Value of different ultrasound elastography techniques in patients with venous malformations prior to and after sclerotherapy

Abstract

AIM: Comparison of different ultrasound elastography techniques for detection of changes after sclerotherapy within venous malformations.

MATERIAL AND METHODS: In patients with venous malformations sonography was executed at exactly the same position prior to and after ethanol-gel sclerotherapy. Both examinations included B-Mode, vascular sonography with Color-Coded Duplex Sonography, and additional sonography with different elastography techniques (strain, qualitative and quantitative Acoustic Radiation Force Impulse (ARFI) elastography) with a linear transducer (6–9 MHz). Qualitative elastograms were read in consensus and scored. Differences of elasticity scores were statistically analyzed, p-values <0.05 were regarded significant.

RESULTS: Elasticity scores of strain and qualitative ARFI elastography in 25 patients (21 females, averagely 24.4 years old) were comparable before treatment (p = 0.69). After therapy qualitative ARFI scores changed significantly compared to pre-treatment scores (p = 0.0017), whereas strain elastography scores revealed no significant changes (p = 0.13). Quantitative ARFI values obtained after sclerotherapy within the venous malformations were significantly higher compared to pre-treatment values (p = 0.049), and significantly higher to values obtained in surrounding tissue (p = 0.030). Comparison of pre- and post-treatment ARFI values of the surrounding tissue was not significant (p = 0.67).

CONCLUSION: Elasticity scores of qualitative ARFI elastography reliably detect ethanol-gel induced changes in venous malformations. Quantitative ARFI may be a tool for therapy planning, and for monitoring sclerotherapy outcome as well as the effect of sclerosing agents on malformation and surrounding tissue in patients with venous malformations.

Keywords

Venous malformation elastography elasticity score Acoustic Radiation Force Impulse (ARFI) imaging therapy control

1 Introduction

Vascular malformations occur in approximately 0.3–0.5% of the population [19]. Based on Mulliken’s findings and according to the International Society for the Study of Vascular Anomalies (ISSVA), vascular anomalies are divided into vascular tumors and vascular malformations [7, 17]. According to their flow pattern, fast-flow and slow-flow vascular malformations are differentiated. The latter ones account for 80% of all vascular malformations and consist of venous, lymphatic, capillary, and combined malformations [7]. Venous malformations (VM) are the most common ones [4].

Diagnosis is based on anamnesis, thorough clinical examination, sonography, and contrast-enhanced MR imaging. To date, sclerotherapy is the preferred treatment for VM [5, 15]. Sclerotherapy can be performed using highly concentrated ethanol or different foam agents (polidocanol or STS). Liquid ethanol is a highly effective sclerosing agent but regularly leads to local and systemic side effects [13]. Gelified ethanol seems to be a save alternative [8 , 21].

Elastography is a newly but more and more established ultrasound technique. It relies on diverging plasticity and deformability of different types of tissue, visualizes tissue stiffness, and may contribute to the discrimination of different tissue types [10, 18]. Recently it has been widely accepted for diagnosis and grading of liver fibrosis. Its application now includes assessment of breast lesions, thyroid nodules, lymph nodes, and many others [1 , 10].

Technically it either works via external compression (manual compression, movement from blood vessels, breathing) known as strain imaging or automatically in a region of interest (ROI) with an acoustic push pulse (Acoustic Radiation Force Impulse Technology, Siemens, Erlangen, Germany). ARFI technology is able to initiate and trace the propagation of shear waves. The mean shear wave velocity within a region of interest (ROI) can be measured, resulting in a quantitative measurement and an absolute value. There is a close correlation between increasing shear wave velocity and increasing tissue stiffness [3 , 16].

In the framework of a prospectively planned clinical study we evaluated 1) whether different elastography techniques are able to differentiate between untreated and treated VM, and which technique is the most reliable one. Furthermore we wanted to determine 2) if quantitative ARFI elastography is a diagnostic tool for distinguishing untreated and treated VM from surrounding tissue. And finally we compared 3) quantitative ARFI values of the surrounding tissue prior to and after sclerotherapy aiming on ethanol-gel’s local effect with or without diffusion into healthy, surrounding tissue.

