Non thermal HIFU treatment for vein occlusion: Histological evidence from a pre-clinical animal study

Abstract

Objective

The aim of this preclinical study is to investigate in-vivo veins effects including histological analysis following endovenous cavitation by Non-Thermal High-Intensity Focused Ultrasound (NT-HIFU).

Methods

Four lateral saphenous veins (LSV) in two sheep were sonicated using ultrasound guided NT-HIFU device inducing cavitation (Veinsound, Lyon, France) during 30s in 2 LSV (protocol A), 20s in 1 LSV (protocol B), and 10s in 1 LSV (protocol C). Clinical and ultrasound follow-up was performed at 3 and 7 days, venous sampling at 10 days. Four vein sections were available in each LSV. Histological analysis considered percentage of vein lumen obliteration. Vein wall damage was qualified as destruction, fibrosis, and immune infiltration.

Results

Obliteration was observed by Duplex Ultrasound in all LSV. Histologic analysis showed total or partial obliteration of the LSV in respectively 15 (93.8%) and 1 case (6.2%). Destruction was observed in the intima in all 16 histology sections, in the media in 11 sections (68.8%) and in the adventitia in five sections (31.2%). Fibrosis was observed in the media in all 16 sections, in the intima in seven sections (43.8%) and in the adventitia in five sections (31.2%). Immune infiltration was observed in the adventitia in eight sections (50%), in the media in five sections (31.2%) and in the intima in one section (6.2%). Resolutive skin damage was observed in one case (protocol A) and mild perivenous tissue remodeling in 3 cases (protocol A and B).

Conclusion

This study demonstrated that endovenous cavitation allows to cause vein occlusion in sheep with significant destruction lesions of intima and fibrosis of the media. Appropriate sonication time avoids side effects.

Keywords

non-thermal high-intensity focused ultrasound cavitation animal study histology

Introduction

A wide range of technological options currently exists for the treatment of varicose veins. Though thermal endovenous treatments (using laser, radiofrequency or microwave) are usually the first choice of treatment, these catheter-based procedures require local anesthesia and vein catheterization. This is also the case for pharmaco-mechanical ablation and glue embolization techniques. While anesthesia can be avoided by opting for non-thermal endovenous treatments, these techniques require the intravenous injection of a sclerosing agent or cyanoacrylate.¹

High-intensity focused ultrasound (HIFU) is an emerging therapeutic technique that enables the non-invasive transcutaneous treatment of varicose veins. HIFU achieves localized tissue ablation through two main mechanisms: thermal or non-thermal (mechanical tissue remodeling induced by cavitation). The thermal approach is employed by Sonablate (Charlotte, USA) and Theraclion (Malakoff, France). Limitations are related to thermal energy diffusion (potentially painful and requiring adjunctive local anesthesia), in addition to the prolonged duration of treatment involving robotized material needed to administer thermal energy to the entire vein wall circumference through the application of HIFU at multiple targeted points.²

Non-thermal HIFU (NT-HIFU) involves a rapid change in tissue pressure that triggers the generation of cavitation microbubbles. The targeted tissue is mechanically damaged by the shear stress generated by the subsequent implosion of these cavities. This technique, known as histotripsy, is already used for the treatment of liver tumors and aortic valve narrowing, with investigations underway for thrombolysis.^3–5

The potential of NT- HIFU for the non-invasive treatment of varicose veins has been demonstrated by a preclinical study involving the use of a dedicated device designed by Veinsound (Lyon, France). In-vivo vein occlusion was successfully achieved by cavitation using NT-HIFU inside the vein lumen of a collateral saphenous vein of a sheep without the injection of microbubble contrast agents.⁶

The objective of this additional preclinical study is to further investigate in-vivo veins occlusion by NT- HIFU, including comprehensive histological analysis of the vein following endovenous cavitation, enabling optimization of the therapeutic protocol for subsequent clinical trials to evaluate the approach for the non-invasive transcutaneous treatment of varicose veins in patients.

