An Adult Female With Resistance to Thyroid Hormone Mediated by Defective Thyroid Hormone Receptor α

Context:

The first human cases (female, age 6 y; father and daughter, ages 47 and 11 y, respectively) with growth retardation/short stature, skeletal dysplasia, constipation, and defective thyroid receptor α (TRα) have been recently described.

Objective:

A 45-year-old, short, overweight female with cognitive impairment, epilepsy, and constipation was investigated.

Design and Intervention:

Clinical, biochemical, and radiological assessment and THRA sequencing were undertaken. The patient's thyroid status and her biochemical and physiological parameters were evaluated at baseline and after T₄ therapy.

Results:

The patient exhibits disproportionate short stature, macrocephaly, low free T₄/free T₃ ratio and rT₃ levels, together with subnormal heart and basal metabolic rate. She is heterozygous for a novel frameshift/premature stop (Ala382ProfsX7) THRA mutation, generating a mutant TRα with constitutive corepressor binding and negligible coactivator recruitment, which inhibits its wild-type counterpart in a dominant-negative manner—both in vitro and in mutation-containing patient blood mononuclear cells studied ex vivo. Her alertness and constipation responded to T₄ therapy, which readily suppressed TSH levels, raised basal metabolic rate, and normalized elevated muscle creatine kinase, but cardiac parameters (heart rate, contractility) remained relatively refractory. The patient and a previous childhood case showed reduced red cell mass with macrocytosis unresponsive to T₄ therapy.

Conclusions:

Clinical (short stature, macrocephaly, constipation) and biochemical (low free T₄/free T₃ ratio, subnormal rT₃) findings that are congruent with previous cases and newly recognized features (epilepsy) in this adult female with defective TRα define a shared phenotype in TRα-mediated resistance to thyroid hormone, with differential tissue responses to T₄ treatment.

The effects of thyroid hormones (THs) on physiological processes are mediated by receptors encoded by separate genes (THRA, TRα1; and THRB, TRβ1, TRβ2) that regulate gene expression in target tissues. TRα1 is most highly expressed in the central nervous system, myocardium, gastrointestinal tract, skeletal muscle, cartilage, and bone; TRβ1 is the predominant receptor subtype in liver and kidney, with TRβ2 expression being limited to hypothalamus, pituitary, cochlea, and retina (1).

Resistance to TH (RTH), due to heterozygous mutations in TRβ (TRβ-mediated RTH), manifests with elevated TH and nonsuppressed TSH levels; dominant negative inhibition of wild-type receptor by mutant TRβ forms the basis of the disorder (2).

Previously, a 6-year-old child with growth and developmental retardation, skeletal dysplasia, and delayed intestinal transit but only borderline-abnormal thyroid function tests was found to be heterozygous for a premature stop mutation in THRA, generating a dominant negative TRα mutant lacking its carboxyterminal transactivation (AF-2) domain (3); subsequently, an analogous TRα defect, also disrupting the AF-2 domain, was described in an 11-year-old child with growth retardation, delayed bone development, and constipation inherited from her father, who has a similar phenotype (4).

Here, we describe the first adult female with a novel dominant negative TRα mutation. Her phenotype includes previously identified (short stature, macrocephaly, constipation) and newly recognized (eg, epilepsy) clinical features and biochemistry (low free T₄ [FT4]/free T₃ [FT3] ratio, subnormal rT₃), that are in common with other described cases. With baseline and serial measurement of physiological and biochemical parameters after T₄ therapy, we have documented differential tissue responses to TH in this disorder.

Patients and Methods

Case description

A 45-year-old female with obesity, epilepsy, and learning disability was investigated.

Neonatal and infancy

The patient was the fourth child of healthy parents and apparently normal siblings (all now deceased). Ten days postpartum, intermittent cyanosis and constipation were noted. At 3 months she presented with macroglossia, sleepiness, and poor feeding. Coarse facial features, an umbilical hernia, bradycardia, slow-relaxing reflexes, and abnormal lower femoral epiphysis on x-ray were noted. These features and low serum protein-bound iodine (PBI; 3.8 μg/100 mL; normal, 5.3–7.3) (5) suggested congenital hypothyroidism; T₄ (0.05 mg three times daily) was commenced. At 5 months she was hospitalized with restlessness, anorexia, and irregular tachycardia (120 bpm); tachycardia improved after stopping T₄ briefly, and it was restarted (0.05 mg twice daily). At age 3.2 years, raised PBI (14.8 μg/100 mL) and advanced bone age prompted T₄ discontinuation; T₄ was restarted at age 6 years for 7 months. T₄ (initially 0.1 mg daily) was recommenced at age 10 years and continued until adulthood.

