Combination of conventional ultrasonography and virtual touch tissue imaging quantification for differential diagnosis of breast lesions smaller than 10 mm

Abstract

OBJECTIVE:

To determine the value of combining conventional ultrasonography with virtual touch tissue imaging quantification (VTIQ) for differential diagnosis of breast lesions smaller than 10 mm.

METHODS:

A total of 98 breast lesions smaller than 10 mm were examined by conventional ultrasound and VTIQ using a Siemens ACUSON S3000 ultrasound machine. Pathologic diagnosis was established after surgery or fine needle biopsy.

RESULTS:

Malignant lesions were characterized by taller-than-wide shape, poorly circumscribed margin, and marked hypoechogenicity. The mean VTIQ shear wave velocity (SWV) value of malignant lesions was 4.88±1.87 m/s (range, 1.75–9.34 m/s), significantly higher than that of benign lesions (2.68±1.02 m/s; range, 1.18–4.67 m/s). The optimal cutoff SWV value was 3.27 m/s, with sensitivity, specificity, diagnostic accuracy, positive predictive value (PPV), and negative predictive value (NPV) of 86.20%, 95.65%, 92.86%, 89.29%, and 94.29%, respectively. The combination of SWV >3.27 m/s plus the US feature of poorly circumscribed margin had the highest sensitivity (93.33%) and specificity (100%) for diagnosis of malignant breast lesions.

CONCLUSION:

Features such as taller-than-wide shape, poorly circumscribed margin, and marked hypoechogenicity on conventional US, and SWV >3.27 m/s on VTIQ, are indicators of malignancy in breast lesions with diameter <10 mm. The combination of poorly circumscribed margin and SWV >3.27 m/s provides the highest specificity and diagnostic accuracy.

Keywords

Conventional ultrasonography virtual touch tissue imaging quantification breast lesions

1 Introduction

Breast cancer is one of the most common malignancies in women and the second most common cause of cancer-related deaths in developed regions such as Europe and North America. According to a report of the American Cancer Society, although the prevalence is relatively low in developing regions, it is increasing [1, 2].

Ultrasonography (US) is widely used for differentiating malignant from benign breast lesions [3], but interpretation of US is very subjective and it has poor specificity [4 –6]. The Breast Imaging Reporting and Data System (BIRADS) criteria was proposed by the American College of Radiology for standardizing evaluation of breast tumors, but some lesions cannot be diagnosed even by these criteria [7].

In 1991, Ophir et al. developed the technique of ultrasound elastography—a noninvasive method for determining tissue mechanical properties [8]. Elastography evaluates tissue strain in response to compression and provides an estimate of the stiffness of a lesion. The elasticity information from US elastography is encoded in color and superimposed on US images to provide a clear picture of the hardness of the lesion [9].

A new elastography technique—acoustic radiation force impulse (ARFI)—has been recently developed that can provide qualitative and quantitative clinical information. Three types of diagnostic evaluations are possible with ARFI: virtual touch tissue imaging (VTI), virtual touch tissue quantification (VTQ), and virtual touch tissue imaging quantification (VTIQ) [10 –12].

ARFI is one type of shear wave elastography (SWE), different manufactures have different SWE techniques. SWE can be used for the examination of various organs, and valuable study results were showed [13, 14].

This study was aimed to determine the value of a combination of conventional US and VTIQ for differential diagnosis of breast lesions less than 10 mm in diameter.

2 Materials and methods

2.1 Patients and breast lesions

This prospective study was approved by the ethical and scientific review board of the Tenth People’s Hospital of Tongji University (approval number: SHSY-IEC-KY-4.0/16-20/01). Verbal informed consent was obtained from all participating patients. The flowchart of the study design is shown in Fig. 1.

Fig.1

Flowchart showing the process of selection of study participants.