Material and methods

Ultrasound-guided HIFU device

The ultrasound-guided HIFU device, previously described, consists of a 10 MHz linear probe (LC1038V, Quantel Medical, Cournon d’Auvergne, France) for ultrasound guidance and an in-house therapeutic HIFU probe composed of four ultrasound emitters that focus high-intensity ultrasound up to 25 mm into the tissue.⁶ Ultrasound treatments are performed with an emission frequency of 1.11 MHz with a negative peak pressure at the focus of −23 MPa. This level of pressure is consistent with the pressure threshold necessary to trigger cavitation cloud inside soft tissue which in this case corresponds to the interface between blood and vein wall.⁷ When in use, a latex membrane (EDM Imaging, Domont, France) is attached to the end of the device and filled with demineralized and degassed water, creating a water bag. The focal depth, indicated by a cross on the ultrasound imaging system (EvoTouch+, Quantel Medical, Cournon d’Auvergne, France), is aligned with the target by adjustment of the water balloon volume.

Preclinical study

The following protocol has been approved under APAFIS #50841-2024061815097801 v3 by the VetAgro Sup Ethics Committee (CEEA No. 18 Project No. 2447). On D0, the sheep was anesthetized and positioned laterally to target the veins of the hind leg, which was carefully shaved with a razor to optimize ultrasound coupling. The leg was imaged using ultrasound to map the course of the lateral saphenous vein (LSV) between 5 and 25 mm below the skin surface (L3-12T probe, Alpinion, Seoul, Republica of Korea).

Degassed ultrasound gel (EDM Imaging, Domont, France) was applied to the skin to ensure proper coupling between the device and skin. NT- HIFU device was then positioned on the hind leg to achieve a cross-section view of the vein. For the two sheep, HIFU was applied to the left and right LSV with pulsed ultrasound of 100μs, a PRF ranging from 250 Hz to 500 Hz. The peak electrical power, adapted to the depth of the vein to be treated, ranged from 900 W to 2000 W (corresponding to a peak negative pressure ranging from −17 to −23 MPa), consistent with typical mechanical HIFU parameters. Three different sonication duration were tested leading to three different protocols (A, B or C) as described in Table 1. HIFU treatment was applied to the portion of each LSV accessible, with the aim of distributing treatment points with a spacing of 5 mm along the targeted vein. Cavitation was monitored in real-time using ultrasound imaging.

Table 1.

Non-thermal HIFU treatment protocol.

Sample	Sheep	LSV	Protocol	Pulse duration (μs)	PRF (Hz)	Sonication duration (s)
1	1	Left	A	100	250	30
2	1	Right	A	100	250	30
3	2	Left	B	100	250	20
4	2	Right	C	100	500	10

Macroscopic examinations were conducted before and after treatment on awake animals to assess skin changes. B-mode and Color Doppler imaging (L3-12T probe, Alpinion, Seoul, Republica of Korea) were carried out before, during and after treatment to analyze the veins and assess perivenous tissue remodeling. These clinical and ultrasound examinations were performed on D3 without anesthesia and on D7 with anesthesia.

Euthanasia was performed at D10. The LSV marked with anatomical paint, were sampled over their entire accessible length on either side of the treated area and fixed in 4% buffered formaldehyde. Histological analysis was achieved on 9 to 12 sections of 3-4 μm thickness by steps of 8 mm all along the sampled vein, stained with hematoxylin-eosin (H&E) and digitalized with VS 200 (Evident) slide scanner to assess tissue injuries. Because the treated area was difficult to identify precisely, consecutive sections presenting with consistent abnormalities were considered corresponding to the effective treated area and selected for histological reporting.

Percentage of obliteration of the vein lumen was quantified from 0 to 100% in each section. For each vein wall layer (intima, media and adventitia), damage was qualified according to destruction, fibrosis and immune infiltration. The intensity of the lesions was assessed using a semi-quantitative score ranging from 0 to 4: 0: none, 1: slight, 2: moderate, 3: intense, 4: severe, in accordance with Gibson-Corley histological scoring proposal.⁸ Cellularity of the thrombus were similarly quantified from 0 to 4 for each histology section.

