Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Oct 8.
Published in final edited form as: Ultrasound Med Biol. 2026 Sep 26;52(12):2855–2865. doi: 10.1016/j.ultrasmedbio.2026.08.023

Investigation of High PRF Single-Cycle Histotripsy for Rapid Tissue Ablation

Alex Simon 1,*, Lauren Ruger 2,*, Elliana Vickers 3, Leah Holmes 2, Tamalika Paul 4, Aislynn Macco 2, Sheryl Coutermarsh-Ott 4, Adam Maxwell 2, Eli Vlaisavljevich 2
PMCID: PMC13644694  NIHMSID: NIHMS2213081  PMID: 42800774

Abstract

Objective:

Single-cycle histotripsy was investigated for the rapid ablation of porcine liver using high pulse repetition frequency (PRF) pulsing methods. Recent evidence from our group has suggested that single-cycle histotripsy does not experience a decrease in ablation efficiency across a wide range of PRFs.

Methods:

Histotripsy ablation efficiency was quantified in a 1% agarose gel phantom containing a layer of red blood cells (RBCs) using a 500 kHz transducer with an F# of 0.61. Lesions were generated using PRFs of 500–2500 Hz at peak negative pressures of 15, 24, and 42 MPa with an 800 pulse exposure, with results compared to a low PRF control (1 Hz). Second, high PRFs were used for the volumetric ablations of porcine liver using doses of 400–800 pulses per point. Tissue damage was assessed via ultrasound imaging and histologically using Masson’s trichrome staining.

Results:

Ablation experiments in RBC gel phantoms showed no significant differences in ablation efficacy as PRF increased from 500 Hz to 2500 Hz, supporting the hypothesis that single-cycle histotripsy can achieve efficient tissue ablation at high PRFs. Volumetric tissue ablation experiments showed that single-cycle histotripsy resulted in effective tissue ablation for all tested PRFs, with near complete or complete ablation observed after 600–800 applied pulses. Partial ablation in some samples at lower doses (400 pulses for 500–1500 Hz PRF, 600 pulses for 2000–2500 Hz PRF, and 800 pulses for 2500 Hz PRF) suggested a slight decrease in ablation effectiveness with increasing PRF.

Conclusion:

This study supports the hypothesis that high PRF single-cycle histotripsy can achieve faster tissue ablation while maintaining precise volumetric lesion generation.

Keywords: Histotripsy, Ablation, Pulse Repetition Frequency, Volumetric Ablation, High-Rate

Introduction

Histotripsy is a cavitation-based ablation modality that can be used to non-invasively disintegrate targeted tissues, such as malignant tumors. Using high pressure focused ultrasound pulses, a dense histotripsy bubble cloud can be non-invasively generated inside the tissue when sufficiently high focal pressures are achieved [1]. The formation of this bubble cloud, which consists of numerous microscopic cavitation bubbles that rapidly expand and collapse, results in the mechanical breakdown of the tissue structures in the immediate vicinity of the bubble cloud [2, 3]. With successive pulses, the expansion and collapse of these bubbles break down the tissue to an acellular liquid homogenate [4]. By moving the focal point of the transducer through physically manipulating the array or electronical focal steering [5, 6], a three-dimensional volume can be targeted for ablation. Histotripsy has shown promise for many clinical applications and was approved by the FDA for the treatment of liver tumors in 2023 [7], with additional clinical trials for pancreatic [8] and kidney [9] cancers underway. Histotripsy is also being explored for many other applications preclinically including the non-invasive treatment of musculoskeletal tumors [10], brain tumors [11], uterine fibroids [12], benign prostatic hyperplasia [13], and deep vein thrombosis [14] where debulking or complete removal of the tissue is desired.

The rate at which the histotripsy therapy is delivered at a given focal location is marked by the number of pulses sent per second or the pulse repetition frequency (PRF). Prior investigations into shock-scattering histotripsy showed that increasing the PRF decreased ablation efficiency due to the cavitation memory effect, which is the restimulation of cavitation events in approximately the same spatial locations across pulses [15, 16]. This process limits the ability for subsequent pulses to nucleate cavitation events in new locations, leaving portions of the focus without cavitation and reducing treatment efficiency [17]. Recently, our group conducted a systematic investigation into the effects of PRF on single-cycle histotripsy [18]. Results from that study showed that higher PRFs did not experience a decrease in ablation efficiency as was seen in shock-scattering histotripsy even though cavitation memory effects were still observed. This finding was explained in large part to be due to the densely generated bubble clouds from single-cycle pulsing when using a low f-number transducer (0.5–0.9) as is typical for single-cycle histotripsy pulsing methods; in contrast, shock-scattering histotripsy typically employs transducers with a higher f-number (0.7–1.0) [19]. When treating at higher PRFs in that study, complete ablation of the focus was achieved in tissue phantoms for all tested PRFs with a similar ablation efficiency, providing evidence that higher PRFs could be used for enhancing tissue ablation rates when using single-cycle histotripsy and suggesting further investigation may be warranted for high PRF shock-scattering histotripsy. Another key finding from our prior work, which has also been observed with shock-scattering histotripsy [20], was that the pressure required to generate a bubble cloud significantly decreased at higher PRFs. For low PRFs (<10 Hz), the cavitation threshold was repeatedly measured around the intrinsic threshold, pt, of 26.0±1.8 MPa [21, 22]. At higher PRFs, the cavitation cloud threshold decreased, reaching a level of 15.0±1.9 MPa at 1000 Hz PRF. High speed optical imaging helped to explain the process of cloud formation in these cases, with results showing that initial cavitation events occurred at sub-intrinsic threshold pressures (<25 MPa) for all PRFs, with the minimum pressure required to generate any cavitation bubbles, pmin, shown to be ~12–14 MPa and consistent across all PRFs. At pressure levels between pmin and pt, cavitation occurred at low probabilities <50% (i.e., bubbles were not formed on every pulse). At PRFs <10 Hz, results showed that these low probability cavitation bubbles only lasted for a single pulse, or a few pulses, before being extinguished. At higher PRFs, however, these cavitation bubbles were restimulated with successive pulses, resulting in the formation of additional bubbles on subsequent pulses until a full histotripsy bubble cloud was produced. This finding helps to explain results from prior in vivo histotripsy treatments using single-cycle histotripsy to ablate tissues at pressures as low as ~20 MPa when using PRFs of 1000–1500 Hz [10, 23]. The findings of this study also suggest that high PRF single-cycle histotripsy methods may be developed to reduce volumetric ablation treatment times, which is essential to advancing histotripsy towards the treatment of large tumor volumes or difficult to treat tissues (i.e., fibrous tissues) requiring higher doses to destroy the tissue [12]. Reducing the pressure required to produce and maintain a histotripsy bubble cloud may also enable robotically-guided histotripsy treatments to be applied using smaller transducers and with less risk of generating off-target effects.

While previous studies using multi-cycle histotripsy methods suggest there is a point of diminishing return when increasing the PRF [15, 20], we hypothesized that single-cycle histotripsy ablation will remain efficient on a per pulse basis even at very high PRFs that have not been previously explored for single-cycle histotripsy (i.e., PRFs > 1000 Hz). To test this hypothesis, this study evaluated histotripsy ablation for PRFs of 500–2500 Hz using a 700 kHz histotripsy transducer and single-cycle pulsing. First, the ability to ablate tissue at high pulsing rates was assessed in vitro using porcine red blood cell phantoms. Next, the pressure needed to generate a bubble cloud was measured ex vivo in porcine liver. Finally, we evaluated the volumetric ablations generated at high PRFs in ex vivo porcine liver at treatment doses of 400–800 pulses/point.