The study participants were selected from among the 125 patients with breast lesions who received conventional ultrasound examination followed by VTIQ at the Tenth People’s Hospital of Tongji University between June 2014 and December 2014. Patients were excluded if 1) maximum lesion diameter was >10 mm; 2) they had received any treatment for the lesion; or 3) they were unwilling or unsuitable to undergo fine needle aspiration/surgery to obtain pathological specimen. Eventually, 97 patients (with a total of 98 breast lesions) were included in the study. In all cases diagnosis was confirmed by pathological examination. All pathological diagnoses were made by the same experienced pathologist.

2.2 Examination

All lesions were evaluated by conventional ultrasound and VTIQ. The examinations were performed by one of three radiologists (each with over 5 years experience in breast ultrasound examination) who were blinded to the pathologic results. Conventional US was performed with a standard 10–15 MHz linear transducer, while VTIQ was performed using a 9–16 MHz linear transducer. The SWE mode is available on the Siemens ACUSON S3000 ultrasound machine that was used. The software generates color-coded quantitative maps based on the shear wave velocity (SWV) of the tissue, with red color representing high SWV and green representing low SWV. Color-coded maps displaying shear wave quality (green for good quality, yellow for marginal quality, red for poor quality) and a color map illustrating induced tissue displacement (dark blue for low displacement, light blue for high displacement) are also generated. VTIQ measures the velocity of the perpendicular shear waves by a detection pulse. The velocity of the shear waves propagating through the tissue is related to the stiffness of the tissue, and the color map gives provides a picture of tissue stiffness in the region of interest (ROI) [15, 16] (Figs. 2 and 3). With the Siemens ACUSON S3000, it is possible to measure the SWV within the ROI in meters per second (m/s), up to 10 m/s (corresponding to 300 kPa). Depending on the apparatus and the technique used, the SWV value is given in kPa or in m/s.

Fig.2

Ultrasound and pathological findings of an adenopathy. (A) Conventional ultrasonography and (B) VTIQ quality map showing defined areas in green. (C) VTIQ velocity map showing SWV values. SWV was measured at the center of the lesion (mean, 2.32 m/s). (D) Histopathological image confirming diagnosis of adenopathy.

Fig.3

Ultrasound and pathological findings of an invasive ductal carcinoma. (A) Conventional ultrasonography and (B) VTIQ quality map showing defined areas in green. (C) VTIQ velocity map displaying relative shear wave velocities. SWV was measured at the center of the lesion (mean = 8.50 m/s). (D) Histopathological image confirming diagnosis of invasive ductal carcinoma.

Patients were examined in the supine position, with their arms abducted at right angles to the body and the head resting on the palms of the hand. A probe was applied to the breast and focused on the target lesion. Compression of tissue increases its stiffness [17], and therefore care was taken to minimize any pressure on the breast during the examination. Representative transverse and longitudinal images obtained by conventional US and by elastography were saved in a picture archiving and communication system as bitmap files.

2.3 Statistical analysis

Statistical analysis was performed using SPSS 19.0 (IBM Corp., Armonk, NY, USA). SWV values were expressed as mean±standard deviation. Receiver-operating characteristic (ROC) curves were constructed to determine the optimal cutoff value for SWV. The chi square test was used for comparison of normal distribution variables. P≤0.05 was considered statistically significant.

3 Results

3.1 Patient characteristics

A total of 98 lesions in 97 female patients were included in the study (mean age, 44.74±14.77 years). There were 29 (29.59%) malignant lesions and 69 (70.41%) benign lesions. Among the 29 malignant lesions, 20 were invasive ductal carcinomas. Among the 69 benign lesions, 47 were adenopathy. The histological findings are presented in Table 1.