Childhood

The patient had global developmental delay, walking at 3 years and with a limited (50 word) vocabulary at 4 years. At age 6 years she spoke in full sentences with a slow, nasal, and dysarthric quality, but she could not read. Hearing was normal, but she was hypermetropic. Poor coordination and “clumsiness” were noted. In the Wechsler Adult Intelligence Scale (WAIS) III, she scored below the first centile for processing visual and verbal information; her IQ was measured at 52, and an unusually placid affect was noted. She had five grand mal seizures before age 2 years. Seizures recurred at 20 years. Electroencephalogram showed bilateral theta waves during hyperventilation, supporting a diagnosis of epilepsy. Sodium valproate treatment was commenced, but when discontinued 4 years later, seizures recurred; restarting sodium valproate reduced seizure frequency (one fit in the last decade).

Growth, adolescence, and adult life

Height velocity improved with T₄ treatment (off treatment at age 10 years, height was 9 cm below the 0.4 centile; at age 15 years, after continuous T₄ treatment, height was 3 cm below the 0.4 centile). Menarche occurred at age 16 years, with menses becoming heavy and irregular later. Ultrasound and hysteroscopy showed normal uterine cavity and endometrium; her menorrhagia is controlled with a progestogen-releasing intrauterine device. Gonadotropin (LH, 2.1–9.3 U/L; FSH, 2.8–13.7 U/L) and estradiol (101–522 pmol/L) levels were normal. As an adult, she is cognitively impaired and lives in a care home. Her affect remains placid, but she is lethargic and poorly motivated, with previous episodes of depression and hallucinations. She has chronic constipation, with consequent rectal prolapse and hemorrhoids.

Methods

All investigations were part of an ethically approved protocol and/or clinically indicated, being undertaken with consent from the patient and her next of kin.

Serial biochemical (thyroid function tests, markers of bone turnover) and physiological (resting energy expenditure [REE], echocardiographic indices, spectral analysis of heart rate variability [HRV]) measurements were made in the patient. Molecular genetic analysis of THRA and functional characterization of mutant TRα was undertaken. Detailed methods are provided in Supplemental Data (published on The Endocrine Society's Journals Online web site at http://jcem.endojournals.org).

Results

Clinical investigation

The patient's height is 148 cm (−2.34 SD score [SDS]), and her weight is 98.3 kg (body mass index, 45 kg/m²). She is disproportionately short, with a normal sitting height (+0.29 SDS) but severely reduced subischial leg length (−3.87 SDS) (Figure 1A). She has coarse facial features and skin tags (Figure 1B). Her head circumference (65 cm) is markedly increased (approximately +9 SDS; Figure 1C) (6); a skull radiograph confirmed macrocephaly, thickening of the vault bones (cranial hyperostosis), and prominent frontal bone (Figure 2A). Skeletal survey showed normal femoral epiphyses (Figure 2B), together with cortical thickening in her long bones, which are shortened (Figure 2C). Bone mineral density (BMD) was normal at the hip (BMD, 1.043 g/cm²; total hip T score, +0.4) and lumbar spine (BMD, 1.086 g/cm²; T score, −0.9). Abdominal imaging was consistent with constipation (Figure 2D).

Figure 1.

Phenotypic features of the patient. A, A photograph of the patient illustrates short stature (−2.34 SDS) that is disproportionate, with normal (+0.29 SDS) sitting height but reduced (−3.87 SDS) subischial leg length. B, Her facies are coarse, with full lips, short nose with anteverted nares, a flattened nasal bridge, long philtrum, midface hypoplasia and multiple skin tags. C, Her head circumference (red symbol) for height and gender is markedly increased.

Figure 2.

Radiological and biochemical investigation of patient. A, A lateral skull radiograph of the patient shows thickened skull vault (black arrow) and frontal bone prominence (white arrow). B, Pelvic radiograph shows normal femoral epiphyses. C, Cortical thickening in her long bones (eg, ulna, arrow) is present. D, Abdominal radiograph (arrow) shows colonic dilatation with faecal loading. E, Comparison of FT4/FT3 ratio in the patient with hormone ratios in control females (n = 82, age 40–50 y; horizontal bar represents the mean).

