The Journal of Clinical Endocrinology & Metabolism, 2024, 109, 1094–1108

https://doi.org/10.1210/clinem/dgad681
Advance access publication 21 November 2023
Approach to the Patient

Approach to the Patient With Raised Thyroid Hormones

and Nonsuppressed TSH
Carla Moran,1,2,3 Nadia Schoenmakers,4 David Halsall,5 Susan Oddy,5 Greta Lyons,4
Sjoerd van den Berg,6,7 Mark Gurnell,4 and Krishna Chatterjee4
1
Endocrine Section, Beacon Hospital, Dublin, D18 AK68, Ireland
2

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

Endocrine Department, St. Vincent's University Hospital, Dublin, D04 T6F4, Ireland
3
School of Medicine, University College Dublin, Dublin, D04 V1W8, Ireland
4
Wellcome Trust-MRC Institute of Metabolic Science, University of Cambridge, Cambridge CB2 0QQ, UK
5
Department of Clinical Biochemistry, Addenbrooke's Hospital, Cambridge CB2 0QQ, UK
6
Department of Clinical Chemistry, Erasmus Medical Center, 3015 GE Rotterdam, The Netherlands
7
Department of Internal Medicine, Erasmus Medical Center, 3015 GE Rotterdam, The Netherlands
Correspondence: Krishna Chatterjee, B.M.B.Ch., Metabolic Research Laboratories, Wellcome-MRC Institute of Metabolic Science, Addenbrooke's Hospital,
Level 4, Box 289, Cambridge CB2 0QQ, UK. Email: kkc1@medschl.cam.ac.uk.

Abstract
Measurement of free thyroid hormones (THs) and thyrotropin (TSH) using automated immunoassays is central to the diagnosis of thyroid
dysfunction. Using illustrative cases, we describe a diagnostic approach to discordant thyroid function tests, focusing on entities causing
elevated free thyroxine and/or free triiodothyronine measurements with nonsuppressed TSH levels. Different types of analytical interference
(eg, abnormal thyroid hormone binding proteins, antibodies to iodothyronines or TSH, heterophile antibodies, biotin) or disorders (eg,
resistance to thyroid hormone β or α, monocarboxylate transporter 8 or selenoprotein deficiency, TSH-secreting pituitary tumor) that can
cause this biochemical pattern will be considered. We show that a structured approach, combining clinical assessment with additional
laboratory investigations to exclude assay artifact, followed by genetic testing or specialized imaging, can establish a correct diagnosis,
potentially preventing unnecessary investigation or inappropriate therapy.
Key Words: thyroid function tests, thyroid hormone action, assay interference
Abbreviations: CH, congenital hypothyroidism; FDH, familial dysalbuminemic hyperthyroxinemia; FT3, free triiodothyronine; FT4, free thyroxine; HPT,
hypothalamic–pituitary–thyroid; MCT8, monocarboxylate transporter 8; PEG, polyethylene glycol; SHBG, sex hormone–binding globulin; SRL, somatostatin
receptor ligand; T3, triiodothyronine; T4, thyroxine; TBG, thyroxine binding globulin; TFT, thyroid function test; TH, thyroid hormone; TSH, thyrotropin;
TSHoma, thyrotropinoma; TT4, total T4; TTR, transthyretin.

Measurement of circulating free thyroid hormones (THs)
(thyroxine, T4; triiodothyronine, T3) and thyrotropin (TSH)
using immunoassays is an essential step when assessing thy­
roid status, and produces characteristic patterns of thyroid
function tests (TFTs) that correlate with classical thyrotoxi­
cosis (raised THs, suppressed TSH) and hypothyroidism
(raised TSH, subnormal THs), or deviate to generate anomal­
ous or discordant TFTs due to different underlying causes (1).
Here, we describe our approach to the investigation of pa­
tients with different patterns of biochemical hyperthyroidism:
isolated, elevated free T4 [FT4] [hyperthyroxinemia]; iso­
lated, raised free T3 [FT3] [hypertriiodothyroninemia]; com­
bined elevation of FT4 and FT3 with nonsuppressed TSH.
We review different categories of assay interference (eg, due
to abnormal TH binding proteins, hormone displacement
from binding proteins, antihormone (iodothyronines, TSH)
or anti-assay reagent antibodies, biotin) causing spuriously ab­
normal hormone measurements. We consider some physio­
logical (eg, T4 replacement), pathological (eg, nonthyroidal
or acute psychiatric illness) and drug treatment (eg,
amiodarone) contexts that are associated with this biochemical
pattern, with other entities, outside the scope of this review,
being discussed elsewhere (2). We outline genetic or acquired
conditions that are associated with genuine hyperthyroxinemia
(eg, genetic or functional deficiency of deiodinase enzymes), hy­
pertriiodothyroninemia (dyshormonogenesis, resistance to thy­
roid hormone α, monocarboxylate transporter 8 [MCT8]
deficiency) or both raised FT4 and FT3 (Resistance to
Thyroid Hormone β [RTHβ], TSH-secreting pituitary tumor).
To exclude assay interference, we describe additional, sim­
ple tests that can be undertaken in many laboratories, even in
resource-limited settings, and also complex investigations that
are best undertaken in specialist centers. We discuss molecular
genetic tests used to diagnose heritable causes of biochemical
hyperthyroidism and nonsuppressed TSH.
Prismatic clinical cases, exhibiting different patterns of non­
suppressed TSH and biochemical hyperthyroidism, have been
used to illustrate our diagnostic approach, which combines
clinical, biochemical, and (if appropriate) genetic and/or
radiological investigation.

