Technical aspects of thyroid ultrasound

INTRODUCTION — Ultrasonography permits "real-time" identification of structures as small as 2 mm in diameter, thereby allowing the visualization of very small tumors of the thyroid and parathyroid glands. In the case of the thyroid, certain ultrasound features of a nodule are clinically useful because they may correlate with histology, thereby enhancing a management decision. These methods also permit estimates of overall and regional blood flow to the thyroid. This topic review will discuss the basic technology of thyroid ultrasonography. The clinical use of thyroid ultrasonography is reviewed elsewhere. (See "Overview of the clinical utility of ultrasonography in thyroid disease".)

GENERAL PRINCIPLES — Thyroid ultrasonography is a technique that requires experience, a basic understanding of physics, and familiarity with neck anatomy and pathology. It cannot be optimally performed by a technician who simply takes pictures and then submits the photographs for later interpretation. Instead, the ideal imaging requires input from both the ordering clinician and the interpreting clinician. The ordering clinician should specify the precise clinical question that must be answered.

Some endocrinologists prefer to perform the sonogram themselves, while others refer the patient to a sonographer or a radiologist. In some cases, improved efficiency may be obtained when the endocrinologist or surgeon actually performs the sonogram [1-4]. No matter who does the sonogram, results can sometimes be equivocal and potentially misleading.

The endocrinologist, surgeon, or radiologist who performs thyroid sonograms must:

●Master thyroid palpation, anatomy, and pathophysiology

And have:

●Special training in ultrasound procedures

●Meticulous attention to detail and adequate time to perform a full examination

●Ability to invest in costly and sophisticated equipment and perform an adequate number of studies to achieve expertise

STANDARD METHODS

Gray-scale ultrasonography

Technique and limitations — Gray-scale ultrasonography involves the intermittent generation of a pulse of sound energy and the reception of reflected echoes to produce an image of the structures traversed by the sound. This technology produces high-resolution thyroid images by using sound frequencies between 5 and 15 million cycles per second (megahertz [MHz]). These frequencies are well above the range audible by humans and are safe. Sound waves of this frequency pass through air poorly and, therefore, the transducer (the device that generates the sound and receives the echo) must be coupled to the skin with a medium (in most cases, gel) that excludes air.

The sound waves penetrate tissues, and a portion of the energy is reflected at tissue interfaces up to a depth of 5 cm. The thyroid gland is well within this distance in most patients. Modern transducers provide high resolution of structures as small as 2 mm, providing their echodensity is different from surrounding tissues or they are separated by an interface. In contrast, other imaging methods such as computed tomography, magnetic resonance imaging, and radionuclide scanning only detect nodules that are considerably larger, almost to the centimeter range.

●Standard examination – Images are best obtained with the patient lying supine with the neck hyperextended. Care should be taken to avoid overextension in patients with limited range of motion in the neck. Landmarks such as the trachea, sternal notch, and the thyroid gland and thyroid abnormalities should be carefully palpated and their locations noted. The entire region should be examined completely in the transverse plane, starting at the chin and extending to the sternal notch, and in the longitudinal (or "sagittal") plane, starting in the midline and extending well laterally.

A 10 to 15-MHz transducer is used to determine the size of the thyroid gland and to assess the regional anatomy and lymph nodes (image 1). Although a 10- to 15-MHz transducer may delineate nodules that are 2 to 3 mm in diameter, the value of detecting nodules of this size is debatable because they are too small to aspirate for cytologic examination, and their image may be difficult to reproduce. However, some information may be obtained, notably the presence of microcalcifications, reproducible changes in size, or the emergence of a multinodular pattern. (See "Overview of the clinical utility of ultrasonography in thyroid disease", section on 'Thyroid nodules'.)

There are protocols to assemble a mosaic of images to depict a very large thyroid lobe or goiter. A modification of conventional ultrasound, panoramic ultrasound (or virtual convex imaging is available on some newer machines), has been reported to produce images with a large anatomic field of view that displays both lobes of the thyroid gland on a single image (image 1) [5].

●Limitations – Some of the technical factors that limit the usefulness of ultrasonography include:

•Attenuation (or decrease in amplitude and intensity) of the high-frequency sound waves in deeper tissues, which may make thyroid evaluation difficult in patients with a large body habitus. Evaluation of very large goiters may also be difficult.