2 Material and methods

Ethical board permission for the prospective study and written informed consent of all patients was obtained.

2.1 Patients

All patients presented at our tertiary care Vascular Anomalies Center (VAC) and were diagnosed for VM after a thorough clinically examination. A subgroup of patients with VM was included in a prospective study; their VM were treated once to twice percutaneously with gelified ethanol (ScleroGel, ab medica, Düsseldorf, Germany). In accordance with the interventional radiologist this group of patients was examined with sonography on exactly the same, most painful position prior to and after treatment. In some of these patients different elastography techniques were executed on exactly the same position before and after treatment. Their data was evaluated off-line after finishing the complete study.

2.2 Sonography

Ultrasound prior to and after treatment was performed by two clinically experienced radiologists (together more than 3000 examinations every year for more than 10 years) with a linear 6–9 MHz multi-frequency transducer (ACUSON S2000, Siemens, Erlangen, Germany). The whole affected body region was examined. Acute deep vein thrombosis was ruled out.

Vascular pattern of the malformation was analyzed with Color-Coded Duplex Sonography (CCDS) and Power Doppler (PD). In accordance with the intended treatment location, elastography then was focused on the most painful area. Transducer position was marked on the skin and documented with a photograph.

2.3 Elastography

The same superficial 6–9 MHz linear multi-frequency transducer was used. Examination was obtained with the parameters of the Breast preset on the marked localization.

Basically two types of elastography were applied: strain elastography (eSie Touch Elasticity Imaging; Siemens, Erlangen, Germany), and Acoustic Radiation Force Impulse (ARFI) Imaging (Virtual Touch Imaging; Siemens, Erlangen, Germany).

Strain elastography is a qualitative method and based on compression –from an external source (ultrasound transducer) or from body movement (i.e. breathing) [2]. After mechanical stress, tissue moves in direction of the ultrasound beam and the displacement of the tissue is detected. Tissue elasticity as well as user-generated manual pressure influence the response. Results are presented qualitatively. The tissue displacement is calculated in relation to the surrounding area, the software displays the different tissue types in color-codes [3, 11], depending on the system setting. Elastograms visualizing hard tissue in blue color and soft tissue in red were evaluated (Fig. 1).

Fig.1

Strain elastographies of a 16.1 year old patient with VM on the right thigh. a) Prior to sclerotherapy mixed picture with mainly soft (red) and intermediate (green) colors in the center of the VM. b) After the second treatment the preexisting malformation portions are completely blue (hard).

Qualitative ARFI elastography uses a special push pulse in a certain ROI resulting in a wave propagating transverse to the ultrasound beam and deforming tissue with a defined velocity, which is deemed proportional to the tissue stiffness. This technique is less examiner-dependent and improves reproducibility. Again images are color-coded according to relative tissue stiffness, hard tissue is visualized in red, soft tissue is bluely colored [3] (Fig. 2).

Fig.2

Same patient as in Fig. 1, qualitative ARFI elastograms. a) Before sclerotherapy malformation portions are bluely (soft) colored. b) After interventional therapy the malformation is yellow (hard) coded.

Quantitative ARFI elastography measures the shear wave propagation speed in a sample volume ROI. Absolute measure values in m/s are calculated [11].

Strain and qualitative ARFI elastograms were derived and stored at exactly the same position prior to and after sclerotherapy. Finally two to three quantitative ARFI measurements in the VM center and 5–6 measurements in the surrounding tissue were documented before and after treatment.

2.4 Analysis

Two for the examination date blinded, in sonography experienced radiologists analyzed the qualitative elastograms in consensus.

Lesions were classified according to the Tsukuba elasticity score of breast lesions [9, 14] (Table 1). Thereby VM were assessed for the ratio of soft and stiff tissue and categorized into 5 grades. The grade increases with a reduction of tissue softness. Results of both qualitative elastography techniques were compared.