Received: 23 August 2023. Editorial Decision: 18 November 2023. Corrected and Typeset: 21 December 2023
© The Author(s) 2023. Published by Oxford University Press on behalf of the Endocrine Society.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
Table 1. Case vignette 3: Thyroid function test results in family with raised TSH levels
Daughter Son
Age Birth 7 years 10 days 1 month
TSH mU/L >100 1.2 104 80
RR <10 0.4-4 <10 0.4-4.0
TT4 nmol/L n/a 127 109 106
RR — 55-135 55-135 55-135
Figures in bold denote abnormal values.
Abbreviations: RR, reference range; TSH, thyrotropin; TT4, total thyroxine.
Clinical Cases
Case 1
A 19-year-old woman, with a constellation of symptoms of
thyrotoxicosis (anxiety, palpitations, insomnia), was found to
have abnormal thyroid function tests (TSH 1.8 mU/L
[RR 0.35-5.5], FT4 24 pmol/L [RR 6.3-14] or 1.86 ng/dL [RR
0.48-1.08]), with similar results when her thyroid function
was retested on two further occasions. Her mother, investigated
for fatigue, showed similarly abnormal thyroid function (TSH
3.5 mU/L [RR 0.35-5.5], FT4 24 pmol/L [RR 6.3-14]
or 1.86 ng/dL [RR 0.48-1.08]). Her maternal grandfather
(deceased) was known to have had a thyroid problem of un­
defined nature.
Case 2
A diagnosis of hypothyroidism (TSH 9.7 mU/L [RR 0.35-5.5])
in a 67-year-old man with known type 2 diabetes mellitus and
cardiomyopathy prompted treatment with 100 µg of T4 daily.
On this dose, discordant TFTs (TSH 13.7 mU/L [RR 0.35-5.5],
FT4 69 pmol/L [RR 10.5-21] or 5.36 ng/dL [RR 0.81-1.63]),
led to discontinuation of therapy. Subsequently, a rise in circu­
lating TSH (45.6 mU/L), together with a strongly positive an­
tithyroid peroxidase antibody measurement (>1300 IU/mL
[RR 0-60]), prompted recommencement of T4 (100 µg daily).
Puzzlingly, TFTs after T4 was restarted remained discordant
(TSH 22.1 mU/L [RR 0.35-5.5]; FT4 61 pmol/L [RR
10.5-21] or 4.74 ng/dL [RR 0.81-1.63]).
Case 3
Neonatal screening in a female infant showed TSH >100 mU/L
(RR <10), prompting commencement of thyroxine therapy. Six
years later, her brother was also found to have a raised TSH
(104 mU/L) after birth (day 10), but levothyroxine therapy was
withheld because his circulating total T4 (TT4) (109 nmol/L
[RR 55-135] or 8.46 μg/dL [RR 4.27-10.48]) and thyroid isotope
scan were normal. Subsequent serial measurements recorded a
spontaneous, progressive fall in TSH that normalized by age 18
months (Table 1) and he developed normally. This prompted a
trial of levothyroxine withdrawal in his sister (age 7 years), fol­
lowing which her TFTs (Table 1) and thyroid isotope scan
were normal. Although clinically euthyroid with no goiter, mater­
nal TSH was raised (60 mU/L) with normal TT4 (121 nmol/L,
9.40 μg/dL) and negative thyroid autoantibody (antithyroid per­
oxidase, thyroglobulin, TSH receptor) measurements (Table 1).
Case 4
Investigation of a 64-year-old man with mitral regurgitation,
atrial fibrillation, and impaired left ventricular function
1095
Mother
3 months 6 months 18 months 28 years
37 18 2 60
0.4-4.0 0.4-4.0 0.4-4.0 0.4-4.0
117 131 120 121
55-135 55-135 55-135 55-135
showed elevated circulating free TH levels (FT4 31 pmol/L
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
[RR 9-20] or 2.40 ng/dL [RR 0.69-1.55]; FT3 8.3 pmol/L
[RR 3-7.5] 5.40 pg/mL [RR 1.95-4.88]), with nonsuppressed
TSH concentration (1.2 mU/L [RR 0.4-4.0]). Magnetic reson­
ance imaging visualized a low attenuation abnormality in
the pituitary gland, prompting commencement of somatostatin
receptor ligand (SRL) therapy for a presumed TSH-secreting pi­
tuitary microadenoma. A fall in circulating THs following
treatment with octreotide 50 µg subcutaneously 3 times a day
for 7 days (Table 2) prompted continuation of therapy with
monthly administration of depot SRL (Lanreotide Autogel
90 mg intramuscularly). However, following five months of
this treatment, his TFTs remained abnormal (Table 2).
Approach to Diagnosis
Considering the clinical context is helpful when evaluating a
patient with discordant thyroid function. For instance, object­
ive abnormalities (eg, goiter, clinical signs of hyperthyroidism,
increased uptake on thyroid isotope scan) can suggest that the
observed biochemical hyperthyroidism is genuine. Sedation
with chloral hydrate (often undertaken in children) prior to
blood sampling is a recognized cause of transient, reversible
hyperthyroxinemia (3). Potential confounding factors, includ­
ing altered physiological states (eg, pregnancy), intercurrent
illness or concurrent medication should be considered (also re­
viewed in detail in (2)).
Amiodarone Therapy
Changes in TFTs are almost universal in amiodarone-treated
patients, with 2% to 24% developing overt thyroid dysfunction
(4). The drug inhibits TH biosynthesis and pituitary type 2
deiodinase activity, prompting an initial rise in circulating
TSH (1 week after commencement), which typically normalizes
by 12 weeks (5, 6). Subsequently, the drug blocks hepatic, type
1 deiodinase activity, inhibiting the conversion of T4 to T3 and
breakdown of reverse T3, resulting in a TFT profile of raised/
high normal T4, low/low-normal T3, and markedly elevated re­
verse T3. This pattern may be evident as early as 2 weeks after
commencement of treatment and can persist after drug with­
drawal (5), with the changes in TFTs being dose dependent.
Due to inhibition of T4 to T3 conversion, patients with
amiodarone-induced hypothyroidism may require T4 replace­
ment in supraphysiologic dosage to normalize TSH and there­
fore exhibit concurrently raised FT4 and TSH levels (5, 7).
Nonthyroidal Illness
Prolonged or severe critical illness is often associated with ab­
normal TFTs; typically, circulating FT3 and FT4 fall (usually
1096
Table 2. Case vignette 4: Thyroid function test results in case 4 at baseline, after treatment with octreotide, and after treatment with lanreotide
Baseline Octreotide 50 µg SC tid,
day 4 of administration
TSH mU/L 1.9 1.0
RR 0.4-4.0 0.4-4.0
FT4 pmol/L 35 25
RR 9-20 9-20
FT3 9.6 6.8
RR 3-7.5 3-7.5
Figures in bold denote abnormal values.
Abbreviations: FT3, free triiodothyronine; FT4 free thyroxine; IM, intramuscularly; RR, reference range; SC, subcutaneously; tid, 3 times daily; TSH,
thyrotropin.
to low-normal levels), reverse T3 rises, and TSH values are
low or low-normal. The degree of excursion in hormones is re­
lated to illness severity (6). Recovery is heralded by a rise in
TSH with subsequent normalization of T4 and T3 (8, 9).
Notably, changes in TFTs are not restricted to critical illness,
with a low FT3, raised TSH and FT4 pattern being recorded in
patients undergoing elective abdominal surgery (10) and in
the weeks following myocardial infarction (11).
Acute Psychiatric States
True hypothyroidism or thyrotoxicosis can be associated with
psychiatric abnormalities (12). Conversely, transient hormo­
nal changes (typically elevated circulating FT4 with nonsup­
pressed TSH) are seen in patients (∼15%) admitted with
acute psychiatric states (13, 14), with this pattern being short
lived, reversible, and not associated with a particular psychi­
atric condition (12, 14).
Causes of Artefactually Raised Thyroid
Hormones
Familial Dysalbuminemic Hyperthyroxinemia
This dominantly inherited condition, due to heterozygous
variants (Arg218His, Arg218Ser, Arg218Pro, Arg222Ile) in
circulating albumin which alter its binding affinity for TH,
causes artefactual elevation of FT4 and FT3 measurements
with nonsuppressed TSH levels in euthyroid individuals,
raising the possibility of Familial Dysalbuminemic
Hyperthyroxinemia (FDH) being misdiagnosed as RTHβ or a
TSH-secreting pituitary tumor. The commonest albumin vari­
ant (Arg218His), with a prevalence ranging from 1 in 10 000
in Caucasians to 1 in 100 in Hispanic populations (15), causes
measurement interference in most currently available free TH
immunoassay methods (16). A different variant at this position
in albumin (Arg218Pro) binds steroid as well as iodothyro­
nines, causing artefactual hyperthyroxinemia and hypercorti­
solemia (17). An albumin variant (Leu66Pro) with selectively
increased affinity for T3, causing isolated hypertriiodothyroni­
nemia, has been identified in a Thai kindred (18).
Transthyretinemic Hyperthyroxinemia
Heterozygous variants (Gly26Ser, Ala129Thr, Ala129Val,
Thr139Met) that increase the T4 binding affinity of transthyr­
etin (TTR, also termed thyroxine-binding prealbumin), a pro­
tein that forms circulating, iodothyronine-binding tetramers,
can also cause raised FT4 (and sometimes FT3) measurements
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
Octreotide 50 µg SC tid, After 5 months treatment with
day 7 of administration Lanreotide Autogel 90 mg IM
1.7 1.4
0.4-4.0 0.4-4.0
27 32
9-20 9-20
7.4 9.4
3-7.5 3-7.5
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
in several, current immunoassay platforms (Table 3) (19), thus
mimicking pathological conditions in euthyroid individuals.
Confusingly, the occurrence of hyperthyroxinemia in genetical­
ly confirmed cases can be intermittent, with such variability
being attributed to nonthyroidal illness or other factors affect­
ing formation of heterotetramers by mutant TTR (15).
Thyroxine Binding Globulin Deficiency
As expected, complete or partial deficiency of thyroxine
binding globulin (TBG), due to defects in its gene on the
X-chromosome, causes low circulating TT4 concentrations
in affected males (15). However, although concurrent FT4
measurements are usually normal in these euthyroid individu­
als, this entity can also cause spuriously raised FT4 measure­
ments in some immunoassay methods (Table 3) (20).
Antibody-Mediated Interference
Categorization of antibody mediated assay interference is use­
ful in designing strategies to overcome this rare but clinically
significant issue. Interfering antibodies can either be directed
against the hormone to be measured (autoantibodies), or to
reagents within an immunoassay (antireagent antibodies).
Antireagent antibodies can be further classified into either spe­
cific/high affinity or polyvalent/low affinity (heterophile) types
(21). Due to their polyvalent nature, heterophile antibodies
can cross link both the capture and detection antibodies in a
2-site immunoassay, causing a false positive signal in the ab­
sence of analyte. Rheumatoid factor, which can be classed
as a type of autoimmune heterophile, can cause interference
in immunoassays, including for THs or TSH (22, 23).
Antireagent antibodies (which can be specific or heterophile)
typically bind to the antibodies that either capture or detect
the analyte in an immunoassay, but can also be directed
against other assay elements (eg, the analogue tracer, linkers
joining assay components). Current immunoassay designs
are increasingly sophisticated, containing nonspecific Ig to ad­
sorb anti-Ig antibodies or Fab fragment or chimeric antibodies
(either lacking or with a humanized Fc portion), which reduce,
but do not eliminate, this type of interference. Conversely, the
increasing therapeutic use of monoclonal antibodies, particu­
larly those which are not fully humanized, has the potential to
increase the incidence of antibody interference.
Anti-Iodothyronine Autoantibodies
Antibodies directed at iodothyronines (anti-T4 and anti-T3
autoantibodies), occurring at low prevalence (1.8%) in the
Table 3. Representative examples of entities which cause interference in free thyroid hormone measurements in different immunoassay platformsa

ENTITYb Perkin–Elmer Beckman Abbott Roche Siemens Siemens ADVIA Ortho Fujirebio
DELFIA ACCESS/DXi ALINITY COBAS ATELLICA Centaur VITROS LUMIPULSE

Transthyretinemic TSH 2.19 2.28 1.96 4.10 2.35 2.20 2.34 2.1
Hyperthyroxinemia RR mU/L 0.40-4.0 0.34-5.6 0.35-4.94 0.27-4.20 0.35-5.5 0.35-5.5 0.50-4.88 0.56-4.27
T139M TTR mutation FT4 18.1 21.0 17.0 28.5 25.1 23.0 25.8 31.3
RR pmol/L 9.0-20.0 7.7-15.1 7.5-21.1 12.0-22.0 10.0-19.8 10.5-21.0 9.8-18.9 13.5-24.3
FT3 NA 4.5 3.3 7.6 5.3 4.9 6.1 5.1
RR pmol/L 4.3-6.8 3.8-6.0 3.1-6.8 3.5-6.50 3.5-6.5 2.9-6.8 3.2-6.7
TT4c 167.0
RR nmol/L 69.0-141.0
TBG deficiency TSH 1.03 1.10 0.98 1.19 1.14 1.13 1.18 1.00
F391S TBG mutation RR mU/L 0.40-4.0 0.34-5.6 0.35-4.94 0.27-4.20 0.35-5.5 0.35-5.5 0.50-4.88 0.56-4.27
FT4 28.0 13.6 10.5 19.3 29.3 20.7 21.5 18.2
RR pmol/L 9.0-20.0 7.7-15.1 7.5-21.1 12.0-22.0 10.0-19.8 10.5-21.0 9.8-18.9 13.5-24.3
FT3 NA 3.8 2.9 3.5 4.52 4.2 6.6 3.9
RR pmol/L 4.3-6.8 3.8-6.0 3.1-6.8 3.5-6.5 3.5-6.5 2.9-6.8 2.7-6.0
TT4 44.4
RR nmol/L 69.0-141.0
TBG <3.5
RR 14.0-31.0
Familial Dysalbuminemic TSH 2.06 2.12 1.75 2.09 2.14 2.04 2.03 1.91
Hyperthyroxinemia RR mU/L 0.40-4.00 0.34-5.6 0.35-4.94 0.27-4.20 0.35-5.5 0.35-5.5 0.50-4.88 0.56-4.27
R218H ALB mutation FT4 20.8 39.8 18.4 31.7 32.2 31.9 10.8 31.3
RR pmol/L 9.0-20.0 7.7-15.1 7.5-21.1 12.0-22.0 10.0-19.8 10.5-21.0 9.8-18.9 13.5-24.3
FT3 NA 6.8 6.3 7.7 8.46 8.6 6.9 7.1
RR pmol/L 4.3-6.8 3.8-6.0 3.1-6.8 3.5-6.50 3.5-6.5 2.9-6.8 3.2-6.7
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4

