Comparison of Virtual Touch Tissue Imaging & Quantification (VTIQ) and Toshiba shear wave elastography (T-SWE) in diagnosis of thyroid nodules: Initial experience

Abstract

OBJECTIVE: The aim of this study was to compare the diagnostic performance of two different 2D shear wave speed imaging techniques of Virtual Touch Tissue Imaging & Quantification (VTIQ) and Toshiba shear wave elastography (T-SWE) in predicting malignant thyroid nodules (TNs).

MATERIALS AND METHODS: 75 TNs in 75 patients which were subject to both VTIQ and T-SWE examinations were enrolled and analyzed. Shear wave speed (SWS) values on VTIQ and T-SWE were computed (SWS_max, min, mean and median). Area under the receiver operating characteristic (AUROC) curve was obtained to assess the diagnostic performance.

RESULTS: The AUROC for VTIQ was the highest with SWS_min whereas for T-SWE was SWS_max (0.774 versus 0.851; p > 0.05). The AUROC, sensitivity and negative predictive value (NPV) corresponding to SWS_max for VTIQ were significantly lower than those for T-SWE (0.717 versus 0.851, 61.5% versus 92.3% and 78.7% versus 94.3%; all p < 0.05). However, no significant differences were found between AUROC with SWS_min, SWS_mean, or SWS_median for VTIQ and SWS_max for T-SWE (all p > 0.05).

CONCLUSION: In general, VTIQ is equal to T-SWE for diagnosis of TNs. In the clinical practice, the selection of SWS_max should be avoided in VTIQ whereas should be selected in T-SWE.

Keywords

Acoustic radiation force impulse shear wave elastography ultrasound thyroid nodules fine-needle aspiration diagnostic performance

1 Introduction

Thyroid nodule (TN) is one of the most common endocrine diseases and presents increasing incidence year by year [1]. In the recent 20 years, the incidence of papillary thyroid cancer (PTC) increased sharply on the average annual rate of 7.8% in men and 7.2% in women [2]. Therefore, it is particularly important to differentiate malignant from benign TNs.

Conventional ultrasound (US), a widely used noninvasive diagnostic method, plays an important role in efficiently checking nodules for patients with thyroid diseases [3, 4]. Nevertheless, US alone provides variable diagnostic information for differentiating malignant from benign TNs [5, 6]. In addition to conventional US, US elastography has recently emerged to detect malignant TNs by assessing the stiffness of tissue and malignant lesions are usually stiffer than benign ones [7 –9]. All elatography technologies assess the tissue stiffness by observing the deformation which occurs in response to an applied force. Then the observed deformation is converted to a suitable form for display [10]. Currently, many studies have proven that US elatography is a promising tool that can be performed to assist conventional US in distinguishing malignant TNs [11 –13].

According to the recent guidelines, the US elastography methods can be divided into 3 categories [10].

The first is displacement or strain imaging includes quasi-static strain imaging and acoustic radiation force impulse (ARFI) imaging (i.e. Virtual Touch Tissue Imaging, VTI; Siemens Medical Solutions, Mountain View, CA, USA). The former evaluates tissue elasticity under an external force (i.e. manual compression with the probe, cardiovascular or respiratory pulsation) by measuring the shape deformation. However, the former only provides qualitative or semi-quantitative information but not quantitative stiffness information. Moreover, high examiner dependence and poor reproducibility are also the limitations [14 –20]. For VTI, though it is less examiner-dependent and more reproducible using the excitation of acoustic impulse, it cannot provide quantitative elastic information.

The second category names shear-wave speed (SWS) measurement includes transient elastography (TE) and ARFI quantification (i.e. Virtual Touch Tissue Quantification, VTQ; Siemens Medical Solutions, Mountain View, CA, USA), which can quantitatively calculate SWS for assessing tissue stiffness. However, as a point shear wave elastography (p-SWE), VTQ only measures SWS in a target region whereas is failed to obtain the SWS distribution information in the whole lesion.

The third category is two-dimensional (2D) SWS imaging, in which 2D map of shear wave (SW) distribution in the lesion is available [5 , 21–24]. Both p-SWE and 2D-SWE is based on capturing the propagation of SWs induced by displacement of local tissues under the excitation of acoustic impulses from probes. The elastic property is reflected by shear or Young’s modulus that is estimated from SWS. According to different types of US machines, the elastic property can be displayed as SWS expressed in meters per second (m/s) or Young’s modulus expressed in kilopascals (kPa).

