the period of time over which the exposure took
place.
The average radiation exposure to members of the public from natural sources
of radiation (radionuclide in the soil and air, our food and bodies) is about 300
mrem per year. The occupational exposure limit is 5,000 rems per year to the
body. The radiation dose to patients from lymphoscintigraphy would be less than
100 mrem.
The dose to workers from radionuclides is related to the amount of activity,
the type of radiation emitted, the distance and the time spent near the source of
radiation. Lead aprons are not as effective at decreasing the dose from most
radionuclides as they are from x-rays because of the relatively higher photon en-
ergy of the gamma rays. Approximately 75% of the gamma rays from Tc-99m will
be stopped by a standard 0.5 mm-thick lead apron compared to 95% of diagnos-
tic x-rays.3 The dose to patients is also related to the amount of activity and the
2 type of radiation emitted, however patients have little control over the distance
and the time spent near the source of radiation. The physical and biological half
lives and the biodistribution of the radiopharmaceutical determine how long the
source of radiation is in the patient.
RADIATION DETECTION
A variety of instruments can be used to detect gamma rays, including gas de-
tectors (ionization chambers, proportional counters and Geiger-Mueller counters),
organic liquid scintillation, x-ray film and solid detectors. Intraoperative gamma
probes utilize solid detectors which efficiently absorb gamma rays due to high
mass density and high atomic number.
Two basic types of intraoperative gamma probes are in current use. Sodium
iodide crystals have been used for many years to detect radiation and are used
extensively in current day gamma cameras. When radiation interacts with a so-
dium iodide crystal, the energy of the gamma ray is converted into a flash of light
called a scintillation. This flash of light is converted to an electrical pulse of cur-
rent by a photomultiplier tube that is optically coupled to the crystal. The inten-
sity of the scintillation and therefore the size of the current pulse is proportional
to the energy of the photon that was detected. This type of detector is called a
scintillation detector. Approximately 85% of 140 keV photons from Tc-99m will
be absorbed by a 3/8-inch-thick (9.5 mm) NaI crystal. Only a few of the absorbed
photons reach and are detected by the photomultiplier tube, such that approxi-
mately 3 light photons are detected for each keV of energy absorbed.1
A newer type of solid detector uses a semiconductor to directly convert the
energy of the detected photons to an electrical pulse of current. The semiconduc-
tor used in currently available intraoperative probes is cadmium telluride (CdTe).
Approximately 60% of 140 keV photons from Tc-99m will be absorbed by 2-mm-
thick CdTe. Charge carriers are produced for each 4.4 eV of energy absorbed. The
energy resolution of semiconductor detectors is better than with scintillators since
many more charge carriers are created for each gamma ray absorbed than are
light photons detected for each gamma ray absorbed in a scintillation detector.1,4-6
GAMMA PROBE TESTING
Prior to using a new gamma probe in the operating room, a number of tests
should be performed. This testing serves two purposes. First, testing ensures that
the probe is functioning properly. Second, testing helps the surgeon become fa-
miliar with the use of the probe. Tests should be performed to determine the di-
Radiation Safety 27
rectionality (isocount lines), shielding of the probe and count rate capability
(linearity).
Directionality can be determined by recording the count rate detected by the
probe as a radioactive source is moved in precise locations around the probe. These
2
measurements define the volume of tissue in which activity is detected by the
probe. In the simplest of terms, this volume of tissue can be visualized as a cone
with the apex directed towards the probe (see Fig. 2.2). The count rate detected
from the same amount of activity will be greatest in the center of each slice through
the cone and in the slices closest to the probe. In vivo, the count rate will dramati-
cally decrease with depth due to the inverse square law and to attenuation of ac-
tivity by the overlying tissue. The size of the cone of tissue sampled (field of view)
by the probe can be readily modified by use of a collimator. A collimator is a
metallic tube that limits the direction that gamma rays must come from in order
to strike the detector. Most collimators have been optimized for use with Tc-99m.
If radionuclides with higher energy photons (e.g., In111-173 and 245 keV; I13-364
keV; F18-511 keV) are used, the collimator wall thickness is increased, shielding
the detector from higher energy photons. The diameter of the collimator is usu-
ally similar to the diameter of the detector. A smaller field of view can be created if
the detector is moved back from the tip of the collimator tube, or if the tube’s
length is increased relative to its diameter (see Fig. 2.2). Small fields of view offer
the potential advantage of improved spatial localization and improved signal-to-
noise ratio, but require a more careful search pattern to avoid missing the radioac-
tive tissue. In addition, if the diameter is smaller, fewer counts will be detected. If
Fig. 2.2. Volume of detection for intraoperative gamma probes is a cone-shaped region. The
volume is broad for a short collimator (left) and narrow for a long collimator (right). The latter
offers the potential advantage of more precise localization, but disadvantages include increased
probability of a sampling error and, under some circumstances, a lower count rate.
the diameter is unchanged, but the collimator is elongated, fewer counts may still
be detected if the target does not fit inside the smaller cone of detection. How to
optimize the tradeoff between maximizing localizing ability and minimizing sam-
pling error is not clear and is dependent upon the technique utilized by the indi-
2 vidual.
Another important characteristic of probe design that can be simply tested is
the adequacy of side shielding. With lymphoscintigraphy, most of the injected
activity remains at the site of injection. If the injection site is close to the site of the
sentinel nodes, detection of the sentinel nodes could be compromised if there is
inadequate side shielding. The adequacy of side shielding can be simply tested by
measuring the count rate when a source of activity is in the center of the field of
view compared to when the source is adjacent to the side of the detector. For the
probes that we tested, the count rate from a Tc-99m source at the side of the
detector was about 1% of the count rate when the source was in the center of the
field of view. If the activity at the injection site is 100 times greater than the activ-
ity in the lymph node (a likely possibility) then equal numbers of counts may be
detected from the injection site and the lymph node. Additional side shielding can
readily be obtained when necessary by placing a small piece of lead (or a bulky
metallic surgical instrument) between the side of the probe and the injection site.
Additional side-shielding will particularly be necessary if the probe is used with
radionuclides that have higher energy gamma rays. Based on tests performed on
current probes, optimized for imaging with Tc-99m, the count rate from an I-131
source at the side of the detector can be as high as 50% of the count rate when the
source was in the center of the field of view.
Typically radiation detectors will only have a linear response over a limited
range of activity. When too much activity is used, the count rate that is detected
reaches a plateau (non-paralyzable electronics) or can actually decrease (paralyzable
electronics). Determining the range of count rates over which the probe responds
linearly helps to define the useful operating range of the probe. The linearity of
detector response with respect to activity level can be tested using two different
methods. For either method, a small vial is filled with an amount of activity
(1-10 mCi Tc-99m would be a good choice) that likely would greatly exceed the
maximum amount of activity that the probe would ever be expected to detect in a
discrete focus. The first method for measuring linearity uses the physical decay
(t1/2p) of the radionuclide in order to place varying but exactly known amounts of
activity in front of the probe. For this method, the vial is placed in front of the
probe, in exactly the same position, twice a day for several days. The date, time of
day and count rate detected by the probe is recorded and plotted on a semilog
graph. The graph will be a straight line over the linear response range of the probe.
A second method of determining linearity uses metallic plates (e.g., 10 copper
plates, 1/16 inch (1.6 mm) thick and approximately 3 inches (8 cm) square) to
precisely vary the amount of activity that reaches the probe. This latter method is
more convenient because it permits this test to be made more quickly. The copper
plates are placed directly between the source and probe, one at a time and the
count rate recorded. Since each copper plate absorbs a constant fraction of the
Radiation Safety 29
activity, a semilog graph of count rate vs. the number of copper plates should be
linear. The test can be repeated with a variety of probe-to-source distances in or-
der to extend the count rate range over which the probe is tested. All probes tested
demonstrated a linear response over a range of activity levels likely to be encoun-
2
tered during surgery.
The energy of a gamma ray can be partially absorbed in the body by a process
termed Compton scatter (Fig. 2.3). The resulting gamma ray is of reduced energy
and altered direction. It may therefore provide unreliable or confusing localizing
information intraoperatively. Ideally this can be compensated for by manual or
factory-set adjustments of the energy detection threshold of the gamma probe
device, eliminating detection of the lower gamma radiation from Compton scatter.
PROPER GAMMA PROBE USE
If images are obtained prior to surgery, marking the location of the sentinel
node on the patient’s skin may help the surgeon locate the node. Marking the
Fig. 2.3. Compton scatter. The injection site in lymphoscintigraphy contains a high level of
activity relative to the lymph node. To localize the lymph node, the probe must have a collima-
tor to block gamma rays from entering the side. Compton scatter, resulting from interaction of
a primary photon from the injection site with an atom in the patient’s tissue, may produce a
photon that is directed into the probe detector, making detection of the lymph node difficult. A
Compton scattered photon has a lower energy than the primary photon, and can be distin-
guished on this basis. The energy of the scattered photon is related to the scatter angle (the
energy is nearly equal to that of the primary photon at low scatter angles). Energy resolution of
the detector is therefore an important factor in scatter rejection and target-to-background ratio.
location of the node on the skin is subject to considerable error unless the follow-
ing steps are taken. First, the skin should be marked with the patient positioned as
he or she will be in during surgery in order to maintain the relationship between
the relatively mobile skin and the underlying structures. Second, because of the
2 problem of parallax, the skin mark site is dependent on the position of the gamma
camera (see Fig. 2.4). Therefore, the person marking the skin must tell the sur-
geon the position of the gamma camera relative to the patient when the skin was
marked.
Skin marking is performed by holding a container with a small amount of a
radionuclide. The radionuclide is placed between the patient and gamma camera
while watching a monitor that displays the node and the external source of activ-
Fig. 2.4. Localizing a lesion intraoperatively after the skin site has been marked in the nuclear
medicine department during lymphoscintigraphy. It is important to orient the probe in the
same direction as the gamma camera to first localize the lymph node. The imaging plane of the
gamma camera is commonly (although not always) oriented parallel to the floor. On a rounded
body part, this will usually not be parallel to the skin surface. If the gamma probe is oriented in
the same direction as the gamma camera (middle figure), the lymph node should be directly
beneath the probe, on a line perpendicular to the probe tip face. The most direct approach, with
the shortest distance from the skin surface, can then be determined by moving the probe away
from the skin mark, perpendicular to the skin surface until the maximum count rate is ob-
tained. The lymph node should be located at the intersection of the first line obtained with the
probe at the skin mark and the line obtained with the probe perpendicular to the skin surface at
the maximum count rate. If the gamma probe is oriented perpendicular to the skin surface
initially, the lymph node will not appear directly beneath the skin mark, as shown in the dia-
gram on the right. It is also important to take into account body positioning differences be-
tween that used during scintigraphy and intraoperatively. For example, if the neck is extended
during surgery for a parathyroid adenoma, then the neck should also be extended during scin-
tigraphy if the skin site is to be marked.
Radiation Safety 31
ity in real time. When the location of the external source corresponds to the sen-
tinel node, the skin site is marked. Because these images consist of counts ac-
quired over only a brief period of time, the quality of the images are poor and it
may not be possible to locate some nodes which contain little activity.
2
Additional localizing information, may be obtained by marking anatomic land-
marks with a radioactive marker or by using a large uniform source of activity
(flood source). Most nuclear medicine facilities have a solid flood source that is
used daily to check that the nuclear medicine camera is functioning properly. When
a patient is placed between the uniform source of activity and the gamma camera,
a transmission image is obtained that outlines the body (see Fig. 2.5).
Some surgeons use only the probe to find the sentinel node. The probe only
approach may work in many patients and has the advantage of scheduling sim-
plicity and reduced cost. However, there are a number of disadvantages in not
obtaining pre-operative imaging. First, the lymphatic drainage of some lesions
(e.g., a midline melanoma) may be uncertain. Second, in transit nodes may be
missed in the case of an extremity melanoma. Third, sentinel nodes are not always
identified with radioactive tracers. If nodes are not seen on preoperative images,
dye based methods may be more useful in identifying the node intraoperatively.
The sentinel node is identified with a gamma probe by the fact that it has a
higher amount of the radioactive tracer per gram of tissue (target) than does the
surrounding tissue (background). For lymphoscintigraphy, the target/background
ratio of activity is very high and therefore the sentinel node can usually be de-
tected without much difficulty. Problems occur when the probe is not pointed in
R Anterior L R Anterior L
Fig. 2.5. Transmission image. Lymphoscintigraphy examination in a 42-year-old woman with

an anterior thoracic wall melanoma. Left: Anterior view, emission image. The gamma ray pho-
tons “emitted” from the radiopharmaceutical are imaged by the gamma camera to create an
“emission” image. The intense activity at the injection site is in the center and uptake in three
lymph nodes is seen to the patient’s left. There is little anatomic information in such an image.
Right: Anterior view, combined emission and transmission image. A Cobalt-57 sheet source
was placed behind the patient during image acquisition. The patient’s body blocks the majority
of gamma rays from the sheet source (any that pass through the patient are described as having
been “transmitted” through the patient). The unobstructed portion of the sheet source beyond
the edge of the patient’s body are transmits freely to the gamma camera, outlining the patient’s
body. One can clearly see that the uptake is in the patient’s left axilla.
the right direction and is therefore not sampling the correct volume of tissue or
includes some counts from the injection site. Additionally, the injection site con-
tains much more activity than the node itself. If the node is near the injection site,
additional shielding to block counts from the injection site may be helpful.
2 The count rate that is detected by the probe is greatly affected by the distance
between the probe and the node. To the extent that the node is a point source
relative to the probe, the count rate will go down as a function of 1/distance2. In
addition, the tissue between the probe and the nodes absorbs some of the gamma
rays. The net effect of these two factors is that the count rate detected from the
probe greatly increases as the probe moves closer to the node. If the node contains
very small amounts of activity, it may be difficult to confirm that it is a sentinel
node in situ; however, once it is removed from the body, it can be counted outside
the body and the count rate can be compared to that from a similarly sized piece
of non-sentinel node tissue.
Currently available probes convey the count rate on a display (either an analog
rate meter or digital display) and by sound. Sound provides immediate feedback
and is most useful when easily detectable amounts of activity are present. Use of
quantitative timed counts may be helpful in locating areas with smaller amounts
of activity or poor target to background ratios.
From a purely hypothetical perspective, statistically different counts should
vary by an amount greater than that expected simply due to chance alone, or at
least three times the standard deviation. For radioactivity (defined by a Poisson
statistical process), one standard deviation is the square root of the total counts. A
common error is to use a time counting period that is too short (for example, by
only using a default mode of counts per one second). To adequately distinguish
the sentinel node from background activity, counting for a longer period of time
may be necessary. For example, if the counts over region “A” is 100 in one second
and that over an adjacent region, “B,” is 85, there is no statistically significant dif-
ference. In this example, the difference is less than three standard deviations, where
the standard deviation is 10 (the square root of 100) and three standard devia-
tions is 30. If counts for 10 seconds yields 1000 over region “A” and 850 over an
adjacent region, “B,” this is a statistically significant difference (since the standard
deviation is 33.3 and three times this is 100).7 Other authors have used more op-
erational definitions of the difference in counts defining a sentinel node, as de-
scribed in other chapters.
Other common errors include moving the probe over the operative site too
rapidly and varying the distance between the probe and the patient. Both of these
errors will cause variations in count rate that are due to technical factors and
therefore the count rate changes will not provide good localizing information.
The easiest way to ensure a consistent distance is to hold the probe against the
skin. When a lymph node is found but not yet resected, it may be lifted partially
out of the surgical bed such that the probe can be angled toward the node and
away from the injection site and other lymph nodes more effectively.
Radiation Safety 33
RADIATION SAFETY AND REGULATIONS
Radiation is frightening to many people. The reality is much less sensational.

The potential harm from radiation exposure is proportional to the dose. The harm
of greatest concern from occupational exposure to radiation is that the radiation
2
will cause cancer. Although it has been possible to measure small (8%) increases
in cancers in large populations instantaneously exposed to large doses of radia-
tion (e.g., atomic bomb survivors), it has been difficult to measure any increase in
cancers in occupationally exposed persons.
Since the harm from radiation exposure is proportional to dose, it is helpful to
have some examples of the possible magnitudes of radiation doses. A useful bench-
mark is the amount of radiation we are all exposed to every year due to naturally
occurring radioactive materials in our body and in our environment. Natural back-
ground radiation exposes every person in the U.S. to about 300 mrem per year.
There is considerable variation in the amount of background radiation
(± 100 mrem) from place to place and people rarely consider exposure to natural
background radiation when choosing a place to live. There are some places in the
world where normal background radiation exceeds the 300 mrem/year average by
1000s of mrem.
The U.S. Nuclear Regulatory Commission annual occupational dose limits for
a radiation worker are 5 rem effective dose equivalent to the body (deep dose
component of body badge dose), 15 rem to the eye, or 50 rem to skin or an indi-
vidual organ. Occupational dose limits to pregnant women (who have declared
the pregnancy) is 0.5 rem during the pregnancy. Typical whole body doses to pa-
tients who are injected for nuclear medicine studies is a few hundred mrem to a
few thousand mrem. The average dose to nuclear medicine technologists is ap-
proximately 180 mrem/year.8 The maximum allowable radiation dose to a mem-
ber of the general public (from man-made sources other than medical exams ben-
efiting the subject) is 100 mrem/year.
Table 2.1 lists estimated whole-body effective dose equivalent and finger doses
from sentinel node biopsy, minimally invasive radioguided parathyroidectomy
(MIRP) and fluoroscopy (e.g., during central venous catheter insertion). Based
on the estimated per case radiation dose, the number of procedures that could be
performed before exceeding three different annual dose limits are listed. The
first limit (fourth column in Table 2.1) is the number of cases that results in the
maximum allowable dose to a member of the general public (deep dose–
100 mrem/year). If it is anticipated that personnel may approach or exceed this
level in a year, formal training as a radiation worker is necessary. This entails training
in basic radiation safety concepts and may require a short written exam, adminis-
tered by the institution’s radiation safety office. Personnel are only required to
wear a radiation badge if their annual whole-body radiation dose is likely to ex-
ceed 500 mrem, or if the finger dose exceeds 5 rem (10% of the annual occupa-
tional limit), as shown in the fifth column in Table 2.1. In practice, however,
institutions may choose to issue radiation badges at a lower threshold (e.g.,

100 mrem/year).
Note that the estimates in Table 2.1 are based upon worst-case calculations
(see footnote for Table 2.1). The actual dose received is dependent on the condi-
2 tions specific to each case, such as the activity administered, the time since injec-
tion, duration of the procedure, length of time spent near the source and the dis-
tance from the source. Based on the administered activity alone (20 mCi for mini-
mally invasive radioguided parathyroidectomy [MIRP] and 500 mCi for sentinel
lymph node biopsy), one can predict that the whole body radiation dose from
MIRP would be substantially higher than the whole body dose from sentinel node
biopsies. Personnel are only required to wear a radiation badge if their annual
whole-body radiation dose is likely to exceed 500 mrem (collar badge), or if the
finger dose exceeds 5 rem (finger dose). These thresholds represent 10% of the
annual occupational limit, as shown in the fifth column in Table 2.1. However,
since the surgeon and his staff likely have many sources of radiation exposure, a
more conservative approach is suggested. If the annual number of procedures
performed approaches or exceeds the number listed for the limit for a member of
the public (100 mrem/year), monitoring is recommended. Although it is likely
that the measured doses will be considerably less than the per case doses listed in
Table 2.1, the most practical way to measure radiation doses accurately is with
radiation badges. Because of possible radiation exposures from other sources, the
per case dose can only be measured accurately if the radiation badge is only worn
during the procedures of interest.
Good radiation safety practice encourages workers to take simple, easy to imple-
ment steps to make even small radiation doses smaller. This philosophy is referred
to as the ALARA (an acronym for “As Low As Reasonably Achievable”) principle.
Unfortunately what is reasonable and what is simple and easy to do are very am-
biguous and hence subject to considerable local variations. Local decisions should
be based on facts and the benefits (small decreases in radiation dose) of invoking
the ALARA principle must be weighed against the harms (e.g., delays in diagnosis,
diversion of limited resources, distraction of the surgical staff).
There are three ways (i.e., time, distance and shielding) to significantly de-
crease radiation doses. The radiation dose is linearly related to time. Halving the
time spent close to the patient would halve the radiation dose. The radiation dose
is related to the square of the distance. Doubling the distance to the source will
decrease the radiation dose by a factor of 4. Finally, placing a shield between the
patient and operating room personnel can significantly decrease the radiation dose
to the personnel. For lymphoscintigraphy, the injection site can be easily shielded
by covering the injection site with a small thick (1-2 mm) lead sheet when access
to the primary site is not needed. For MIRP, the radiopharmaceutical is much
more widely distributed and more difficult to shield. Key operating room person-
nel could wear lead aprons (containing shielding of 0.5 mm thick lead equivalent)
that would absorb 75% of the radiation; however, aprons are cumbersome and
therefore impractical.3
Table 2.1. Estimated dose during procedure and regulatory limits for surgical and pathology personnel
Radiation Safety
Number of procedures per year
Radiation worker; Radiation worker;
Per case dose Limit to member level requiring maximum
Exposure Personnel (mrem) of general public monitoring allowable
Sentinel node biopsy Surgeon, body 0.29 350 1,700 17,000

(following Surgeon, finger 6.6 – 750 7,500
lymphoscintigraphy)
(500 µCi Tc-99m) Pathologist, body 0.052 1,900 9,600 96,000
Pathologist, finger 4.4 – 1,100 11,000
Minimally invasive Surgeon, body 2.1 48 240 2,400
radioguided Surgeon, finger 10 – 500 5,000
parathyroidectomy
(MIRP) Pathologist, body 0.00094 110,000 530,000 5,300,000
(20 µCi Tc-99m) Pathologist, finger 0.85 – 5,900 59,000
Fluoroscopy, Body, no lead apron 3.6-14 7 35 350
3 minutes Body,with lead apron 0.18-0.72 140 690 6,900
Body with lead apron 0.090-0.36 280 1,400 14,000
and thryoid shield
Estimated whole-body effective dose equivalent and finger doses for a single procedure are based on generalized worst-case assumptions, including no biological clearance,
no attenuation by overlying soft tissues, and high (5%) uptake by the sentinel lymph node, and a large (2 g) parathyroid adenoma with uptake 15 times the average uptake
of the remainder of the body. Typical doses to personnel during fluoroscopy on an adult at a distance of 3 feet are shown for comparison. A lead apron of 0.5 mm lead
equivalent, worn by the operator, absorbs approximately 99% of x-rays in the diagnostic energy range and will reduce the deep whole-body effective dose equivalent by
approximately 95%. It will absorb only 75% of gamma rays from Tc-99m at 140-keV energy.3 It is assumed that, during surgery following lymphoscintigraphy, the surgeon
will work at the lymph node site between hours 2 and 3 after injection (at a distance of 30 cm from the lymph node site and 60 cm from the primary site) and that he will
then work at the primary tumor site between hours 3 and 4 after injection (at a distance of 30 cm from the primary site and 60 cm from the lymph node site). Surgeon
finger dose is based uon a point source at 5 cm for 30 minutes at the primary site and at 1 cm for 15 minutes at the lymph node site. Following parathyroid scintigraphy, it
is assumed that surgery is performed at hours 2.5 to 3.5 after injection. Surgeon dose for parathyroid scintigraphy is based upon the dose rate from patients 10 minutes
after injection of 20 µCi of a Tc-99m labeled bone scintigraphy raiopharmaceutical, at a distance of 50 cm for body dose and at a distance of 10 cm for finger dose.9 The
pathologist is assumed to handle samples for 15 minutes immediately following surgical removal. Pathologist dose is calculated as that from a point source at 30 cm for
body dose, at 5 cm for finger dose from the tumor primary site sample and at 1 cm for sentinel lymph node and parathyroid adenoma.
35
2
Alternatively a lead apron could be used to shield the patient’s abdomen and
lower chest. Sestamibi is excreted by the hepatobiliary system into bowel and by
the kidneys into bladder. Therefore most of the activity will be in the abdomen at
the time of surgery. Use of a lead apron over the patient’s abdomen and lower
2 chest would significantly decrease (perhaps 50% or more) the radiation dose to
the surgical staff. Whether this should be done is controversial. A very small per
case reduction of radiation dose (possibly 2.1 mrem to 1.0 mrem) may be achieved;
however, there also may be some very small increased risk to the patient from
respiratory insufficiency (due to the weight of the apron) or hyperthermia. Local
decisions vary considerably due to the difficulty in assessing very small competing
risks. An overriding principle is that radiation safety measures should never in-
crease the surgical risk to the patient by increasing operating room time or other-
wise interfering with the operation.
All fissionable materials and radionuclides derived from fission products fall
under the auspices of the United States Nuclear Regulatory Commission (US NRC)
in many states. In the remaining states, referred to as “agreement states,” the regu-
latory control has been transferred to the state. The US NRC licenses radiation
safety committees at individual institutions to oversee radiation safety functions.
The US NRC also regulates disposal facilities. Typically the regulations in agree-
ment states are patterned after those of the NRC. The pertinent regulations are
found in Title 10, Chapter 1, Code of Federal Regulations –Energy, parts 20 and
35 (Nuclear Regulatory Commission, Washington DC 20555).
A recurring radiation safety question is whether surgical pathology specimens
need to be treated differently if they are radioactive. Unfortunately, the answer is
complex, confusing and at times, seemingly irrational. Generally, patients’ excreta,
body fluids and tissue are exempt from regulation. That is, following an injection
of Tc-99m sestamibi, a patient can use a public toilet and eliminate several milli-
curies of Tc-99m into the sewer system. This exemption has been made so that the
public is not denied the medical benefits provided by radionuclides because of
cumbersome, impractical regulations. In contrast to a patient, a nuclear medicine
technologist cannot empty a syringe containing several mCi of Tc-99m into a
toilet. Technologists and other radiation workers must hold radioactive waste until
it decays enough so no more radioactivity can be detected. This practice is called
“decay in storage” or “DIS”. For Tc-99m, DIS is not particularly burdensome since
Tc-99m has a half-life of only 6 hours. After 10 half lives (60 hours), only 1/1000th
of the original activity remains. In most cases, this small residual activity can no
longer be detected and the waste can be disposed of in the appropriate nonradio-
active waste stream. Standard procedures for DIS require that written records are
kept documenting that waste was surveyed before disposal.
The exception to the exemption of patient excreta, body fluids and tissues from
regulation is when the material is specifically collected to be assayed for radioac-
tivity. For example, urine which is collected specifically for radioassay can no longer
be disposed of by flushing it down the toilet. Once it has been assayed, the urine
must be held for DIS.
Radiation Safety 37
Similar rules apply for surgical pathology specimens. If a specimen is radioac-

tive because the patient happened to have a nuclear medicine study shortly before
surgery, the specimen is exempt from regulations. On the other hand, if the pa-
tient was injected with a radioactive tracer prior to surgery for the purpose of
2
radioassay, the specimen is subject to regulation. The exact requirements govern-
ing the handling of the specimen vary depending on the local regulations and the
local interpretation of the regulations.
In states regulated by the Nuclear Regulatory Commission (NRC), the legal
labeling requirements are specified in NRC regulation 10 CFR part 20.1904, with
activity levels requiring labeling listed in appendix C. For Tc-99m this level is
1.0 mCi, while for some radionuclides it is more restrictive (this is especially true
for I-131, which must be labeled if activity exceeds 1.0 mCi). If this activity level is
exceeded, the container must be labeled with a radioactivity trefoil symbol, the
radionuclide, amount of activity, time and date of the measurement, and the ini-
tials of the person who made the measurement. The majority of surgical pathol-
ogy specimens contain insufficient radioactivity to require labeling by law. Speci-
mens likely requiring detailed labeling include breast tissue from the injection site
obtained after lymphoscintigraphy (the specimen may contain 1.0 to 2.0 mCi of
Tc-99m) Tissue from the injection site in patients with melanoma are unlikely to
require detailed labeling since usually less than 1 mCi of Tc-99m is injected.
For specimens containing less than 1 mCi of Tc-99m, no labeling may be nec-
essary if the procedures for processing the specimens are reviewed by the radia-
tion safety committee and they conclude that labeling is not necessary. Labeling
would not be necessary if anticipated doses to the pathology staff are small
(< 100 mrem/year; almost certainly true) and if DIS is already implemented since
specimens are not discarded for a minimum of 5 days after surgery.
As shown in Table 2.1, the dose to a pathologist from handling radioactive
tissue samples is very low and should not prevent initial processing of samples
and preliminary histological determinations. Nor is it necessary to take special
precautions for more complete processing. However, in the interest of taking simple
measures that can make small doses even smaller (ALARA principle), personnel
at some institutions delay final processing of the sample for 24 hours. The activity
will have decayed to one-sixteenth its original value (for Tc-99m). This should
only be considered if such a delay will not adversely affect patient care.
Based on the amounts of activity used in lymphoscintigraphy, it is not legally
necessary to survey operating rooms that are used for lymphoscintigraphy. Simple
calculations show that any anticipated dose rate to personnel from blood in the
operating room is well below that allowed to members of the general public. The
cleaning that is done to protect personnel and patients from significant biological
hazards is more than sufficient for protection against radiation hazards.
Overall, it is important to emphasize that radioactivity exposure in the hospi-
tal setting must be monitored and approved by the radiation safety committee at
that institution, particularly when it involves new procedures such as sentinel lymph
node biopsy and MIRP. The guidelines presented in this chapter cannot be adopted
at individual institutions without the guidance of these monitoring bodies. How-

ever, clinical data and radiation dose estimates clearly suggest that these proce-
dures may be done safely at all hospitals with experience in handling these
substances.
2 References
1. Sorenson JA, Phelps ME. Physics in Nuclear Medicine, 2nd edition. Philadel-
phia: W.B. Saunders 1987.
2. Thrall, Ziessman. Nuclear Medicine, The Requisites. St. Louis: Mosby 1995.
3. Huda W, Boutcher S. Should nuclear medicine technologists wear lead aprons? J
Nucl Med Tech 1989; 17:(1) 6-11.
4. Serreze HB, Entine G, Bell RO, Wald FV. Advances in CdTe gamma-ray detectors.
IEEE transactions on Nuclear Science. 1974; NS-21(1) 404-407.
5. Siffert P, Cornet A, Stuck R. Cadmium telluride nuclear radiation detectors. IEEE
transactions on Nuclear Science. 1975; NS-22(1):211-225.
6. TerPogossian MM, Phelps ME. Semiconductor detector systems. Semin Nucl Med
1973; 3:343-365.
7. Gulec SA, Moffat FL, Carroll RG, Krag DN. Gamma probe guided sentinel node
biopsy in breast cancer. Quarterly J Nucl Med 1997; 41(3):251-261.
8. Huda W, Bews J, Gordon K, Sutherland JB, Sont WN, Ashmore JP. Doses and
population irradiation factors for Canadian radiation technologists (1978 to
1988). Can Radiol Assoc J 1991; 42:(4)247-252.
9. Harding LK, Mostafa AB, Roden L, Williams N. Dose rates from patients having
nuclear medicine investigations. Nucl Med Commun 1985; 6:191-194.
Training and Credentialing 39
Training and Credentialing Physicians

in Radioguided Surgery
Eric D. Whitman 3
Introduction .......................................................................................................... 39
Training ................................................................................................................ 39
Didactic Learning ................................................................................................. 41
Experiential learning ............................................................................................ 42
Credentialing ........................................................................................................ 42
Summary .............................................................................................................. 45
INTRODUCTION
As with any new surgical procedure, there are many technical aspects of
radioguided surgery that must be mastered. In addition to the surgical techniques,
this handbook also describes the multidisciplinary aspects of sentinel node pro-
cedures, specifically the essential roles of nuclear medicine and pathology in fa-
cilitating the identification of the sentinel nodes, and the optimal examination of
the nodes after excision, respectively. This chapter approaches the evolving clini-
cal applications of intraoperative lymphatic mapping from a more pragmatic view-
point: how physicians are trained and subsequently credentialed to perform
radioguided surgery.
TRAINING
Since the first reports of sentinel node procedures in the early 1990s, continu-
ing medical education (CME) courses have been given at various institutions to
train physicians to safely and correctly perform sentinel node mapping and bi-
opsy. These courses were initially given at some of the first centers to publish their
results. More recently, courses have been held at multiple other institutions. The
quality and content of these courses is relatively unregulated; while most are struc-
tured to legally provide CME credits, there are no national standards or educa-
tional goals for training in sentinel node biopsy.
A typical course generally covers the following topics:
• Surgical techniques
- Melanoma (sentinel lymph node, i.e., SLN)
- Breast cancer (SLN)
- Other skin cancers (SLN)

- Parathyroid (adenoma resection)
• Nuclear Medicine
• Pathologic examination
- Melanoma
- Breast cancer, including touch prep
3 • Conscious sedation
• Observation of live surgical cases
• Animal laboratory of sentinel node biopsy (optional)
Newer additions to courses include the utilization of telemedicine to project
interactive live audio and video content directly from the operating room to a
large, remote, auditorium. Many courses appear to lack extensive discussion of
some issues that are particularly relevant to the practicing surgeon, such as how
surgeons should be credentialled in these techniques, how an institution might
actually organize and promote a radioguided surgery program, and how these
procedures are appropriately coded for billing purposes.
Empirically, what should the ideal training program for radioguided surgery
include? Note first that I use the word, “program,” instead of “course.” In current
usage, a course is generally an instructional event with a defined time frame, held
at a place remote to where the actual clinical practice will occur. Courses, from a
practical standpoint, have to condense information to fit the large amount of nec-
essary clinical data into the time span of the course. A program would account for
and emphasize the ongoing nature of any learning situation, where training takes
place over a period of time and, potentially, places. A program would include a
course component, but extend the effectiveness of the course format by including
different educational modalities and allowing for proctoring or “post-training”
evaluation of trainees.
The ideal training program for radioguided surgery would instruct interested
physicians in the most up-to-date clinical techniques and outcomes data regard-
ing these procedures, as is currently the practice. However, the ideal program would
also include sessions on organizational aspects, credentialing issues, reimburse-
ment, and data collection, all topics of crucial importance to practicing clinicians
but often glossed over by course instructors, generally physicians at large aca-
demic medical centers with readily available capabilities in these areas. Most of
the physicians attending courses will need to consider these issues carefully when
they return to their home institution, as the infrastructure necessary to process
and/or organize these essential components may not be in place. Therefore, the
ideal training program should provide the clinicians in attendance with informa-
tion that will assist them in the logistical aspects of initiating a radioguided surgi-
cal clinical service. This may include sample protocol documents, consent forms,
data collection sheets, database program files, press releases, and letters to refer-
ring physicians, in short providing the program’s students with a business plan
describing how a radioguided surgical program is initiated and maintained.
Lastly, the ideal training program should enable the faculty or appointed sur-
rogates to personally supervise the first few radioguided surgical procedures
performed by the student. Current courses make no provision for this; a physician
is trained for a day or two, then returns to his or her institution without being
“checked-out” on the new procedures or necessary skills (on humans), putting
pressure on the physician to make a “leap of faith” that the procedure(s) will pro-
ceed appropriately in his or her hands, and also putting pressure on the home
institution to judge if this physician should be permitted to put patients at risk
during performance of the procedure after such a limited instructional experi- 3
ence (i.e., credentialing, see below). Fixing this problem may involve the use of
portable telemedicine modules, to lessen the travel needs of the faculty. Alterna-
tively, “certified” or previously trained physicians in radioguided surgery may be
commissioned by the course directors to act as surrogate instructors, completing
forms that are mailed back to the training program director. This proctoring may
also involve continued education and student evaluation via world wide web sites,
where a student may need to correctly evaluate a set of images or plan patient care
based on hypothetical patient histories posted on a web site, then submit their
responses to the program faculty. This internet-based “post-testing” might also
include instructional audio or video files that could be downloaded for a certain
amount of time after official course attendance, based on password access with a
90 day expiration, for example.
To address these concerns, a newer education program is currently being de-
signed, involving many of the authors of chapters in this textbook. The goal of
this program is to comprehensively prepare all attendees to return to their home
institutions and immediately be able to implement a clinically superior, well-or-
ganized, radioguided surgery program. The program will utilize telemedicine hard-
ware and software to create a “virtual” learning experience, where class attendees
could be at any one of a number of geographically separate sites and have an
identical educational experience. Tentative course structure includes:
DIDACTIC LEARNING
• Surgical techniques: SLN for melanoma, breast cancer, other skin can-
cers, and other malignancies with emerging applications
• Surgical techniques: minimally invasive parathyroidectomy
• Nuclear medicine
• Radiation safety
• Lymphoscintigraphy
• Pathology
• Melanoma
• Breast cancer
• Coding/Billing/Reimbursement
• Credentialing issues
• Implementation of a radioguided surgical program: a business plan
EXPERIENTIAL LEARNING
• Observation of live surgery via telemedicine