TT4 248.5
RR nmol/L 69.0-141.0
Anti-T4 TSH 5.87 5.75 5.20 6.25 6.10 5.97 6.2 5.59
antibody RR mU/L 0.40-4.0 0.34-5.6 0.35-4.94 0.27-4.20 0.35-5.5 0.35-5.5 0.50-4.88 0.56-4.27
FT4 14.7 15.0 15.8 35.0 100.7 90.8 37.2 47.8
RR pmol/L 9.0-20.0 7.7-15.1 7.5-21.1 12.0-22.0 10.0-19.8 10.5-21.0 9.8-18.9 13.5-24.3
FT3 NA 5.2 5.9 5.7 2.95 3.4 5.5 5.7
RR pmol/L 4.3-6.8 3.8-6.0 3.1-6.8 3.5-6.5 3.5-6.5 2.9-6.8 2.7-6.0
TT4 347.0
RR nmol/L 69.0-141.0

FT3 measurement using the DELFIA method is no longer available. Values for FT4, FT3 and TT4 are provided in SI units. Equations to convert to non SI units are; FT4 (pmol/L)×0.077688 = ng/dL; FT3 (pmol/L)/
1.536 = pg/mL, TT4 (nmol/L)/12.87 = μg/dL.
Figures in bold denote abnormal values.
Abbreviations: NA, not available; RR, reference range; TBG, thyroxine-binding globulin; TT4, total T4.
a
Two-step immunoassay methods are italicized, with other methods being one-step.
b
For each entity, hormone measurements in the same sample from a case were undertaken in different immunoassays.
c
Total T4 was measured by the DELFIA immunoassay method.
1097

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

1098 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
general population but more frequently in patients with
thyroid autoimmunity (especially hypothyroidism), can also
bind labeled analogues of T4 or T3 used in immunoassays,
leading to falsely elevated estimation of hormone values (24).
Typically, assays with a two-step architecture, preventing con­
tact between patient's serum and labeled analogue, are not sus­
ceptible to autoantibody-mediated interference, whereas
one-step measurement methods are affected (25).
Antireagent (Ruthenium) Antibodies
Heterophile antibodies directed against ruthenium, a tracer
used by one manufacturer in electrochemiluminescence im­
munoassays, are a relatively common (up to 0.24%) (26)
cause of falsely raised FT4 and FT3 or falsely low TSH meas­
urements in assays of this type (24). The addition of blocking
agents to ruthenium-based assays has greatly reduced, but not
eliminated, this class of interference (26).
Paraprotein-Associated Iodothyronine Assay
Interference
Spuriously elevated total T3 but normal TSH and FT4 meas­
urements, due to IgG paraprotein–mediated assay interfer­
ence, has been recorded in cases of myeloma (27).
Interference With Assay Architectures
The high affinity of interaction between biotin and streptavi­
din has prompted manufacturers to use this molecular link
as part of the architecture of some, or all, of their immunoas­
says. While intake of 50μg of this B vitamin daily is sufficient
to meet nutritional requirements, higher doses (5 to 30 mg)
are used in neonates with suspected inborn errors of metabol­
ism and in adults with multiple sclerosis. It is also widely mar­
keted as a health supplement. Biotin ingestion in such high
dosage causes interference with measurement of multiple thy­
roid (28) and nonthyroid analytes, with previous Roche,
Beckman and Siemens Immulite immunoassays reported to
be susceptible (29). Falsely elevated THs, with spuriously sup­
pressed TSH and positive anti-TSH receptor antibody test re­
sults can mimic Graves disease (29). To avoid measurement
interference, a recent study recommended withholding biotin
intake for 24 hours prior to TSH, FT4, and FT3 measurement,
or for longer to measure thyroglobulin reliably (30). The bio­
tin–streptavidin link is also a target for antibody interference,
and antibodies directed against streptavidin which interfere
with susceptible TH and TSH assays have been described (31).
Displacement From Binding Proteins
Heparin (polymeric or fractionated, low molecular weight),
when administered intravenously or by subcutaneous injec­
tion, can activate endothelial lipoprotein lipase in vivo, releas­
ing free fatty acids from circulating triglyceride, with such free
fatty acids displacing free THs from binding proteins in vitro.
This artifact is most evident when a blood sample is not proc­
essed for TH measurement for some time after being taken, or
with laboratory assays requiring long incubation periods, and
in patients with hypertriglyceridemia or hypoalbuminemia
(32, 33). Blood sampling more than 10 hours after heparin ad­
ministration and/or laboratory analysis without delay can
minimize the risk of spurious free hyperthyroxinemia. In add­
ition, concurrent normal TT4 values in the same sample are in­
dicative of a displacement effect and can confirm the patient's
euthyroid status. Administration of furosemide in high dosage
(250-500 mg, orally or intravenously) is known to inhibit
interaction of THs with its binding proteins, increasing meas­
urements of FT4 (34), with even modern FT4 assays being vari­
ably susceptible to such interference (35). In such cases, TSH
measurement may assess thyroid status more reliably.
Causes of Spuriously Abnormal TSH
Macro-TSH
Macro-TSH, an autoimmune, TSH–Ig complex of larger mo­
lecular size than TSH itself (36), is detected to varying extents
by different TSH immunoassays, accounting for its variable
(0.6-1.6%) estimated prevalence (24). Due to its high molecu­
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
lar mass, the half-life of circulating macro-TSH is prolonged;
however, as this species is also restricted to the intravascular
compartment, it is thought to be biologically inactive.
Heterophile Anti-TSH Antibodies
As two-site immunometric assays are used almost universally
to detect circulating TSH in clinical practice, they are suscep­
tible to heterophile antibody interference (reported prevalence
0.4%) (37), which typically causes falsely high, or (less com­
monly) falsely low TSH measurements, depending on whether
the antibodies are crosslinking or blocking. In practice, deter­
mining whether a spuriously raised TSH is due to interference
from a crosslinking heterophile antibody or a macro TSH–Ig
complex can be difficult, although once such assay interfer­
ence is detected, the need to make this distinction is less clin­
ically relevant.
Rare Amino Acid Change (Arg75Gly) in TSHβ
Subunit
A single nucleotide change in TSHB, encodes a variant TSHβ
subunit polypeptide containing an amino acid change
(Arg75Gly) that does not alter TSH function, but prevents rec­
ognition of TSH by monoclonal antibodies used in some im­
munoassays. The Arg75Gly TSHβ variant is particularly
prevalent (1%) in South Asian populations, with heterozy­
gotes and homozygotes exhibiting falsely low or even sup­
pressed TSH measurements in some assay platforms (eg,
Siemens ADVIA Centaur, IMMULITE 2000), sometimes re­
sulting in misdiagnosis of hyperthyroidism (38), with other
methods (eg, Roche Elecsys, Abbott ARCHITECT, or
Beckman Coulter DxI) recording normal TSH (39, 40).
Paraprotein-Associated TSH Assay Interference
Rare cases of falsely subnormal (Abbott AxSYM assay) or
raised (Beckman-Coulter DxC 880i assay) TSH measure­
ments due to interference from circulating IgG paraproteins
have been recorded (41, 42), with interference resolving
upon disappearance of the monoclonal gammopathy.
True Biochemical Hyperthyroidism With
Nonsuppressed TSH
Genuinely Raised (Free) T4
Thyroxine therapy, including with poor compliance
Variable compliance with T4 therapy in hypothyroidism can
cause a biochemical pattern of normal or elevated FT4 and
raised TSH concentrations (43). In this context, a seemingly
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4 1099

Table 4. Causes of isolated raised (free) T4 and normal TSH

Disorder Thyroxine replacement Nonthyroidal Acute psychiatric Amiodarone therapy Selenoprotein

therapy illness states deficiencya

FT4 High High High High High

FT3 Normal Normal or low Can be normal Normal Normal or low
TSH Normal or raised Normal Normal Normal Normal
Reverse Normal Raised Not known Very Raised Raised
T3
Clinical Some individuals with Sepsis or other Acute admissions Raised FT4 can be an effect of Growth
context acquired or congenital clinical states with psychosis or amiodarone therapy, with persistent, retardation
hypothyroidism mood disorders isolated elevation not indicative of Muscle
thyroid disease weakness

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

Abbreviations: FT3, free triiodothyronine; FT4 free thyroxine; TSH, thyrotropin.
a
Low plasma selenium levels are characteristic of selenoprotein deficiency due to mutations in SECISBP2 or TRU-TCA1-1.

