The normal thyroid examined by ultrasound shows the parenchyma to be finely granulated (ground-glass appearance), and the echogenicity is similar to the salivary glands but of higher echogenicity compared to the adjacent strap muscles or sternocleidomastoid muscle [6] (Fig. 6.6). HT can be identified with ultrasound by hypoechoic changes in the thyroid (Figs. 6.5b and 6.6b) prior to thyroid dysfunction or elevation of the TPO antibody [14, 15]. Autoimmune thyroid disease may change the thyroid volume (normal values: women 12 mL, men 14 mL) and echogenicity of the thyroid gland. Raber quantified thyroid echogenicity into five grades (Table 6.3) and correlated the degree of change with thyroid function and TPO antibody levels [15]. The extent of echogenicity decrease accurately predicted autoimmune thyroiditis and thyroid dysfunction with a 96% probability. The positive predictive values (PPV) of grade 3 pattern (thyroid parenchyma is iso- or hypoechoic compared to strap muscles) for detection of autoimmune thyroiditis were 94% (with overt hypothyroidism) and 96% (with any degree of hypothyroidism). The negative predictive value (NPV) of grade 1 pattern (thyroid echogenicity similar to submandibular gland, hyperechoic to strap muscles) for detection of euthyroid and TPO antibody negative subjects was 91%. In this study, goiter (larger than normal) or atrophy of the gland did not correlate with TPO antibody status.

The sonographic appearance of HT is extremely variable and depends on the degree of autoimmune involvement and the resulting patient-specific amount of follicular destruction, lymphocytic infiltration, and fibrosis. Figure 6.5 illustrates the ultrasound changes associated with the histopathology (follicular destruction, lymphocyte infiltration, and fibrosis) of HT. Early in the course of thyroiditis, the gland may be normal in size with a diffuse fine micronodular change, with increased diffuse vascularity by Doppler, with short echogenic fibrous bands within the parenchyma, and with an overall normal echogenicity (Figs. 6.7 and 6.8). Frequently, even with early HT and normal thyroid function, there may be benign reactive nodes located around the isthmus and inferior to each thyroid lobe (Figs. 6.7 and 6.9). The micronodular change in a normal size thyroid gland (Figs. 6.7, 6.8, and 6.9) or an enlarged thyroid (Fig. 6.10) with a diffuse increase in vascular flow by CDI can be seen before the overall thyroid echogenicity is significantly reduced and is an important signal that the patient has autoimmune thyroiditis.

As thyroiditis progresses, the gland enlarges and becomes more hypoechoic (Figs. 6.10 and 6.11). The entire gland increases in size with HT including the isthmus and pyramidal lobe. The enlarged pyramidal lobe may appear to be a nodule on transverse imaging (Figs. 6.12c and 6.13), but if the isthmus is scanned in transverse view from superior to inferior,the apparent mass (Fig. 6.12b) will be contiguous with the isthmus to create the HT pseudonodule type 1 (Table 6.4) without a distinct nodule margin.

As HT progresses, the thyroid ultrasound develops more obvious fibrous bands, and increased parenchymal vascularity is seen with Doppler color imaging (Fig. 6.14). The thyroid parenchyma with HT becomes progressively coarser with patchy hypoechoic micronodular change (Fig. 6.14). The micronodular changes are poorly defined hypoechoic patches measuring 1–7 mm in size that can become confluent resulting in the appearance of pseudonodules (HT pseudonodule type 2) of the relatively normal isoechoic tissue surrounded by the hypoechoic micronodules (Fig. 6.15 and Table 6.4) [14, 16–18]. The hypoechoic patches can be relatively large (up to 7 mm) in a regular pattern that resembles the spots on a giraffe. The “giraffe” pattern has been reported as “innumerable” small hypoechoic nodules, but this is the type 3 pseudonodule (Figs. 6.9, 6.10, 6.16, and 6.17 and Table 6.4). The micronodular change is highly specific for HT with a PPV 95% [16]. The amount of fibrosis is variable with HT. Some HT glands have extensive fibrosis in the thyroid parenchyma. The fibrosis is seen on ultrasound as hyperechoic strands that often have vascular flow along them by CDI (Figs. 6.18 and 6.19). Fibrous bands may separate the thyroid parenchyma giving a false appearance of a nodule. Generally, when the area of concern is examined in orthogonal directions, a discrete nodule is not seen. The hypoechoic parenchyma of HT allows visualization of the hyperechoic thyroid capsule separating a normal anatomic variant, the tubercle of Zuckerkandl. The tubercle of Zuckerkandl is normal thyroid tissue derived from the fourth branchial cleft and the ultimobranchial body extending posteriorly from the posterior thyroid (Fig. 6.20). The tubercle of Zuckerkandl can be mistaken for a thyroid nodule on US imaging, but it is an important caveat that thyroid nodules are not surrounded by hyperechoic fibrous bands (Figs. 6.21 and 6.22). Care should be taken to examine the tubercle of Zuckerkandl for a true nodule that can be distinguished by a difference in echogenicity from the thyroid lobe, well-formed nodular margins, and peri-nodular vascular flow (Fig. 6.23).

