Dual-frequency boiling histotripsy in an ex vivo bovine tendinopathy model

Molly Smallcomb; Julianna C Simon

doi:10.1121/10.0019630

. 2023 Jun 6;153(6):3182–3191. doi: 10.1121/10.0019630

Dual-frequency boiling histotripsy in an ex vivo bovine tendinopathy model

Molly Smallcomb ^1,^a),^✉, Julianna C Simon ¹

PMCID: PMC10247224 PMID: 37279386

Abstract

Histotripsy fractionates most soft tissues; however, healthy tendons have shown resistance to histotripsy fractionation. Prior work has shown that pre-heating tendons increases susceptibility to histotripsy fractionation; combining multiple driving frequencies may also allow successful fractionation of tendons. Here, we evaluate single- and dual-frequency histotripsy in four healthy and eight tendinopathic ex vivo bovine tendons. First, we evaluated single-frequency (1.07, 1.5, and 3.68 MHz) and dual-frequency (1.07 and 1.5 MHz or 1.5 and 3.68 MHz) bubble dynamics with high-speed photography in a tissue-mimicking phantom. Then, tendons were treated with histotripsy. Cavitation activity was monitored with a passive cavitation detector (PCD) and targeted areas were evaluated grossly and histologically. Results in tendinopathic tendons showed 1.5 MHz or 3.68 MHz single-frequency exposure caused focal disruption, whereas 1.5 and 3.68 MHz dual-frequency exposures caused fractionated holes; all treatments caused some thermal denaturation. Exposure to 1.07 MHz alone or combined with 1.5 MHz did not show fractionation in tendinopathic tendons. In healthy tendons, only thermal necrosis was observed for all tested exposures. PCD showed some differences in cavitation activity in tendinopathic tendons but did not predict successful fractionation. These results suggest that full histotripsy fractionation is possible using dual-frequency exposures in tendinopathic tendons.

I. INTRODUCTION

High intensity focused ultrasound (HIFU) is a non-invasive acoustic therapy that can create thermal and/or mechanical bioeffects within a well-defined focal volume (ter Haar, 2007). Histotripsy is a subset of HIFU that mechanically fractionates most soft tissues through the creation, oscillation, and collapse of bubbles (Parsons et al., 2006; Khokhlova et al., 2014); however, highly collagenous tissues like tendon have shown resistance to histotripsy fractionation. Previous work shows that full tendon fractionation only occurs after pre-heating tendons to 90 °C; however, the extreme heating causes complete collagen hydrolysis before the histotripsy exposure (Vlaisavljevich et al., 2015a). Recently, we showed that a small parameter space exists where microdamage in the form of fiber separation and fraying occurs in healthy, ex vivo rat tendons (Khandare et al., 2021; Smallcomb et al., 2021); however, we have yet to achieve full histotripsy fractionation in healthy or tendinopathic tendons. In this work, we evaluate whether dual frequency histotripsy can cause full mechanical fractionation in healthy and tendinopathic bovine tendons.

Tendon injuries, or tendinopathies, account for 30% of the 102 billion musculoskeletal injuries occurring annually in the United States (Kaux et al., 2011). Therapies like dry needling (DN) and extracorporeal shock wave therapy (ESWT) can trigger a healing response in chronic tendinopathies by inducing microdamage, which can increase perfusion and release healing factors (APTA, 2013; Notarnicola and Moretti, 2012). However, mixed success rates arise from these procedures likely due to off-target effects, lack of dose standardization, and inconsistencies in user variability and image guidance. We recently showed that HIFU-induced mild mechanical microdamage, or partial histotripsy, performed as well or better than the conventional DN in an in vivo murine tendinopathy model (publication in review). While the mild microdamage might be ideal for treating tendinopathies, it remains unknown whether full mechanical fractionation by histotripsy is achievable in tendinopathic tendons.