Currently, the widely used 2D SWS imaging techniques include SuperSonic Imagine (SSI; Aix en Provence, France), Virtual Touch Tissue Imaging & Quantification (VTIQ; Siemens Medical Solutions, Mountain View, CA, USA) and Toshiba SWE (T-SWE; Toshiba Medical System, Tochigi, Japan). Several studies have demonstrated promising application of SSI in diagnosis of TNs [24 –28]. However, so far publications about VTIQ in thyroid are few and no literature about T-SWE has been found [29]. Comparisons between different 2D SWS imaging techniques are relevant in clinical practice under the background of increasing controversy over the usefulness of SWE in diagnosing TN, which is helpful to make a clarification whether the controversy is originated from the techniques used or the specific conditions for TNs. The aim of this study was to compare the diagnostic performance of VTIQ and T-SWE in the prediction of malignancy in TNs.

2 Materials and methods

2.1 Patient enrollment

This study was approved by the institutional review board and informed consent was obtained from all participating patients to include their data for analysis. From July 2015 to August 2015, a total of 122 consecutive patients with TNs checked by conventional US underwent both VTIQ and T-SWE examinations in the university hospital. The inclusion criteria for TNs were as follows: (1) TNs detected by conventional US; (2) solid TNs or predominant solid nodules (cystic part <25%); (3) with sufficient thyroid tissue surrounding the nodule; (4) without any previous treatment for the TNs. TNs were excluded owing to the following reasons: (1) cytology or histology results of nodules were not available (36 were excluded); (2) TNs with incomplete elastic images (11 were excluded). For patients with multiple nodules, only the most suspicious or the largest lesion was analyzed. Eventually, 75 patients with 75 lesions were included in this study. All the patients underwent both VTIQ and T-SWE examinations in the same setting.

2.2 Measurement protocol of VTIQ

Before VTIQ and T-SWE, all the nodules underwent a thorough conventional US examination with a linear transducer for which the frequency was usually above 10 MHz. The 2D-SWE examinations were carried out after the target nodule was determined with conventional US. VTIQ was performed by a radiologist with more than 2 years of experience in US elastography with the Acuson S3000 US system (Siemens Medical Solutions, Mountain View, CA, USA), which was equipped with a 9L4 linear array transducer (frequency range, 4–9 MHz). Patients were asked to lie on the bed in a supine position with head leaning back slightly, so as to fully expose the thyroid. Afterward, the transducer was placed perpendicular to the target nodule. The unconscious compression to the probe by the operator may result in inaccuracy of elastic values [25, 30], so, the operator should hold the transducer as slightly as possible in order to minimize pre-compression distortion. SW images were obtained while patients were told not to swallow or breathe for a few seconds. As a 2D_color image, SW-quality map which indicates whether the SW is of sufficient magnitude and signal-to-noise ratio (SNR) can be obtained firstly to evaluate the quality of SW propagation in the nodule [23]. In SW-quality map, green represents high quality, orange and yellow are for low quality and marginal quality respectively (Fig. 1a). The SW-velocity map was then obtained to display SWS distribution in a 2D_color map. Different colors in SW-velocity map represent the SWS by colors of red for high, blue for low and yellow or green for intermediate (Fig. 1b). SWS was subsequently measured by placing a small SW region of interest (SW-ROI) over the SW-velocity map and the square of the SW-ROI was around 2 mm×2 mm. Cystic, calcified and necrotic area were avoided when placing the SW-ROI on SW-velocity map because these factors may confound measuring results of SWE. The areas corresponding low or marginal quality on SW-quality map were also avoided when measuring SWS. The measurement of SWS on each lesion was repeated for five times. Five SW-ROIs can be placed arbitrarily in the lesions with homogeneous SW distribution; while in the lesions with heterogeneous SW distribution, two SW-ROIs were respectively placed on the highest stiffness area and lowest stiffness area corresponding to the SW distribution on SW-velocity map and the other three SW-ROIs were placed in the remaining areas at random. With SW-ROI, the SWS can be quantitatively measured in m/s. The scale of SWS ranges from 0.5 to 10 m/s and was not adjusted during the whole measuring. Five SWS values were then obtained to compute the value of SWS_max, min, mean and median for the following comparison with T-SWE.

Fig.1

Two display modes in Virtual Touch Tissue Imaging & Quantification (VTIQ). (a) Shear wave-quality map. (b) Shear wave-velocity map.

