HUMANA PRESS
Handbook of
Diagnostic
Endocrinology
Edited by
Janet E. Hall,
MD
Lynnette K. Nieman,
MD
CONTEMPORARY ENDOCRINOLOGY
™
Handbook of
Diagnostic
Endocrinology
Edited by
Janet E. Hall,
MD
Lynnette K. Nieman,
MD
HUMANA PRESS
HANDBOOK OF DIAGNOSTIC ENDOCRINOLOGY
00/FM/CE/Hall/F 12/4/02, 8:48 AM1
Androgens in Health and Disease, edited by
C
ARRIE BAGATELL AND WILLIAM J.
B
REMNER, 2003
Endocrine Replacement Therapy in Clinical
Practice, edited by A. W
AYNE MEIKLE,
2003

Challenging Cases in Endocrinology,
edited by M
ARK E. MOLITCH, 2002
Selective Estrogen Receptor Modulators:
Research and Clinical Applications,
edited by A
NDREA MANNI AND MICHAEL
F. VERDERAME, 2002
Transgenics in Endocrinology, edited by
M
ARTIN MATZUK, CHESTER W. BROWN,
AND T. RAJENDRA KUMAR, 2001
Assisted Fertilization and Nuclear Transfer
in Mammals, edited by D
ON P. WOLF
AND
MARY ZELINSKI-WOOTEN, 2001
Adrenal Disorders, edited by A
NDREW N.
M
ARGIORIS AND GEORGE P. CHROUSOS,
2001
Endocrine Oncology, edited by S
TEPHEN P.
E
THIER, 2000
Endocrinology of the Lung: Development
and Surfactant Synthesis, edited by
C
AROLE R. MENDELSON, 2000

CHARLES T. ROBERTS, JR., 1999
Neurosteroids: A New Regulatory Function
in the Nervous System, edited by
E
TIENNE-EMILE BAULIEU, PAUL ROBEL,
AND MICHAEL SCHUMACHER, 1999
Autoimmune Endocrinopathies, edited by
R
OBERT VOLPÉ, 1999
Hormone Resistance Syndromes, edited by
J. L
ARRY JAMESON, 1999
Hormone Replacement Therapy, edited by
A. W
AYNE MEIKLE, 1999
Insulin Resistance: The Metabolic Syndrome
X, edited by G
ERALD M. REAVEN AND
AMI LAWS, 1999
Endocrinology of Breast Cancer, edited by
A
NDREA MANNI, 1999
Molecular and Cellular Pediatric
Endocrinology, edited by S
TUART
HANDWERGER, 1999
Gastrointestinal Endocrinology, edited by
G
EORGE H. GREELEY, JR., 1999
CONTEMPORARY ENDOCRINOLOGY

information published and to describe generally accepted practices. The contributors herein have carefully checked
to ensure that the drug selections and dosages set forth in this text are accurate and in accord with the standards
accepted at the time of publication. Notwithstanding, as new research, changes in government regulations, and
knowledge from clinical experience relating to drug therapy and drug reactions constantly occurs, the reader is
advised to check the product information provided by the manufacturer of each drug for any change in dosages or
for additional warnings and contraindications. This is of utmost importance when the recommended drug herein is
a new or infrequently used drug. It is the responsibility of the treating physician to determine dosages and treatment
strategies for individual patients. Further it is the responsibility of the health care provider to ascertain the Food and
Drug Administration status of each drug or device used in their clinical practice. The publisher, editors, and authors
are not responsible for errors or omissions or for any consequences from the application of the information presented
in this book and make no warranty, express or implied, with respect to the contents in this publication.
This publication is printed on acid-free paper.
∞
ANSI Z39.48-1984 (American National Standards Institute)
Permanence of Paper for Printed Library Materials.
Production Editor: Robin B. Weisberg
Cover Design: Patricia Cleary
Photocopy Authorization Policy:
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is
granted by Humana Press Inc., provided that the base fee of US $20.00 per copy is paid directly to the Copyright
Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a
photocopy license from the CCC, a separate system of payment has been arranged and is acceptable to Humana Press
Inc. The fee code for users of the Transactional Reporting Service is: [0-89603-757-6/03 $20.00].
Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
Handbook of diagnostic endocrinology /edited by Janet E. Hall and Lynnette K. Nieman.
p. ; cm.—(Contemporary endocrinology)
Includes bibliographical references and index.
ISBN 0-89603-757-6 (alk. paper); 1-59259-293-7 (e-book)
1. Endocrine glands—Diseases—Diagnosis—Handbooks, manuals, etc. I. Hall, Janet E.
II. Nieman, Lynnette, K. III. Contemporary endocrinology (Totowa, NJ).