Materials and Methods

Histotripsy Platform and Transducer Calibration

The pre-clinical treatment cart (Cloud Break, Histosonics, Inc., Plymouth, MN, USA) used in this study, shown in Figure 1A, consisted of an amplifier (Histosonics, Inc., Plymouth, MN, USA), power supply (GENH1500W, TDK-Lambda, Tokyo, Japan), robotic positioning arm (UR5e, Universal Robots USA, Inc., Novi, MI, USA), and a custom built 700 kHz histotripsy transducer (ImaSonic SAS, Voray-sur-l’Ognon, France) with a coaxially aligned US imaging probe (BK Ultrasound 5C1e, BK Medical, Burlington, MA, USA) (Figure 1B). The focal length of the 700 kHz histotripsy transducer was 11 cm, and the aperture size was 18 cm, yielding an F# of 0.61.

Figure 1. High PRF Ablation Setup.

Figure 1.

Open in a new tab

(A) The histotripsy treatment cart consisted of a user-controlled interface, a robotic treatment arm, and a 700 kHz single-cycle transducer with coaxially aligned US imaging (B.) capable of pulsing at very high PRFs (C.). Tissue samples were placed at the bottom of the tank with the focus of the transducer in the center of the tissue sample. Ex vivo ablation of porcine liver was monitored using real time US imaging (E.). Damage assessment was performed using US imaging and histologically.

The 700 kHz single-cycle histotripsy transducer was calibrated using a fiber optic hydrophone (HFO-50, ONDA, Sunnyvale, CA, USA). Figure 2 shows an example single-cycle histotripsy waveform (Figure 2A), corresponding beam profiles from the transducer (Figure 2 B–D), and an example bubble cloud generated in a 1% agarose phantom (Figure 2E). The geometric focus of the transducer was measured to be 3 × 3 × 8 mm (18.85 mm2) at −6dB. The high pressures listed in this study were estimated based on a linear extrapolation from direct measurements at low pressures (<15.5 MPa) at which cavitation did not occur.

Figure 2. Calibration Results.

Figure 2.

Open in a new tab

The 700 kHz histotripsy transducer was calibrated using a fiber optic hydrophone in degassed water. The focal waveform (A) is a characteristic single-cycle histotripsy pulse with a peak negative pressure of 15 MPa. The beam profiles of the transducer in the transverse (B), azimuthal (C), and axial (D) directions were scaled to 24 MPa to match the pressure required to generate a bubble cloud in a 1% agarose gel phantom (E).

Tissue Sample Collection and Preparation

For in vitro analysis of high PRF ablations, porcine blood was collected from local harvests (VT Meat Center, Blacksburg, VA & Firehouse Farms, Floyd, VA). During harvest, blood was immediately mixed with 1/10 solution of CPD and refrigerated while in storage. On the day of experiments, 0.5% red blood cell (RBC) gel phantoms were made following a previously defined protocol15, 17. Briefly, for each gel phantom, the bottom half of a 2 × 1 × 1.5 in rectangular silicon mold was filled with 1% agarose, and solidified. While the gel was cooling, the blood and CPD solution was taken out of refrigerated storage to reach room temperature. The solution was then centrifuged at 3000rpm for 10min to separate the blood cells from plasma. Once separated, 0.5mL of RBCs were mixed with 9.5mL of degassed 1% agarose. The gently mixed solution was poured onto the bottom layer and balanced to spread the RBC solution evenly across the surface of the gel. Excess RBC solution was dumped out of the holder quickly before the thin layer of agarose could solidify. The RBC and bottom layer were then refrigerated for approximately 10min to solidify enough to pour the top layer of 1% agarose gel without disturbing the RBC layer. The finished RBC gel phantoms were then refrigerated for 1.5 hours prior to treatments to ensure the gel had completely solidified. All RBC experiments were performed within 12 hours after RBCs were suspended.

For bubble cloud generation testing and ex vivo volumetric ablations, whole porcine livers were also collected from each harvest and treated within 48 hours. Liver samples were submerged in Dulbecco’s Modified Eagle Medium (Dulbecco’s Modified Eagle Medium, Gibco, NC, USA), and refrigerated while in storage. Prior to treatments, tissues were sectioned into 3 × 5 × 3 cm rectangular blocks and degassed in 0.9% saline for 2-hours inside a desiccator at 75 kPa. Degassed tissue samples were encapsulated in degassed 7.5% porcine gelatin phantoms (G1890, Sigma-Aldrich, MO, USA) and refrigerated for roughly 2 hours to solidify prior to testing.

Red Blood Cell Ablation and Analysis

Ablation efficiency was assessed in vitro by measuring the destruction of the thin layer of the RBC gel phantoms, following previously published methods [24, 25]. A high speed camera (FastCAM Nova S12, Photron, WA, USA; Zhong Yi Optics, Beijing, China) with an 85 mm lens (Mitakon Zhongyi Creator 85 mm f/2.8 1–5x Super Macro Lens) at 3:1 and a high-intensity light array (GS Vitec Multi-LED QT Light, MultiLED G8 controller, 320 W power supply, Soden-Salmünster, Germany) at 100% intensity was used to visualize the destruction of the layer. The exposure time was set to 5μs with an image resolution of 1024 × 1024 and pixel size of 8.5 μm/pixel.

The high speed camera was aligned to the focus of the transducer by focusing on bubble clouds in water (Figure 2E). Using optical imaging, the center of the RBC layer was positioned to the focus of the transducer. Three different treatment pressures were used for this study corresponding to a relative low pressure (15 MPa), an intermediate pressure (24 MPa), and a higher pressure (42 MPa). These correspond to pressures that are below, near, and above the histotripsy intrinsic threshold [21, 22]. For each pressure, PRFs of 500, 1000, 1500, 2000, and 2500 Hz (duty cycle (DC) = 0.07%, 0.14%, 0.21%, 0.29%, 0.36%, respectively) were used to deliver 800 pulses to a single point without moving the transducer array (n = 3/group). For comparison, an additional set of experiments was conducted at 1 Hz PRF (DC = 0.0001%) as a control representing the intrinsic threshold condition. All experiments were conducted at room temperature (~20–22°C).

Images were taken at 6.5 and 300 μs after pulse arrival to capture the bubble cloud and corresponding lesion, respectively. Images were analyzed for total ablation area by calculating the removal of contrast from the targeted region using a previously defined analysis [15] and plotted as a function of the pulse number. Ablation efficiency for each condition was calculated by determining how many pulses were required to achieve 95% of the final lesion area. At each pressure, ablation areas and efficiencies were compared across PRFs using one-way ANOVA with Tukey’s multiple comparisons or the nonparametric equivalent test (Kruskal-Wallis test with Dunn’s multiple comparisons) (α = 0.05). To determine the effect of treatment pressure on ablation efficiency, results from all PRFs were averaged and compared using a one-way ANOVA with Tukey’s multiple comparison (α = 0.05). Normality of residuals was assessed via the Shapiro-Wilk test for this analysis and all statistical tests in this study unless otherwise noted. Datasets were considered to follow a normal (Gaussian) distribution if p > 0.05.