Table 1
Histological findings

Lesion findings Number Percentage (%)

Malignant 29 29.59

Invasive ductal carcinoma 20 20.40

Intraductal carcinoma 4 4.08

Invasive carcinoma 2 2.04

Invasive lobular carcinoma 1 1.02

Hybrid invasive cribriform carcinoma 1 1.02

Papillary carcinoma 1 1.02

Benign 69 70.41

Intraductal papilloma 6 6.12

Adenopathy 47 47.96

Fibroadenoma 14 14.29

Benign phyllodes tumor 1 1.02

Adenoma 1 1.02

Lesion findings	Number	Percentage (%)
Malignant	29	29.59
Invasive ductal carcinoma	20	20.40
Intraductal carcinoma	4	4.08
Invasive carcinoma	2	2.04
Invasive lobular carcinoma	1	1.02
Hybrid invasive cribriform carcinoma	1	1.02
Papillary carcinoma	1	1.02
Benign	69	70.41
Intraductal papilloma	6	6.12
Adenopathy	47	47.96
Fibroadenoma	14	14.29
Benign phyllodes tumor	1	1.02
Adenoma	1	1.02

3.2 Features identified on conventional US and VTIQ

Table 2 shows the features of the benign and malignant breast lesions on conventional US. The breast lesions ranged in size from 1.3 mm×1.5 mm to 9.9 mm×9.9 mm. Lesion shape, margin, and echogenicity were significant different between benign and malignant lesions; however, there was no significant difference in the posterior echo, internal echo, color Doppler findings, and incidence of calcification. The mean SWV value of malignant lesions was obviously higher than that of benign lesions (4.88±1.87 m/s vs. 2.68±1.02 m/s; P = 0.002).

Table 2
Conventional ultrasonographic features of breast lesions smaller than 10 mm

Feature Malignant Benign P

Shape <0.001

Taller than wide 17 7

Oval to round 12 62

Margin <0.001

Poorly circumscribed 17 7

Well circumscribed 12 62

Echogenicity <0.001

Marked hypoechogenicity 10 4

Hypoechogenicity 1 2

Isoechogenicity 9 60

Hyperechogenicity 9 3

Posterior echo 0.144

Posterior enhancement 5 40

Posterior normal 8 2

Posterior attenuation 16 27

Internal echo 0.092

Homogeneous 24 64

Heterogeneous 5 4

Calcification 0.611

Microcalcification 2 8

Macrocalcification 9 12

No calcification 18 39

Color Doppler 0.650

Color Doppler signal 12 32

No color Doppler signal 17 37

Feature	Malignant	Benign	P
Shape			<0.001
Taller than wide	17	7
Oval to round	12	62
Margin			<0.001
Poorly circumscribed	17	7
Well circumscribed	12	62
Echogenicity			<0.001
Marked hypoechogenicity	10	4
Hypoechogenicity	1	2
Isoechogenicity	9	60
Hyperechogenicity	9	3
Posterior echo			0.144
Posterior enhancement	5	40
Posterior normal	8	2
Posterior attenuation	16	27
Internal echo			0.092
Homogeneous	24	64
Heterogeneous	5	4
Calcification			0.611
Microcalcification	2	8
Macrocalcification	9	12
No calcification	18	39
Color Doppler			0.650
Color Doppler signal	12	32
No color Doppler signal	17	37

3.3 Diagnostic performance of VTIQ

ROC curves were constructed to determine the optimum cutoff SWV values for differentiating between malignant and benign breast lesions (Fig. 4). For all lesions, the optimum cutoff value of SWV for differentiating tumor tissue from normal tissue was 3.27 m/s. This cutoff value provided sensitivity of 86.20%, specificity of 95.65%, and accuracy of 92.86%, the AUC was 0.888 (95% CI: 0.791–0.985; Table 3). When VTIQ was combined with the feature of “taller-than-wide shape,” the overall sensitivity, specificity, accuracy, PPV, and NPV were 88.24%, 100%, 97.37%, 100%, and 96.72%, respectively. When VTIQ was combined with the conventional US finding of “pooly circumscribed margin,” the overall sensitivity, specificity, accuracy, PPV, and NPV were 93.33%, 100%, 98.65%, 100%, and 98.33%, respectively. When VTIQ was combined with the conventional US finding of “marked hypoechogenicity,” the overall sensitivity, specificity, accuracy, PPV, and NPV were 75%, 100%, 95.95%, 100%, and 95.38%, respectively (Table 4).

