*6.1.1 Thyroid function tests*

All patients with suspected thyrotoxicosis should undergo detailed thyroid function tests, including TSH, free thyroxine (FT4) and preferably total triiodothyronine (T3). Measurement of total T3 is preferred over free T3 (FT3) as the FT3 assays are less robust and less validated as compared to the FT4 assays. Measurement of free hormones makes the assay reports more reliable in the presence of conditions affecting the concentrations of thyroid binding globulin (TBG) like liver disease, nephrotic syndrome. Serum TSH is more sensitive the changes in thyroid hormone levels due to the log-linear relationship of TSH with the free thyroid hormone levels.

Assays have been constantly evolving from the older manual radioimmunoassays (RIA) to the modern day fully automated chemiluminescent immunoassay (CLIA) and electrochemiluminescent immunoassay (ECLIA) platforms. Results have to interpreted in the light of age and assay-specific reference ranges. Lack of consistency between biochemical reports and clinical presentation should alert the physician to consider assay interferences like presence of heterophile antibodies and excess biotin consumption.

Thyroid function tests typically reveal elevated free thyroxine (FT4) and T3 with a suppressed TSH in overt thyrotoxicosis. Patients with milder thyrotoxicosis may exhibit only elevated T3 values and suppressed TSH with normal T4 values, a state known as T3 toxicosis, a common presentation in pediatric age groups. Higher T3 values have been noted in prepubertal children as compared to post-pubertal children, and have been observed to be a negative predictor of the likelihood of remission in pediatric GD [25, 32].

## *6.1.2 Markers of thyroid autoimmunity*

TRAbs are specific for GD, and are found in majority of the patients at diagnosis, with a reported prevalence of >70% in pediatric GD in some studies [33, 34]. TRAb levels can be higher in younger patients <5 years of age and in clinically severe disease [2].

Traditionally, TRAb concentrations were measured using either bioassays or receptor assays. Estimation of TRAb levels by bioassays was based on the ability of TRAbs to increase cyclic adenosine monophosphate (cAMP) production either directly from thyroid follicular cells in vitro, or indirectly via TSH-R transfected Chinese hamster ovary (CHO) cells. These also enabled detection of functional subtypes of TRAbs, including inhibitory TRAb antibodies. Receptor assays, on the other hand, provide an estimate of TRAb levels by measuring the ability of TRAb to inhibit the binding of labelled TSH to thyroid membranes, and provided enhanced sensitivity [1].

TRAb assays subsequently evolved with the advancements in immunoassays techniques. The "liquid phase" first generation immunoassays were competitive immunoassays based on inhibition of binding of radionuclide or enzyme labelled TSH. These provided excellent specificity of 97.5–100%, but suffered from a suboptimal sensitivity ranging from 52 to 94%. Subsequently, 2nd generation "solid phase" immunoassays utilized monoclonal antibodies and human or porcine TSH-R immobilized on a solid surface, improving the sensitivity to 87–100%. The 3rd generation immunoassays utilized a stimulating biotinylated monoclonal antibody

(M22) to bind to immobilized TSH-R. In recent years, fully automated platforms have been developed using the ECLIA and fluoroenzymatic immunoassay principles, with excellent sensitivity of 95–100% and specificity of 97–100% [35].

GD may also be associated with other anti-thyroid antibodies like thyroid peroxidase (TPO) autoantibodies, anti-thyroglobulin antibodies, as well as other autoantibodies like antimicrosomal antibodies (AMA) and antinuclear antibodies (ANA) [33].

#### **6.2 Thyroid imaging**

#### *6.2.1 Thyroid scintigraphy*

Thyroid scans utilize a gamma camera to provide a planar image, which provides anatomical information in addition to the functional status, whereas uptake scans are used to measure the % radioiodine uptake (RAIU), typically measured by placing a non-imaging gamma scintillation probe detector over the neck. Most commonly used radiopharmaceuticals for nuclear thyroid imaging are iodine-123 (I-123) and 99 m-technetium (Tc99m-pertechnetate). Drugs which can potentially interfere the tracer uptake like thionamides and sources of excess iodine have to be stopped atleast 3–7 days and 2–4 weeks prior to the study respectively. The doses for diagnostic imaging should be weight-based rather than fixed doses especially in children. The scintigraphic images and uptake studies are typically performed around 4 hours after I-123 intake, or 20 minutes after the iv injection of Tc-99 m pertechnetate. While technetium scans result in a low radiation exposure, they result in a high background noise and a low range of normal uptake in the thyroid gland [36].

GD is characterized by a homogenous increase in tracer uptake in the thyroid scans and an increased %RAIU on uptake studies. These studies are not done routinely for diagnosis in pediatric GD. Their primary role is in differential diagnosis of thyrotoxicosis in cases with inconclusive clinical and biochemical findings, and in assessing the radioiodine uptake in order to calculate the dose of therapeutic I-131 when radioiodine ablation is planned as therapy. However, germline TSH-R activating mutations can also give rise to a diffuse and homogenously increased uptake in the thyroid gland. On the other hand, McCune Albright syndrome, TA and TMNG are typically associated with focal increases in uptake with suppressed uptake in the remainder of the gland, whereas autoimmune thyroiditis, iodine excess and thyrotoxicosis factitia are associated with decreased to absent uptake in the thyroid gland [22, 37].

#### *6.2.2 Thyroid ultrasonography*

A thyroid ultrasound and doppler study provides a safe and non-invasive modality for differential diagnosis of thyrotoxicosis. Thyroid gland is classically diffusely enlarged in GD, and may display normal echogenicity or hypoechogenicity like in thyroiditis. GD is characterized by a diffuse increase in parenchymal vascularization, often referred to as a "thyroid inferno". Autoimmune thyroiditis may be associated with a lesser degree of increase in parenchymal vascularity as well. Quantitative measurements of the thyroidal blood flow can also be vital in diagnosing GD, with superior and inferior thyroidal artery's mean peak systolic velocities of more than 45–50 cm/second suggestive of a diagnosis of GD, providing a sensitivity and specificity of 81–83% and 92–96% respectively.

Ultrasound should also be performed in the case of thyroid asymmetry or a palpable nodule [2, 18, 38–40].
