The effectiveness of shear wave elastography in the diagnosis of PTMC

Abstract

OBJECTIVE:

The present study explored the efficiency of shear wave elastography (SWE) in diagnosing papillary thyroid microcarcinoma (PTMC).

METHODS:

This retrospective study constituted of 446 patients with 510 thyroid nodules. SWE values and ultrasound parameters of each nodule were used for comparison and analysis.

RESULTS:

The ultrasonography findings between malignant and benign nodules differed significantly ( $P<$ 0.05). The benign and PTMC nodule groups showed a remarkable difference in the stiffness, i.e., higher SWE values.

CONCLUSION:

SWE is useful in diagnosing micro thyroid nodules, and the values were significantly correlated with tumor size and thyroid capsule invasion of PTMC.

Keywords

Shear wave elastography ultrasound thyroid nodules papillary thyroid microcarcinoma

1. Introduction

Papillary thyroid microcarcinoma (PTMC) is defined as a papillary thyroid carcinoma with largest dimensions $\leqslant$ 1.0 cm [1]. Some types of thyroid benign nodules and PTMC appeared to be similar during ultrasound (US) screening, Therefore, identifying an inspection method for the evaluation of PTMC before surgery as a supplementary method to conventional US is essential. Shear wave elastography (SWE) provides a quantitative measurement of stiffness of thyroid nodules in kPa and can be performed during routine thyroid US examinations. The SWE values of thyroid cancer are demonstrated as variable from benign thyroid nodules because of the histological features and specific US parameters [2].

2. Methods

510 nodules (maximum diameter was $<$ 10 mm) from 446 patients (64 males and 382 females, mean age 45.7 $\pm$ 8.1 years, range 24–74 years) were included. The pathological results of all the malignant nodules revealed PTMC. Conventional US (including B-mode US and color Doppler US) and SWE examination were performed using the same scanner (Aixplorer, SuperSonic Imagine, Aix en Provence, France) with a 15–4 MHz linear transducer. Maximum Young’s modulus, Mean Young’s modulus and Minimum Young’s modulus in the region of interest (ROI) were calculated as SWE_max, SWE_mean, and SWE_min independently. The ratio between the SWE_mean of the thyroid nodule and the elastic value of the normal thyroid tissue was calculated as SWE_ratio.

Table 1
Basic characteristics and US features of the nodules and patients selected for the study

Parameters	Overall nodules	Malignant	Benign	$\chi^{2}/$ W	$P$ values
No. of patients	446	349	97
No. of nodules	510	407	103
Sex (male/female)	64/382	50/299	14/83	0.000	1
Age range	24–74	28–66	44–74
Mean age	45.7 $\pm$ 8.1	44.6 $\pm$ 7.7	50.1 $\pm$ 8.4	28817	$<$ 0.0001 ${}^{\ast}$
Size (US)	6.8 $\pm$ 2.2	6.8 $\pm$ 2.2	7.2 $\pm$ 2.2	23284	0.079
Location (Left/right/isthmus)	225/265/20	176/215/16	49/50/4		0.694
Shape (Regular/irregular)	112/398	53/354	59/44	91.4	$<$ 0.0001 ${}^{\ast}$
Margin (Well/poorly defined)	103/407	40/367	63/40	131.2	$<$ 0.0001 ${}^{\ast}$
Internal echo (Uniform/nonuniform)	442/68	363/44	79/24	10.042	0.002 ${}^{\ast}$
Height and width ( $<$ 1/ $>$ 1)	197/313	122/285	75/28	61.84	$<$ 0.0001 ${}^{\ast}$
Calcification (No-/macro-/micro-)	160/101/249	123/63/221	37/38/28	32.30	$<$ 0.0001 ${}^{\ast}$
Vascularity (No-/rare-/rich internal flow)	301/180/29	242/148/17	59/32/12	8.8	0.012 ${}^{\ast}$
SWE_max(kPa)	68.6 $\pm$ 27.2	73.2 $\pm$ 27.6	50.5 $\pm$ 15.6	9675	$<$ 0.0001 ${}^{\ast}$
SWE_mean(kPa)	46.8 $\pm$ 16.3	49.5 $\pm$ 15.9	36.0 $\pm$ 12.8	10500	$<$ 0.0001 ${}^{\ast}$
SWE_min(kPa)	1.9 $\pm$ 0.6	1.9 $\pm$ 0.6	1.5 $\pm$ 0.5	12701	$<$ 0.0001 ${}^{\ast}$
SWE_Ratio(kPa)	25.9 $\pm$ 7.7	26.5 $\pm$ 7.4	23.5 $\pm$ 8.4	15891	$<$ 0.0001 ${}^{\ast}$
SWE_SD(kPa)	10.0 $\pm$ 6.5	10.8 $\pm$ 6.7	6.6 $\pm$ 3.8	11266	$<$ 0.0001 ${}^{\ast}$

${}^{\ast}P$ -value $<$ 0.05 was considered statistically significant.