• Multiple sites
• Multiple procedures
• Multiple surgeons
3 • All interactive with all audiences in real-time
• Animal laboratory of sentinel node mapping (remains optional)
This course is available beginning January, 1999 and information can be ob-
tained by calling 1-888-456-2840.
The importance of excellent training in these evolving techniques cannot be
overemphasized, particularly since most of the patients that are candidates to un-
dergo radioguided surgery have cancer. Inaccurate or inappropriate use or imple-
mentation of these procedures could lead to devastating results in 5-10 years, as
patients suffer the long-term ill effects of inadequate or misguided treatment of
their cancers. Responsibility for this issue falls to both the leading experts in the
field and the clinicians who introduce these techniques and procedures into their
clinical practice. The experts must continue their investigations to optimize the
technical and procedural aspects of radioguided surgery. They must also continue
to dedicate their time and energy towards providing superior instruction for other
clinicians in these techniques and towards the design and standardization of edu-
cational programs and credentialing criteria. The clinicians in some ways shoul-
der the more arduous task: they ultimately will treat the majority of patients with
these techniques, and therefore it is imperative that all physicians receive adequate
instruction, both didactic and experiential, and also establish guidelines at their
own institutions for the safe and correct way to employ this technology.
CREDENTIALING
Although physicians of multiple specialties may (and do) attend the existing
CME courses, the majority of attendees seem to be surgeons, and it appears that
the majority of credentialing issues will center on the actual performance of the
surgical procedure, not the lymphoscintigraphy and pathologic examination.
Lymphoscintigraphy is a well-described technique,1 and exact protocols to per-
form the radionuclide scan for radioguided surgical procedures are available from
multiple published sources. Similarly, the histopathologic techniques are also well
described in the medical literature.2 The technical aspects of performing the nec-
essary pathologic examinations are utilized by pathologists for other indications
(or can easily be obtained through an outside specialty laboratory) and it there-
fore does not seem necessary in most cases to train pathologists in these tech-
niques. The most important aspects of developing a radioguided surgical pro-
gram at an institution from both the radiologic and pathologic perspectives will
be the implementation by both departments of a consistent, reproducible approach
to all patients undergoing these procedures (see chapter 1).
The emergence of radioguided surgical techniques and procedures is analo-

gous in many ways to the rapid adoption of laparoscopic surgery less than a de-
cade ago. In both cases, new technology and creative surgical perspectives com-
bined to create a set of revolutionary surgical procedures. With laparoscopic sur-
gery, almost overnight most if not all general surgeons in the United States desired
training, and wanted to begin performing minimally invasive surgical procedures.
Many hospitals were ill-prepared for the inevitable credentialing issues: how to 3
regulate the conditions under which surgeons would be allowed to utilize these
new techniques.
Unfortunately, when surgeons began performing laparoscopic procedures with-
out formal training or without firm hospital or governmental regulatory policies,
the incidence of cholecystectomy-related complications, particularly bile duct in-
juries, increased. In response to these problems, most hospitals established mini-
mal criteria to allow surgeons to perform laparoscopic surgery. In New York state,
a law was passed defining the training and experience necessary for a surgeon to
be allowed to perform laparoscopic surgery. Despite these precautions, the inci-
dence of bile duct injury remains 2.5-4.0 x higher after laparoscopic cholecystec-
tomy than in the open cholecystectomy “era”, in a recent study.3
The issues for radioguided surgery are also somewhat more complex, because
the surgery itself is probably less morbid, or with less morbid potential, than the
situation with laparoscopic vs. open cholecystectomy. However, radioguided sur-
gery is to date used primarily to stage malignancies, so that errors in technique
when performing these procedures might have a delayed adverse outcome, when
the patient develops an unexpected recurrence months or years later. In this situ-
ation, inaccurate or inappropriate use of radioguided surgical techniques might
have altered or deferred adjuvant therapy, with resulting life-threatening and/or
medicolegal implications.
The American College of Surgeons is taking a proactive position on other
emerging surgical technologies, specifically stereotactic breast biopsy,4 and it ap-
pears likely physician qualifications for radioguided surgery will also be proposed
by that organization. In a recent issue of the Bulletin of the American College of
Surgeons (ACS), the Board of Regents announced that they had approved a “pro-
cess by which its Fellows and Associate Fellows could be verified for the use of
emerging technologies.”5 The intent of this laudable process is to provide sur-
geons with documentation that they have the necessary training and experience
to be allowed by local credentialing facilities to utilize the new technology in pa-
tient care. Although not specifically mentioned in the announcement, the rapid
emergence of radioguided surgery for multiple clinical indications suggests that
this will be one of the early areas to be addressed by the Committee on Emerging
Surgical Technology and Education (CESTE), the group charged with developing
verification programs and standards.
At present, there are no universally adopted credentialing criteria for surgeons
to be permitted to perform radioguided surgery. At national meetings and in major
publications,6 the responsibility for credentialing surgeons has been transferred
to the individual hospitals, yet as previously noted there are no consistent training
guidelines. Pending the establishment of such criteria and guidelines, particularly

through the efforts of CESTE and ACS, the provisional regulations used at my
home institution, Missouri Baptist Medical Center of BJC Health System in St.
Louis, Missouri serve as an excellent example of how a hospital might control
usage of radioguided surgery.
Our criteria are based on experience, training, and proctoring. Initially, only
3 surgeons with prior experience at other hospitals or who had taken a CME-ac-
credited course were given “Sentinel Node Privileges.” This requirement was made
very clear to both the surgeons and the operating room staff: except for a limited
list of surgeons, no other surgeons would be allowed to perform radioguided sur-
gery unless they:
1) Took a CME-accredited course in radioguided surgery, or provide docu-
mentation of residency training experience in radioguided surgical
techniques;
2) Or, they could be proctored for a minimum of five radioguided surgical
procedures by an experienced surgeon (on a voluntary basis by a local
experienced surgeon; in all cases to date, the author); or through the
mentoring service offered at Moffitt Cancer Center. In this case, the proc-
tor would then send a letter to the Chief Surgeon of the hospital indi-
cating that the trainee surgeon in question had been proctored on a
certain number of cases and was proficient to perform radioguided sur-
gical procedures independently.
In this evolving policy, one partially unresolved issue is what would happen if
a surgeon went and took a CME-accredited course now, and then returned to the
hospital seeking radioguided surgical privileges. It is not clear if lack of supervised
experiential training in these techniques should then require one or two cases of
local proctoring by an experienced surgeon to ensure that the trainee surgeon in
question is indeed ready for independent performance of radioguided surgery.
Our operating room administrators and physician leaders are currently address-
ing that issue, with plans to publish detailed guidelines for credentialing at our
hospital.
What has happened to date at our institution is similar to what I understand is
happening elsewhere: the local surgeons are taking advantage of the experience
and interest of a solitary motivated colleague who is willing and able to train them
in the techniques during surgical procedures on the trainee’s own patients. For
example, if a surgeon without radioguided surgical privileges is referred a patient
who may be eligible for sentinel node biopsy, that surgeon would schedule the
patient for surgery, coordinating the operative schedule with my calendar to en-
sure my full availability. The nonprivileged surgeon explains to the patient that
this is a new technique and that their surgeon will be enlisting the aid of another
local surgeon who is more experienced in the technique. In this way, the doctor-
patient relationship is maintained, and the radioguided surgery may still be per-
formed. This method of obtaining privileges in radioguided surgery has been more
attractive to my colleagues than the time, expense, and travel necessary for an
official CME-accredited course. As the courses themselves (see earlier section)
and ACS/CESTE verification criteria evolve, it may become necessary for any fu-
ture surgeons seeking radioguided surgical privileges to enroll in more formal
training courses.
SUMMARY
3
As with any new technology, proper training and documented credentialing
standards are essential to the safe and clinically prudent implementation of
radioguided surgical procedures by physicians at individual hospitals. The lessons
learned during the advent of laparoscopic cholecystectomy, where the incidence
and type of operative biliary complications changed from the “open” era, should
be built upon to prevent similar occurrences from radioguided surgery, particu-
larly since most of the applications developed to date involve the treatment of
malignancies, where adverse outcomes may not become clinically apparent for
months or years.
Currently, training programs provide varying levels of didactic and experien-
tial learning activities. Supervision or oversight/review (i.e., proctoring) of newly
trained physicians after the course is complete is nonexistent. At our institution,
we have implemented guidelines that ensure that patients only receive radioguided
surgical procedures by those surgeons with documented training or experience in
these techniques, or under the direct, hands-on, supervision of another surgeon
with the necessary experience.
Ultimately, it is my hope that a large governing body such as the American
College of Surgeons will step forward and take steps to ensure patients, third party
payors, and other physicians that a given surgeon has obtained adequate training
to allow him or her to utilize radioguided surgical techniques in clinical practice.
This desirable level of credentialing can only be possible with standardization of
educational objectives and experience by existing or future training courses.
ACKNOWLEDGMENTS
The author wishes to thank Dr. Douglas Reintgen, Dr. Charles Cox and Dr.
James Norman, all of H. Lee Moffitt Cancer Center and the University of South
Florida, Gene Theslof, of Infomedix Communications, Inc., and Eric C. Miller, of
United States Surgical Corporation, for their invaluable thoughts and ideas on the
subject of training programs for radioguided surgery.
REFERENCES
1. Reintgen D, Cruse CW, Wells K et al. The orderly progression of melanoma nodal
metastases. Ann Surg 1994; 220(6):759-767.
2. Turner RR, Ollila DW, Krasne DL, Giuliano AE. Histopathologic validation of
the sentinel lymph node hypothesis for breast carcinoma. Ann Surg 1997; 226:
271-278.
3. Strasberg SM, Hertl M, Soper NJ. An analysis of the problem of biliary injury
during laparoscopic cholecystectomy. J Am Coll Surg 1995; 180:101-125.
4. Physician qualifications for stereotactic breast biopsy: A revised statement. Bull

Amer Coll Surg 1998; 83 (5):30-33.
5. Verification by the American College of Surgeons for the use of emerging tech-
nologies. Bull Amer Coll Surg 1998; 83 (5):34-35.
6. Reintgen D. The credentialing of American surgery. Ann Surg Oncol 1997;
4:99-101.
3
SLN Melanoma: Scientific Support and Biological Significance 47
Melanoma Lymphatic Mapping:

Scientific Support for the Sentinel
Lymph Node Concept and Biological
Significance of the Sentinel Node
4
Merrick I. Ross, Jeffrey E. Gershenwald
Introduction .......................................................................................................... 47
Scientific Support for the Sentinel Node Concept ................................................. 48
Technical Advances ............................................................................................... 50
Biolo gic Significanc e of the Sentinel Node ............................................................ 52
Accurate Nodal Staging ........................................................................................ 55
Clinical Trials ....................................................................................................... 57
Concluding Comments ......................................................................................... 60
INTRODUCTION
The lifetime incidence of melanoma will reach an all time high of 1 in 75 indi-
viduals by the turn of the century.1 Fortunately, as a result of increased public
awareness and screening programs, the vast majority of newly diagnosed patients
will present with clinically localized (stage I and II) disease. While patients with
thin melanomas are likely to be cured following appropriate surgical excision of
the primary tumor alone, a significant percentage of individuals with tumors that
are thicker or exhibit other unfavorable prognostic factors, harbor clinically un-
detectable regional lymph node metastases. Such disease may serve as a potential
source or predictor of subsequent nodal failure and distant dissemination. The
routine use of lymphadenectomy targeted to those patients possessing an increased
risk of microscopic nodal disease—an approach popularized as elective lymph
node dissection (ELND)—has not consistently been shown to improve survival.2,3
Accordingly, the role for ELND in the initial surgical management of these pa-
tients has been controversial.4 Historically, the motivation to perform ELND was
strong, primarily because surgery was the most effective melanoma treatment
modality, since no effective systemic adjuvant therapy had yet been identified.
The major obstacles which prevent widespread acceptance of ELND are the
following: 1) conflicting results from clinical trials which assessed survival of pa-
tients who received ELND compared to patients who only underwent lym-
phadenectomy as treatment for clinically detectable regional lymph node failure
(therapeutic lymph node dissection),2 2) unnecessary morbidity and costs associ-

ated with formal lymphadenectomy in patients without evidence of microscopic
nodal metastases, and 3) the common (but never proven) belief that lymphadenec-
tomy in patients with undetectable metastases may alter the regional immune
system and render the patient vulnerable to recurrences.5 Despite these arguments,
those promoting ELND remain concerned about the delay in treatment of lym-
phatic disease until it is clinically palpable, when the tumor burden and the num-
ber of lymph nodes involved is greater and overall prognosis is worse.6-8
Amidst this long-standing controversy has emerged a rational approach to the
4 clinically negative regional lymph node basin known as “selective lymphadenec-
tomy”, using the techniques of lymphatic mapping and sentinel node biopsy
(LM/SNB), initially introduced in the surgical literature by Don Morton.9 This
approach is based on the concepts that the first draining lymph node—“the sen-
tinel node”—is the lymph node most likely to contain melanoma metastases, and
that a minimally invasive surgical technique can consistently and accurately iden-
tify this lymph node. While the initial motivation for LM/SNB was to perform
lymph node dissections only in patients with proven microscopic disease (early
therapeutic dissection) and spare those patients without nodal metastases the risks
and cost of unnecessary surgery, the recently reported improvement in survival of
node positive patients treated with high-dose interferon (Intron A, Schering In-
corporated, Kenilworth, NJ) in the adjuvant setting has promoted accurate lymph
node staging as another and probably more important motivation.10
SCIENTIFIC SUPPORT FOR THE SENTINEL NODE CONCEPT
Lymphatic mapping, as it applies to melanoma, relies on the hypothesis that

the dermal lymphatic drainage from specific cutaneous areas to the regional lymph
node basin is an orderly and definable process. These lymphatic drainage patterns
should mimic how melanoma cells spread within the lymphatic compartment
such that the first lymph node(s) receiving lymphatic drainage are the most likely
to contain metastatic disease. In theory, each lymph node within a formal basin
may potentially function as a sentinel node, but for different and finite regions of
the skin. In addition, in a patient whose melanoma arises from a cutaneous site
with potential drainage to more than one lymph node basin (i.e., trunk), sentinel
nodes exist in each basin to which the primary tumor drains.
To test this hypothesis, a reliable method of SLN identification had to be es-
tablished. Pioneers of this technique performed preclinical animal studies which
evaluated a variety of dyes that could be intradermally injected and transported
through the lymphatic system to the regional lymph node basin, thereby provid-
ing a visualization of the sentinel node upon surgical exploration of the lymph
node basin. Morton and colleagues demonstrated that two dyes were most effec-
tive: isosulfan blue (Lymphazurin) and Patent Blue V.9 Initial clinical studies were
performed in melanoma patients to determine the following: 1) sentinel node
identification rates, and 2) the accuracy of the sentinel node in establishing the
presence or absence of regional nodal metastases.

The first report published in 1992 evaluated 237 consecutive patients using
intradermal injections of Lymphazurin around the intact primary melanoma or
excisional biopsy site.9 The authors demonstrated an 82% sentinel node identifi-
cation rate and an average of 1.3 SLNs were removed per basin. Subsequent stud-
ies from the M. D. Anderson Cancer Center, Moffitt Cancer Center, and the Sydney
Melanoma Unit in Australia reported similar findings.11-13 Accuracy assessment
was similarly accomplished at all of these centers through the use of synchronous
ELND performed at the time of the sentinel node biopsy in order to ascertain the
false negative rate. For these studies, a false-negative event was defined as the de- 4
tection of microscopic disease in a non-sentinel node when the sentinel node from
the same basin was histologically negative. Accordingly, the false-negative rate was
then calculated as the number of false-negative events divided by the total num-
ber of patients with microscopic nodal disease. Collectively, these initial studies
evaluated 472 patients, 86 of whom were found to have regional node metastases.
A false negative rate of 5% and an overall accuracy of 95% was established. These
data strongly support the sentinel lymph node concept (Table 4.1).
Additional evidence that regional lymph node metastasis is an orderly and
non-random event is provided from studies performed at the M. D. Anderson
Cancer Center. Investigators found that the sentinel node was the only node in-
volved in 83 (79%) of the basins with at least one metastasis, while additional
disease was identified in only 21% of the lymphadenectomy specimens.14a A sub-
sequent analysis demonstrated that 68% of all sentinel nodes and only 1.8% of all
non-sentinel nodes in 105 therapeutic lymphadenectomy specimens contained
metastatic disease when at least one sentinel node was positive for melanoma
(Table 4.2).
As a result of the tremendous interest generated from these studies, many cen-
ters subsequently adopted the concept of selective lymphadenectomy when man-
aging newly diagnosed intermediate and high risk Stage I and II melanoma pa-
tients.27 Prospective clinical programs were established to enroll patients with the
following primary tumor characteristics: Breslow tumor thickness of at least 1 mm
or Clark’s level of invasion greater than III, or the presence of histologic ulceration.
Table 4.1. Sentinel node identification and accuracy of initial studies: dye injection alone
Study Year Patients SLN ID +SLN +ELND False-
N Rate N N(%) Negative Rate
N (%) N(%)
UCLA 1992 237 194 (82%) 38 40 (21%) 2 (5%)

MDACC/MCC*, 1993, 1994 117 103 (86%) 23 24 (20%) 1 (4%)
Sydney Melanoma 1995 118 105 (89%) 20 22 (21%) 2 (%)
Unit
Total 472 402 (85%) 81 86 (21%) 5 (5.8%)
*M.D. Anderson Cancer Center, Moffitt Cancer Center

Table 4.2. Additional metastases in therapeutic lymph node dissection after at least 1
positive SLN identified
Total No. positive lymph node basins 105

Basins with additional positive SLNs, N (%) 22 (20)
Total No. LSNs harvested 183
Positive SLNs, N (%) 125 (68)
No. non-SLNs harvested 1883
Positive non-SLNs, N (%) 34 (1.8)
4 Adapted from Gershenwald et al. Improved sentinel lymph node localization in primary
melanoma patients with the use of radiolabeled colloid. Surgery 1998; (In press).
Patients with a histologically positive SLN underwent therapeutic lymphadenec-

tomy, while those with a negative SLN were observed. This large clinical experi-
ence yielded significant improvements in SLN localization techniques, generated
additional findings which supported the sentinel lymph node concept, and pro-
vided valuable insights into the biologic significance of the SLN.
TECHNICAL ADVANCES
Although initial sentinel node biopsy experiences using the blue dye only tech-
nique provided a promising beginning, it was clear that room for improvement
existed since initial sentinel lymph node identification rates were only 80% to
85%.9,11-13 Prior experience with preoperative cutaneous lymphosintography has
also improved the technique of SLN localization. This approach has provided a
reliable and accurate way to identify regional lymph node basins at risk for me-
tastases in patients whose melanoma arose in anatomic regions of ambiguous
lymphatic drainage.14,15 Application of this technology to lymphatic mapping
clearly delineates the lymphatic drainage patterns from primary tumor injection
sites to the regional nodal basin(s) at risk.16 Afferent lymphatic vessels are well-
visualized and relative location and number of sentinel nodes within the regional
nodal basin(s) are established. Importantly, these techniques also helped to iden-
tify the uncommon but clinically relevant sentinel nodes that are located outside
of the formal regional nodal basin, termed “in-transit” sentinel nodes.16 Further-
more, the presence, if any, of direct lymphatic drainage to the popliteal or epitro-
chlear regions can also be established for primary melanomas arising distal to the
knee or elbow, respectively. These high-resolution scans provide preoperative road
maps for the surgeon and aid in the intraoperative identification of the blue-stained
sentinel node(s).
The intraoperative use of a handheld gamma probe, capable of detecting the
accumulation of intradermally injected radiolabeled colloid within sentinel nodes,
provided the next important technical advance.14a,17,18 This technique provided
the surgeon with a sensitive instrument with which to transcutaneously localize
the sentinel node prior to the skin incision and with an adjunct to the visual cues
required to localize the blue-stained sentinel node at the time of surgical explora-
tion. Preclinical animal models of lymphatic drainage performed by Nathanson
provided useful information concerning the kinetics of different colloid prepara-
tions.19 Clinical studies from the Moffitt and M. D. Anderson Cancer Centers dem-
onstrated that the ratios of radioactivity within the sentinel node compared to
neighboring non-sentinel nodes actually increased with time, providing surgeons
a window of opportunity during which to perform the SLN biopsy after injection
of radioactive colloid in the nuclear medicine department.18 These data demon-
strated that the particles in sulfur colloid preparations were large enough to be
actively phagocytized by macrophages within a lymph node. The majority of ma- 4
terial reaching the nodal basin was concentrated within the first draining lymph
node—the SLN—-with only minor passage of colloid to secondary echelon (non-
sentinel) nodes. Improved SLN identification rates using combined modality tech-
niques compared to blue dye alone have been obtained from the Netherlands20
and the M. D. Anderson Cancer Center.14a In these reports, identification rates
increased to 99% with the largest incremental improvement in the cervical (neck)
and axillary regions (Table 4.3).
In a multicenter trial reported by Krag et al,17 identification rates of greater
than 95% were obtained when injections of radiolabeled colloid alone were pri-
marily employed. Unfortunately, concomitant ELNDs were not performed in this
or other SLN studies in which radiolabeled colloid injections were utilized. As a
result, questions have emerged whether or not the SLN localized in this manner
was definitively the first lymph node of drainage and thus an accurate indicator of
regional nodal metastases. The 16% incidence of SLN involvement reported by
Krag was similar to the 20% incidence previously demonstrated with blue dye
injections, and alleviated some of the concerns.17 A recent study from the M. D.
Anderson Cancer Center using both localization techniques more convincingly
supported the concept that the node identified with the radiolabeled colloid is the
sentinel node: 1) in patients with microscopic nodal disease and more than one
sentinel node identified, the metastases were present within the lymph node con-
taining the highest radioactivity 92% of the time, and 2) co-localization of blue
dye and radioactivity occurred in at least 93% of the sentinel lymph nodes
identified.14a
Table 4.3. SLN identification rate according to basin mapped and SLN biopsy technique
Basin All Patients Dye Alone Dye & Colloid
N (%) N (%) N (%) P-Value
Cervical 42/47 (89) 5/9 (56) 37/38 (97) 0.002

Axillary 413/445 (93) 142/169 (84) 217/276 (98) < 0.00001
Inguinal 222/227 (98) 93/98 (95) 129/129 (100) < 0.03
Total 766/719 (94) 240/276 (87) 437/443 (99) < 0.00001
Adapted from Gershenwald et al. Improved sentinel lymph node localization in primary
melanoma patients with the use of radiolabeled colloid. Surgery 1998 (In press).
Interestingly, the number of sentinel nodes identified per nodal basin was
greater when both modalities were employed compared to blue dye alone (1.74
versus 1.31)(Table 4.4).14a The following explanation is offered to account for this
finding. The first blue lymph node encountered within a lymph node basin may
be only one of the sentinel nodes present in that basin, or may actually be a sec-
ondary echelon node that received blue dye via the efferent lymphatic channels of
a sentinel node located in a deeper, or perhaps different region of the basin. Fail-
ure to detect the sentinel node(s) or additional sentinel nodes may result from use
of the blue dye alone, since it is difficult to completely examine the entire basin if
4 one relies solely on visualization of blue dye. In contrast, intraoperative use of the
gamma probe provides a method of detection that is independent of and more
sensitive than visual cues. Blue-stained sentinel nodes that are initially undetected
or sentinel nodes which did not receive a sufficient amount of blue dye to be
visualized because of technical problems are likely to be located with the gamma
probe (Fig. 4.1).
BIOLOGIC SIGNIFICANCE OF THE SENTINEL NODE
Studies have demonstrated that the incidence of SLN metastases correlates

directly with increasing tumor thickness (Table 4.5).14a This strong correlation is
nearly identical to that of the incidence of microscopic nodal disease in ELND
specimens when stratified by tumor thickness.5 Other known primary tumor fac-
tors such as anatomic location, ulceration, and Clark’s level of invasion, also pre-
dict sentinel node involvement.14a In a multivariate analysis, the two variables that
independently predicted SLN involvement were tumor thickness and the pres-
ence of ulceration (Table 4.6). The finding that the most powerful primary tumor
prognostic factors were also the best predictors of SLN metastases offers further
evidence that SLN involvement is a biologically important and non-random event.
Further support for the SLN concept is provided from long-term follow-up of
patients who have undergone a negative SLN biopsy and no further surgical inter-
vention in the lymph node basin. In a cohort of almost 250 patients followed for
over three years, only 3% of the patients have died of recurrent disease, 10% of the
Table 4.4. SLNs identified according to SLN biopsy technique

All Patients Dye Alone Dye & Colloid P-Value
Mean 1.59 1.31 1.74 < 0.00001
Median 1 1 2 –
Range 1–7 1–4 1–7 –
Secondary
Echelon Nodes
(Non-Sentinel)
Efferent Lymphatic
Sentinel Channels
Nodes
In-transit 4
Sentinel Nodes
Afferent Lymphatic
Channels
Primary
Melanoma
Fig. 4.1. Potential lymphatic drainage patterns. By definition, the first node(s) of drainage identified fol-
lowing injection of blue dye or radiolabeled colloid that travels through afferent lymphatics from the
primary tumor are termed sentinel nodes. Some of the blue dye as well as small amounts of colloid may
pass to secondary echelon nodes via efferent lymphatic channels.
Table 4.5. Incidence of positive sentinel lymph nodes according to tumor thickness
Tumor Thickness Patients Patients with Positive SLN
(mm) N N %
< 1.00 123 4 3.2

1.01–2.00 242 25 10.3
2.01–3.00 86 19 22.1
3.01–4.00 69 26 38.5
4.01 + 70 31 44.0
Table 4.6. Prognostic factors influencing SLN pathologic status

Multiple Covariate
Prognostic Factor Univariate P-Value Odds Ratio CI P-Value
Tumor Thickness < 0.0001 1.23 1.11–1.36 0.0001

Clark Level > III 0.01 1.38 0.83–2.31 0.21
Axial Location 0.17 1.31 0.83–2.09 0.25
Ulceration < 0.001 2.50 1.53–4.07 0.0002
Abbreviations: CI, 95% confidence interval for odds ratio

patients have had some type of recurrence, and 4% developed failure within the
previously mapped regional nodal basin as the sole site or a component of the
first site of failure.21 Such regional nodal basin failures represent another type of
false-negative event. Three mechanisms have been proposed to explain nodal fail-
ure after a negative sentinel node biopsy:
1) TECHNICAL FAILURE
The lymphatic-mapping technique did not identify the primary node
of drainage and a non-sentinel node was removed rather than the sen-
4 tinel node. This unidentified sentinel node contained microscopic dis-
ease and provided the ultimate source of failure within that basin.
2) PATHOLOGIC FAILURE
One of the sentinel nodes was removed and contained microscopic dis-
ease undetected by conventional histologic techniques. An additional
sentinel node or a non-sentinel node with disease remained as the source
of subsequent clinical failure.
3) BIOLOGIC OR NATURAL HISTORY-TYPE FAILURE

At the time of the initial surgery, the correct sentinel node was removed
and no metastatic disease was present; however, subclinical intra-
lymphatic in-transit disease present since diagnosis provided the ulti-
mate source of subsequent nodal failure.
A more careful histologic examination of the SLNs in patients who developed
regional nodal failure as a component of the first site of recurrence after a nega-
tive SLN biopsy was also performed. In these studies, the original blocks were
subjected to hematoxylin and eosin examination (H&E) of step sections and im-
munohistochemical staining using antibodies to HMB-45 and S-100. Interestingly,
microscopic disease initially undetected by conventional histologic techniques was
identified in 8 of the 10 patents who had failed in the nodal basin as a component
of the first site of recurrence.21 Therefore, only two of these nodal failures can be
truly classified as false negative events resulting in an actual overall false negative
rate of <1%. The most common mechanism for failure after a negative sentinel node
biopsy was actually related to the pathologist’s inability to detect microscopic disease
using only a conventional approach (H&E staining of bivalved lymph nodes), rather
than the surgeon’s failure to identify the first node of drainage. These data therefore
provide additional scientific support that the SLN, as defined by these minimally
invasive techniques, is the most likely node to contain metastatic disease.
Long-term follow-up of large numbers of patients who underwent SLN bi-
opsy and selective lymphadenectomy has also revealed valuable prognostic infor-
mation. With a median follow-up of more than three years, prognostic factor analy-
ses of patients from a large database from the M. D. Anderson and Moffitt Cancer
Centers demonstrated that the histologic status of the sentinel node was the most
powerful predictor of overall survival when compared to previously described
primary tumor factors (Table 4.7).22 Using this information, prognostically
Table 4.7. Univariate and multivariate analyses of prognostic factors influencing disease-
specific survival in stage I & II melanoma patients
Prognostic Factors Univariate P-Value Multivariate P-Value
Age .93 .62

Gender .14 .68
Thickness < .0001 .02
Clark Level > III .005 .07
Axial Location .003 .01
Ulceration < .003 .33 4
SLN Status < .0001 .0001
Adapted from Gershenwald et al. Patterns of failure in melanoma patients after successful
lyjmphatic mapping and negative sentinel node biopsy. 49th Annual Meeting of The
Society of Surgical Oncology, Atlanta, GA; 1996.
disparate subsets of stage I and II patients can be defined; combined with the
availability of effective adjuvant systemic therapy in node positive patients, stag-
ing has emerged as a critical motivation for performing a sentinel lymph node
biopsy (Fig. 4.2).
ACCURATE NODAL STAGING
The importance of accurate nodal staging is 2-fold:

1) It effectively determines which patients harbor clinically occult nodal
metastases and who can therefore benefit from earlier therapeutic node
dissections, both from the standpoint of improved local regional con-
trol (lymphadenectomy is performed when the tumor burden is low)
and potentially improved survival.
2) It identifies those clinical stage I and II patients who are actually occult
stage III and who are therefore at higher risk for distant failure. These
patients may benefit from earlier administration of systemic adjuvant
therapy or may be offered participation in adjuvant therapy trials ear-
lier in the course of their disease. Components of accurate nodal stag-
ing include proper and accurate identification of the sentinel nodes as
well as careful histologic examination of the nodal specimen. Although
the pathologic examinations which define careful histologic examina-
tion continue to evolve, it is clear that analysis of one or two special
nodes (the sentinel nodes) can be more rigorous compared to the histo-
logic evaluation of all lymph nodes submitted following an ELND.
Historically, the standard approach to the evaluation of SLNs was to bivalve a
clinically negative node and stain a section from each half with H&E. This tech-
nique probably samples less than 5% of the entire volume of the lymph node and
likely represents the most important reason for the underestimation of regional
Fig. 4.2. Disease-specific survival. Kaplan-Meier survival for patients undergoing successful lymphatic
mapping and SLN biopsy stratified by SLN status. Disease-specific survival was significantly better for
patients with a negative SLN biopsy (p < 0.0001).
nodal disease in stage I and II patients. For example, the incidence of nodal failure
following surgical excision alone for primary melanomas 1.5-4.00 mm (stage IIa)
is approximately 40-50%, while the incidence of microscopic nodal disease in
ELND or sentinel node biopsy specimens is approximately one-half that predicted
by natural history data alone (Table 4.8).18,23 While subsequent nodal failure may
in part result from microscopic in-transit disease, several lines of evidence sup-
port the concept that nodal disease is more often present than is demonstrated by
conventional histologic techniques:
1) step sectioning studies (i.e., better sampling) improve the ability to de-
tect microscopic disease,24-26
2) 80% of patients who develop nodal basin failure after a negative senti-
nel node biopsy initially assessed by routine pathology are actually node
positive following more careful analysis of the paraffin blocks (see above)
and,22
3) prospective evaluation of sentinel lymph nodes using the polymerase
chain reaction (PCR) to detect the presence of messenger RNA (mRNA)
specific for the production of melanoma markers (i.e. tyrosinase) pro-
vides results more similar to the natural history data (Table 4.8).27,28
The obvious advantage of PCR analysis is that the entire lymph node can be
potentially evaluated. The clinical relevance of PCR findings are still under inves-
tigation, but preliminary data from the Moffitt Cancer Center suggest that the
PCR positive-H&E negative lymph node patients have survival rates intermediate
between those patients who are PCR positive-H&E positive and those who are
Table 4.8. Incidence of microscopic nodal disease comparison of histologic technique to

natural history
AJCC Stage Nodal Failure ELND + SLN + SLN +
(Natural History) H&E H&E PCR
I (< 1.5 mm) 10–12% 5% 5% 33%