Bubble Cloud Generation Testing

The pressure required to initiate a bubble cloud at high PRFs was tested ex vivo in porcine liver. Prior to testing, a treatment tank was filled with water and degassed to <30% dissolved O2 at room temperature (~20–22°C). The histotripsy transducer was positioned in the tank and solidified liver phantoms were aligned to the focus of the transducer using US imaging. For each test (n=6/PRF), 5 points within the central slice of a targeted 1.5 cm diameter spherical volume (center, xmin, xmax, ymin, ymax) at equal depth (axial-direction) were selected for bubble cloud generation testing at PRFs of 500, 1000, 1500, 2000 and 2500 Hz. The depth was held consistent for each treatment point to minimize attenuation or aberration changes between samples. For each point, the pressure output from the transducer was increased by 1–2 MPa every 3 s until a bubble cloud formed at the focus inside the liver sample on US imaging, following a protocol similar to what is done for test pulsing on the HistoSonics clinical system [10, 26]. Once the hyperechoic bubble cloud was detected on ultrasound imaging, the pressure was recorded. To compare the cavitation threshold between PRFs and across different liver samples, two-way ANOVA (factors of PRF and sample) with Tukey’s multiple comparisons was used (α = 0.05).

Volumetric Ablation at High PRFs

The ability to ablate tissue volumetrically at high PRFs using single-cycle histotripsy was investigated ex vivo in porcine liver. In a tank of room temperature degassed water (~20–22°C), solidified liver gel phantoms were fixed at the bottom of the tank with a 3D printed holder (Figure 1D). Using US imaging, the focal point of the transducer was lowered into the center of the tissue sample. In pre-treatment planning, a ~1.75 cm3 spherical volume (1.5 cm diameter) was targeted for ablation. For each excised liver block, two isolated treatment volumes were targeted with no overlap between targeted regions. Treatment volume locations in the tissue were selected to ensure the bubble cloud was encapsulated in the tissue for the entirety of treatment. Treatment point spacing was 0.72 mm in the lateral and azimuth directions and 3.36 mm in the axial direction which corresponded to approximately half of the dimensions of the bubble cloud measured in 1% agarose phantoms using 1000 Hz PRF at an estimated peak negative pressure of 24 MPa. This packing method has been previously shown to generate complete ablations in liver [27–29]. Treatments (n=3/group) were performed at a pressure of 24 MPa. For PRFs of 500, 1000, and 1500 Hz, treatment dosages of 400 and 600 pulses per point were tested. For the higher PRF treatments in this study (2000 and 2500 Hz), treatment doses of 600 and 800 pulses per point were used. Minimum treatment doses at high PRFs (2000 and 2500 Hz) were limited by the maximum allowable speed of the robotic positioner on our treatment cart. Treatments were monitored in real time using ultrasound imaging (Figure 1E). Post treatment US scans were used to clearly identify and mark the ablation zone for histological analysis.

Histological Analysis

After treatment, porcine liver samples were immediately scanned using US imaging to locate the ablation zone. Pins were used to mark the extremities of the affected areas. Tissue samples were then removed from the gelatin phantom and submerged in 10% formalin. Fixed tissue samples were sliced down the center of the ablation zone as marked by the pins and submitted in cassettes for routine embedding in paraffin. Then, samples were cut into 5 μm sections and stained with Masson’s trichrome (ViTALS, Virginia-Maryland College of Veterinary Medicine, Blacksburg, VA, USA) to show the full extent of damage to the connective structures in the liver.

The percentage of intact tissue remaining in the targeted area after histotripsy treatment was digitally quantified for each sample using a previously defined algorithm and reported as mean percent ablation ± standard deviation [30]. Two-way ANOVA (factors of PRF and dose) with Tukey’s multiple comparisons was used to compare the percentage of ablated cells on histopathology across PRFs and low and high dose groups (α = 0.05). Additionally, the resulting two-dimensional lesion size was measured using histological analysis (Motic DSAssistant, Motic Digital Pathology, Emeryville, CA, USA) and compared against the planned target volume as well as across PRFs and doses. Two-way ANOVA (factors of PRF and dose) with Tukey’s multiple comparisons was used to compare ablation diameter on histopathology across PRFs and doses. Then, a two-tailed, one sample t-test was used to evaluate the difference between achieved and planned ablation diameters across all samples (α = 0.05).

Results

Red Blood Cell Ablation in vitro

High speed optical imaging of bubble clouds generated in a thin layer of RBCs showed that ablation area scaled with pressure, with higher pressures resulting in larger bubble clouds and lesions. At each pressure level tested, results showed no clear trends in ablation efficiency as a function of PRF. For instance, for RBC ablations using pressures at 15 MPa, the cavitation behavior observed in high speed imaging showed a consistent pattern of a single or small grouping of cavitation events that persisted for the duration of the test (Figure 3). The formation of these initial cavitation events occurred after a variable number of pulses, particularly for PRFs of 500–1500 Hz, resulting in a high degree of variability in the ablation area from pulse #1–400 at this pressure level (Figure 3). Results for these 15 MPa experiments showed the final lesion area to be 0.66±0.09 mm2 at 500 Hz, 0.79±0.13 mm2 at 1000 Hz, 0.28±0.27 mm2 at 1500 Hz. At the two highest PRFs tested at 15 MPa, 2000–2500 Hz PRF tests resulted in the largest lesion areas of 1.08±0.11 mm2 at 2000 Hz and 0.95±0.41 mm2 at 2500 Hz by the end of the 800-pulse exposure. At 15 MPa, there was a significant effect of PRF on ablation area [F(4,10) = 5.030, p = 0.0175], with a significantly increased area measured at PRFs of 2000 Hz and 2500 Hz compared to 1500 Hz (p < 0.05). No bubbles formed at 15 MPa at the lowest PRF of 1 Hz; as a result, this PRF was excluded from analysis.

Figure 3. RBC Ablation Imaging @ 15 MPa.

Figure 3.

Open in a new tab

High speed optical imaging of bubble clouds generated at 1, 500, 1000, 1500, 2000, and 2500 Hz PRFs inside 1% Agarose RBC gel phantoms at 15 MPa on pulse # 10, 200, 400, and 800. Images were captured at 6.5 and 300 us after pulse arrival for each pulse.

At 24 MPa (Figure 4), high PRFs generated isolated cavitation events in the initial pulses of the exposure. As more pulses were delivered, these cavitation events extended into the formation of bubble clouds with more cavitation events covering a greater area. At the end of the test, the final lesion areas were 3.46±1.00 mm2 at 500 Hz, 3.24±1.31 mm2 at 1000 Hz, 4.24±1.14 mm2 at 1500 Hz, 5.78±1.58 mm2 at 2000 Hz, and 4.64±1.08 mm2 at 2500 Hz. In each case, a complete removal of RBC contrast was observed. At 24 MPa, there was a significant effect of PRF on ablation area [F(5,12) = 8.373, p = 0.0013], with a significantly smaller area observed at 1 Hz compared to 500 Hz, 1500 Hz, 2000 Hz, and 2500 Hz (p < 0.05 for all comparisons). While cavitation was consistently generated for all PRFs, 1 Hz pulsing rates only generated single bubbles at 24 MPa, as expected from the intrinsic threshold curve, which resulted in minimal removal of contrast and less efficiency in ablation in comparison with the higher PRFs.

Figure 4. RBC Ablation Imaging @ 23 MPa.

Figure 4.