It has been theorized that bubble expansion in histotripsy and other therapies is impeded by the high ultimate stress of connective tissues like tendon (Allen and Silverman, 2000; Vlaisavljevich et al., 2014). Moreover, the high ultimate fractional strain found in elastic tissues contributes to erosion resistance as it allows the tissue to withstand large strain without rupture (Vlaisavljevich et al., 2014; Simon et al., 2015). Thus, techniques that can increase the ultrasound energy and/or maximize bubble expansion may allow for successful histotripsy fractionation of these collagenous tissues. Previous work has shown that combining two frequencies allows for tailoring of bubble expansion (Lin et al., 2014; Vlaisavljevich et al., 2015b) and creating denser bubble clouds in histotripsy therapy (Lin et al., 2014). Furthermore, the dual-frequency approach has been shown to increase erosion due to larger shock scattering and stronger inertial cavitation (Zhou and Lei, 2020). For example, Lin et al. (2014) applied a low-frequency “pump” pulse to help a sub-threshold high-frequency “probe” pulse exceed the intrinsic cavitation threshold in highly attenuative tissues. Similarly, Vlaisavljevich et al. (2015b) used a dual-frequency transducer to theoretically and experimentally demonstrate that adjustment of pressure ratios allows for bubble size to be manipulated between the two single-frequency bubble sizes. Increasing the frequency (1–3 MHz) and pulse duration (3–20 ms) in single-frequency histotripsy has been shown to initiate boiling bubbles on the order of millimeters in diameter in boiling histotripsy (Canney et al., 2010; Khokhlova et al., 2014). Taken together, this prior work suggests that dual-frequency boiling histotripsy may allow for tuning of bubble dynamics to allow for fractionation of highly collagenous tissues such as tendon.

This study evaluated single- and dual-frequency histotripsy regimes in healthy and tendinopathic ex vivo bovine tendons. First, bubble cloud activity for dual-frequency histotripsy was monitored in a tissue-mimicking phantom using high-speed photography. Then, tendinopathy was induced in some tendons by injecting collagenase, an enzyme that breaks the triple helix bonds of collagen. Single- and dual-frequency histotripsy was compared through inertial cavitation activity and gross and histological analysis between healthy and tendinopathic tendons. It was hypothesized that the dual-frequency approach would result in higher inertial cavitation in tendinopathic tendons compared to single-frequency exposures or healthy tendons and that this higher cavitation activity would cause fractionation in tendinopathic tendons.

II. METHODS

A. Dual frequency arrangement

A custom-designed fixture was built to hold two HIFU transducers with the geometric foci oppositely and coaxially aligned [Fig. 1(a)]. Dual-frequency histotripsy was applied with an f# = 1 (geometric focus = 51.74 mm) HIFU transducer (H-102, Sonic Concepts, Bothell, WA) driven at the fundamental frequency, 1.07 MHz, or 3rd harmonic frequency, 3.68 MHz, and an f# = 0.7 (geometric focus = 28.18 mm) HIFU transducer (H-234 with modified 42-mm central opening, Sonic Concepts, Bothell, WA) driven at 1.5 MHz. Two waveform generators (33600 A Series and 33500B Series, Keysight, Santa Rosa, CA) were synced to produce two, 10-ms boiling histotripsy pulses, applied either simultaneously (10-ms total pulse duration) or with the middle 5 ms (5–10 ms) overlapping [15-ms total pulse durations; Fig. 1(b)]. Five millisecond delays were based upon the volumetric rate of heat generation to reach 100 °C in tendon using shocks from the 1.5 MHz transducer (Duck, 2013; Szabo, 2016). These pulses were repeated at 1 Hz for a total treatment time of 60 s.

For dual-frequency exposures, the 1.07/3.68 MHz transducer was connected to a Class A radio frequency power amplifier (60 dB, A500, Electronic Navigation Industries, Rochester, NY) and the 1.5 MHz transducer was connected to a Class AB radio frequency power amplifier (55 dB, 1040 L, Electronics and Innovations, Ltd., Rochester, NY). For single-frequency histotripsy, all transducers were connected to the Class A amplifier (60 dB, A500) to achieve equal acoustic output and higher-pressure output than the Class AB amplifier (55 dB, 1040 L). Focal pressures of the transducer-amplifier combinations were measured individually using a fiber optic probe (FOPH) (HFO-690, ONDA, Sunnyvale, CA) in free-field degassed, de-ionized water. Single-frequency peak pressures for 1.07 and 3.68 MHz were measured with the Class A amplifier as p₊ = 75 MPa, p₋ = 15 MPa, and p₊ = 78 MPa, p₋ = 21 MPa, respectively [Figs. 2(a) and 2(b)]. Single-frequency peak pressures for 1.5 MHz were measured with the Class AB amplifier as p₊ = 49 MPa, p₋ = 19 MPa [Fig. 2(c)]. Dual-frequency waveforms were estimated by superimposing these waveforms with respective time delays [Figs. 2(d) and 2(e)]. Of note, superposition of waveforms is only valid under linear conditions, so the resulting dual-frequency waveforms were only estimations of the combined pressure field. Peak negative pressures exceeding ∼25 MPa caused cavitation damage to the fiber tip, so HIFU-beam Simulator (Yuldashev et al., 2021) was utilized to predict single-frequency focal pressures for the 1.5 MHz transducer with the Class A amplifier pairing (p₊ = 136 MPa, p₋ = 36 MPa).