Fig.4

Receiver operating characteristic curves of VTIQ.

Table 3

Diagonistic efficiency of useful ultrasonographic features

Feature	Sensitivity	Specificity	Accuracy	PPV	NPV
	(%)	(%)	(%)	(%)	(%)
Taller-than-wide shape	58.62	89.86	80.61	70.83	83.78
Poorly circumscribed margin	58.62	89.86	80.61	70.83	83.78
Marked hypoechogenicity	34.48	94.20	76.53	71.43	77.38
VTIQ SWV >3.27 m/s	86.20	95.65	92.86	89.29	94.29

Table 4

Diagnostic efficiency of VTIQ combined with useful ultrasonographic features

Feature (%)	Sensitivity	Specificity	Accuracy	PPV	NPV
	(%)	(%)	(%)	(%)	(%)
A or B	88.24	100	97.37	100	96.72
A or C	93.33	100	98.65	100	98.33
A or D	75.00	100	95.95	100	95.38

Note: A = VTIQ SWV >3.27 m/s; B = taller-than-wide shape; C = poorly circumscribed margin; D = marked hypoechogenicity.

4 Discussion

The incidence and prevalence of breast cancer has increased rapidly worldwide in the past few decades [18, 19]. Conventional US has been used for differentiating malignant breast lesions from benign ones, but it has poor specificity [20 –22]. Better modalities are needed for noninvasive diagnosis. One such modality is the newly developed method of elastography, which uses tissue stiffness to differentiate bewteen maligant and benign lesions [23, 24]. VTIQ is a new elastography-based technique that is less investigator dependent than US. It is a noninvasive imaging modality with high diagnostic specificity and sensitivity [25 –27]. In this study we found that malignant lesions have significantly higher SWV than benign lesions. Further, we obtained seven repetitive measurements of each lesion and confirmed the reproducibility and reliability of VTIQ. Our results were consistent with previous reports on the use of VTIQ for breast lesions [28 –30].

This is the first study to demonstrate the good diagnostic performance of VTIQ for diagnosis of breast lesions smaller than 10 mm. Yoon et al. [31] and Yao et al. [32] found that the diagnostic sensitivity of VTQ for breast lesions smaller than 10 mm was relatively low (33.33%), which was probably because it is difficult to interpret the data when the ROI is fixed. To overcome this limitation, we used VTIQ instead of VTQ. The ROI in VTIQ has the minimum area as 1 mm×1 mm and it can be adjusted according to the lesion size. Normal tissue surrounding the lesion can be mostly excluded if the ROI is placed properly. Moreover, VTIQ provides two-dimensional color-coded maps for the ROI and multi-point repetitive velocity measurements, whereas VTQ has a fixed ROI of 5 mm×6 mm and only provides single-point velocity value measurement.

Diagnosis with conventional US is mainly based on morphological findings related to lesion margin and echogenicity [33]. A typical benign lesion, such as a fibroadenoma, usually presents a well circumscribed margin, low echogenicity, and an expansive growth pattern [34]. In contrast, malignant lesions usually have a poorly circumscribed margin. The findings of the present study were consistent with previous reports: taller-than-wide shape, poorly circumscribed margin, and marked hypoechogenicity were features associated with breast malignancy in lesions smaller than 10 mm. Hypoechogenicity had high specificity (94.20%) but poor sensitivity (34.48%); taller-than-wide shape and poorly circumscribed margin both had the same sensitivity (58.62%). Although these features were associated with malignant lesions, they had low sensitivity; however, if they were associated SWV of >3.27 m/s in the concerned lesion, the sensitivity increased markedly. Yao et al. [35] have also reported that combining conventional US with VTIQ can improve diagnostic accuracy for breast lesions.

This study has some limitations. The sample size was small, only 98 lesions were included in the study. Our results should be confirmed in a larger sample. The performance of VTIQ depends on multiple factors, which makes it difficult to obtain consistent image quality.

