Comparative Efficacy of Ultrasonography and Acoustic Radiation Force Impulse (ARFI) Elastography in Prediction of Malignancy in Thyroid Nodules

Abstract

Objective:

The incidence of malignancy in thyroid nodules is infrequent, but this trend may be reversing. The present study was conducted to emphasize the diagnostic accuracy of acoustic radiation force impulse (ARFI) imaging, in addition to conventional gray-scale ultrasonography (US), for differentiating benign and malignant thyroid nodules.

Methods:

A total of 141 patients with thyroid nodules (≥10 mm) were included in the study and were evaluated with US, Doppler, and ARFI elastography using Siemens S2000 Acuson ultrasound equipment.

Results:

The sonographic patterns most predictive and indicative of malignancy included irregular margins and presence of microcalcifications. The Doppler findings in isolation were not extremely sensitive in the detection of malignancy. The shear wave velocity cutoff value on ARFI imaging using receiver operating characteristic curves for differentiation of benign and malignant nodules were noted at 2.87 m/s. ARFI imaging performed better than US and Doppler with sensitivity of 75%, specificity of 96%, and accuracy of 94%.

Conclusion:

ARFI elastography could be utilized as a reliable initial screening test for detection of malignancy in thyroid nodules.

Keywords

ultrasonography elastography acoustic radiation force impulse thyroid malignancy

Thyroid nodules are often incidentally detected when clinically examining a sample of the general population.¹ These nodules are more commonly noted among females.² While the prevalence of thyroid nodules is relatively high at 12.2% in the Indian community, thyroid cancer is rare, with an increasing trend over the years.³ Compared to all other cancers in India, the prevalence of thyroid cancer is only 0.1% to 0.2%.⁴ Therefore, in a patient presenting with thyroid swelling, the primary task is to differentiate benign from malignant lesions. There is currently no perfect test available; however, palpation by an experienced clinician along with biochemical parameters (serum thyroxine and TSH levels) can be extremely helpful.⁵

Ultrasonography (US) is the preferred method for detecting thyroid nodules.⁶ The American Thyroid Association recommends that a thyroid US be part of the initial workup for patients presenting with one or more thyroid nodules.⁵ It is a rapid, noninvasive procedure with remarkable ability to differentiate between benign and malignant nodules as well as guide selection of nodules for pathologic examination.⁵ However, its usefulness in prediction of nodule malignancy varies depending on the skill of US operator, with sensitivity between 42% and 91% and specificity ranging from 50% to 91%.⁶

Malignancy may be reliably determined due to a nodule’s firm consistency on palpation.⁷ Dynamic elastography was the first tool to depart from the subjective observation of the examiner to a reproducible qualitative assessment of tissue consistency.⁸ It allowed for in vivo estimation of the parenchyma’s physical properties, as induced by mechanical compression. A loss of elasticity in the malignant nodule could be detected compared to surrounding tissue.⁹ Modern techniques allow for a quantitative evaluation of tissue stiffness, which enhances reproducibility and specificity.¹⁰

Acoustic radiation force impulse (ARFI) is a new technique that is shown to be highly reproducible and reliable for parenchymal organs.¹¹ ARFI uses short bursts of acoustic radiation force to assess the tissue properties.¹¹ It works by inducing ultrasound pulses to the target tissue and detecting slight displacements in the tissue, on the order of less than 1 μM.¹¹ ARFI allows for operator-independent, reproducible, and quantitative evaluation of tissue elasticity and avoids external compression artifacts.¹¹

Pathologic examination is the gold standard for thyroid lesion characterization, with a diagnostic predictive accuracy of over 90% for malignancy.¹² However, this test is invasive, which can lead to patient reluctance.¹³ Given the background, this study emphasized the diagnostic accuracy of ARFI imaging, in addition to conventional US and Doppler, for differentiating benign and malignant thyroid nodules. An additional goal of the study was to compare the efficacy of US and ARFI elastography in predicting malignant thyroid nodules.

Materials and Methods

This study was conducted in a tertiary care hospital in northern India in association with its department of pathology from July 2018 to June 2019. Prior approval from the College Research Committee and Institutional Ethics Committee was obtained. The study was done on patients referred to the hospital’s department of radiology with thyroid nodule/nodules. The design was a prospective observational study with a sample size of 150 patients.