Open in a new tab

High speed optical imaging of bubble clouds generated at 1, 500, 1000, 1500, 2000, and 2500 Hz PRFs inside 1% Agarose RBC gel phantoms at 24 MPa on pulse # 10, 200, 400, and 800. Images were captured at 6.5 and 300 us after pulse arrival for each pulse.

For experiments conducted at 42 MPa (Figure 5), high PRFs generated larger lesions compared to the 1 Hz PRF ablations, which showed ablations resembling characteristic single-cycle histotripsy treatments with a high fidelity to the ellipsoidal focal region. Minimal cavitation was observed outside the elliptical ablation area (14.2±0.40 mm2) at 1 Hz PRF. For higher PRFs (500–2500 Hz), while pulse one generated similar bubble clouds to the 1 Hz tests, as more pulses were delivered, cavitation occurred throughout the field reaching well beyond the predicted focal area. The high PRF cases at these pressures showed large variations in lesion geometry between samples (Figure 5), but generated consistent ablation areas in the central region of the focus despite a decrease in focal fidelity. The average final lesion area, shown in Figure 5, for 500 Hz PRF was smaller than 1 Hz ablations (14.2±0.40 mm2 at 1 Hz and 11.1±0.72 at 500 Hz), but as PRF increased, the final lesion area increased to 15.4±1.33 mm2 at 1000 Hz, 23.5±10.4 mm2 at 1500 Hz, 21.4±1.33 mm2 at 2000 Hz, and 21.8±1.55 mm2 at 2500 Hz. At 42 MPa, there was a significant effect of PRF on ablation area (Kruskal-Wallis test, p = 0.0185), with no significant post-hoc differences in area between PRFs. Contributing to these larger areas was the numerous cavitation events observed throughout the image, demonstrating a decrease in precision and more irregular geometries with higher PRFs.

Figure 5. RBC Ablation Imaging @ 42 MPa.

Figure 5.

Open in a new tab

High speed optical imaging of bubble clouds generated at 1, 500, 1000, 1500, 2000, and 2500 Hz PRFs inside 1% Agarose RBC gel phantoms at 42 MPa on pulse # 10, 200, 400, and 800. Images were captured at 6.5 and 300 us after pulse arrival for each pulse.

When normalized by the final lesion area, results showed no significant differences in the pulses required to achieve 95% of the total lesion area for 500–2500 Hz PRFs (743±29 pulses for 500 Hz and 707±32 pulses at 2500 Hz, at 15 MPa [F(4,10) = 0.5353, p = 0.7133]). Cavitation was not consistently observed for 1 Hz tests and no ablation was achieved in the RBC layer (Figure 6A). Furthermore, no significant difference in ablation efficiency was observed between PRFs at 24 MPa, with 95% of the final ablation area achieved in 584±103 pulses at 500 Hz, 660±87 pulses at 1000 Hz, 624±36 pulses at 1500 Hz, 694±52 pulses at 2000 Hz, and 674±18 pulses at 2500 Hz [F(5,12) = 2.281, p = 0.1123]. At 42 MPa, there was a significant effect of PRF on the number of pulses needed to achieve 95% ablation [F(5,12) = 14.35, p = 0.0001]. Ablations performed at 1 Hz PRF had the highest ablation efficiency, with complete ablation of the target area occurring in 536±24 pulses in comparison with 691±22 pulses at 500 Hz, 670±37 pulses at 1000 Hz, 689±11 pulses at 2000 Hz, and 648±16 pulses at 2500 Hz (p < 0.05 for all comparisons). There was also a significant increase in efficiency at 1500 Hz (586±47 pulses) compared to PRFs of 500 Hz, 1000 Hz, and 2000 Hz (p < 0.05). Furthermore, there was significant effect of pressure on the average number of pulses needed to achieve 95% ablation [F(2,48) = 9.865, p = 0.0003], with a significant increase in the average ablation efficiency at the medium (24 MPa) and high (42 MPa) pressure levels (664±76 pulses and 637±64 pulses, respectively) compared to the low pressure level (730±30 pulses at 15 MPa) (p < 0.05). There was no change in the ablation efficiency between the 24 MPa and 42 MPa cases. The average ablation area and number of pulses need to achieve 95% ablation for each condition are summarized in Table 1.

Figure 6. RBC Ablation Area vs Pulse Number.

Figure 6.

Open in a new tab

Ablation area at (A) 15, (B) 24, and (C) 42 MPa in RBC gel phantoms after an 800 pulse exposure for 1, 500, 1000, 1500, 2000, and 2500 Hz PRFs. The ablation area (mm2) was measured by calculating the area of the removed contrast in the lesion images for each pulse.

Table 1.

Findings from red blood cell phantom ablations and associated statistical analyses. Results are reported as mean ± standard deviation. A p-value < 0.05 was considered to be statistically significant.

Pressure PRF Mean Ablation Area (mm2) ANOVA Results Post-hoc Findings (p < 0.05) 95% Ablation (# Pulses) ANOVA Results Post-hoc Findings (p < 0.05)
15 MPa 500 Hz 0.66 ± 0.09 F(4,10) = 5.030
p = 0.0175		 2000 Hz – 1500 Hz
2500 Hz – 1500 Hz		 743 ± 29 F(4,10) = 0.5353
p = 0.7133		 N/A
1000 Hz 0.79 ± 0.13 707 ± 32
1500 Hz 0.28 ± 0.27 735 ± 44
2000 Hz 1.08 ± 0.11 727 ± 33
2500 Hz 0.95 ± 0.41 736 ± 17
24 MPa 1 Hz 0.20 ± 0.22 F(5,12) = 8.373
p = 0.0013		 500 Hz – 1 Hz
1500 Hz – 1 Hz
2000 Hz – 1 Hz
2500 Hz – 1 Hz		 746 ± 49 F(5,12) = 2.281
p = 0.1123		 N/A
500 Hz 3.46 ± 1.00 584 ± 103
1000 Hz 3.24 ± 1.31 660 ± 87
1500 Hz 4.24 ± 1.14 624 ± 36
2000 Hz 5.78 ± 1.58 694 ± 52
2500 Hz 4.64 ± 1.08 674 ± 18
42 MPa 1 Hz 14.2 ± 0.40 p = 0.0185* N/A 536 ± 24 F(5,12) = 14.35
p = 0.0001		 500 Hz – 1 Hz
1000 Hz – 1 Hz
2000 Hz – 1 Hz
2500 Hz – 1 Hz
1500 Hz – 500 Hz
1500 Hz – 1000 Hz
2000 Hz – 1500 Hz
500 Hz 11.1 ± 0.72 691 ± 22
1000 Hz 15.4 ± 1.33 670 ± 37
1500 Hz 23.5 ± 10.4 586 ± 47
2000 Hz 21.4 ± 1.33 689 ± 11
2500 Hz 21.8 ± 1.55 648 ± 16
Open in a new tab
*

Due to the dataset’s non-normal distribution, a Kruskal-Wallis test with Dunn’s multiple comparisons was used in place of ANOVA.