Results

The mean diameter of the treated LSV was 4.6 mm (min. 4.1, max. 5.0 mm). The length of the portion of LSV accessible for treatment was 5 cm. For each LSV, the number of HIFU treatment points and their estimated spacing are detailed in Table 2.

Table 2.

Non-thermal HIFU treatment patterns.

Sample	Sheep	LSV	Protocol	Nb. treatment pts	Spacing (mm)
1	1	Left	A	8	6.2
2	1	Right	A	12	4.1
3	2	Left	B	16	3.1
4	2	Right	C	5	10.0

Immediately after HIFU treatment, the retraction and obliteration of the treated vein were observed using ultrasound imaging in all LSV (Figure 1(a) and (b)). In one case (sample 2) partial recanalization was observed on D3 with secondary obliteration visible on D7. In all other cases, vein occlusion persisted until D7. Post-treatment perivenous tissue remodeling was observed with duplex examination in two cases (sample 1 and 2) as a hyperechoic perivenous area and in one case as a hypoechoic area suggestive of a hematoma and partially resolved on D7 (sample 3).

Figure 1.

Ultrasound images of the LSV (longitudinal view).

A localized skin damage with dermal thickening and induration was observed at D0 immediately after treatment in one case corresponding to overtreatment (protocol A sample 1) and was resolved by the follow-up on D7. For all other samples, no skin alteration was visible.

Among all LSV sections prepared for histologic analysis, 23 were available on two legs for protocol A, 11 for protocol B and nine for protocol C. Partial or total obliteration was observed in all sections. Vein wall abnormalities were present in 14 sections for protocol A (61%), in eight sections for protocol B (72%) and in four sections for protocol C (44%).

Considering the four selected histology sections of each treated LSV allowing a comparative analysis, a total obliteration of the vein was confirmed in each of the four LSV with total obliteration observed in 15 sections (93.8%) and partial obliteration in one section (6.2%) (Figure 2(a) and (b)). Cellular infiltration of the thrombus was observed in 15 sections (93.8%): graded at least as moderate in eight sections (50%) and at least as intense in seven sections (43.8%).

Figure 2.

Representative images of the LVS. (a) Untreated (b) after NT-HIFU treatment: Complete obstruction of the LSV. Scale bar 250 µm. (The small patents area are artefactual, due to tissue and clot retraction related to the tissue fixation process).

Detailed analysis of the damage observed to the vein wall layers is presented in Table 3. Overall, the main lesions observed for each vein wall layer were destruction for the intima layer, fibrosis for the media layer and inflammation for the adventitia. Destruction of the intima was observed in all 16 histology sections (100%), graded as intense or sever in nine sections (56.2%). Destruction was observed in the media in 11 sections (68.8%) and in the adventitia in five sections (31.2%) (Figure 3). Fibrosis was observed in the media in all 16 sections (100%), in the intima in seven sections (43.8%) and in the adventitia in five sections (31.2%) (Figure 4(a) and (b)). Immune infiltration was observed in the adventitia in eight sections (50%), in the media in five sections (31.2%) and in the intima in one section (6.2%) (Figure 5).

Table 3.

Detailed analysis of the damage observed to the vein wall layers.

Vein wall layer	Lesion	N (%)	Grade 3 to 4 N (%)
Intima	Destruction	16 (100)	9 (56.2)
	Fibrosis	7 (43.8)	3 (18.7)
	Immune infiltration	1 (6.2)	1 (6.2)
Media	Destruction	11 (68.8)	3 (18.7)
	Fibrosis	16 (100)	6 (37.5)
	Immune infiltration	5 (31.2)	1 (6.2)
Adventitia	Destruction	5 (31.2)	0
	Fibrosis	5 (31.2)	0
	Immune infiltration	8 (50)	0

Figure 3.

Representative picture of intimal destruction. Between dotted line is the place where endothelial cells should be present in a normal vessel.

Figure 4.

Representative images of media remodeling. (a) The whole media thickness is replaced by a fibrotic reaction, while smooth muscle cells are disorganized. (b) A hyalin reaction occurred between smooth muscle cells.