After giving written informed consent, all patients who were diagnosed with a thyroid nodule and those with nodules at or greater than 10 mm on US underwent further nodule characterization as to benign or malignancy. Patients without cytological or histopathological follow-up were excluded. Cystic lesions without any mural nodule were also excluded from this study. Conventional US with Doppler and ARFI elastography was performed using a Siemens S2000 Acuson ultrasound equipment (Siemens Medical Systems). US was preformed using a 4-9 MHz linear transducer. In the cases of multiple thyroid nodules, the nodule with malignant features or the largest nodule was selected.

ARFI-based eSie Touch Elasticity Index (see Figure 1), virtual touch tissue imaging (VTI; see Figure 2), and virtual touch tissue quantification (VTTQ; see Figure 3) elastography techniques were used. The eSie Touch EI method provides a color elastogram, which demonstrates increasing stiffness of tissues, depicted by a range of red, intermediate, and overlapping to violet hues.

Figure 1.

Color elastogram showing predominantly green tint suggestive of soft (benign) texture of the nodule.

Figure 2.

Virtual touch tissue imaging in the normal thyroid gland showing light-gray homogeneous tint in the region of interest, suggesting a soft texture.

Figure 3.

Acoustic radiation force impulse imaging (virtual touch tissue quantification) with region of interest placed within the normal thyroid parenchyma measures a shear wave velocity of 1.84 m/s.

The elasticity index (EI) was obtained by assigning numbers 0 to 4 for a specific color (e.g., 0: violet = hardest; 1: blue = medium hard; 2: green = medium soft; 3: yellow = soft; 4: red = softest). In cases with overlapping color, the softest alternative hue was considered. Tissue stiffness on VTI elastography was scored from 1 to 6 (soft to hard; see Figure 4). The VTTQ model employs ARFI imaging as well as a region of interest (ROI) box, placed on the tissue of interest. It was important to determine whether cystic or calcified areas were to be skipped. The shear wave velocity (VTTQ) value was recorded in m/s.⁶ Extreme care was taken while gently placing the transducer on the skin surface with minimal pressure to avoid adding external compression to the elasticity of the material. The patient was asked to hold their breath, and the transducer was kept motionless during image acquisition. All patients were subjected to cytology/histopathological examination. Patients with follicular neoplasms on cytology underwent biopsy for confirmation.

Figure 4.

Isoechoic nodule with well-defined margins and complete peripheral halo showing bright gray tint on virtual touch tissue imaging, suggestive of benign nodule.

Statistical Analysis

All statistical analysis was done using Microsoft Excel Spreadsheet with SPSS software (Version 22). When conducting the Mann-Whitney U test and chi-square test, a P value < .05 was considered statistically significant. A confusion matrix (2 × 2 contingency table) was plotted in Excel software to check the performance of US and ARFI. A variety of matrices were calculated, such as sensitivity (true positive rate), specificity (true negative rate), positive predictive value (precision), negative predictive value, false positive rate (1 – specificity), and accuracy. The receiver operating characteristic (ROC) curve was created by plotting the sensitivity (x axis) against the false positive rate (Y axis). The maximum value of area under the curve (AUC) was considered as the best threshold (cutoff value) for differentiation of benign from malignant nodules. Descriptive analysis was used, and the data were represented in the form of charts, tables, and graphs. The charts and tables were created in Microsoft Excel and Word.

Results

This prospective observational study conducted over a period of 1 year. A total of 150 subjects with thyroid nodules were recruited and evaluated. Eleven subjects (7.3%) were excluded from further analysis because of lack of cytological/histopathological diagnosis.

Thus, a total of 141 subjects were included in the study. There were 120 (85%) female patients and 21 (14.8%) males. The mean age of males presenting with a thyroid nodule was 49.43 years, and the mean age of females presenting with a nodule was 46.63 years.

Nodule characterization on US and Doppler sonography

Among the 141 patients that were studied with thyroid nodules, the most common finding was an echotexture that depicted a spongiform nodule, with 46 patients having this classification. Hyperechoic nodules were a secondary sonographic classification, with 35 patients having this imaging appearance. There were 27 nodules that had coarse calcifications, 10 internal mural nodules, and 11 echogenic punctate foci (EPI). Septations were seen in two cystic nodules, and cystic areas were noted in two otherwise hyperechoic nodules (see Table 1).

Table 1.

Gray-Scale Characterization of Thyroid Nodules on High-Resolution Ultrasonography.