IIa (1.5–4.0 mm) 40–45% 20% 20% 50%
IIb (> 4.0 mm) 60–70% 35% 35–45% 80%
From Reintgen D, Balch C, Kirkwood J, Ross M. Recent advances in the care of the patient 4
with malignant melanoma. Ann Surg 1997; 225:1-14.
Table 4.9. Clinical correlation of nodal status with reverse transcription polymerase chain
reaction (RT-PCR)
Nodal Status Number of Patients Recurrences Local Regional
Histology– 14 6 (42%) 2 4
RT-PCR +
Histology– 27 6 (22%) 3 3
RT-PCR +
Histology– 33 2 (6.6%) 1 1
RT-PCR +
From Reintgen D, Balch C, Kirkwood J, Ross M. Recent advances in the care of the patient
with malignant melanoma. Ann Surg 1997; 225:1-14.
both PCR and H&E negative (Table 4.9).27 Since the volume of material submit-
ted to the pathologist is not only small (1 or 2 nodes), but also the most likely to
contain microscopic disease, this approach provides an efficient way to prospec-
tively evaluate the importance of these more sensitive histologic techniques—PCR,
step sectioning and immunohistochemical analysis. Such staging information will
allow us to define prognostically homogenous subgroups and better stratify pa-
tients for adjuvant therapy trials.
CLINICAL TRIALS
LM/SNB provides an opportunity to ask important questions in the context of

clinical trials. Such endeavors will greatly impact our understanding of the fol-
lowing issues:
1) the natural history and staging of clinical stage I and II patients,
2) the local-regional control and/or survival benefits for early therapeutic
node dissection,
3) the long-term regional nodal basin failure rate following a negative sen-
tinel node biopsy, and
4) the role for systemic adjuvant therapy in patients with subclinical nodal
disease.
The Multicenter Selective Lymphadenectomy Trial sponsored by the National
Cancer Institute is an ongoing prospective randomized study with a target accrual
of 1200 patients. The treatment algorithm for patients with melanomas > 1.0 mm
thick is depicted in Figure 4.3. Eligible patients are randomized to LM/SNB versus
wide local excision (WLE) alone. Patients with positive sentinel nodes undergo
therapeutic lymphadenectomy while sentinel node negative patients are observed.
Patients who are initially managed with WLE alone will undergo a therapeutic
4 dissection if clinical regional nodal failure develops. The following questions will
be specifically addressed:
1) Is there a survival benefit for patients who undergo early lymphadenec-
tomy?,
2) What, if any, therapeutic value exists for sentinel node biopsy alone?,
and
3) What is the false negative rate as determined by failure in nodal basin
after a negative SLN biopsy?
Another multi-institutional study—the “Sunbelt Melanoma Trial”—addresses
issues related to the relative clinical significance of microscopic nodal disease as
determined by different histologic methods:
1) conventional histology,
2) step sectioning and immunohistology, and
3) PCR analysis using four molecular markers (MAGE III, MART I, GP
100 and Tyrosinase). This study will not only elucidate the natural his-
tory of these subsets of patients, but will also examine the potential ben-
efit of high dose interferon administered to patients with low nodal
burden metastatic disease (Fig. 4.4).
Fig. 4.3. Treatment algorithm for Multi-Center Selective Lymphadenectomy Trial. WLE, wide local exci-
sion; LM/SNB, lymphatic mapping/sentinel node biopsy.
Fig. 4.4. Algorithm for “Sunbelt

Melanoma Trial”. (A) Tech-
niques for histologic examina-
tion of each SLN removed. (B)
Treatment according to results
of histologic evaluation.
B
Future clinical trials will stratify patients into homogenous prognostic groups
based on sentinel node evaluation. In this way, one can better select the highest
risk patients for aggressive adjuvant therapy, determine if less toxic adjuvant regi-
mens may be effective in lower risk groups, and spare the most favorable groups
(H&E negative, step section negative, PCR negative) the morbidity and cost of
any adjuvant therapy.
CONCLUDING COMMENTS
4
Numerous studies support that the minimally invasive technique of LM/SNB
accurately stages the clinically negative regional lymph node basin in primary
melanoma patients and confirms the orderly progression of melanoma metastases
within the lymphatic compartment. Recent studies also demonstrate that the his-
tologic status of the SLN is the most powerful prognostic factor in clinical stage I
and II patients. While this approach emerged out of ELND controversy as a way to
perform therapeutic dissections early in the course of nodal disease, the establish-
ment of high dose interferon as the first effective adjuvant therapy, particularly in
node positive patients, has created another motivation for accurate nodal staging.
The components of accurate nodal staging include the following:
1) proper and reliable identification of the sentinel node, and
2) careful histologic examination of those sentinel node(s) identified. The
combined modality approach of blue dye and radiolabeled colloid in-
jections can best accomplish the first component, while a better under-
standing of the clinical relevance of various histologic techniques ap-
plied to the sentinel node will establish new standards of care for the
second component. The ultimate success of LM/SNB depends on the
integration of multiple disciplines. The long-term goals of a better un-
derstanding of the natural history of melanoma patients and the estab-
lishment of more defined guidelines for surgical and adjuvant therapy
will be accomplished by using the information obtained from LM/SNB
as stratification criteria in the design of clinical trials.
REFERENCES
1. Parker SL, Tong T, Bolden S, Wings, PA. Cancer Statistics, CA: Cancer J Clin
1997; 47:5-27.
2. Ross MI. Surgical management of stage I and II melanoma patients: Approach to
the regional lymph node basin. Seminars in Surgical Oncology 1996; 12:394-401.
3. Balch C, Soong S, Bartolucci A et al. Efficacy of an elective regional lymph node
dissection of 1 to 4 mm thick melanomas for patients 60 years of age and younger.
Ann Surg 1996; 224:255-263.
4. Balch CM. The role of elective lymph node dissection in melanoma: Rationale,
results and controversies. J Clin Oncol 1988; 6:163-172.
5. Slingluff CL Jr, Stidham KR, Ricci WM et al. Surgical management of regional
lymph nodes in patients with melanoma: Experience with 4682 patients (see
comments). Ann Surg 1994; 219:120-130.
6. Buzaid AC, Tinoco LA, Jendiroba D et al. Prognostic value of size of lymph node
metastases in patients with cutaneous melanoma. J Clin Oncol 1987; 39:139-147.
7. Slingluff CJ Jr, Vollmer R, Seigler H. Stage II malignant melanoma: Presentation
of a prognostic model and an assessment of specific active immunotherapy in
1,273 patients. J Surg Oncol 1987; 39:139-147.
8. Balch CM, Murad TM, Soong SJ et al. Tumor thickness as a guide to surgical
management of clinical stage I melanoma patients. Cancer 1979; 43:883-888.
9. Morton D, Wen D, Wong J et al. Technical details of intraoperative lymphatic
mapping for early stage melanoma. Archives Surg, 1992;127:392-399Krag D,
Meijer S, Weaver D et al. Minimal-access surgery for staging of malignant mela-
noma. Arch Surg 1995; 130:654-658.
4
10. Kirkwood J, Strawderman M, Ernstoff M, Smith T, Bordern E, Blum R. Inter-
feron-alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The
Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 14:7-17.
11. Reintgen D, Cruse C, Wells K et al. The orderly progression of melanoma nodal
metastases. Ann Surg 1994; 220:759-767.
12. Ross M, Reintgen D, Balch C. Selective lymphadenectomy: Emerging role for
lymphatic mapping and sentinel ode biopsy in the management of early stage
melanoma. Semin Surg Oncol 1993; 9:219-223.
13. Thompson J, McCarthy W, Bosch C et al. Sentinel lymph node status as an indi-
cator of the presence of metastatic melanoma in regional lymph nodes. Mela-
noma Res 1995; 5:255-260.
14. Berger DH, Feig BW, Podoloff D, Norman J, Cruse CW, Reintgen DS, Ross MI.
Lymphoscintigraphy as a predictor of lymphatic drainage from cutaneous mela-
noma. Ann Surg Oncol 1997; 4:247-251.
14a. Gershenwald JE, Tseng CH, Thompson W, Mansfield PF, Lee JE, Bouvet M, Lee J,
Ross MI. Improved sentinel lymph node localization in primary melanoma pa-
tients with the use of radiolabeled colloid. Surgery 1998, in press.
15. Norman J, Cruse CW, Espinosa C et al. Redefinition of cutaneous lymphatic
drainage with the use of lymphoscintigraphy for malignant melanoma. Am J
Surg 1991; 162:432-437.
16. Uren R, Howman-Giles R, Thompson J et al. Lymphoscintigraphy to identify
sentinel lymph nodes in patients with melanoma. Melanoma Res 1994; 4:395-399.
17. Krag D, Meijer S, Weaver D et al. Minimal-access surgery for staging of malig-
nant melanoma. Arch Surg 1995; 130:654-658.
18. Albertini J, Cruse C, Rapaport D et al. Intraoperative radiolymphoscintigraphy
improves sentinel lymph node identification for patients with melanoma. Ann
Surg 1996; 223:217-224.
19. Nathanson SD, Anaya P, Karvelis KC, Eck L, Havstad S. Sentinel lymph node
uptake of two different technetium-labeled radiocolloids. Annals of Surgical
Oncology 1997; 4(2):104-110.
20. Kapteijn B, Nieweg O, Liem I et al. Localizing the sentinel node in cutaneous
melanoma: gamma probe detection versus blue dye. Ann Surg Oncol 1997;
4(2):156-160.
21. Gershenwald J, Colome M, Lee J et al. Patterns of recurrence following a negative
sentinel lymph node biopsy in 243 patients with stage I or II melanoma. J Clin
Oncol 1998; 16:2253-2260.
22. Gershenwald J, Thompson W, Mansfield P et al. Patterns of failure in melanoma
patients after successful lymphatic mapping and negative sentinel node biopsy.
49th Annual Meeting of The Society of Surgical Oncology, Atlanta, GA; 1996.
23. Schneebaum S, Briele HA, Walker MJ et al. Cutaneous thick melanoma. Progno-
sis and treatment. Arch Surg 1987; 122:707-711.
24. Robert MR, Wen DR, Cochran AJ. Pathological evaluation of the regional lymph
nodes in malignant melanoma. Semin Diagn Pathol 1993; 10:102-115.
25. Lane N, Lattes R, Malm J. Clinicopathologic correlations in a series of 117 malig-
nant melanomas of the skin of adults. Cancer 1958; 11:1025-1043.
26. Das Gupta TK. Results of treatment of 269 patients with primary cutaneous
melanoma: a five-year prospective study. Ann Surg 1977; 186:201-209.
27. Reintgen D, Balch C, Kirkwood J, Ross M. Recent advances in the care of the
patient with malignant melanoma. Ann Surg 1997; 225:1-14.
4 28. Wang X, Heller R, VanVoorhis N et al. Detection of submicroscopic lymph node
metastases with polymerase chain reaction in patients with malignant melanoma.
Ann Surg 1994; 220(6):768-774.
Sentinel Lymph Node Biopsy for Melanoma: Surgical Technique 63
Sentinel Lymph Node Biopsy

for Melanoma: Surgical Technique
Fadi F. Haddad, Douglas Reintgen
Introduction .......................................................................................................... 63
The Role o f Nuclear Medicine—S urgical Perspective ........................................... 63
Descr iption of the Technique of Intraoperative Lymphatic Mapping ................... 65
Role o f Patholo gy .................................................................................................. 68 5
Updated Results o f Lymphatic Mapping f or Melanoma from the MCC .............. 70
Conclusion ............................................................................................................ 70
INTRODUCTION
The care of patients with melanoma has changed in the last 5 years with the
development of new lymphatic mapping techniques to reduce the cost and mor-
bidity of nodal staging, the emergence of more sensitive assays for occult mela-
noma metastases, and the identification of interferon alfa-2b as an effective adju-
vant therapy for the treatment of patients with melanoma at high risk for recur-
rence. The accurate staging of melanoma patients has become more important in
light of the recent report of a multi center, prospective, randomized trial that shows
a survival benefit for patients with T4 (tumor thickness > 4.0 mm) or Stage 3
(nodal metastases) melanoma who are treated with adjuvant interferon alfa-2b
(Intron A®, Schering Corporation, Kenilworth, New Jersey).1 The lymphatic map-
ping technology is the least morbid and costly method to obtain nodal status of
the patient with melanoma. Surgical technique is important, but it must be em-
phasized that the surgeon needs excellent nuclear medicine and pathology sup-
port to perform this technique.
THE ROLE OF NUCLEAR MEDICINE—SURGICAL PERSPECTIVE
Lymphoscintigraphy has been shown to be indispensable in predicting lym-

phatic basins at risk for the development of metastatic disease in patients with
cutaneous malignant melanoma.2 Preoperative lymphoscintigraphy serves as a
road map for the surgeon planning the surgical procedure in the following ways:

Fig. 5.1. 50 year old white male

with an “intermediate thickness”
melanoma on the right shoulder.
Preoperative lymphoscintigraphy
with Technetium Sulfur Colloid
shows bidirectional drainage to
sentinel nodes in both the left
neck (open arrow) and left axilla
(closed arrow). Both basins are at
risk for metastatic disease, while
the clinical prediction would be
for drainage to just the left neck.
Intraoperative lymphatic map-
5 ping harvested the SLN in both
locations with the neck being
negative and the axilla having
micro metastatic disease.
1. To identify all nodal basins at risk for metastatic disease (Fig. 5.1).
2. To identify any in-transit nodes that can be tattooed by the nuclear medi-
cine colleague for later harvesting. In-transit metastases occur in 5% of the
melanoma population and may, by definition, be considered the SLN (Fig. 5.2).
3. To identify the location of the SLN in relation to the rest of the nodes in the
basin.3 The location of the SLN may be variable in a basin and ideally the surgeon
needs a mark of the position of the SLN in reference to other nodes in the basin, in
order to perform the harvest under local anesthesia with a minimal incision. Pre-
operative lymphoscintigrams can do this quite well. Twenty-nine patients with
clinically negative nodes and melanomas thicker than 0.76 mm had preoperative
lymphoscintigrams in two planes to mark the location of the SLN prior to opera-
tion. Thirty-three percent of the time the clinician could not predict the approxi-
mate location of the SLN within 5 cm, but the lymphoscintigraphy was accurate
in the identification of the location of the SLN within 1.0 cm 100% of the time.3
The technique was most accurate in the groin and the head and neck where the
lymph nodes are more superficial. The axilla is the most difficult area to map and
the best the preoperative lymphoscintigram can do in this basin is to tell the surgeon
whether the node is located anterior, posterior, superior or inferior in the basin.
4. To estimate the number of SLN in the regional basin that will need to be
harvested.
Preoperative lymphoscintigraphy is a simple, accurate test that is the most ef-
ficient, cost-effective method of identifying the lymph nodes at highest risk for
metastatic disease.
Fig. 5.2. 35 year old white male with a 2.0 mm malignant melanoma excised from the left upper abdomen.
Clinically the predicted cutaneous lymphatic flow would be to the left axilla. With preoperative
lymphoscintigraphy, the left axilla is found not to be at risk for metastatic disease, but lymphatic flow was
noted to an in-transit node in the left upper abdomen within 5 cm of the primary site (closed arrow) and
to a lower internal mammary node (open arrow). Both nodes were mapped intraoperatively and biopsied.
Pathology was negative.
DESCRIPTION OF THE TECHNIQUE OF INTRAOPERATIVE

LYMPHATIC MAPPING
There is great variation from center to center on the technique of intraopera-

tive mapping. This section will discuss the nuances of our technique, detailing the
steps we believe important for successful mapping. However, similar success rates
are possible with slight variations in technique, as shown by multiple centers.
Patients are scheduled in the nuclear medicine suite early the day of the sur-
gery and undergo preoperative lymphoscintigraphy with the injection of
technetium around the primary site (see chapter 8 for details). We do not obtain
lymphoscintigraphy before the day of surgery; this is redundant and costly. Dy-
namic scans are performed 5-10 minutes after the injection of the radiocolloid
and the location of the SLN is marked in the basin with an intradermal tattoo.
Delayed images (2 hours) for trunk and head and neck melanoma may be neces-
sary to ensure there is not delayed drainage to more than one regional basin.
The patient is then taken to the OR 2-6 hours after radiocolloid injection and
1 cc of 1% lymphazurin (isosulfan blue) is injected intradermally (per number of
directions of drainage, i.e., 1 cc of dye per lymphatic basin identified by scan)

around the primary site. After prepping and draping the primary site and regional
basin and allowing 10 minutes for the vital blue dye to travel to the SLN, attention
is directed initially to the regional basin. With a hand-held gamma probe (e.g.,
Navigator™, United States Surgical Corporation, Norwalk, CT), the “hot” spot in
the regional basin is identified and the hot spot/background ratio is noted. If “shine
through” from the primary site is a problem, the wide local excision (WLE) of the
primary may be performed first. An incision is made over the hot spot of the
sentinel node and small flaps are created in all directions to allow identification of
the blue stained afferent lymphatics. Surgical dissection is aided by both visual-
ization of the stained afferent lymphatic down to the blue-stained node (Fig. 5.3)
5 and by the use of the hand held gamma probe. At times, the surgeon can be con-
fused as to what is proximal or distal on the afferent lymphatic and the probe can
be used to identify the direction of the dissection. The SLN is identified and re-
moved with sharp or electrocautery dissection. After removal of the entire SLN,
afferent and efferent lymphatics from the SLN are controlled with hemoclips, since
the electrocautery does not seal lymphatics. This technique decreases the chance
of postoperative wound seroma.
Lymph nodes are confirmed in vivo and ex vivo to be SLN with gamma probe,
by computing localization ratios. In vivo radioactivity of the suspected SLN is
compared to non-SLN tissue. The ratio between these two numbers should be at
least 3x. After removal of a SLN, the ex vivo measured cps of the SLN compared to
excised non-SLN tissue should be at least 10x. For blue-stained lymph nodes, lo-
calization ratios confirm them as SLN by each mapping technique. Non-blue
stained lymph nodes are identified as SLN by their localization ratios.
Studies4,5 have shown that the localization ratios double if the harvest occurs
2-6 hours after the injection of the radiocolloid compared to performing the map-
ping immediately after the injection of the radiocolloid. This data supports the
timing intervals we use, to maximize localization ratios and hence facilitate SLN
identification. After removal of the SLN(s), measured radioactivity in the lym-
phatic basin should decrease, without persistent focal “hot spots.” Dissection and
identification of more SLN continues until this global decrease is documented.
The use of the radiocolloid for intraoperative mapping allows for excision of
the SLN in unusual locations, such as the in-transit nodes. Figure 5.2 illustrates
the preoperative lymphoscintigraphy of a patient with an intermediate thickness
melanoma of the left abdominal wall. Experience would predict drainage to the
left axilla, but preoperative lymphoscintigraphy shows drainage to an in-transit
node located within 5 cm of the primary site and further drainage noted to a
lower left internal mammary lymph node.
The radiocolloid and vital blue dye mapping techniques are complimentary
and we believe they should be used simultaneously to increase the success rate of
localization of the SLN. The different mapping techniques are important depend-
ing on the location of the primary in relation to the regional basin. If the primary
site is close, overlying or in a direct line to the basin so that a hand held gamma
probe will detect “shine through” from the residual radioactivity at the primary
Fig. 5.3. Colored radiocolloid injection into the dermis of the skin in 4 quadrants around a nodular mela-
noma. Within minutes of the injection, tracer is seen being taken up by the cutaneous lymphatics. It is
rare to have the primary lesion intact at the time of the mapping, and most of the data generated for
success rates for melanoma mapping is after an excisional biopsy.
Fig. 5.4. A blue-stained afferent lymphatic is shown entering a blue-stained node. This node will be hot
with the gamma probe, will be the SLN and will be the first site of metastatic disease.
site, use of the vital blue dye may be the only technique that allows for successful
mapping. Even performing the WLE first, enough radioactivity may still be present
at the primary site. In contrast, in patients with a fatty axilla or in head and neck
mapping, it may be impossible to follow a wisp of blue-stained afferent lymphatic
to the SLN. Particularly in the head and neck area, because of the presence of
surrounding vital structures, large flaps are to be avoided and the ability of the
gamma probe to locate the “hot” spot through the skin is a tremendous advan-
tage, since it directs both the location of the (smaller) incision and also the dissec-
tion vector, once the incision is made.
SLNs are defined as those nodes either in an in-transit location or in the major
regional basin that initially and/or primarily receive lymphatic flow from the pri-
5 mary melanoma site. They can be identified by following a blue-stained afferent
lymphatic to a blue stained node (blue node), as originally described by Morton6
or with a gamma probe determined localization ratio. Using these techniques and
definitions, the success rate of SLN mapping and biopsy should approach 100%
for melanoma patients.4
ROLE OF PATHOLOGY
After harvesting, the SLN is then submitted for a detailed histologic examina-
tion that may include serial sectioning, immunohistochemical staining with S-100
and/or HMB-45 monoclonal antibodies, and perhaps newer assays using molecu-
lar biology techniques for occult metastases.7 See the chapter 10 for further details.
INCLUSION AND EXCLUSION CRITERIA

It is evident that lymphatic mapping technology has changed the standard of
surgical care for melanoma. A crucial issue is which patients should or should not
be offered the procedure. We use the following indication for SLN mapping:
1. Cutaneous malignant melanoma with thickness 0.76 mm or greater.
(Stage Ib or above)
2. Ulcerated melanomas
3. Thinner lesions may be considered if the melanoma had three or more
of the following characteristics: a) located on the trunk, head or neck,
b) male patients, c) the lesion is Clark level III or greater, d) regression
on histopathologic examination. These variables may indicate that the
melanoma is a more aggressive lesion or at one time was a thicker lesion.
Our exclusion criteria are:
1. Clinically positive regional nodal basins or evidence of in-transit, dis-
tant skin or systemic metastases.
2. Prior wide local excision (relative), particularly on the head and neck
and trunk areas. The data on this are unclear.
3. Previous Z-plasty or rotational flap. This primary site reconstruction,
particularly in the head and neck, is too likely to disrupt the cutaneous
lymphatics, and is not reflective of cutaneous lymphatic flow when the

primary was intact.
4. Preoperative lymphoscintigraphy showing drainage to more than three
basins (relative). The question would arise whether this is too extensive
a staging surgical procedure to perform on an elective basis.
5. Other medical problems (relative). However, it is unlikely that a patient’s
other medical problems would be severe enough to prohibit the perfor-
mance of a low morbidity procedure such as SLN biopsy.
6. Patients with an allergy to isosulfan blue or sulfur colloid (extremely
rare).
Patients with melanoma of tumor thickness between 0.76-1.0 mm (5.3% chance
of harboring metastatic disease) are given the choice of having the SLN harvested 5
vs observation. Invariably, patients elect to have the SLN biopsy because the mor-
bidity of the procedure is low and the treatment decisions if one has metastatic
disease in the regional basin is radically different than if the SLN is negative. Like-
wise, patients with thick melanomas (greater than 4.0 mm in tumor thickness)
are recommended to undergo lymphatic mapping for staging since T4N0 patients
have a better survival than T4N1 patients. In addition, even though interferon alfa
(Intron A®) is approved for patients with T4N0 disease, that subgroup was small
and showed no benefit in the randomized trial. Many medical oncologists are
hesitant to treat this subgroup with the toxic Intron A®. Ongoing national trials
are addressing this question. In the meantime, it is reasonable to perform lym-
phatic mapping and SLN harvest on patients with melanomas greater than 4.0 mm
and if nodal disease is found, consider them for adjuvant therapy. Overall, when
given a choice, almost all patients will choose to “know” rather than “watch and
wait,” especially given the option of the low morbidity SLN procedure.
Patients with a positive SLN should undergo a completion lymph node dissec-
tion (CLND). This is considered the standard of care for all patients with meta-
static disease in any SLN. However, a high percentage of patients will have no
further disease within the nodal basin (i.e., in the non-SLN). This phenomenon,
termed “exclusivity” elsewhere in this handbook, may occur in more than 90% of
patients, with at least one positive SLN. This finding would seem to validate the
principle that the SLN is the primary draining lymph node of the tumor site, and
that tumor cells (and mapping agents) may be “trapped” within that SLN for a
period of time before migrating to other lymph nodes.
The important question of whether one can do lymphatic mapping after a
wide local excision was examined retrospectively at our institution. One hundred
and seventy patients underwent lymphatic mapping and SLN biopsy after the WLE
of the primary site, and 399 patients with melanoma underwent mapping prior to
the WLE. The patients that had the procedure after the WLE had an increased
incidence of “skip” metastases, had an increased number of SLN removed/basin
(1.7 vs 1.9 SLN/basin) and had a significantly increased number of multiple (26%
vs 15%, p < 0.05) basins. These data suggest that WLE may disrupt and/or alter
cutaneous lymphatic flow, rendering subsequent SLN mapping less effective. This
is a controversial area; pathologically positive SLN can be located after WLE, and
as SLN technology is adopted around the country, many patients will be referred
to SLN-experienced surgeons after WLE.
Radioguided SLN mapping minimizes surgical dissection by directing the op-
erator to the site of the lymph node in vivo. This permits the use of smaller
incisions and facilitates biopsy of deep nodes, such as in the axilla. Also, SLN in
previously poorly accessible areas can be identified and biopsied, including scapular,
internal mammary, and intra-parotid nodes.
Complications of the procedure are few. Drains are not placed, and in follow-
up there is a 10% incidence of seroma formation that can be handled easily with
aspiration. The side effects of the vital blue dye is staining of the tissue (up to a
period of 12 months) if all the blue dye is not removed with the WLE. This does
5 not seem to be a problem in the head and neck area probably due to the rich
lymphatic and vascular supply in this area. The urine turns blue/green for 24-48
hours and patients should be informed of this likelihood. Clinicians should be
aware that the oxygen saturation reading on the pulse oximeter may decrease sec-
ondary to the blue dye. Rarely, an allergy to the blue dye (with hives and a blue
discoloration of the skin) may be encountered.
UPDATED RESULTS OF LYMPHATIC MAPPING FOR MELANOMA

FROM OUR INSTITUTION
A total of 693 patients with melanoma have undergone lymphatic mapping

and SLN biopsy at our institution, Moffitt Cancer Center. The SLN was success-
fully identified and harvested in 688 patients, for a 99% success rate. One hundred
patients (14.5%) had evidence of nodal metastases. The rates of SLN involvement
for primary tumors < 0.76 mm, 0.76 mm-1.0 mm, 1.0-1.5 mm, 1.5-4.0 mm and
greater than 4.0 mm in thickness was 0%, 5.3%, 8%, 19% and 29%, respectively.
Eighty-one patients underwent a completion lymph node dissection (CLND) af-
ter positive SLN biopsy and only six patients (7.4%) with positive SLN demon-
strated metastatic disease beyond the SLN. The tumor thickness of these 6 pa-
tients ranged from 2.8-6.0 mm. No patients with a tumor thickness less than
2.8 mm was found to have evidence of metastatic disease beyond the SLN upon
CLND. It is hypothesized that a melanoma has to reach a certain thickness before
it sheds off enough cells to involve more than the SLN with disease. This hypoth-
esis was recently supported by data from MD Anderson that showed no non-SLN
involved with metastatic disease after a positive SLN biopsy for melanomas less
than 2.5 mm in thickness.11
CONCLUSION
Radioguided surgery may be the next revolution in general surgery, following

in the footsteps of the laparoscopic procedures a decade ago. With potential appli-
cations to colon and breast cancers, other skin tumors like the Merkel cell carcinoma
or poorly differentiated squamous cell carcinomas, parathyroid localizations, bone

tumor localizations and vulvar or vaginal lesions, there are 350,000 new cases of
cancer diagnosed each year in the United States to which one can apply the tech-
niques. For melanoma, extensive data from multiple institutions confirm that the
SLN are the lymph nodes at highest risk for metastatic disease; that these nodes
may successfully be mapped preoperatively with lymphoscintigraphy; and, utiliz-
ing state-of-the-art technology, the SLN may be located intraoperatively.
REFERENCES
1. Kirkwood JM, Strawderman MH, Ernstoff MS, Smith TJ, Borden EC, Blum R.
Interferon Alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma:
The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 1996; 5
14:7-17.
2. Norman J, Cruse CW, Espinosa C et al. Redefinition of cutaneous lymphatic
drainage with the use of lymphoscintigraphy for malignant melanoma. Am J
Surg 1991; 162:432-437.
3. Godellas CV, Berman C, Lyman G et al. The identification and mapping of mela-
noma regional nodal metastases: minimally invasive surgery for the diagnosis of
nodal metastases. Am Surg 1995; 61:97-101.
4. Albertini JJ, Cruse CW, Rapaport D et al. Intraoperative radiolymphoscintigraphy
improves sentinel lymph node identification for patients with melanoma. Ann
Surg 1996; 223:217-224.
5. Nathanson SD, Anaya P, Eck L. Sentinel lymph node uptake of two different ra-
dionuclides. The Society of Surgical Oncology, 49th Cancer Symposium, Atlanta,
March, 1996 (Abstract).
6. Morton DL, Wen DR, Cochran AJ. Management of early-stage melanoma by
intraoperative lymphatic mapping and selective lymphadenectomy or “watch
and wait.” Surgical Oncology Clinics of North America 1992; 1:247-259.
7. Wang X, Heller R, VanVoorhis N et al. Detection of submicroscopic metastases
with polymerase chain reaction in patients with malignant melanoma. Ann Surg
1994; 220:768-774.
8. Reintgen DS, Cruse CW, Berman C, Ross M, Rapaport D, Glass F, Fenske N,
Messina J. An orderly progression of melanoma nodal metastases. Ann Surg 1994;
220:759-767.
9. Ross M, Reintgen DS, Balch C. Selective lymphadenectomy: Emerging role of
lymphatic mapping and sentinel node biopsy in the management of early stage
melanoma. Seminars in Surgical Oncology 1993; 9:219-223.
10. Gershenwald J, Thompson W, Mansfield P. Lee J, Colome M, Balch C, Reintgen
D, Ross M. Patterns of failure in melanoma patients after successful lymphatic
mapping and negative sentinel node biopsy. Society of Surgical Oncology, 1996
(Abstract).
11. Haddad FF, Stall A, Messina J, Brobeil A, Ramnath E, Glass F, Cruse CW, Berman
C, Reintgen DS. The orderly progression of melanoma nodal metastases is de-
pendent on tumor thickness of the primary lesion. Ann Surg Oncol, Submitted.
Technique for Lymphatic Mapping

in Breast Carcinoma
Charles E. Cox
Introduction: Economic Imperatives .................................................................... 72

Status o f Axillary Node Dissection: Historical Overview ..................................... 73
Historical Perspective of the Development of Lymphatic Mapping
for Breast Cancer .......................................................................................... 73
Technical Aspects of Lymphatic Mapping f or Breast Carcino ma ......................... 74
Operative Technique for Lymphatic Mapping o f Breast Carcino ma .................... 76
6 Gamma Detection ................................................................................................ 78
Summary .............................................................................................................. 80
INTRODUCTION: ECONOMIC IMPERATIVES
The surgical treatment of breast cancer is rapidly evolving. Some factors influ-
encing these changes include altered disease demographics, advances in technol-
ogy, governmental and reimbursement controls, and increasing public expecta-
tions. The epidemiology of breast cancer as “baby boomers” turn 50 years of age
marks a dramatic rise in the prevalence of breast cancer. Although the relative risk
of developing breast cancer has remained constant, the incidence of breast cancer
is projected to increase from 185,000 new cases annually to 420,000 new cases
annually over the next 20 years. Advances in surgical technology and the develop-
ment of minimally invasive surgical techniques have heralded a new era in sur-
gery.3 The new technology of sentinel lymph node (SLN) mapping for breast car-
cinoma, as with any other new surgical technique in this era, must meet the bur-
den of not only improved efficiency and reduced risk but also diminished cost
and resource utilization.4 The added burden of economic (reimbursement) pres-
sure and increased public expectations for better cancer therapies place additional
scrutiny on surgeons utilizing these new techniques.
I believe that safe implementation of SLN biopsy for breast cancer requires
two things: adequate training programs and initial performance only under an
approved IRB protocol. A recent publication introduced several guidelines regard-
ing these issues.5 Training, credentialing, IRB protocols and outcome measures
are discussed in other chapters of this book. Our program has developed our own
flow-tech, coded and secure forms for recording outcomes. These forms can be
faxed to a secure internet repository <http://mapping.rad.usf.edu> allowing sur-
geons to enter, review and monitor their own outcomes.5 We are excited about
Technique for Lymphatic Mapping in Breast Carcinoma 73
this new program, because it will provide confidential access to multi-institutional

clinical research data to users around the world.
STATUS OF AXILLARY NODE DISSECTION:

HISTORICAL OVERVIEW
The current standard of care for the management of invasive breast cancer is
the complete removal of the tumor and documentation of negative margins by
either mastectomy or lumpectomy followed by complete axillary lymph node dis-
section.6-8 Lymphatic mapping of the breast is clearly changing this long held
paradigm.
The role of axillary dissection may be currently the most controversial topic in
the treatment of breast cancer. Nearly 100 years ago, Halsted demonstrated the
curative potential of radical mastectomy. Fifty years later, Patey proved that modi-
6
fied radical mastectomy could yield similar survival with limited morbidity. The
controversy now rages over the current role of axillary dissection in the manage-
ment of operable breast cancer. Since the time of Halsted, the status of the re-
gional nodal basin remains the single most important independent variable of pre-
dicting prognosis.9 Advocates of axillary dissection contend that there is a benefit
for breast cancer patients since axillary dissection renders regional control of ax-
illary disease. They propose that surgical removal of microscopic nodal metastases
is curative without adjuvant chemotherapy in certain patient populations. Never-
theless, recent critics of axillary dissection maintain that overall survival depends
on the development of distant metastases and is not influenced by axillary dissec-
tion in most patients.10-12 They contend that patients with microscopic axillary
metastases might be cured with adjuvant chemotherapy with or without nodal
irradiation without the need for axillary dissection. Many have even advocated
the abandonment of axillary dissection in early breast cancer. Sentinel node bi-
opsy for breast cancer may eliminate this controversy. This low morbidity proce-
dure appears to obviate the need for routine axillary node dissection because it
correctly identifies those patients with nodal metastases who require completion
node dissection while also defining that subset of patients without any nodal dis-
ease who are unlikely to benefit from further node removal.
HISTORICAL PERSPECTIVE OF THE DEVELOPMENT

OF LYMPHATIC MAPPING FOR BREAST CANCER
Lymphatic mapping was initially developed for the treatment of malignant

melanoma and has subsequently been adopted for breast cancer lymph node evalu-
ation.14-18 There are obvious differences between melanoma and breast cancer.
Melanoma generally is faster growing and declares itself quickly. Therefore, the
magnitude of the delay for clinical detection of recurrent nodal metastases fol-
lowing surgical and histologic examination of sentinel nodes for metastatic breast
cancer could take between 3-5 years versus the 8-18 months required for the same
clinical detection of melanoma metastases. Furthermore, these delays could be
further prolonged by the broad application of adjuvant therapies for breast can-
cer in node negative patients, especially if detection of lymph node metastases is
minimized during the initial surgical evaluation. The possibility also exists, as with
melanoma, that some patients may be spared the use of adjuvant chemotherapy if
no nodal metastases can be found with these highly sensitive techniques. (See
elsewhere in this handbook for details on pathologic examination.)
The purpose of this chapter is to describe the current technique for lymphatic
mapping of breast carcinoma. Covered in this section will be a current descrip-
tion of the methods to perform lymphatic mapping using both blue dye and ra-
diocolloid. Included will be sections on: the characteristics of both blue dye and
radiocolloid and injection techniques for both, operative techniques, gamma map-
ping techniques and finally a section on intraoperative measurement recording
6 and documentation of results.
TECHNICAL ASPECTS OF LYMPHATIC MAPPING