Off T₄ treatment, her FT4 and FT3 levels were normal, but with a subnormal FT4/FT3 ratio (Figure 2E) and low rT₃ levels (Table 1). She was bradycardic (sleeping heart rate, 51 bpm) with low REE and raised skeletal muscle creatine kinase (CK) isoenzyme (CK-MM) levels.

Table 1.

Biochemical and Metabolic Measurements in the Patient

Variable	Daily Dose of T4				Reference Values
	Nil	75 μg	100 μg	125 μg
REE	−3.35	−3.07	−2.61	−1.47	Z scorea
CK-MM, U/L	387	358	191	111	26–192
TSH, mU/L	5.8	0.6	0.04	<0.03	0.35–5.5
FT4, pmol/L	10.0	14.6	14.7	19.6	10–19.8
Total T4, nmol/L	85	120	121	138	69–141
FT3, pmol/L	4.9	6.2	7.7	9.6	3.5–6.5
Total T3, nmol/L	1.7	2.2	2.7	3.3	0.9–2.8
rT3, ng/dL	10	19	18	22	11–32
SHBG, nmol/L	48.3	44.2	50	64.9	14.1–82.9b
Total cholesterol, mmol/L	5.6	4.7	4.6	4.6
LDL cholesterol, mmol/L	3.7	3.11	2.92	2.8
IGF-I, nmol/L	9.2	10.9	9.5	7.9	7.5–30
Copper, μmol/L	12.5	16.3	16.5	18.2	14.2–24.4b
Selenium, μmol/L	0.93	0.88	0.85	0.85	0.76–1.44b
Bone turnover markers
Formation
Bone-specific alkaline phosphatase, ng/mL	11.3	11.7	11.7	13.8	9.1–27.5
Osteocalcin, ng/mL	12.52	19.25	18.77	24.55	6.8–26.5
P1NP, ng/mL	17.2	25.2	30.37	75.73	16.2–60.9
Resorption
CTx, ng/mL	0.23	0.3	0.3	0.68	0.1–0.62
NTx/Cr, nmol BCE/mmol Cr	28.4	25.9	28.5	51.13	13.9–55.9
RBC mass, ×1012/L	3.51	3.53	3.51	3.5	3.8–5.3
MCV, fL	107.1	107	105	106.2	80–100
Hemoglobin, g/dL	12.7	12.6	11.5	11.7	11.5–16

Abbreviations: P1NP, procollagen type 1 N propeptide; CTx, C-terminal cross-linking telopeptide of type 1 collagen; NTx/Cr, N-terminal cross-linking telopeptide of type 1 collagen to creatinine ratio; MCV, mean corpuscular volume; BCE, bone collagen equivalents.

^a

Her REE is expressed as a Z score relative to measurements in 135 healthy control subjects, as described in Supplemental Data.

^b

Values are from 24 healthy control subjects who were matched with the patient for age, gender, and body mass index.

Molecular genetic studies

THRA sequencing indicated heterozygosity for a nucleotide deletion (c1144delG), shifting the reading frame at codon 382, which alters six subsequent residues and then introduces a premature stop codon (Ala382ProfsX7), deleting 22 carboxyterminal amino acids; the mutation does not affect other transcripts (TRα2, Rev-erbα) generated from the THRA locus (Supplemental Figure 1) and is not present in normal genome databases (dBSNP, 1000 Genomes, NHLBI exome server).

Functional properties of Ala382ProfsX7 mutant TRα

Ala382ProfsX7 mutant TRα1 is unable to bind radiolabeled T₃ (Figure 3A) or to mediate hormone-dependent transactivation (Figure 3B); however, the mutant receptor retains DNA binding, with repression of basal reporter gene activity (Figure 3B, inset). Structural modeling suggests that the Ala382ProfsX7 removes a carboxyterminal α-helix, exposing a hydrophobic cleft that accommodates corepressor, facilitating its recruitment (Figure 3C); conversely, deletion of this helix also removes residues required for coactivator recruitment (Figure 3D). Consonant with this, in protein-protein interaction assays, Ala382ProfsX7 mutant TRα1 fails to dissociate from corepressor (Figure 3C) and recruit coactivator (Figure 3D) in a T₃-dependent manner. Furthermore, when coexpressed, Ala382ProfsX7 mutant TRα1 inhibits hormone-dependent transactivation by WT receptor (Figure 3E); this correlates with markedly reduced basal and T₃-induced expression of known TH-responsive target genes (KLF9, HR) (7, 8) in patient-derived peripheral blood mononuclear cells (Figure 3F).