Nodule Features	No.	%
Echogenicity
Spongiform	46	32.62
Hyperechoic	35	24.82
Hypoechoic	27	19.15
Iso-hyper	13	9.22
Cystic	12	8.51
Iso-hypo	7	4.96
Peripheral iso-central hypo	1	0.71
Total	141	100.00
Width vs. height
Round	10	7.09
Taller than wider	11	7.80
Wider than taller	120	85.11
Total	141	100.00
Margins
Ill defined	6	4.26
Smooth	135	95.74
Total	141	100.00
Present
Coarse calcification	27	19.15
Mural nodules	10	7.09
Echogenic Punctate Foci	11	7.80
Cystic area	3	2.13
Septations	2	1.42

There were 120 of the 141 nodules that were further classified as taller than wide, and 10 nodules were graded round in shape. This left a total of 11 nodules that were taller than wide, which was suggestive of malignancy. There were 6 out of 142 nodules that demonstrated an ill-defined margin, and of these, 2 were malignant on fine needle aspiration cytology (FNAC). There were 8 out of 142 that were noted to have mural nodules, and of those, 2 were proven malignant with FNAC. Of the 11 nodules that possessed the taller than wide appearance, FNAC demonstrated that 6 were malignant (see Table 1).

Among the Doppler findings, the most sensitive indicator for malignancy was the presence of both peripheral and internal vascularity in the nodule. Peripheral vascularity and the presence of both peripheral and internal vascularity were combined and seen in 5 of the 12 nodules. Again, the Doppler findings in isolation were not extremely sensitive in the detection of malignancy (see Tables 2 and 3).

Table 2.

Distribution of Color Doppler Findings in Thyroid Nodules.

Doppler Findings in Nodules	No.	%
Mild internal vascularity	83	58.87
Minimal internal vascularity	17	12.06
Mild peripheral vascularity	16	11.35
Mild internal and peripheral vascularity	14	9.93
Increase internal vascularity	6	4.26
Increase internal and peripheral vascularity	3	2.13
Increase internal and peripheral vascularity with encased left common carotid artery	1	0.71
Mild compression on internal jugular vein	1	0.71
Total	141	100.00

Table 3.

Correlation of Doppler Ultrasound Findings in Thyroid Nodules With FNAC.

Doppler Findings	FNAC Results
Doppler Findings	Benign	Benign With Inflammation	High Probability of Malignant	Probably Malignant	No.
Mild internal	77(93%)	2(2%)	4(5%)	0	83
Minimal internal	15(88%)	1(6%)	1(6%)	0	17
Mild peripheral	15(94%)	0	0	1^a (6%)	16
Mild internal and Peripheral	10(71%)	0	4(29%)	0	14
Increase internal	1(16.7%)	3(50%)	1(16.7%)	1^b (16.7%)	6
Increase internal and peripheral	1(33%)	2(67%)	0	0	3
Increase internal and peripheral with encased left CCA	1(100%)	0	0	0	1
Mild compression on IJV	1100%	0	0	0	1
grand Total	121(86%)	8(6%)	10(7%)	2(1%)	141

Abbreviations: CCA, Common Carotid Artery; FNAC, fine needle aspiration cytology; IJV, Internal Jugular Vein.

Follicular variant of papillary carcinoma confirmed on histopathology.

Hemato-lymphoid neoplasm of thyroid confirmed on histopathology.

Nodule characterization on ARFI elastography

The shear wave velocity (SWV) cutoff value on ARFI imaging utilized ROC curves for differentiating benign and malignant nodules, and this value was noted as 2.87m/s (see Figure 5).

Figure 5.

Receiver operating curve (ROC) analysis of shear wave velocity (SWV) cutoff values for differentiation of benign and malignant nodules (a) ROC curve with various acoustic radiation force impulse imaging (ARFI) thresholds (b) ROC curve showing area under the curve (AUC) at SWV cutoff value of 2.87 m/s (AUC = 0.36, false-positive rate = 0.04; sensitivity = 0.75; specificity = 0.96).

ARFI detected a total of 14 nodules that met the hardness criteria, indicative of malignancy (see Figures 6 –8). Among these 14 nodules labeled as malignant on ARFI imaging, 9 nodules were suspicious for malignancy based on FNAC (see Table 4). Out of 129 pathologically proven benign nodules, ARFI elastography demonstrated good specificity by indicating a soft tissue texture in 124 nodules (see Figure 9).