occur in only 10% of hypertensive subjects, this small fraction represents a large
number of patients. Jennifer Lawrence and Robert Dluhy present a thoughtful
and efficient approach to the diagnosis of these disorders. Anastassios Pittas and
Stephanie Lee then update the approach to the diagnosis of thyroid disease,
including an important discussion of currently available assays. Regina Castro
and Hossein Gharib present a cost-effective approach to evaluation of the com-
mon problem of thyroid nodules and discuss the preoperative evaluation and
postoperative followup of patients with thyroid cancer. Allison Goldfine pro-
vides an overview of the diagnosis of the various forms of diabetes and a com-
prehensive discussion of long-term monitoring of the primary disease and its
complications. Robert Ratner reviews the important area of gestational diabetes,
both its diagnosis and consequences, and John Service presents a clear approach
to the diagnosis of hypoglycemia. William Donahoo, Elizabeth Stephens, and
00/FM/CE/Hall/F 12/4/02, 8:48 AM5
vi Preface
Robert Eckel provide a thorough discussion of the modalities that are currently
available for the assessment of dyslipidemia and obesity. Benjamin Leder and
Joel Finkelstein then review calcium metabolism and present a logical approach
to the diagnosis of hyper- and hypocalcemia. Patrick Doran and Sundeep Khosla
discuss the spectrum of osteoporosis and an approach to diagnosis and long-term
followup based on the currently available tools. Margaret Wierman reviews the
normal physiology of the hypothalamic–pituitary–testicular axis and the physi-
ology of erection, both of which are critical to the approach to evaluation of
hypogonadism and erectile dysfunction. Drew Tortoriello and Janet Hall then
discuss the physiology of normal menstrual function as a backdrop to their
approach to the evaluation and long-term followup of women with disorders of
menstrual function. Ricardo Azziz completes this volume with a practical and
focused discussion of the evaluation of androgen excess in women.
In Handbook of Diagnostic Endocrinology we provide the reader with a
concise approach to the diagnosis of endocrine disorders that is based on an

Allison B. Goldfine
9 Gestational Diabetes: Where Do We Look for It
and How Do We Find It? 179
Robert E. Ratner
10 Hypoglycemic Disorders 193
F. John Service
11 The Evaluation of Dyslipidemia and Obesity 213
William T. Donahoo, Elizabeth Stephens,
and Robert H. Eckel
12 Hyper- and Hypocalcemia 239
Benjamin Z. Leder and Joel S. Finkelstein
00/FM/CE/Hall/F 12/4/02, 8:48 AM7
viii Contents
13 Osteoporosis 257
Patrick M. Doran and Sundeep Khosla
14 Hypogonadism and Erectile Dysfunction 277
Margaret E. Wierman
15 Menstrual Dysfunction 295
Drew V. Tortoriello and Janet E. Hall
16 Differential Diagnosis and Evaluation
of Hyperandrogenism 323
Ricardo Azziz
Index 331
00/FM/CE/Hall/F 12/4/02, 8:48 AM8
ix
CONTRIBUTORS
RICARDO AZZIZ, MD, MPH, MBA, Department of Obstetrics and Gynecology,
Cedars-Sinai Medical Center, Los Angeles, CA
M. R
EGINA CASTRO, MD, Department of Medicine and Endocrinology, Albany