Figure 5.

Representative image of adventitia remodeling. Moderate alteration of adventitia: Red stars show little hemorrhages, yellow star shows a tissue remodeling, green stars show tissue disruption.

The mean extent of vein lumen obliteration, the mean intensity of thrombus cellularity and the mean intensity of the main lesions for each vein wall layer according to the treatment protocol (A, B or C) are presented in Table 4, calculated as the mean percentage of the vein affected across the relevant histology sections. The mean extent of obliteration decreases from 100% in protocol A and B to 87.5% in protocol C. In comparison, the mean intensity of thrombus cellular infiltration remains relatively stable at 31.2–37.5% across protocols. The mean intensity of destruction in the intima layer decreases from 75% in protocol A, to 68.8% in protocol B, and to 43.8% in protocol C. No decrease in the efficacy with respect to decreasing sonication duration is seen for fibrosis in the media layer and inflammation in the adventitia.

Table 4.

Distribution and intensity of the lesions according to the three protocols.

	Protocol A	Protocol B	Protocol C
Obliteration^a	400 (100%)	400 (100%)	350 (87.5%)
Thrombus cellular infiltration^b	5 (31.2%)	6 (37.5%)	6 (37.5%)
Intima (destruction)^b	12 (75%)	11 (68.8%)	7 (43.8%)
Media (fibrosis)^b	8.5 (53%)	11 (68.8%)	9 (56.2%)
Adventitia (immune infiltration)^b	2 (12.5%)	1 (6.3%)	3 (18.8%)

^aPercentage of lumen obliteration (maximum score = 400).

^bMean intensity of the lesion (maximum score = 16).

Discussion

The objective of this additional preclinical study is to further investigate in-vivo vein occlusion by NT- HIFU, including comprehensive histological analysis of the vein following endovenous cavitation. Several parameters have been tested to shorten the treatment time while obliterating efficiently the vein without complication, enabling optimization of the therapeutic protocol for subsequent clinical trials.

Considering the four LSV from two sheep treated in this study, vein occlusion was observed using ultrasound imaging in all samples immediately after treatment on D0 and remained visible in ultrasound images on D7 irrespective of the protocol used.

Histological analysis revealed that cavitation can induce significant alteration of the three layers of the vein wall. The most regularly observed lesions are destruction, followed by fibrosis and then inflammation. This can be explained by the short time between processing and sampling, inflammation and fibrosis appearing to be the result of initial lesions.

The histological analysis was limited in this study to four sections by LSV. The reason for it is that the precise identification of the treated segment was difficult to assess after vein sampling, furthermore significant vein retraction occurs after sampling and paraformaldehyde conservation. That is why four consecutive sections presenting with consistent histological abnormalities were considered corresponding with high confidence to the effective treated area.

The sheep included in our study, recommended in ISO 10993-6 for the evaluation of vascular treatments, has already been questioned by Boersma for mechanochemical endovenous ablation (MOCA) evaluation to explain the discordance of the results obtained in animal and human evaluation.⁹ Sheep have saphenous veins that are relatively like humans and thus represent a proven model for the preclinical evaluation of endovenous treatments. The mean LSV diameter in this study was 4.6 mm, which is like the mean diameter of saphenous veins in patients referred for suspected superficial vein insufficiency in clinical practice in France.¹⁰ However, due to the course of the LSV in sheep, only a short portion of the LSV, located 5–25 mm below the skin surface, was accessible for NT-HIFU treatment. An additional point concerns the significant difference in the perivenous tissue of sheep and humans: the former with muscle and strong fascia interposed between the skin and LSV which could perturb sound wave propagation, compared to subcutaneous fat tissue in the latter. Furthermore, a very high venous flow velocity was measured in the LSV of the sheep included in this study that could disrupt the cavitation effect. For all these reasons, we expect a higher efficacy of cavitation in humans than in sheep.