2.3 Measurement protocol of T-SWE

Afterwards, an Aplio500 US machine (Toshiba Medical System, Tochigi, Japan) was used by the same radiologist for measuring SWS with a 14L5 liner array transducer (frequency range, 5–14 MHz). The patient position, breath holding and no swallowing, probe location, placement and size of SW-ROI, the number of measurement and SWS computing were the same for T-SWE as for VTIQ. T-SWE displays not only an SW-velocity mode in m/s (range, 0–8 m/s) but also an SW-elasticity mode in kPa (range, 10–100 kPa) to show the stiffness distribution of lesions. For the two modes, the velocities increase along with the square root of Young’s modulus. We used SW-velocity mode in this study. The obtained SW-velocity mode was under the “continuous shot scan” mode in which frame rate can be selected among three levels (0.4, 0.7 and 1.0 frames per second) in real time. When a well color pigmented image occurred under the “continuous shot scan” mode, the image was frozen for following elastic measurement. In addition, SW propagation mode (arrival time contour) is available which is shown as contour lines and can be used to verify the quality of SW propagation by observing whether the contour lines are smooth and parallel to each other (Fig. 2). The higher SWS is (i.e. the harder the lesion is), the wider distance of contour lines is depicted, and vice versa. Different intervals between contour lines can be displayed by different density distribution of curves in propagation mode. For areas with parallel contour lines, the reliability of image and data obtained are both high. For areas with distorted and unparallel contour lines, another SWE image acquisition is needed by reason of poor reliability. The propagation mode is used to guide the SW-ROI placement in the nodule and the SW-ROI is generally placed on areas with parallel contour lines.

Fig.2

Three display modes in Toshiba shear wave elastography (T-SWE). (a) Shear wave-velocity mode. (b) Shear wave-elasticity mode. (c) Propagation (arrival time contour) mode.

2.4 Inter-operator and intra-operator consistency of VTIQ and T-SWE

Another consecutive thirty patients with TNs who didn’t participate in the diagnostic performance analysis were enrolled for the test of inter-operator consistency of VTIQ and T-SWE which were performed by another two independent operators with the same experience on SWE. To evaluate intra-operator consistency, US elastography was conducted by the same operator and the time interval between the two procedures was one day. All the measurement was the same as the above mentioned method.

2.5 Statistical analysis

All the statistical analyses were conducted with the SPSS 20.0 software (SPSS, Chicago, IL) and MedCalc software (Mariakerke, Belgium). Means and standard deviations were used for continuous data while counts and percentages for categorical data. The differences in SWS between benign and malignant TNs, as well as between VTIQ and T-SWE were performed with independent-samples or paired-samples t test. SWS values referring to pathology results were collected to construct receiver operating characteristic (ROC) curve which can be used to obtain area under ROC (AUROC), sensitivity, specificity, positive and negative predictive values (PPV and NPV). The best cutoff value was calculated when Youden index (YI) was maximum (sensitivity + specificity –1). Comparisons of AUROC were conducted at two levels: at a parallel level, the comparisons among four SWS groups (SWS_max, min, mean, and median) of VTIQ or T-SWE and at a vertical level, the comparisons between VTIQ and T-SWE. Comparisons of AUROC were evaluated by a univariate Z score test. The AUROC corresponding to four SWS groups can be expressed in AUROC_max, min, mean, and median, respectively. Because of the paired_samples in this study, differences in sensitivity, specificity, accuracy, PPV and NPV between VTIQ and T-SWE were compared using the McNemar test or chi-square test. Inter-operator and intra-operator consistency of VTIQ and T-SWE were evaluated with Intra-class correlation coefficient. Two-tailed p values < 0.05 were considered to be statistically significant for all analyses in this study.

3 Results

3.1 Basic characteristics

The included 75 patients consisted of 12 men and 63 women, with a mean age of 49.6±13.6 (range, 22–79 years). The mean age was 46.7±13.6 (range, 22–79 years) for female patients and 49.8±13.7 (range, 28–68 years) for male patients. The diameter of TNs ranged from 3 mm to 50 mm (mean size, 12.6±9.7 mm). Pathology diagnoses of TNs were confirmed by fine-needle aspiration biopsy (FNAB) or surgery in which 49 (65.3%) benign lesions and 26 (34.7%) malignant lesions were confirmed. For malignant TNs, 20 out of 26 TNs were confirmed by surgery while 6 by FNAB (Bethesda category VI) [31]. For benign TNs, 40 out of 49 TNs were confirmed by FNAB (Bethesda category II) [31] in which the interval of follow_up with US imaging was 6 months except for the other 9 TNs confirmed by surgery. All the malignant lesions were PTCs.

3.2 VTIQ and T-SWE

The SWS_max, min, mean and median in VTIQ and T-SWE were all significantly higher in malignant nodules than those in benign ones (Table 1) (Figs. 3 and 4). The AUROC_max, min, mean and median in VTIQ and T-SWE were compared: at a parallel level, the comparisons of AUROC among the four groups showed no significant differences both in VTIQ and T-SWE (all p > 0.05) and at a vertical level, the comparison of AUROC_max between VTIQ and T-SWE showed statistically significant difference (p = 0.002) (Fig. 5). However, the best performances of AUROC_min in VTIQ and AUROC_max in T-SWE were also compared and there was no statistical significance (p > 0.05). In addition, no significant differences were found between AUROC_mean or AUROC_median in VTIQ and AUROC_max in T-SWE (both p > 0.05) (Table 2). The cut-off values of SWS_max for all TNs were 3.77 m/s in VTIQ and 3.51 m/s in T-SWE respectively and there was statistically significant difference in SWS_max for malignant TNs between VTIQ and T-SWE (p = 0.015) (Fig. 6). With regard to SWS_max, the sensitivity, specificity, accuracy, PPV and NPV were 61.5% versus 92.3%, 75.5% versus 67.3%, 70.7% versus 76.0%, 57.1% versus 60.0% and 78.7% versus 94.3% in VTIQ and T-SWE, respectively. Among them, sensitivity and NPV in terms of AUROC_max showed significant differences between VTIQ and T-SWE (p < 0.05) (Table 3).