Bubble Cloud Generation Testing

The results from the bubble cloud initiation testing in ex vivo porcine liver showed no significant effect of PRF [F(4,120) = 0.6433, p = 0.6326] or sample [F(5,120) = 0.4613, p = 0.8044] on the cavitation threshold. Additionally, no significant effect of the interaction between PRF and sample on cavitation threshold was observed [F(20,120) = 1.485, p = 0.0989]. The peak negative pressure pressures required to generate a bubble cloud for each PRF were as follows: 500 Hz (15.5±3.15 MPa), 1000 Hz (15.4±2.45 MPa), 1500 Hz (16.0±2.73 MPa), 2000 Hz (15.4±2.38 MPa), and 2500 Hz (16.3±3.57 MPa) (Figure 7). It was noted that during threshold testing, when cavitation did occur, the bubble cloud started from initially detectable cavitation events and grew larger on US imaging into a robust hyperechoic bubble cloud as the exposure continued during the 3 second interval.

Figure 7. Bubble Cloud Generation Testing.

Figure 7.

Open in a new tab

The onset of cavitation was measured ex vivo in porcine liver and in vitro in 1% agarose gel phantoms for each PRF (500 – 2500 Hz). The peak negative pressure was increased by 1–2 MPa every 3 s until a well-defined bubble cloud was observed on US imaging. Results showed consistent cavitation pressures for each PRF around 15.5 MPa.

Ex Vivo Volumetric Ablation of Porcine Liver

Volumetric ablations of ex vivo porcine liver showed that all PRFs were capable of generating precise volumetric ablations and were fully visualized in real-time with ultrasound imaging despite an increase in image artifacts from the faster pulsing rates. The treatment time for 600 pulses per point treatments decreased from 17 minutes 09 seconds at 500 Hz to 4 minutes 18 seconds at 2500 Hz. Post treatment US imaging showed some hyperechoic shading at 400 pulses per point and dark hypoechoic shading corresponding to a liquefaction of the targeted volume after 600 pulses per point for PRFs of 500–1500 Hz. At 2000 and 2500 Hz, 600 pulses per point treatments resulted in the targeted volume showing hyperechoic shading outlining the treatment volume instead of the characteristic hypoechoic shading, which suggested some tissue structures may have remained intact after treatment. At the highest dose of 800 pulses, both PRFs generated a hypoechoic shading of the targeted volume, indicating effective ablation.

Assessed histologically, lesion analysis showed that for PRFs of 500–1500 Hz, partial ablation was observed at 400 pulses per point (Figure 8), while complete tissue destruction was observed for the 600 pulses per point group (Figure 9). Conversely, PRFs of 2000–2500 Hz did not achieve a complete ablation at 600 pulses per point. At the lowest dose of 400 pulses per point (Figure 8), lesions still had many cell nuclei and collagenic structures intact in the treated region, yielding a percent ablation of 52.7±4.9% at 500 Hz, 64.0±15.2% at 1000 Hz, and 62.4±9.5% at 1500 Hz. At 600 pulses per point (Figure 9), there were few remaining cell nuclei and a near complete removal of the collagen matrix in the treated region for PRFs of 500–1500 Hz, resulting in >90% ablations for each PRF (500 Hz: 96.4±3.1%, 1000 Hz: 94.9±1.0%, 1500 Hz: 92.9±2.2%). For the highest PRFs tested (2000–2500 Hz), there were partially damaged and intact collagen fibers with clumps of liver cells remaining in the treatment volume at 600 pulses per point (82.6±8.3% at 2000 Hz and 85.3±6.3% at 2500 Hz). As the dose was increased to 800 pulses per point, lesion analysis showed that a PRF of 2000 Hz generated a near complete ablation of the target volume (93.1±8.0%). At 2500 Hz, islands of liver cells remained throughout the treatment volume, often near intact collagen fibers or along the periphery, similar in number to the 600 pulses per point group (84.6±11.5%). At both PRFs, some partially disintegrated connective structures remained after treatment (Figure 10), but were isolated in regions that had undergone liquefactive necrosis at 2000 Hz. Additionally, in some cases, the hepatic veins were present and undamaged, even at the highest dose and PRF, indicating that these higher pulsing rates may still keep the tissue selectivity characteristic to histotripsy therapies. For all treatments, distinct treatment boundaries were observed, sharply delineating treated versus non-treated tissue, with a transition zone typically <100 μm. No off-target damage was observed pre-focally, and only minor ablative damage (1–2 mm extension) was seen in the post-focal direction in some samples, likely due to the increase in the length of the lesion at high PRFs as observed in the RBC lesion characterization. No visual differences in off-target damage were noted between pressure or PRF conditions.

Figure 8. Trichrome Staining of Lesions @ 400 pulses/point.

Figure 8.

Open in a new tab

Histological evaluation of ex vivo porcine liver ablations using H&E staining assessed the damage to liver cells and connective tissues. For PRFs 500–1500 Hz, a clear circular ablation volume can be seen with intact cellular structures or fibers in the targeted area. Images are marked to highlight remaining cells (*) and intact or partially damage collagen fibers (↑).

Figure 9. Trichrome Staining of Lesions @ 600 pulses/point.

Figure 9.

Open in a new tab

Histological evaluation of ex vivo porcine liver ablations using trichrome staining assessed the damage to liver cells and connective tissues. For PRFs 500–1500 Hz, a clear circular ablation volume can be seen with no remaining intact cellular structures or fibers in the targeted area. At 2000–2500 Hz PRFs there were some remaining cells (*) intact bundled by partially damage collagen fibers (↑).

Figure 10. Trichrome Staining of Lesions @ 800 pulses/point.

Figure 10.

Open in a new tab

For PRFs 2000 and 2500 Hz, trichrome staining showed some partially damaged collagen fibers (←) and cellular structures remaining (*) in the targeted region after 800 pulses/point. Increasing the dose to 800 pulses/point resulted in a complete ablation at 2000 Hz. Some detached liver cells (*) and broken collagen fibers (←) persisted in the field at 2500 Hz.

Across PRFs, samples were stratified into low (400 pulses/point for 500, 1000, and 1500 Hz; 600 pulses per point for 2000 and 2500 Hz) and high (600 pulses/point for 500, 1000, and 1500 Hz; 800 pulses per point for 2000 and 2500 Hz) dose groups for statistical analyses. Results revealed that there was a significant effect of dose [F(1,20) = 59.14, p < 0.0001] and a near significant effect of PRF [F(4,20) = 2.651, p = 0.0634] on the percentage of ablation on histopathology, as well as a significant interaction between PRF and dose on percent ablation [F(4,20) = 7.073, p = 0.0010]. Post-hoc tests revealed significantly lower percent ablation for samples treated at 500 Hz and the low dose as compared to low dose samples at 2000 and 2500 Hz as well as high dose samples at all PRFs (p < 0.05 for all comparisons). Additionally, samples treated at 1000 Hz and the low dose had significantly lower percent ablation as compared to samples treated at the high dose at all PRFs except 2500 Hz (p < 0.05). Lastly, samples treated at 1500 Hz and the low dose had significantly lower percent ablation as compared to samples treated at the high dose at all PRFs except 2500 Hz (p < 0.05).

No significant effect of PRF [F(4,20) = 1.391, p = 0.2730] or dose [F(1,20) = 0.9086, p = 0.3519] on ablation diameter was measured. There was also no significant interaction between PRF and dose [F(4,20) = 0.5679, p = 0.6889] on ablation diameter. When compared against the targeted ablation diameter of 1.5 cm, the average achieved ablation diameter was significantly smaller (p < 0.0001), with a mean 0.098 cm difference between achieved and targeted ablation diameters, highlighting the precision of the histotripsy treatment, even at high PRFs.