FOR BREAST CARCINOMA
RADIOCOLLOID INJECTION
In the United States, the radiocolloid employed for SLN mapping is techne-
tium sulfur colloid, although other compounds are used elsewhere in the world.19
For example, antimony sulfide, no longer available in the U.S., has successfully
been used for SLN mapping in Australia and other countries.
In some of the initial pilot studies of mapping at our institution, subdermal
injections were used due to the richness of the subdermal plexus of lymphatics.
However, though mapping was excellent it was unclear if this is a true representa-
tion of the lymphatic flow of the tumor bed. We abandoned this method due to
this uncertainty. Guiliano’s work and our current series of 700 mapping cases
have less than a 1% skip rate which would argue strongly that intraparenchymal
injection of the mapping agents is the optimal route.20 This low skip rate is consis-
tent with the clinical imperative of SLN mapping for breast cancer: that the SLN
be pathologically representative of the nodal metastatic status of the patient. It
has been our observation that deep parenchymal injections of blue dye have ap-
peared in the skin with time. Does the lymph flow to the skin then to the nodes in
some cases? Are deep lymphatics draining to different nodes than skin lymphat-
ics? Further studies perhaps using PET scanning may more carefully elucidate the
actual lymphatic flow of the breast and definitively resolve the question of where
mapping agents should be injected.
All studies performed within the U.S. employ Tc99m labeled sulfur colloid ei-
ther in a filtered or unfiltered state injected in the amount of 0.450 mCi-1.0 mCi
of specific activity bound to the sulfur colloid respectively. Krag and his group
have routinely used unfiltered Tc99m labeled sulfur colloid and injected 1.0 milli-
curies of radioactivity. We utilize 450 microcuries of 0.22 micron filtered sulfur

colloid, (in combination with intraoperative Lymphazurin Blue dye). The radio-
nuclide is injected in a 6 cc volume at six different sites intraparenchymally sur-
rounding the tumor or the tumor bed following excisional biopsy.
BLUE DYE TECHNIQUE

There are two agents referred to in the mapping literature trademarked as
Lymphazurin and Patent blue dye. Essentially, these are biochemically the same
agents. Lymphazurin 1% (isosulfan blue) is a sterile aqueous solution for subcu-
taneous administration. Phosphate buffer and sterile, pyrogen-free water is added
in sufficient quantity to yield a final pH of 6.8 -7.4. Each cc of solution contains
10 mg isosulfan blue, 6.6 mg sodium monohydrogen phosphate and 2.7 mg po-
tassium bihydrogen phosphate. The solution contains no preservatives. It is a con-
trast agent for the identification of lymphatic vessels without known pharmaco-
logic action. Following subcutaneous administration isosulfan blue is selectively
6
picked up by the lymphatic vessels, giving them a bright blue-color discernible
from surrounding tissue. There is some evidence that 50% of the isosulfan blue,
from aqueous solution, is weakly bound to serum protein (albumin). Since inter-
stitial protein is presumed to be carried almost exclusively by lymphatics and in
view of evidence of binding of dye to proteins, visualization may be due to protein
binding phenomenon. Up to 10% of the subcutaneously administered dose of
Lymphazurin 1% is excreted unchanged in the urine in 24 hours in man. Presum-
ably, 90% is excreted through the biliary route.
Lymphazurin 1% (isosulfan blue) has demonstrated a 1.5% incidence of ad-
verse reactions. All the reactions were of an allergic type. Localized swelling at the
site of administration and mild hives of hands, abdomen and neck have been
reported within several minutes following administration of the drug. Reports of
mild to severe reactions have appeared in the literature for compounds similar to
isosulfan blue. There have been no reported deaths from the administration of
isosulfan blue dye. However, a death has been reported following intravenous ad-
ministration of a similar compound employed to estimate the depth of a severe
burn. Severe reactions may be manifested by edema of the face and glottis, respi-
ratory distress or shock. Such reactions may prove fatal unless promptly controlled
by such emergency measures as maintenance of a clear airway and immediate use
of oxygen and resuscitative drugs.
To date in our current series of over 700 patients, approximately 1% have dem-
onstrated allergic reactions to isosulfan blue. These have been manifested by an
initial wheal reaction at the injection site followed by the development of blue
hives scattered about the neck, groin, and other intertriginous areas. These have
generally responded to IV Benadryl, and have disappeared rapidly. It is advisable
to observe the patient at least 30-60 minutes following the administration of
isosulfan blue since severe or delayed reactions may occur.
The admixture of Lymphazurin with local anesthetics, i.e., lidocaine, in the
same syringe prior to administration results in an immediate precipitation of 4-9%
of the drug complex. Similar precipitation of sulfur colloid occurs with the mixture
of these two compounds. This technique is not recommended. Is in the best inter-
est of the patient to give other medications such as local anesthetics or sulfur
colloid via separate syringe and at separate time intervals.
Lymphazurin has no known carcinogenic, mutagenic, or teratogenic effects
but should be given to pregnant women only if clearly needed. Safety and effec-
tiveness of Lymphazurin for children has not been established. It is unknown
whether the drug is excreted in human milk and should be administered with
caution to nursing mothers.
INJECTION TECHNIQUE FOR LYMPHATIC MAPPING WITH BLUE DYE

There are occasional reports of lymphatic mapping for breast carcinoma with
intradermal injection of blue dye. The same high percentage uptake in lymph
nodes occurs due to the rich subdermal lymphatic channels; however, the same
arguments apply to this technique as are described above with the intradermal
6 injection of sulfur colloid. The authors contributing to this handbook recom-
mend intraparenchymal, not the intradermal, injection of Lymphazurin for the
lymphatic mapping of breast carcinoma.
In our preferred method, an intraparenchymal injection of 5 cc of blue dye is
performed in multiple sites around the tumor or excisional biopsy site. For le-
sions in the upper outer quadrant of the breast which assuredly drain to the axilla,
the dye may be injected through a 27 gauge needle and fanned through a single
injection site into the upper outer aspect of the breast towards the axilla. Follow-
ing this injection, manual compression of the breast and gentle massage for a
sustained time of five minutes by the clock is performed to insure proper migra-
tion of the blue dye into the lymphatic channels. This is performed just prior to
skin preparation and draping of the patient for operative intervention. Immedi-
ately following this preparation of the patient the sentinel lymph node can be
found by making an accurate incision approximately 1 cm inferior to the hair line
of the axilla. Dissection may proceed quickly to the clavi-pectoral fascia after which
care must be taken to avoid damage to any lymphatic channels seen beyond that
point. Disruption of the lymphatic channels at this level will preclude the ability
to find a SLN. Channels may appear superficial to this area and may lead to the
SLN; however, other channels at a deeper level are more likely to carry the major-
ity of the blue dye to the SLN.
OPERATIVE TECHNIQUE FOR LYMPHATIC MAPPING

OF BREAST CARCINOMA
In the ideal circumstance the patient would have been injected at least 2 hours
previously with 450 microcuries of filtered Tc 99m labeled sulfur colloid.
Lymphazurin 1% (isosulfan) blue dye, 5 cc would have been injected, compressed
for five minutes and the skin of the breast, chest wall and axilla would have been
prepared for operation with appropriate scrubs and disinfectant solutions
(Hibiclens and Hibistat respectively preferred by the author). The patient is ap-
propriately draped to expose the breast, chest wall, and axilla. The gamma probe
to be utilized is sterilely sheathed and available for axillary node mapping.
Before proceeding to the actual mapping, it is important to mention several
general considerations. Since most surgeons are just learning the new technique,
they will experience an initial learning curve. Since new operative techniques may
require additional time, it is important that the patients be scheduled to allow
enough time for the operation to proceed unencumbered by time constraints. For
example, most surgeons can easily perform axillary dissection quicker than they
will be able to do their initial sentinel lymph node biopsies. The procedure should
be done in a calm environment and it is encouraged as one begins the mapping
and selective lymphadenectomy that the operator and assistant sit down to operate.
Visualization of small lymphatic channels requires good exposure and retrac-
tion in a bloodless field. Therefore, is encouraged that one uses the electrocautery
for dissection with a self-retaining retractor in the incision and if using local anes-
thetic, epinephrine should be included. Careful dissection is mandatory and the
6
author has found the use of small fine tipped clamps (mosquito, classic delicate,
6", Codman 30-4470) for dissection to be extremely useful. One should proceed
with caution. Lymphatic channels should be clipped. However, it is imperative
that one not cut nor clip the blue channel containing the dye until a node or
nodes have been isolated. Do not cut the blue channel, as this will result in loss of
drainage of the blue dye into the SLN, making it very difficult to localize without
the gamma probe. Furthermore, the blue dye will stain the surrounding tissues
further complicating the ability to detect the blue lymph node. If inadvertently
the blue dye channel is cut then the use of the gamma detection probe is critical in
location of the SLN.
Mapping the axillary SLN requires a properly calibrated machine in which the
sensitivity is properly set to allow detection of the gamma counts. Certain anatomic
considerations will be helpful in SLN localization. If one were to draw a line along
the lateral border of the pectoralis major muscle and lateral border of the latissi-
mus dorsi muscle in the axilla, these would mark the outer borders of the axillary
limits for the dissection. Careful marking of the axillary hairline is useful. One
should place a tangential line at the axillary hairline in a perpendicular fashion
anterior to posterior. A line is then drawn through the axis of the axilla, through
the center point of the hairline. Those intersecting lines mark the center of a 5 cm
circle, which can be drawn on the axilla. Within this 5 cm circle are 94% of the
SLN (see Fig. 6.1). The remaining 6% will be found in the level II location. Using
the gamma detection probe, this starting point of the 5 cm circle may be useful for
identifying the actual location of the sentinel lymph node. Once the sentinel lymph
node is found, an accurate incision may be made overlying the area of highest
activity as determined by the gamma probe. The incision falls generally at or be-
low the hairline as indicated above. Care should be taken to extend the dissection
towards the chest perpendicular to the chest wall. The tendency of most novice
surgeons is to make the incision and dissect in a cephalad direction. This should
be avoided. Internally, the landmarks which localize the SLN are the central axillary
vein and the third branch of the intercostal nerve. These anatomic structures are
External Anatomic
Localization of SLN
jor ine
Ma yL
lar
alis xil
tor al A
Pec ntr
Tangent Ce
to hairline
5 cm circle in which 94% of the

Hairline
sentinel nodes will be found.
Remaining 6% of Sentinel nodes found
in Level II and very rarely in Level III.
Dorsi
Latissimus
Fig. 6.1 Mapping the axillary SLN. If one were to draw a line along the lateral border of the pectoralis
major muscle and lateral border of the latissimus dorsi muscle in the axilla, these would mark the outer
borders of the axillary limits for the dissection. One should place a tangential line at the axillary hairline
in a perpendicular fashion anterior to posterior. A line is then drawn through the axis of the axilla, through
the center point of the hairline. Those intersecting lines mark the center of a 5 cm circle, which can be
drawn on the axilla. Within this 5 cm circle are 94% of the SLN.
identifiable beneath the clavi-pectoral fascia. The central axillary vein can be found
easily with careful dissection. Where the nerve crosses over the vein four quad-
rants are defined, as were seen on the external anatomy for the 5 cm circle. This
internal anatomical point of intersection will define one of four quadrants, which
collectively will contain 94% of the SLN, as noted above, found in breast lym-
phatic mapping.
GAMMA DETECTION
Intraoperative use of a gamma radiation detector can be complicated by sev-

eral technical issues, including “shine through”, and others, which are further dis-
cussed in another chapter.
“SHINE THROUGH” EFFECT

“Shine through” occurs when the gamma probe detects radiation energy from
the injection site while looking for a SLN. It is generally due to a large deposit of
radioactivity in a close proximity to the dissection area. It is critical to keep in
mind exactly where the injection site is located in relation to where one is point-
ing the probe. An example of this is in a melanoma of the mid-back with an axil-
lary SLN. Pointing the probe towards the axilla also aims it at the injection site in
the back. Thus, one has to direct the aim of the probe towards the head or foot
while searching the axilla. The same possibility exists with breast cancer in that
the patient may have an ipsilateral lesion in the upper inner quadrant of the breast
or bilateral breast cancers and have bilateral mapping performed with a contralat-
eral lesion in the upper half of the breast. Each of these conditions may result in
difficulty avoiding the “shine through” effect.
Another classic “shine through” problem is finding a hot node only to remove
it and find that there is actually only minimal radioactivity in the removed node.
This is usually due to another node containing the radioactivity directly behind
the removed lymph node. Blue channels may surround the removed node, as well,
but may actually be traveling directly to another node deep to the one removed.
This reinforces the necessity of checking background counts after removal of the
node to be sure all hot nodes were removed, and measuring the ex vivo counts of
SLN removed to verify that indeed the node removed is hot. The hot nodes can
sometimes be stacked so that even by removing the hot node another hot node
6
lies directly behind. Again, this emphasizes the need to check the background as
well as the node removed. Finally, another situation can make node detection with
the probe difficult. Counts are drained through the blood vessels and occasionally
the central axillary vein will be hot. This is verified by noting the course of the vein
and measuring counts along the vein.
INTRAOPERATIVE MEASUREMENTS
There are several specific measurements that are important to validate the lym-
phatic mapping procedure. Included here will be information that is obtain-
able through the National Breast Lymphatic Mapping Database <http://
surgonc.rad.usf.edu>. It is important to record the type of gamma probe used,
the probe size, the time of injection and the dose of injection material. It is also
important to measure the number of counts obtained at the incision site pre- and
postexcision and the central axillary bed count.
The following data should be entered for each SLN harvested. By definition,
SLN must be “hot “, “blue” or “hot and blue”. Hot nodes are defined as those with
in vivo counts which are 2-3x those of the surrounding basin, or those with ex vivo
counts which are 10x those of an excised nonsentinel node. Blue nodes are de-
fined as those that are at least faintly stained by the isosulfan blue dye. A sentinel
node in situ count should be obtained. The SLN should be exposed and the gamma
probe should be placed directly on the exposed node before harvesting. The time
of harvest should be entered as well. SLN ex vivo counts should also be recorded.
For a measurement the node should be placed on a plain sterile towel located well
away from the lymphatic basin and primary injection site. A postexcisional radio-
activity count in the basin where the lymph node was removed should likewise be
obtained. This measurement should be taken just after removal of the sentinel
node at the site where the node once was. This count determines if the SLN has
indeed been removed and detects the presence of additional SLN in the basin.
Avoid directing the gamma probe toward the primary tumor site, which will result
in “shine through” extraneous counts. Any non-SLN removed should have an
ex vivo count recorded, in cps. Furthermore, a completion node dissection, if done,

should be recorded and the reason for this procedure indicated. Reasons for com-
plete lymph node dissection include: 1) node grossly positive (i.e., a node was
grossly positive for metastatic breast cancer), 2) lymphatic mapping failed (i.e.,
lymphatic mapping failed to locate any hot or blue sentinel nodes), or 3) phase
one patient (i.e., you are using a “phase one” protocol in which complete lymph
node dissection is performed on all patients after SLN biopsy). Finally, the probe
should be utilized to localize any counts extraneous to the injection site or axillary
lymph node basin such as internal mammary nodes.
SPECIAL CASE: INTERNAL MAMMARY SLN

To date the incidence of internal mammary node involvement based on lym-
phatic mapping of the breast has been reported as 1-6% based on detected radio-
active counts at the internal mammary location. Any central breast lesion or inner
6 quadrant lesions of the breast should be evaluated with preoperative
lymphoscintigraphy. Furthermore, when mapping any lesion of the breast it is
critical to evaluate the internal mammary locations, specifically the second and
third parasternal interspaces. Should “hot spots” be identified in these locations, it
may be feasible to excise these internal mammary nodes. This has been accom-
plished by division of the intercostal muscles at the parasternal location thereby
exposing the internal mammary nodes. These can be removed with careful dissec-
tion and application of the gamma probe to localize the node. This is one draw-
back to the blue dye method since it is nearly impossible to localize a node in the
internal mammary location with the blue dye. To date, if a sentinel lymph node is
located in the internal mammary area and cannot be removed easily because of
the primary tumor location, especially in the event of a lumpectomy procedure,
then we have postoperatively scanned the patient and tattooed the location of the
internal mammary node in preparation for internal mammary lymph node ra-
diation. If a mastectomy is being performed and the internal mammary lymph
node is identified with the gamma probe, then the node is removed and submit-
ted for histologic evaluation. If the internal mammary node returns positive on
pathologic evaluation, then internal mammary radiation would be given.
SUMMARY
SLN mapping for breast cancer is technically feasible but there is a significant
learning curve for all surgeons, even to an extent for surgeons experienced in SLN
biopsy for melanoma. Most breast SLN are located in the axilla, rarely in the inter-
nal mammary chain. With appropriate gamma detection device selection and setup,
attention to the technical details described here, and a little patience all surgeons
should be able to learn this technique.
A national network of training centers is being established for radioguided
surgery. This network will provide an opportunity for surgeons, nuclear medicine
physicians and pathologists to come together and learn about this new technol-
ogy. Training will include didactic sessions, live surgery, and hands-on experience
with animal models. The faculty will consist of leading experts from acoss the
country. Participating centers include the H. Lee Moffitt Cancer Center and Re-
search Institute, John Wayne Cancer Institute, and the M.D. Anderson Cancer
Center. Training sites will also be available in Durham, NC, Pittsburgh, PA, Se-
attle, WA, Little Rock, AR, and St. Louis, MO. In addition, a mentoring program
will be available to assist the multidisciplinary lymphatic mapping team as they
return to their institutions to implement this new procedure into their practices.
Information on this national training and support network can be obtained by
telephone (888-456-2840) and will soon be available at the web site <http://
teleconmed.com>.
REFERENCES
1. U.S. Census Bureau, Population Division, Series P-25.
2. Kosary CL, Ries LAG, Hankey BF et al. SEER Cancer Statistics Review, 1973- 6
1992: Tables and Graphs. National Cancer Institute. NIH Pub 1995; No. 95-2789.
3. Way LW. General surgery in evolution: Technology and competence. Am J Surg
1996; 171(1):2-9.
4. Geis WP, Kim HC, McAffe PC et al. Synergisitc benefits of combined technolo-
gies in complex, minimally invasive surgical procedures. Clinical experience and
educational processes. Surg Endosc 1996; 10(10):1025-1028.
5. Cox CE, Pendas S, Cox JM et al. Guidelines for sentinel node biopsy and lym-
phatic mapping of patients with breast cancer. Ann Surg 1998; 227( 5): 645-653.
6. Frazier TG, Copeland EM, Gallaher HS, Paulus DD Jr, White EC. Prognosis and
treatment in minimal breast cancer. Amer J Surgery 1977; 133(6): 697-701.
7. Silverstein MJ, Rosser RJ, Gierson ED et al. Axillary lymph node dissection for
intraductal carcinoma: Is it indicated? Cancer 1987; 59(10):1819-1824.
8. Balch CM, Singletary ES, Bland KI. Clinical decision-making in early breast can-
cer. Ann Surg 1993; 217:207-225.
9. Moffatt FL, Senofsky GM, Davis K et al. Axillary node dissection for early breast
cancer: some is good but all is better. J Surg Oncol 1992; 51(1):8-13.
10. Silverstein MJ, Gierson ED, Waisman JR, Senofsky GM, Colburn WJ, Gamagami
P. Axillary lymph node dissection for TIa breast carcinoma: is it indicated? Can-
cer 1994; 73(3):664-667.
11. Fisher B, Wolmak N, Bauer M et al. The accuracy of clinical nodal staging and of
limited axillary dissection as a determinant of histologic nodal status in carci-
noma of the breast. Surg Gynecol Obstet 1981; 152(6):765-772.
12. Cady B. The need to reexamine axillary lymph node dissection in invasive breast
cancer. Cancer 1994; 73(3):505-508.
13. Morton DL, Wen DR, Wong JH et al. Technical details of intraoperative lym-
phatic mapping for early stage melanoma. Arch Surg 1992; 127(4):392-399.
14. Guiliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and
sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391-401.
15. Krag DN, Weaver DL, Alex JC et al. Surgical resection and radiolocalization of
the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993;
2:335-339.
opsy in the patient with breast cancer. JAMA 1996; 276:1818-1822.
17. Borgstein PJ, Pijpers R, Comans EF et al. Sentinel lymph node biopsy in breast
cancer: Guidelines and pitfalls of lymphoscintigraphy and gamma probe detec-
tion. J Amer Coll Surg 1998; 186(3):275-283.
18. Barnwell JM, Arredondo MA, Kollmorgen D et al. Sentinel node biopsy in breast
cancer. Ann Surg Oncol 1998; 5(2:126-130.
19. Veronesi U, Paganelli G, Galimberti V et al. Can axillary dissection be avoided in
breast cancer? Lancet 1997; 349:1864-1867.
20. Giuliano AE, Jones RC, Brennan M et al. Sentinel lymphadenectomy in breast
cancer. J Clin Oncol 1997; 15:2345-2350.
21. Schreiber, SH, Pendas S, Ku NN, Reintgen DS, Shons AR, Berman C, Boulware D,
Cox CE. Microstaging of breast cancer patients using cytokeratin staining of the
sentinel lymph node. Ann Surg Oncol (in press: 1998).
22. Tiourina T, Arends B, Huysmans D, Rutten H, Lemaire B, Muller S. Evaluation of
surgical gamma probes for radioguided sentinel node localization. European
Journal of Nuclear Medicine (in press).
6
Technique for Intraoperative Lymphatic Mapping in Breast Carcinoma 83
The Technique of Intraoperative

Lymphatic Mapping and Sentinel
Lymphadenectomy in Breast Cancer
Using Blue Dye Alone
Philip I. Haigh, and Armando E. Giuliano
Introduction .......................................................................................................... 83
Detailed Technique ............................................................................................... 84
Complicat ions ....................................................................................................... 88
Summary .............................................................................................................. 89
7
INTRODUCTION
The status of the axillary lymph nodes remains the most important factor as-
sociated with survival of patients with primary breast cancer. The standard method
for staging is complete axillary lymph node dissection (ALND). Whether there is
a survival benefit or local-regional benefit from ALND is controversial, with the
debate far beyond the scope of this chapter. ALND to stage curable breast cancer
patients will be performed in large numbers of patients without nodal disease and
will subject them to the risks of nerve injury and lymphedema unnecessarily. The
procedure using intraoperative lymphatic mapping and sentinel lymphadenec-
tomy (SLND), as developed by Morton1 to identify regional metastases from a
primary cutaneous melanoma, was adapted by our group in an effort to identify
axillary metastases in patients with breast cancer. The intent was to determine if
the technique could be a viable alternative to routine ALND, accurately predicting
the presence or absence of axillary metastases, without the associated morbidity
of ALND. Using 1% isosulfan vital blue dye (Lymphazurin®) as the lymphatic
mapping agent, 174 consecutive SLND procedures were performed in our initial
feasibility study, followed by completion axillary lymph node dissection.2 One
goal was to establish the optimal technique necessary to identify the sentinel node.
The sentinel node was identified in 114 of the 174 procedures (66%) and it accu-
rately predicted the axillary nodal status in 109 (96%) cases. With the SLND pro-
cedure standardized, in a follow-up study using immunohistochemistry (IHC)
stained on sections of the sentinel node, we found that SLND significantly in-
creases the accuracy of detecting metastases in the axilla and also increases the

rate of detection when compared to ALND with routine histopathologic process-

ing of random lymph nodes.3 Furthermore, to validate the power of the sentinel
node in predicting the entire axilla, 1087 nonsentinel nodes were examined in 60
patients whose sentinel nodes were tumor-free by both IHC and hematoxylin/
eosin staining. Only one additional tumor-positive node was identified, indicat-
ing that the sentinel node is the most likely axillary node to contain metastases.4
In our last report of 107 SLND procedures, using all refinements except preopera-
tive lymphoscintigraphy for medial lesions, the technique was successful in iden-
tifying a sentinel node in 94% of cases, and was 100% predictive of axillary sta-
tus.5 Currently we identify the sentinel node in more that 99% of cases. The fol-
lowing is a description of this refined technique in detail, with key features of each
step presented, which will hopefully serve to augment the accuracy of the proce-
dure utilized by those surgeons who have been trained in the technique.
DETAILED TECHNIQUE
7
After induction of general anesthesia or deep IV sedation, the patient is pre-
pared and draped with the arm free to allow for its movement. Under sterile con-
ditions, 1% isosulfan blue vital dye is injected peritumorally into the breast pa-
renchyma on the axillary side of the tumor (Fig. 7.1). It helps if the breast is el-
evated to prevent subfascial injection. If the patient has already had an excisional
biopsy, then the dye is injected into the wall of the cavity and surrounding tissue.
Biopsy cavity injection must be avoided, as lymphatic uptake will be minimal. If
the patient had a core biopsy for a nonpalpable carcinoma, and is undergoing
lumpectomy with pre-operative mammographic localization, then the injection
is made at the appropriate depth and location around the wire, or if a needle is
Fig. 7.1. Injection of isosulfan blue in the wall of the excisional biopsy cavity.
used then the dye may be injected using the same needle. The volume used ranges
between 3-5 cc, with more dye used the further the lesion is from the axilla. The
smaller volume is important for lesions high in the upper outer quadrant, because
the axilla may be flooded with blue dye and make identification of the blue node
particularly difficult.
After injection, massage or manual compression of the breast is essential to
enhance lymphatic uptake of the dye. This compression is begun immediately
after injection, continued for a specific time interval, and followed immediately
by an axillary incision. The timing of the sentinel lymphadenectomy after dye
injection depends on the location of the primary lesion. For high upper-outer
quadrant lesions, the dissection should begin no sooner than three minustes after
injection, whereas lower inner-quadrant lesions require seven to ten minutes for
dye transit to the axilla. For all other areas, five minutes is the optimal time for
sentinel lymph node dissection. This timing is crucial, for too short an elapsed
time does not allow dye to reach the sentinel node, while a delay may stain nodes
blue “beyond” just the sentinel node, or allow egress of dye from the sentinel node
making identification troublesome. 7
Sentinel lymphadenectomy is performed using a small transverse incision about
1 cm inferior to the hair-bearing area of the axilla, and slightly towards the ante-
rior axillary line (Fig. 7.2). The subcutaneous tissue is incised with electrocautery,
and dissection is continued directly perpendicular to the skin. There are often
superficial subcutaneous blue-stained lymphatics, which can be transected; these
should not be confused with lymphatics that drain the primary tumor. At this
point, the shoulder should be abducted and flexed which often helps to bring the
axillary contents anteriorly closer to the incision. With meticulous dissection using
Fig. 7.2. Axillary incision about 1 cm below hair bearing area (indicated with
purple markings).
fine Crile forceps, a blue lymphatic can be identified and easily traced to a blue
lymph node (Figs. 7.3a,b). The field must be kept dry as any bleeding rapidly
obscures the contrast of blue dye to the yellow color of fat. If a blue lymphatic is
identified, carefully follow this superiorly and inferiorly by incising surrounding
tissue and clavi-pectoral fascia to identify a blue node. This method simplifies
Fig. 7.3a. Blue lymphatic entering into blue-stained sentinel lymph node.
Fig. 7.3b. Another example of a blue sentinel

lymph node in vivo.
identification of a blue node, which may only have a small portion of its hilum
stained blue near the afferent lymphatic. If a blue lymphatic channel is not identi-
fied, blindly searching for a blue node is tedious, difficult, and often not success-
ful. Similarly, the blue lymphatic must be kept intact to facilitate identification of
the blue node. After excising the blue node, it is imperative that the afferent blue
lymphatic is followed towards the breast to ensure there is not another blue lymph
node more proximally representing the true sentinel lymph node (Fig. 7.4).
The excised sentinel node(s) is then bisected and frozen section obtained. If
negative, then the wound is irrigated and closed without drainage. According to
our current research protocol, a completion axillary lymph node dissection is not
performed if the sentinel lymph node is free of metastases. If frozen sections are
postive for metastases, then complete axillary lymph node dissection is performed
immediately. The sentinel lymph node is processed for permanent section evalu-
ation by hematoxylin and eosin staining, and if negative, further levels of the bi-
sected node are stained for cytokeratin using immunohistochemistry. If metastases
are identified after permanent sections are reviewed, then the completion lym-
phadenectomy is performed within a week after SLND. After sentinel lymphadenec- 7
tomy, the primary lesion or biopsy cavity is removed with standard breast conser-
vation surgery or total mastectomy. The significance of IHC-detected tumor
deposits is however unknown and therefore should not be a standard procedure
outside of a research protocol.
Fig. 7.4. Excised blue sentinel

lymph node.
COMPLICATIONS
Complications of this surgical procedure can occur from either the isosulfan
blue dye, or the surgery itself. In our experience, there have been no injuries to the
long thoracic or thoracodorsal nerves as a result of sentinel node biopsy. Other
complications, including infection, lymphocele, or other local or regional prob-
lems have occurred rarely, as they might from any axillary lymph node biopsy
procedure, without any apparent increase associated with sentinel node surgery
specifically.
Side effects from the blue dye may occur. After intra-parenchymal breast injec-
tion the dye typically migrates to the dermal lymphatics, causing the skin overly-
ing the injection site to become blue-stained (although less so than with intrader-
mal injection). This generally fades rather quickly, but the occasional patient may
have residual faint bluish discoloration of the breast skin for months. This almost
never occurs following total mastectomy or re-excision of the breast cancer at the
same time as sentinel node biopsy, but is more likely when the sentinel node bi-
7 opsy is performed after breast lumpectomy. All patients should be informed that
their urine will turn a green color for 24-48 hours after the procedure, a predict-
able but self-limited process as residual dye is cleared from the patient’s body.
Other complications of blue dye injection have been reported in up to 1.5% of
patients.6 Most of these reactions were of an allergic type, consisting of localized
swelling at the injection site and mild pruritus of the hands, abdomen, and neck
within minutes of injection. More rarely, more severe reactions have been re-
ported,7,8 although at least some of these may have been due to mixing the dye
with local anesthetic, which is known to cause a precipitate. These have consisted
of facial and glottic edema which has progressed in extremely rare cases to respi-
ratory distress or shock, with a reported fatality in a related compound used to
estimate the depth of a burn.6
The blue dye may also interfere with the function of percutaneous oximetry
probes. Chemically related dyes, (i.e., Patent Blue V), have been shown to absorb
light at a wavelength similar to deoxygenated hemoglobin, resulting in dimin-
ished oxygen saturation readings by standard oximetry probes. Technically, the
patient’s blood absorbance of red light is increased by tri-aryl methane dyes (such
as Patent Blue V or Lymphazurin) relative to its absorbance of infra-red light. As
this ratio increases, the calibrated (calculated) oxygen saturation linearly decreases.
This appears more likely in anemic patients and at higher intravascular dye con-
centrations.8,9
Overall, the incidence of these complications with current techniques and dos-
ages is unknown but appears low. The manufacturer6 recommends avoiding us-
age in patients with known hypersensitivity to Lymphazurin or related compounds,
pregnant women, nursing mothers, or children, due to unknown risks in those
latter three groups. The carcinogenic potential of Lymphazurin is unknown, as is
its effect, if any, on male or female fertility. Lymphazurin should never be mixed
with local anesthetics because a precipitate immediately forms. Finally, due to the
side effects described above, administration and observation by trained/informed
medical personnel, and the availability of emergency medical facilities, is recom-

mended for 30-60 minutes after administration, a requirement easily met in the
clinical context of sentinel node mapping as currently practiced at our institution
and others.
SUMMARY
The SLND technique using blue dye alone as developed over the last seven
years at our institution has been proven to precisely predict the status of the axil-
lary lymph nodes. The procedure is well tolerated, and axillary staging can be
achieved with minimal morbidity. At first the procedure can be annoyingly frus-
trating, but as with any operation, patience, keen attention to detail, and adher-
ence to sound surgical principles will lead to success. The detailed description of
the procedure and technical hints will hopefully aid surgeons who have adopted
the procedure into their own practice in identifying the sentinel node with confi-
dence. Surgeons should validate the technique at their own insitutions to ensure 7
accuracy; mastery in removing the true sentinel node will not help if the patholo-
gists cannot detect metastases with proficiency.
REFERENCES
1. Morton DL, Wen D-R, Wong JH et al. Technical details of intraoperative lym-
phatic mapping for early stage melanoma. Arch Surg 1992; 127:392-399.
2. Giuliano AE, Kirgan DM, Guenther JM, Morton DM. Lymphatic mapping and
sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391-401.
3. Giuliano AE, Dale PS, Turner RT et al. Improved axillary staging of breast cancer
with sentinel lympyhadenectomy. Ann Surg 1995; 222:394-401.
4. Turner RT, Ollilla DW, Krasne DL et al. Histopathologic validation of the senti-
nel lymph node hypothesis for breast carcinoma. Ann Surg 1997; 226(3):271-278.
5. Biuliano AE, Joens RC, Brennan M, Statman R. Sentinel lymphadenectomy in
breast cancer. J Clin Onc 1997; 15(6):2345-2350.
6. Product Insert: Lymphazurin 1% (isosulfan blue). Norwalk, CT: United States
Surgical Corporation, 1997.
7. Hietala SO, Hirsch JI, Faunce HF. Allergic reaction to patent blue violet during
lymphography. Lymphology 1977;10:158-160.
8. Longnecker SM, Guzzardo MM, Van Voris LP. Life-threatening anaphylaxis fol-
lowing subcutaneous administration of isosulfan blue 1%. Clin Pharm
1985;4:219-221.
9. Larsen VH, Freudendal-Pedersen A, Fogh-Andersen N. The influence of Patent
Blue V on pulse oximetry and haemoximetry. Acta Anaesthesiol Scand 1995;39
Suppl 107:53-55.
10. Saito S, Fukura H, Shimada H, Fujita T. Prolonged interference of blue dye “patent
blue” with pulse oximetry readings. Acta Anaesthesiol Scand 1995;39:268-269.
Lymphoscintigraphy
Claudia G. Berman
Introduction .......................................................................................................... 90
Dose Administ ration ............................................................................................. 93
Imaging Protocol ................................................................................................... 96
Radioguided Surgery .......................................................................................... 101
Summary ............................................................................................................ 105
INTRODUCTION
A diagnostic strategy utilizing selective lymph node sampling directed by

lymphoscintigraphic methods has been advocated in recent years. It has been pro-
posed that meticulous pathological study of the sentinel lymph node (SLN), the
primary lymph node draining the malignant lesion, will accurately predict the
8 pathologic status of the remainder of the draining lymph node basin. This hy-
pothesis has been shown to be true, in both melanoma and more recently carci-
noma of the breast.1 Not only nodes of the draining basin but in-transit lymph
nodes, situated between the injection site and the anatomically recognized re-
gional lymph node groups, have been found to accurately predict the pathologic
status of the regional nodal basin as a whole (Fig. 8.1). The prospect of less mor-
bid and more accurate tumor staging is apparent and appealing.2
Lymphoscintigraphy is the injection of radioactive particles around a tumor
in an effort to identify the lymph nodes that have afferent drainage from that
tumor. Lymphoscintigraphy is usually performed using technetium, which is an
ideal radionuclide for imaging. All equipment in modern nuclear medicine de-
partments is optimized for the technetium energy peak. Technetium provides a
low radiation dose to the patient(s) and staff. It is a gamma emitter only, unlike
other agents such as iodine-131 and gold, which are beta emitters. Technetium
has a short half-life of approximately 6 hours. Again, this means that the radiation
dose for the patient and staff members is lessened, and it also gives the ability to
perform repeat studies over a short period of time if this is desirable or necessary.
The choice of the particle to which the technetium is to be labeled varies with
each diagnostic problem. It is by no means certain what is the optimal particle for
lymphoscintigraphy. If a particle is too small, it will be taken up preferentially into
the capillaries which results in less of the dose going to the lymph nodes and a low
target to background ratio. If a particle is too large, it will not be taken up into
lymphatics but will be phagocytized. This results in less uptake in the lymph nodes
and more uptake at the injection site.
Lymphoscintigraphy 91
8
Fig. 8.1. Anterior view, right arm and axilla, shows uptake in the antecubital fossa (<) which represents an
in-transit lymph node which by definition is a SLN. There is also uptake in the axilla (<<).
Human serum albumin is commonly used for lymphoscintigraphy because of