5 Conclusion

The US features of taller-than-wide shape, poorly circumscribed margin, and marked hypoechogenicity, along with VTIQ finding of SWV >3.27 m/s, are valuable for differential diagnosis of breast lesions smaller than 10 mm. When VTIQ is combined with conventional US, the sensitivity of these features is improved significantly. The best sensitivity is achieved with SWV >3.27 m/s plus poorly circumscribed margin. Thus, VTIQ can be a useful supplement to conventional US for diagnosis of breast lesions.

Declaration of conflict of interest

None.

Footnotes

Acknowledgments

This work was supported in part by grants from the Shanghai Hospital Development Center [grant number SHDC12014229] and the Science and Technology Commission of Shanghai Municipality [grant number 14441900900].

References

Mohey

and Hassan

T.A.

, Value of mammography and combined grey scale ultrasound and ultrasound elastography in the differentiation of solid breast lesions, Egyptian J Radiol NuclearMed 45 (2014), 253–261.

Siegel

, Naishadham

and Jemal

, Cancer statistics, CA Cancer J Clin 63 (2013), 11–30.

Stavros

A.T.

, Thickman

, Rapp

C.L.

, Dennis

M.A.

, Parker

S.H.

and Sisney

G.A.

, Solid breast nodules: Use of sonography to distinguish between benign and malignant lesions, Radiology 196 (1995), 123–134.

Berg

W.A.

, Supplemental screening sonography in dense breasts, Radiol Clin North Am 42 (2004), 845–851.

Corsetti

, Ferrari

, Ghirardi

, et al., Role of ultrasonography in detecting mammographically occult breast carcinoma in women with dense breasts, Radiol Med 111 (2006), 440–448.

Houssami

, Irwig

, Simpson

J.M.

, McKessar

, Blome

and Noakes

, Sydney Breast Imaging Accuracy Study: Comparative sensitivity and specificity of mammography and sonography in young women with symptoms, AJR Am J Roentgenol 180 (2003), 935–940.

Athanasiou

, Tardivon

, Tanter

, et al., Breast lesions: Quantitative elastography with supersonic shear imaging-preliminary results, Radiology 256 (2010), 297–303.

Ophir

, Cespedes

, Ponnekanti

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

Fischer

, Peisker

, Fiedor

, et al., Significant differentiation of focal breast lesions: Raw data-based calculation of strain ratio, Ultraschall in der Medizin 33 (2012), 372–379.

10.

Tozaki

, Saito

, Joo

, et al., Ultrasonographic elastography of the breast using acoustic radiation force impulse technology: Preliminary study, Jpn J Radiol 29 (2011), 452–456.

11.

Tozaki

, Isobe

and Fukuma

, Preliminary study of ultrasonographic tissue quantification of the breast using the acoustic radiation force impulse (ARFI) technology, Eur J Radiol 80 (2011), 182–187.

12.

Marcon

, Trottmann

, Rübenthaler

, Stief

C.G.

, Reiser

M.F.

and Clevert

D.A.

, Shear wave elastography of the testes in a healthy study collective - Differences in standard values between ARFI and VTIQ techniques, Clin Hemorheol Microcirc 64(4) (2016), 721–728.

13.

Ren

W.W.

, Li

X.L.

, He

Y.P.

, et al., Two-dimensional shear wave elastography of breast lesions: Comparison of two different systems, Clin Hemorheol Microcirc 66(1) (2017), 37–46.

14.

Marcon

, Trottmann

, Rübenthaler

, et al., Shear wave elastography of the testes in a healthy study collective - Differences in standard values between ARFI and VTIQ techniques, Clin Hemorheol Microcirc 64(4) (2016), 721–728.

15.

Weismann

, Mayr

, Egger

, et al., Breast sonography-2D, 3D, 4D ultrasound or elastography? Breast Care (Basel, Switzerland) 6 (2011), 98–103.

16.

Barr

R.G.