Figure 6.

Ill-defined round to oval hypoechoic nodule with absent halo, showing acoustic radiation force impulse imaging (ARFI) measurement of 4.3m/s with regions of interest (ROI) placed in different parts of nodule, suggestive of stiff nodule-malignant etiology. Histopathological examination revealed papillary carcinoma.

Figure 7.

Ill-defined round to oval hypoechoic nodule showing 2.50 m/s on acoustic radiation force impulse imaging (ARFI) elastography, suggestive of benign etiology, although histopathology showed follicular carcinoma.

Figure 8.

Ill-defined hypoechoic nodule in right lobe of thyroid with high velocity x.xx m/s (>7 m/s), suggestive of malignant etiology. Histopathology showed medullary carcinoma.

Table 4.

Comparison of ARFI Findings in Thyroid Nodules With FNAC.^a

	FNAC
ARFI Findings	Benign	Benign With Inflammation	High Probability of Malignant	Probably Malignant	No.
Benign	91(92%)	6(6%)	1(1%)	1^b (1%)	99
Probably benign	26(93%)	1(3.5%)	1(3.5%)	0	28
Probably malignant	4(29%)	1(7%)	8(57%)	1^c (7%)	14
Total	121(86%)	8(6%)	10(7%)	2(1%)	141

χ² = 75.41; df = 9. ARFI, acoustic radiation force impulse imaging; FNAC, fine needle aspiration cytology.

Follicular variant of papillary carcinoma confirmed on histopathology.

Hemato-lymphoid neoplasm of thyroid confirmed on histopathology.

Figure 9.

Virtual touch tissue quantification (VTTQ) elastography in a nodule showing low velocity (0.85 m/s), suggestive of benign nodule.

The US and Doppler data correctly identified 121 benign nodules (out of 129) and 5 malignant nodules (out of 12). The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy came out to be 42%, 94%, 38%, 95%, and 89%, respectively (see Table 5).

Table 5.

Correlation of Gray-Scale Ultrasound and Doppler Findings in Thyroid Nodules With FNAC.^a

		FNAC
		Positive (for Malignancy)	Negative
US and Doppler	Positive (for malignancy)	5	8	13
US and Doppler	Negative	7	121	128
		12	129	141

Sensitivity = 42%; specificity = 94%; positive predictive value (precision) = 38%; negative predictive value = 95%; accuracy = 89%; false-positive rate = 6%. FNAC, fine needle aspiration cytology; US, ultrasound.

ARFI imaging performed better and was able to correctly pick up 9 malignant nodules (5 more than US and Doppler) and 124 benign nodules (3 more than US and Doppler). The ARFI elastography with a cutoff value of SWV at 2.87 m/s yielded a sensitivity of 75%, specificity of 96%, and accuracy of 94% (see Table 6).

Table 6.

Correlation of ARFI Findings in Thyroid Nodules With FNAC.^a

		FNAC
		Positive (for Malignancy)	Negative
ARFI (2.87 m/s)	Positive (for malignancy)	9	5	14
ARFI (2.87 m/s)	Negative	3	124	127
		12	129	141

Sensitivity = 75%; specificity = 96%; positive predictive value (precision) = 64%; negative predictive value = 98%; accuracy = 94%; false-positive rate = 4%. ARFI, acoustic radiation force impulse imaging; FNAC, fine needle aspiration cytology.

Discussion

As much as the thyroid nodules are common in the general population, malignant nodules are infrequent and constitute only 5% of those detected.¹ “Population-based screening studies” have shown clinically detectable nodules in only 5% adults, whereas US examinations and autopsies demonstrated thyroid nodules in 50% of women and 20% of men over 50 years.⁵ The main objective of evaluating a thyroid nodule is to exclude malignancy given that a low proportion is found to be malignant.¹ In spite of various newer techniques and multiple investigation choices, there may be diagnostic drawbacks that make the ultimate diagnosis elusive.