The micronodular change and overall hypoechogenicity observed on ultrasound is thought to reflect lymphocytic infiltration [1, 19]. Diffuse decrease in thyroid echogenicity (Figs. 6.6b, 6.12, 6.13, 6.21, and 6.23) has a PPV for AITD of 88.3% [95% CI, 85–91%]. Hashimoto’s thyroiditis can be detected by ultrasound before clinical disease or the presence of serum TPO antibodies [14]. A prospective study demonstrated that diffuse hypoechogenicity was a better sensitivity (100% vs 63.3%) at predicting current or future thyroid dysfunction than TPO antibody level [20]. However, the ability of observers to reliably detect degrees of thyroid hypoechogenicity is poor (generalized κ = −0.002–0.781) and depends on the US equipment, equipment settings, and experience of the observer [21, 22]. One attempt to normalize detection is to use histogram analysis that does not depend on equipment or US settings. The degree of hypoechogenicity assessed by this standardized method called the histogram analysis of gray scale is associated with higher TSH levels [21, 23].

Clinically, hypothyroidism can be categorized as goitrous or atrophic thyroiditis (Ord’s thyroiditis ; Fig. 6.24) [9, 24]. Although it is assumed that the US changes of thyroiditis starts with an enlarged gland with hypoechoic micronodules, progresses to the diffusely hypoechoic gland with fibrous bands, and ends with a small hypoechoic atrophic gland, this progression has never been specifically documented. Recent investigations suggest that HT is a multifaceted disease including various subtypes which exhibit distinct clinicopathological characteristics [25]. Several subclassification schemes of HT have been introduced [11]. The fibrous variant of HT is regarded as end stage of HT, but IgG4 thyroiditis may overlap the so-called fibrous variant of HT to a large extent because of the close histologic similarities [25, 26] (Fig. 6.25). Recently, the enlarged, very hypoechogenic gland (Figs. 6.6b, 6.11, and 6.12) has been associated with a new syndrome of IgG4-related sclerosing systemic disease. IgG4-related sclerosing disease is a systemic disease that is characterized by elevated gamma globulins with a predominant increase in IgG4 levels with lymphoplasmacytic infiltration, fibrosis, obliterative phlebitis, and increased numbers of IgG4-positive plasma cells in affected organs [27]. It has suggested there are four types of IgG-related thyroid disease including IgG4-related HT, fibrosing variant of HT, Graves’ disease with elevated IgG4, and Riedel thyroiditis [28]. IgG4 Hashimoto’s thyroiditis showed higher grade of stromal fibrosis, lymphoplasmacytic infiltration, and follicular cell degeneration than non-IgG4 thyroiditis [25]. IgG4 thyroiditis is associated with younger age, male gender, subclinical hypothyroidism with a large and diffusely hypoechoic thyroid gland on ultrasound, and higher level of circulating thyroid antibodies [28, 29]. Prospective studies will need to be performed to determine if the variable sonographic appearances of HT is a continuum or whether they represent different variants of Hashimoto’s thyroiditis.