Table 1
Elasticity score

Grade % of soft tissue in the lesions % of stiff tissue in the lesion

0 Evaluation not possible Evaluation not possible

1 100% 0%

2 75% 25%

3 50% 50%

4 25% 75%

5 0% 100%

Grade	% of soft tissue in the lesions	% of stiff tissue in the lesion
0	Evaluation not possible	Evaluation not possible
1	100%	0%
2	75%	25%
3	50%	50%
4	25%	75%
5	0%	100%

2.5 Statistics

Qualitative elastography scores before and after treatments were compared among each other using Wilcoxon signed-rank test. Quantitative ARFI-data was log transformed, ANOVA analysis with Bonferoni Holm posthoc tests was performed; p-values <0.05 were considered significant.

3 Results

3.1 Patients

In 25 patients elastography was performed. 21 patients (84.0%) were female; mean patients’ age was 24.4±12.2 years (6.6–46.5 years). The VM involved the lower extremity (n = 15, 60.0%), the upper extremity (n = 6, 24.0%), and the face (n = 4, 16.0%). All patients had qualitative ARFI elastographies before and after successful sclerotherapy (Fig. 3). In one patient strain elastography did not work, thus 24 patients were examined with strain elastography twice. In all 25 patients quantitative ARFI measurements were carried out 2–3 times in the lesions’ center. In two patients ARFI measurements in the surrounding tissue were not repeated 5–6 times resulting in 23 patients with ARFI examinations in the surrounding tissue before and after treatment with ethanol-gel.

Fig.3

Same patient as in Fig. 1, fluoroscopy after contrast medium injection into the venous malformation on the corresponding area of the elastographies. a) Delineation after puncture of one malformation compartment. b) Progressive filling of the malformation with contrast medium after puncturing a second compartment.

3.2 Sonography

On average the final ultrasound examination was executed 141±17 (114–170 days) days after the first examination. One patient had an old thrombosis at both examinations. In CCDS neither calcified aneurysms nor an acute thrombophlebitis were detected. According to the inclusion criteria none of the patients suffered from fast-flow malformations with arteriovenous shunts.

3.3 Strain-elastography

Prior to treatment elasticity score in the lesion was 2.79±0.93 averagely (2–4), after therapy its mean value was 3.48±1.02 (2–5). Comparing scores before and after treatment revealed no significant difference (p = 0.13) (Fig. 4).

Fig.4

Strain elastography of a 39.25 year old patient. a) Prior to sclerotherapy the tissue adjacent to the bone is blue with single green islands. b) After sclerotherapy there is no difference detectable.

3.4 Qualitative ARFI elastography

The average elasticity score in the VM was 2.68±0.99 before (1–5), and 4.12±0.97 after treatment (2–5). The comparison of results prior to and after treatment differed significantly (p = 0.0017) (Fig. 5).

Fig.5

Qualitative ARFI elastograms of the same patient as in Fig. 4. a) Prior to sclerotherapy: adjacent to the bone the tissue is displayed greenish. b) After sclerotherapy the sclerotized malformation is clearly detectable in terms of almost homogenously redly colored tissue.

3.5 Different qualitative elastography techniques

Comparison of elasticity scores of strain and qualitative ARFI elastography among each other in the malformation before sclerotherapy revealed no significant difference (2.79±0.93 vs. 2.68±0.99; p = 0.69). After therapy the difference was significant (3.48±1.02 vs. 4.12±0.97; p = 0.038).

3.6 Analysis of quantitative ARFI measurements

Comparing ARFI values in the center of the malformations prior to and after treatment results were significantly higher after treatment (1.80±0.77 m/s vs. 2.29±0.96 m/s, p = 0.049) (Fig. 6).

Fig.6

Quantitative ARFI measurements in the patient of Fig. 1. a) Prior to sclerotherapy ROI was placed in the center of a dysplastic venous vessel resulting in a value of 0.57 m/s. b) After sclerotherapy at the same area the values were 3.32 m/s.