Endocrinology, Massachusetts General Hospital, Boston, MA
S
UNDEEP KHOSLA, MD, Division of Endocrinology and Metabolism, Mayo
Clinic, Rochester, MN
J
ENNIFER E. LAWRENCE, MD, Valdosta Specialty Clinic, Valdosta, GA
B
ENJAMIN Z. LEDER, MD, Division of Endocrinology, Massachusetts General
Hospital, Boston, MA
S
TEPHANIE L. LEE, MD, Division of Endocrinology, Diabetes and Metabolism,
Boston Medical Center, Boston, MA
L
YNNETTE K. NIEMAN, MD, Pediatric and Reproductive Endocrinology Branch,
National Institute of Child Health and Human Development, National
Institutes of Health, Bethesda, MD
A
NASTASSIOS G. PITTAS, MD, Division of Endocrinology, Diabetes, Metabolism
and Molecular Medicine, New England Medical Center, Boston, MA
00/FM/CE/Hall/F 12/4/02, 8:48 AM9
ROBERT E. RATNER, MD, MedStar Research Institute, Washington, DC
F. J
OHN SERVICE, MD, PhD, Division of Endocrinology and Internal Medicine,
Mayo Clinic, Rochester, MN
P
ATRICK M. SLUSS, PhD, Reproductive Endocrine Unit, Division of
Endocrinology, Massachusetts General Hospital, Boston, MA
E
LIZABETH STEPHENS, MD, Division of Endocrinology, Diabetes, and Clinical
Nutrition, Oregon Health and Science University, Portland, OR

INTRODUCTION
Over the past 30 years, immunoassays have become a valuable and widely
available tool among the repertoire of clinical methods to assess hormonal func-
tion (1,2). The majority of hormonal proteins and steroids are bioactive at
extremely low concentrations in the peripheral circulation, and hormone
metabolites often play a significant physiological role at target tissue sites. There-
fore, evaluation of a patient’s endocrine status with a given immunoassay is
dependent upon both the sensitivity and specificity of that assay. Hormone levels
outside a given reference range can indicate a pathologic disease process. When
clinical decisions regarding a patient’s hormonal status or their response to a
therapeutic intervention are based on hormone testing, it is important that due
consideration be given to issues relating to the validity and utility of the assay.
In practice, the performance of an immunoassay can be influenced by a variety
of physiological, pre-analytical and analytical factors, which may result in devia-
tion from the “true” value, i.e., that which reflects the actual physiological status
of the patient. Indeed, it is important to bear in mind that the clinical definition
of a “true” value for many hormones is relative only to the reference ranges
established for the individual assay methods. Furthermore, these issues often
vitiate the ability to compare values generated by different laboratories.
1
01/Sluss/1-22/F 12/2/02, 8:53 AM1
2 Gill et al.
The intent of this chapter is to provide the reader with an appreciation of how
an endocrine immunoassay can be an effective tool provided that the methods are
carefully chosen, monitored, and applied in the appropriate clinical setting.
CURRENT IMMUNOASSAY METHODS
Principles
Immunoassays were originally developed as competitive binding assays that
utilized immunoglobulins as the binding protein. The first clinically important
immunoassay developed for hormones was the radioimmunoassay (RIA). RIA

binding both in solution and when immobilized to a solid surface. Such systems
01/Sluss/1-22/F 12/2/02, 8:53 AM2
Chapter 1/Endocrine Immunoassay 3
are referred to as immunoradiometric assays (IRMA), rather than RIA, to empha-
size the differences between the design of these two approaches. In both cases,
the amount of hormone present in a patient specimen is determined by comparing
the amount of radioactive hormone bound to antibody in the specimen and in the
standards. In addition to the standards and patient specimens, assays also include
quality control specimens, which are standardized concentrations of hormone
that are tested in every assay to confirm its accuracy and precision.
While RIA is still considered a “gold standard” in reference laboratories, it has
now been largely replaced by more rapid, nonisotopic methods, primarily enzyme-
linked immunosorbent assays (ELISAs). There are many different design strat-
egies used for ELISAs (2,13,14). Early forms utilized competition between
hormone tagged with enzyme and unlabeled hormone. This design is analogous
to RIA but substitutes enzyme activity for radioactivity for hormone detection.
Enzyme activity is detected by washing away unbound reactants and then incu-
bating the antibody–hormone complex with a substrate that produces a light-
absorbing product. More recently, substrates that produce chemiluminescence
upon enzymatic digestion have also been used effectively (15). Enzyme turnover
is then measured in a spectrophotometer or luminometer and is directly propor-
tional to the amount of enzyme-linked hormone that is bound. As in RIA, the
amount of labeled hormone is inversely proportional to the unlabeled hormone
in the standards and unknown specimens.
Immunometric immunoassays have now found widespread clinical utility,
primarily for measurement of protein hormones. These assays utilize multiple
antibody systems based on noncompetitive, e.g., “immunometric” designs (12).
In these systems, it is not necessary for the hormone to be chemically manipu-
lated by labeling with radioactivity or enzyme. Instead, endogenous hormone is
captured with one antibody and quantitatively detected with another, which is