Discussion

In this study, single-cycle histotripsy was investigated for the rapid ablation of porcine liver using high PRF pulsing methods. Prior research in multi-cycle histotripsy has shown that higher PRFs lead to less efficient tissue ablation [15, 31], but recent evidence by our group has shown that single-cycle histotripsy does not experience this decrease in efficiency as long as the bubble cloud is formed with a high density of bubbles from the initial bubble cloud [18]. The results from this prior work also showed that the pressure required to generate a bubble cloud at 500 Hz (15.5±3.15 MPa) decreased below the pressure for generating consistent bubble clouds at low PRFs ≤1 Hz. In this work, cavitation clouds were similarly generated in porcine liver at pressure ranges of 14–17 MPa, which was significantly below the intrinsic threshold pressure required to generate cavitation at 1 Hz, confirming that the results from our prior studies held up to 2500 Hz. The pressure required for the onset of cavitation clouds was consistent across the PRFs tested in this study, suggesting that further decreases in threshold are not expected at higher PRFs beyond those used in this study. Based on these results, three pressure levels were selected for ablation experiments in agarose RBC gel phantoms, covering a range of pressures that corresponded to the lowest pressure range for generating a bubble cloud at high PRF (15 MPa), an intermediate pressure close to what is required to generate a single bubble at low PRF (24 MPa), and a high pressure case where a full bubble cloud could be generated at both low and high PRFs (42 MPa). Results from these RBC ablation experiments demonstrated no significant differences in ablation efficiency as the PRF was increased from 500–2500 Hz, supporting our hypothesis that single-cycle histotripsy can generate efficient tissue ablation at high PRFs. However, treatment precision was observed to decrease at higher PRF, particularly at the highest pressure levels, due to bubbles being formed beyond the central region of the cloud and no longer generating the precise and predictable ellipsoidal ablation area characteristic of single-cycle histotripsy at low PRFs (≤1 Hz). Despite the decrease in focal precision for the single treatment point in our RBC experiments, volumetric tissue ablation experiments demonstrated that single-cycle histotripsy produced effective tissue ablation covering a predetermined volume for all tested PRFs, with near complete or complete ablation observed for all samples after 600–800 applied pulses, which is in line with the required dose required for complete ablation using similar pulsing parameters (250–1000 pulses/point) [32]. The regions of partially ablated tissue observed in some samples at the lower doses (400 pulses for 500–1500 Hz PRF, 600 pulses for 2000–2500 Hz PRF, and 800 pulses for 2500 Hz) may suggest a small decrease in ablation effectiveness with increasing PRF. This may also have resulted from differences in lesion size at the targeted dose for each PRF. As seen in the RBC lesion characterization, the lesion sizes differed for a given pulse number, indicating that the spacing between points may not have provided enough coverage to fully break down the tissue before moving to the next treatment point. However, these results demonstrated that higher PRFs are capable of achieving complete ablation in much faster treatment times while maintaining adequate precision for volumetric lesion generation (i.e., close agreement between planned and measured ablation area). Overall, results from this study support our hypothesis that high PRF single-cycle histotripsy can generate rapid and efficient tissue ablation, which may be used to reduce treatment times for clinical histotripsy procedures in the future.

In previous work published by our team, it was shown that the cavitation cloud threshold decreased as a function of PRF, suggesting that single-cycle histotripsy ablation could be generated at pressures below the histotripsy intrinsic threshold of ~25–28 MPa [18]. Results from that study showed that the bubble cloud collapsed prior to the subsequent pulse, even at high PRFs, but cavitation was still generated due to the restimulation of residual nuclei which had not yet dissolved out of the field. Consequently, when cavitation occurred even at low probabilities at sub-intrinsic threshold pressures, the residual nuclei were restimulated until a bubble cloud was formed. This was further supported by the results from this work. In the bubble cloud generation testing, a robust bubble cloud was formed in ex vivo liver at an average pressure of 15.7±3.24 MPa, with no statistical difference between cloud generation pressures across PRFs of 500–2500 Hz, suggesting that further increases in PRF may not experience a decrease in the pressure required to generate a bubble cloud. US imaging showed initial cavitation events would occur and maturate into a robust hyperechoic signal over the 3 second interval once a sufficiently high pressure level was reached. It should be noted that cavitation events may have occurred that were not detectable on US imaging at pressures below those required for generating the bubble cloud. Therefore, the pressures reported here do not represent the lowest pressure at which cavitation occurred, but the pressure at which a robust hyperechoic signature representing the bubble cloud was observed. However, the findings from our tissue experiments were in good alignment with our in vitro RBC analysis, which showed a bubble cloud forming from initial cavitation events before growing into a full cavitation bubble cloud in this same pressure region (data not shown). Future work should explore the mechanisms underlying histotripsy bubble cloud formation at sub-intrinsic threshold pressure levels when using single-cycle histotripsy to better understand and predict bubble cloud formation under these conditions.

Although RBC ablation results showed a decrease in focal precision compared to 1 Hz conditions, results from volumetric tissue ablation studies demonstrate that this approach can achieve sufficient precision to closely match the planned ablation volumes required for histotripsy applications, such as the treatment of large tumors (>1 cm3 in diameter) where high precision ablation of each focal point is not required for the entirety of the treatment. Additional improvements in scanning patterns, step size, and robotic assistance could lead to an even further decrease in treatment time as well as improvements in treatment precision; however, even with the limitations in this study, high PRF treatments at 2000 Hz achieved >2.5x faster ablation (~6min) than 500 Hz PRF tests (~17min) at the PRF-specific treatment dose yielding >90% ablation (800 pulses/point and 600 pulses/point, respectively). Since 500 Hz is close to the reported values for most preclinical and clinical histotripsy tumor ablation studies [26, 33, 34], this result suggests that increasing the PRF can result in significant improvements for future volumetric ablation procedures with histotripsy. Overall, analysis of the histological results showed that all PRFs generated tightly bounded spherical lesions with minimal damage reported in the margins outside of the targeted volume. This finding is promising and suggests there was no observed risks of additional off-target cavitation when increasing the PRF from 500 Hz to 2500 Hz. Future studies should further explore the role of PRF and other pulsing parameters on the potential for generating pre-focal or off-target cavitation in vivo, particularly for applications near critical structures in which higher precision may be required.

The results from this work support our hypothesis that high PRFs can be used for volumetric tissue ablation at pressures well below the pressure required to generate bubble clouds at low PRFs (<1 Hz). The ability to increase the pulsing rate and effectively decrease treatment times while maintaining treatment precision is a key finding of this work. However, additional studies are warranted in order to further explore the safety and efficacy of ultra-high PRF histotripsy for specific clinical applications. For instance, while the results of this study showed examples of tissue selective ablation around vessels for all PRFs, future work should conduct a more robust analysis of the effects of PRF on tissue selectively near vessels, ducts, and other critical structures. Additional work is also warranted to explore the safety profile of high PRF single-cycle histotripsy treatments, particularly with respect to potential thermal effects. Future work should investigate the thermal effects induced by single-cycle histotripsy as a function of PRF and pressure level for different clinical applications. As thermal effects were not investigated in this study, and the volumetric ablations were performed at room temperature in a water bath, additional studies are warranted to characterize both thermal effects and ablation efficiency in vivo with perfusion. Likewise, while the reduction in the amplitude required to generate a bubble cloud reduces with increasing PRF up to a point, thereby reducing the expected thermal effects induced during treatment, the consistent thresholds observed in this study suggest there will be a limitation to how high the PRF can be raised before generating thermal effects at the target or on overlying tissue. This effect may also be application specific as some applications have a higher risk of thermal injury (i.e., transcostal vs. transabdominal histotripsy). While no visible signs of thermal degradation were noted on visual inspection of the tissue after treatment or on histological analysis, it should be noted that this was not a focus of this study, and it is expected that there will be an increased risk of inducing thermal effects with increasing PRF. Future work should explore this potential heating in different tissue types and setups. Since these effects will be application specific, it is likely that there will be an upper limit in the PRF that can be used for certain histotripsy applications.