appropriate increase in thyroxine dosage (perhaps to a supra­
physiological level) can normalize TSH levels but be associ­
ated with concomitantly elevated circulating THs, raising
the possibility of underlying RTHβ (43). Distinguishing this
clinical situation from true coincidence of RTHβ with auto­
immune hypothyroidism (which occurs rarely), is discussed
further below.
Separate to this, it is also recognized that a subset of hypo­
thyroid patients, compliant with thyroxine therapy in physio­
logical dosage, can exhibit raised circulating FT4 but normal
FT3 and TSH concentrations (44, 45). This phenomenon has
been attributed to diminished activity of type 2 deiodinase, re­
ducing the generation of T3 (from its T4 precursor) that is
available to inhibit pituitary TSH secretion (45).
Deficiency of selenocysteine-containing proteins, including
deiodinase enzymes
Over 25 human proteins, including the thyroid deiodinase
enzymes, contain the amino acid selenocysteine (Sec). The incorp­
oration of Sec into selenoproteins during their translation is de­
pendent on a unique cellular pathway that includes its own
transfer RNA (Sec-tRNA(Ser)Sec, encoded by TRU-TCA1-1) and
a protein (SECIS binding protein 2, encoded by SECISBP2),
which binds a specific sequence (SElenium Cysteine Insertion
Sequence [SECIS]) located in the 3′-untranslated region of all
selenoprotein mRNAs. Biallelic mutations in SECISBP2 cause
a rare multisystem disorder, often presenting in childhood
with growth retardation and developmental delay. Diminished
activity of deiodinase enzymes and low circulating selenoproteins
cause a distinctive biochemical signature of raised FT4, normal or
low FT3, normal or raised TSH, elevated reverse T3 and low
circulating selenium concentrations in all patients (Table 4);
this biochemical pattern has also been documented in 2 patients
with selenoprotein deficiency due to a homozygous nucleotide
substitution in TRU-TCA1-1 (46).
Loss of deiodinase enzyme function
In a family with dyshormonogenic congenital hypothyroidism
(CH), heterozygosity for an additional mutation (Arg132His)
in the type 1 deiodinase was associated with hyperthyroxine­
mia and raised reverse T3 concentrations, irrespective of
whether or not they were on thyroxine replacement (47).
Similarly, a combination of homozygosity for a common vari­
ant (Thr92Ala) in the type 2 deiodinase gene, together with
heterozygosity for loss-of-function TSHR variants in two dif­
ferent hypothyroid patients, has been associated with raised
circulating TSH and FT4 but normal total T3 in one individ­
ual and normalization of TSH only after addition of liothyro­
nine therapy in the other patient, suggesting impaired
conversion of T4 to T3 in both cases (48).
Genuinely Raised (Free) T3 and T3/T4 Ratio With
Nonsuppressed TSH
Some acquired or genetic conditions can be associated with
high-normal/high circulating FT3 and low-normal/low FT4
concentrations, computing to a raised T3/T4 ratio, with nor­
mal TSH levels.
Dyshormonogenesis
Iodine deficiency, typically caused by a restrictive diet with in­
adequate intake of this element, can result in low circulating
FT4 with slightly increased FT3 (due to preferential produc­
tion of T3 over T4) and normal TSH concentrations, with
chronic deficiency causing goiter formation (49). Progressive
goiter, normal TSH, and raised circulating FT3 or an elevated
T3/T4 ratio has been reported in the context of mild, dyshor­
monogenic CH due to partial, loss of function mutations in
thyroid peroxidase (50) or thyroglobulin gene defects, with
hypersecretion of T3 being attributed to increased thyroidal
activity of pituitary type 2 deiodinase (51, 52). Increased thy­
roidal deiodinase enzyme activity in cases of metastatic fol­
licular (53) or poorly differentiated thyroid cancer (54) also
mediates higher circulating FT3 concentrations and an ele­
vated T3/T4 ratio.
Resistance to thyroid hormone α
Heterozygous mutations in the thyroid hormone receptor α
gene cause a disorder characterized by features of hypothyroid­
ism, but associated with normal circulating TSH, low-normal/
low FT4, high-normal/high FT3, and a raised T3/T4 ratio (55).
Monocarboxylate transporter 8 deficiency
Mutations in an X-linked gene (SLC16A2), encoding a mem­
brane protein (MCT8) required for transport of THs into the
central nervous system, cause a disorder (Allan–Herndon
Dudley syndrome) with psychomotor retardation, elevated
circulating FT3, low or low-normal FT4, and normal TSH
concentrations (56).
1100 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4

Table 5. Causes of raised (free) T3 and elevated T3/T4 ratio

Disorder Iodine deficiency Mild dyshormonogenic Follicular or poorly RTHαb Monocarboxylate

congenital hypothyroidisma differentiated thyroid transporter 8 deficiencyc
cancer

Occurrence Commoner in some Not known Rare case reports ∼1 in 20 1 in 70 000 males
world regions 000-40 000
FT4 Low-normal or Low Low-normal or Low Low-normal or Low Low-normal or Low-normal or Low
Low
FT3 High-normal or High-normal or High High-normal or High High-normal or High-normal or High
High High
T3/T4 ratio High High High High High
TSH Normald Normald Normal Normald Normald
Reverse T3 Low Not known Low Low or Normal Low

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

Thyroglobulin Raised Raisede Raised Normal Normal
Urinary iodine Low Normal Normal Normal Normal

Abbreviations: FT3, free triiodothyronine; FT4 free thyroxine; RTHα, Resistance to Thyroid Hormone alpha.
a
Due to mutations in genes (eg, TPO, TG) mediating thyroid hormone biosynthesis.
b
Due to mutations in THRA.
c
Due to mutations in SLC16A2.
d
TSH can be slightly raised in severe iodine deficiency, mild dyshormonogenic CH, RTHα or some cases of MCT8 deficiency.
e
Thyroglobulin is not raised in dyshormonogenic CH due to defects in TG.