•Inadequate coupling of the transducer to the skin (along its full length) in patients with certain anatomical characteristics (eg, significant scarring due to radiation or extensive surgery, thin necks [in older adults], presence of a tracheostomy, or a deep sternal notch).

•Distortion by air-filled structures such as the trachea.

•Blockade of the signals posterior to calcific deposits within the thyroid, cartilage, or bone.

•Nonvisualization of portions of the thyroid gland that lie retrosternally.

•User experience, as the ability to accurately interpret ultrasound findings is highly dependent on the expertise of the sonographer. Furthermore, there is substantial interobserver variability even among experienced thyroid sonographers when interpreting thyroid ultrasound images and describing individual features.

Estimation of thyroid gland volume — Gray-scale ultrasonography may be used to answer questions about the size and anatomy of the thyroid gland and adjacent structures in the neck. Using two-dimensional (2D) images, a skilled ultrasonographer is able to reconstruct the thyroid in three dimensions. The structural information becomes optimally useful in management when it is integrated with other clinical data.

The dimensions of each lobe should be determined in the sagittal and transverse planes to determine the length (L), anterior-posterior depth (D), and transverse width (W) of the gland. The volume of the thyroid gland, lobe, or nodule can then be calculated using a simplified formula for a prolate ellipse:

In healthy adults without iodine deficiency, a normal thyroid lobe is approximately 4 to 4.8 by 1 to 1.8 by 0.8 to 1.6 cm in size, with a mean total volume for both lobes of 8 to 10 mL (range 3 to 20 mL) [6,7]. The size of the thyroid is positively correlated with body surface area [8,9]. Sex also impacts thyroid volume; adult males have larger thyroids than females [8-11].

Characterization of thyroid nodules — Ultrasonography provides a detailed map of the location of thyroid nodules and helps to characterize the nodules.

●Size – The size of a nodule can be measured fairly accurately based on a scale on the sonographic screen or the sonographer can place marks on the image to calculate the dimensions electronically. One study reported imperfect concordance between the dimensions of thyroid nodules as revealed by ultrasound examination when compared with surgical excision. Concordance was 78.5 percent with nodules equal to or smaller than 1 cm and significantly less for larger nodules [12].

●Echogenicity – Solid nodules should be described by using normal thyroid tissue as a reference. Among solid nodules, the intensity of the echo pattern relative to the extranodular thyroid lobe depends upon the transmission or reflection of the ultrasonic signal by the nodule. Hypoechoic, solid nodules transmit the sound energy with minimal reflection, while hyperechoic nodules reflect a greater proportion of the signal back to the ultrasound probe. Isoechoic nodules interact with the ultrasound similarly to normal thyroid tissue. The clinical use of ultrasound to identify and characterize nodules is reviewed in detail separately. (See "Overview of the clinical utility of ultrasonography in thyroid disease", section on 'Thyroid nodules'.)

●Change over time – Ultrasound measurements are used to compare size changes in a nodule over time. However, comparisons based on ultrasonic assessments should be interpreted cautiously. It is difficult to reproduce identical 2D image planes for follow-up studies, and there is variation in ultrasonic thyroid nodule volume determinations even among experienced thyroid ultrasonographers [13,14]. As an example, in a prospective, blinded trial of 42 patients (8 men, 34 women) with 25 uninodular and 17 multinodular thyroid glands, the interobserver variation for the determination of thyroid nodule volume (n = 38) was 48.96 percent for the ellipsoid method of assessing size and 48.64 percent for the planimetric method [13]. These data have been used as the basis for the recommendation that a 50 percent increase in volume is the minimally significant detectable change in size, which translates into an increase of at least 20 percent in at least two of the three measured nodule diameters. (See "Diagnostic approach to and treatment of thyroid nodules", section on 'Benign nodules (Bethesda II)'.)

Color-flow Doppler ultrasonography — Color-flow Doppler is useful to assess the general vascularity of a gland and to differentiate small cystic nodules from vascular structures. Additionally, it can provide insight into the distinction of various diffuse thyroid disorders. (See "Diagnosis of hyperthyroidism", section on 'Determining the etiology'.)