Conclusion

The results from this study show that high PRFs are capable of generating precise volumetric ablations of soft tissue using single-cycle histotripsy. In bubble cloud generation testing, each PRF consistently generated bubble clouds below the reported intrinsic threshold. Furthermore, there was no statistical difference in the cloud formation pressure between PRFs, indicating that it is not expected to decrease with higher pulsing rates. In vitro analysis in 1% agarose red blood cell gel phantoms showed no difference in ablation efficiency from 500 – 2500 Hz PRF. Furthermore, these results showed that the pressure required to generate a focal ablation at 1 Hz was higher than for the 500–2500 Hz PRF treatments, with a decrease in focal precision at the high pressure level (42 MPa). Consequently, using the same pressure at which a single bubble is generated at 1 Hz PRF, ex vivo volumetric ablation testing showed precise destruction of the targeted tissue with minimal changes in efficiency across PRFs. Even for 2000–2500 Hz tests, volumetric ablations were limited to the target volume with sharply demarcated boundaries between treated and untreated tissue. The findings and conclusions from this work provide further evidence for the use of high PRF pulsing due to the ability to treat faster and at lower pressures.

Acknowledgements

This work was supported by grants from the NIH (R01CA262474, 1R21HD109631-01, 1R21EB035278-01) and the Focused Ultrasound Foundation. The authors of this study would like to thank the Virginia Tech Meat Center for supplying the tissue and blood used in this study. Furthermore, we would also like to acknowledge the Virginia Tech Animal Laboratory Service (ViTALS) for processing the treated tissue samples for histology.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest Statement

EV is a co-founder and stockholder of Sound Blade Medical. LR has an ongoing consulting relationship with Theraclion. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Ethics Statement

Porcine blood and liver samples used in this study were collected from local abattoirs (Virginia Tech Meat Center, Blacksburg, VA, USA; Firehouse Farms, Floyd, VA, USA) after scheduled harvests for retail sale. All animal-derived specimens were used in accordance with 3R research principles (Replacement, Reduction, and Refinement) under an approved Virginia Tech Institutional Biosafety Committee (IBC) protocol (#21-033).