3. Results

Out of the 510 nodules 103 were benign and 407 were malignant. The basic characteristics and US features of all the nodules and patients selected for the study are summarized in Table 1. The patients with PTMC were younger than the patients with benign mirconodules ( $P<$ 0.05). PTMC is inclined towards an irregular shape, poorly defined margins, non-uniform internal echo, height and width $>$ 1, micro-calcification, and absence of internal flow. SWE values of PTMC were higher than the benign micronodules with a statistical significant difference (all $P<$ 0.05).

The diagnostic efficacy of SWE in distinguishing PTMC and benign thyroid micro-nodules is shown in Fig. 1 and Table 2. The AUC of SWE_max, SWE_mean, and SWE_SD were $>$ 0.7. The optimal cut-off values of SWE_max, SWE_mean, and SWE_SD were set at 64.5 kPa, 34.5 kPa and 7.5 kPa, respectively.

Multiple linear regression indicated that SWE_max, SWE_mean, and SWE_SD values displayed a conspicuous linear relationship with several specific tumor features. The tumor size (postoperative and US) and capsule invasion were the robust pathological determinants of SWE_max and SWE_SD (all $P<$ 0.05). Moreover, the tumor size, which was measured postoperative, significantly influenced the SWE_mean ( $P<$ 0.05), as shown in Table 3. The Wilcoxon signed-rank test that was used to compare the nodule size of PTMC between the US and postoperative measurements, which demonstrated that the size measured by US was significantly higher than the postoperative ( $P<$ 0.0001); the Pearson’s correlation coefficient between the two groups was 0.83 ( $P<$ 0.0001), indicating a strong correlation. The scatter plot and curve fitting of the two sets of data are shown in Fig. 2.

Table 2
Diagnostic efficacy and optimal cut-off values for SWE parameters

Parameters	AUC	95% CI	Cut-off value	Sensitivity (%)	Specificity (%)
SWE_max	0.769 ${}^{\ast}$	(0.722–0.816)	64.5	56.9	83.5
SWE_mean	0.749 ${}^{\ast}$	(0.696–0.803)	34.5	87.2	51.5
SWE_ER	0.697	(0.638–0.757)	1.45	85.5	45.6
SWE_min	0.621	(0.557–0.685)	23.5	64.0	56.3
SWE_SD	0.731 ${}^{\ast}$	(0.677–0.785)	7.5	64.3	83.8

${}^{\ast}$ AUC $>$ 0.7 show a better diagnostic efficiency. CI, confidence interval.

Figure 1.

ROC curves for SWE values.

Table 3

Linear correlation between PTMC features and SWE parameters

Parameters	Coefficient (B)			$P$ value
	SWE_max	SWE_mean	SWE_SD	SWE_max	SWE_mean	SWE_SD
Size (US)	2.74	0.08	1.00	0.009 ${}^{\ast}$	0.902	0.000 ${}^{\ast}$
Size (Postoperative)	3.42	1.90	0.57	0.002 ${}^{\ast}$	0.007 ${}^{\ast}$	0.034 ${}^{\ast}$
Capsule invasion	9.61	3.21	2.35	0.002 ${}^{\ast}$	0.102	0.002 ${}^{\ast}$
Lymphatic metastasis	1.64	3.17	$-$ 0.34	0.734	0.301	0.774
Thyroid function	1.97	0.48	0.16	0.447	0.770	0.796
Shape	1.03	2.22	0.27	0.658	0.131	0.634
Margin	3.55	0.26	0.53	0.158	0.872	0.390
Height and width	4,37	$-$ 0.71	1.32	0.154	0.710	0.077
Echo	6.00	2.97	0.16	0.184	0.300	0.888
Calcification	1.04	0.07	0.58	0.510	0.934	0.131
Vascularity	$-$ 1.47	$-$ 0.64	0.44	0.522	0.660	0.437

${}^{\ast}P$ -value $<$ 0.05 was considered statistically significant.

Figure 2.

Scatter plots and the fitting curves of the size both measured by US and postoperative. Note: The figure shows the linear correlation.

PTMC may occur as early cervical lymph node metastasis or distant metastasis, and some patients may display the metastatic symptoms as the main suit when visiting the medical practitioner [3]. A large number of patients with suspicious thyroid nodules are referred to a diagnostic surgery in the case of indeterminate UG-FNA results. These patients often undergo thyroidectomy to discover a benign pathological outcome of the nodules [4]. Elastography was used as a US-based tool in detecting the stiffness of lesion tissues, which varied structurally from the normal organic tissue. SWE technology reflects the hardness of the sample area by calculating the Young’s modulus value (kPa) of the local tissue; a high value indicates stiff tissue. The SWE images demonstrate a homogeneous blue pattern, representing low Young’s modulus for a maximum number of the benign nodules. On the other hand, an inhomogeneous pattern with the yellow or red area is observed, representing high Young’s modulus for most of the malignant nodules [5].