ARFI or point shear wave elastography is the latest effort to quantify tissue elasticity by generating a push pulse into the tissue, which assesses the speed of sound reflecting back from displaced tissue. This diagnostic technique quantifies the hardness of the target such that the harder the tissue, the higher the velocity. Now it is possible to assess, in advance, the velocities in different tissues by comparing to normal controls based on large sample sizes (e.g., the cutoff value in case of a normal liver is 3.0 m/s, and normal thyroid is <2 m/s).^14,15

In the present study, a total of 141 nodules were imaged; 12 (8.5%) were found malignant, and 129 (91.5%) nodules were classified benign. Among the malignant nodules, papillary carcinoma was most common, whereas colloid goiter accounted for the majority of the benign nodules. This is in comparison with the data published by the National Council on Radiation Protection and Measurements in 1999.¹⁶

This study’s age range of patients was 17 to 80 years. The mean age of males was 49.43 years, and females’ mean age was 46.63 years. Only 21 (14.8%) out of 141 were male patients, and the remaining 120 (85%) were females. This cohort’s gender distribution for thyroid pathology may be indicative of the female predominance of this disease. Tunbridge et al.² reported similar findings in the Whickham survey regarding the gender distribution of thyroid nodules.

Nodule Characterization on High-Resolution US

Each lesion was evaluated with high-resolution US and color Doppler imaging. The mean size of malignant nodules on US (30.9 ± 9.67 mm) was not significantly different from the mean size of benign nodules (29.6 ± 12.43 mm).

For 7 out of 12 malignant nodules, there was a hypoechoic echotexture compared to the normal surrounding thyroid parenchyma. Most of these nodules showed an ill-defined margin. One papillary carcinoma was found in the solid mural component of a predominantly large cystic lesion. The large malignant nodules depicted internal vascularity or had internal and peripheral vascularity.

These findings are in concordance with Cappelli et al.,¹⁷ who reported microcalcifications (P < .001), blurred margins (P < .001), solid hypoechoic appearance (P < .001), and increased intranodular vascular flow (P < .001) in a majority of malignant rather than indolent nodules.¹⁷

Hypoechogenicity of the nodule was not the most significant feature to label it as a malignant nodule. In fact, hypoechogenicity of the nodule was also found to be indicative of the presence of some inflammation. This was proven by the presence of increased vascularity on color Doppler and later with FNAC evaluation. In this study, only 34 (24%) nodules were found to be hypoechoic. Out of these, only 7 (20%) were found malignant, 4 (~11.8%) showed inflammation, and the rest (23, ~67%) were benign.

Several authors have reported similar findings of thyroid malignancy on conventional US and have reported on the distinction between benign and malignant nodules. Solbiati et al.¹⁸ observed that the chances of a solid nodule being malignant was only 4% if it appeared hyperechoic; however, it was as high as 26% if the lesion appeared hypoechoic. Papini et al¹⁹ observed that irregular margins, intranodular vascularity, and microcalcifications were independent risk factors of malignancy. Kim et al.²⁰ demonstrated that the sonographic findings of a hypoechoic nodule with microcalcifications and irregular or microlobulated margins were highly suggestive of malignancy.

The results of a study done by Kovacevic and Smetana Škurla²¹ revealed that hypoechogenicity, ill-defined margins, calcific foci, and absence of a peripheral halo were significantly associated with malignancy. The findings of present study are in concordance with those studies, suggesting the presence of ill-defined margins as the most significant single marker for reporting the lesion as malignant.

Ivanac et al.²² reported that a statistically significant proportion of malignant nodules showed a peripheral flow along with mild to moderate amount of internal vascularity or marked internal vascular flow with or without peripheral ring.

The majority of nonmalignant lesions (93%) were consistently uniform in shape and showed regular and smooth margins on sonography. A finding of ill-defined margins was more typically indicative of malignancy. This was in concordance with study done by Cappelli et al.¹⁷

Out of the 129 benign nodules, the most commonly found echotexture was spongiform, with 46 patients having spongiform nodule texture. Hyperechoic nodules were second, with 35 patients having hyperechoic nodules on a sonogram. There were at least 12 nodules that were isoechoic to hyperechoic compared to the surrounding normal thyroid tissue. The spongiform pattern suggesting benignity is in accordance with a 2009 study by Bonavita et al.²³

There were 27 nodules that had coarse or chunky calcifications, indicative of nodule benignity. There were no calcifications observed in 81 of the benign nodules.

A large number of benign nodules (83) showed mild internal flow pattern (58%). This finding was in complete discordance with the results of the study by Rago et al²⁴ and was more in concordance with a 2007 study by Ivanac et al.²² Doppler parameters in isolation were not extremely sensitive for diagnosis of malignancy in present study.