Figure 3.

Functional properties and molecular modeling of A382PfsX7 TRα. A, Radiolabeled T₃ binding to in vitro translated proteins in the absence (white bars) or presence (black bars) of excess (100 nm) unlabeled hormone is shown, with specific hormone binding to wild-type TRα1 but not A382PfsX7 mutant TRα or unprogrammed reticulocyte lysate. The inset shows an EMSA with comparable interaction of unliganded or hormone-bound WT/RXR and A382PfsX7 mutant TRα/RXR heterodimers with a direct repeat thyroid response element from the malic enzyme gene. B, JEG-3 cells were transfected with empty, WT, or A382PfsX7 TRα expression vectors together with a TH responsive reporter gene, assaying T₃-dependent activation; the inset shows basal reporter gene activity in the absence of hormone. C, Crystallographic modeling of unliganded TRα ligand binding domain (LBD), highlighting (in red) the region deleted by the A382PfsX7 mutation, favoring recruitment of corepressor peptide (yellow). D, Modeling of TRα LBD bound to T₃ (blue) and coactivator (green) shows that the A382PfsX7 deletion removes residues critical for hormone binding and coactivator interaction. Protein-protein interaction assays in JEG-3 cells cotransfected with a reporter gene (UASTKLUC), vectors expressing GAL4-NCoR (C) or GAL4-SRC1 (D) fusion proteins and constructs expressing a heterologous activation domain linked to TRs (VP16-WT, VP16–A382PfsX7). Western blotting (inset) confirmed comparable VP16-receptor fusion expression. E, Dominant negative inhibition was tested in cells cotransfected with reporter gene and equal combinations of epitope-tagged vectors, with receptor quantitation by Western blotting (inset). F, Quantitative RT-PCR (internal control: 36B4, acidic ribosomal phosphoprotein) showing expression of KLF9 and HR in peripheral blood mononuclear cells from the patient or control subjects with or without T₃ (100 nm) exposure.

Response to T₄ therapy

The patient was treated with increasing doses of T₄ over 6 months, with serial biochemical and physiological measurements.

Thyroid axis

Despite T₄ treatment in comparatively low dosage (1–1.3 μg/kg), TSH levels suppressed readily, with elevated FT3 and high-normal FT4 levels (Table 1).

Metabolic

Her baseline REE was low and rose with T₄ therapy (Table 1), with negligible change in body composition.

Cardiac function

Echocardiographic indices were evaluated before and after T₄ therapy, with comparison to parameters in hypothyroid, control, or thyrotoxic patients (Figure 4, A–E). At baseline, heart rate, systolic (pre-ejection period [PEP]) and diastolic (ratio passive E wave/active A wave velocity through the mitral valve or E/A ratio) parameters and cardiac index were in the hypothyroid range; after T₄ therapy, most parameters (heart rate, cardiac index, left ventricular ejection fraction [%], PEP) remained within either hypothyroid or control subject ranges, despite concomitant elevated TH and suppressed TSH levels (Table 1).

Figure 4.

Cardiac contractility and autonomic parameters in patient. A–E, Comparison of echocardiographic indices in the patient before (yellow triangle) and after (green triangle) T₄ treatment, with ranges for these parameters in hypothyroid, control, and thyrotoxic subjects. F, Spectral analysis shows time (RMSSD) and LF or HF domains of HRV in controls (open bars, n = 7 females; mean age, 47 y) or the patient after T₄ treatment (black bars, 0–125 μg). Error bars represent SEM.

Spectral analysis of HRV computed three parameters: root mean square successive differences (RMSSD) of cardiac intervals, a robust estimate of HRV and time domain measure of cardiac parasympathetic tone; and the high frequency (HF) (reflecting the vagal component), and low frequency (LF) domain (reflecting both sympathetic and vagal components) measures of cardiac autonomic tone. Baseline RMSSD and HF were significantly raised in the patient compared to controls, and both parameters fell progressively with T₄ treatment (Figure 4F).