Method
a
Method
b
Reagent Kits
DPC Manual CT R
DSL Manual P, CT, CW R, C
Linco Manual P, CW R, C
Serotec Manual CW C
Automated Systems
Architect Abbott Continuous Access MP EL
AxSYM Abbott Continuous Access PF EL
IMx Abbott Batch PF EL
ACS180 Bayer Continuous Access MP L
Centaur Bayer Continuous Access MP L
Immuno 1 Bayer Continuous Access MP L
Access Beckman Continuous Access MP C
Immulite DPC Batch CB EL
Immulite 2000 DPC Continuous Access CB EL
Advantage Nichols Continuous Access MP L. EL
Viros Eci Ortho Continuous Access CW EL
Elecsys Roche Continuous Access MP L
a
Codes: CB, coated-bead; CT, coated-tube; CW, coated-well; MP, magnetic particle; P,
precipitation; PF, particle-filtration.
b
Codes: C, colorometric enzyme-linked immunoassay; EL, enzyme-chemiluminescent
immunoassay, L, chemiluminescent immunoassay; R, radioimmunoassay.
01/Sluss/1-22/F 12/2/02, 8:53 AM4
Chapter 1/Endocrine Immunoassay 5

tion into the peripheral blood, can result in additional isoforms of these hor-
mones, which can be either nicked (broken but still complexed into a dimeric
protein) or truncated (protein sequence missing).
These aspects of the biosynthesis and posttranslational processing of gonadot-
ropins are associated with differences in the specificity and accuracy of all cur-
rent immunoassays. The presence of a shared α-subunit among gonadotropins,
TSH, and hCG results in significant cross-reactivity in immunoassays, which
depend on the specificity of the antibody for the α-subunit. Many of the early
RIAs utilized polyclonal antibodies, which primarily recognized the α-subunit.
Almost all current automated immunoanalyzers utilize at least one monoclonal
or polyclonal antibody directed at the α-subunit, often resulting in inconsistent
or conflicting results. Similarly, the high degree of structural homology between
01/Sluss/1-22/F 12/2/02, 8:53 AM5
6 Gill et al.
the LH and hCG β-subunits can lead to the inability to distinguish these two
proteins in assays, which utilize nonspecific β-directed antibodies (16). In addi-
tion, posttranslational processing, by either glycosylation or proteolysis, can
alter antigenic epitopes and thus contribute to differences among immunoassays
utilizing different, although specific antibodies. Finally, the effects of proteoly-
sis on hCG are notorious for confounding differences among hCG immunoas-
says (17).
Inaccuracies between methods, attributable at least in part to specificity issues,
can be quite substantial (18). For example, results of FSH measurements of an
identical specimen (Ligand–Special Series; August, 2000) distributed in aliquots
to different laboratories by the College of American Pathologists, revealed levels
varying by almost 2-fold due to method-based differences in gonadotropin
immunoassays of proven clinical utility (Table 2).
Regulated Secretion of Protein and Steroid Hormones
Physiological variation in serum hormone concentrations can significantly
impact immunoassay utility. Use of hormonal measurements to assess a