The variation in the number of sonication points per LSV is due to the anatomical presentation of the LSV in the lower leg of the sheep that determined the length of the portion of each LSV accessible for cavitation treatment. The objective of regularly spaced treatment points at 5 mm intervals by manual displacement of the HIFU probe was difficult to achieve and we estimate that HIFU was applied at irregularly spaced points ranging from 3.1 to 10 mm (see Table 2) in this study. We expect that the application of HIFU treatment at regularly spaced 5 mm intervals should be easier to achieve in patients due to a more regular course of varicose veins and a more plane contact surface of the leg with the probe.

The mechanical damage to the vein wall caused by cavitation is comparable to the mechanochemical techniques employed by devices such as Clariven™ and Flebogrif ™, both of which have been tested on the LSV of sheep. Vein occlusion was reported in only 66% of cases when the mechanical action of the Clariven™ device was associated with the injection of a sclerosing agent, no obliteration was reported when the device was used alone.⁹ No vein occlusion was reported using the Flebogrif™ device, either alone or in combination with the injection of a sclerosing agent.¹¹ Nevertheless, further studies have demonstrated that both devices are effective when applied to humans. Considering the higher vein occlusion efficacy we have demonstrated using cavitation with the Veinsound device (observed in all four LSV treated, i.e., 100%) compared to the Clariven™ and Flebogrif™ devices in animal models, we therefore expect a similarly higher efficacy to also be achieved in humans.

Thermal HIFU has been tested in sheep before to be used in human. A 1-month follow-up demonstrated vein wall hyalinization in half of the 6 LSV treated and LSV thrombosis in all the six LSV treated.¹² The follow-up in our study was significantly shorter, but it demonstrated that cavitation as another HIFU modality is also able to lead to venous occlusion using a different mechanism of action.

An objective of this preclinical study was to optimize the therapeutic protocol, ensuring the absence of complications, for subsequent clinical trials to evaluate the approach for the non-invasive transcutaneous treatment of varicose veins in patients. Any change or damage to the skin following treatment was thus carefully monitored. Though short-term skin damage occurred in one case (observed on D0 and resolved by D7) no skin injury was observed in other cases. We observed that despite optimal shaving, in opposite to humans, residual pilosity of some sheep can create an interface likely to trigger unappropriated cavitation at the skin level especially in case of over treatment leading to skin injury. This study suggests that effects to the skin can be avoided if appropriate NT-HIFU dose are settled and adequate coupling between the HIFU probe and skin surface is ensured.

Perivenous tissue remodeling was observed by ultrasound imaging following treatment in three cases: two for protocol A and one for protocol B (partially resolved by D7), indicating a possible relationship with sonication duration. These lesions could be due to extra-venous cavitation, potentially caused by the thick triangle-shaped fascia that surrounds the LSV in sheep and creates a significant acoustic interface capable of modifying HIFU propagation. Post-treatment perivenous remodeling has been reported in studies involving techniques such as endovenous laser and thermal HIFU.^12,13 In the latter case, the perivenous muscle and fat necrosis was completely resolved at the 3-months follow-up.¹² A longer follow-up than the 1-week duration of our study could, therefore, have enabled the resolution of the lesions we observed.

Vein occlusion was achieved for all treated LSV in this study irrespective of the protocol (A, B or C). These results thus indicate that a sonication time of less than or equal to 20 s, with a 5 mm spacing between treatment points, should be considered in clinical trials for the evaluation of NT-HIFU for the non-invasive treatment of varicose veins.

Conclusion

This study demonstrated that endovenous cavitation using Veinsound device can obtain immediate and persistent vein occlusion in sheep. At 1 week follow-up, histological analysis as shown in 100% of the cases destruction lesions of intima and fibrosis of the media. Because of the short follow-up of the study the mid- and long-term evolution of veins treated by non-thermal HIFU is still hypothetic but these initial lesions let hope the secondary occurrence of a retractile fibrosis process as it is observed with current endovenous treatments. Further studies will be necessary to confirm this hypothesis in human in which non thermal HIFU treatment should be applied in more suitable conditions than in sheep.

Footnotes

Acknowledgements

We thank the BioVivo laboratory where the in-vivo experiments were performed.