References

  • [1].Xu Z, Khokhlova TD, Cho CS, and Khokhlova VA, “Histotripsy: A Method for Mechanical Tissue Ablation with Ultrasound,” (in eng), Annu Rev Biomed Eng, vol. 26, no. 1, pp. 141–167, Jul 2024, doi: 10.1146/annurev-bioeng-073123-022334. [DOI] [Google Scholar]
  • [2].Bader K, Vlaisavljevich E, and Maxwell A, “For Whom the Bubble Grows: Physical Principles of Bubble Nucleation and Dynamics in Hisotripsy Ultrasound Therapy,” Ultrasound Med. Biol, vol. 45, no. 5, pp. 1056–1080, 2019, doi: 10.1016/j.ultrasmedbio.2018.10.035 [DOI] [Google Scholar]
  • [3].Mancia L, Vlaisavljevich E, Xu Z, and Johnsen E, “Predicting Tissue Susceptibility to Mechanical Cavitation Damage in Therapeutic Ultrasound,” (in eng), Ultrasound Med Biol, vol. 43, no. 7, pp. 1421–1440, Jul 2017, doi: 10.1016/j.ultrasmedbio.2017.02.020. [DOI] [Google Scholar]
  • [4].Vlaisavljevich E, Maxwell A, Mancia L, Johnsen E, Cain C, and Xu Z, “Visualizing the Histotripsy Process: Bubble Cloud-Cancer Cell Interactions in a Tissue-Mimicking Environment,” Ultrasound Med. Biol, vol. 42, no. 10, pp. 2466–2477, 2016, doi: 10.1016/j.ultrasmedbio.2016.05.018. Epub 2016 Jul 9. [DOI] [Google Scholar]
  • [5].Lundt J et al. , “Coalescence of residual histotripsy cavitation nuclei using low-gain regions of the therapy beam during electronic focal steering,” (in eng), Phys Med Biol, vol. 63, no. 22, p. 225010, Nov 12 2018, doi: 10.1088/1361-6560/aaeaf3. [DOI] [Google Scholar]
  • [6].Lundt J, Allen S, Shi J, Hall TL, Cain CA, and Xu Z, “Noninvasive, Rapid Ablation of Tissue Volume Using Histotripsy,” Ultrasound Med. Biol, vol. 43, no. 12, pp. 2834–2847, 2017, doi: 10.1016/j.ultrasmedbio.2017.08.006. [DOI] [Google Scholar]
  • [7].HistoSonics, “FDA Awards HistoSonics Clearance of its First-of-a-Kind Edison® Histotripsy System,” in Company Ramping Up Launch Plans of its “Breakthrough Device” Designated Platform, ed. https://histosonics.com/news/fda-awards-histosonics-clearance-of-its-first-of-a-kind-edison-histotripsy-system-2/:HistoSonics. [Google Scholar]
  • [8].H. Inc. “The HistoSonics Edison™ System for Treatment of Pancreatic Adenocarcinoma Using Histotripsy (GANNON).” ClinicalTrials.gov identifier: NCT06282809. Updated November 14, 2025. Accessed December 16, 2025. https://clinicaltrials.gov/study/NCT06282809. [Google Scholar]
  • [9].H. Inc. “The HistoSonics Edison™ System for Treatment of Primary Solid Renal Tumors Using Histotripsy (#HOPE4KIDNEY) (#HOPE4KIDNEY).” ClinicalTrials.gov identifier: NCT05820087. Updated November 14, 2025. Accessed December 16, 2025. https://www.clinicaltrials.gov/study/NCT05820087. [Google Scholar]
  • [10].Ruger LN et al. , “Characterizing the Ablative Effects of Histotripsy for Osteosarcoma: In Vivo Study in Dogs,” Cancers, vol. 15, no. 3, p. 741, 2023, doi: 10.3390/cancers15030741. [DOI] [Google Scholar]
  • [11].Lu N et al. , “Transcranial Magnetic Resonance-Guided Histotripsy for Brain Surgery: Pre-clinical Investigation,” Ultrasound Med. Biol, vol. 48, no. 1, pp. 98–110, 2022, doi: 10.1016/j.ultrasmedbio.2021.09.008. [DOI] [Google Scholar]
  • [12].Simon A et al. , “Histotripsy for the Treatment of Uterine Leiomyomas: A Feasibility Study in Ex Vivo Uterine Fibroids,” (in eng), Ultrasound Med Biol, vol. 48, no. 8, pp. 1652–1662, Aug 2022, doi: 10.1016/j.ultrasmedbio.2022.04.214. [DOI] [Google Scholar]
  • [13].Khokhlova VA et al. , “Initial Assessment of Boiling Histotripsy for Mechanical Ablation of Ex Vivo Human Prostate Tissue,” (in eng), Ultrasound Med Biol, vol. 49, no. 1, pp. 62–71, Jan 2023, doi: 10.1016/j.ultrasmedbio.2022.07.014. [DOI] [Google Scholar]
  • [14].Hendley SA et al. , “(More than) doubling down: Effective fibrinolysis at a reduced rt-PA dose for catheter-directed thrombolysis combined with histotripsy,” (in eng), PLoS One, vol. 17, no. 1, p. e0261567, 2022, doi: 10.1371/journal.pone.0261567. [DOI] [Google Scholar]
  • [15].Wang TY, Xu Z, Hall TL, Fowlkes JB, and Cain CA, “An efficient treatment strategy for histotripsy by removing cavitation memory,” (in eng), Ultrasound Med Biol, vol. 38, no. 5, pp. 753–66, May 2012, doi: 10.1016/j.ultrasmedbio.2012.01.013. [DOI] [Google Scholar]
  • [16].Duryea AP, Cain CA, Tamaddoni HA, Roberts WW, and Hall TL, “Removal of residual nuclei following a cavitation event using low-amplitude ultrasound,” (in eng), IEEE Trans Ultrason Ferroelectr Freq Control, vol. 61, no. 10, pp. 1619–26, Oct 2014, doi: 10.1109/tuffc.2014.006316. [DOI] [Google Scholar]
  • [17].Duryea AP, Cain CA, Roberts WW, and Hall TL, “Removal of residual cavitation nuclei to enhance histotripsy fractionation of soft tissue,” (in eng), IEEE Trans Ultrason Ferroelectr Freq Control, vol. 62, no. 12, pp. 2068–78, Dec 2015, doi: 10.1109/tuffc.2015.007202. [DOI] [Google Scholar]
  • [18].Simon A, Edsall C, Maxwell A, and Vlaisavljevich E, “Effects of pulse repetition frequency on bubble cloud characteristics and ablation in single-cycle histotripsy,” Phys. Med. Biol, 2023, doi: 10.1088/1361-6560/ad11a1. [DOI] [Google Scholar]
  • [19].Vlaisavljevich E, Gerhardson T, Hall TL, and Xu Z, “Effects of f-number on the histotripsy intrinsic threshold and cavitation bubble cloud behavior,” Phys. Med. Biol, vol. 62, no. 4, pp. 1269–1290, 2017. [Google Scholar]
  • [20].Vlaisavljevich E et al. , “Effects of tissue stiffness, ultrasound frequency, and pressure on histotripsy-induced cavitation bubble behavior,” Phys. Med. Biol, vol. 60, no. 6, pp. 2271–2292, 2015, doi: 10.1088/0031-9155/60/6/2271. [DOI] [Google Scholar]
  • [21].Lin KW et al. , “Histotripsy beyond the instrinsic cavitation threshold using very short ultrasound pulses: microtripsy,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 61, no. 2, pp. 251–265, 2014, doi: 10.1109/TUFFC.2014.6722611. [DOI] [Google Scholar]
  • [22].Vlaisavljevich E et al. , “Effects of ultrasound frequency and tissue stiffness on the histotripsy intrinsic threshold for cavitation,” Ultrasound Med. Biol., vol. 41, no. 6, pp. 1651–67, Jun 2015, doi: 10.1016/j.ultrasmedbio.2015.01.028. [DOI] [Google Scholar]
  • [23].Ruger L et al. , “Mechanical High-Intensity Focused Ultrasound (Histotripsy) in Dogs with Spontaneously Occurring Soft Tissue Sarcomas,” IEEE Trans. Biomed. Eng, 2022, doi: 10.1109/TBME.2022.3201709. [DOI] [Google Scholar]
  • [24].Edsall C, Ham E, Holmes H, Hall TL, and Vlaisavljevich E, “Effects of frequency on bubble-cloud behavior and ablation efficiency in intrinsic threshold histotripsy,” Phys. Med. Biol, vol. 66, no. 22, 2021. [Google Scholar]
  • [25].Maxwell AD, Wang TY, Yuan L, Duryea AP, Xu Z, and Cain CA, “A tissue phantom for visualization and measurement of ultrasound-induced cavitation damage,” (in eng), Ultrasound Med Biol, vol. 36, no. 12, pp. 2132–43, Dec 2010, doi: 10.1016/j.ultrasmedbio.2010.08.023. [DOI] [Google Scholar]
  • [26].Swietlik JF et al. , “Noninvasive thyroid histotripsy treatment: proof of concept study in a porcine model,” (in eng), Int J Hyperthermia, vol. 38, no. 1, pp. 798–804, 2021, doi: 10.1080/02656736.2021.1922762. [DOI] [Google Scholar]
  • [27].Vlaisavljevich E et al. , “Image-Guided Non-Invasive Ultrasound Liver Ablation Using Histotripsy: Feasibility Study in an in vivo Porcine Model,” Ultrasound Med. Biol, vol. 39, no. 8, pp. 1398–1409, 2013, doi: 10.1016/j.ultrasmedbio.2013.02.005. [DOI] [Google Scholar]
  • [28].Vlaisavljevich E et al. , “Non-Invasive Ultrasound Liver Ablation Using Histotripsy: Chronic Study in an In Vivo Rodent Model,” Ultrasound Med. Biol., vol. 42, no. 8, pp. 1890–1902, 2016, doi: 10.1016/j.ultrasmedbio.2016.03.018. [DOI] [Google Scholar]
  • [29].Smolock AR et al. , “Robotically Assisted Sonic Therapy as a Noninvasive Nonthermal Ablation Modality: Proof of Concept in a Porcine Liver Model,” Radiology, vol. 287, no. 2, pp. 485–493, May 2018, doi: 10.1148/radiol.2018171544. [DOI] [Google Scholar]
  • [30].Kumar YN et al. , “A comparative study of histotripsy parameters for the treatment of fibrotic ex-vivo human benign prostatic hyperplasia tissue,” (in eng), Sci Rep, vol. 14, no. 1, p. 20365, Sep 2 2024, doi: 10.1038/s41598-024-71163-2. [DOI] [Google Scholar]
  • [31].Duryea AP, Tamaddoni HA, Cain CA, Roberts WW, and Hall TL, “Removal of residual nuclei following a cavitation event: a parametric study,” (in eng), IEEE Trans Ultrason Ferroelectr Freq Control, vol. 62, no. 9, pp. 1605–14, Sep 2015, doi: 10.1109/tuffc.2014.006601. [DOI] [Google Scholar]
  • [32].Worlikar T et al. , “Development of Histotripsy as a Local-Regional Liver Cancer Therapy: Preclinical to Clinical Translation,” (in eng), Radiol Imaging Cancer, vol. 7, no. 4, p. e240403, Jul 2025, doi: 10.1148/rycan.240403. [DOI] [Google Scholar]
  • [33].Gannon J et al. , “Ultrasound-guided noninvasive pancreas ablation using histotripsy: feasibility study in an in vivo porcine model,” (in eng), Int J Hyperthermia, vol. 40, no. 1, p. 2247187, 2023, doi: 10.1080/02656736.2023.2247187. [DOI] [Google Scholar]
  • [34].Worlikar T et al. , “Insights from in vivo preclinical cancer studies with histotripsy,” Int. J. Hyperthermia, vol. 41, no. 1, p. 2297650, 2024/12/31 2024, doi: 10.1080/02656736.2023.2297650. [DOI] [Google Scholar]

RESOURCES