Table 5 illustrates how entities sharing this pattern of TFTs
can be differentiated, using a combination of clinical features
and key biochemical measurements (eg, urinary iodine, serum
thyroglobulin, and reverse T3).
Genuinely Raised (Free) T4 and (Free) T3 With
Nonsuppressed TSH
After excluding confounding causes (eg, measurement inter­
ference, effects of concomitant comorbidities or drug therap­
ies) outlined above, genuinely raised, circulating FT4 and T3
with nonsuppressed TSH concentrations are compatible
with either a genetic (RTHβ) or acquired (TSH-secreting pitu­
itary tumor or thyrotropinoma, TSHoma) condition. Over
900 families with RTHβ, a dominantly inherited disorder
caused by different (∼230 known), heterozygous mutations
in the thyroid hormone receptor β gene (THRB), have been re­
corded worldwide (57). Greater detection of TSHomas likely
mediates their higher prevalence (2.8 per million) than previ­
ously recorded (58). The age and gender of patients, and
magnitude or pattern of elevation in free THs or TSH concen­
trations does not distinguish between these disorders (59).
Although, in principle, a combination of genetic testing (for
RTHβ) and pituitary imaging (for TSHoma) should readily
differentiate these entities, a number of issues can cause diag­
nostic difficulties, as highlighted below.
Possibly due to sooner ascertainment, a greater propor­
tion (25-30%) of TSHomas now present as microadenomas,
with some pituitary lesions not discernible using standard
magnetic resonance imaging (60) and being associated
with a paucity or even absence of hyperthyroid signs
and symptoms (58, 59, 61). Conversely, clinical features of
hyperthyroidism can also be present in RTHβ (62) and inci­
dental abnormalities on pituitary imaging are found in a sig­
nificant proportion (20%) of genetically-proven RTHβ
cases (59, 60).
Although dynamic endocrine tests (rise in TSH in response
to thyrotropin-releasing hormone; inhibition of TSH follow­
ing T3 administration) can differentiate between RTHβ and
TSHoma, these investigations have limitations. For instance,
thyrotropin-releasing hormone is not available in many coun­
tries. Moreover, although one case series confirmed that the
TSH response is normal or exaggerated in RTHβ but blunted
in TSHoma, with the authors specifying a greater than 5-fold
TSH increase being a cut-off that differentiates these entities
(60), there may be a “gray zone” (3- to 5-fold TSH response
to thyrotropin-releasing hormone) compatible with either
diagnosis (Gurnell, unpublished observations). T3 adminis­
tration is not advisable in elderly individuals or those with
underlying cardiac disease or psychiatric disturbance; further­
more, although circulating TSH is higher in TSHoma than
RTHβ cases following the T3 suppression test, there is an
overlap in TSH values with no discrete cut-off that distin­
guishes between these entities (60).
Circulating levels of sex hormone–binding globulin
(SHBG) are normal in RTHβ (signifying hepatic resistance
to TH action), but elevated in patients with TSHoma with
hepatic hyperthyroidism (63). However, due to inhibition
of its synthesis, SHBG can be normal in patients with mixed
GH/TSH-secreting pituitary tumors or TSHoma with coinci­
dent insulin resistance or in microadenoma cases (59); con­
versely, other factors (oral estrogen therapy, anorexia) can
cause elevated SHBG in RTHβ (2). An elevated molar ratio
of circulating pituitary glycoprotein α-subunit to TSH can
signify TSHoma, but this biomarker can be “falsely” ele­
vated in postmenopausal RTHβ cases or normal in patients
with micro-TSHoma (60).
Although finding genuinely elevated, circulating THs with
nonsuppressed TSH in relatives of an index case is suggestive
of a heritable disorder such as RTHβ, an absence of affected
family members does not exclude the diagnosis, because
15% of RTHβ cases can occur sporadically due to a THRB
mutation arising de novo (64). In 10% to 15% of cases with
clinical and biochemical features consistent with RTHβ,
THRB sequencing shows no abnormality. Here, diagnostic
possibilities include a defect involving an as yet undiscovered
gene mediating TH action, or somatic mosaicism for a THRB
mutation that is not expressed in DNA from peripheral blood
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
mononuclear cells (65). A recent report suggests that next-
generation rather than conventional (Sanger) sequencing of
DNA from patient cells of different embryonic origin is
more sensitive at detecting somatic mosaicism (66).
Apparent Hypothalamic–Pituitary–Thyroid
Axis Resistance
It is well recognized that in some thyroxine-treated patients
with goitrous or dysgenetic CH, but without any additional
defect in genes mediating TH action, circulating TSH remains
raised despite FT4 concentrations being within the normal
range, such that a higher dosage of T4 (with resultant elevated
FT4 concentrations) is required to normalize TSH (67, 68).
This phenomenon has been attributed to CH altering the “set­
point” of the hypothalamic–pituitary–thyroid (HPT) axis dur­
ing development, perhaps epigenetically (68). Previously, such
axis resistance has been shown to lessen with age (69), but a
recent study documented later development of HPT axis re­
sistance in CH cases with previously normal sensitivity to
TH during infancy (70).
Co-occurrence of Different Entities
FDH and Autoimmune Thyroid Disease
When FDH and thyrotoxicosis coexist, a suppressed TSH
is associated with FT4 concentrations that are dispropor­
tionately raised or discordant when measured using differ­
ent methods. Conversely, dramatic elevation in FT4
measurements when TSH is normalized with T4 treatment
can indicate coincidence of FDH and autoimmune hypothy­
roidism (71).
TBG Deficiency and Graves Disease
Investigation of a male patient with low serum TSH and TT4,
but raised FT4 concentrations and thyroid stimulating immu­
noglobulins, identified hemizygosity for a pathogenic mutation
in the TBG (SERPINA7) gene, hence signifying coincident
TBG deficiency and Graves disease (72).
RTHβ and Thyroid Autoimmunity
Graves disease or thyroiditis, with concomitant underlying
RTHβ, is suspected when antithyroid drug treatment of a pa­
tient with conventional features (thyrotoxic symptoms, raised
THs, subnormal or suppressed TSH) results in a marked,
exaggerated rise in TSH concentration in the face of normal,
circulating THs (73). Similarly, having excluded variable com­
pliance, malabsorption or accelerated metabolism of TH (43),
a requirement for thyroxine replacement in supraphysiologi­
cal dosage (above 1.6-1.8 µg/kg body weight) to normalize
circulating TSH in autoimmune hypothyroidism, in conjunc­
tion with raised FT4 and FT3 concentrations, can suggest
underlying, coexistent RTHβ (74).
RTHβ and TSH-Secreting Pituitary Tumor
Although pituitary abnormalities in RTHβ patients are usually
incidental (see above), genuine coincidence of TSH-secreting
pituitary microadenoma with genetically-confirmed RTHβ
was described in a 12-year-old child who had raised free
THs but elevated TSH and blunted responses in a T3 suppres­
sion test (75).
1101
RTHβ and Congenital Hypothyroidism
As discussed above, whilst persistently raised TSH with hyper­
thyroxinemia due to an altered HPT axis setpoint is recog­
nized in thyroxine-treated CH (68), this can rarely be due to
coexistence of CH and RTHβ, with failure to normalize
TSH despite marked elevation in both FT4 and FT3 concen­
trations, or a requirement for levothyroxine in very supraphy­
siologic dosage, being suggestive of a dual disorder (76, 77).
Methods to Detect Analytical Interference in
Assays or Diagnose Entities
The ability to detect assay interference, and to estimate the
true concentration of circulating THs or TSH, will depend
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
on additional methods available in the clinical laboratory.
Some additional tests can be undertaken in a general labora­
tory, whereas other measurements are best performed in a spe­
cialist reference laboratory. As even the simple methods
described here cannot be used on every sample, interpreting
results in conjunction with clinical context (eg, discordance
between TFTs and clinical status of patient) or knowledge
of laboratory systems (eg, an unexpected change in TFT
pattern compared with previous measurements) can be very
helpful. Grossly discordant, widely variable, or extremely de­
ranged results usually raise suspicion of interference; unhelp­
fully, it is more subtle interference, causing plausible
biochemical patterns, which can lead to diagnostic confusion
and possibly unnecessary further investigation or intervention
(24). Furthermore, in some situations, the presence of assay
interference precludes accurate estimation of true TSH or
TH values, requiring management of patients using clinical
thyroid status rather than laboratory measurements.
Simple Methods
Method Comparison
Comparing hormone measurements made using two (or
more), judiciously chosen, comparator assay platforms can
identify most forms of immunoassay interference. For in­
stance, assays using differing capture antibody species or af­
finities, sample dilutions or interference blocking agents, can
expose antireagent or anti-analyte antibody interference.
“Two-step” free TH assays are not susceptible to anti-
iodothyronine antibody mediated interference (25).
Measurement of free THs using assays with differing buffer
or incubation conditions can expose interference due to genet­
ic variants of circulating TH binding proteins (16) (Table 3).
Comparison with an assay architecture that does not use bio­
tin can be used to detect biotin and streptavidin antibody
interference. Current initiatives to harmonize immunoassays
(78) may limit the ability to detect assay interference by ex­
ploiting differences between measurement methods, placing
greater reliance on analytical techniques described below.
Polyethylene Glycol Precipitation
This test relies on the concept that addition of polyethylene gly­
col (PEG) to a serum sample preferentially precipitates either
immunoglobulin-bound analytes or an antireagent antibody
and is widely used to detect these classes of interference (79).
Whilst cost-effective and readily available in most clinical la­
boratories, this method is not infallible and test protocols
need to be carefully controlled. In the absence of antibody
1102 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
interference, PEG can precipitate up to 70% of native TSH,
equating to greater than 30% recovery of this analyte in the
test; in contrast, in cases of significant TSH assay interference,
PEG can precipitate almost all macro-TSH or antireagent anti­
body, equating to less than 5% recovery of native TSH.
However, with this test, there is a gray zone (5-30% TSH recov­
ery) which signifies the presence of assay interference, but
whether the magnitude of such interference has clinical signifi­
cance is uncertain. Furthermore, in cases of TSH assay interfer­
ence, the variability of this test also makes it difficult to
accurately estimate the true TSH value. Elimination of such inter­
ference by incubating serum with protein G or anti-Ig agarose
can confirm the presence of an interfering immunoglobulin (79).
As free TH immunoassays are exquisitely sensitive to ma­
trix effects, the use of a PEG precipitation test to detect
antibody-mediated interference in such assays is not recom­
mended; better methods for measurement of free hormones
(eg, equilibrium dialysis or ultrafiltration) which can over­
come this type of interference are available (see below).
Dilution Studies
When serum containing native TSH is diluted serially, its
measured concentration should fall proportionately; in con­
trast, the presence of antireagent antibody or macro-TSH usu­
ally (but not always) causes a nonlinear reduction in measured
TSH following dilution. As with PEG precipitation, this test
needs to be conducted carefully using an appropriate diluent
to preserve the assay matrix. Similarly, a nonlinear fall in total
TH measurements following serial dilution can signify inter­
ference due to antireagent or anti-iodothyronine antibody.
In contrast, as free THs re-equilibrate with sample dilution,
this test cannot be undertaken using free TH assays (80).
Interference Blocking Agents
Most commercial immunoassays incorporate proprietary re­
agents to neutralize either heterophile or anti-animal anti­
bodies. When serum is assayed in the native (undiluted)
state, high concentrations of interfering antibody may render
these blocking reagents ineffective, generating a false result. In
contrast, when a sample is assayed in dilution, favoring re­
moval of the diluted interfering antibody by blocking agent,
loss of interference can generate an accurate result.
In addition, it is possible to “preabsorb” antibody-mediated
interference using dedicated, commercially available, blocking
tubes or reagents before a sample is immunoassayed. In practice,
this maneuver is a helpful but relatively ineffective way of elimin­
ating antibody interference, possibly because many immunoassays
have already incorporated reagents to block such interference.
More Complex Methods
Direct Methods for Measurement of Free Thyroid
Hormones (Equilibrium Dialysis/Ultrafiltration)
Direct methods, which require physical separation of THs
from their binding proteins prior to measurement of their con­
centration (typically by mass spectrometry) are intrinsically
more accurate than indirect immunoassay methods.
Equilibrium dialysis and ultrafiltration methods, separating
free and protein-bound hormone, are exquisitely sensitive to
analytical conditions which need to be carefully controlled
(81, 82). Consequently, these assays are best undertaken in
reference, rather than routine, clinical laboratories, but remain
methods of choice when interference in free TH immunoassays
is suspected. However, in FDH, even FT4 measured by direct
dialysis can be raised in rare cases (83). It should also be noted
that interference due to displacement of free THs (eg, heparin),
will also affect these methods.
Gel Filtration
A gel filtration method can detect interfering species if they are of
different size to the native analyte and is particularly useful in de­
fining “macro hormone” (eg, hormone–Ig) complexes (36, 84).
This method is labor intensive and best undertaken in a specialist
laboratory. In addition to detecting the macro-hormone complex,
this method can potentially measure the uncomplexed free
hormone—knowledge that could aid patient management.
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
Unfortunately, macro-TSH–Ig complexes are often weakly bound
and can dissociate during gel filtration, often confounding estima­
tion of the true, biologically active, TSH concentration (85).
Radiolabeled Binding Studies
When assayed under specific conditions, the binding of radio­
labeled T4 to serum can detect variant albumin proteins caus­
ing dysalbuminemic hyperthyroxinemia (86). A combination
of radiolabeled T4 and serum electrophoresis can detect ab­
normal binding of this tracer to other circulating T4-binding
proteins (eg, TTR, TBG) (87) with almost complete sensitivity
and specificity. Radiolabeled T4 or T3 bound to antibodies in
serum can be precipitated by PEG, identifying the presence of
anti-iodothyronine antibodies (88, 89). Similarly, radiola­
beled TSH can be used to detect circulating anti-TSH anti­
bodies and this test may be helpful in identifying weak
macro-TSH–-Ig complexes (see above) which are not detected
by gel filtration (85). Although many routine laboratories are
now reluctant to use radiolabeled tracers and readily avail­
able, non-radiolabeled alternatives are yet to be developed,
these techniques are effective screening methods.
Genetic Testing
DNA sequencing to identify variants in genes encoding circu­
lating TH binding proteins (ALB, TBG, TTR) or mediating
TH action (SLC16A2, SECISBP2, TRU-TCA1-1, DIO1,
THRA, THRB), enables definitive diagnosis of genetic entities
and disorders.
In resource-limited settings, we suggest that using particular
patterns of discordant TFTs (as illustrated in Tables 4-6) to
guide and inform the selection of specific candidate genes
for sequencing can be an economical approach, especially
for entities (eg, FDH, transthyretinemic hyperthyroxinemia,
RTHβ) where most cases are associated with a restricted rep­
ertoire of causal or pathogenic genetic variants, localizing to a
few coding exons. However, as costs of this technology con­
tinue to fall, it may become cost-effective to sequence panels
of genes (eg, mediating hyperthyroidism, hypothyroidism)
or analyze the whole exome or genome of an individual.
Irrespective of the approach, it is important to recognize
that gene sequencing can identify variants of unknown signifi­
cance. To avoid misclassification of variants and misdiagnosis
as has been recently highlighted with THRB (90), it is import­
ant to show that the variant genotype cosegregates with ab­
normal thyroid function or phenotype in families. If this is
not possible (eg, in sporadic cases or due to the unavailability
of family members), modeling based on protein structures or
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4 1103

Table 6. Causes of raised (free) T4 and (free) T3 and nonsuppressed TSH

Disorder Transthyretinemic Familial dysalbuminemic Resistance to thyroid hormone TSHoma

hyperthyroxinemia hyperthyroxinemia β

Prevalence 1 in 10-10 000b 1 in 6 000a 1 in 19 000-40 000 1 in 360 000

Etiology TTR mutation ALB mutation THRB mutation Pituitary tumor
FT4 High High High High
FT3 Normal or High Normal or High High High
TSH Normal Normal Normal (or Raised) Normal (or Raised)
SHBG Normal Normal Normal High (or Normal)
Reverse T3 Raised Normal or Raised Raised Raised
Clinical None None Goiter Goiter
features Hyperthyroid features, may be Hyperthyroid features, may be

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

asymptomatic asymptomatic

Abbreviations: FT3, free triiodothyronine; FT4 free thyroxine; TSHoma, thyrotropinoma.

a
Prevalence varies depending on ethnicity.
b
Prevalence of causal TTR variants in UK Biobank.