FIG. 2. — Maximum measurable waveforms for (a) 1.07 MHz, (b) 1.5 MHz, and (c) 3.68 MHz. Superimposed waveforms of (d) 1.07 MHz and 1.5 MHz and (e) 3.68 MHz and 1.5 MHz simultaneous dual-frequency exposures.

To confirm alignment between the transducers' foci and evaluate bubble dynamics from the dual-frequency approach, bubble activity was imaged in tissue-mimicking hydrogels. There is currently no standard to model fibrous tendon in vitro; therefore, a common acoustic polyacrylamide hydrogel (PAA) was used (Lafon et al., 2005). High-speed photography of the second pulse was used to evaluate visual differences between pulsing schemes. Images of all treatment schemes were analyzed shortly after each transducer turned on (2 and 7 ms) and shortly after the first or both transducers turned off (10.5 ms) [Fig. 1(b)].

B. Sample preparation

Twelve deep digital flexor tendons were dissected from bovine limbs obtained from local abattoirs immediately after euthanasia. Tendons were stored in PBS and kept cold until treatment to lessen natural degradation. To mimic chronic tendinopathies, eight tendons were locally injected 3–6 times with a collagenase-fibrin solution and stored for >3 h to allow for degradation (Watts et al., 2012; Groth et al., 2017). The solution was made by first combining all dry ingredients—8 mg collagenase, 50 mg fibrinogen, and 8 mg thrombin—into a syringe mixed with 1 ml PBS, and then immediately injecting 0.1 ml per site to prevent gelation in the syringe. Changes to the fiber pattern after injection was monitored with B-mode ultrasound using a research ultrasound system (Vantage-128, Verasonics^®, Kirkland, WA) and L12-5 imaging transducer (Philips/ATL, Bothell, WA) to measure the degraded area in vitro. The other four tendons were unaltered and treated within 3 h after harvesting. Of the 12 tendons, 45 locations were targeted with single- and dual-frequency histotripsy for a sample size of 3 per pre-treatment and parameter set. Healthy tendons were not treated with single-frequency histotripsy as our previous research found no histotripsy-induced liquefaction in large animal ex vivo tendons (Smallcomb and Simon, 2019). All tendons were treated fully submerged in degassed (<6 mg/mL), de-ionized water.

C. Injury evaluation

Gross morphology was evaluated in five tendinopathic tendons and three healthy tendons. Tendons were sliced in the sagittal plane to visualize the focal location and propagation paths of the HIFU transducers. Targeted areas were evaluated for tissue liquefaction by gently drying with gauze and noting morphological changes including holes and discoloration. The entire treatment volumes in the remaining three tendinopathic tendons and one healthy tendon were flash-frozen in Tissue-Tek^® OCT compound (Sakura Finetek, Torrance, CA) and stored at −80 °C for histological analysis. Representative samples were sectioned at 7-μm thick at 0 mm and 2 mm from the center of the –6 dB focal volume with a cryostat (CM1950, Leica, Wetzlar, Germany) and alternate sections stained with hematoxylin and eosin (H&E) for analysis of cellular morphology and alpha nicotinamide adenine dinucleotide diaphorase (α-NADH-d) for analysis of enzymatic activity.

D. PCD signal processing

A focused PCD (Y-107, Sonic Concepts, Bothell, WA) with a bandwidth of 15 MHz and diameter of 14.7 mm designed to fit within the central opening of the 1.07/3.68 MHz HIFU transducer and coaxially aligned with the geometric focus (Fig. 1) was connected to a digital oscilloscope (InfiniiVision DSOX3034T, Keysight, Santa Rosa, CA) synced to the function generators. Due to the limiting block size of the oscilloscope digitization and high harmonics of the HIFU transducers, PCD waveforms were limited to capturing 2 ms increments at 32 MHz sampling frequency. To capture cavitation activity information within the entire pulse length (maximum 15 ms), the oscilloscope trigger was cumulatively delayed 2 ms up to 16 ms, after which the triggering system reset back to 0 ms.

Cavitation activity was processed from PCD signals similar to the protocol described by Li et al. (2014). Briefly, the single-sided spectral density was calculated from the raw data for all 60 pulses of the exposure. Then, peaks from the fundamental, harmonics, and sum and difference frequencies generated by the HIFU transducers were identified in the frequency spectrum; frequencies below 10 kHz were filtered out to remove transient electrical noise. Broadband energy between the harmonics, or between the full width half maximum (FWHM) of each peak, was integrated to find the total inertial cavitation energy of the signal.