its uniform particle size. I do not advocate the use of albumin because it seems to
pass through the lymph nodes rapidly. Instead of uptake in one or two SLN, we
commonly see six to eight lymph nodes with radioactive uptake following the
injection of radiolabelled albumin. We also worried that since the albumin passes
through the lymph nodes so rapidly, that it may in fact pass through the SLN by
the time the patient is presented to the operating room, thus defeating the pur-
pose of SLN mapping. It is also more difficult to use the hand held gamma probe
and identify SLN when there are a number of lymph nodes which are radioactive
(Fig. 8.2).
Historically, investigators have preferred technetium antimony trisulfide col-
loid for lymphoscintigraphic studies because its particle size, 3-30 nm, is optimal
for transit through lymphatics and localization in nodes without phagocytosis.
Unfortunately, antimony sulfur colloid is no longer produced for use in the United
States. It is still available in Australia and Europe and is used widely there. Hung
and associates have produced acceptable lymphoscintigraphic images with tech-
netium sulfur colloid filtered to a maximum particle size of 100 nm. In practice,
nearly all of their filtered activity was distributed in the 15-50 nm range.2
We too use sulfur colloid. Our filtration technique uses a 0.2 micron (20 nm)
filter and we have found this preparation satisfactory. It is important that the sul-
fur colloid be fresh because over time there is clumping of the colloidal particles,
Fig. 8.2. Left anterior oblique head and neck

lymphoscintigram; (A) immediately after,
(B) 15 minutes and (C) 45 minutes post-
injection. The immediate images show up-
take in a SLN (closest to injection site). The
lymphatics are bifurcated and two non-
SLNs are also seen. After 15 minutes, addi-
tional non-SLNs are present and after 45
minutes many non-SLNs are present mak-
ing the detection of the SLN much more
difficult.
Fig. 8.2.A
Fig. 8.2.B
Fig. 8.2.C
which in effect causes them to be larger in size and less effective. Our 0.2 micron
sulfur colloid preparation appears to work very reliably, consistently identifying
one or two SLN for excision. 8
The equipment necessary for doing lymphoscintigraphy is commonly found
in most nuclear medicine departments. We use a standard large field of view gamma
camera, a high resolution collimator and a 10% imaging window at the 140 keV
energy peak, which is the technetium energy peak.
In most situations we employ a dose of 450 microcuries. This includes cases
involving the breasts, trunk and extremities. If the lesion is very near a draining
lymph node basin, we may use 250 microcuries to minimize artifact from the
injection site. We do this only rarely, for example in breast cases where the lesion is
well within the axilla. I also advocate using 250 microcuries in head and neck
cases. The anatomy of the head and neck is very tight and compact and it is often
difficult to identify SLN because of interference produced by the injection site.
Minimizing both the volume and dose injected seems to improve results.
DOSE ADMINISTRATION
BREAST
Patients with palpable tumors are injected in the nuclear medicine depart-
ment with six injections around the periphery of the tumor at the depth of the
mass. Patients with nonpalpable tumors first undergo mammographic or ultra-
sound needle localization. If imaged by ultrasound six equal aliquots are injected,
with ultrasound guidance if necessary, at the correct depth. If imaged by mam-
mography six equal aliquots are injected equidistant from the tip of the localiza-
tion wire, at the correct depth. In small breasted patients or patients with superficial
tumors, it is often possible to perform the injections without compression. In

large patients or in patients with centrally located lesions it may be necessary to
perform the injections while the breast remains under compression to assure ac-
curate depth.
A third group of patients are studied after lumpectomy or excisional biopsy
has been performed. Most have seromas that are palpable and injections are per-
formed as if the seroma were the tumor. If the seroma is not readily palpable,
ultrasound is used and in almost all cases will identify a small residual seroma or
area of architectural distortion within the breast so that the injectate can be in-
stilled at the correct location. The physician making the injections must be able to
distinguish the tumor seroma or scar from induration or scar related to the skin
incision, which is often placed remote from the tumor to satisfy cosmetic concerns.
Care must be taken not to inject into the tumor or seroma cavity as this im-
pedes lymphatic flow. We have found injection through the localization needle to
be undesirable as the needle acts as a wick causing much of the dose to migrate to
the skin surface. This not only reduces the dose available for imaging but pro-
duces confounding contamination on the skin surface (Fig. 8.3).
Fig. 8.3. Breast lymphoscintigraphy, right anterior oblique projection. This right breast carcinoma shows
axillary drainage. The more diffuse areas of uptake (<) are the result of contamination from the “wick
effect” of a localization needle.
We have obtained good results using both 2 cc and 6 cc total diluent. The ideal
volume of injectate is unknown. On one hand a larger volume should improve
uptake within the lymphatic system. On the other hand larger volumes may dif-
fuse into the axilla, confounding imaging and detection. We are currently per-
forming a randomized study comparing results between injected volumes of 2 cc
versus 6 cc.
MELANOMA
When patients present with an intact melanoma, the radiopharmaceutical is
injected about the periphery of the melanoma in four equal aliquots. One cc of
total volume diluent is used. The injections are performed intradermally and care
is taken to obtain a skin wheal. Many cases will present following excision (biopsy
or wide local excision) of the primary melanoma. The four injections are per-
formed about the center of the scar. Two injections are placed on each side of the
scar. The injections are approximately 1 centimeter apart. It is important that the
injections not be performed at the ends of the scar because drainage from the
ends may not necessarily reflect the drainage pattern of the original tumor.
8
IMAGING PROTOCOL
A caution generally pertinent to the use of lymphoscintigraphy for SLN iden-

tification relates to the importance of confirming that the lymph nodes that pick
up the radiopharmaceutical are indeed SLN. The nuclear physician must track the
afferent drainage channels to see if multiple drainage channels culminate in mul-
tiple SLN (Fig. 8.4) or if the lymphatic channels each terminate in a single SLN
(Fig. 8.5). One must also ascertain whether or not the visualized lymph nodes run
in series, meaning that the first lymph node is the sentinel node, or in parallel,
meaning that each imaged lymph node represents a SLN.
BREAST
The breast carcinoma patient is imaged immediately after injection, positioned
supine under the gamma camera in the anterior oblique lateral projection. The
arm is extended above the head and the hand placed under the head to optimize
axillary exposure. Unlike imaging in melanoma, it is very unusual to see afferent
lymphatics. As the regions of interest are the axilla, clavicular region and internal
mammary nodes, various maneuvers are attempted to remove breast activity from
the camera’s field of view. The injection site in the breast can be shielded with
lead, but this may produce a penumbra effect which can camouflage lymph nodes.
The breast can sometimes be taped out of the field of view. Alternatively, the pa-
tient may be imaged sitting or standing so that gravity will act to remove the injec-
tion site from the field of view. Small breasted women or women with lesions near
the chest wall or axilla often present a challenge and it may not be possible to
separate the injection site from the regional lymph node groups.
8
Fig. 8.4. Anterior pelvic lymphoscintigraphy shows two afferent lymphatics which each end in a lymph
node, resulting in two SLNs (>).
Fig. 8.5. Lateral chest lymphoscintigraphy shows two afferent lymphatics ending in a single axillary lymph
node (<).
The persistence scope is used to identify accumulations of the radiopharma-

ceutical corresponding to lymph nodes. Internal mammary and supraclavicular
lymph nodes can be tattooed in the anterior projection, whilst the axillary lymph
nodes are tattooed with the patient in the lateral position with the arm above the
head. Images are acquired over 8-10 minutes per view to assure high count den-
sity. Images are obtained in the anterior, lateral and oblique positions. The patient
can be imaged on a cobalt flood source to define the body contour. A cobalt flood
source is a sheet of radioactive material that is giving off gamma rays. The patient
lies on the flood source and is between the flood source and the camera. There-
fore, the patient’s body will attenuate gamma rays and we are able to see the body
contour (Fig. 8.6). Alternatively, the body contour and landmarks can be demon-
strated using a Technetium-99m marker. SLN are found with immediate imaging
in approximately 60% of cases. Delayed imaging will be necessary in the remain-
der of cases. In our practice the optimal time for delayed imaging as well as for
surgery with hand held gamma probe guidance is 4-6 hours following injection.
Conventional wisdom, noting that most breast carcinomas arise within the
upper outer quadrant of the breast, holds that nodal drainage will be to the axilla
and only the most medial tumors will drain to the internal mammary chain. Uren
studied 32 patients with antimony sulfur colloid lymphoscintigraphy and found 8
that there was ipsilateral axillary node drainage in 85% of the cases.3 However the
Fig. 8.6. Right anterior oblique view, breast lymphoscintigram, showing uptake in two lymph nodes in the
right axilla. The patient is imaged on a cobalt flood source.
multiplicity and variability of drainage patterns was unexpected. Twenty-eight

percent of patients with outer quadrant tumors showed unexpected drainage to
internal mammary nodes while 33% of inner quadrant tumors showed axillary
drainage. Thus one-third of patients with lateralized tumors had drainage which
crossed the midline of the breast. Twenty percent of patients with upper quadrant
tumors showed direct drainage to supraclavicular or infraclavicular nodes. In one
patient an in-transit intramammary node, lying in the breast parenchyma be-
tween the primary lesion and the axilla, was discovered. This in-transit lymph
node was in fact the SLN and contained metastatic disease (Fig. 8.7). Standard
axillary dissection would not have identified this node and, by implication, the
patient’s need for systemic adjuvant therapy.
Preoperative breast lymphoscintigraphy offers the opportunity for identifica-
tion of the unique pattern of nodal drainage for each malignant lesion. Directing
the surgeon to the site of the SLN minimizes the operative time, the extent of
dissection and likelihood of late morbidity. Identification of in-transit lymph nodes
Fig. 8.7. Breast lymphoscintigraphy, anterior projection, clavicles and sternum are outlined by a Tc-99m
marker. This periareolar cancer shows flow to an intramammary lymph node (<<), and a left internal
mammary lymph node (<). Since afferent lymphatics are not identified, it is not possible to determine
whether one or both lymph nodes are sentinel. Both were marked for excision.
opens up a subset of patients inadequately examined by standard axillary dissec-

tion techniques. Identification of internal mammary SLN allows for rational ra-
diotherapeutic planning in patients who might not otherwise be recognized to be
at high risk (Fig. 8.8). The identification of a supraclavicular lymph node as a SLN
is of unknown significance given the current American Joint Commission on Can-
cer (AJCC) staging system which classifies supraclavicular adenopathy as meta-
static, stage IV, breast carcinoma.
MELANOMA
Melanoma, in distinction to carcinoma of the breast, is a much easier disease
in which to successfully perform lymphoscintigraphy due to the richness of the
cutaneous lymphatics. The patient is placed under the gamma camera immedi-
ately following injection. An effort is made to identify the afferent lymphatics. We
feel, as do others, that immediate (“dynamic”) scanning is an essential compo-
nent of any SLN identification procedure (Fig. 8.9).4 The afferent lymphatics are
followed with the gamma camera using the persistence scope. The SLN and in-
transit lymph nodes are localized, moving the patient in multiple projections, and
they are marked or tattooed. Multiple drainage basins may require additional
imaging. If the multiple drainage basins are in different locations of the body it
8
may be necessary for the patient to have more than one imaging session. Delayed
imaging is seldom required.
Fig. 8.8. Anterior view, breast lymphoscintigraphy, showing flow to right internal mammary chain only in
a lesion at about the 12 o’clock position of the right breast.
8
Fig. 8.9. Anterior chest lymphoscintigram in a patient with melanoma of the right arm. (A) immediate
view-shows drainage to a single lymph node.
Fig. 8.9. (B) ten minute delay-shows uptake in a second lymph node.
Fig. 8.9. (C) fifteen minute delay-shows continuing increase in uptake in both nodes. 8
Localization of the lymph nodes is performed by positioning the patient in

various projections and using the persistence scope and hot marker to triangulate
upon the lymph nodes (Fig. 8.10). We routinely outline the body contour and
body landmarks, such as the sternocleidomastoid and ear, with a hot technetium
marker (Fig. 8.11). The patients are imaged on a cobalt flood source to show the
body contour.
RADIOGUIDED SURGERY
Considerable recent interest has centered around a strategy of injecting tar-

geted radioisotopes systemically. This is to be followed by surgical exploration
using a radioactivity-sensitive hand-held probe to guide the surgeon to otherwise
indiscernible tumor deposits or tissues. Two distinct targeting strategies readily
present themselves. For tumor, the most attractive approach in most settings would
employ a radioactively labeled antibody specific for a definitive tumor antigen.
For metabolically distinctive tissues a radioactively labeled precursor molecule
would represent an ‘ideal’ targeting agent.
Ongoing efforts using various radioactively-labeled preparations in a wide ar-
ray of tumors and clinical situations remain in a developmental stage. At the present
time only a few such techniques are used regularly in our department, recom-
mended by consistent clinical reliability.
We have used Tc-99m-oxydronate, one of several standard bone imaging agents,
to localize bone tumors for excisional biopsies. The patient is injected with the
8
Fig. 8.10. (A) Anterior, (B) left and (C) right lateral chest lymphoscintigrams showing separate drain-
age into both axillae. The body contour is outlined by placing the patient on a cobalt flood source.
Fig. 8.10. (B)

8
Fig. 8.10. (C)
Fig. 8.11. Left lateral head and neck lymphoscintigram showing drainage into the posterior cervi-
cal chain only in a patient with a midline forehead melanoma. The ear and sternocleidomastoid
are outlined with a Tc-99m marker.
radiopharmaceutical using standard techniques. We use a dose of 25-30 mCi. and

routinely use intravenous hydration to increase wash out of soft tissue uptake.
Unlike standard bone imaging protocols, however, the injection is timed to pre-
cede surgery by 5-6 hours in an effort to optimize target to background ratios.
Imaging has invariably been performed previously and is not repeated at the time
of surgery aside from imaging of the specimen to verify removal of the targeted
abnormality.
We also routinely use Tc-99m-Sestamibi to aid the surgeon performing par-
athyroidectomy in patients with hyperparathyroidism from parathyroid adenoma.
Sestamibi is preferentially taken up in the mitochondria of tissues undergoing
oxidative metabolism. Uptake is pronounced in both the thyroid and parathyroid.
However, washout is much more rapid from thyroid tissue. The patient is injected
with 25 mCi. Sestamibi followed at one hour by imaging of the neck and upper
two-thirds of the mediastinum. The patient is positioned supine and the head and
neck are extended and immobilized. Images are acquired in the anterior and 30%
Fig. 8.12. Sestamibi scan in patient with primary hyperparathyroidism. Single focus of intense
uptake represents parathyroid adenoma. Scan was obtained one hour following injection and
residual thyroid uptake is still visible.
right and left anterior oblique projections. Each image is acquired over a 10 minute
interval. The skin overlying the adenoma—normal parathyroid glands are rarely
visualized—is marked with an indelible pen. The optimal differential washout
occurs at approximately 3 hours postinjection. Our early imaging assures that the
surgeon will benefit from the optimal differential washout at the time of actual
dissection (Fig. 8.12).
SUMMARY
Lymphoscintigraphy or targeted radioisotope scans are possible using readily

available nuclear medicine reagents and equipment. High quality scans performed
correctly are essential to a successful radioguided surgical procedure.
REFERENCES
1. Alazraki N. Lymphoscintigraphy and the intraoperative gamma probe. J Nucl
Med 1995; 36:1780-1783.
opsy in the patient with breast cancer. JAMA 1996; 276:1818-1822.
3. Uren RF, Howman-Giles RB, Thompson JF et al. Mammary lymphoscintigraphy 8
in breast cancer. J Nucl Med 1995; 36:1775-1780.
4. Taylor A Jr., Murray D, Herda S, Vansant J, Alazraki N. Dynamic lympho-
scintigraphy to identify the sentinel and satellite nodes. Clin Nucl Med 1996;
21:755-758.
Pathologic Evaluation of Sentinel

Lymph Nodes in Breast Cancer
Ni Ni K. Ku
Introduction ........................................................................................................ 106

Patholo gy Protocol .............................................................................................. 107
Patholo gic Finding s ............................................................................................. 108
Results ................................................................................................................. 111
Conclusion .......................................................................................................... 113
INTRODUCTION
Lymphatic mapping utilizing technetium labeled sulfur colloid and isosulfan

blue dye injections has successfully identified the sentinel lymph nodes (SLNs)
and allowed accurate staging.1,2 This minimally invasive technique may lead to
more conservative lymph node dissection for stages I and II breast cancer patients
9 with decreased morbidity and cost savings for the health care system. For most
cancers, the nodal status is the single most important prognostic indicator and
determines the need in many cases for adjuvant chemotherapy.3 Micrometastases
in the axillary lymph nodes has been reported to be associated with poorer sur-
vival.4 Currently, there is a failure rate of 15% to 20% at five years in node-nega-
tive patients which may be attributed to the low detection rate of micrometastases
using the routine hematoxylin and eosin (H&E) stain.5 With the advent of lym-
phatic mapping, these SLNs may be carefully and thoroughly evaluated for suc-
cessful identification of micrometastases as SLN involvement determines whether
the surgeon will perform a complete lymph node dissection (CLND). Generally,
frozen section is not recommended for lymph node tissue as it causes extensive
freezing artifacts and may obscure the histology of the tumor or lose the micro-
scopic metastatic tumor cells in the cryostat. In addition, traditional staining meth-
ods with H&E alone on sectioned lymph nodes may not detect the micrometastatic
tumor cells in a background of millions of lymphocytes. We have developed a
protocol utilizing an intraoperative imprint cytology (IIC) of the SLNs to con-
firm the metastatic tumor in grossly positive and suspicious nodes and to identify
micrometastases in grossly negative nodes. IIC has therefore become a crucial part
of our intraoperative surgical management, enabling the immediate decision for
CLND. IIC is followed by a standard histopathology protocol in concert with an-
cillary immunohistochemical stain for low molecular weight cytokeratin
(CAM 5.2) on grossly negative SLNs.
Pathologic Evaluation of Sentinel Lymph Nodes in Breast Cancer 107
PATHOLOGY PROTOCOL
The radiolabeled SLNs are submitted to the pathology processing room iden-
tified as SLN 1, SLN 2, SLN 3, etc. SLNs are bisected intraoperatively and exam-
ined for any gross evidence of metastatic disease. About one-quarter of the SLN is
then snap frozen in the operating room for RT-PCR analysis for other protocols.
At our institution, all physicians including surgeons, radiologists, pathologists,
nuclear medicine staff and intraoperative personnel routinely wear radiation
monitoring badges.
The SLNs are dissected free from the surrounding fat and measured. SLNs
with a measurement of 5 mm or less in the maximum dimension are bivalved and
those greater than 5 mm in diameter are serially sectioned at 2-3 mm intervals to
maximize the surface area for IIC. The cut surfaces are carefully examined for
gross evidence of metastases and the size of tumor noted. Frozen sections of the
SLNs are not done in our institution. IIC utilizes glass slides prelabeled with the
patient’s initials and SLN number. Imprints are made with a gentle, single touch
on each cut surface of the SLN. Imprints of multiple cut surfaces may be prepared
on the same glass slide from each SLN without overlapping. The slides are air-
dried and stained with Diff-Quik stain (Table 9.1) followed by immediate (intra-
operative) interpretation. The turn-around time from specimen receipt until cy-
tologic diagnosis is approximately 5-8 minutes for each SLN. The diagnostic ter-
minologies that are utilized for reporting IIC are: negative, indeterminate and 9
positive. Indeterminate is used when atypical cells of undetermined origin (his-
tiocytic versus epithelial) or rare suspicious cells are identified. From the surgical
management point of view, the negative and indeterminate categories are consid-
ered negative and only patients with definitively positive SLNs receive immediate
CLND.
We have recently developed a technique of intraoperative utilization of immu-
nocytochemical cytokeratin stain on cytologic touch imprints of the SLNs as an
adjunct to the Diff-Quik stain. Such technique is extremely useful in supporting
the intraoperative diagnosis or detecting micrometastases in lobular or low grade
ductal cancers. The VECTASTAIN universal Quick Kit (Vector Laboratories,
Burlingame, CA) is used along with a primary monoclonal antibody against low
molecular weight cytokeratin (CAM 5.2) (Becton-Dickinson Immunocytochem-
istry System, San Jose, CA). The turnaround time for such IIC is approximately 16
minutes. Currently, the procedure is performed parallel to the Diff-Quik stain on
all SLNs with known lobular and low grade ductal cancers, when the result of
Diff-Quik stain is indeterminate and when there is a discrepancy between gross
and cytologic findings (grossly positive or suspicious and cytologic imprint nega-
tive). Each of the SLN is then placed in separate formalin containers by the nuclear
medicine staff and quarantined at room temperature for 48 hours along with the
lumpectomy or mastectomy specimen, to allow decay of the technetium labeled
probe used for intraoperative lymphatic mapping. The nuclear medicine staff
monitors these specimens for background emission and resubmits them to the
pathology laboratory when six half-lives of the technetium have dissipated.
Table 9.1. DIFF-QUIK Stain
A. Thoroughly air dry slides.

B. Stain with Diff-Quik solution as follows.
Step 1: Fix for 10-15 seconds in solution I. Drain excess solution onto a paper towel.
Step 2: Dip slides repeatedly for 10-20 seconds in solution II until slides are uniformly
coated and turn red-pink. Drain excess stain.
Step 3: Dip slides repeatedly for 10-20 seconds into solution III. Rinse in tap water.
Step 4: Drain excess water. The slides may be examined wet without coverslipping.
Step 5: Mount in resin and coverslip after intraoperative evaluation.
(Editors’ Note: Quarantine of pathologic specimens, especially the sentinel nodes, is

not universally practiced nor necessarily required, depending on the radionuclide
dosage used and the specific radiation safety guidelines of individual institutions. For
further details, see the chapter 2). The specimens are then accessioned, the SLNs
are submitted entirely and processed for routine histopathological examination
with H&E stain. This usually involves one to two blocks per SLN (average one
block per SLN). Each block is sectioned at one to three levels per slide for H&E
stain. In addition, any SLNs that are grossly and IIC negative for metastases are
immunohistochemically stained with a monoclonal antibody against low-molecu-
lar weight cytokeratin (CAM5.2), clone NCL-5D3, using the avidin-biotin com-
9 plex technique with diaminobenzidene chromogen.
PATHOLOGIC FINDINGS
On cytologic touch imprints stained by Diff-Quik technique, the metastatic

tumor cells are usually arranged in small to large cohesive clusters, syncytial frag-
ments, as well as single cells (Fig. 9.1). Difficulty in interpretation may occur based
upon the low volume of tumor cells regardless of their arrangement or the small
size and bland appearance of the metastatic cells, particularly in metastatic lobu-
lar or low grade ductal carcinomas. Aggregates of reactive histiocytes may look
like epithelial cell clusters on touch imprints, especially following excisional
biopsies.
In the permanent histologic sections stained by H&E and cytokeratin
immunostain (CKI), the clusters of micrometastatic tumor cells can be seen in
the: 1) subcapsular sinuses (most common pattern; 2) medullary sinuses (less
common pattern); and 3) interfollicular areas. Diffuse single cell patterns of
micrometastases are seen commonly in metastatic lobular carcinoma where H&E
is often negative even after multiple levels are examined with scanning magnifica-
tion, but are found with the immunohistochemical stain for cytokeratin (Fig. 9.2).
Micrometastases are arbitrarily defined as tumors of less than 2 mm in size.
Serial sectioning of the lymph node and cutting levels as suggested above from
the paraffin blocks to include the nodal capsule are essential since the most com-
mon location of micrometastases is in the subcapsular sinuses (Figs. 9.3 and 9.4).
Fig. 9.1. Intraoperative imprint cytology of sentinel lymph node with metastatic ductal Carcinoma
as loose clusters and single cells, Diff-Quik stain, X400
Fig. 9.2. Small microscopic tumor cells in single cell pattern distribution demonstrated
only by cytokeratin immunostaining (Hematoxylin & Eosin stain negative), X400
The results of CKI must be interpreted in the context of available data such as
tumor/nuclear grade (low grade or high grade), tumor type (ductal or lobular),
tumor cell arrangement (cohesive clusters or single cells) and tumor cell distribu-
tion. The CKI of SLNs are defined as positive only when the charactistic cytoplas-
mic staining pattern in the form of “donuts’ or “rings” is present, despite the quan-
tity of cells and its arrangement in the given node. Cytokeratin positive cells in the
SLN other than metastatic carcinoma are interdigitating dendritic reticulum cells,
Fig. 9.3. Incomplete sectioning of the lymph nodal capsule missing the subcapsular
micrometastases, hematoxylin & e osin stain, X100
Fig. 9.4. Subcapsular micrometastases highlighted only cytokeratin immunostaining,

X100 (deeper sections of Fig. 9.3)
and benign mammary and adnexal (sweat, sebaceous or apocrine glands) inclu-
sions. Dendritic reticulum cells are distributed around germinal centers in a mesh-
like network, and are single cells with small nuclei and dendritic fibrillary cellular
processes. Epithelial inclusions and benign nevus cells are very rare and usually
located in the nodal capsule or perinodal fat. With experience, these pitfalls can be
avoided by comparison of the positively stained areas with the corresponding H&E
stained slide and confirming the cells to have a benign histology. We have seen
extremely rare cases of a nonspecific staining pattern or aberrant antigen expres-
sion appearing as diffuse nuclear and cytoplasmic staining pattern but with mod-
Table 9.2. Errors and potential pitfalls in interpretation of sentinel lymph nodes
False negative False positive
Imprint Cytology sampling error aggregates of reactive/
atypical histiocytes
interpretative error
low grade ductal or lobular follicle fragments
carcinoma
low tumor volume
single cell distribution
Permanent incomplete section of nodal
Histolgy capsule nevus cell rest
sampling/interpretative error sinus histiocytosis
lobular carcinoma
low tumor volume
single cell distribution
Tissue CKI interpretative error dendritic reticulum cells
mammary inclusions
adenexal inclusions
hemosiderin-laden
macrophages
erately weak intensity. This staining is often in an unusual location such as germi-
nal centers of the nodes and usually involves only one or two single cells. Errors
and potential pitfalls in interpretation of SLNs in breast cancer patients are given 9
in Table 9.2.
RESULTS
During the period of January 1996 to August 1997, 225 patients with the diag-
nosis of breast cancer by fine needle aspiration biopsy, core needle biopsy, ABBI
or excisional biopsy underwent lymphatic mapping with selective sentinel lym-
phadenectomy at our institution. Of 397 SLNs (from 225 patients), 16 (from 15
patients) grossly positive or suspicious SLNs were confirmed to be positive (sensi-
tivity 100%) by IIC and those patients were converted to CLND. The results of IIC
(by Diff-Quik stain), permanent histology (PH) by standard H&E stain and CKI
on the remaining SLNs were compared in Table 9.3, and the significance of CKI
upon detecting micrometastases was determined. Of the remaining 381 SLNs (from
210 patients), 22.4% (47 patients) were positive for metastatic carcinoma by H&E
and/or CKI, and 77.6% (163 patients) were negative by H&E and CKI. CKI de-
tected micrometastases in 35 and 28 SLNs not identified by IIC (Diff-Quik stain)
and H&E, respectively. Seventeen of 47 patients (36.2%) with metastatic disease
were detected by CKI only and therefore metastases in such patients would have
been missed by H & E alone. Out of 180 patients with H&E negative SLNs, 163
(91.6%) had SLNs that were both H&E and CKI negative while micrometastases
was detected in 17 (9.4%) patients by CKI only, enabling to upstage them from
N0 to N1 (Table 9.4).
Table 9.3. Correlation between IIC (Diff-Quik), permanent histology and CKI in 381
SNLs
IIC (Diff-Quik)* Histology (H&E) CKI (tissue)
True positive 15 50 68
True negative 254 313 313
False negative 35 28 –
False positive 1 0 –
*IIC not done in 76 cases
Table 9.4. Statistical data including upstaged nodal status by CKI in 210 patients
The above data was recently updated to April 1998 and showed that out of 478
patients with the retrieval of 972 SLNs, 10.6% of patients were upstaged by CKI.
No false negative or positive cases were seen with CKI in our series. With a recent
introduction of intraoperative CKI on touch imprint of the SLNs in conjunction
with the standard Diff-Quik cytologic stain, the detection of micrometastases has
become more sensitive and were able to upstage in 27% of patients intraopera-
tively enabling immediate conversion to CLND.
One false positive cytology in our initial series was associated with reactive/
atypical histiocytes and false negative cases by intraoperative Diff-Quik stain were
secondary to sampling error, low tumor cell volume or single cell distribution of
lobular or low grade ductal carcinoma cells. Pathological analysis of the nonsentinel
lymph nodes in the CLND by H&E stain reveals additional metastatic disease in
45% of patients with both H&E and CKI positive SLNs and only 5% of patients
with CKI only positive SLNs. In another recent study, Turner demonstrated that if
the SLN is negative by both H&E and CKI, the probability of nonsentinel lymph
node involvement is less than 1.0%.6
CONCLUSION
Lymphatic mapping with selective sentinel lymphadenectomy allows a detailed

pathologic examination of the nodes most likely to contain metastastic disease for
accurate staging and detection of micrometastases. Intraoperative identification
of micrometastases is critical because it enables the immediate conversion of these
patients’ surgical treatment to CLND, rather than the CLND being done during a
second, deferred, surgical procedure. IIC of the lymph nodes has been used rou-
tinely in the diagnosis of hematologic malignancies and also in breast cancer as a
useful method in many series.7,8 Giuliano and Veronesi have reported successful
detection of metastases in SLNs by frozen section,9,10 but we do not recommend
these techniques because of potential loss of material in the cryostat, sampling
error from low tumor volume, and freezing artifact. In our experience, IIC by
Diff-Quik stain is an excellent tool to confirm the SLNs with grossly positive or
suspicious (100% sensitive) metastatic disease. Concurrent utilization of intraop-
erative Diff-Quik stain and CKI appears to be more sensitive in detecting
micrometastases intraoperatively based on our preliminary, as yet unpublished, 9
observations. This group of high-risk, traditionally node-negative patients can be
more accurately staged with lymphatic mapping and more detailed examination
of the SLN. The CKI that we use on the paraffin embedded sections is a low mo-
lecular weight (CAM 5.2), clone NCL-5D3, and consists of cytokeratin proteins 8
and 18. This combination of antigens is a very sensitive indicator and is present
on all epithelial cells including ductal and lobular epithelial cells of the breast. IIC
technique utilizing either Diff-Quik stain or CKI requires intensive training and
experience to avoid potential pitfalls in interpretation.
In conclusion, IIC of SLNs stained by Diff-Quik preparation is a straightfor-
ward, rapid, and accurate approach to immediate evaluation of lymph nodes at
high risk for breast cancer metastases, particularly if those SLN are grossly posi-
tive or suspicious. IIC stained by CKI is an extremely useful ancillary technique as
an adjunct to Diff-Quik stain in detecting micrometastases particularly in meta-
static low grade ductal or lobular carcinoma with low tumor cell volume. Appro-
priate utilization of Diff-Quik and CKI allows immediate intraoperative staging
and cost effective management of breast cancer. SLN mapping technology at our
institution, in conjunction with serial sectioning of the SLN and CKI, detected
micrometastases in 10.6% of patients not detected by IIC (Diff-Quik) and PH,
providing a more accurate nodal staging so that CLND and adjuvant therapy can
be given in a selective fashion. These results have encouraged investigators to pur-
sue even more sensitive techniques to uncover micrometastases such as molecular
biology technique like reverse transcriptase-polymerase chain reaction (RT-PCR).
Experienced cytopathologists and an active cytopathology service are required to
avoid the potential pitfalls of performing and interpreting IIC.
Overall, pathologic examination is a key component of radioguided surgery
and lymphatic mapping. One of the proposed advantages of SLN biopsy for breast
cancer is that women with negative SLN need not receive complete CLND. How-
ever, this surgical decision can only be made with adequate pathology support,
both intraoperatively with IIC and postoperatively (serial permanent sections).
Performing sentinel node biopsy using only standard, bivalved H&E pathology
techniques is an inappropriate application of new technology and is potentially
damaging to the patient.
ACKNOWLEDGMENT
I would like to express my sincere appreciation to Santo V. Nicosia, M.D., Chair-
man of the Department of Pathology at the University of South Florida, Nazeel
Ahmad, M.D., Prudence Smith, M.D. and cytopathology residents for their con-
tinuous support and participation with daily intraoperative activities, Solange
Pendas, M.D. for statistical data, and surgical pathology staff at the Moffitt Cancer
Center for their contributions.
REFERENCES
1. Guiliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and
9 sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391-401.
2. Albertini JJ, Lyman GH, Ku NNK, Cox CE, Nicosia SV, Reintgen DS et al. Lym-
phatic mapping and sentinel node biopsy in patients with breast cancer. JAMA
1996; 276:1818-1822.
3. Statman R, Giuliano AE. The role of the sentinel lymph node in the manage-
ment of patients with breast cancer. Advances in Surgery, Vol. 30, Mosby Year
Book, Inc. 1997.
4. Anderson BO, Austin-Seymore M, Gralow J, Moe RR, Byrd DR. A multidisci-
plinary approach to locoregional management of the axilla for primary operable
breast cancer. Cancer Control 1997.
5. Dowlatshahi K, Fan M, Snider HC. Lymph node micrometastases from breast
carcinoma. Reviewing the dilemma. Cancer 1997; 80(7):1188-1197.
6. Turner RR, Ollila DW, Krasne DL, Guiliano AE. Histopathological validation of
the Sentinel Lymph Node hypothesis for breast carcinoma. Ann of Surg 1997;
226:271-276.
7. Fischer CJ, Boyle S, Burke M, Price AB. Intraoperative assessment of nodal status
in the selection of patients with breast cancer for axillary clearance. Br J Surg
1993; 80:457-458.
8. Hadjiminas J, Burke M. Intraoperative assessment of nodal status in the selec-
tion of patients with breast cancer for axillary clearance. Br J Surg 1994;
81:1615-1616.
9. Guiliano AE, Dale PS, Turner RR et al. Improved axillary staging of breast cancer
with sentinel lymphadenectomy. Ann Surg 1995; 222(3):394-401.
10. Veronesi U, Paganelli G, Galimberti V et al. Sentinel node biopsy to avoid axil-
lary dissection in breast cancer with clinically negative lymph nodes. Lancet 1997;
349:1864-1867.
Pathologic Evaluation of the Sentinel Lymph Node in Malignant Melanoma 115
Pathologic Evaluation of the Sentinel

Lymph Node in Malignant Melanoma
Jane L. Messina
Introduction ........................................................................................................ 115