Figure 1. Molecular (functional) pituitary imaging in a patient with an occult microthyrotropinoma. Standard clinical MRI (coronal T1SE before and after
gadolinium) at diagnosis fails to demonstrate an adenoma. Similarly, volumetric (FSPGR) sequences are unremarkable. However, molecular imaging
with Met-PET coregistered with FSPGR MRI reveals intense focal radiotracer uptake just to the left of the site of insertion of the infundibulum.
Following 3 months of depot SRL therapy, with resultant normalization of thyroid function tests, there is no discernible change in anatomical imaging
findings; however, in marked contrast, there is dramatic diminution in radiotracer uptake, thereby revealing the location of the occult microadenoma
(subsequently confirmed at transsphenoidal surgery). Abbreviations: FSPGR, fast spoiled gradient recalled echo; Gad, gadolinium; Met-PET/MRCR,
11
C-methionine PET/CT coregistered with FSPGR MRI; MRI, magnetic resonance imaging; PET, positron emission tomography; SE, spin echo; SRL,
somatostatin receptor ligand; T3, triiodothyronine; T4, thyroxine; TSH, thyroid-stimulating hormone.

even functional studies of a variant protein to establish its
pathogenicity, may be required (91).
Back to Solving the Cases
Case 1
Further investigation of this patient showed that her circulating
TT4 was raised (249 nmol/L [RR 69-141] or 19.34 μg/dL [RR
5.36-10.95]), but associated with a normal serum TBG
(29.9 μg/mL [RR 14-31]) concentration. Then, using a validated
method (86), we showed that binding of radiolabeled T4 to her se­
rum albumin was increased, raising the possibility of an alterna­
tive binding protein abnormality—FDH. Knowing that FDH is
commonly caused by a restricted repertoire of genetic variants
(15), we sequenced ALB and identified the most common causal
variant (Arg218His) in her case, as well as in her sibling with hy­
perthyroxinemia. Importantly, as we have documented previous­
ly with FDH (16), when her serum was tested using different
immunoassay methods in current use, many recorded falsely
raised FT3 as well as elevated FT4 values (Table 3). This
1104 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4

Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024

Figure 2. Antibody interference in free T4/T3 measurement can be due to anti-iodothyronine or antireagent antibodies, and in TSH measurement due to
anti-TSH (macroTSH) or antireagent antibodies Abbreviations: CH, congenital hypothyroidism; GFC, gel filtration chromatography; PEG, polyethylene
glycol; RTHα, resistance to thyroid hormone α; RTHβ, resistance to thyroid hormone β; SHBG, sex hormone–binding globulin; SRL, somatostatin
receptor ligand.