Table 1
Comparisons of SWS indices between papillary thyroid carcinomas (PTCs) and benign thyroid nodules for VTIQ and T-SWE

SWS indices(m/s)^♦ Elastography PTC (n = 26) Benign (n = 49) p

SWS_max VTIQ 4.11±0.90 (2.84–6.81) 3.56±0.74 (2.62–6.42) 0.006

T-SWE 4.43±0.95 (3.24–6.47) 3.39±0.73 (2.17–5.71) <0.001

SWS_min VTIQ 3.50±0.66 (2.68–5.30) 2.95±0.47 (1.96–4.42) <0.001

T-SWE 3.46±0.90 (2.05–5.83) 2.63±0.60 (1.22–4.44) <0.001

SWS_mean VTIQ 3.81±0.72 (2.78–5.58) 3.24±0.60 (2.12–5.09) <0.001

T-SWE 3.90±0.86 (2.48–6.00) 3.02±0.65 (1.96–5.14) <0.001

SWS_median VTIQ 3.82±0.70 (2.76–5.76) 3.23±0.56 (2.17–5.16) <0.001

T-SWE 3.83±0.83 (2.45–5.88) 3.06±0.69 (1.99–5.26) <0.001

SWS indices(m/s)^♦	Elastography	PTC (n = 26)	Benign (n = 49)	p
SWS_max	VTIQ	4.11±0.90 (2.84–6.81)	3.56±0.74 (2.62–6.42)	0.006
	T-SWE	4.43±0.95 (3.24–6.47)	3.39±0.73 (2.17–5.71)	<0.001
SWS_min	VTIQ	3.50±0.66 (2.68–5.30)	2.95±0.47 (1.96–4.42)	<0.001
	T-SWE	3.46±0.90 (2.05–5.83)	2.63±0.60 (1.22–4.44)	<0.001
SWS_mean	VTIQ	3.81±0.72 (2.78–5.58)	3.24±0.60 (2.12–5.09)	<0.001
	T-SWE	3.90±0.86 (2.48–6.00)	3.02±0.65 (1.96–5.14)	<0.001
SWS_median	VTIQ	3.82±0.70 (2.76–5.76)	3.23±0.56 (2.17–5.16)	<0.001
	T-SWE	3.83±0.83 (2.45–5.88)	3.06±0.69 (1.99–5.26)	<0.001

SWS = shear wave speed; SWS_max = maximum speed index; SWS_min = minimum speed index; SWS_mean = mean speed index; SWS_median = median speed index; PTC = papillary thyroid carcinomas; VTIQ = Virtual Touch Tissue Imaging & Quantification; T-SWE = Toshiba shear wave elastography. ^♦The format of the numbers are mean±SD (range).

Fig.3

A benign nodule confirmed by cytological result and follow up in a 38-year-old man. (a) Gray-scale ultrasound image shows a solid hypoechogenic nodule in the right thyroid lobe (arrows). (b) For VTIQ, the five measurements of shear wave speed (SWS) in the lesion range from 2.36–2.63 m/s and the mean SWS is 2.52 m/s. (c) For T-SWE, the five measurements of shear wave speed (SWS) in the lesion range from 1.97–2.77 m/s and the mean SWS is 2.40 m/s.

Fig.4

Papillary thyroid carcinoma confirmed by histopathological result in a 37-year-old woman. (a) Gray-scale ultrasound image shows a solid hypoechogenic nodule in the right thyroid lobe (arrows). (b) For VTIQ, the five measurements of SWS in the lesion range from 5.30–5.90 m/s and the mean SWS is 5.58 m/s. (c) For T-SWE, the five measurements of SWS in the lesion range from 5.32–6.47 m/s and the mean SWS is 6.02 m/s.

Fig.5

Receiver operating characteristic (ROC) curves corresponding to SWS_max, SWS_min, SWS_mean and SWS_median in VTIQ and T-SWE. VTIQ = Virtual Touch Tissue Imaging & Quantification; T-SWE = Toshiba shear wave elastography; SWS = shear wave speed.