Review of the Literature ...................................................................................... 115
Patholo gic Protocol for Handling Sentinel Nodes ............................................... 116
Results ................................................................................................................. 118
Discussio n ........................................................................................................... 121
INTRODUCTION
Successful lymphatic mapping and sentinel lymph node (SLN) biopsy requires
a specialized but multidisciplinary approach, utilizing the surgeon, nuclear medi-
cine physician, and pathologist. With sentinel lymphadenectomy rapidly becom-
ing the standard of care for patients with melanoma at risk for metastases, pa-
thologists are encountering these specimens with increasing frequency in their
daily practice. The pathologic status of the sentinel lymph node is pivotal to the
patient’s care because it provides staging information that dictates the need for 10
further therapy. Therefore, defining a method for detailed assessment of this tis-
sue is of the utmost importance. Our standard pathology protocol for SLN in-
cludes complete submission of all tissue with routine use of immunohistochemi-
cal staining for S-100 protein, and intraoperative touch preparation cytology as
an adjunct technique in SLN grossly suspicious for metastatic disease. This chap-
ter reviews the literature concerning pathology of sentinel nodes in malignant
melanoma patients and describes our results in over 400 melanoma patients to
date.
REVIEW OF THE LITERATURE
In the presentinel node era, lymph nodes from melanoma patients were ex-
cised as part of complete nodal basin dissections, either for therapeutic purposes
or as an elective procedure. Pathologic evaluation procedures often entailed only
submission of a representative section from each node identified. A limited amount
of prognostic information could be gathered from these dissections. Lymph node
parameters demonstrated to have an adverse effect on disease-free and overall

survival include: the number of positive nodes (> 5 vs 2-4 vs 1), clinical vs micro-
scopic nodal involvement (80% vs 36-50% five year survival), and extracapsular
extension.1-3 However, even in the mid-1980s it was recognized that the incidence
of nodal involvement in elective dissections may be underestimated with H&E
stains alone. Cochran et al demonstrated a 14% increase in the incidence of
metastatic melanoma detected in elective node dissections using S-100
immunostaining.4
With the advent of selective lymphadenectomy, pathologists have recognized
that a detailed method for assessing these nodes is needed, but there is little con-
currence in the general literature as to exactly what constitutes an ideal protocol.
The standard method of evaluating lymph nodes for metastatic tumor involves
submitting 1-2 representative sections from the center of the lymph node for H&E
staining. This method is sensitive enough to detect 1 abnormal cell in a back-
ground of 104 normal cells.5 This sensitivity can be increased 10-fold by using
immunohistochemical staining.6 Immunohistochemical stains which have been
particularly useful in melanoma diagnosis include S-100 and HMB-45. S-100 is
an acidic, calcium-binding protein so named because of its solubility in 100%
ammonium sulfate. It was first identified in mammalian brain tissue, and subse-
quently identified in tumors of neural origin. Extensive studies have now shown
this protein to be present in a wide variety of non-neurogenic tissues, as well are
their corresponding tumors.7 S-100 antibody is 95-100% sensitive in detecting
both primary and metastatic melanoma.8 HMB-45, a marker of melanosomes, is
92% sensitive in diagnosing primary melanoma9 but in our experience it is ap-
10 proximately 50% sensitive in diagnosing micrometastatic malignant melanoma
(unpublished observations).
The usefulness of immunohistochemistry in studying sentinel lymph nodes
was demonstrated in an early study of 15 sentinel node negative patients with
nodal basin recurrence. Retrospective evaluation of these patient’s sentinel nodes
using serial sectioning, S-100, and HMB-45 immunostains confirmed the pres-
ence of micrometastases in 66% of these “negative” sentinel nodes.10 A more re-
cent study confirmed these findings.11 In this abstract, the immunostains were
instrumental in identifying the occult metastases, while serial sections were useful
in only one case.
PATHOLOGIC PROTOCOL FOR HANDLING SENTINEL NODES
With these issues in mind, we instituted the following protocol in October

1995 (Table 10.1). Following harvesting from the patient, each SLN is placed in a
separate formalin container and held at room temperature for at least 48 hours
(six 1/2 lives) to allow decay of the technetium-labeled sulfur colloid used for
intraoperative mapping. Specimens are then accessioned and processed in the
pathology laboratory in the same manner as routine, nonradioactive specimens.
Gross examination is performed, and the presence or absence of vital blue dye,
Table 10.1. Algorithm for pathologic evaluation of sentinel node
Sentinel node(s) obtained
Place in formalin for 48 hours

Transport to pathology
▲
Serially section entire node at 2-3 mm
intervals, submit entirely
▲
▲
Gross metastatic disease No gross disease
▲ ▲
H&E stain only S-100 all sections
▲
▲
▲
Metastatic Malignant Negative Benign
melanoma cytology for melanoma cytology
▲ ▲
Metastatic
melanoma
Nevus cell
aggregates
10
melanin-like pigment, and tumor nodules are noted. The lymph node is sectioned
at 2-3 mm intervals and entirely submitted for paraffin embedding. This usually
entails submission of one to eight cassettes of tissue per SLN (average of
1.5 blocks/node). One glass slide per block is cut and stained with H&E; this slide
may contain 1-3 lymph node profiles, depending on the size of tissue in the block.
Additionally, all blocks of lymph nodes that are grossly negative for melanoma
metastases are stained with S-100 protein antibody using the avidin-biotin com-
plex immunoperoxidase technique with diaminobenzidine chromogen. When
completion dissections are performed (after a positive SLN biopsy), the lymph
nodes are entirely embedded and sections stained with H&E. We have studied a
group of 20 complete node dissections with S-100 immunostaining as well (see
results).
Occasionally, sentinel nodes may contain small areas of brown-black pigmen-
tation without frank tumor nodularity, and as such be suspicious for metastatic
melanoma. In this selected group of patients, we may perform touch preparations
for cytologic examination before fixation at the time of lymphadenectomy. The
SLN is bivalved and each half is touched with a slide, which is stained with
Diff-Quik.
RESULTS
During the period of October 1995 to March 1998, 1005 SLN from 405 pa-
tients at our institution were studied with this technique. Metastatic melanoma
was detected in 92 lymph nodes from 72 patients (18% of patients).
Four patterns of lymph node involvement by metastatic malignant melanoma
are seen: 1) The most common distribution of tumor within lymph nodes is a
subcapsular nodule or several nodules of tumor cells (Fig. 10.1); 2) Tumor may
less commonly be seen as a dominant nodule in the medullary area of the lymph
node; 3) Diffuse nodal involvement by small aggregates of tumor cells is fairly
uncommon (Fig 10.2); 4) Micrometastatic disease, which may not be visible on
initial H&E sections, is encountered in two situations. In the first scenario, the
Fig. 10.1. Subcapsular

aggregate of metastatic
malignant melanoma.
10
Fig. 10.2. Diffuse distri-

bution of metastatic
malignant melanoma.
tumor volume is so small that it is not present in initial sections but only detected
after deeper sections through the block are performed for immunostaining. In the
second situation, the tumor is only identified on immunostaining, but on retro-
spective review is present in initial sections in amounts too low to be appreciated
on scanning magnification (Figs. 10.3, 10.4). The techniques described above are
especially useful in evaluating micrometastatic disease.
Cytologic features of the typical melanoma tumor cell are well known: the
cells are large, epithelioid, and display ample eosinophilic to brown-black
cytoplasm, a prominent nucleolus, and occasional nuclear pseudoinclusions.
10
Fig. 10.3. Rare clumps of strongly S-100 positive cells of metastatic malignant mela-
noma, scattered weakly positive single dendritic cells.
Fig. 10.4. Histologic section corresponding to Fig. 3, showing rare aggregates of pig-
mented, cytologically malignant cells.
Multinucleated tumor cells and bizarre mitotic figures are frequent. Cytoplasmic
melanin formation is not often readily visible on H&E sections. The diagnosis of
micrometastatic malignant melanoma rests on finding clumps or aggregates of
strongly S-100 positive cells within the lymph node, which on retrospective re-
view of the original H&E section or deeper H&E sections are found to demon-
strate these malignant cytologic features.
In our study population metastatic melanoma was detected in 95 lymph nodes
from 72 patients. In 62 of these lymph nodes (from 49 patients), metastatic dis-
ease was detected on the H&E stain alone. S-100 immunostaining revealed an
additional 31 positive nodes from 23 patients, increasing the yield of detected
disease by 33% (32% of all node positive patients). In three lymph nodes from
three of these latter patients (4%), the tumor was not present on the original H&E
slide and was only seen on the deeper section cut for immunostaining (Table 10.2).
Thus, in 23 patients (32% of patients with positive sentinel lymph nodes), the
diagnosis of metastatic melanoma would have been missed on H&E sections alone
and was only picked up by S-100 immunostaining. Overall, 19/405 patients (4.7%
of total study population) were upstaged using these special techniques. These 19
patients comprised 7% of the node-negative population.
Touch preparations were performed on 23 lymph nodes from 13 patients and
results compared to the final pathology using the above techniques. In three lymph
nodes from three patients, the cytology was interpreted as negative or suspicious
but not diagnostic for metastasis, while the corresponding histology revealed meta-
static melanoma. In the remaining 19 lymph nodes (10 patients), there was agree-
10 ment between the cytology and histology (sensitivity of 62%, specificity of 100%).
In all patients with a positive sentinel lymph node, a completion nodal dissec-
tion of the involved basin was offered, and it was performed in 61 patients (85%
of SLN positive patients). Pathologic analysis of these lymph nodes revealed addi-
tional disease in 6 (10%) of these patients. Overall, the incidence of metastatic
melanoma within nonsentinel lymph nodes of the complete dissection was 1.4%.
This is in contrast to the frequency of metastatic disease in the SLN, which is 18%.
No additional disease in the completion node dissection specimens was uncov-
ered using S-100 immunostaining in the 20 cases subjected to this technique.
Table 10.2. Histologic findings in sentinel lymph nodes

# Patients % Patients # Lymph % Lymph
Nodes Nodes
H&E first 47 65% 62 65%
H&E only 2 3% 2 2%
S-100 first 23 32% 31 33%
Total 72 100% 95 100%
DISCUSSION
Identification of metastatic disease in SLNs plays a critical role in the manage-

ment of melanoma patients, and the techniques used must be both sensitive and
specific. The specificity of S-100 is not high; S-100 antigen is present on a wide
variety of normal and abnormal lymph node constituents (Table 10.3). The most
numerous of these are the interdigitating reticulum cells (IRC) of the interfollicular
regions. These cells form a mesh-like network surrounding germinal centers. They
are seen as single cells with small nuclei, the same size as lymphocyte nuclei, and
dendritic cellular processes (Fig. 10.5). IRC stain lightly with S-100. In contrast,
metastatic melanoma forms cohesive cellular aggregates of strongly S-100-positive
tumor cells, which are much larger than the surrounding lymphocytes. These cells
Table 10.3. Distribution of S-100-positive cells in lymph nodes

S-100 Positive Cells Distribution
Interdigitating Reticulum Cells Around Germinal Centers,

Fine Net-like Distribution
Adipose Tissue Perinodal Fat
Peripheral Nerves Intranodal and Perinodal
Connective Tissue
Benign Nevus Cell Aggregates Small, Nested Groups within
Lymph Node Capsule 10
or Fibrous Trabeculae
Metastatic Malignant Melanoma See Text
Fig. 10.5. Typical pattern of interdigitating reticulum cells.

must be positively identified as having malignant cytologic features on the corre-
sponding H&E stain before they are diagnosed as melanoma.
Another S-100 positive cell found in lymph nodes is the benign nevus cell
aggregate (NCA). Small groups of S-100 positive, cytologically benign melano-
cytes are found in the lymph node fibrous capsule or trabeculae in < 1-22% of
lymphadenectomy specimens.12 The reported frequency of benign NCA in the
literature ranges from 33% to 7.3%.13,14 Numerous theories have been expounded
concerning the origins of these cells, from errant migration from the neural crest
to dislodgment and “benign metastasis” from cutaneous nevi. These cells can be
distinguished from malignant melanoma by their completely benign cytology
(Figs. 10.6, 10.7). NCA were found in 6.3% of sentinel lymph nodes in our study
population.
10
Fig. 10.6. Intracapsular aggregate of S-100 positive cells of a benign nevus cell aggregate.
Fig. 10.7. Histologic section corresponding to Fig. 10.6, showing benign cytology of
nevus cell aggregate.
In contrast to S-100, the HMB-45 antibody, while quite specific (95%), has
been less sensitive (45-65%) in detecting micrometastatic disease in our patients.
HMB-45 immunostaining is useful in cases where S-100 positive cells are found
within the lymph node, but demonstrate equivocal cytologic atypia. If question-
able areas are found to be HMB-45 positive, they are considered to represent meta-
static melanoma.
Since histologic techniques are not 100% sensitive in detecting all
micrometastatic disease, several other methods have been investigated to this end.
A cell culture technique using portions of lymph nodes from melanoma nodal
dissections, followed by immunohistochemical staining, increased detection of
occult nodal metastases by 31%.15 The reverse-transcriptase polymerase chain re-
action technique for detection of tyrosinase revealed that 47% of histologically
negative lymph nodes contained the RNA signal for tyrosinase.5 This has recently
been shown to be of prognostic significance; the recurrence rate for these patients
is 13%, compared to 2% for lymph nodes negative both histologically and by PCR
analysis.16 Serial sectioning has also been advocated, increasing detection of meta-
static melanoma by up to 24%.17
Our results are in concurrence with these data. Immunohistochemical stain-
ing of SLNs significantly increases the detection of melanoma metastases. The
antibody of choice for this procedure is S-100, because with a sensitivity of
95-100%, it is a perfect screening tool. Pathologic interpretation of S-100 stained
lymph nodes may be difficult because the antibody lacks specificity, staining a
number of normal lymph node counterparts. With experience, however, such pit-
falls can be avoided by comparison of the positively-stained areas with the corre- 10
sponding H&E-stained slide and confirming the benign cytology of the cells in
question.
Other pitfalls in the diagnosis of metastatic melanoma include melanin-laden
macrophages and foamy macrophages. Melanin-laden macrophages (melano-
phages) are frequently found in lymph nodes draining the sites of chronic cutane-
ous inflammation (dermatopathic lymphadenopathy). Their existence, therefore,
in lymph nodes draining the site of a malignant melanoma is not surprising. The
presence of large nodal aggregates of melanophages is a cause for interpretative
caution. S-100 is extremely useful in detecting the presence of viable melanoma
cells in such lesions, as the antibody does not stain most melanophages. In heavily
pigmented lymph nodes, however, a red immunoperoxidase chromogen
(aminoethylcarbazine) is used in lieu of the standard brown end product. Expan-
sion of the subcapsular sinuses by histiocytic infiltrates may be especially promi-
nent in lymph nodes draining regions of joint prostheses or silicone breast im-
plants. This may occasionally be so exuberant as to resemble metastases of bal-
loon cell melanoma. However, careful high-power examination will reveal that
these histiocytes display benign nuclear features.
In summary, the technique of lymphatic mapping and SLN biopsy is an effec-
tive method of accurately staging melanoma patients and identifying those who
may benefit from further surgery and/or adjuvant chemotherapy. The typical SLN
pathology specimen consists of 1-2 lymph nodes, which are the most likely to
contain metastatic tumor and a detailed pathologic examination using immuno-
histochemistry is both feasible and practical. We have demonstrated that thor-
ough evaluation of sentinel lymph nodes requires full utilization of the tissue sub-
mitted to the pathologist. This evaluation includes complete embedding of the
entire lymph node as well as routine immunohistochemical staining for S-100 in
all nodes grossly negative for tumor. These techniques will increase the yield of
detected disease by 39%. Although our study sample was small, the technique of
touch preparation of these lymph nodes appears to be useful in nodes, which are
grossly suspicious for metastases. Of our study population of 405 patients, only
six patients with positive sentinel lymph nodes (1.4%) had additional metastatic
disease in the completed resected lymphatic basin. All of these patients had a pri-
mary melanoma > 3.00 mm in thickness. Because of this low incidence of addi-
tional disease in the basin outside of the sentinel lymph node, at this time S-100
immunostaining of the entire dissection specimen does not appear to be war-
ranted. Further studies will be needed to determine the prognostic significance
and clinical correlation of such occult, micrometastatic disease in patients with
malignant melanoma.
REFERENCES
1. Buzaid AC, Tinoco LA, Jendiroba D et al. Prognostic value of size of lymph node
metastases in patients with cutaneous melanoma. J Clin Oncol 1995; 13:2361-8.
2. Coit DG, Rogatko A, Brennan MF. Prognostic factors in patients with melanoma
metastatic to axillary or inguinal lymph nodes. A multivariate analysis. Ann Surg
1991; 214(5):627-636.
10 3. Callery C, Cochran AJ, Roe DJ et al. Factors prognostic for survival in patients
with malignant melanoma spread to the regional lymph nodes. Ann Surg 1982;
196:69.
4. Cochran AJ, Wen DR, and Morton DL. Occult tumor cells in the lymph nodes of
patients with pathological stage I malignant melanoma. Am J Surg Pathol 1988;
12(8):612-618.
5. Wang X, Heller R, Cruse CW, VanVoorhis N, Glass LF, Fenske NA, Berman C,
Leo-Messina J, Rapapport D, Wells K, DeConti R, Moscinski L, Stankard C, Puleo
C, Reintgen DS. Detection of submicroscopic lymph node metastases with poly-
merase chain reaction in patients with malignant melanoma. Annals Surg 1994;
220(6):768-774.
6. Morton DL, WEN DR, Wong J et al. Technical details of intraoperative lym-
phatic mapping for early stage melanoma. Arch Surg 1992; 127:392-399.
7. Kahn HJ, Marks A, Thom H, Baumal R. Role of antibody to S100 protein in
diagnostic pathology. Am J Clin Pathol 1983; 79:341.
8. Gaynor RB, Herschman HR, Irie R et al. S-100 protein: A marker for malignant
melanoma. J Clin Pathol 1985; 38:7-15.
9. Fernando SS, Johnson S, Bate J. Immunohistochemical analysis of cutaneous
malignant melanoma: comparison of S-100 protein, HMB-45 monoclonal anti-
body and NKI/C3 monoclonal antibody. Pathology 1994; 26(1):16-19.
10. Gershenwold J, Thompson W, Mansfield P et al. Patterns of failure in melanoma
patients after successful mapping and negative sentinel node biopsy. Ann Surg,
in press.
11. Colome-Grimmer MI, Gershenwald J, Ross MI et al. The negative sentinel lymph
node in Stage I-II melanoma patients. Retrospective study of 20 cases after re-
currence and of 41 cases prospectively. Abstract. J Cutan Pathol 1997; 24(2):91.
12. Carson KF, Wen DR, Li P, Lana A, Bailly C, Morton DL, Cochran AJ. Nodal Nevi
and Cutaneous Melanomas Am J Surg Pathol 1996; 20(7):834-840.
13. Bautista NC, Cohan S, Anders KH. Benign melanocytic nevus cells in axillary
lymph nodes. A prospective incidence and immunohistochemical study with lit-
erature review. Am J Clin Pathol 1994; 102:102.
14. Yazdi HM. Nevus cell aggregates associated with lymph nodes. Immunohis-
tochemical Observations. Arch Pathol Lab Med 1985; 102:1044.
15. Heller R, King B, Baekey P, Cruse W, Reintgen D. Identification of submicro-
scopic lymph node metastases in patients with malignant melanoma. Semin Surg
Oncol 1993; 9:285.
16. Shivers S, Wang X, Cruse W, Fenske N, DeConti R, Messina J, Glass LF, Berman
C, Reintgen DS. Molecular Staging of Melanoma, JAMA, submitted.
17. Cochran AJ, Wen DR, Morton DL. Occult tumor cells in lymph nodes of patients
with pathological stage I malignant melanoma. Am J Surg Path 1988; 12:612.
10

Radioguided Parathyroidectomy
(MIRP)
James Norman
Patient Selection and Anticipated Results .......................................................... 127

Optimal Timing B etween Sestamib i and Operative Explo ration ....................... 127
Use o f the Gamma Probe in the O perating Room .............................................. 130
Summary ............................................................................................................ 132
Primary hyperparathyroidism (HPTH) is the result of a single adenoma in

87-92% of all cases and is cured long-term following the removal of this one gland.1
Despite this fact, most surgeons continue to perform a complete bilateral neck
exploration for patients with primary HPTH. The purpose of this comprehensive
dissection is to identify and even biopsy each gland so all hyperfunctional par-
athyroid tissue is identified and resected while normal glands are left behind. This
conviction is a historical reflection of the inability of preoperative testing to accu-
rately distinguish those patients harboring a single diseased gland from the 8 to
13% with multiple adenomas or four gland hyperplasia.
The advent of the sestamibi scan in the early 1990s has changed the manage-
ment of primary HPTH for many surgeons. A recent meta-analysis of more than
11 6000 patients has shown preoperative sestamibi scanning to have a sensitivity of
approximately 90% and a specificity approaching 100% in identifying patients
with a single adenoma.1 This suggests that the vast majority of all parathyroid
adenomas will be identifiable using this technique. More important is the fact
that when a single adenoma is identified, the scan is almost always correct.
Using the technique of sestamibi scanning followed immediately by surgical
exploration, we have shown that the radioactive gland can be rapidly identified by
using a miniature gamma detection device intraoperatively to guide the dissec-
tion.15 This technique allows a smaller more directed approach to the adenoma
which is easily completed using local anesthesia in a true outpatient setting. The
advantages of this technique are greater than they appear on the surface. The smaller
operative field dictates that a smaller incision can be used which has a tremen-
dous influence on patients and referring physicians alike. The use of local anes-
thesia also has a positive influence on the patient’s view of surgery preoperatively,
but it also more readily allows for them to be sent home almost immediately after
the procedure. Use of the probe in the operating room allows the surgeon to pro-

The Technique of Minimally Invasive Radioguided Parathyroidectomy (MIRP) 127
ceed with confidence toward the adenoma allowing him/her to close the wound
and call for the next patient without waiting for frozen section histologic diagno-
sis or searching for other glands. The cost savings of this approach has been exam-
ined by our group previously, noting that it becomes cost-effective to perform
sestamibi scans on all patients when more than 50% of patients can undergo a
unilateral neck exploration (without radioguidance). When patients can undergo
a MIRP, this break-even point decreases to only 35%. Our experience shows that
approximately 84% of all patients with sporadic primary HPT can have a MIRP.
PATIENT SELECTION AND ANTICIPATED RESULTS
Once the diagnosis of primary HPTH has been confirmed, patients are coun-
seled regarding the standard versus minimally invasive approach. We do not per-
form a sestamibi scan or any other localizing studies prior to the day of the opera-
tion. Patients are scheduled for an operation and the operative technique used
(standard vs MIRP) will be dictated by the results of the sestamibi scan an hour or
two prior to surgery. Because of the critical nature of timing (discussed below),
we suggest that surgeons learning this technique begin by choosing their patients
for MIRP through the use of preoperative scanning a week or two prior to the
operation. Then, on the morning of surgery, a second sestamibi scan is performed
such that the patient is in the operating room about 2-2.5 hours after the reinjec-
tion of the radiopharmaceutical. Once the nuclear medicine department, the sur-
geon, and the entire OR team realize the critical nature of timing, then the screen-
ing sestamibi scans should be eliminated. The goal for surgeons using this tech-
nique is to scan patients only once—1-2 hours prior to surgery.
Since not all patients will demonstrate a single adenoma on sestamibi (89% 11
have an adenoma and the typical sestamibi sensitivity in the literature is 90%),
they are informed that some will not be candidates for minimal resection. Fig-
ure 11.1 details the selection of all patients with primary HPTH for MIRP versus
standard bilateral exploration. This diagram is given to each patient when coun-
seling them preoperatively and explained in detail. If a single adenoma is found
on the preoperative scan, then a MIRP is performed. If no localization occurs, a
standard bilateral exploration will be performed at that time. Because of the dra-
matically increased referral pattern associated with the use of this minimally inva-
sive approach, an occasional patient will elect not to undergo a standard explora-
tion if a single adenoma fails to visualize (the morbidly obese, extremes of age
with confounding medical problems).
OPTIMAL TIMING BETWEEN SESTAMIBI

AND OPERATIVE EXPLORATION
There are two critical determinants of successful MIRP. The first is the quality
of the preoperative sestamibi which allows the surgeon to operate with conviction
Fig. 11.1. Selection of operative procedure for patients with primary hyperparathyroidism for MIRP. Fol-
lowing clinical conformation of primary HPTH, the patient is scheduled for the operating room with a
sestamibi scan to precede the OR by 3 hours. The majority of patients will localize to a single adenoma
and are candidates for out-patient minimally invasive resection under local anesthesia. Those who do not
11 localize on scanning are explored at that time using standard techniques. No other localizing studies are
performed prior to the morning of surgery.
on a single area of the neck without the need for identification of other “normal”
glands while making the chances of persistent disease due to a missed adenoma
extremely low. The second critical determinant for success became obvious only
after we had accumulated sufficient experience with this technique to realize that
there was a “window” of optimal timing between injection of the radiopharma-
ceutical and using the probe in the operating room.
Several factors dictate the optimal time between the sestamibi scan and intra-
operative nuclear mapping/resection. The first consideration is the half-life of the
radiopharmaceutical which is approximately 6.0 hours. This dictates that the two
procedures must be carried out on the same day. The most important consider-
ation, however, is the relative speeds at which the thyroid and parathyroid “wash
out” their nuclear tag. Since the thyroid will lose its initial uptake of sestamibi-
Tc99 at a faster rate than a hyperactive parathyroid gland, the optimal situation
occurs when the thyroid has washed out and the parathyroid remains relatively
radioactive. Only when there is differential activity between the thyroid and the
Fig. 11.2. Operative Window for MIRP. The optimal timing for performing radioguided parathyroidec-
tomy is determined by several factors but none is more important than the length of time which has
expired since the patient was injected with radiopharmaceutical. The gamma probe will be most accurate
at a time when the thyroid has lost more of its radioactivity than the parathyroid. Once beyond 3.5 hours,
the radioactivity in the parathyroid has washed out to such a degree that it is difficult for the probe to
distinguish it from background. 11
parathyroid adenoma can the gamma probe be helpful. We have found that this
situation occurs within a window between 1.5 and 3.0 hours. Typically, 1.5-2.5
hours is ideal for the vast majority of patients and we discourage an elapsed time
over 3.5 hours. We have had the occasion to try this technique several times be-
tween 4.0-6.5 hours after technetium injection. In this setting, no differential ra-
dioactivity could be found and a true MIRP was not possible. We typically sched-
ule the sestamibi injection for 07:30 with the operating room scheduled for 09:30.
If a second case will follow, the injection should follow the first case by about
1.0-1.5 hours. The details of how we perform sestamibi scanning prior to MIRP
can be found on-line at www.EndocrineWeb.com using the keyword sestamibi.
We believe that the risk of recurrent nerve injury is not increased and may
even be less for the MIRP procedure. Dissection around the ipsilateral nerve is
limited because the adenoma is localized with the gamma probe. The contralat-
eral side is obviously without risk. We have not had this complication in our sev-
eral hundred case experience, but inform our patients that the expected incidence
is comparable to standard exploration at about 1%.
USE OF THE GAMMA PROBE IN THE OPERATING ROOM
When positioned on the operating room table, a small (9 to 14 mm) gamma

probe is used to measure radioactivity in four quadrants of the neck defined by
the upper and lower poles of the thyroid on each side. Note that the first generation
of probes which were large, unshielded, and noncolumnated are not very useful for
this operation. The new small probes are required for successful implementation
of this technique. There is typically a difference of approximately 100-400 counts
per second (varies with dosage, time after injection, size of the patient, depth of
the adenoma from skin surface, and the size/design of the probe) overlying the
adenoma, although this often is not appreciated for a deep lesion in a thick neck.
If this difference is seen, the case should go very quickly because of the large dif-
ferential radioactivity which will become apparent as the probe is placed deeper
in the neck. If a single source was seen on sestamibi, cases should still proceed
even if a significant difference cannot be detected at the skin level since the radio-
activity will increase dramatically as the dissection nears the radioactive source.
The skin and subcutaneous tissues are infiltrated with local anesthesia and the
patient is given IV sedation (we prefer Propofol™). The initial incision is placed
according to the expected location of the adenoma as determined by both sestamibi
scanning and measurement of gamma emissions on the skin. This will necessitate
that the incision is slightly higher or lower than usual, but all should be oriented
transversely to allow extension as needed, or even conversion to bilateral explora-
tion if necessary. Superficial adenomas (at the level of the thyroid lobe) can be
removed through a 2.0-2.5 cm incision. Those adenomas lying in the tracheo-
esophageal groove, however, usually require a 3 cm incision.
Sub-platysma flaps are created 1.0-1.5 cm in all directions and held open with
11 a small self-retaining retractor. Radioactivity is again quantitated in all four quad-
rants. The strap muscles are now separated along the midline and another self-
retaining retractor is placed at 90° to the original to hold these muscles apart. The
dissection is carried deeper as directed by increasing gamma counts to locate the
radioactive gland. Beyond this point, blunt dissection should be used exclusively
to prevent damage to small vascular or nervous structures. Cautery below the
strap muscles can usually be avoided completely, but when needed, any deep cau-
tery should be of the bipolar type. The recurrent laryngeal nerve is examined if
the radioguided dissection includes it in the operative field. We do not make a
specific point of locating the nerve during every case, but are constantly aware of
its anatomical relationships.
The adenoma is located by continued and frequent use of the probe to direct
the dissection. When placing the probe deep in the neck, it must be remembered
not to aim it directly at the heart as the sestamibi-Tc99 is also used as a cardiac
imaging agent and will give false-positive readings. Once identified, the adenoma
is teased from its surrounding tissues bluntly and elevated to reveal its single pedicle
which is ligated with a hemoclip and transected. A drain is not needed. At no time
should safety be compromised by a hesitancy to extend the incision or even
conversion to general anesthesia if necessary. When using the probe to its full
potential there will be very little unnecessary dissection. Our experience has shown
that as many as 35% of all parathyroid adenomas can be removed without even
exposing the thyroid gland.
We routinely send MIRP patients home within an hour or two of the proce-
dure. Those patients with significant underlying medical problems are kept over-
night, but this has been necessary only 9 times in over 200 cases. Advanced age
alone is not reason enough to preclude an outpatient approach if all has gone
well.
There are several important points following removal of the adenoma which
combine to acknowledge it as the anticipated source of radioactivity.2,3 The first is
that the excised adenoma emits radioactivity at least 20%, and occasionally higher
than 50% of postexcision background. Fat, lymph nodes, and even thyroid nod-
ules will never show this level of radioactivity (typical ex vivo emissions less than
3% of background). Ex vivo radioactivity has proved to be 100% accurate in dis-
tinguishing parathyroid tissue from fat and lymph nodes when the excised tissue
emits 20% of postexcision background radioactivity. This has reduced the num-
ber of “diagnostic” frozen sections dramatically with less than 5% of all adenomas
falling below this threshold radioactivity level and thereby necessitating a frozen
section. We would not recommend eliminating frozen sections until the surgeon
has gained significant experience in using the probe in this regard. Because of the
systemic administration of the radiopharmaceutical, ex vivo counts must be per-
formed several feet from the patient with the probe aimed away. Operations per-
formed more than 3 hours after sestamibi injection will result in a less radioactive
adenoma so this must be kept in mind when measuring ex vivo radioactivity. If in
doubt, a frozen section should be performed.
The second important observation is that removal of the radioactive gland 11
will be associated with dramatically decreased gamma emissions within that quad-
rant of the neck. The loss of this main focus of radioactivity within the neck gives
rise to the third observation: the radioactivity in all four “quadrants” of the neck
will equalize.2,3
The use of the probe allows the resection to be performed rather quickly. Ad-
enomas are identified an average of 11 minutes after incision. We have had five
patients with completely intrathyroidal parathyroids each of which were found
within 30 minutes of making the skin incision. Their hidden position was dis-
closed by a decline in radioactivity behind the thyroid and the demonstration of
emissions several thousand per-second higher in one specific portion of the thy-
roid compared to the remainder of the gland. Our average operative time for ad-
enoma resection is now under 30 minutes. Although the speed at which an opera-
tion is performed is not important in itself, we believe these times are a reflection
of the simplicity of this technique.
Detailed monitoring of the potential radiation hazards has shown this proce-
dure to pose no significant risk to operating room personnel, surgeon, or patholo-
gist.2 The surgeon’s exposure is relatively insignificant and we have reported that
the cumulative radiation dose acquired over 15 cases is less than 1% of acceptable
yearly exposure (5 rem) as determined by the Nuclear Regulatory Commission.
(Radiation safety issues are discussed more fully in another chapter). Similarly,
the radioactive adenoma sent to pathology contains only slightly more radioac-
tivity than background (0.04 mR/hr) and therefore poses no exposure hazard to
frozen section personnel and does not contaminate the cryostat or other instru-
ments/liquids. The soiled linens and sponges do not require special handling and
can be discarded as routine.
SUMMARY
When selected appropriately, a majority of patients with primary HPTH may

be successfully treated through a minimally invasive technique. The use of local
anesthesia and the limited scope of the dissection afforded by intraoperative nuclear
mapping may decrease the incidence of failed explorations and other potential
complications associated with this operation. Although intraoperative nuclear
mapping can be a very useful tool to minimize the efforts necessary to localize a
diseased parathyroid, prior experience in parathyroid surgery is still mandatory
since clinical judgment will continue to play the dominant role in determining
when the operation is complete and when a bilateral exploration is required.
REFERENCES
1. Denham D, Norman J. Cost-effectiveness of preoperative sestamibi scan for pri-
mary hyperparathyroidism is dependent solely upon surgeon’s choice of opera-
tive procedure. J Am Coll Surg 1998; 186:293-304.
2. Norman J, Chheda H. Minimally invasive parathyroidectomy facilitated by in-
11 traoperative nuclear mapping. Surgery 1997; 122:998-1004.
3. Norman J, Denham D. Minimally invasive radioguided parathyroidectomy in
the reoperative neck. Surgery 1998; in press.
4. Norman J. The technique of intraoperative nuclear mapping to facilitate mini-
mally invasive parathyroidectomy. Cancer Control, 1997; 4:500-504.
Emerging Applications in Other Skin Cancers 133
Emerging Applications
in Other Skin Cancers
C. Wayne Cruse
Introduction ........................................................................................................ 133