4. Discussion

The current study indicated that patients with PTMC tend to be younger than those with benign nodules. With respect to the occurrence and development of the thyroid nodule, the biological characteristics of the malignant tumor growth differs distinctly from the benign nodules. These differences are noted as specific manifestations in the ultrasound. As most of PTMC growth is invasive and some lack the tumor-associated fiber coating, it can be characterized by irregular shape or unclear edges [6]. PTMC histology showed the papillary structure of hyperplasia instead of normal thyroid follicular tissue, as well as thyroid cancer cells with overlapping growth; therefore, most of the PTMC can be manifested as an uneven internal echo. Thus, we observed a PTMC characteristic height and width $>$ 1 mm and micro-calcification in standard US examination, which was in accordance with the guidelines for thyroid cancer published by the American Thyroid Association [7]. PTMC also showed a lack of blood for performance resulting in slow growth. Within the two micro-nodule status groups (benign and malignant), a huge difference in the nodules stiffness was seen, indicating that essential tumor biological factors effectuate the change in nodule stiffness.

In the study, we found that SWE_max, SWE_mean, and SWE_SD had a higher diagnostic efficiency. SWE_max64.5-kPa, SWE_mean34.5-kPa, and SWE_SD7.5-kPa could be used as a reference standard for distinguishing between benign and malignant thyroid micro nodules. This outcome was similar to that of Sebag et al. [8], who applied SWE as the measure of thyroid disease and established SWE_max 65 kPa as the threshold for differential diagnosis of benign and malignant thyroid nodules. During the occurrence of PTMC, the normal follicles of the tissue were damaged or replaced by abnormal tumor tissue, interstitial components were reduced, PTMC nodules received insufficient blood supply causing tissue necrosis, and partial nodules with calcium showed salt deposition. All the above features could cause the increase of PTMC hardness compared with the benign nodules or normal thyroid tissue. SWE_max and SWE_SD had a higher specificity (0.835 and 0.838 respectively). SWE_mean had higher sensitivity (0.872), which was in agreement with that of Duan et al. [9].

We found a significant linear correlation between the SWE value and some features or alterations of the nodules. SWE_max and SWE_SD were significantly correlated with the size of the nodules (statistical results showed that the size of the nodules measured by US was positively correlated with the size of the nodules measured postoperative), whereas SWE_mean was correlated with the actual size of PTMC. This may be due to the increase in the size of the nodule that causes a gradual change in its internal components, resulting in a gradual increase of the hardness, and the difference between the surrounding tissue or benign nodules are eventually apparent.

However, this study had several limitations. We did not explore the correlation between the values of nodules’ stiffness and other critical tumor characteristics, such as thyroid hormones and PTMC gene information. In addition, we did not consider the cases of specific nodules in the process of SWE examination and data analysis, such as the position deep or close to the isthmus, which might give rise to the differences between the data and the real value [10].

In conclusion, for micro thyroid nodules, the sonographic findings of benign and malignant nodules were significantly different. SWE exhibits specific diagnostic performances while identifying PTMC, especially SWE_max and SWE_SD. These two values were linearly correlated with the size of PTMC lesions, and could be more accurate in preoperative prediction of the presence of PTMC capsule invasion. Therefore, shear wave elastography can be a valuable and complementary tool in identifying PTMC from thyroid micronodules with extensive application prospects.

Footnotes

Conflict of interest

None to report.

References

Cooper

Doherty

Haugen

, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid. 2009; 19: 1167-1214.

Kim

Son

Youk

. Quantitative assessment of shear-wave ultrasound elastography in thyroid nodules: diagnostic performance for predicting malignancy. Eur Radiol. 2013; 23: 2532-2537.

Wada

Duh

Sugino

, et al. Lymph node metastasis from 259 papillary thyroid microcarcinoma: frequency, pattern of occurrence and recurrence, and optimal strategy for neck dissection. Ann Surg. 2003; 237: 399-407.

Liu

Zhang

, et al. Acoustic radiation force impulse elastography for differentiation of benign and malignant thyroid nodules with concurrent Hashimoto’s thyroiditis. Med Oncol. 2015; 32, 50.

Liu

Liang

Zheng

, et al. Two-dimensional shear wave elastography as promising diagnostic tool for predicting malignant thyroid nodules: a prospective single-centre experience. European Society of Radiology. 2015; 25: 624-634.

Wang

Cao

Shao

, et al. Retrospectively analyzed and literature review in 1239 cases of thyroid nodules. Modern Oncology. 2015; 22: 921-925.

Haugen

Alexander

Bible

, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016; 26: 1-133.

Sebag

Vaillant-Lombard

Berbis

, et al. Shear wave elastography: a new ultrasound imaging mode for the differential diagnosis of benign and malignant thyroid nodules. Clin Endocrinol Metab. 2010; 95: 5281-5288.

Duan

, et al. Diagnostic value of two-dimensional shear wave elastography in papillary thyroid microcarcinoma. OncoTargets and Therapy. 2016; 9: 1311-1317.

10.

Bhatia

Tong

Cho

, et al. Shear wave elastography of thyroid nodules in routine clinical practice: Preliminary observations and utility for detecting malignancy. Eur Radiol. 2012; 22: 2397-2406.