Table 2

Comparison of AUROCs at a parallel level and a vertical level for VTIQ and T-SWE

Elastography	AUROC_max	AUROC_min	AUROC_mean	AUROC_median	p ^★
VTIQ	0.717	0.774^♣	0.767^♣	0.764^♣	>0.05^♦
	(0.593, 0.841)	(0.667, 0.882)	(0.654, 0.879)	(0.652, 0.876)
T-SWE	0.851^♣	0.796	0.829	0.803	>0.05^♦
	(0.766, 0.935)	(0.687, 0.906)	(0.734, 0.924)	(0.701, 0.905)
p ^★★	0.002	0.614	0.099	0.371

VTIQ = Virtual Touch Tissue Imaging & Quantification; T-SWE = Toshiba shear wave elastography; AUROC = area under the receiver operating characteristic curve. p^★ values indicate comparisons of AUROC among AUROC_max, min, mean and median for VTIQ or T-SWE at a parallel level. ^♦There are all no significant differences among the four groups in VTIQ or T-SWE at a parallel level. P^★★ values indicate comparisons of AUROC between VTIQ and T-SWE at a vertical level, namely AUROC_max of VTIQ versus AUROC_max of T-SWE, AUROC_min of VTIQ versus AUROC_min of T-SWE, AUROC_mean of VTIQ versus AUROC_mean of T-SWE, AUROC_median of VTIQ versus AUROC_median of T-SWE. ^♣Indicates no statistically significant difference.

Fig.6

Box-and-whisker plot of maximum shear wave speed (SWS_max) was estimated using VTIQ and T-SWE methods. The difference of SWS_max in benign lesions between VTIQ and T-SWE is not statistically significant (p > 0.05), whereas in the case of malignant lesions, the difference between VTIQ and T-SWE is statistically significant (p = 0.015). Boxes represent the values from lower to upper quartiles; central lines represent medians; dots represent outliners; and the whiskers extend from minimal to maximal values.

Table 3

Diagnostic performances of VTIQ versus T-SWE

ROC indices	Elastography	Sensitivity%	Specificity%	Accuracy%	PPV%	NPV%	Cut-off (m/s)
ROC_max	VTIQ	61.5^♦	75.5	70.7	57.1	78.7^♦	3.77
	T-SWE	92.3^♦	67.3	76.0	60.0	94.3^♦	3.51
ROC_min	VTIQ	80.8	63.3	69.3	53.8	86.1	3.06
	T-SWE	76.9	73.5	74.7	60.6	85.7	2.86
ROC_mean	VTIQ	65.4^♦	83.7	77.3	68.0	82.0	3.51
	T-SWE	92.3^♦	69.4	77.3	61.5	94.4	3.91
ROC_median	VTIQ	65.4^♦	81.6^♦	76.0	65.4	81.6	3.51
	T-SWE	96.2^♦	61.2^♦	73.3	56.8	96.8	3.03

VTIQ = Virtual Touch Tissue Imaging & Quantification; T-SWE = Toshiba shear wave elastography. ^♦There were statistically significant differences between VTIQ and T-SWE.

3.3 Inter-operator and intra-operator consistency of VTIQ and T-SWE in SWS_max

SWS_max was used to assess the inter-operator and intra-operator consistency of VTIQ and T-SWE. The correlation coefficients were 0.853 and 0.814 for inter-operator while 0.921 and 0.922 for intra-operator corresponding to VTIQ and T-SWE respectively.

4 Discussion

Since US elastography was firstly proposed as a noninvasive tool to estimate the stiffness of tissue by Ophir et al. in 1991 [32], the elastography has been developed for decades. The p-SWE technology such as VTQ was firstly emerged to quantitatively assess the stiffness of tissue. On the basis of VTQ, VTIQ was designed as a 2D-SWE technology. After that, T-SWE was also developed as a 2D-SWE. Both VITQ and T-SWE can provide quantitative and qualitative information for better evaluating the stiffness of tissue.

Based on the latest SWE technologies, if quantitative data were applied such as SWS values in detecting malignant thyroid lesions, a promising diagnostic performance can be obtained. As is known to us, VTQ and VTIQ, the two types of quantitative measurement, both belong to SWE [33]. According to previous study conducted by Xu et al., SWS_mean of at least 2.87 m/s on VTQ was the best cutoff value in distinguishing malignant from benign TNs [17]. However, in the present study the cutoff value of SWS_mean on VTIQ was 3.06 m/s which was higher than 2.87 m/s. The difference made by VTIQ and VTQ are believed to be due to the following factors: First, difference might be present between the two subject groups. Second, a 2D-SWS map can be provided by VTIQ in which the stiffest region can be selected according to SWS distribution expressed in various colors, whereas there is no corresponding 2D map as a reference in VTQ when SWS is measured [5]. Third, the size of SW-ROI in VTQ is different from that in VTIQ. With size of 6 mm×5 mm that cannot be adjusted, the SW-ROI in VTQ represents an average SWS value and the stiffness of most hardest region may be reduced by the softer portions in the lesion while with size of SW-ROI is 2 mm×2 mm in our study, the stiffness distribution of SW-ROI in VTIQ is more concentrated [34]. Additionally, as mentioned above, different degrees of pre-compression to transducer may be inevitable which might cause wide variation in measuring the stiffness of TNs [25, 30].