Comparison between ARFI values in the center of the malformation and the surrounding tissue prior to treatment suggests the tissue in the center of the malformation to be slightly softer than the surrounding tissue, but without significant difference (1.76±0.79 m/s vs. 1.83±0.43 m/s; p = 0.70). After treatment ARFI values in the center of the malformations were significantly higher than in the surrounding tissue (2.33±0.97 m/s vs. 1.89±0.69 m/s, p = 0.030).

Comparison between ARFI values of surrounding tissue prior to and after treatment revealed no significant differences (1.83±0.43 m/s vs. 1.89±0.69 m/s; p = 0.67).

4 Discussion

Elastography is a newly established, prosperous ultrasound technique. It already is an easily applicable and important diagnostic tool for early detection of liver fibrosis and cirrhosis, and a more accurate classification of breast lesions. The guidelines of the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) depict a broad range of indications [6]. For soft tissues the EFSUMB guidelines recommend indication for breast tumor imaging and lymph nodes.

The value of quantitative ARFI measurements was investigated in liver disease and results correlated with biopsies [10, 12]. Values of the liver are influenced by many factors, for example chemotherapy or portal vein embolization [10], indicating the method’s sensitivity.

Quantitative elastography already is an important tool in the differentiation between benign and malignant breast lesions [11]. It even seems to have a prognostic impact, as data of a certain breast tumor type (triple-negative breast cancer) showed that patients with higher velocities are at a higher risk for occurrence of metastases [18].

Until now additional indications of quantitative ARFI measurements have been increasing and recently results of soft tissue, for example standard values in the testes, were published [6, 16]. Regarding vascular changes US elastography has been used for the detection of tissue necrosis as one risk after interventional therapy so far [24].

On the background of this data we wanted to investigate the role of ultrasound elastography in therapy monitoring after percutaneous sclerotherapy of VM.

To our knowledge to date nobody has compared pre- and post-treatment results of strain, qualitative, and quantitative ARFI elastography methods in VM. Furthermore neither values of quantitative ARFI elastography performed within the center of the VM have yet been compared to measurements within the surrounding tissue, nor pre- and post-treatment values within surrounding tissue.

Whereas strain elastography is not able to differentiate between treated and untreated VM (Fig. 4), results of qualitative ARFI elastography after sclerotherapy with gelified ethanol get significantly higher compared to pre-treatment results (p = 0.0017) (Fig. 5).

An explanation might be, that qualitative ARFI elastography works with a certain push pulse in a ROI instead of depending on compression generated from body movement or the examiner, and thus may be more reliable

The softness and compressibility of untreated VM in comparison to surrounding tissue are not displayed on quantitative ARFI measurements. Though values in the lesions’ center indicate a slightly softer tissue compared to surrounding tissue, results are not statistically significant (p = 0.70). A reason could be, that slow blood flow and micro-phleboliths might raise ARFI values and thus counteract low values induced by blood vessels’ softness. Nevertheless, quantitative ARFI measurement is able to differentiate between untreated and treated malformation portions (p = 0.049) (Fig. 6). Apparently ethanol-gel provokes a hardening of VM and quantitative ARFI elastography is a perfect tool for the detection of these changes. Furthermore quantitative ARFI elastography can differentiate between sclerotized VM and healthy, surrounding tissue as values in treated VM and surrounding tissue differ significantly after sclerotherapy (p = 0.030).

As ARFI values in the surrounding tissue do not change after treatment, gelified ethanol seems to have an only local effect without diffusion into the surrounding tissue.

Limitations of the study are the availability of reliable quantitative ARFI measurements only to a depth at maximum up to 4 cm for linear probes [3] and the small patient group. Furthermore results regarding long-term changes of sclerotized malformation portions are lacking.