Nodule characterization on ARFI imaging

The interactions between shear waves and transducer can be quantified by VTTQ, where stiffer tissues show higher velocities.¹¹

The ROC-optimized SWV cutoff value for malignant nodules in this study came out to be 2.87 m/s. There were only three false negatives and five false positives out of the total 141 nodules. This was the least possible score given a value of 2.87 m/s as the cutoff point. So, with this cutoff value and using ROC curves, the sensitivity and specificity of SWV for prediction of malignancy was 75% and 96%. The negative predictive value, positive predictive value, and accuracy with this cutoff value were 98%, 64%, and 94%, respectively.

In the current study, SWV values were notably toward the higher side in malignant thyroid nodules compared to benign nodules. This was very comparable to the findings reported by Friedrich-Rust et al.²⁵ In the present study, malignant nodules had shear wave velocities that ranged from 2.43 to 4.56 m/s (M = 3.25 ± 0.63) whereas in the benign nodules, the velocities ranged from 0.50 to 2.7 m/s (M = 2.00 ± 0.80). The difference in velocities between benign and malignant nodules came out to be notably significant with P < .0001.

The current study aligns with the results provided by Bojunga et al.²⁶ They demonstrated a significant difference between SWV values in malignant thyroid nodules and normal thyroid tissue (P = .0019) versus benign thyroid nodules and normal thyroid tissue (P = .0039).²⁶ The SWVs of the malignant nodules in their study were in the range of 0.6 to 8.4 m/s (average = 3.41 ± 2.37 m/s), which proved to be statistically higher than that of the benign nodules (2.02 ± 0.95 m/s; P = .0039).²⁶ They found that the most appropriate cutoff with the highest sum of specificity and sensitivity (Youden cutoff) was 2.57 m/s.²⁶ The area under the ROC curve for ARFI imaging with this cutoff value for diagnosing malignant thyroid nodules was 0.69 (95% CI, 0.53–0.85; P = .0043).²⁶

Friedrich-Rust et al.²⁷ also found that the median SWV value of malignant nodules (4.30 m/s) was conclusively higher compared to nonmalignant nodules (2.02 m/s; P = .014). In the study by Gu et al.,²⁸ the average shear wave velocity value of the malignant nodules (3.94 ± 1.393 m/s) was significantly higher compared with benign lesions (2.005 ± 0.485 m/s; P < .001). The cutoff of SWV value for carcinoma in their study was 2.55 m/s. Thus, the prediction of malignancy with high sensitivity and specificity became possible using this cutoff value for the ROC curves.

In a subsequent study, Friedrich-Rust et al.²⁵ found that the median SWV value of malignant nodules (3.63 m/s) was really higher compared to benign nodules (1.91 m/s; P = .001). The negative predictive value of 94% was achieved using a cutoff of 3.0 m/s for ARFI for diagnosing malignancy in thyroid nodules.²⁵

In the current study, two of the thyroid nodules that were malignant were classified as nonmalignant (false negative) on ARFI because the SWV values were less than 2.82 m/s. After the pathological investigation was completed, both the lesions were established and documented as follicular carcinomas. Average SWV value recorded with lesions having follicular carcinomas (2.313 ± 0.984 m/s) turned out to be less than that of lesions with papillary carcinomas (3.134 ± 0.443 m/s; P = .0766). Analogous conclusions were described by Gu et al.²⁸ in their study. They discovered that the average VTTQ value of papillary carcinomas (4.112 ± 1.413 m/s; P = .308) was higher than that of follicular carcinomas (2.857 ± 0.612 m/s). This led to the conclusion that papillary carcinomas are harder compared to follicular carcinomas.²⁸ On histopathological analysis, in specimens of follicular carcinomas, copious tumor cells are detected; however, “microcalcifications” and “fibrosis” are absent from the interstitial tissue that might be the cause for the “low SWV values.”²⁸

The average VTTQ value in this study for papillary carcinoma was found to be 3.25 ± 0.63 m/s. This is in accordance with the values reported by Gu et al.²⁸ On pathological examination, the majority of the papillary carcinomas demonstrated microcalcifications, papillary tubercles, and considerable fibrosis in the interstitium. This may indicate that high SWV values may be attributable to presence of calcifications and fibrosis.²⁸

SWV values of nonmalignant nodules in the research oscillated between 0.50 and 2.7 m/s (M ± SD = 2.00 ± 0.80). The majority of the benign nodules (124/129) had SWV values below 2.87 m/s. Only 5 out of 129 nonmalignant nodules (false-positive cases) exhibited values more than the cutoff values of 2.87 m/s and were documented as malignant on VTTQ imaging. However, three nodules showed signs of inflammation on histopathological examination. Areas of coarse calcifications also may be responsible for producing high SW velocities.