Sustained cavitation activity was evaluated in two parts: across the treatment time in sections of 8 pulses (i.e., pulse numbers 1–8, 9–16, 17–24, 25–32, 33–40, 41–48, 49–56) and across the pulse duration in 2 ms increments (i.e., 0–2 ms, 2–4 ms, 4–6 ms, 6–8 ms, 8–10 ms, 10–12 ms, 12–14 ms, 14–16 ms); treatment times from pulses 57–60 were not included in the sustained treatment time analysis to ensure all eight pulsing sections included the same number of pulses and 2 ms increments. Sustained activity during the treatment time was quantified by averaging all pulses within the respective quartile section or 2-ms interval, and the mean and standard deviation was calculated for all independent targeted treatments with the same treatment group. Total cavitation activity was quantified by summing the cavitation activity across all pulses. Statistical analysis was performed using ANOVA (p < 0.05) to compare differences in cavitation for tendinopathic tendons exposed to single frequencies, and a paired t-test (p < 0.05) between tendinopathic and healthy tendons exposed to the same dual-frequency histotripsy parameters.

III. RESULTS

A. Bubble dynamics

Observations in PAA hydrogels exposed to 1.07 and 1.5 MHz simultaneously or with a delay in either order showed similar bubble cloud activity extending 3–4 mm × 9.5–10 mm at 2 ms into the second pulse (Fig. 3). However, when 1.5 MHz led with 1.07 MHz beginning after a 5-ms delay, the bubble cloud moved 2 mm closer to the 1.5 MHz transducer [Fig. 3(c)]. When the PAA gels were treated with 3.68 and 1.5 MHz simultaneously or with 1.5 MHz leading with 3.68 MHz delayed 5-ms, the bubble cloud was much smaller at 2 ms [1 mm × 4–5 mm; Fig. 3(d)] compared to the same pulsing scheme with the 1.07 MHz transducer [Fig. 3(d)]. Yet when PAA gels were treated with 3.68 MHz with 1.5 MHz beginning after a 5-ms delay, bubble activity was slightly larger (2 mm × 6 mm), which was substantially reduced after 10 ms when only the 1.5 MHz was active (1 mm × 2 mm) [Fig. 3(f)].

FIG. 3. — High-speed photographs captured during the second pulse of dual-frequency exposure with the 1.07/3.68 MHz transducer on the left and the 1.5 MHz transducer on the right. PAA gels were targeted with 10-ms pulses of 1.07 and 1.5 MHz (a) simultaneously, (b) 1.07 MHz with 1.5 MHz delayed by 5 ms, and (c) 1.5 MHz with 1.07 MHz delayed by 5 ms. PAA gels were also targeted with 10 ms pulses of 3.68 MHz and 1.5 MHz (d) simultaneously, (e) 1.5 MHz with 3.68 MHz delayed by 5 ms, and (f) 3.68 MHz with 1.5 MHz delayed by 5 ms. Of note, high speed photography windowing in (c) was moved slightly left toward the 1.07 MHz to capture more of the bubble cloud.

B. Tendon treatments

All tendinopathic tendons were successfully injected with collagenase; B-mode imaging found degenerated fibers extended at least 10 mm axially and 5 mm transversely to the transducers. Successful mechanical fractionation was confirmed in 12 of 45 targeted sites, which is summarized in Table I. Full fractionation was found only in tendinopathic tendons exposed to 1.5 and 3.68 MHz combinations (simultaneous and 1.5 MHz leading), which was accompanied by thermal necrosis. Partial fractionation, or multiple small holes, was found in tendinopathic tendons exposed to dual-frequency histotripsy of 3.68 MHz leading or single-frequency histotripsy of 1.5 MHz or 3.68 MHz, accompanied by some thermal necrosis. Tendinopathic tendons exposed to 1.07 MHz (single- or dual-frequency) as well as all healthy tendons did not display tissue fractionation grossly or histologically although some thermal necrosis was observed.

TABLE I.

Summary of tendinopathic samples displaying fractionation for each single- and dual-frequency histotripsy exposure.