secreted in a pulsatile fashion by hypothalamic neurons, which in turn stimulates
pulsatile secretion of LH and FSH by the gonadotrope cells of the anterior pitu-
itary. In young healthy men, pulses of LH are secreted approx every 120 min, and
given this pulsatile secretion, concentrations of LH can vary 3–4-fold (19). In
women, there are dynamic changes in the hypothalamic-pituitary-ovarian axis
across the menstrual cycle (20–22). LH pulses occur every 90 min in the early
follicular phase, increasing to every 60 min in the mid-follicular phase and then,
following the mid-cycle LH surge and ovulation, slow progressively to 1 pulse
every 4 h in the late luteal phase. Although secretion of most gonadal steroids is
apulsatile, progesterone concentrations have been shown to fluctuate rapidly in
the luteal phase increasing from 2 to 40 ng/mL within minutes (22,23). Other
pituitary hormones, such as GH and TSH, are also secreted in a pulsatile manner,
regulated by their respective hypothalamic releasing hormones, growth hor-
mone-releasing hormone (GHRH) and thyrotropin-releasing hormone (TRH),
respectively. Such variability in serum hormone levels makes it difficult to
interpret a single blood test.
D
IURNAL AND CYCLIC VARIATION IN HORMONE LEVELS
The secretion of many hormones, including cortisol, TSH, GH, and prolactin,
follows a circadian pattern. The nocturnal rise of TSH begins in the early evening,
peaks near midnight, and reaches nadir levels by early morning. Conversely,
cortisol levels peak in the morning and decline throughout the day, with lowest
levels achieved in the late afternoon–early evening. GH and prolactin concentra-
tions increase at night at the onset of sleep and decline by early morning. Glucose
response to food also increases in the evening, which, along with a decline in
insulin secretion, results in decreased glucose tolerance in the late evening. An
increase in the counter-regulatory hormones at this time of day may contribute
to this relative insulin resistance.
During puberty, there is a nocturnal rise in LH pulse frequency and amplitude.
In adult men, mean gonadotropin levels remain relatively constant. However,

Women experience significant fluctuations in their menstrual cycle patterns
as they get older (30). After menopause, there is a gradual slowing of GnRH
pulse frequency, and mean gonadotropin concentrations in older postmeno-
pausal women are approx 40% lower than younger postmenopausal subjects
(31). In addition, the half-lives of LH and FSH are prolonged in postmenopausal
women, due to a shift to more acidic isoforms (32). During the perimenopausal
transition, peak FSH levels are achieved by d 3, as opposed to d 6 of the cycle,
and mean levels are higher than those of younger controls. Follicular phase
estradiol levels are also higher in older than younger women, whereas proges-
terone levels are lower (33).
Unlike menopause, reproductive aging in the male is a gradual process. Serum
total testosterone levels begin to decline at the age of 40 yr. Cross-sectional
studies suggest an annual decline in total testosterone of 0.4% and in free test-
osterone of 1.2% (34). The proportionately greater decline in free testosterone
could be attributed to the increase in sex hormone-binding globulin (SHBG) that
may occur with aging. Furthermore, the diurnal variation in testosterone secre-
tion is attenuated in older men (24).
C
LEARANCE AND METABOLISM OF PROTEIN AND STEROID HORMONES
Factors that alter the metabolism and clearance of hormones are another source
of specific interassay and intra-assay variation, with respect to reference ranges.
Hormones undergo structural and/or functional changes during metabolism,
01/Sluss/1-22/F 12/2/02, 8:53 AM8
Chapter 1/Endocrine Immunoassay 9
which can generate forms that are less biologically active yet still capable of
binding antibody. Therefore, different levels of hormone reflect differences in
the secretion of the biologically active fraction only if the metabolism and clear-
ance of the hormone remain constant.
Additional complications exist for hormones that circulate bound to binding
globulins. These physiological carrier proteins, such as thyroxine-binding globu-