Recently, a study documented that the diagnostic performance of VTIQ for TNs in which the cutoff value corresponding to SWS_max concurred with the result in our study (3.54 m/s versus 3.77 m/s) [5]. Moreover, with regard to cutoff value, the diagnostic performance of AUROC for VTIQ in that study was also similar to our study (0.78 versus 0.71) [5]. Consequently, the results acquired from our study can be validated by the previous research of VTIQ for thyroid.

Our study showed well inter- and intra- operator consistency of VTIQ and T-SWE which implies that both VTIQ and T-SWE are robust to assess thyroid elasticity with high reproducibility. What is more, the diagnostic performances of AUROCs among VTIQ or T-SWE related to four SWS groups were compared and there were no significant differences among the four AUROCs, indicating a stable diagnostic efficiency for VTIQ or T-SWE per se. Nevertheless, at a vertical level, the comparisons of AUROC between VTIQ and T-SWE showed a statistical difference in AUROC_max_. Furthermore, with regard to AUROC_max_, it revealed inferiority for VTIQ in sensitivity and NPV compared with T-SWE. The underlying causes might be associated with the following reasons: First, there are two scan modes can be selected for T-SWE: “continuous shot scan” for evaluating areas that are likely to be affected by cardiovascular motion and “one shot scan” for quantification which is the same as VTIQ. In our study, “one shot scan” was used in VTIQ, while “continuous shot scan” was used in T-SWE to avoid the influence of carotid pulsation and choose a better pigmented elastic image in real time. Second, Chang et al. [35] documented different frequency of probe can cause the variability of SWS values on liver and a low frequency probe can get higher SWS at the same depth. Consequently, the same principle may be applied to thyroid for VTIQ with 4–9 MHz and T-SWE with 5–14 MHz probes, which can also explain the cutoff SWS_max value of VTIQ was slightly higher than that of T-SWE with 3.77 m/s versus 3.51 m/s. Third, as discussed above, pre-compression to probe may also be an important factor leading to discrepancy of diagnostic performances between VTIQ and T-SWE. However, no significant differences were found between AUROC_min, AUROC_mean or AUROC_median in VTIQ and AUROC_max in T-SWE. Therefore, in the clinical practice, perhaps the selection of SWS_max in VTIQ should be avoided whereas should be recommended in T-SWE.

For evaluating the diagnostic performance of TNs, SWS_max, min, mean and median in VTIQ and T-SWE were all calculated in our study which provided more complete elastic information of thyroid compared with the previous study for VTIQ in thyroid [5]. However, several limitations need to be addressed in this study. The first was small sample size, with only 26 malignant and 49 benign thyroid lesions, which may cause suboptimal accuracy of diagnostic results. Second, selection bias may exist because only patients with cytology or histology results were included. Third, six out of 26 malignant TNs were confirmed by FNAB instead of surgery, which may make pathology results not be so precise because false positive rate of a malignant FNAB interpretation is 1–3% [31]. Fourth, all malignant thyroid lesions were PTCs and other types of carcinomas were not included. Hence, the diagnostic efficiency of our study may not be applied to other research and future studies contain various thyroid carcinomas are needed. Because all the above factors would affect VTIQ and T-SWE equally, it may not bias the comparison between VTIQ and T-SWE. However, more prospective studies are necessary for supporting or overthrowing our results for comparison between VTIQ and T-SWE.

To conclude, VTIQ and T-SWE are comparable in diagnosing TNs. Reproducibility of VTIQ and T-SWE are both favorable. However, with SWS_max, VTIQ showed inferior diagnostic performance in comparison with T-SWE in AUROC, sensitivity and NPV. Considering the application in clinic, the selection of SWS_max in VTIQ should be avoided whereas should be recommended in T-SWE. For further research, not only multi-center and large scale studies are needed, but also assessment of VTIQ or T-SWE combined with conventional US is also necessary.

Footnotes

Acknowledgments

Supported in part by the Shanghai Hospital Development Center (Grant SHDC 12014229), the Science and Technology Commission of Shanghai Municipality (Grants 14441900900 and 15411969000), the Shanghai Municipal Human Resources and Social Security Bureau (Grant 2012045) and the National Natural Scientific Foundation of China (Grants 81401417 and 81501475).

References

Sprague

B.L.

and Warren

, Andersen and A. Trentham-Dietz, Thyroid cancer incidence and socioeconomic indicators of health care access, Cancer Causes & Control: CCC 19(6) (2008), 585–593.

Colonna

, Uhry

, Guizard

A.V.

, Delafosse

, Schvartz

, Belot

, et al. Recent trends in incidence, geographical distribution, and survival of papillary thyroid cancer in France, Cancer Epidemiology 39(4) (2015), 511–518.