Color-flow Doppler imaging is based upon the change in pitch (wave frequency) of a sound that increases when it approaches a listener (the ear or, in the case of ultrasonography, a transducer) and decreases as it departs. Doppler imaging provides information about blood flow and, when superimposed on the real-time gray-scale image, indicates both the direction and the velocity of blood flow [15,16]. The assignment of color is designed to depict flow toward or away from the ultrasound probe, rather than to signify arteries and veins. The direction of the flow can be discerned by comparing with the color bar on the side of the ultrasound image. The color on the top of the bar is indicative of flow toward the ultrasound probe, whereas the color on the bottom indicates flow away from the probe. The shade or intensity of the color is a semiquantitative indicator of flow velocity. By convention, darker shades indicate slow flow and lighter shades high flow; black signifies no flow or an excessively oblique angle of the Doppler signal to the vascular structure.

Power Doppler ultrasonography — Power Doppler may be useful to distinguish the vascularity of a small structure, such as a lymph node, to detect the presence of abnormal peripheral blood flow that would indicate malignant transformation. One study found that the detection of peripheral vascularization had the best combination of sensitivity and specificity for identification of malignant transformation in a lymph node [17].

Power Doppler ultrasonography is more sensitive than color-flow Doppler for detection of blood flow, and it is less dependent on the scanning angle. In addition to detecting blood flow toward and away from the probe (like color-flow Doppler), it also detects motion that is parallel to the ultrasound probe. However, it does not provide information regarding velocity of flow as is available with color-flow Doppler.

INVESTIGATIONAL METHODS

Three-dimensional ultrasonography — Three-dimensional (3D) ultrasonography is a newer technology that has been applied to the evaluation of thyroid nodules [18-21]. The accuracy of thyroid nodule assessment, interobserver variability, and repeatability appears to be improved with 3D compared with conventional two-dimensional (2D) echography [20,22].

This compounded imaging aims to provide insight into the 3D shape of nodules to enhance the detection of malignancy. Specifically, this technology seeks to assess the presence of extracapsular and extrathyroidal extension. At present, 3D ultrasound remains investigative as limited numbers of studies have examined its diagnostic performance. Furthermore, the need for additional software/equipment and time for image processing limit the widespread applicability of its use.

Contrast-enhanced ultrasonography (CEUS) — Contrast-enhanced ultrasound (CEUS) technology aids in the evaluation of tissue microvascularization. This technology is being applied to thyroid nodules to uncover tumor vascularity, an important component of neoplastic growth.

Gas-filled microbubbles with a mean diameter less than that of a red blood corpuscle act as the contrast agents [23]. Due to their small size (1 to 10 micrometer in diameter), the microbubbles are able to pass into capillary blood vessels. The contrast agent resonates upon application of the 7 to 15-MHz ultrasound probe, which is then detected by the ultrasound probe because its acoustic characteristics are different from the surrounding tissues [24]. The contrast agent is typically in a powdered form that must be reconstituted into normal saline for intravenous injection. Each injection of the contrast agent requires rapid examination of the area of interest as the microbubble perfusions can only be observed for two minutes or less. For this reason, cine-loop examinations of the region of interest are recorded for later interpretation by the sonographer.

The normal thyroid gland is rich in vascularity. The healthy parenchyma, therefore, demonstrates rapid, uniform enhancement after contrast injection. In contrast, thyroid nodules exhibit neovascularization that is distinct from the surrounding tissue. These patterns of vessel formation have been investigated for their potential to predict malignancy. With CEUS, thyroid nodules can be classified as having different enhancement patterns, based on:

The distribution of contrast:

●Homogenous

●Heterogeneous

The degree of contrast uptake:

●Low (little or no contrast agent in a lesion)

●Iso (intensity similar to the surrounding thyroid tissue)

●High (intensity higher than the surrounding thyroid tissue)

And the pattern of contrast uptake:

●Ring enhancement

●Centripetal enhancement

●Central enhancement

Benign nodules tend to demonstrate homogeneous iso- or high-contrast enhancement, or a peripheral ring of enhancement. Malignancies, on the other hand, tend to display heterogeneously low-contrast enhancement [25,26].

There are conflicting reports on the diagnostic efficacy of CEUS. A meta-analysis of 37 studies calculated a pooled sensitivity and specificity of 87 and 83 percent, respectively [27]. However, there was high heterogeneity of included studies, likely due to a lack of unified standards for quantitative or qualitative studies [27,28]. Existing studies have not uniformly established which parameters are optimal for determining the diagnostic efficacy of this technology. Additional limitations of this technology include uncertain performance characteristics for nodules that are largely cystic, small (<5 mm), or have coarse calcifications as these lesions are typically excluded from analysis. More detailed and standardized methodology, as well as a more inclusive nodule population, is needed to design optimal studies before this technology would be suitable for widespread clinical use.