Immulite 2000 DPC 44 698.0 920.0 815.10 49.20 6.0%
IRMA (CAC) DPC 64 508.0 796.0 650.30 58.50 9.0%
Elecsys Roche 41 826.0 1127.0 972.60 77.20 7.9%
Eci Vitros 30 737.0 991.0 843.20 55.30 6.6%
01/Sluss/1-22/F 12/2/02, 8:53 AM9
10 Gill et al.
tin secretion. GH secretion increases in response to exercise and fasting and is
decreased in obesity. In women, extreme stress, caloric restriction, and intensive
exercise can suppress endogenous GnRH secretion leading to hypothalamic
amenorrhea.
Various drugs can interfere with hormone levels in the absence of any glan-
dular dysfunction. Anticonvulsants stimulate the hepatic drug-metabolizing
system augmenting the metabolism of hormones, such as cortisol and thyroid
hormones. Alterations in the binding affinity or concentrations of binding globu-
lins can alter total hormone concentrations. Estrogen-containing drugs, such as
oral contraceptives, can increase TBG and thus total thyroxine levels, while
androgens and glucocorticoids decrease TBG concentrations. Drugs, such as
dopamine, dobutamine, and glucocorticoids can decrease TSH levels and con-
found the interpretation of TSH levels.
IMMUNOASSAY ISSUES RELATED
TO PRE-ANALYTIC VARIATION
Artifacts associated with the collection, handling, or storage of blood may alter
test results by producing changes in the specimen that do not reflect the physi-
ologic state of the patient. Many hormones are very stable in blood or serum as
illustrated in Fig. 1. In contrast, some hormones are extremely sensitive to pro-
teolytic degradation, which results in a loss of recognition by antibodies and an
underestimation of the true hormone level in blood. For example, insulin is par-
ticularly sensitive to degradation by enzymes released from hemolyzed red blood
cells, and thus, insulin measurements in hemolyzed specimens are invalid (Fig. 2).
IMMUNOASSAY ISSUES RELATED TO ANALYTIC VARIATION

two ways. Analytical sensitivity is often referred to as the limit of detection (LOD),
i.e., the lowest concentration that can be distinguished from no hormone. For
immunometric assays with a positive dose–response curve, the LOD is the dose
equivalent of the upper 95% confidence limit of replicate determinations (usually
at least 10) of a “zero” dose standard (Fig. 3). For competitive immunoassays that
01/Sluss/1-22/F 12/2/02, 8:53 AM11
12 Gill et al.
Fig. 2. Effect of pre-analytical hemolysis on insulin measurements.
have a negative dose–response curve, the lower 95% confidence limit is used.
Alternatively, the minimum detectable concentration of hormone can be defined
based upon the variance of repeated measurements of unknown specimens, referred
to as the functional sensitivity or limit of quantitation (LOQ). The LOQ takes into
account the fact that the variance of repeated measurements is related to the con-
centration of hormone being measured and is always nonuniform across the dose–
01/Sluss/1-22/F 12/2/02, 8:53 AM12
Chapter 1/Endocrine Immunoassay 13
Fig. 3. Principles of immunoassay.
response curve of an immunoassay. Typically, the coefficient of variation (CV),
defined as the standard deviation expressed as a percentage of the mean of repli-
cated measurements, is used to represent variance. As shown in the insert of
Fig. 3,
the CV of repeated measurements is nonuniform across the dose–response
curve of an assay. At very low or very high concentrations of hormone, there is no
dose–respsonse relationship, and the CV approaches infinity. Thus, a functional
sensitivity can be defined as the level associated with a minimally acceptable CV.
In the absence of specific clinical requirements for greater precision, a CV of 20%
is generally used to define functional sensitivity (40,41).
The range of concentrations between the limit of detection (or functional sen-
sitivity) and the maximum measurable concentration is called the reportable range
of the assay. Generally, the maximum measurable concentration is defined based

men. Artifacts associated with lack of antibody specificity are fairly intuitive and
generally result in an overestimation of hormone levels, to the extent that the other
compounds bind to the antibodies. However, it is also important to bear in mind
that immunoassays, which utilize one or more highly specific monoclonal anti-
bodies, can be too specific and measure only a subpopulation of biologically
active hormone (16,44,45). This point is best illustrated in the published case of
a 31-yr-old women with a history of two previous pregnancies, whose LH levels
were inappropriately low due to an immunological variant of LH that was poorly
recognized or unrecognized by two monoclonal antibodies, but had normal bio-
activity (46).
Precision
Precision is a measure of the reproducibility of repeated measurements and
varies tremendously among hormone immunoassays (47,48). The typical approach
to defining this parameter of immunoassay performance is based upon repeated
measurements of stable aliquots of the same specimen. Intra-assay or within
assay precision refers to repeated measurements in the same immunoassay, while
interassay or “between assay” precision is the term used when the measurements
are repeated in separate immunoassays. The precision associated with repeated
01/Sluss/1-22/F 12/2/02, 8:53 AM14