Frequency (MHz)	Healthy		Tendinopathic
Frequency (MHz)	Partial fractionation	Full fractionation	Partial fractionation	Full fractionation
1.07	—	—	0/3	0/3
1.5	—	—	3/3	0/3
3.68	—	—	3/3	0/3
1.07 and 1.5	0/3	0/3	0/3	0/3
1.07, 1.5	0/3	0/3	0/3	0/3
1.5, 1.07	0/3	0/3	0/3	0/3
3.68 and 1.5	0/3	0/3	0/3	3/3
1.5, 3.68	0/3	0/3	0/3	2/3
3.68, 1.5	0/3	0/3	1/3	0/3

Open in a new tab

C. Single-frequency exposures

Single frequency treatments of tendinopathic tendons showed no evidence of gross disruption or discoloration for 1.07 MHz [Fig. 4(a)], although histology indicated spotty areas of disruption on H&E-stained samples. However, exposures to 1.5 MHz [Fig. 4(b)] and 3.68 MHz [Fig. 4(c)] displayed partial fractionation of tendon, or multiple small holes rather than one discrete hole. This observation was supported histologically with larger areas of frayed fibers observed in tendinopathic tendons treated with 3.68 MHz compared to tendons treated with 1.5 MHz [Fig. 4(c)]. In all treatments, tissue discoloration was observed grossly and histologically, as evidenced by lack of stain uptake on α-NADH-d samples (Fig. 4).

When comparing cavitation activity observed by PCD, 1.07 MHz showed a significantly higher (p = 0.006) total cavitation activity compared to 1.5 or 3.68 MHz. Cavitation activity throughout the treatment was not different for the 1.07 MHz (p = 0.999) or 3.68 MHz (p = 0.954) exposures, whereas cavitation activity decreased as the treatment progressed for the 1.5 MHz exposure (p = 0.001) [Figs. 5(a)–5(c)]. Cavitation activity was greatest at the beginning of the pulse for 1.5 MHz (p = 0.021) [Figs. 5(d) and 5(f)]; 1.07 MHz (p = 0.092) and 3.68 MHz (p = 0.228) showed no changes in cavitation activity throughout the pulse duration [Fig. 5(e)]. While more cavitation was observed for the 1.07 MHz, partial fractionation of tendinopathic tendons were only observed for the 1.5 MHz and 3.68 MHz exposures.

FIG. 5. — Cavitation activity throughout (a)–(c) the treatment time and (d)–(f) the pulse duration for tendinopathic tendons exposed to (a), (d) 1.07, (b), (e) 1.5, and (c), (f) 3.68 MHz. (b), (e) As the treatment and pulse duration progressed, cavitation activity significantly decreased in tendons exposed to 1.5 MHz. (a), (c), (d), (f) Conversely, cavitation activity was sustained throughout the treatment and pulse duration in tendons exposed to 1.07 and 3.68 MHz. Note that 1.07 MHz graphs are not to scale with 1.5 and 3.68 MHz. (An asterisk indicates significance of p < 0.05.)

D. 1.07 MHz and 1.5 MHz dual-frequency histotripsy

Healthy and tendinopathic tendons exposed to 1.07 and 1.5 MHz dual-frequency histotripsy (simultaneous or 1.07 MHz delayed) displayed no gross damage [Figs. 6(a) and 6(c)]. However, tendinopathic tendons exposed to 1.07 MHz with 1.5 MHz after a 5 ms delay showed some focal tissue discoloration with no visible hole [Fig. 6(b)]. This observation was supported histologically with more frayed and disorganized fibers in H&E slides in tendinopathic tendons [Fig. 6(d)—right] compared to healthy tendons [Fig. 6(d)—left]. Thermal necrosis was observed histologically for all exposures as indicated by lack of stain uptake on α-NADH-d samples.

FIG. 6. — (Color online) (a)–(c) Gross morphology and (d) representative histological slides of (left) healthy and (right) tendinopathic tendons after exposure to (a) 1.07 and 1.5 MHz simultaneous, [(b), (d)—left] 1.07 MHz with 1.5 MHz after a 5-ms delay, and [(c), (d)—right] 1.5 MHz with 1.07 MHz after a 5-ms delay. No disruption was noted in healthy tendons nor in tendinopathic tendons after (a) simultaneous frequency exposure and (c) 1.5 MHz leading. However, (b) tendinopathic tendons treated with 1.07 MHz leading displayed discoloration, circled in blue. [(d)—left] Healthy tendons displayed sparse morphological changes in H&E-stained samples, with respective areas in α-NADH-d-stained samples showing lack of stain uptake on edges around the disrupted fibers. [(d)—right] Tendinopathic tendons displayed more fiber disruption with no evidence of blue stain at the targeted area.