Estimation of tissue elasticity by ultrasound — The role of sonoelastography in the evaluation of thyroid nodules is uncertain. The clinical utility of elastography to augment gray-scale sonography for malignancy risk prediction remains intriguing, and it may be beneficial in certain circumstances (eg, low or intermediate suspicion sonographic pattern and/or indeterminate [Bethesda III/IV] cytology). Large-scale, prospective, multi-institutional studies are needed to determine the exact role for this adjunctive procedure.

The hardness of a nodule is assessed on physical examination by palpation; firm nodules are associated with an increased risk of malignancy [29]. Not all nodules are palpable, however. Elastography circumvents this issue of nonpalpable nodules while also providing a more objective way to examine the degree of tissue stiffness and, by extension, further stratify malignancy risk. The greater the compressibility of a nodule as measured by elastography, the lower the likelihood of malignancy. Conversely, nodules that are stiffer (or less compressible) are more likely to harbor a malignancy.

There are two main forms of elastography: external compression of the probe applied by the user (also called strain ultrasound elastography [SUSE]) and shear waves generated by the ultrasound machine (called shear wave elastography [SWE]).

●Strain ultrasound elastography – The SUSE is performed by the sonographer manually pressing the probe into the patient's skin, which creates a perpendicular strain. The area of interest is captured in a box on the ultrasound screen. Images are obtained before and after the tissue compression. Software contained within the ultrasound machine measures the tissue displacement and superimposes a color map on the ultrasound images that provides a qualitative measure of the tissue elasticity. The colors range from red (highest degree of elasticity and lowest risk of malignancy) to green (intermediate elasticity) to blue (least elasticity and highest risk of malignancy). For the purposes of quantitative analysis, these colors may be converted to an elasticity score, typically between 1 and 5 (or 1 to 4), with 1 being the most elastic (easily compressible) and 5 the least elastic (not compressible).

In a systematic review of 12 studies (817 benign and 363 malignant thyroid nodules) examining the diagnostic accuracy of SUSE for predicting malignant nodules, the highest sensitivity (98.3 percent) was achieved using a threshold elasticity score of between 1 and 2, but did so at the expense of very low specificity (19.6 percent) [30]. The sensitivity and specificity were 86 and 66.7 percent, respectively, using a threshold elasticity score between 2 and 3.

SUSE is limited in its utility by operator experience; the ability to apply uniform force with the probe from one examination to the next requires a significant learning curve. Additionally, SUSE is a semiquantitative technology.

●Shear wave elastography – SWE technology generates particle motion at a tangential angle to the skin surface utilizing an acoustic radiation force impulse (ARFI) from the ultrasound probe. The shear waves cause transient displacement of the soft tissues. The wave velocities are directly related to the tissues' elastic properties; stiffer tissues allow for shear waves to propagate faster. By measuring how fast the shear wave travels to reach certain positions, the degree of tissue stiffness is able to be quantified.

The performance of this technology for the distinction of benign and malignant nodules is inconsistently reported, likely related to small sample sizes, retrospective nature of some studies, and differing inclusion/exclusion criteria. A larger sample size of nodules found a modest performance of this technology; a meta-analysis of 13 studies including 1854 thyroid nodules in 1641 patients found a pooled sensitivity and specificity of 81 and 84 percent, respectively [31].

In theory, SWE may be more objective, more reproducible, and less operator-dependent than SUSE [31]. However, direct comparison of SUSE and SWE in one meta-analysis including 54 studies and 2621 malignant nodules revealed that SUSE had improved diagnostic capacity compared with SWE for differentiating benign and malignant nodules. SUSE had a pooled sensitivity and specificity (83 and 81.2 percent, respectively), which was significantly higher than SWE (78.7 and 80.5 percent, respectively) [32].

●Limitations of elastography – There are noteworthy limitations to the routine use of elastography in clinical practice.

•Indeterminate cytology – A major limitation of many elastography studies is the exclusion of nodules with indeterminate cytology. This is a particularly important limitation as the nodules with indeterminate cytology (Bethesda III and IV) also are frequently associated an indeterminate sonographic appearance [33]. In a pooled analysis of 20 studies (1734 indeterminate nodules undergoing elastography), the sensitivity and specificity of elastography were 76.6 and 86.7 percent, respectively [34].