Results of this research exhibit that for distinguishing malignant from nonmalignant thyroid nodules, ARFI imaging had 75% sensitivity, 96% specificity, and 94% accuracy and a cutoff value of 2.87 m/s. Thus, ARFI elastography may prove to be a reliable and more concrete diagnostic test to exclude benign lesions with confidence. This has the potential to prevent these patients from an unnecessary FNAC procedure.

Study Limitations

The present study has limitations, and the most significant is the convenient sample of patients that were evaluated. This may also have affected the number of malignant cases that were fewer and hampered adequate statistical power. The technical limitations of US elastography such as motion sensitivity and an inability to place the ROI in a nodule less than 1 cm in size are limitations of this study. In addition to patients’ breathing movements, there were limitations due to the close proximity of the carotid artery and associated pulsations that may have hampered elastographic accuracy.

Conclusions

Thus, ARFI has a fairly high detection rate of malignant nodules (~75%) and may be used for initial screening of thyroid nodules. The addition of further US features of malignancy, such as high-resolution imaging and Doppler, may be utilized to augment elastography and avoid a needle biopsy.

This study reinforced that the sonographic patterns indicative of malignancy included nodular hypoechogenicity, presence of microcalcifications, irregular margins, and presence of intranodular vascular flow. Based on these, irregular margins and presence of microcalcifications were the most predictive of malignancy. This study reassessed the fact that stiffer tissues produced greater shears wave velocities. Additionally, on VTTQ imaging, the mean SWV of malignant nodules is undoubtedly higher than that of benign ones. The ROC curve analysis of SWVs for differentiation of benign and malignant nodules demonstrated a cutoff velocity of 2.87 m/s. However, with this cutoff value, there may be overlap of elasticity indices between malignant and nonmalignant nodules of the thyroid gland. The inflammation, calcifications, or fibrosis may result in higher velocities in otherwise indolent nodules and may interfere with the diagnostic interpretation. Some follicular carcinomas, probably due to the lack of fibrosis and calcifications, may be misinterpreted as soft nodules. Thus, a careful assessment of such a nodule taking in account other features of malignancy on US and Doppler should be completed to detect malignant nodules.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

Prior ethical approval was received from the Institutional Ethics Committee (IEC) and Institutional review Board (IRB), Teerthanker Mahaveer Medical College & Research Centre (TMMC&RC), Moradabad (Ethical clearance certificate Ref No. TMMC&RC/IEC/17-18/049; dated 12/7/2018).

ORCID iD

Shruti Chandak

References

Guth

Theune

Aberle

Galach

Bamberger

: Very high prevalence of thyroid nodules detected by high frequency (13 MHz) ultrasound examination. Eur J Clin Investig 2009;39:699–706.

Tunbridge

Evered

Hall

, et al: The spectrum of thyroid disease in a community: the Whickham survey. Clin Endocrinol 1977;7:481–493.

Usha

Sundaram

Unnikrishnan

Jayakumar

Nair

Kumar

: High prevalence of undetected thyroid disorders in an iodine sufficient adult south Indian population. J Indian Med Assoc 2009;107:72–77.

Unnikrishnan

Menon

: Thyroid disorders in India: an epidemiological perspective. Indian J Endocr Metab 2011;15(suppl S2):78–81.

Cooper

Doherty

Haugen

: Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association (ATA) guidelines taskforce on thyroid nodules and differentiated thyroid cancer. Thyroid 2009;19:1167–1214.

J-M

X-H

H-X

, et al: Conventional US, US elasticity imaging, and acoustic radiation force impulse imaging for prediction of malignancy in thyroid nodules. Radiology 2014;272:577–586.

Pacini

Schlumberger

Dralle

Elisei

Smit

Wiersinga

: European consensus for the management of patients with differentiated thyroid carcinoma of the follicular epithelium. Eur J Endocrinol 2006;154:787–803.

Bojunga

Herrmann

Meyer

Weber

Zeuzem

Friedrich-Rust

: Real-time elastography for the differentiation of benign and malignant thyroid nodules: a meta-analysis. Thyroid 2010;20:1145–1150.