observation, together with the familial nature of FDH, emphasizes Case 2

the possibility of this entity being misdiagnosed as RTHβ or a After brief discontinuation of T4 to assess the HPT axis in its
TSH-secreting pituitary tumor, leading to unnecessary further in­ natural state, TFTs showed congruent TSH measurements in
vestigation or inappropriate treatment (16). different assays (Perkin-Elmer DELFIA: 43.8 mU/L [RR
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
0.4-4]; Siemens CENTAUR 45.6 mU/L [0.35-5.5]), but dis­
cordant FT4 values using a two-step (Perkin-Elmer DELFIA:
7 pmol/L [RR 9-20] or 0.54 ng/dL [RR 0.69-1.55]) vs one-
step (Siemens CENTAUR 60 pmol/L [RR 10.5-21] or
4.66 ng/dL [RR 0.81-1.63]) immunoassay methods, with con­
firmation of positive antithyroid peroxidase antibody status.
After restarting thyroxine therapy (125 µg daily), testing his
thyroid function using different platforms showed highly con­
cordant, near-normal TSH values, with normal FT4 measure­
ments in two-step assays but variably discordant values in
one-step methods (Table 3). As others have documented pre­
viously (25), this pattern is highly suggestive of measurement
interference due to an anti-iodothyronine antibody. Ongoing
titration of thyroxine replacement in this patient was guided
by measurement of TSH and, if necessary, FT4 using a two-
step immunoassay method.
Case 3
When assayed in serial dilution, TSH values in maternal se­
rum did not fall linearly (neat serum, 22.4 mU/L; actual val­
ues on dilution 1 in 2: 23.2 mU/L; 1 in 4: 19.3 mU/L; 1 in 8:
12.2 mU/L; 1 in 16: 6.4 mU/L; 1 in 32: 3.6 mU/L), raising the
possibility of assay interference. When corrected for the dilu­
tion factor, TSH concentrations appear to increase (neat:
22.4 mU/L; 1 in 2: 46.4 mU/L; 1 in 4: 77.2 mU/L; 1 in 8:
97.6 mU/L; 1 in 16: 102.4 mU/L; 1 in 32: 115.2 mU/L)—a
pattern highly suggestive of a weakly bound macro-TSH–Ig
complex which dissociates on dilution. As we have docu­
mented in an unrelated case (92), gel filtration studies showed
that immunoreactive TSH eluted at a higher molecular mass
peak which was not present after adsorption with protein
G-Sepharose, consistent with maternal serum containing a
TSH–IgG complex. We surmise that transplacental passage
of this antibody, causing interference in neonatal TSH
screening, led to misdiagnosis of CH in both her children.
Progressive disappearance of this interfering antibody from
their circulation likely accounts for spontaneous reduction in
(younger child) or normalization of (older sibling) TSH meas­
urements in her children with time. A recent report docu­
mented the presence of macro-TSH in 0.43% of neonates
screened (85).
Case 4
Although administration of a short-acting SRL (octreotide)
lowered circulating THs in this patient, as documented previ­
ously (93) this response does not distinguish between RTHβ
and thyrotropinoma. On the other hand, the inability of
long-acting (depot) SRL treatment to lower TH concentrations
was highly suggestive of a diagnosis of RTHβ (93). THRB se­
quencing, identifying a heterozygous mutation (Arg438His)
known to cause the disorder, confirmed this diagnosis. Thus,
in patients with genuinely elevated free THs and nonsup­
pressed TSH, this case illustrates how coincidental anomalies
on pituitary imaging can be misleading, and also the utility of
dynamic investigation with administration of long-acting
SRL in differential diagnosis. Indeed, as we have documented
previously (94), combining long-acting SRL administration
with 11C-methionine PET pituitary imaging (Fig. 1), enables
localization of TSH-secreting pituitary microadenomas that
have eluded detection with conventional magnetic resonance
imaging.
1105
Conclusions
We have reviewed patterns of discordant thyroid function
(raised [free] T4 and/or [free] T3 and nonsuppressed TSH)
due to entities causing interference with measurement meth­
ods or disorders associated with genuinely altered hormone
concentrations. While we have illustrated the susceptibility
of most current TH immunoassay methods to measurement
interference, we recognize that this is a dynamic situation.
Specifically, with manufacturers continually changing the
architecture and biochemical conditions of their measurement
methods, it is quite possible that their susceptibility to some
types of interference will be eliminated, whilst exposing vul­
nerability to new or different interference mechanisms.
In our overall approach to evaluation of raised THs and
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
nonsuppressed TSH, we have outlined an algorithm for differ­
ential diagnosis (Fig. 2), which we hope is both economical
and applicable in resource-limited settings.
Finally, with one review suggesting that over 50% of cases
with discordant thyroid function were associated with mis­
diagnosis, inappropriate investigation and management
(24), the potential for preventing unnecessary costs and ad­
verse health outcomes via a structured approach to this bio­
chemical entity cannot be underestimated.
Acknowledgments
We thank colleagues in other laboratories (Norfolk &
Norwich University Hospital, East & North Herts NHS
Trust, University Hospital Southampton, UK) for method
comparison studies.
Funding
Our research is supported by the Wellcome Trust (210755/Z/18/
Z to K.C.; 219296/Z/19/Z to N.S.), Medical Research Council
(MRC_MC_UU_00014/40), and the NIHR Cambridge
Biomedical Research Centre (N.S., M.G., K.C.).
Disclosures
The authors (C.M., N.S., D.H., S.O., G.L., S.v.B., M.G., K.C.)
have nothing to disclose and no conflicts of interest in relation
to this manuscript.
Data Availability
Data sharing is not applicable to this article as no datasets
were generated or analyzed during the current study.
References
1. Gurnell M, Halsall DJ, Chatterjee VK. What should be done when
thyroid function tests do not make sense? Clin Endocrinol (Oxf).
2011;74(6):673-678.
2. Koulouri O, Moran C, Halsall D, Chatterjee K, Gurnell M. Pitfalls
in the measurement and interpretation of thyroid function tests.
Best Pract Res Clin Endocrinol Metab. 2013;27(6):745-762.
3. Buck ML. The use of chloral hydrate in infants and children.
Paediatr Pharmacother. 2005;11:1-9.
4. Trohman RG, Sharma PS, McAninch EA, Bianco AC. Amiodarone
and thyroid physiology, pathophysiology, diagnosis and manage­
ment. Trends Cardiovasc Med. 2019;29(5):285-295.
5. Martino E, Bartalena L, Bogazzi F, Braverman LE. The effects of
amiodarone on the thyroid. Endocr Rev. 2001;22(2):240-254.
1106 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
6. Melmed S, Nademanee K, Reed AW, Hendrickson JA, Singh BN,
Hershman JM. Hyperthyroxinemia with bradycardia and normal
thyrotropin secretion after chronic amiodarone administration. J
Clin Endocrinol Metab. 1981;53(5):997-1001.
7. Figge J, Dluhy RG. Amiodarone-induced elevation of thyroid
stimulating hormone in patients receiving levothyroxine for pri­
mary hypothyroidism. Ann Intern Med. 1990;113(7):553-555.
8. Hamblin PS, Dyer SA, Mohr VS, et al. Relationship between thyro­
tropin and thyroxine changes during recovery from severe hypo­
thyroxinemia of critical illness. J Clin Endocrinol Metab.
1986;64(4):717-722.
9. Langouche L, Jacobs A, Van den Berghe G. Nonthyroidal illness
syndrome across the ages. J Endocr Soc. 2019;3(12):2313-2325.
10. Michalaki M, Vagenakis AG, Makri M, Kalfarentzos F,
Kyriazopoulou V. Dissociation of the early decline in serum T3 con­
centration and serum IL-6 rise and TNFα in nonthyroidal illness
syndrome induced by abdominal surgery. J Clin Endocrinol
Metab. 2001;86(9):4198-4205.
11. Wiersinga WM, Lie KI, Touber JL. Thyroid hormones in acute
myocardial infarction. Clin Endocrinol (Oxf). 1981;14(4):
367-374.
12. Radhakrishnan R, Calvin S, Singh JK, Thomas B, Srinivasan K.
Thyroid dysfunction in major psychiatric disorders in a hospital
based sample. Indian J Med Res. 2013;138(6):888-893.
13. Arem R, Cusi K. Thyroid function testing in psychiatric illness: use­
fulness and limitations. Trends Endocrinol Metab. 1997;8(7):
282-287.
14. Lambert TJ, Davidson R, McLellan GH. Euthyroid hyperthyroxi­
naemia in acute psychiatric admissions. Aust N Z J Psychiatry.
1987;21(4):608-612.
15. Pappa T, Ferrara AM, Refetoff S. Inherited defects of thyroxine-
binding proteins. Best Pract Res Clin Endocrinol Metab.
2015;29(5):735-747.
16. Khoo S, Lyons G, McGowan A, et al. Familial dysalbuminaemic hy­
perthyroxinaemia interferes with current free thyroid hormone im­
munoassay methods. Eur J Endocrinol. 2020;182(6):533-538.
17. Moran C, Seger C, Taylor K, et al. Hyperthyroxinemia and
hypercortisolemia due to familial dysalbuminemia. Thyroid.
2020;30(11):1681-1684.
18. Sunthornthepvarakul T, Likitmaskul S, Ngowngarmratana S, et al.
Familial dysalbuminemic hypertriiodothyroninemia: a new, domin­
antly inherited albumin defect. J Clin Endocrinol Metab.
1998;83(5):1448-1454.
19. Dieu X, Sueur G, Moal V, et al. Apparent resistance to thyroid hor­
mones: from biological interference to genetics. Ann Endocrinol
(Paris). 2019;80(5-6):280-285.
20. Moeller LC, Appiagyei-Dankah Y, Köhler B, Biebermann H,
Janssen OE, Führer D. Two novel mutations in the Serpina7 gene
are associated with complete deficiency of thyroxine-binding globu­
lin. Eur Thyroid J. 2015;4(Suppl. 1):108-112.
21. Frost SJ. More on heterophile and human anti-animal antibodies.
Clin Chem. 1999;45(11):2042-2043.
22. Norden AG, Jackson RA, Norden LE, Griffin AJ, Barnes MA, Little
JA. Misleading results from immunoassays of serum free thyroxine
in the presence of rheumatoid factor. Clin Chem. 1997;43(6):
957-962.
23. Georges A, Charrie A, Raynaud S, Lombard C, Corcuff J-B.
Thyroxin overdose due to rheumatoid factor interferences
in thyroid-stimulating hormone assays. Clin Chem Lab Med.
2011;49(5):873-875.
24. Favresse J, Burlacu M-C, Maiter D, Gruson D. Interferences with
thyroid function immunoassays: clinical implications and detection
algorithm. Endoc Rev. 2018;39(5):830-850.
25. Zouwail SA, O’Toole AM, Clark PM, Begley JP. Influence of
thyroid hormone autoantibodies on 7 thyroid hormone assays.
Clin Chem. 2008;54(5):927-928.
26. Sapin R, Agin A, Gasser F. Efficacy of a new blocker against anti-
ruthenium antibody interference in the elecsys free triiodothyronine
assay. Clin Chem Lab Med. 2007;45(3):416-418.
27. Mandal K, Ashorobi D, Lee A, Liao H, Kumar SC, Rosenthal DS.
Factitiously elevated total triiodothyronine in a euthyroid patient
with multiple myeloma. Case Rep Endocrinol. 2021;2021:
8479193.
28. Arya VB, Ajzensztejn M, Appleby G, et al. High-dose biotin in in­
fants mimics biochemical hyperthyroidism with some commercial
assays. Clin Endocrinol (Oxf). 2018;88(3):507-510.
29. Elston MS, Sehgal S, Du Toit S, Yarndley T, Conaglen JV.
Factitious graves’ disease due to biotin immunoassay interference-
A case and review of the literature. J Clin Endocrinol Metab.
2016;101(9):3251-3255.
30. Ylli D, Soldin SJ, Stolze B, et al. Biotin interference in assays for thy­
roid hormones, thyrotropin and thyroglobulin. Thyroid.
2021;31(8):1160-1170.
31. Ricci V, Esteban MP, Sand G, Menises MM. Interference of anti-
streptavidin antibodies: more common than we thought? In relation
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
to six confirmed cases. Clin Biochem. 2021;90:62-65.
32. Mendel CM, Frost PH, Kunitake ST, Cavalieri RR. Mechanism of
the heparin-induced increase in the concentration of free thyroxine
in plasma. J Clin Endoc Metab. 1987;65(6):1259-1264.
33. Stevenson HP, Archbold GP, Johnston P, Young IS, Sheridan B.
Misleading serum free thyroxine results during low molecular
weight heparin treatment. Clin Chem. 1998;44(5):1002-1007.
34. Newnham HH, Hamblin PS, Long F, Lim C-F, Topliss DJ, Stockigt