Compared to healthy tendons, tendinopathic tendons exposed to 1.07 and 1.5 MHz simultaneous showed differences in cavitation activity at pulses 17–24 [p = 0.096; Fig. 7(a)]. When transducers were turned off at 10-ms, tendinopathic tendons had more residual cavitation activity at 10–12 ms (p = 0.009), compared to healthy tendons [Fig. 7(d)]. Cavitation activity was not different between healthy and tendinopathic tendons throughout the treatment time or pulse duration for 1.07 MHz leading (p > 0.455, p > 0.274, respectively) or 1.5 MHz leading (p > 0.245, p > 0.067) [Figs. 7(b) and 7(c)], despite more fraying and disorganized fibers in tendinopathic tendons.

FIG. 7. — Cavitation activity throughout (a)–(c) the treatment time and (d)–(f) the pulse duration for healthy and tendinopathic tendons exposed to (a), (d) 1.07 and 1.5 MHz simultaneous, (b), (e) 1.07 MHz with 1.5 MHz on a 5-ms delay, and (c), (f) 1.5 MHz with 1.07 MHz on a 5-ms delay. (d) As the pulse duration progressed, cavitation activity significantly increased at 10 ms in tendinopathic tendons compared to healthy tendons exposed to simultaneous exposure. (a)–(c) No statistical difference in cavitation activity throughout the treatment time was found between tendinopathic and healthy tendons exposed to all frequencies, and (e), (f) no difference in cavitation activity throughout the pulse was found between tendinopathic and healthy tendons exposed to 1.07 MHz leading or 1.5 MHz leading. (An asterisk indicates significance of p < 0.05 between healthy and tendinopathic tendons.)

E. 3.68 MHz and 1.5 MHz dual-frequency histotripsy

After exposure to 3.68 and 1.5 MHz dual-frequency histotripsy (simultaneous or 1.5 MHz with 3.68 MHz after a 5-ms delay), tissue fractionation was observed in tendinopathic tendons [Figs. 8(a) and 8(b)]. Specifically, distinct 1.5–3 mm diameter holes were observed after blotting in 3/3 samples in the simultaneous exposure and in 2/3 samples treated with 1.5 MHz with 3.68 MHz after a 5 ms delay. These cases also exhibited disrupted fibrous structures on H&E histology [Figs. 8(a) and 8(b)]. For tendinopathic tendons treated with 3.68 MHz with 1.5 MHz after a 5 ms delay, a less-defined hole was observed after blotting in 1/3 samples; these samples also showed less fiber disruption on H&E histology [Fig. 8(c)]. In all tendons (healthy and tendinopathic), thermal injury was observed as evidenced by lack of stain uptake on α-NADH-d slides (Fig. 8).

FIG. 8. — (Color online) (Left) Gross morphology and (right) representative histological slides of (left) healthy and (center, right) tendinopathic tendons after exposure to (a) 1.5 and 3.68 MHz simultaneous, (b) 1.5 MHz with 3.68 MHz after a 5-ms delay, and (c) 3.68 MHz with 1.5 MHz after a 5-ms delay. [(a)–(c), left] Discoloration indicative of thermal damage occurred in the pathway of the 3.68 MHz (right of images). [(a)–(c), center] Holes were present after exposures in tendinopathic tendons, circled in blue, but included thermal discoloration on the edge near to the 3.68 MHz transducer (right of images). [(a), (b), right] Simultaneous and 1.5 MHz leading tendons displayed frayed fibers, circled in blue, with lack of stain uptake in α-NADH-d slides. [(c), right] 3.68 MHz leading tendons showed similar lack of stain uptake in α-NADH-d slides but less fiber fraying in H&E slides.

Cavitation activity between healthy and tendinopathic tendons was not significantly different throughout the treatment time for 3.68 and 1.5 MHz simultaneous (p > 0.615), 1.5 MHz leading (p > 0.212), or 3.68 MHz leading (p > 0.203) exposures [Figs. 9(a)–9(c)]. Similarly, no differences were found throughout the pulse duration for tendinopathic tendons exposed to 1.5 MHz with 3.68 MHz beginning after a 5 ms delay (p > 0.240) or 3.68 MHz with 1.5 MHz beginning after a 5 ms delay (p > 0.066) compared to healthy tendons. While some differences were observed in PCD results, they could not be used to predict successful fractionation.