, Sonographic breast elastography: A primer, J Ultrasound Med 31 (2012), 773–783.

17.

Barr

R.G.

and Zhang

, Effects of precompression on elasticity imaging of the breast: Development of a clinically useful semiquantitative method of precompression assessment, J Ultrasound Med 31 (2012), 895–902.

18.

Fox

B.L.

, Surviving triple negative breast cancer: Hope, treatment, and recovery, Library Journal 137 (2012), 121–124.

19.

Kubista

, Breast cancer: Figures and facts, Wien Med Wochenschr 151 (2001), 548–551.

20.

Thomas

, Degenhardt

, Farrokh

, et al., Significant differentiation of focal breast lesions: Calculation of strain ratio in breast sonoelastography, Acad Radiol 17 (2010), 558–563.

21.

Navarro

, Ubeda

, Vallespi

, et al., Role of elastography in the assessment of breast lesions: Preliminary results, J Ultrasound Med 30 (2011), 313–321.

22.

Regini

, Bagnera

, Tota

, et al., Role of sonoelastography in characterising breast nodules.: Preliminary experience with 120 lesions, Radiol Med 115 (2010), 551–562.

23.

Parker

K.J.

and Lerner

R.M.

, Sonoelasticity of organs: Shear waves ring a bell, J Ultrasound Med 11 (1992), 387–392.

24.

Garra

B.S.

, Cespedes

E.I.

, Ophir

, et al., Elastography of breast lesions: Initial clinical results, Radiology 202 (1997), 79–86.

25.

Zhi

, Ou

, Xiao

X.Y.

, et al., Ultrasound elastography of breast lesions in chinese women: A multicenter study in China, Clin Breast Cancer 13 (2013), 392–400.

26.

Cosgrove

D.O.

, Berg

W.A.

, Dore

C.J.

, et al., Shear wave elastography for breast lesions is highly reproducible, Eur Radiol 22 (2012), 1023–1032.

27.

Evans

, Whelehan

, Thomson

, et al., Differentiating benign from malignant solid breast lesions: Value of shear wave elastography according to lesion stiffness combined with greyscale ultrasound according to BI-RADS classification, Br J Cancer 107 (2012), 224–229.

28.

Golatta

, Schweitzer-Martin

, Harcos

, et al., Evaluation ofvirtual touch tissue imaging quantification, a new shear wavevelocity imaging method, for breast lesion assessment byultrasound, Biomed Res Int 2014 (2014), 960262.

29.

Tozaki

, Saito

, Benson

, et al., Shear wave velocity measurements for differential diagnosis of solid breast lesions: A comparison between virtual touch quantification and virtual touch IQ, Ultrasound Med Biol 39 (2013), 2233–2245.

30.

Chen

Y.P.

, Han

, Wu

, et al., Comparison of Virtual Touch Tissue Quantification and Virtual Touch Tissue Imaging Quantification for diagnosis of solid breast tumors of different sizes, Clin Hemorheol Microcirc 64(2) (2016), 235–244.

31.

Yoon

J.H.

, Kim

M.J.

, Kim

E.K.

, et al., Discordant elastography images of breast lesions: How various factors lead to discordant findings, Ultraschall Med 34 (2013), 266–271.

32.

Yao

, Wu

, Zou

, et al., Diagnostic value of virtual touch tissue quantification for breast lesions with different size, Biomed Res Int 2014 (2014), 142504.

33.

Zhang

Y.F.

, Xu

H.X.

, He

, et al., Virtual touch tissuequantification of acoustic radiation force impulse: A newultrasound elastic imaging in the diagnosis of thyroid nodules, Plos One 11 (2012), 49094.

34.

J.Y.

, Du

L.F.

, 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 31 (2012), 763–771.

35.

Yao

M.H.

, Wu

, Xu

, et al., A novel two-dimensional quantitative shear wave elastography to make differential diagnosis of breast lesions: Comprehensive evaluation and influencing factors, Clin Hemorheol Microcirc 64(2) (2016), 223–233.