Magri

Chytiris

Capelli

, et al: Shear wave elastography in the diagnosis of thyroid nodules: feasibility in the case of coexistent chronic autoimmune Hashimoto’s thyroiditis. Clin Endocrinol 2012;76:137–141.

10.

Lippolis

Tognini

Materazzi

, et al: Is elastography actually useful in the presurgical selection of thyroid nodules with indeterminate cytology? J Clin Endocrinol Metab 2011;96:E1826–E1830.

11.

Nightingale

Soo

Nightingale

Trahey

: Acoustic radiation force impulse imaging: in vivo demonstration of clinical feasibility. Ultrasound Med Biol 2002;28:227–235.

12.

Nachiappan

Metwalli

Hailey

Patel

Ostrowski

Wynne

: The thyroid: review of imaging features and biopsy techniques with radiologic-pathologic correlation. Radiographics 2014;34:276–293.

13.

Liu

B-J

H-X

, et al: The diagnosis value of acoustic radiation force impulse (ARFI) elastography for thyroid malignancy without highly suspicious features on conventional ultrasound. Int J Clin Exp Med 2015;8:15362–15372.

14.

Zhang

Mao

D-F

Zhang

M-W

Fan

X-X

: Clinical value of liver and spleen shear wave velocity in predicting the prognosis of patients with portal hypertension. World J Gastroenterol 2017;23:8044–8052.

15.

Palabiyik

İnci

Cakır

EDP

Hocaoglu

: Evaluation of normal thyroid tissue and autoimmune thyroiditis in children using shear wave elastography. J Clin Res Pediatr E 2019;11:132–139.

16.

Rao

: Epidemiological observations of thyroid cancer, in Shah

Samuel

Rao

(eds): Thyroid Cancer-An Indian Perspective. Mumbai, Quest Publications, 1999, pp 3–16.

17.

Cappelli

Castellano

Pirola

, et al: The predictive value of ultrasound findings in the management of thyroid nodules. Q J Med 2007;100:29–35.

18.

Solbiati

Volterrani

Rizzatto

, et al.: The thyroid gland with low uptake lesions: evaluation by ultrasound. Radiology 1985;155:187–191.

19.

Papini

Guglielmi

Bianchini

Crescenzi

Taccogna

Nardi

, et al.: Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clinic Endocrinol Metab 2002;87:1941–1946.

20.

Kim

E-K

Park

Chung

, et al: New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR Am J Roentgenol 2002;178:687–691.

21.

Kovacevic

Smetana Škurla

: Sonographic diagnosis of thyroid nodules: correlation with the results of sonographically guided fine-needle aspiration biopsy. J Clin Ultrasound 2007;35:63–67.

22.

Ivanac

Brkljacic

Ivanac

Huzjan

Skreb

Cikara

: Vascularisation of benign and malignant thyroid nodules: CD US evaluation. Eur J Ultrasound / Ultraschall Med 2007;28:502–506.

23.

Bonavita

Mayo

Babb

, et al: Pattern recognition of benign nodules at ultrasound of the thyroid: which nodules can be left alone? AJR Am J Roentgenol 2009;193:207–213.

24.

Rago

Di Coscio

Basolo

, et al: Combined clinical, thyroid ultrasound and cytological features help to predict thyroid malignancy in follicular and Hurthle cell thyroid lesions: results from a series of 505 consecutive patients. Clin Endocrinol 2007;66:13–20.

25.

Friedrich-Rust

Dauth

Berner

, et al: Differentiation of thyroid nodules using acoustic radiation force impulse imaging. Presented at 15th International & 14th European Congress of Endocrinology 2012, Florence, Italy. Endocrine Abstracts 2012; 29:1814.

26.

Bojunga

Dauth

Berner

, et al: (2012) Acoustic radiation force impulse imaging for differentiation of thyroid nodules. PLoS ONE 7(8):e42735. doi:10.1371/journal.pone.0042735

27.

Friedrich-Rust

Romenski

Meyer

, et al: Acoustic radiation force impulse-imaging for the evaluation of the thyroid gland: a limited patient feasibility study. Ultrasonics 2012;52:69–74.

28.

Bai

, et al: Preliminary study on the diagnostic value of acoustic radiation force impulse technology for differentiating between benign and malignant thyroid nodules. J Ultrasound Med 2012;31:763–771.