FIG. 9. — Cavitation activity throughout (a)–(c) the treatment time and (d)–(f) the pulse duration for healthy and tendinopathic tendons exposed to (a), (d) 1.5 and 3.68 MHz simultaneous, (b), (e) 1.5 MHz with 3.68 MHz delayed, and (c), (f) 3.68 MHz with 1.5 MHz delayed. No statistical differences in cavitation activity were found between tendinopathic and healthy tendons exposed to all exposures.

IV. DISCUSSION

Dual-frequency histotripsy at 1.5 and 3.68 MHz combinations (simultaneous and 1.5 MHz leading) produced a fractionated hole in tendinopathic bovine tendons. Exposure to dual-frequency with 3.68 MHz leading or single-frequency histotripsy of 1.5 MHz or 3.68 MHz also produced partial fractionation in the tendinopathy model; exposure to 1.07 MHz alone or 1.07 and 1.5 MHz combined did not show any fractionation in tendinopathic tendons. All healthy tendons exposed to any of the tested parameters did not result in mechanical fractionation.

Obvious holes in tendinopathic tendons occurred with simultaneous 3.68 and 1.5 MHz or 1.5 MHz with 3.68 MHz delayed by 5 ms. Histologically, loss of fibrillar structure and thermal necrosis at the focal area was also observed for both exposures. Exposure to 1.5 MHz with 3.68 MHz delayed by 5 ms in vitro also displayed the shortest extent of bubble activity (1 mm × 4–5 mm). On gross evaluation, thermal necrosis was significant from the surface of tendon nearer the 3.68 MHz transducer all the way through to the focus. These dual frequency parameters that produced the most mechanical disruption also showed the largest extent of thermal necrosis, suggesting heating may have further weakened the tissue. Collagen denaturation by heat has been shown to increase susceptibility to histotripsy (Vlaisavljevich et al., 2015a). Healthy tendons also displayed thermal necrosis as did other tendinopathic tendons exposed to 1.07 MHz or a combination of 1.07 and 1.5 MHz, but the appearance of thermal necrosis was sparse within the focal volume compared to 3.68 MHz combinations. It is also possible that the large boiling bubbles may induce atomization and/or larger stresses on the weakened fibers (Simon et al., 2015; Vlaisavljevich et al., 2015b). It is likely a combination of collagenase and thermal degradation is needed to weaken the tissue for successful histotripsy fractionation by the dual-frequency boiling histotripsy.

Partial fractionation occurred in tendinopathic tendons exposed to 3.68 MHz with 1.5 MHz on a 5-ms delay, and single-frequency treatments of 1.5 MHz or 3.68 MHz. These results are supported by histology with varying degrees of fiber separation depending on the exact exposure parameters. While thermal necrosis was observed in all these exposures, 1.5 MHz single-frequency histotripsy (which produced partial fractionation) showed similar sparse thermal necrosis to 1.07 MHz single- and dual-frequency combinations that did not result in any fractionation. It is unclear how thermal necrosis contributes to the partial fractionation (as compared to complete fractionation) that occurs for these exposure parameters. In vitro exposure to 3.68 MHz leading with 1.5 MHz delayed by 5 ms resulted in greater extent of bubble activity (2 mm × 6 mm) compared to 1.5 MHz with 3.68 MHz delayed by 5 ms (1 mm × 4–5 mm), which displayed full fractionation in tendons. However, extent of bubble activity was smaller than the bubble extent of 1.07 and 1.5 MHz exposures in vitro (3 mm × 9.5 mm), which did not display any fractionation.

PCD analysis resulted in no significant differences between tendinopathic tendons and healthy tendons despite only tendinopathic tendons displaying holes grossly. Cavitation emissions during histotripsy have been used to monitor treatment progression as bubble activity is dependent on the properties of the tissue (Canney et al., 2010; Vlaisavljevich et al., 2015b). However, in this study cavitation activity was not able to predict the bioeffect of full or partial fractionation. More work is needed to parse out differences in cavitation energies and the resulting bioeffect between healthy and tendinopathic tendons.

This pilot study is limited in sample size; however, differences in histotripsy fractionation and cavitation activity were still evident. It is unclear why exposure to 3.68 MHz leading dual-frequency histotripsy did not display complete fractionation, but rather partial fractionation, despite exposure to the same temporal peak intensity as the 1.5 MHz leading dual-frequency histotripsy. It is important to note that pure mechanical fractionation has not yet been achieved in tendons; successfully fractionated samples still displayed collateral thermal necrosis. Future work should be focused on multiple frequency and bubble simulations that include a viscoelastic surrounding medium in simulations and tissue-mimicking phantom models to begin to understand bubble dynamics in the tendons.