Peripheral markers of TH action

Baseline SHBG levels in the patient (48.3 nmol/L) were higher than controls (37 nmol/L) and rose (64 nmol/L) with T₄ treatment; total and low-density lipoprotein (LDL) cholesterol fell with T₄ therapy; serum copper levels rose but selenium levels were unchanged with T₄ treatment. Her low-normal serum IGF-I did not change with T₄ therapy. Initially raised CK-MM levels normalized after T₄ therapy (Table 1). Baseline levels of bone turnover markers were normal. Despite correction of baseline vitamin D insufficiency (14.6 nmol/L; deficiency < 25 nmol/L) and elevated PTH (87 ng/mL; normal range, 14–72) levels with cholecalciferol supplementation (after supplementation vitamin D 89.5 nmol/L; PTH, 35 ng/mL), her bone turnover markers rose progressively, with some (osteocalcin, N-telopeptide of type I collagen) being high-normal and other markers (procollagen type 1 N-propeptide, C-telopeptide cross-linked collagen type I) becoming frankly elevated (Table 1).

Hematological phenotype

Our patient showed mild reduction in red cell mass with raised mean corpuscular volume (Table 1), but normal white blood cell count and slightly reduced platelet count (Supplemental Table 1). Reticulocyte count, hematinic indices (ferritin, B12, folate), and hemolytic parameters (haptoglobin, lactate dehydrogenase, bilirubin) and erythropoietin levels (Supplemental Table 1) are normal. Notably, our childhood mutant TRα case (3) also has low red blood cell (RBC) mass, raised mean corpuscular volume, and normal indices (Supplemental Table 1).

Discussion

Clinical features in our patient (eg, hoarse cry, macroglossia, umbilical hernia, growth retardation, constipation) suggested hypothyroidism but were dissociated from overt thyroid dysfunction with either variable PBI levels in childhood or only marginally abnormal TSH levels in adult life. However, similar to previous cases (3, 4), her circulating FT4/FT3 ratio is low, with subnormal rT₃ levels. She is heterozygous for a nonfunctional TRα mutation, which is a potent dominant negative inhibitor of WT receptor action in vitro; such dominant negative activity in vitro is mirrored by markedly reduced TH target gene responses in mutation-containing patient peripheral blood mononuclear cells studied ex vivo.

Skeletal features in our adult patient are recognized in congenital hypothyroidism (9) and resemble findings in our previous childhood case (3): she is macrocephalic, possibly due to associated cranial hyperostosis, as has been described in other disorders (10). Although her adult epiphyses are normal, abnormal femoral epiphyses noted in infancy may have resolved after TH treatment because this manifestation is related to degree of hypothyroidism (11). Likewise, the characteristic wormian bone appearance of cranial sutures in hypothyroidism, noted in our childhood TRα case (3), can disappear by age 5 years (12), possibly explaining its absence in our adult patient. Consistent with disproportionate short stature, her long bones are shortened; cortical thickening (sclerosis) is recognized in congenital hypothyroidism, but the mechanism is unclear (13). Overall, the skeletal phenotype in our patient is very similar to TRα mutant (TRa1PV) mice (14), highlighting the importance of TRα in bone development; whether skeletal dysplasia in particular sites reflects regional variation in TRα importance remains to be elucidated.

Reduced RBC mass in our adult female and previous childhood case accords with mild anemia documented in two other patients with defective TRα (15). Macrocytic anemia and hypocellular bone marrow are recognized in hypothyroidism (16). Erythropoiesis is compromised in TRα null mice (17), and reduced hematocrit was noted in TRα1PV mutant mice (18). Taken together, it seems plausible to link the anemia in our patients with underlying loss of TRα function.

Despite intermittent TH therapy, childhood neurodevelopmental delay was documented in our patient with marked cognitive impairment. Her impaired motor coordination and placid effect resemble neurocognitive features of our childhood case (3). Her history of childhood seizures, persisting as chronic epilepsy into adult life, is distinctive and may be linked to the underlying TRα defect. TRα1L400R mutant mice exhibit spontaneous generalized seizures, particularly after photic stimulation (Ref. 19; and F. Flamant, personal communication); susceptibility to audiogenic seizures was noted in different, TRα1R384C mutant mice, with TH treatment either exacerbating (female) or preventing seizures (males) (Ref. 20; and B. Vennstrom and B. Cannon, personal communication). These mice also exhibit locomotor ataxia with aberrant development of cortical GABAergic interneurons (21); perhaps similar deficiency of inhibitory neuron activity in the human context mediates propensity to seizures. TRα1-PV mutant mice exhibit impaired fertility in both genders (22); however, known transmissibility of the TRα defect from father to daughter (4) suggests that affected human males are fertile; and our adult female patient has normal cyclical ovarian function.