Contrast-enhanced ultrasound (CEUS) is an alternative ultrasound technique for the assessment of vascular anomalies. In combination with perfusion analysis it can be a diagnostic tool for distinguishing between different vascular malformation types as well as between vascular malformations and surrounding tissue [22]. Furthermore time intensity curves prior to and after treatment of vascular malformations are able to measure therapy-induced changes [23].

Thus for the assessment of risk, success and follow up of interventional therapy in patients with VM complementary imaging techniques in terms of CEUS and elastography might be advisable. This should be investigated in future studies.

5 Conclusion

Our study shows that in the detection of post-treatment changes in VM elasticity scores of qualitative ARFI elastography exceed strain elastography.

Quantitative ARFI elastography is able to differentiate between untreated and treated compartments of the VM. Therefore it could be an important tool in planning interventional therapy in patients with VM, in monitoring the success of sclerotherapy, and in displaying the effect of sclerosing agents within the malformation as well as surrounding tissue.

References

Bamber

C.D.J.

and Dietrich

C.F.

, EFSUMB Guidelines andRecommendations on the Clinical Use of Ultrasound Elastography, Part 1: Basic prinicples and technology, Ultraschall in der Medizin 34 (2013), 169–184.

Barr

R.G.

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

Benson

F.L.J.

, Understanding ARFI and New ElastographyQuantification Technologies, in https://www.healthcare.siemens.com/siemens_hwem-hwem_ssxa_websites-contextroot/wcm/idc/groups/public/@global/@imaging/@ultrasound/documents/download/mday/njcz/edisp/2e7016_siemens_vtq_whitepaper_20_hood05162001588827-02673689.pdf

Berenguer

, Burrows

P.E.

, Zurakowski

and Mulliken

J.B.

, Sclerotherapy of craniofacial venous malformations: Complications and results, Plast Reconstr Surg 104 (1999), 1–11.

Burrows

P.E.

and Mason

K.P.

, Percutaneous treatment of low flow vascular malformations, J Vasc Interv Radiol 15 (2004), 431–445.

Cosgrove

P.F.D.

and Bamber

, EFSUMB Guidelines andRecommendations on the Clinical Use of Ultrasound Elastography,Part 2: Clinical applications, Ultraschall in der Medizin 34 (2013), 238–253.

Dasgupta

and Fishman

S.J.

, ISSVA classification, Semin Pediatr Surg 23 (2014), 158–161.

Dompmartin

, Blaizot

, Theron

, Hammer

, Chene

, Labbe

, Barrellier

M.T.

, Gaillard

, Leroyer

, Chedru

, Ollivier

, Vikkula

and Boon

L.M.

, Radio-opaqueethylcellulose-ethanol is a safe and efficient sclerosing agent for venous malformations, Eur Radiol 21 (2011), 2647–2656.

Fischer

, Peisker

, Fiedor

, Slowinski

, Wedemeyer

, Diekmann

, Grigoryev

and Thomas

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

10.

Frulio

, Laumonier

, Carteret

, Laurent

, Maire

, Balabaud

, Bioulac-Sage

and Trillaud

, Evaluation of liver tumors using acoustic radiation force impulse elastography and correlation with histologic data, J Ultrasound Med 32 (2013), 121–130.

11.

Goddi

, Bonardi

and Alessi

, Breast elastography: A literature review, J Ultrasound 15 (2012), 192–198.

12.

Hanquinet

, Courvoisier

D.S.

, Rougemont

A.L.

, Dhouib

, Rubbia-Brandt

, Wildhaber

B.E.

, Merlini

, McLin

V.A.

and Anooshiravani

, Contribution of acoustic radiation force impulse (ARFI) elastography to the ultrasound diagnosis of biliary atresia, Pediatr Radiol 45 (2015), 1489–1495.

13.

Horbach

S.E.

, Lokhorst

M.M.

, Saeed

, de Gouyon Matignon de Pontouraude

C.M.

, Rothova

and van der Horst

C.M.

, Sclerotherapy for low-flow vascular malformations of the head and neck: A systematic review of sclerosing agents, J Plast Reconstr Aesthet Surg 69 (2016), 295–304.