Squamous C ell Carcino ma ................................................................................. 133
Merkel Cell Carcino ma (MCC) .......................................................................... 137
Technical Considerations .................................................................................... 139
Summary ............................................................................................................ 140
INTRODUCTION
The use of lymphatic mapping and sentinel lymphadenectomy has at many

institutions profoundly changed the management of patients with malignant mela-
noma. The number of patients with melanoma is increasing and the death rate for
patients with non-thin lesions is significant. This technique allows identification
of regional nodal disease at an early stage.
Non-melanoma skin cancer far surpasses the incidence of malignant mela-
noma, and is the leading cause of cancer in the United States. More than 600,000
cases occur annually. Basal cell carcinoma comprises the majority of these and
only very rarely involves the regional lymph nodes. Squamous cell carcinoma (SCC)
comprises approximately 20% and metastases to regional lymph nodes gravely
influence the prognosis. Lymphatic mapping and sentinel lymphadenectomy may
be useful to identify the patients with micrometastatic SCC in the regional lymph 12
nodes. Merkel cell carcinoma (MCC) is an uncommon skin malignancy, has a
poor prognosis, and a significant chance for regional lymph node metastases. These
patients may also benefit from lymphatic mapping and sentinel lymphadenec-
tomy. Overall, any patient with any skin cancer with a significant chance of lym-
phatic spread should be considered for this technique.
SQUAMOUS CELL CARCINOMA
Squamous cell carcinoma accounts for approximately 20% of all skin cancers.
The incidence in the United States is 41.4 cases per 100,000 and is increasing.1 The
most powerful predictor of survival in patients with SCC is the presence or ab-
sence of metastatic disease. Approximately 80% of all metastatic disease initially
occurs in the regional lymph nodes. The overall incidence of regional lymph node
metastases in SCC is relatively small, reported between 0.5% and 5%2 but carries
a poor prognosis. Regional metastatic SCC is associated with a 5-year survival
rate of 34% and a 10-year survival rate of 16%.3
The use of lymphatic mapping and sentinel lymphadenectomy will be benefi-
cial only to the patients at significant risk of lymph node disease. Factors influenc-
ing SCC metastatic rate include histological parameters, anatomic site, prior treat-
ment, diameter, immunosuppression, and etiology.
Histological factors are important in predicting metastatic potential as a poorly
differentiated histological pattern carries a 3-fold chance of metastases over a well-
differentiated pattern. Poorly differentiated lesions comprise only 19.6% of all
cutaneous SCC, but account for 51% of all metastasizing lesions.4 The histologic
classification is also important. Adenoid SCC, a histological variant, exhibits a
metastatic rate up to 67% in extensive lesions, whereas verrucous SCC has a very
low rate of metastases.5
Histological depth of invasion according to Clark has been investigated in SCC.
The risk of recurrence and metastasis increases greatly with invasion to or below
the reticular dermis, which is Clark level IV and V, respectively. Tumor thickness
has been noted to be important in determining metastases. In a review, all cases
that metastasized were at least 4 mm thick and all deaths were in lesions over
10 mm thick.6
The presence of an inflammatory response with a lymphocytic infiltration sur-
rounding the lesion has been noted to diminish the chance for metastases. Histo-
logical evidence of perineural invasion occurs in approximately 5% of patients
and studies have shown an increased metastatic rate and poor prognosis.2
Location of the tumor is important as tumors of the scalp, temple, ears, nos-
trils and extremities have been reported to be especially prone to metastases. The
overall metastatic rate for the ear is 11% and lip is 13.7%.4
Prior treatment and subsequent local recurrence of SCC carries a worse prog-
12 nosis. Recurrent tumors overall have a 25% metastatic rate and recurrent tumors
of the ear carry a 45% metastatic rate.2
Increased size of the primary tumor is associated with a decreased overall sur-
vival and increased rate of regional metastasis. For lesions over 2 cm. in diameter,
the metastatic rate is 23.4%. The overall 5-year survival rate for lesions less than
2 cm is 98% and is 72% for lesions over 2 cm.2
Patients with an altered immune status are also at higher risk for cancer in
general and specifically SCC. Organ transplant patients on immunosuppression
have an increased incidence of SCC which increases with time after transplanta-
tion. Patients with some lymphoproliferative disorders develop SCC that behaves
aggressively.
The etiology influences the frequency of metastases. Carcinomas arising from
chronic lesions have a relatively high metastatic rate, generally more than 20%,
and are associated with a poor prognosis. These include squamous cell carcino-
mas arising in burn scars, radiation ulceration, chronic ulcers, osteomyelitis si-
nuses, traumatic wounds, fistulas, and calluses.7
Patients with palpable lymph nodes are treated with a regional lymph node
resection. Patients without clinically suspicious regional disease are generally treated
without prophylactic regional node resection, but are monitored closely. If re-
gional disease develops clinically later, then it is treated with regional lymph node
resection.
Ideally, the above risk factors could identify SCC patients who would benefit
from SLN biopsy. Unfortunately, no long term studies with large numbers of pa-
tients are available to provide specific information as to which patients would
benefit from lymphatic mapping and selective lymphadenectomy. For selected
patients at-risk we use lymphatic mapping to identify occult micrometastatic re-
gional nodal SCC. If the sentinel lymph node is negative then no further treat-
ment is necessary. Patients with micrometastatic SCC in the SLN undergo comple-
tion node dissection. Some patients may not be candidates for further surgery,
and radiation therapy may be considered for treatment of the micrometastatic
SCC of the lymph nodes. For example, radiation therapy has been used exten-
sively to treat micrometastatic SCC to the lymph nodes of the neck from primary
lesions.
Early experience with SLN biopsy exists in patients wtih SCC of the penis,
even predating SLN techniques for melanoma or breast cancer. The “sentinel”
lymph node was identified by preoperative lymphangiogram of the penis.8 Penile
cancer spreads in a predictable pattern with inguinal nodes as the primary site of
metastases, and bilateral involvement is not uncommon.
If invasion of the shaft or corpora cavernosa occurs, then 33% of patients with
clinically negative nodes develop inguinal metastases.9 The pathological status of
the regional lymph nodes is an independent prognostic factor in predicting sur-
vival. In addition, the extent of nodal involvement is itself of prognostic impor-
tance. Solitary or unilateral nodal metastases are associated with improved sur-
vival, relative to bilateral or extensive regional disease.10 The identification of pa-
tients with micrometastatic squamous cell carcinoma in the sentinel lymph node 12
and subsequent early treatment of the inguinal nodal basin may improve survival.
Because the lymphatic drainage of the penis is ambiguous, preoperative
lymphoscintigraphy is essential to identify the site(s) of lymph nodes at highest
risk for metastases. Intraoperative SLN mapping proceeds as with melanoma,
avoiding the high morbidity associated with inguinal node dissection.
Lymphatic mapping and sentinel lymphadenectomy has been successfully used
to identify micrometastatic regional squamous cell carcinoma in a patient with an
extensive squamous cell carcinoma of the upper extremity. The micrometastatic
regional disease was subsequently treated with regional node dissection and the
patient had no regional recurrence (Figs. 12.1 and 12.2).11
Squamous cell carcinoma of the lip is the most common malignancy of the
oral cavity. The incidence of lymphatic involvement is related to the size and loca-
tion of the primary and to a poor histological grade of differentiation. The cure
rate is overall very good, but regional metastases occurs in 2-15% of patients and
the 5-year survival rate is dramatically decreased with regional spread of the cancer.
Fig.12.1. 72 year old female with 4 x 3 cm SCC of left wrist involving radius
12
Fig. 12.2. Preoperative radiolymphoscintigram demonstrating two sentinel lymph nodes.
Some have advocated bilateral suprahyoid neck dissections for at-risk patients,
then proceed with a complete neck dissection if positive.12,13
In patients with lesions over 2 cm, poorly differentiated histologically, and lo-
cated on the upper lip or at the commissures, the chance for metastases is signifi-
cant. These patients may benefit from lymphatic mapping and selective lym-
phadenectomy. In the excision and reconstruction of large lip lesions, extensive
face and neck incisions and dissection is necessary, and any lymph nodes located
by the mapping techniques could be harvested through incisions used for the ex-
cision and reconstruction. The patients that have micrometastatic disease would
therefore be identified earlier in the course of disease. The death of patients with
squamous cell carcinoma of the lip is usually from uncontrolled local and re-
gional disease and earlier treatment may improve survival.
MERKEL CELL CARCINOMA (MCC)
Merkel cell carcinoma of the skin is an uncommon skin malignancy derived

from primitive epithelial cells. The cell was first described by Merkel in 187514 and
the tumor was first described by Toker in 1972.15 It has been called neuroendo-
crine carcinoma of the skin, cutaneous APUDoma, and anaplastic carcinoma of
the skin. The Merkel cell is located near the basal layer of the epidermis and func-
tions as a receptor of mechanical stimuli. Clusters of cells form sensitive mechan-
oreceptors and are found throughout the skin.
Since 1972, over 600 cases have been reported. Patients are predominately over
65, and there are more males than females. The tumor frequently arises in sun-
exposed areas such as the head and neck. It begins usually as a smooth dermal
nodule, and may become ulcerated. Merkel cell carcinoma is thought by some to
be increasing in frequency.
The tumor carries an overall poor prognosis. Local recurrence rates are re-
ported between 26-44%. Regional nodal metastatic disease is present in 50-75%
of patients sometime in the course of the disease. Between 7-31% of patients present
with clinically evident regional lymph node disease, particularly tumors of the
head and neck. Systemic disease is usually preceded by regional lymph node
disease.16
There is little data on long-term follow-up with adequate numbers of patients.
Some studies have shown a 50% regional recurrence rate at 2 years and between
30-55% long term survival rate.
Treatment of the primary tumor consists generally of surgical excision with a
2-3 cm margin. For patients with clinically suspicious regional lymph node dis- 12
ease, a therapeutic lymph node dissection is recommended. In patients with re-
gional metastases, the prognoses is poor.
For patients without clinically suspicious regional disease, the role of prophy-
lactic lymph node resection remains controversial. There are no reliable primary
tumor characteristics to predict lymph node metastatic disease. The most impor-
tant predictor of survival has been the presence of lymph node metastases, and
because of the poor prognosis, some authors have recommended prophylactic
lymph dissection in all patients.17
A hypothetical alternative to prophylactic node dissection is selective lym-
phadenectomy for all patients with Merkel cell carcinoma.18 Routine application
of SLN mapping to this patient group eliminates the morbidity of the more ex-
tensive surgical procedure, while identifying those patients with regional metastases
who may benefit from complete node dissection.
We have reported lymphatic mapping and selective lymphadenectomy in 12
patients with Merkel cell carcinoma. 19 Two patients were found to have
Fig. 12.3. 69 year old female with 4 x
5 cm MCC of nasal dorsum.
12
Fig. 12.4. Preoperative radiolymphoscintigram demonstrating sentinel lymph nodes in

bilateral submandibular areas.
micrometastatic disease in the sentinel nodes, and complete dissection of the in-
volved nodal basin revealed additional positive nodes. Neither patient has devel-
oped regional recurrence. For patients with negative sentinel nodes, no patient
developed regional disease at a median follow-up time of 10.5 months. Figures 12.3
and 12.4 illustrate a patient with Merkel cell carcinoma of the nose treated with
lymphatic mapping and selective lymphadenectomy. Bilateral cervical sentinel
nodes were positive, she subsequently underwent bilateral modified radical neck
dissections and is now disease-free at 2 years.20
Lymphatic mapping and selective lymphadenectomy has been used in only a
few patients with skin cancer other than malignant melanoma. However the ana-
tomical and physiological parameters that make the technique effective in mela-
noma are present for all skin cancers. No long term data is available in large num-
bers of patients with other skin cancers, but patients at high-risk for developing
regional lymph node disease may benefit from the procedure. The morbidity is
relatively low and much benefit may be obtained.
TECHNICAL CONSIDERATIONS
The techniques of lymphatic mapping and selective lymphadenectomy for skin

cancers other than malignant melanoma are much the same as those used for
malignant melanoma. Preoperative lymphoscintigraphy and intraoperative map-
ping and biopsy are performed as described in multiple references and elsewhere
in this handbook.21-23
The head and neck is the predominate site for Merkel cell carcinoma and squa-
mous cell carcinoma. When performing lymphatic mapping in the head and neck,
a few specific points should be noted. All incisions for lymphatic mapping and
selective lymphadenectomy should be made with considerations for the possibil-
ity of subsequent radical neck surgery. Upper neck incisions should be placed 12
transversely to correspond with a later upper neck transverse incision for radical
neck dissection. Preauricular incisions can be used liberally to expose the parotid
gland and lymph nodes in the area. The sentinel lymph node(s) are found in the
usual Level I-V locations and much care must be taken to preserve the many neck
structures at risk. Small initial incisions may need to be extended for adequate
exposure if difficulty arises in finding the sentinel lymph node(s).
In the head and neck, it is often necessary to excise the primary tumor site in
order to perform lymphatic mapping. The radioactivity from the primary tumor
site may be very close to the lymph node basin and interfere with intraoperative
localization with the hand-held gamma probe. Removing the primary tumor site
eliminates this proximity problem.
Primary tumor site excision may require rotation flaps or other reconstructive
coverage procedures. Planning for the reconstruction of the primary tumor site
must be done prior to making an incision for lymphatic mapping, as a poorly
placed incision may interfere with subsequent reconstruction needs.
The preoperative lymphoscintigram should be available for examination dur-
ing surgery. Often the preoperative lymphoscintigram will identify one predomi-
nate sentinel lymph node and the overlying skin will be tattooed, but additional
lymph nodes may show uptake also. These secondary lymph nodes may also be
identified at surgery and removed.
In conclusion, SLN mapping for non-melanoma skin cancers is performed
with near identical techniques as for melanoma cases, with additional require-
ments as a consequence of the predominately head and neck location of these
lesions.
SUMMARY
The use of lymphatic mapping and selective lymphadenectomy in patients with

skin cancers other than malignant melanoma needs to be validated with larger
groups of patients. However, preliminary data suggest that selective lymphadenec-
tomy may be beneficial in those patients with squamous cell carcinoma at higher
risk for metastatic disease, based on published histopathologic variables. Sentinel
node mapping is also likely to prove useful in all cases of Merkel cell carcinoma,
because of the high reported rate of regional metastases in these patients. As with
other malignancies, the application of radioguided surgical techniques allows pa-
tients with regional metastatic disease to be identified without the morbidity of
prophylactic node dissections, reserving node dissection only for those patients
with proven regional metastases.
REFERENCES
1. Silverberg E, Boring CC, Squires TS. Cancer statistics, 1990. CA 1990:40:9-26.
2. Johnson TM, Rowe DE, Nelson BR et al. Squamous cell carcinoma of the skin
excluding lip and oral mucosa). J Am Acad Dermatol 1992; 26:467-84.
12 3. Epstein E, Epstein NN, Bragg K et al. Metastases from squamous cell carcinoma
of the skin. Arch Dermatol 1968; 97:245-9.
4. Rowe DE, Carroll RJ, Day CL. Prognostic factors for local recurrence, metastasis,
and survival rates in squamous cell carcinoma of the skin, era, and lip. J Am Acad
Dermatol 1992; 26:976-90.
5. Johnson WC, Helwig EB. Adenoid squamous cell carcinoma (adenocanthoma).
A clinicopathologic study of 155 patients. Cancer 1966; 19:1639-150.
6. Friedman H, Friedman HI, Cooper PH et al. Prognostic and therapeutic use of
microstaging of cutaneous squamous cell carcinoma of the trunk and extremi-
ties. Cancer 1985; 56:1099-105.
7. Kwa RE, Campana K, Moy RL. Continuing medical education biology of cutane-
ous squamous cell carcinoma. J Amer Acad Derm 1992; 26:1-26.
8. Cabanas RM. An approach for the treatment of penile carcinoma. Cancer. 1997;
2:456-466.
9. Persky L, deKernion J. Carcinoma of the Penis. CA–A Cancer Journal for Clini-
cians 1986; 36:258-273.
10. Gillenwater JY, Grayhack JT, Howards SS et al (eds). Adult and Pediatric Urology
3rd Edition, St. Louis: Mosby, 1996:2002-2042.
11. Stadelmann WK, Javaheri S, Cruse CW et al. The use of selective lymphadenec-
tomy in squamous cell carcinoma of the wrist: A case report. The Journal of
Hand Surgery. 1997; 22A:726-731.
12. Marshall KA, Edgerton MT. Indications for Neck Dissection in Carcinoma of the
Lip. Amer J Surg 1977; 133:216-217.
13. Koc C, Akyol MU, Cekic A et al. Role of suprahyoid neck dissection in the treat-
ment of squamous cell carcinoma of the lower lip. Ann Otol Rhino Laryngol
1997; 106:787-789.
14. Merkel F. Tastazellen und Tastkorperchen bei den Hausthieren und beim
Menschen. Arch Mikr Anat 1875; 11:635-52.
15. Toker C. Trabecular carcinoma of the skin. Arch Dermatol 1972; 105:107-110.
16. Haag ML, Glass LF, Fenske NA. Merkel cell carcinoma diagnosis and treatment.
Dermatol Surg. 1995; 21:669-683.
17. Victor NS, Morton B, Smith JW. Merkel cell cancer: Is prophylactic lymph node
dissection indicated? The American Surgeon 1996; 62:870-882.
18. Cruse CW, Reintgen D, Glass F et al. Neuroendocrine carcinoma of the skin.
Cancer Control 1997; 4:346-348.
19. Messina JL, Reintgen D, Cruse CW et al. Selective lymphadenectomy in patients
with merkel cell (cutaneous neuroendocrine) carcinoma. Annals of Surgical
Oncology. 4(5):389-395.
20. Javaheri S, Cruse CW, Stadelmann WK. Sentinel node excision for the diagnosis
of metastatic neuroendocrine carcinoma of the skin: A case report. Annals of
Plastic Surgery 1997; 39(3):299-302.
21. Morton DL, Wen D, Wong JH et al. Technical details of intraoperative lymphatic
mapping for early stage melanoma. Arch Surg 1992; 127:392-399.
22. Reintgen DS, Rapaport DP, Tanabe KK, Ross M. Lymphatic mapping and senti-
nel lymphadenectomy. In: Balch CM, Houghton AN, Sober AJ, Soong S eds. Cu-
taneous Melanoma 3rd Edition. St. Louis: Quality Medical Publishing Inc.
1998:227-244.
23. Norman J, Wells K, Kearney R et al. Identification of lymphatic basins in patients
with cutaneous melanoma. Semin Surg Oncol 1993; 9:224-7.
12
Radioguided Surgery
and Vulvar Carcinoma
Edward C. Grendys, Jr., James V. Fiorica
Introduction ........................................................................................................ 142

Epidemiolo gy ...................................................................................................... 143
Anatomy and Lymphatic Drainage .................................................................... 144
Surgical Management and Staging o f Vulvar Carcino ma .................................. 144
Intraoperative Lymphatic Mapping f or Vulvar Carcino ma ............................... 147
Technique of Vulvar Lymphoscint igraphy .......................................................... 147
Vulvar Melanoma ............................................................................................... 148
Conclusion and the Future ................................................................................. 148
INTRODUCTION
Overall, carcinoma of the vulva is a relatively rare malignancy representing

approximately 5% of all female genital tract cancers1 and 1% of all malignancies
in women. The incidence has risen slightly over the past 50 years, perhaps because
of an overall rise in the average age of women in the U.S.,2 or an increase in the
prevalence of human papillomavirus (HPV) genital tract infection. The reported
increase in vulvar carcinoma in situ, a precursor lesion, has nearly doubled from
1.1-2.1 per 100,000 woman–years between 1973 and 1987,3,4 mostly in younger
women.
Fortunately, the majority of lesions are diagnosed at a relatively early stage and
are thus amenable to local surgical extirpation with high overall cure rates. Gener-
ally, these tumors are readily visible on the external vulvar surface and produce
13 the typical symptom of pruritus in over 90% of patients.5 Other signs and symp-
toms associated with vulvar carcinoma include a visible, palpable vulvar mass,
pain, bleeding, ulceration, dysuria and/or an abnormal vaginal, perineal discharge.
Unfortunately, even given the high prevalence of early symptoms, patient reporting
and subsequent diagnostic evaluation is often delayed, most commonly due to
patient denial or embarrassment. A histologic diagnosis should always be consid-
ered before initiating any topical therapy for women with these complaints.
Vulvar malignancies with the rare exception of vulvar sarcomas appear most
commonly in women between the ages of 65 and 75 while the median age of the
precursor in situ lesion is approximately 45-50. However, 15% of vulvar carcino-
mas occur in women less than 40 years of age6 with multifocal disease being more
prevalent.
Radioguided Surgery and Vulvar Carcinoma 143
The vulva is covered by keratinized squamous epithelium and the majority of
these neoplasms are of squamous cell (epidermoid) histology (see Table 13.1).
Though relatively uncommon, vulvar melanoma represents about 5-10% of ma-
lignant vulvar neoplasms and overall about 17% of melanomas diagnosed in fe-
males will originate on the vulva. The aggressive biopsy of hyperpigmented vulvar
lesions using standard criteria should be considered (See Table 13.2).
EPIDEMIOLOGY
Several infectious agents have been proposed as possible etiologic factors in

the development of vulvar carcinoma, including various granulomatous infec-
tions, herpes simplex virus and human papillomaviruses (HPV). The molecular
role of the human papillomavirus in cervical dysplasia and carcinoma as well as
the association between viral induced vulvar condylomata and subsequent devel-
opment of vulvar carcinoma has been well established.7 Many investigators have
identified HPV DNA in both invasive and precursor carcinoma in situ vulvar le-
sions.8 With the increasing incidence of HPV related dysplasias in the younger
female population and the reported trend toward a younger age of diagnosis9 there
is certainly cause for concern that an increase in related vulvar cancer may be
forthcoming. In a study by Mitchell et al10 169 women with invasive vulvar carci-
nomas were noted to have a secondary genital squamous neoplasm in 13% of
Table 13.1.Histologic distribution of malignant vulvar neoplasms

Tumor type Percent
Epidermoid 86.2
Melanoma 4.8
Sarcoma 2.2
Basal cell 1.4
Bartholin gland 1.2 13
Squamous 0.4
Adenocarcinoma 0.6
Undifferentiated 3.9
Plentl, AA, Friedman, EA, Lymphatic system of the Female Genitalia Philadelphia, 1971,
WB Saunders
Table 13.2. Indications for excisional biopsy of vulvar nevi
Change in surface area of nevus

Change in lesion contour or surface
Change in lesion color
Change in sensation
cases. Brinton et al11 concluded that women with a history of genital warts, previ-
ous abnormal Papanicolaou smears as well as a history of smoking are at increased
risk for vulvar cancer. Chronic immunosuppression has also been linked to an
increased incidence of both vulvar and cervical disease.9 Other observational as-
sociations have been made between hypertension, diabetes mellitus, and obesity.12
ANATOMY AND LYMPHATIC DRAINAGE
The vulva, including the mons pubis, labia majora and minora, clitoris, vagi-
nal vestibule, perineal body and the supporting subcutaneous structures, devel-
ops embryologically from the genital tubercle of the cloacal membrane, as does
the distal vagina. Because of their embryologic derivation the vulva and distal
vagina share common routes of lymphatic drainage. The vulvar lymphatics run
anteriorly through the labia majora, turn laterally at the mons pubis and drain
primarily into the superficial inguinal lymph nodes. Previous lymphatic dye map-
ping studies by Parry-Jones13 demonstrated that the vulvar lymphatic channels do
not cross to the contralateral lymph nodes unless the dye is injected in the midline
structures (clitoris or perineal body). There may be some minimal, direct drain-
age to the pelvic lymph nodes though the clinical relevance of these communica-
tions appears negligible. Most studies indicate an orderly progression of lymphatic
drainage from the vulva to the superficial inguinal lymph nodes then directly to
the deep inguinal nodes prior to proceeding to the pelvic nodal structures (See
Fig. 13.1).
SURGICAL MANAGEMENT AND STAGING OF VULVAR CARCINOMA
Vulvar cancers have three modes of spread: 1) direct extension into adjacent
organs; 2) embolization into locoregional lymph nodes; and 3) hematogenous
spread. Fortunately well documented clinical investigations have revealed that early
13 spread is almost always confined to the local inguinal lymphatic basins. Although
lymphatic drainage usually proceeds from the superficial to the deep inguinal
(femoral) lymph nodes, care must be taken before adopting too cavalier an ap-
proach in that deep nodal involvement has been reported without evidence of
superficial inguinal lymph node disease.15-17
Few radical procedures in gynecologic oncology have been as successful and
yet continue to change as much as the surgical approach to vulvar carcinoma.
Original radical procedures as described by Taussig18 and Way19 defined the pro-
cedure that dramatically reduced the mortality from vulvar carcinoma. These tra-
ditional “Longhorn” resections included the entire vulva, mons and laterally ex-
tending over and including the inguinal lymph nodes and inferiorly to the uro-
genital diaphragm. Until recently, this procedure was the treatment of choice for
all resectable vulvar lesions regardless of clinical size or location. The conven-
tional radical vulvectomy and bilateral inguinofemoral lymphadenectomy had been
Fig. 13.1. Lymphatic drainage from
the external genitalia to the inguinal
lymph nodes. Reprinted with permis-
sion from DiSai and Creasman, Clini-
cal Gynecologic Oncology, Copyright
1997; 206-207. © 1997 Mosby-Year
Book Inc.
13
extremely successful from a curative standpoint as has been well documented in

the literature.12,20,21 The overall uncorrected 5-year survival following this proce-
dure is approximately 65% and operability rates in this generally older popula-
tion about 95%.22 Unfortunately postoperative morbidity remained extremely
substantial. Reported operative and perioperative mortality rates in modern se-
ries range from 1-5%.20-22 The incidence of acute wound breakdown approaches
85%22 and often disabling chronic lower extremity edema in 30-70%.23 Genital
prolapse, urinary incontinence, and vaginal stricture occurs in 15-20% of women
postoperatively with Hacker et al reporting a reoperation rate of 15% for correc-
tion of postsurgical complications.20
The radical vulvectomy with bilateral inguinofemoral lymphadenectomy and
often pelvic lymph node sampling has helped define local patterns of lymphatic
dissemination, incidence of nodal metastasis and pathologic features of primary
lesions which help predict the risk of nodal spread. These studies have guided
gynecologic oncologists in redefining this radical procedure. For many years the
addition of a pelvic node dissection to the radical vulvectomy/inguinofemoral
lymphadenectomy was standard practice at many centers.24 Recent studies indi-
cate that fewer than 6% of vulvar cancers have metastatic disease in the pelvic
nodes and essentially no patient without evidence of spread to the groin nodes
had pelvic lymph node metastasis. This discovery has essentially eliminated the
routine use of pelvic lymph node sampling in this situation.
Most gynecologic oncologists have moved away from the classical radical vul-
vectomy with complete en bloc resection through a single incision. Current trends
are moving towards less extensive vulvar surgery with more customization based
on lesion size and location, and separate groin incisions for inguinofemoral lym-
phadenectomy. Based on surgical-pathologic indicators surgical procedures can
be uniquely customized based on lesion size, location, and depth of invasion as
well as clinical suspicion of lymph node metastasis. Especially true of small (< 2 cm)
lateralized lesions with < 1 mm depth of invasion, a wide radical excision with
approximately 10 mm free margins (fresh, nonfixed tissue)25 can be performed
with no need to perform a lymphadenectomy thus sparing the patient the poten-
tial morbidity. The risk of groin node metastasis as related to depth of invasion,
lesion diameter and clinical suspicion and has been well documented by the
Gynecologic Oncology Group (GOG)26 (see Table 13.3) and clearly defines the
rationale for more conservative therapies in the early lesion.
The decision to perform inguinal-femoral lymphadenectomy is based upon
tumor size, location and depth of invasion as well as clinical suspicion of lymph
node metastasis. Recalling that the only patient without risk of nodal spread has
tumor invasion of < 1 mm, the decision to perform groin node dissection is rela-
tively straightforward. In general the prevailing thought is that a small, well later-
alized lesion can be managed with ipsilateral inguinofemoral lymph node dissec-
13 tion unless a metastatic deposit is noted in which case the contralateral lym-
phadenectomy should also be performed. Midline lesion resection should include
bilateral lymphadenectomies. Controversy remains over whether a superficial and
deep groin node dissection should be performed routinely. There is little data to
support the notion that superficial dissection alone decreases resultant lymphe-
dema. Metastases to the femoral nodes without involvement of superficial nodes
has been reported15-17 and the Gynecologic Oncology Group study revealed an
unexpectedly high incidence of ipsilateral groin recurrences following superficial
dissection alone.27 With the high mortality associated with recurrent groin me-
tastasis, the possible therapeutic benefit of groin node dissection and no docu-
mented benefit in limited dissection, our institution continues to perform both
superficial and deep lymphadenectomies in patients not placed on study protocol.
INTRAOPERATIVE LYMPHATIC MAPPING
FOR VULVAR CARCINOMA
In 1979 DiSaia et al28 described the eight to ten inguinal nodes above the crib-
riform fascia as “sentinel nodes.” The concept of sentinel lymph node identifica-
tion using intraoperative mapping with lymphoscintigraphy offers the unique
potential of decreasing the morbidity associated with inguinal lymphadenectomy.
Previous studies have examined lymphatic mapping using lesionally injected tech-
netium99 sulfur colloid in breast carcinoma29,30 as well as cutaneous melanoma.31
These investigators defined the sentinel lymph node as the first node in the lym-
phatic basin that receives primary lymphatic flow from the suspect lesion and
therefore should be the first site of metastatic disease. These and other studies in
breast carcinoma and cutaneous melanoma have revolutionized the surgical treat-
ment of these diseases while dramatically reducing associated morbidity, hospital
stay and subsequent cost.
Given the basic understanding of vulvar lymphatic drainage investigators have
begun to examine the role of selective sentinel lymph node biopsy in early vulvar
carcinomas and melanomas. Levenback et al32 described a method of intradermal
injection of isosulfan blue dye which is a triphenylmethane dye transported via
lymphatic vessels and concentrated in lymph nodes. Subsequent groin dissection
was able to visualize a sentinel node in 7 of 12 samples. In no patient so studied
was a positive nonsentinel node identified in the presence of a negative sentinel
lymph node. A modification of this approach using technetium99 sulfur colloid
lymphoscintigraphy provides a relatively simple method of intraoperative identi-
fication of the sentinel lymph nodes.
TECHNIQUE OF VULVAR LYMPHOSCINTIGRAPHY
After appropriate consent is obtained patients are placed in the dorsal litho-
tomy position in an outpatient setting approximately 2 hours prior to surgery. No
sedation is generally required. The vulvar lesion (or biopsy site) is appropriately 13
identified, prepped and approximately 400 microcuries of technetium99 sulfur col-
loid (reconstituted in approximately 8 ml total) is injected intralesionally in a cir-
cumferential manner using a 25 gauge spinal needle (see Fig. 13.2). Minimal dis-
comfort is to be anticipated and no local analgesia is usually required and in fact
may potentially decrease uptake of the radioactive colloid.
The patient is taken to the operating room about 2 hours after injection, anes-
thetized, positioned in modified lithotomy position to facilitate inguinofemoral
lymph node dissection and prepped and draped in the usual manner. A standard
incision is made approximately 2 cm below and parallel to the inguinal ligament.
The dissection is carried down through the level of the superficial fascia and a flap
is dissected both cephalad and caudad in the usual manner. A intraoperative gamma
counter is next brought onto the field. Using the gamma counter the area of great-
est activity is identified and careful dissection is used to identify the sentinel lymph
Table 13.3. Risk of lymph node metastasis in relation to depth of tumor invasion
Invasion (mm) % Ipsilateral node + %Contralateral only + % Bilateral +
<2 6.8% 0.0 0.0

3-5 20.4 1.9 2.8
6-10 28.8 3.8 11.3
>11 36.7 2.5 50
Homesley H, Prognostic factors for groin node metastasis in squamous cell carcinoma of
the vulva (a Gynecologic Oncology Group Study). Gynecol Oncol 1994; 49:279.
node. After excision it is important to recheck the excised sample to be certain the
activity was precisely within the removed lymph node. In our experience, reex-
amination of the nodal basin can occasionally identify a second or third node
with significant radioactive uptake. These nodes should also be excised and iden-
tified as sentinel lymph nodes. Given the experimental nature of this procedure,
at present we then continue with a complete inguinofemoral lymphadenectomy.
We have published the results of a pilot study of SLN mapping in ten patients
with vulvar cancer. 33 Since this was a pilot study, subsequent complete
inguinofemoral lymphadenectomy was then undertaken. Of the 10 patients stud-
ied, 3 were noted to have metastatic disease all of which were correctly identified
in the sentinel node. No patient with a negative sentinel lymph node was found to
have metastatic spread to other areas (false negative rate of 0%).
VULVAR MELANOMA
The major treatment strategy for vulvar melanoma remains surgical. As with
squamous cell carcinoma there has been a trend away from the complete radical
vulvectomy towards a more conservative wide radical resection. Based on work by
Trimble et al,34 a 2 cm margin with excision deep to the level of the inferior uro-
13 genital diaphragm appears adequate with similar survival. Inguinal lymph node
metastases are rare in patients with Clark’s level I or II melanoma35 and lym-
phadenectomy can most likely be avoided in these patients. Although regional
lymph node dissection in more advanced lesions serves more of a prognostic value
rather than a therapeutic one,36 most investigators still recommend inguinofemoral
lymphadenectomy for lesions greater than Clarks level II.37
CONCLUSION AND THE FUTURE
In general the role of ultra-radical vulvar resection and inguinofemoral lym-

phadenectomy has undergone extensive revision in the past 10-15 years with the
ultimate goal being to decrease physical as well as psychosocial and psychosexual
morbidity. With continuing advances in technology and refinement in surgical
Fig. 13.2a. Posterior squamous cell carcinoma of the vulva.
13
Fig. 13.2b. Insertion of the 25 gauge spinal needle.