After T₄ therapy, small increases in FT4 and FT3 readily suppressed TSH levels in our patient, suggesting preservation of TH sensitivity within the hypothalamic-pituitary feedback axis. Likewise, her SHBG rose, and total/LDL cholesterol levels fell with T₄ treatment, suggesting retention of hepatic TH responsiveness. Serum copper levels rose after TH treatment in our patient, consistent with synthesis of ceruloplasmin, the major circulating copper-transport protein, being TRβ-regulated in both mice and humans (23); in contrast, her serum selenium levels did not change with T₄ treatment, analogous to a lack of induction of selenium levels by TH in TRα1R384C mutant mice (24). Interestingly, similar to the previous adult male, mutant TRα case (15), low-normal baseline IGF-I levels in our patient did not change after T₄ treatment, whereas IGF-I synthesis in both children with TRα defects was TH responsive (5, 22). Raised baseline CK-MM levels in our patient fell with T₄ treatment; slightly raised CK levels had been noted (226 IU/L at age 1 y; 254 IU/L at age 2 y) in our previous childhood case (3), and T₄ withdrawal caused CK elevation in the older child with a TRα defect (15). Elevated CK-MM is recognized in hypothyroidism (25), suggesting that skeletal muscle is refractory to TH action in our patient.

Baseline echocardiographic systolic and diastolic parameters are in the hypothyroid range in our patient. Although these parameters normalized after T₄ treatment, some indices (heart rate, PEP, cardiac index) did not reach the thyrotoxic range, suggesting that cardiac responsiveness to TH may be blunted. Baseline spectral analysis in our patient showed raised RMSSD and HF components, indicating increased parasympathetic (vagal) tone. In contrast, reflecting increased sympathetic activity and reduced vagal tone in the disorder, RMSSD and HF are known to be reduced in human hyperthyroidism (26, 27); this observation also suggests baseline hypothyroid, cardiac autonomic status in our patient. Our patient's REE, an aggregate index of metabolic rate in tissues including liver, skeletal muscle, and myocardium, was subnormal at baseline with an attenuated response to T₄ therapy. Elevation in bone turnover markers after TH treatment in our patient likely reflects enhanced osteoblast and osteoclast activity. In this context, TRβ expression has been documented in human osteoblasts (28), and serum ICTP (C-terminal cross-linked telopeptide of type 1 collagen generated by matrix metalloproteinases) levels are known to be normal in TRβ-mediated RTH compared to hyperthyroid (primary thyrotoxicosis, TSHoma) patients (29), suggesting that some effects of TH on the adult skeleton may be TRβ-mediated.

T₄ therapy has had a beneficial effect in our patient, raising basal metabolic rate and improving her constipation and general alertness, perhaps justifying its continuation. However, the changes in bone turnover markers seen after T₄ treatment may preclude its use in higher dosage, and careful monitoring for reduction in BMD is warranted. This adverse effect suggests that treatment with a TRα-specific agonist (30) might represent a more rational therapeutic approach.

Acknowledgments

Our research is supported by funding from the Wellcome Trust (100585/Z/12/Z, to N.S.; and 095564/Z/11/Z, to K.C.), the National Institute for Health Research Cambridge Biomedical Research Centre (to C.M. and M.G.), and the Great Ormond Street Children's Charity (to M.D.).

Disclosure Summary: The authors have nothing to disclose.

^*

C.M. and N.S. are co-equal first authors.

Abbreviations

• BMD

  bone mineral density

• CK

  creatine kinase

• CK-MM

  skeletal muscle isoenzyme of CK

• FT3

  free T₃

• FT4

  free T₄

• HF

  high frequency

• HRV

  heart rate variability

• LDL

  low-density lipoprotein

• LF

  low frequency

• PBI

  protein-bound iodine

• PEP

  pre-ejection period

• RBC

  red blood cell

• REE

  resting energy expenditure

• RMSSD

  root mean square successive differences

• RTH

  resistance to TH

• SDS

  SD score

• TH

  thyroid hormone.