Liu

B.X.

, Xie

X.Y.

, Liang

J.Y.

, Zheng

Y.L.

, Huang

G.L.

, Zhou

L.Y.

, et al. Shear wave elastography versus real-time elastography on evaluation thyroid nodules: A preliminary study, European Journal of Radiology 83(7) (2014), 1135–1143.

Kim

, Kim

J.A.

, Son

E.J.

and Youk

J.H.

, Quantitative assessment of shear-wave ultrasound elastography in thyroid nodules: Diagnostic performance for predicting malignancy, European Radiology 23(9) (2013), 2532–2537.

Azizi

, Keller

J.M.

, Mayo

M.L.

, Piper

, Puett

, Earp

K.M.

, et al., Thyroid Nodules and shear wave elastography: A new tool in thyroid cancer detection, Ultrasound in Medicine & Biology 41(11) (2015), 2855–2865.

Liu

, Liang

, Zheng

, Xie

, Huang

, Zhou

, et al., Two-dimensional shear wave elastography as promising diagnostic tool for predicting malignant thyroid nodules: A prospective single-centre experience, European Radiology 25(3) (2015), 624–634.

Sun

, Cai

and Wang

, Real-time ultrasound elastography for differentiation of benign and malignant thyroid nodules: A meta-analysis, Journal of Ultrasound in Medicine: Official Journal of the American Institute of Ultrasound in Medicine 33(3) (2014), 495–502.

Monpeyssen

, Tramalloni

, Poiree

, Helenon

and Correas

J.M.

, Elastography of the thyroid, Diagnostic and Interventional Imaging 94(5) (2013), 535–544.

Zaleska-Dorobisz

, Kaczorowski

, Pawlus

, Puchalska

and Inglot

, Ultrasound elastography - review of techniques and its clinical applications, Advances in Clinical and Experimental medicine: Official Organ Wroclaw Medical University 23(4) (2014), 645–655.

10.

Bamber

, Cosgrove

, Dietrich

C.F.

, Fromageau

, Bojunga

, Calliada

, et al., EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography, Part 1: Basic principles and technology, Ultraschall in der Medizin (Stuttgart, Germany: 1980) 34(2) (2013), 169–184.

11.

Park

A.Y.

, Son

E.J.

, Han

, Youk

J.H.

, Kim

J.A.

and Park

C.S.

, Shear wave elastography of thyroid nodules for the prediction of malignancy in a large scale study, European Journal of Radiology 84(3) (2015), 407–412.

12.

Fukuhara

, Matsuda

, Izawa

, Fujiwara

and Kitano

, Utility of Shear Wave Elastography for Diagnosing Chronic Autoimmune Thyroiditis, Journal of Thyroid Research 2015 (2015), 164548.

13.

Hekimoglu

, Yildirim Donmez

, Arsla

, Ozdemi

, Demi

and Yazic

, The role of shear wave elastography in the diagnosis of chronic autoimmune thyroiditis, Medical Ultrasonography 17(3) (2015), 322–326.

14.

Cantisani

, Lodise

, Grazhdani

, Mancuso

, Maggini

, Di Rocco

, et al., Ultrasound elastography in the evaluation of thyroid pathology, Current status, European Journal of Radiology 83(3) (2014), 420–428.

15.

Cantisani

, Grazhdani

, Drakonaki

, D’Andrea

, Di Segni

, Kaleshi

, et al., Strain US elastography for the characterization of thyroid nodules: Advantages and limitation, International Journal of Endocrinology 2015 (2015), 908575.

16.

Lin

, Chen

, Liu

, Wang

and Li

, Diagnostic performance of shear wave elastography in the identification of malignant thyroid nodules: A meta-analysis, European Radiology 24(11) (2014), 2729–2738.

17.

J.M.

, Xu

X.H.

, Xu

H.X.

, Zhang

Y.F.

, Zhang

, Guo

L.H.

, et al., Conventional US, US elasticity imaging, and acoustic radiation force impulse imaging for prediction of malignancy in thyroid nodules, Radiology 272(2) (2014), 577–586.

18.

J.M.

, Xu

X.H.

, Xu

H.X.

, Zhang

Y.F.

, Guo

L.H.

, Liu

L.N.

, et al., Prediction of cervical lymph node metastasis in patients with papillary thyroid cancer using combined conventional ultrasound, strain elastography, and acoustic radiation force impulse (ARFI) elastography, European Radiology 26(8) (2016), 2611–2622.

19.

Zhan

, Jin

J.M.

, Diao

X.H.

and Chen

, Acoustic radiation force impulse imaging (ARFI) for differentiation of benign and malignant thyroid nodules–A meta-analysis, European Journal of Radiology 84(11) (2015), 2181–2186.

20.

Chen

B.D.

, Xu

H.X.