JR. Effect of oral frusemide on diagnostic indices of thyroid func­
tion. Clin Endocrinol (Oxf). 1987;26(4):423-431.
35. Hawkins RC. Furosemide interference in newer free thyroxine as­
says. Clin Chem. 1998;44(12):2550-2551.
36. Loh TP, Kao SL, Halsall DJ, et al. Macro-thyrotropin: a case report
and review of literature. J Clin Endocrinol Metab. 2012;97(6):
1823-1828.
37. Ismail AA, Walker PL, Barth JH, Lewandowski KC, Jones R, Burr
WA. Wrong biochemistry results: two case reports and observational
study in 5310 patients on potentially misleading thyroid-stimulating
hormone and gonadotropin immunoassay results. Clin Chem.
2002;48(11):2023-2029.
38. Drees JC, Stone JA, Reamer CR, et al. Falsely undetectable TSH in a
cohort of south Asian euthyroid patients. J Clin Endocrinol Metab.
2014;99(4):1171-1179.
39. Pappa T, Johannesen J, Scherberg N, Torrent M, Dumitrescu A,
Refetoff S. A TSHβ variant with impaired immunoreactivity but in­
tact biological activity and its clinical implications. Thyroid.
2015;25(8):869-876.
40. Shaki D, Eskin-Schwartz M, Hadar N, et al. TSHB r75g is a founder
variant and prevalent cause of low or undetectable TSH in Indian
Jews. Eur Thyroid J. 2022;11(1):e210072.
41. Luzzi VI, Scott MG, Gronowski AM. Negative thyrotropin assay
interference associated with an IgGk paraprotein. Clin Chem.
2003;49(4):709-710.
42. Imperiali M, Jelmini P, Ferraro B, et al. Interference in thyroid-
stimulating hormone determination. Eur J Clin Invest. 2010;
40(8):756-758.
43. Morris JC. How do you approach the problem of TSH elevation in
a patient on high-dose thyroid hormone replacement. Clin
Endocrinol (Oxf). 2009;70(5):671-673.
44. Jonklaas J, Bianco AC, Bauer AJ, et al. Guidelines for the treatment
of hypothyroidism: prepared by the American thyroid association
task force on thyroid hormone replacement. Thyroid. 2014;24(12):
1670-1751.
45. Jonklaas J, Davidson B, Bhagat S, Soldin SJ. Triiodothyronine levels
in athyreotic individuals during levothyroxine therapy. JAMA.
2008;299(7):769-777.
46. Schoenmakers E, Chatterjee K. Human genetic disorders resulting
in systemic selenoprotein deficiency. Int J Mol Sci. 2021;22(23):
12927.
47. Furman A, Hannoush Z, Echegoyen FB, Dumitrescu A, Refetoff S,
Weiss RE. Novel DIO1 gene mutation acting as phenotype modifier
for novel compound heterozygous TPO gene mutations causing
congenital hypothyroidism. Thyroid. 2021;31(10):1589-1591.
The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
48. Park E, Jung J, Araki O, et al. Concurrent TSHR mutations and
DIO2 T92A polymorphism result in abnormal thyroid hormone
metabolism. Sci Rep. 2018;8(1):10090.
49. Sasidharan Pillai S, Quintos JB. Thyroid function tests of iodine de­
ficiency goiter can mimic thyroid hormone resistance alpha. J
Pediatr Endocrinol Metab. 2022;35(11):1433-1436.
50. Zhang J, Han R, Shen L, et al. Mild TPO deficiency characterized
by progressive goiter and normal serum TSH level. Endocrine.
2020;68(3):599-606.
51. Kanou Y, Hishinuma A, Tsunekawa K, et al. Thyroglobulin gene
mutations producing defective intracellular transport of thyro­
globulin are associated with increased thyroidal type 2 iodothyro­
nine deiodinase activity. J Clin Endocrinol Metab. 2007;92(4):
1451-1457.
52. Hermanns P, Refetoff S, Sriphrapradang C, et al. A clinically eu­
thyroid child with a large goiter due to a thyroglobulin gene defect:
clinical features and genetic studies. J Pediatr Endocrinol Metab.
2013;26(1-2):119-123.
53. Kim BW, Daniels GH, Harrison BJ, et al. Overexpression of type 2
deiodinase in follicular carcinoma as a cause of low circulating free
thyroxine levels. J Clin Endocrinol Metab. 2003;88(2):594-598.
54. Okazaki-Hada M, Maruoka A, Yamamoto M, et al. Poorly differ­
entiated thyroid carcinoma coexisting with Graves’ disease involv­
ing T3 thyrotoxicosis due to increased D1 and D2 activities.
Thyroid. 2021;31(10):1592-1596.
55. Erbas IM, Demir K. The clinical spectrum of resistance to
thyroid hormone alpha in children and adults. J Clin Res Pediatr
Endocrinol. 2021;13(1):1-14.
56. Groeneweg S, van Geest FS, Abaci A, et al. Disease characteristics of
MCT8 deficiency: an international, retrospective, multicentre co­
hort study. Lancet Diabetes Endocrinol. 2020;8(7):594-605.
57. Moran C, Schoenmakers N, Visser WE, Schoenmakers E, Agostini
M, Chatterjee K. Genetic disorders of thyroid development, hor­
mone biosynthesis and signalling. Clin Endocrinol (Oxf).
2022;97(4):502-514.
58. Gurnell M, Bashari WA, Senanayake R, MacFarlane J, Koulouri O.
Thyroid-stimulating hormone-producing pituitary tumours. In:
Paul Robertson R, Giudice LC, Grossman AB, et al., eds.
DeGroot's Endocrinology. Vol. 1. 8th ed. Elsevier; 2022:145–154.
59. Beck-Peccoz P, Persani L, Lania A, et al. Thyrotropin-secreting pi­
tuitary adenomas. In: Feingold KR, Anawalt B, Blackman MR, et
al., eds. Endotext. Updated 2022 Oct 13. MDText.com Inc.;
2000. https://www.ncbi.nlm.nih.gov/books/NBK278978/.
60. Campi I, Covelli D, Moran C, et al. The differential diagnosis of dis­
crepant thyroid function tests: insistent pitfalls and updated flow-
chart based on a long-standing experience. Front Endocrinol
(Lausanne). 2020;11:432.
61. Socin HV, Chanson P, Delemer B, et al. The changing spectrum of
TSH-secreting pituitary adenomas: diagnosis and management in
43 patients. Eur J Endocrinol. 2003;148(4):433-442.
62. Beck-Peccoz P, Chatterjee VK. The variable clinical phenotype in thy­
roid hormone resistance syndrome. Thyroid. 1994;4(2):225-232.
63. Beck-Peccoz P, Roncoroni R, Mariotti S, et al. Sex hormone-
binding globulin measurement in patients with inappropriate secre­
tion of thyrotropin (IST): evidence against selective pituitary thy­
roid hormone resistance in nonneoplastic IST. J Clin Endocrinol
Metab. 1990;71(1):19-25.
64. Adams M, Matthews C, Collingwood TN, Tone Y, Beck-Peccoz P,
Chatterjee K. Genetic analysis of 29 kindreds with generalized and
pituitary resistance to thyroid hormone. Identification of thirteen
novel mutations in the thyroid hormone receptor beta gene.
J Clin Invest. 1994;94(2):506-515.
65. Mamanasiri S, Yesil S, Dumitrescu AM, et al. Mosaicism of a thy­
roid hormone receptor-beta gene mutation in resistance to thyroid
hormone. J Clin Endocrinol Metab. 2006;91(9):3471-7347.
66. Donnars A, Leplat A, Grosheny C, et al. Clinically symptomatic re­
sistance to thyroid hormone β syndrome because of THRB gene
mosaicism. J Clin Endocrinol Metab. 2022;107(9):e3548-e3552.
1107
67. Kempers MJ, van Trotsenburg AS, van Tijn DA, et al. Disturbance
of the fetal thyroid hormone state has long-term consequences for
treatment of thyroidal and central congenital hypothyroidism. J
Clin Endocrinol Metab. 2005;90(7):4094-4100.
68. Russo M, Gullo D, Tumino D, Leonardi D, Malandrino P, Frasca F.
Different FT3/TSH correlation in acquired and congenital hypothy­
roid patients reveals a different hypothalamic set-point. Clin
Endocrinol (Oxf). 2023;98(1):117-122.
69. Fisher DA, Schoen EJ, La Franchi S, et al. The hypothalamic-
pituitary-thyroid negative feedback control axis in children with
treated congenital hypothyroidism. J Clin Endocrinol Metab.
2000;85(8):2722-2727.
70. Radetti G, Rigon F, Salvatoni A, et al. Tissue sensitivity to thyroid
hormones may change over time. Eur Thyroid J. 2022;11(2):
e210054.
71. Khoo S, Lyons G, Solomon A, et al. Familial dysalbuminemic hy­
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024
perthyroxinemia confounding management of coexistent auto­
immune thyroid disease. Endocrinol Diabetes Metab Case Rep.
2020;2020:19-0161.
72. Berger HR, Creech MK, Hannoush Z, Watanabe Y, Kargi A, Weiss
RE. A novel mutation causing complete thyroid binding globulin
deficiency (TBG-CD MIA) in a male with coexisting Graves disease.
AACE Clin Case Rep. 2017;3(2):e134-e139.
73. Wang TWM, Chatterjee VKK. Clinical case: case No. 29—answer.
The Association of Clinical Biochemists News Sheet. 1996;398:22.
74. Fukata S, Brent GA, Sugawara M. Resistance to thyroid hormone in
Hashimoto's thyroiditis. N Engl J Med. 2005;352(5):517-518.
75. Teng X, Jin T, Brent GA, Wu A, Teng W, Shan Z. A patient with a
thyrotropin-secreting microadenoma and resistance to thyroid hor­
mone (P453T). J Clin Endocrinol Metab. 2015;100(7):2511-2514.
76. Salas-Lucia F, França MM, Amrhein JA, Weir JE, Dumitrescu AM,
Refetoff S. Severe resistance to thyroid hormone Beta in a patient
with athyreosis. Thyroid. 2022;32(3):336-339.
77. Lauffer P, Bikker H, Garrelfs MR, et al. Defective levothyroxine re­
sponse in a patient with dyshormonogenic congenital hypothyroid­
ism caused by a concurrent pathogenic variant in thyroid hormone
receptor-β. Thyroid. 2021;31(11):1757-1762.
78. Thienpont LM, Van Uytfanghe K, Van Houcke S, et al. A progress
report of the IFCC committee for standardization of thyroid func­
tion tests. Eur Thyroid J. 2014;3(2):109-116.
79. Fahie-Wilson M, Halsall D. Polyethylene glycol precipitation: pro­
ceed with care. Ann Clin Biochem. 2008;45(3):233-235.
80. Stockigt JR. Free thyroid hormone measurement. A critical apprais­
al. Endocrinol Metab Clin North Am. 2001;30(2):265-289.
81. Thienpont LM, Van Uytfanghe K, Poppe K, Velkeniers B.
Determination of free thyroid hormones. Best Pract Res Clin
Endocrinol Metab. 2013;27(5):689-700.
82. Gu J, Soldin OP, Soldin SJ. Simultaneous quantification of free tri­
iodothyronine and free thyroxine by isotope dilution tandem mass
spectrometry. Clin Biochem. 2007;40(18):1386-1391.
83. Refetoff S, Scherberg NH, Yuan C, Wu W, Wu Z, McPhaul M. Free
thyroxine concentrations in sera of individuals with familial dysal­
buminemic hyperthyroxinemia: a comparison of three methods of
measurement. Thyroid. 2020;30(1):37-41.
84. Hattori N. The frequency of macroprolactinemia in pregnant wom­
en and the heterogeneity of its etiologies. J Clin Endocrinol Metab.
1996;81(2):586-590.
85. Hattori N, Aisaka K, Yamada A, Matsuda T, Shimatsu A.
Prevalence and pathogenesis of macro-thyrotropin in neonates:
analysis of umbilical cord blood from 939 neonates and their moth­
ers. Thyroid. 2023;33(1):45-52.
86. Stewart MF, Ratcliffe WA, Roberts I. Thyroid function tests in pa­
tients with familial dysalbuminaemic hyperthyroxinaemia (FDH).
Ann Clin Biochem. 1986;23(Pt 1):59-64.
87. Hay ID, Klee GG. Thyroid dysfunction. Endocrinol Metab Clin
North Am. 1988;17(3):473-509.
88. Desai RK, Bredenkamp B, Jialal I, Omar MA, Rajput MC, Joubert
SM. Autoantibodies to thyroxin and triiodothyronine. Clin Chem.
1988;34(5):944-946.
1108 The Journal of Clinical Endocrinology & Metabolism, 2024, Vol. 109, No. 4
89. Teti C, Nazzari E, Galletti MR, et al. Unexpected elevated free thyroid
hormones in pregnancy. Thyroid. 2016;26(11):1640-1644.
90. Korwutthikulrangsri M, Dosiou C, Dumitrescu AM, Refetoff S.
A novel G385E variant in the cold region of the T3-binding do­
main of thyroid hormone receptor beta gene and investigations
to assess its clinical significance. Eur Thyroid J. 2019;8(6):
293-297.
91. Refetoff S. Suggested guidelines for preparation of brief reports on
new cases of resistance to thyroid hormone. Thyroid. 1994;4(1):
139.
92. Halsall DJ, Fahie-Wilson MN, Hall SK, et al. Macro thyrotropin-IgG
complex causes factitious increases in thyroid-stimulating hormone
screening tests in a neonate and mother. Clin Chem. 2006;52(10):
1968-1969.
93. Mannavola D, Persani L, Vannucchi G, et al. Different responses to
chronic somatostatin analogues in patients with central hyperthy­
roidism. Clin Endocrinol (Oxf). 2005;62(2):176-181.
94. Gillett D, Senanayake R, MacFarlane J, et al. Localization of
TSH-secreting pituitary adenoma using 11C-methionine image
subtraction. EJNMMI Res. 2022;12(1):26.
Downloaded from https://academic.oup.com/jcem/article/109/4/1094/7439666 by guest on 26 March 2024