ORCID iD

Olivier Pichot

Ethical considerations

In-vivo experiments were conducted in the Biovivo laboratory, accredited by AAALAC International. The protocol was approved under APAFIS #50841-2024061815097801 v3 by the VetAgro Sup Ethics Committee (CEEA No. 18 Project No. 2447).

Author contributions

OP, PC, AG, and CM participated in the animal study. VD performed the histological analysis. OP drafted the article, all authors have made substantial contribution to the analysis and interpretation of data, revising the article, approved the manuscript, and agree with its submission to Phlebology.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Focused Ultrasound Foundation (FUS) provided financial support to this study (FUSF-RAP-S-18-00022).

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Olivier Pichot, Pierre Combes and Cyril Lafon own shares of the company Veinsound. Alice Ganeau and Charlotte Millet are employees of Veinsound.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*

Guarantor

OP.

References

De Maeseneer

Kakkos

Aherne

, et al. Editor’s choice - European society for vascular surgery (ESVS) 2022 clinical practice guidelines on the management of chronic venous disease of the lower limbs. Eur J Vasc Endovasc Surg 2022; 63(2): 184–267. Epub 2022 Jan 11. Erratum in: Eur J Vasc Endovasc Surg. 2022 Aug-Sep;64(2-3):284-285.

Whiteley

. High intensity focused ultrasound (HIFU) for the treatment of varicose veins and venous leg ulcers - a new non-invasive procedure and a potentially disruptive technology. Curr Med Res Opin 2020; 36(3): 509–512.

Mendiratta-Lala

Wiggermann

Pech

, et al. The #HOPE4LIVER single-arm pivotal trial for histotripsy of primary and metastatic liver tumors. Radiology 2024; 312(3): e233051.

Goudot

Khider

Del Giudice

, et al. Non-invasive recanalization of deep venous thrombosis by high frequency ultrasound in a swine model with follow-up. J Thromb Haemostasis 2020; 18(11): 2889–2898.

Trifunović-Zamaklar

Karan

Kovačević-Kostić

, et al. Non-invasive ultrasound therapy for severe aortic stenosis: early effects on the valve, ventricle, and cardiac biomarkers (A case series). J Clin Med 2024; 13(16): 4607.

Ganeau

Charrel

Pichot

, et al. Preclinical demonstration of in-vivo vein occlusion by mechanical HIFU without contrast agent injection: paving the way for the non-invasive treatment of varicose veins. Ultrasound Med Biol 2025; 51: S0301562925001218.

Khokhlova

Cho

, et al. Histotripsy: a method for mechanical tissue ablation with ultrasound. Annu Rev Biomed Eng 2024; 26(1): 141–167.

Gibson-Corley

Olivier

Meyerholz

. Principles for valid histopathologic scoring in research. Vet Pathol 2013; 50(6): 1007–1015.

Boersma

van Haelst

van Eekeren

, et al. Macroscopic and histologic analysis of vessel wall reaction after mechanochemical endovenous ablation using the ClariVein OC device in an animal model. Eur J Vasc Endovasc Surg 2017; 53(2): 290–298. Epub 2016 Dec 23.

10.

Hamel-Desnos

De Maeseneer

Josnin

, et al. Great saphenous vein diameters in phlebological practice in France: a report of the DIAGRAVES study by the French society of phlebology. Eur J Vasc Endovasc Surg 2019; 58(1): 96–103.

11.

Rybak

Janeczek

Dobrzynski

, et al. Study of Flebogrif^®-a new tool for mechanical sclerotherapy-effectiveness assessment based on animal model. Nanomaterials 2021; 11(2): 544.

12.

Barnat

Grisey

Gerold

, et al. Vein wall shrinkage induced by thermal coagulation with high-intensity-focused ultrasound: numerical modeling and in vivo experiments in sheep. Int J Hyperther 2020; 37(1): 1238–1247.

13.

Vuylsteke

Van Dorpe

Roelens

, et al. Intraluminal fibre-tip centring can improve endovenous laser ablation: a histological study. Eur J Vasc Endovasc Surg 2010; 40(1): 110–116.