, Zhang

Y.F.

, Liu

B.J.

, Guo

L.H.

, Li

D.D.

, et al., The diagnostic performances of strain elastography (SE), acoustic radiation force impulse (ARFI) and point shear-wave speed (pSWS) for non-calcified thyroid nodules, Clinical Hemorheology and Microcirculation (2016). [Epub ahead of print].

21.

Han

, Li

, Wang

, Ying

and Zhang

, Virtual touch tissue quantification (VTQ) in the diagnosis of thyroid nodules with coexistent chronic autoimmune Hashimoto’s thyroiditis: A preliminary study, European Journal of Radiology 84(2) (2015), 327–331.

22.

Cheng

K.L.

, Choi

Y.J.

, Shim

W.H.

, Lee

J.H.

and Baek

J.H.

, Virtual touch tissue imaging quantification shear wave elastography: Prospective assessment of cervical lymph nodes, Ultrasound in Medicine & Biology 42(2) (2016), 378–386.

23.

Tozaki

, Saito

, Benson

, Fan

and Isobe

, Shear wave velocity measurements for differential diagnosis of solid breast masses: A comparison between virtual touch quantification and virtual touch IQ, Ultrasound in Medicine & Biology 39(12) (2013), 2233–2245.

24.

Vlad

, Golu

, Bota

, Vlad

, Timar

, et al., Real-time shear wave elastography may predict autoimmune thyroid disease, Wiener Klinische Wochenschrift 127(9-10) (2015), 330–336.

25.

Lam

A.C.

, Pang

S.W.

, Ahuja

A.T.

and Bhatia

K.S.

, The influence of precompression on elasticity of thyroid nodules estimated by ultrasound shear wave elastography, European Radiology 26(8) (2016), 2845–2852.

26.

Slapa

R.Z.

, Piwowonski

, Jakubowski

W.S.

, Bierca

, Szopinski

K.T.

, Slowinska-Srzednicka

, et al., Shear wave elastography may add a new dimension to ultrasound evaluation of thyroid nodules: Case series with comparative evaluation, Journal of Thyroid Research 2012 (2012), 657147.

27.

Sebag

, Vaillant-Lombard

, Berbis

, Griset

, Henry

J.F.

, Petit

, et al., Shear wave elastography: A new ultrasound imaging mode for the differential diagnosis of benign and malignant thyroid nodules, The Journal of Clinical Endocrinology and Metabolism 95(12) (2010), 5281–5288.

28.

Bhatia

K.S.

, Tong

C.S.

, Cho

C.C.

, Yuen

E.H.

, Lee

Y.Y.

and Ahuja

A.T.

, Shear wave elastography of thyroid nodules in routine clinical practice: Preliminary observations and utility for detecting malignancy, European Radiology 22(11) (2012), 2397–2406.

29.

Yang

Y.P.

, Xu

X.H.

, Xu

H.X.

, Bo

X.W.

, Liu

B.J.

, Guo

L.H.

,et al., Comparison of virtual touch tissue imaging & quantification (VTIQ) and Virtual touch tissue quantification (VTQ) for diagnosis of thyroid nodules, Clinical Hemorheology and Microcirculation (2016). [Epub ahead of print].

30.

Lyshchik

, Higashi

, Asato

, Tanaka

, Ito

, Hiraoka

, et al., Elastic moduli of thyroid tissues under compression, Ultrasonic Imaging 27(2) (2005), 101–110.

31.

Cibas

E.S.

and Ali

S.Z.

, The Bethesda System for Reporting Thyroid Cytopathology, Thyroid 19(11) (2009), 1159–1165.

32.

Ophir

, Cespedes

, Ponnekanti

, Yazdi

and Li

, Elastography: A quantitative method for imaging the elasticity of biological tissues, Ultrasonic Imaging 13(2) (1991), 111–134.

33.

D.D.

, Xu

H.X.

, Guo

L.H.

, Bo

X.W.

, Li

X.L.

, Wu

, et al., Combination of two-dimensional shear wave elastography with ultrasound breast imaging reporting and data system in the diagnosis of breast lesions: A new method to increase the diagnostic performance, European Radiology 26(9) (2016), 3290–3300.

34.

Zhang

Y.F.

, Xu

J.M.

, Xu

H.X.

, Liu

, Bo

X.W.

, Li

X.L.

, et al., Acoustic radiation force impulse elastography: A useful tool for differential diagnosis of thyroid nodules and recommending fine-needle aspiration: A diagnostic accuracy study, Medicine 94(42) (2015), e1834.

35.

Chang

, Kim

M.J.

, Kim

and Lee

M.J.

, Variability of shear wave velocity using different frequencies in acoustic radiation force impulse (ARFI) elastography: A phantom and normal liver study, Ultraschall in der Medizin (Stuttgart, Germany: 1980) 34(3) (2013), 260–265.