14.

Itoh

, Ueno

, Tohno

, Kamma

, Takahashi

, Shiina

, Yamakawa

and Matsumura

, Breast disease: Clinical application of US elastography for diagnosis, Radiology 239 (2006), 341–350.

15.

Lee

B.B.

, Do

Y.S.

, Byun

H.S.

, Choo

I.W.

, Kim

D.I.

and Huh

S.H.

, Advanced management of venous malformation with ethanol sclerotherapy: Mid-term results, J Vasc Surg 37 (2003), 533–538.

16.

Marcon

, Trottmann

, Rubenthaler

, 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 (2016), 721–728.

17.

Mulliken

J.B.

and Glowacki

, Hemangiomas and vascular malformations in infants and children: A classification based on endothelial characteristics, Plast Reconstr Surg 69 (1982), 412–422.

18.

, Zhao

L.X.

, Yao

M.H.

, Xu

, Liu

, Xu

and Wu

, Conventional US combined with acoustic radiation force impulse imaging (ARFI) for prediction of triple-negative breast cancer and the risk of lymphatic metastasis, Clin Hemorheol Microcirc 65(4) (2017), 335–347. doi: 10.3233/CH-16196.

19.

Redondo

, Aguado

and Martinez-Cuesta

, Diagnosis and management of extensive vascular malformations of the lower limb: Part I. Clinical diagnosis, J Am Acad Dermatol 65 (2011), 893–906.

20.

Sannier

, Dompmartin

, Theron

, Labbe

, Barrellier

M.T.

, Leroyer

, Toure

and Leroy

, A new sclerosing agent in the treatment of venous malformations. Study on 23 cases, Interv Neuroradiol 10 (2004), 113–127.

21.

Schumacher

, Dupuy

, Bartoli

J.M.

, Ernemann

, Herbreteau

, Ghienne

, Guibaud

, Loose

D.A.

, Mattassi

, Petit

, Rossler

J.K.

, Stillo

and Weber

, Treatment of venous malformations: First experience with a new sclerosing agent–a multicenter study, Eur J Radiol 80 (2011), 366–372.

22.

Teusch

V.I.

, Wohlgemuth

W.A.

, Piehler

A.P.

and Jung

E.M.

, Color-coded perfusion analysis of CEUS for pre-interventional diagnosis of microvascularisation in cases of vascular malformations, Clin Hemorheol Microcirc 58 (2014), 183–193.

23.

Wiesinger

, Schreml

, Wohlgemuth

W.A.

, Stroszczynski

and Jung

E.M.

, Perfusion quantification of vascular malformations using contrast-enhanced ultrasound (CEUS) with time intensity curve analysis before and after treatment: First results, Clin Hemorheol Microcirc 62 (2015), 283–290.

24.

Wiggermann

, Jung

E.M.

, Glockner

, Hoffstetter

, Uller

, Vasilj

, Schreyer

and Stroszczynski

, Real-time elastography of hepatic thermal lesions in vitro: Histopathological correlation, Ultraschall Med 33 (2012), 170–174.

Value of different ultrasound elastography techniques in patients with venous malformations prior to and after sclerotherapy

Abstract

Keywords

1 Introduction

2 Material and methods

2.1 Patients

2.2 Sonography

2.3 Elastography

Table 1 Elasticity score Grade % of soft tissue in the lesions % of stiff tissue in the lesion 0 Evaluation not possible Evaluation not possible 1 100% 0% 2 75% 25% 3 50% 50% 4 25% 75% 5 0% 100%

3 Results

3.1 Patients

3.3 Strain-elastography

3.6 Analysis of quantitative ARFI measurements

5 Conclusion

References

Table 1
Elasticity score

Grade % of soft tissue in the lesions % of stiff tissue in the lesion

0 Evaluation not possible Evaluation not possible

1 100% 0%

2 75% 25%

3 50% 50%

4 25% 75%

5 0% 100%