Fig. 32.2c. Injection of technetium99 sulfur colloid.
techniques it is hoped that sentinel lymph node biopsy will someday replace the
complete lymphadenectomy as currently practiced. As with other malignancies,
sentinel lymphadenectomy for vulvar carcinoma may refine staging and surgical
therapy, eliminating (or at least reducing) the need for prophylactic lymphadenec-
tomy and converting all lymphadenectomies to therapeutic. However, at present,
we believe that SLN mapping for patients with vulvar cancer should only be per-
formed under a protocol setting, especially since prior GOG studies suggest a thera-
peutic benefit to routine complete inguinofemoral lymphadenectomy.
REFERENCES
13 1. DiSai PH, Creasman WT. Invasive Carcinoma of the Vulva. Clinical Gynecologic
Oncology. St. Louis: Mosby-Year Book, Inc., 1997.
2. Green T. Carcinoma of the vulva; A reassessment. Obstet Gynecol 1978; 52:462.
3. Eifel P, Berek J, Thigpen J. Cancer of the cervix, vagina, and vulva. In: Devita V,
Hellman S, Rosenberg S eds. Cancer: Principles and Practice of Oncology. Phila-
delphia: Lippincott-Raven Publishers 1997.
4. Sturgeon S, Brinton L, Devesa S et al. In situ and invasive vulvar cancer incidence
trends. Am J Obstet Gynecol 1992; 166:1482.
5. Burke T, Eifel P, McGuire W et al. Vulvar Carcinoma. In: Hoskins W, Perez C,
Young R eds. Principles and Practice of Gynecologic Oncology. Philadelphia:
Lippincott-Raven Publishers 1997.
6. Rutledge F, Mitchell M, Munsell M et al. Prognostic indicators for invasive carci-
noma of the vulva. Gynecol Oncol 1991; 47:239.
7. Brinton L, Nasca P, Mallin K et al. Case-controlled study of cancer of the vulva.
Obstet Gynecol 1990; 75.
8. Downey G, Okagaki T, Ostrow R et al. Condylomatous carcinoma of the vulva
with special reference to human papillomavirus DNA. Obstet Gynecol 1988; 72:68.
9. Carter J, Carlson J, Fowler J et al. Invasive vulvar tumors in young women: A
disease of the immunosuppressed? Gynecol Oncol 1993; 51:307.
10. Mitchell M, Prasad C, Silva E et al. Second genital primary squamous neoplasms
in vulvar carcinoma: Viral and histopathologic correlates. Obstet Gynecol 1993;
81:13.
11. Brinton L, PC N, Mallin K et al. Case-controlled study of cancer of the vulva.
Obstet Gynecol 1990; 65:859.
12. Franklin E, Rutledge F. Epidemiology of epidermoid carcinoma of the vulva.
Obstet Gynecol 1972; 39;165.
13. Parry-Jones E. Lymphatics of the vulva. J Obstet Gynecol Br Empire 1963; 70:751.
14. Shimm D, Fuller A, Orlow E et al. Prognostic variables in the treatment of squa-
mous cell carcinoma of the vulva. Gynecol Oncol 1986; 24:343.
15. Parker R, Duncan I, Rampone J et al. Operative management of early invasive
epidermoid carcinoma of the vulva. Am J Obstet Gynecol 1975; 123:349.
16. Hacker N, Nieberg R, Berek J et al. Superficially invasive vulvar cancer with nodal
metastases. Gynecol Oncol 1983; 15:65.
17. Chu J, Tamimi H, Figge D. Femoral node metastases with negative superficial
inguinal nodes in early vulvar cancer. Am J Obstet Gynecol 1981; 140:337.
18. Taussig F. Carcinoma to the vulva: An analysis of 155 cases. Am J Obstet Gynecol
1970; 40:764.
19. Way S. Carcinoma of the vulva. Am J Obstet Gynecol 1960; 79:692.
20. Hacker N, Leuchter R, Berek J et al. Radical vulvectomy and bilateral inguinal
lymphadenectomy through separate groin incisions. Obstet Gynecol 1981; 58:574.
21. Iverson T Aalders J, Christensen A et al. Squamous cell carcinoma of the vulva: A
review of 424 patients, 1957-1974. Gynecol Oncol 1980; 9:271.
22. Podratz K, Symmonds R, Taylor W. Carcinoma of the vulva: Analysis of treat-
ment failures. Am J Obstet Gynecol 1982; 143:340.
23. Andreasson B, Bock J, Weberg E. Invasive cancer in the vulvar region. Acta Obstet
Gynecol Scand 1982; 61:113.
24. Monaghan J, Hamond I. Pelvic node dissection in the treatment of vulvar carci-
noma, is it necessary? Br J Obstet Gynecol 1984; 91:270.
25. Heaps J, Fu Y, Montz R et al. Surgical-pathologic variables predictive of local
recurrence in squamous cell carcinoma of the vulva. Gynecol Oncol 1990; 38:309. 13
26. Homesley H. Prognostic factors for groin node metastasis in squamous cell car-
cinoma of the vulva (a Gynecologic Oncology Group Study). Gyencol Oncol
1994; 49:279.
27. Stehman F, Bundy B, Dvoretsky P et al. Early stage I carcinoma of the vulva treated
with ipsilateral superficial inguinal lymphadenectomy and modified radical
hemivulvectomy: A prospective study of the Gynecolgic Oncolgy Group. Obstet
Gynecol 1992; 79:490.
28. DiSaia P, Creasman W, Rich W. An alternate approach to early cancer of the vulva.
Am J Obstet Gynecol 1979; 133:825.
29. Drag D, Weaver D, Alex J et al. Surgical resection and raidolocalization of the
sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993;
2:335.
30. Guiliano A, Kirgan D, Guenther J et al. Lymphatic mapping and sentinel lym-
phadenectomy for breast cancer. Ann Surg 1994; 220:391.
31. Morton D, Wen D, Wong J et al. Technical details of intraoperative lymphatic
mapping for early stage melanoma. Arch Surg 1992; 127:392.
32. Levenback C, Burke T, Gershenson D et al. Intraoperative lymphatic mapping
for vulvar cancer. Obstet Gynecol 1994; 84:163.
33. DeCsare S, Fiorica JV, Roberts W et al. A pilot study utilizing intraoperative
lymphoscintigraphy for identification of the sentinel lymph nodes in vulvar can-
cer. Gynecol Oncol 1997; 66:425.
34. Trimble E, Lewin J, Williams L. Management of vulvar melanoma. Gynecol Oncol
1992; 45:254.
35. Clark W, Bernardino E, From L et al. This histogenesis and biologic behavior of
primary human malignant melanomas of the skin. Cancer Res 1969; 29:705.
36. Phillips G, Bundy B, Okagaki T et al. Malignant melanoma of the vulva treated
by radical hemivulvectomy: A prospective study of the Gynecologic Oncology
Group. Cancer 1994; 73:2626.
37. Homesley H. Management of vulvar cancer. Cancer 1995; 76:2159.
13
Bone Lesion Localization 153
Bone Lesion Localization
Lary A. Robinson
Introduction ........................................................................................................ 153

Bone Metastases .................................................................................................. 153
Radioisotope Bone Scan ...................................................................................... 154
Techniques o f Bone Lesio n Localization ............................................................. 155
Conclusions ......................................................................................................... 161
INTRODUCTION
Accurate staging plays a critical role in the evaluation of any malignancy be-
cause it has a pivotal role in determining therapy. Since stage and survival are
strongly correlated, the prognosis is also determined by staging. Metastases to bone
from any solid tumor are classified as distant, blood-born metastases and the tu-
mor is considered Stage IV.1 Documented bone metastases eliminates surgery as a
curative option, generally mandating a chemotherapy approach occasionally with
radiotherapy added for symptomatic lesions.
In one study in 1976, as many as 15% of primary extraosseous malignancies
have solitary abnormalities on radioisotope bone scan,2 yet from 36%2-71%3 of
these bone lesions were benign on biopsy. With modern, more sensitive imaging
equipment, it is likely that more routine bone scans will have at least one abnor-
mal focus leading to an even higher false positive rate of suspected bone metasta-
sis. This high false positive rate in solitary bone scan abnormalities could poten-
tially lead to over-staging of a malignancy. Therefore, it is imperative that there is
histologic confirmation of suspected osseous metastases, especially when they are
discovered only on radioisotopic bone scans.
BONE METASTASES 14
Despite the fact that primary bone malignancies are rare, metastases to the
bone are relatively common and tend to be found in sites of persistent red mar-
row, particularly the axial skeleton.4 The most common sites of metastases in a
series of 2001 patients with known bone metastases was (in decreasing frequency)
the vertebrae, pelvis and sacrum, femur, ribs, skull, humerus, scapula, and sternum.5
Metastases to bone most commonly are seen in breast and in prostate carcino-
mas. Kidney, thyroid and lung cancer bony metastases occur less frequently.4
Cancers of the gastrointestinal tract, sarcomas, and genital tract rarely metastasize
to bone. The overall frequency of osseous metastases is 67% in breast cancer pa-
tients, 50% in prostate cancer, 25% for lung cancer and kidney cancer, and less
than 10% in all other malignancies.4
Most bony metastases present with symptoms, usually new onset bone pain
and occasionally swelling over the metastatic site. Rarely, a pathologic fracture is
the first sign of metastatic disease to the bone. An elevation of the serum alkaline
phosphatase may suggest the presence of bony metastases, but this enzyme may
be elevated due to many other causes such as hyperparathyroidism, osteomalacia,
osteitis deformans, osteogenic sarcoma, rickets, pregnancy, healing fractures, nor-
mal growth, and a variety of hepatobiliary conditions.6 The finding of an elevated
serum calcium is even more uncommon and less specific as an indicator of bony
metastases.
Clinically-silent bone metastases are infrequent and are suspected from an
abnormal staging radioisotopic bone scan. For some malignancies, a bone scan is
routine in the initial workup and asymptomatic abnormalities may be found. For
other cancers, such as nonsmall cell lung cancer, a bone scan is only recommended
if there is some definite clinical indicator such as new onset bone pain or an el-
evated serum calcium or serum alkaline phosphatase. Osseous metastases are rarely
found in nonsmall cell lung cancer in the absence of definite clinical indicators.7,8
The usual plain radiographic finding suggestive of bony metastases is decreased
density at the site of the metastasis. Most bone metastases cause bone destruction
giving this osteolytic picture. Much less common are osteoblastic or osteoscle-
rotic changes from increased bone density that occasionally is caused by prostate
cancer or from hormone treatment in metastatic breast cancer.5 However, the plain
bone radiograph is a relatively insensitive indicator of metastatic disease since at
least 50% of trabecular bone must be destroyed before it is radiographically visible.
RADIOISOTOPE BONE SCAN
A much more sensitive indicator of metastatic disease is the radioisotopic bone

scan, which is generally positive when as little as 5-15% of trabecular bone is de-
stroyed by tumor.4 Therefore, the bone scan may well demonstrate a bony me-
14 tastasis much earlier than a plain bone radiograph. Indeed, in only 3% of patients
will bone metastases be visible on a plain radiograph when the bone scan is normal.
The mechanism allowing for uptake of the radiopharmaceutical, usually
99m
technetium-labeled diphosphonate, on a bone scan is incompletely understood.
Areas of increased bone formation and/or increased blood flow localize the radio-
isotope9 to cause increased uptake (“hot spot”) on bone scan. Metastatic tumor
will cause local bone destruction but also there will generally be synchronous bone
formation. The radiopharmaceutical appears to bind to the surface of the hydroxya-
patite crystal of the newly forming bone, and not to the tumor cells themselves.
A focal decrease in the uptake of the radioisotope, a “photopenic lesion,” is
unusual and is generally seen with areas of decreased blood flow or rarely a meta-
static tumor with decreased or absent new bone formation.9 A rapidly-growing,
destructive metastatic tumor may cause bone resorption before new bone miner-
alization occurs. Therefore, the radioisotope is not taken up in this localized area
of metastatic tumor. However, these instances are rare and most bone scan lesions
demonstrate increased uptake.
Despite the high sensitivity of the bone scan, it is not particularly specific in
demonstrating metastatic tumor in the bone. The bone scan may have localized
areas of increased uptake of radioisotope from a wide variety of normal or benign
causes listed in Table 14.1, which may lead to a false-positive result in the staging
of the primary malignancy. The nonspecific nature of a positive bone scan, espe-
cially when plain radiographs are normal, mandates biopsy and histologic confir-
mation of the suspected metastatic tumor to definitively stage the primary malig-
nancy and determine subsequent treatment.
TECHNIQUES OF BONE LESION LOCALIZATION
Metastatic disease to the bone may be readily apparent in some patients who
have localizing symptoms. When local bone pain or swelling is present, plain ra-
diographs often will demonstrate the lesion. A bone scan may then be obtained if
these plain radiographs are equivocal, but also a bone scan may be helpful to search
for other metastatic sites particularly on weight-bearing bones such as the femur,
so that they can be treated before there is a pathologic fracture. Other radiological
techniques such as computed tomography (CT) or particularly magnetic reso-
nance imaging (MRI) with gadolinium contrast appear to be quite sensitive, espe-
cially with the spine and pelvis, to locate metastases when the plain radiographs
are normal and the bone scan is abnormal.4 An MRI of the spine may provide
important information in patients with neurologic symptoms or vertebral body
collapse. The MRI will delineate the extent of the tumor mass to determine if
spinal cord impingement is imminent or present, and to allow planning for pos-
sible neurosurgical intervention. However, stepping straight to an initial “screen-
ing” MRI of the spine in the patient with advanced cancer and new back pain to
locate and delineate the extent of metastatic disease to the spine may be more
efficient and cost-effective than the conventional approach of first obtaining the
plain radiographs, the bone scan, and then the spine MRI.10 14
In the patient with a soft-tissue mass or a large lytic lesion in the bone, a percu-
taneous needle biopsy usually will provide histologic confirmation of the meta-
static tumor. A CT-directed needle biopsy of a spine lesion is also often successful.
However, because of the potential for sampling error when a needle biopsy is
nondiagnostic, a open surgical biopsy may be advisable when staging and treat-
ment depends on obtaining a malignant histologic result. With local bone symp-
toms and a corresponding plain radiographic lesion, intraoperative localization
of the bone abnormality for surgical biopsy is usually not difficult, especially if the
target bone is a rib and the patient is thin. Nevertheless, in the obese or very mus-
cular patient, it still may be difficult to find the correct rib without a large incision
and multiple intraoperative radiographs or fluoroscopy.
Table 14.1. Conditions causing increased uptake of tracer on bone scan*
•Normal structures:
Sternomanubrial and corpus-manubrial joints
Base of skull, facial bones, inferior tip of scapula
Alae of sacrum
Kidneys and bladder
Variant anatomy
•Increased blood flow and bone formation:
Sudeck’s atrophy
Hyperostosis frontalis interna
Osteitis pubis
Renal osteodystrophy
Eosinophilic granuloma
Fibrous dysplasia
Paget’s disease
Aseptic necrosis, cysts
Osteoid osteoma
•Bony abnormalities with increased blood flow:
Osteomyelitis, osteitis
Fracture (recent or healing)
•Soft-tissue abnormalities:
Calcific tendinitis or myositis
Hydronephrosis or hydroureter
Dental abscess
Injection site
Postoperative scar
Soft tissue osseous metaplasia
•Benign bone tumors:
Fibroma
Chondroma
•Primary malignant bone tumors:
Chondrosarcoma
Osteosarcoma
Ewing’s sarcoma
•Metastatic tumors
*Adapted from Wahner HW and Brown ML,9 with permission.
When there is a bone scan abnormality, especially if there is more than one,
14 and the plain radiographs are normal in an asymptomatic patient with a known
cancer, most radiologists feel that this indicates metastatic disease in bone.2 Fig-
ure 14.1 illustrates a typical patient with a positive bone scan with suspected meta-
static cancer but the plain bone radiographs were normal and the patient (quite
obese) had no localizing symptoms. However, it appears that this commonly-held
belief by radiologists is incorrect since there is a bone biopsy-documented, very
high incidence of false-positive bone scan results in this setting, ranging from
47-71% in various studies.3,8,11,12 That is, benign lesions frequently account for
these bone scan abnormalities and a confirmatory open biopsy is mandatory. Still,
there are increased technical problems for the surgeon to find the exact lesion in
the bone when only the bone scan serves as a guide.
Fig. 14.1. Left antero-oblique view of a radioisotope bone scan (coned down view of
the thorax only) of a patient with suspected carcinoma metastatic to bone. An area of
increased tracer activity (white arrow) is seen in the anterolateral aspect of the left 5th
rib. The rib detail radiographs of this area of the 5th rib were normal. This small area
of the rib was biopsied using gamma probe guidance and it proved histologically to be
a benign chondroma.
CONVENTIONAL SURGICAL BIOPSY

The surgeon faces a difficult clinical problem when asked to do a biopsy of an
area of increased uptake on bone scan in an asymptomatic patient with normal
plain bone radiographs. When positive on bone scan, ribs are technically the easi- 14
est to biopsy but localization of the exact site is challenging. Depending upon the
quality and definition of the image on the bone scan, it may be difficult just to
define exactly which rib is “hot” and even more challenging to localize the precise
anterior-posterior location of the lesion.
In this setting, the surgeon may be forced to make a substantial incision to
excise a large area of one or even two ribs which are usually grossly normal in
order to be sure that the abnormality is included in the specimen. Equally frus-
trating is the difficulty at surgery of actually counting which is the target rib for
resection when the patient is obese or muscular. As a result, one or often many
more intraoperative radiographs (usually cross-table lateral radiographs) with a
radio opaque marker imbedded through the wound in the rib are necessary to
verify the correct numbered rib to biopsy. Since it is not possible to obtain an
intraoperative frozen section examination of the excised rib to verify that an ab-
normal area was excised, the final pathologic diagnosis awaits 7-10 days of decal-
cification and processing. And if the final diagnosis is “normal” bone, then there is
more uncertainty whether the correct site was biopsied since some abnormality
should be present to cause the localized uptake of radioisotope.
METHYLENE BLUE MARKING OF THE TARGET RIB

In view of the technical difficulty and uncertainty of conventional rib biopsy
techniques using only the bone scan image as a guide, Little and associates in 1993
published their results of a technique they developed to mark the target rib pre-
operatively in the nuclear medicine department. 11 In this technique, the
99m
technetium diphosphonate radioisotope is injected 6-12 hours earlier and the
patient is brought to the nuclear medicine department. An external technetium
point source is positioned over the abnormal area imaged by the gamma camera.
Once the point source is properly positioned and corresponds in several views to
the abnormal area on the bone scan, the skin is marked with a pen. The area is
anesthetized with lidocaine and a needle is inserted to the underlying rib where a
small amount of radioisotope is injected locally to mark the rib. Further nuclear
imaging is performed to verify that the area marked with the radioisotope corre-
sponds to the hot spot on the bone scan. Several injections may be necessary to
verify positioning. This is followed by an injection of methylene blue into the
same needle left in place to stain the periosteum of the rib and the local soft tissue
up to the skin. At this point, the patient is promptly taken to the operating room
for an open biopsy of the blue-stained rib before the stain diffuses further into the
surrounding tissues.
From this study, Little and associates11 biopsied bones in 15 cancer patients
(13 ribs, 1 skull and 1 scapula), and had a positive histologic result in all biopsies.
However, only 8 of 15 (53%) had metastatic cancer and the other were benign
conditions thereby yielding a false positive bone scan rate of 47%. Moores and
collegues12 in 1990 published a follow-up series of 33 bone biopsies using the
same localization technique and a positive histologic abnormality was seen in 97%
of biopsies. But once again only 52% of the entire series had a malignant focus
14 found in the bone, so that the false positive rate was 48%. However in the 17
patients in this series with an abnormal scan but normal plain radiographs, the
false positive rate was even higher at 71%.12
Although this methylene blue bone staining technique appears to be effective
in experienced hands, it is somewhat time-consuming and cumbersome. The
nuclear medicine department must carefully coordinate their time with an avail-
able operating room so that the patient may undergo immediate surgery after
injection of the methylene blue to avoid spread of the dye to other ribs. In view of
the considerable experience and coordination needed in the nuclear medicine
department to allow smooth performance of this technique, the surgeon perform-
ing only an occasional bone biopsy using this method of localization might find it
difficult to duplicate the excellent results of Little11 and Moores.12
INTRAOPERATIVE GAMMA PROBE LOCALIZATION
The development of hand-held, high resolution gamma probes that provide
real-time gamma counting intraoperatively has enabled the evolution of a much
simpler, more direct technique to localize bone lesions for biopsy. These gamma
radiation detection devices have been used quite successfully for intraoperative
lymphatic mapping to locate and biopsy sentinel lymph nodes in melanoma13
and breast cancer14 patients. This technique has subsequently been adapted for
use at our cancer center for the intraoperative localization of areas of increased
uptake of radioisotope in ribs and sternum to guide the surgeon in open bone
biopsy in patients suspected of having bone metastases.3
Technique
1. Three to twelve hours prior to surgery (preferably 4-6 hours for the
strongest counts), each patient receives an intravenous injection of
28 mCi technetium 99m oxidronate, the standard dose for a radioiso-
topic bone scan.
2. General anesthesia is induced and the patient is positioned, prepped,
and draped. The hand-held gamma probe in a sterile plastic sleeve is
then used to locate the area of greatest tracer activity (measured in real
time in counts per second) beginning in the anatomic area suspected
from the bone scan. There is a moderate amount of background activ-
ity from the tracer in all of the nearby bones. The target “hot spot” on
the bone scan when encountered by the probe will have a noticeably
increased amount of tracer activity.
3. A 3-4 cm incision is made over the point of maximum activity and the
bone is exposed down to the periosteum. Now the probe in the sterile
sleeve is used to locate precisely the point of greatest activity in the sur-
gical wound with comparison to the background counts on nearby ribs
and elsewhere on the same rib. The point of greatest activity is then
marked directly on the surface of the rib with the electrocautery just
prior to removal of that section. Though this is a small wound, the tis-
sues can easily be moved around enough to perform counting on the
adjacent ribs immediately above and below the target ribs as well as
further away from the hot spot on the same rib. The hot spot with its
increased counts is usually very localized. In our series of biopsies of 18 14
ribs, the mean ratio of hot spot counts on the target rib to the back-
ground counts on adjacent ribs was 1.68. That is, the area of greatest
tracer activity was an average 68% “hotter” by counts than the counts
on surrounding bones, a readily discernible difference.
4. If desired, an intraoperative cross-table lateral chest radiograph may
then be taken with a radio opaque marker on the rib (a spinal needle
imbedded in the periosteum is my preference) to verify that the target
rib is the correct numbered rib based on the bone scan. Usually, several
radiographs are necessary to get the correct view so that the 1st or 12th
rib is included in the image to enable counting of the ribs from above or
below. After experience with only a few biopsies and after gaining
confidence in this radioguided technique, we eliminated this time-con-
suming and costly step.
5. A 3 cm portion of rib is then removed subperiosteally or the outer table
of sternum or other bone is removed. The remaining ribs are then
checked again in the wound with the gamma probe to verify that the
hot spot target was removed. Of note, in the setting of a hot bone scan
and normal rib detail plain radiographs, the resected piece of rib usu-
ally is grossly normal. The resected rib is placed in a solution to decal-
cify it and the specimen may be isolated for three to four days to allow
the radioactive tracer to decay, depending upon the radiation safety re-
quirements of your institution. Then, the rib is studied histologically
with particular attention directed to the area of rib scored on the sur-
face with the electrocautery.
6. The small wound is filled with saline to check to see if the pleural cavity
was inadvertently entered, which occurred about 15% of the time in
our experience. If the pleura was violated, a small 24 Fr chest tube is
inserted through a separate skin incision and it is removed in the recov-
ery room after the postoperative chest radiograph is shown to be nor-
mal (without a pneumothorax or pleural effusion). The wound is closed
in layers with absorbable sutures, with a final subcuticular skin closure.
Almost all patients are then discharged the same day from the recovery
room, with only quite debilitated patients requiring observation over-
night. A postoperative chest radiograph documents which numbered
rib was biopsied.
Results
In our initial series3 which is now updated and enlarged, 15 patients (9 men
and 6 women) with a variety of known or suspected underlying primary cancers
had positive radioisotopic bone scans but with normal plain radiographs, who
were either asymptomatic or had minimal, nonlocalizing symptoms. A total of 19
bones (18 ribs and 1 sternum) were biopsied using this radioguided technique. All
(100%) bones biopsied contained a pathologic process that accounted for the
positive bone scan. However, only 4/19 (21%) of the bones demonstrated meta-
static cancer (1 squamous cell carcinoma of the lung, 2 lymphoma, 1 carcinoma
14 of the prostate). The other 15 bones (79% false positive rate for diagnosing meta-
static cancer in the bone) showed a variety of benign pathologic processes:
hypercellular marrow (4 ribs), fibrous dysplasia (1 rib), Paget’s disease of bone (2
ribs), localized fibrosis with granulation tissue (1 rib), enchondroma (3 ribs), and
chondroma (3 ribs and 1 sternum).
With the gamma probe directed biopsies, detection of the bone scan hot spot
was easily accomplished. Intraoperatively, the bone for resection always appeared
grossly normal to the surgeon. However, the mean ratio of measured hot spot
activity on the target rib compared to adjacent ribs or the same rib away from the
hot spot was 1.68 ± 0.32 (median 1.61, range 1.31-2.69), and the difference was
easily discernible intraoperatively.
The most common benign abnormality in this series was a chondroma or en-
chondroma, accounting for 7/19 (37%) of biopsy results. This benign cartilagi-
nous tumor is relatively common, representing 13.4% of all benign bone tumors
although the actual incidence is unknown since they occur sporadically and are
asymptomatic.15 They tend to occur most commonly in the small bones of the
hands and feet, but also occur in long thin bones such as the ribs. Unless they
become very large, they remain asymptomatic and are not generally visible on
plain bone radiographs. They are usually found incidentally as a hot spot on a
bone scan performed during the evaluation of a patient with a malignancy. A
subsequent biopsy usually follows and the diagnosis of a benign chondroma is
made.
There was no morbidity or mortality specifically associated with the intraop-
erative gamma probe technique or the actual rib biopsies. With experience, the
total operative time for these cases decreased to only 20-40 minutes. Intraopera-
tive radiographs were eliminated because the counting technique is precise, rapid,
and reproducible, making the radiograph unnecessary and/or redundant for lo-
calization of the abnormal section of bone.
CONCLUSIONS
The finding of an asymptomatic area of increased uptake on a radioisotopic

bone scan with normal plain radiographs in a patient with a known or suspected
malignancy does not necessarily indicate that bone metastases are present. Rather,
there is a high false positive rate and as many as 75% or more will be benign
lesions on biopsy. Histologic confirmation is mandatory in this setting.
Until recently, the surgeon faced with the prospect of accurately locating a
bone lesion with only the bone scan image as a guide faced a real challenge. The
technique of methylene blue “tattooing” of the target bone11,12 appears to be effec-
tive but it is cumbersome and time consuming. Radioguided surgical techniques3
offer real advantages in terms of decreased operative time, less interdepartmental
coordination and 100% sensitivity. With the advent of multiple applications and
widespread use of the gamma probe in melanoma and breast cancer surgery, this
instrument is becoming a common fixture in most active operating rooms and it 14
will be available to guide surgeons in performing bone biopsies.
The gamma probe was used in our study3 to guide the biopsy of asymptomatic
rib and sternal lesions. However, this same technique could no doubt be easily
adapted by orthopedic surgeons to biopsy subtle lesions in other bones in the
appendicular skeleton. Likewise, the gamma probe could be used to aid in the
biopsy of symptomatic and/or plain radiographically visible bone lesions when
body habitus or bone lesion location might make precise conventional localiza-
tion difficult intraoperatively. This technique has no side effects, it is easy to learn
and apply, and should be strongly considered in the future for use by the surgeon
in guiding the open biopsy of suspected asymptomatic bone metastases.
REFERENCES
1. Beahrs OH, Henson DE, Hutter RVP, Kennedy BJ. Handbook for staging of can-
cer. 1st ed. Philadelphia: JB Lippincott Co 1993:3-14.
2. Corcoran RJ, Thrall, JH, Kyle RW, Kaminski RJ, Johnson MC. Solitary abnor-
malities in bone scans of patients with extraosseous malignancies. Radiology
1976; 121:663-7.
3. Robinson LA, Preksto D, Muro-Cacho C, Hubbell DS. Intraoperative gamma
probe-directed biopsy of asymptomatic suspected bone metastases. Ann Thorac
Surg 1998; 65:1426-32.
4. Rogers LF. Secondary malignancies of bone. In: Juhl JH, Crummy AB, eds. Paul
and Juhl’s essentials of radiological imaging. 6th ed, Philadelphia: J.B. Lippincott
Company 1993:164-5.
5. Clain A. Secondary malignant disease of bone. Brit J Cancer 1965; 19:15-29.
6. Henry JB, ed. Clinical diagnosis and management by laboratory methods. 19th
ed. Philadelphia:WB Saunders Co 1996:277.
7. Michel F, Solèr M, Imhof E, Perruchoud AP. Initial staging of nonsmall cell lung
cancer: Value of routine radioisotope bone scanning. Thorax 1991; 46:469-73.
8. Ichinose Y, Hara N, Ohta M et al. Preoperative examination to detect distant
metastasis is not advocated for asymptomatic patients with Stages 1 and 2
nonsmall cell lung cancer. Chest 1989; 96:1104-9.
9. Wahner HW, Brown ML. Role of bone scanning. In: Sim FH, ed. Diagnosis and
management of metastatic bone disease. 1st ed. New York: Raven Press 1988:51-67.
10. Ruchdeschel JC. Rapid cost-effective diagnosis of spinal cord compression due
to cancer. Cancer Control 1995; 2:320-23.
11. Little AG, DeMeester TR, Kirchner PT et al. Guided biopsies of abnormalities on
nuclear bone scans. J Thorac Cardiovasc Surg 1983; 85:396-403.
12. Moores DWO, Line B, Dziuban SW, Jr, McKneally MF. Nuclear scan-guided rib
biopsy. J Thorac Cardiovasc Surg 1990; 90:620-1.
13. Albertini JJ, Cruse CW, Rapaport D et al. Intraoperative radiolympho-scintigra-
phy improves sentinel lymph node identification for patients with melanoma.
Ann Surg 1996; 223:217-4.
opsy in the patient with breast cancer. J Amer Med Assoc 1996; 276:1818-22.
15. Unni KK. Chondroma. In: Unni KK. Dahlin’s Bone tumors. 5th ed, Philadelphia:
Lippincott-Raven 1996:25-45.
14
Index 163
Index
A I
Accuracy 48, 49, 83, 84, 89 Immunohistochemistry 116, 124
ACS (American College of Surgeons) Internet 6, 41, 72
4, 15, 43-45 Intraoperative imprint cytology (IIC)
106-108, 111-114
B Intron A® 3, 63, 69
Biopsy 39, 40, 43, 44, 49-52, 54-58, L

68-70, 72, 73, 75, 76, 80, 84, 87, 88,
94, 95, 111, 120, 124, 125, 127, 131, Localization ratio 66, 68
134, 135, 139, 145, 146, 156-160, Lymphazurin 11, 48, 49, 65, 75, 76,
162 83, 88
Bone 35, 71, 101, 104, 153-161 Lymphoscintigraphy 5, 6, 11, 12, 17,
Breast cancer 3, 4, 6-8, 11, 12, 17, 18, 20, 25, 28-31, 34, 35, 37, 41, 42, 90,
22, 40, 41, 70 91, 93, 95, 97,-99, 105, 135, 139,
147
C
M
CESTE (Committee on Emerging
Surgical Technology and Educ) Melanoma 3, 6, 7, 9, 11, 12, 16-18, 20,
43-45 21, 31, 37, 63-66, 68-71, 73, 74, 78,
Credentialing 2, 18, 40-45, 72 80, 83, 90, 95, 99, 100, 103, 115-
Cytokeratin 82, 87, 106-110, 113 124, 133, 135, 139, 140, 143, 147,
148, 152, 159, 161
D Merkel Cell 70
Multicenter selective lymphadenec-
Diff-Quik 107-109, 111-113 tomy trial 58
E N
Exclusivity 17, 18, 69 Navigator™ 66
G P
Gamma camera 26, 30, 31, 93, 95, 99, Parathyroid 4, 6, 16, 18, 19, 21, 30, 35,
158 40, 71, 104, 105, 126, 128, 129, 131,
Gamma probe 1, 5, 6, 23, 26, 29, 31, 132
38, 50, 52, 66,-68, 77-80, 91, 97, PCR (polymerase chain reaction)
129, 130, 139, 157, 159-162 56-58, 60, 107, 113, 123
H
HMB-45 54, 68, 116, 123
Hyperparathyroidism 11, 104, 126,
128, 154
Index R Sestamibi scan 11, 19, 104, 126-130,

132
Radiation Squamous cell 71, 133-135, 137, 139,
detector 13, 28, 78 140, 143, 148, 149, 160
safety 23, 24, 33, 34, 36-38, 41, 108, Step sectioning 56-58
132, 160 Sunbelt melanoma trial 58, 59
exposure 25, 33, 34
Reimbursement 19, 21, 40, 41, 42, 72 T
Rib 155, 157-161
Technetium 25, 65
S Telemedicine 40-42
S-100 54, 115-117, 119-124 V

Sensitivity 3, 17, 77, 111, 116, 120,
123, 126, 127, 155, 161 Vulvar cancer 143, 144, 146, 148, 150
Serial sections 7, 9, 116
LANDES LANDES
BIOSCIENCE V ade me c um BIOSCIENCE V ade me c um
Table of contents
1. Logistics and Organizational Aspects of
Radioguided Surgery Programs
8.
9.
Lymphoscintigraphy

Radioguided
2. Intraoperative Gamma Radiation Detection Nodes in Breast Cancer
3.
and Radiation Safety
Training and Credentialing Physicians in

Radioguided Surgery
10.
11.
Nodes in Malignant Melanoma

Surgery
4. Melanoma Lymphatic Mapping: Scientific Radioguided Parathyroidectomy (MIRP)
Support for the Sentinel Lymph Node
12. Emerging Applications in Other Skin
Concept and Biological Significance of the
Sentinel Node Cancers
5. Sentinel Lymph Node Biopsy for 13. Radioguided Surgery and Vulvar
Carcinoma
Melanoma: Surgical Technique
14. Bone Lesion Localization Intraoperative
6. Technique for Lymphatic Mapping in
Breast Carcinoma Deaths
7. The Technique of Intraoperative Lymphatic

Mapping and Sentinel Lymphadenectomy
in Breast Cancer Using Blue Dye Alone
This is one of a new series of medical handbooks.

It includes subjects generally not covered in other handbook series, especially
many technology-driven topics that reflect the increasing influence of technology
in clinical medicine.
The name chosen for this comprehensive medical handbook series is Vademecum,
a Latin word that roughly means “to carry along”. In the Middle Ages, traveling
clerics carried pocket-sized books, excerpts of the carefully transcribed canons,
known as Vademecum. In the 19th century a medical publisher in Germany, Samuel
Karger, called a series of portable medical books Vademecum.
The Landes Bioscience Vademecum books are intended to be used both in the
training of physicians and the care of patients, by medical students, medical house
staff and practicing physicians. We hope you will find them a valuable resource.
Eric D. Whitman
All titles available at
www.landesbioscience.com
Douglas Reintgen

Whitman, Reintgen - Radio Guided Surgery (Landes Bio Science Vademecum

Uploaded by

Copyright:

Available Formats

Whitman, Reintgen - Radio Guided Surgery (Landes Bio Science Vademecum

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Whitman, Reintgen - Radio Guided Surgery (Landes Bio Science Vademecum

Uploaded by

Copyright:

Available Formats

LANDES LANDES

BIOSCIENCE V ade me c um BIOSCIENCE V ade me c um

Pathologic Evaluation of Sentinel Lymph

Training and Credentialing Physicians in

The Technique of Minimally Invasive

7. The Technique of Intraoperative Lymphatic

This is one of a new series of medical handbooks.

Eric D. Whitman, M.D.

Douglas Reintgen, M.D.

Copyright © 1999 Landes Bioscience

Please address all inquiries to the Publishers:

Development of this book was supported in part by an unrestricted educa-

Library of Congress Cataloging-in-Publication Data

Douglas Reintgen, M.D.

James V. Fiorica, M.D. Fadi F. Haddad, M.D.

Logistics and Organizational Aspects 1

Radioguided Surgery, edited by Eric D. Whitman and Douglas Reintgen.

and consistent pathology evaluation algorithm employing the techniques described

The most important organizational issues that I have encountered in my in-

Lymphoscintigraphy for SLN procedures is performed using standard tech-

minimizing potential disruption of surgical schedule and acquisition of unneces-

1 Sentinel Node Pathologic Evaluation

Place in saline and

Serial sections cut ("Bread

Scrape prep or touch prep Results conveyed to

H & E exam of central slides

Exam no cytokeratin immunohistochemistry

Positive Exam Exam

Fig. 1.1. Sentinel node pathologic evaluation.

To schedule a RGS procedure, it is necessary to coordinate the availability and

1) We never obtain a radionuclide localization study before the day of sur-

Fig. 1.3. Example of an intraoperative gamma probe, specifically

As your institution progresses in its development of a RGS program, it is im-

DATA ACQUISITION, STORAGE AND ANALYSIS

Our information is stored in computerized, password protected, data regis-

100 = Sensitivity + False Negative Rate

BILLING AND REIMBURSEMENT

reimbursement are the most intently listened-to sections of my presentation.

Table 1.1. Billing for RGS procedures

SLN Biopsy, Melanoma

* Amount billed is usual and customary rate, except as otherwise noted

We believe that in the absence of an established all-encompassing code for

Table 1.2. Reimbursement for RGS procedures*

Insurance Co #2 100% HMO

Insurance Co #4 100% HMO

Insurance Co #6 100% small private insurer

Insurance Co #7 60% HMO

Insurance Co #8 Under appeal “Not part of benefit package”

Insurance Co #9 Under appeal “Included as part of wide

This chapter has described the implementation of a successful RGS program,

Intraoperative Gamma Radiation

David A. Hillier, Henry D. Royal

Effective intraoperative use of a gamma probe requires familiarity with basic

BASIC RADIATION TERMINOLOGY

Transition from high to low energy state

Alpha particles Neutrons

Fig. 2.1. Types of radiation.

radionuclide, Technetium-99m (Tc-99m), has a t1/2p of 6 hours. This means that

GAMMA PROBE TESTING