•Partially cystic or calcified nodules – Elastography has very limited value for evaluating partially cystic or calcified thyroid nodules; most studies excluded nodules with either sonographic feature. Additionally, many studies only examined patients with a solitary nodule, limiting the real-life clinical utility of this technology.

•Nodule location – The location of a nodule may impact whether it is amenable to evaluation with elastography. Nodules located in the isthmus or near the trachea may not be amenable to evaluation by elastography due to the artefactual interference introduced by the rigid trachea. In addition, proximity to the carotid may also impact the accuracy. One study found that nodules located within 15 mm of the carotid resulted in an underestimation of the elasticity contrast index, which can translate into a false lowering of the malignancy risk [35]. For those with a smaller thyroid size (between 15 and 20 mm in the transverse direction), which would include most women and children, the majority of nodules will be positioned within this proximity to the carotid to misregister the tissue elasticity.

In a retrospective, multicenter study (498 patients) analyzing whether the addition of SUSE or SWE improved the diagnostic accuracy of the American Thyroid Association (ATA) and American College of Radiology Thyroid Imaging Reporting and Data System (ACR-TIRADS) sonographic risk stratification systems (SRSS) for prediction of malignancy and reduced unnecessary biopsy rates, the two SRSS had similar sensitivity but the specificity was significantly higher for the ACR-TIRADS [36]. Both SRSS had a significantly higher sensitivity than SUSE or SWE. The combination of each SRSS and elastography (either type) improved the sensitivity compared with the SRSS alone but the difference was not statistically significant. The specificity decreased, however, after the combination of SRSS and elastography. Unnecessary biopsy rates decreased with the addition of elastography to ACR-TIRADS but was not significantly impacted for the ATA SRSS.

SUMMARY

●Gray-scale ultrasonography

•Gray-scale ultrasonography involves the intermittent generation of a pulse of sound energy (sound frequencies between 5 and 15 million cycles per second [megahertz (MHz)]) and the reception of reflected echoes to produce an image of the structures traversed by the sound.

•Modern transducers provide high resolution of structures as small as 2 mm, providing their echodensity is different or they are separated by an interface.

•A 10 to 15 MHz transducer is typically used to determine the size of the thyroid gland, to assess the regional anatomy (image 1), and to detect any abnormalities of the thyroid parenchyma.

●Estimation of thyroid gland size – Thyroid ultrasonography is used to answer questions about the size and anatomy of the thyroid gland. In healthy adults without iodine deficiency, a normal thyroid lobe is approximately 4 to 4.8 by 1 to 1.8 by 0.8 to 1.6 cm in size, with a mean total volume for both lobes of 8 to 10 mL (range 3 to 20 mL). (See 'Estimation of thyroid gland volume' above.)

●Characterization of thyroid nodules – Thyroid ultrasonography is used to identify the location, size, and characteristics of thyroid nodules and to estimate changes in nodule size or characteristics over time. Nodules are called isoechoic if their texture closely resembles that of normal thyroid tissue (image 2), hyperechoic if more echogenic (or brighter) (image 3), and hypoechoic if less echogenic (or darker) (image 4). (See 'Characterization of thyroid nodules' above.)

●Color-flow Doppler – Color-flow Doppler is useful to assess the general vascularity of a gland and to differentiate small cystic nodules from vascular structures. Color-flow Doppler imaging is based upon the change in pitch (wave frequency) of a sound that increases when it approaches a listener (the ear or, in the case of ultrasonography, a transducer) and decreases as it departs. Doppler imaging provides information about blood flow to a static gray-scale image and, when superimposed on the real-time gray-scale image, indicates both the direction and the velocity of blood flow. (See 'Color-flow Doppler ultrasonography' above.)

●Power Doppler ultrasonography – Power Doppler may be useful to distinguish the vascularity of a small structure, such as a lymph node, to detect the presence of abnormal peripheral blood flow that would indicate malignant transformation. In addition to detecting blood flow toward and away from the probe (like color-flow Doppler), it also detects motion that is parallel to the ultrasound probe. However, it does not provide information regarding the speed of flow as is available with color-flow Doppler. (See 'Power Doppler ultrasonography' above.)

●Investigational methods – Elastography and contrast-enhanced ultrasonography (CEUS) are being investigated for their role in further delineating malignancy risk of thyroid nodules. (See 'Investigational methods' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Manfred Blum, MD, FACP (deceased), who contributed to earlier versions of this topic review.