V. CONCLUSION

This study evaluated single and dual frequency ultrasound in a tendinopathy model. Tendinopathic tendons exposed to histotripsy at 1.5 and 3.68 MHz simultaneous or 1.5 MHz with 3.68 MHz delayed by 5 ms were successfully fractionated in combination with thermal necrosis. Tendinopathic tendons exposed to histotripsy at 1.5, 3.68, or 3.68 MHz with 1.5 MHz delayed by 5 ms were partially fractionated. Healthy tendons of the same exposures did not result in fractionation. It is likely that successful fractionation was caused by weakening of collagen fibers from the collagenase injection; thermal absorption may have further weakened the focal area and contributed to the successful disruption. These results are the first such evidence of localized histotripsy fractionation of an intact tendinopathic tendon.

ACKNOWLEDGMENTS

The authors would like to thank members of the Biomedical Acoustics Simon Laboratory (BASiL), especially Grace Wood for support in experiments. Thank you also to Penn State Meats Lab and Bierly's Meat Market for supplying tendons. This work was funded by NSF GRFP (Grant No. DGE1255832) and NIH NIBIB (Grant No. R21EB027886).

References

1. Allen, L. M. , and Silverman, R. K. (2000). “ Prenatal ultrasound evaluation of fetal diastematomyelia: Two cases of type I split cord malformation,” Ultrasound Obstet. Gynecol. 15(1), 78–82. 10.1046/j.1469-0705.2000.00019.x [DOI] [PubMed] [Google Scholar]
2.APTA (2013). Description of dry needling in clinical practice: An educational resource paper.
3. Canney, M. S. , Khokhlova, V. A. , Bessonova, O. V. , Bailey, M. R. , and Crum, L. A. (2010). “ Shock-induced heating and millisecond boiling in gels and tissue due to high intensity focused ultrasound,” Ultrasound Med. Biol. 36(2), 250–267. 10.1016/j.ultrasmedbio.2009.09.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Duck, F. A. (2013). Physical Properties of Tissues: A Comprehensive Reference Book ( Academic, New York: ). [Google Scholar]
5. Groth, K. , Berezhanskyy, T. , Aneja, M. K. , Geiger, J. , Schweizer, M. , Maucksch, L. , Pasewald, T. , Brill, T. , Tigani, B. , Weber, E. , Rudolph, C. , and Hasenpusch, G. (2017). “ Tendon healing induced by chemically modified MRNAS,” Eur. Cells Mater. 33, 294–307. 10.22203/eCM.v033a22 [DOI] [PubMed] [Google Scholar]
6. Kaux, J. F. , Forthomme, B. , le Goff, C. , Crielaard, J. M. , and Croisier, J. L. (2011). “ Current opinions on tendinopathy,” J. Sport Sci. Med. 10(2), 238–253. [PMC free article] [PubMed] [Google Scholar]
7. Khandare, S. , Smallcomb, M. , Klein, B. , Geary, C. , Simon, J. C. , and Vidt, M. E. (2021). “ Comparison between dry needling and focused ultrasound on the mechanical properties of the rat Achilles tendon: A pilot study,” J. Biomech. 120, 110384. 10.1016/j.jbiomech.2021.110384 [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Khokhlova, T. D. , Wang, Y.-N. , Simon, J. C. , Cunitz, B. W. , Starr, F. , Paun, M. , Crum, L. A. , Bailey, M. R. , and Khokhlova, V. A. (2014). “ Ultrasound-guided tissue fractionation by high intensity focused ultrasound in an in vivo porcine liver model,” Proc. Natl. Acad. Sci. U.S.A. 111(22), 8161–8166. 10.1073/pnas.1318355111 [DOI] [PMC free article] [PubMed] [Google Scholar]
9. Lafon, C. , Zderic, V. , Noble, M. L. , Yuen, J. C. , Kaczkowski, P. J. , Sapozhnikov, O. A. , Chavirier, F. , Crum, L. A. , and Vaezy, S. (2005). “ Gel phantom for use in high-intensity focused ultrasound dosimetry,” Ultrasound Med. Biol. 31(10), 1383–1389. 10.1016/j.ultrasmedbio.2005.06.004 [DOI] [PubMed] [Google Scholar]
10. Li, T. , Chen, H. , Khokhlova, T. , Wang, Y.-N. , Kreider, W. , He, X. , and Hwang, J. H. (2014). “ Passive cavitation detection during pulsed HIFU exposures of ex vivo tissues and in vivo mouse pancreatic tumors,” Ultrasound Med. Biol. 40(7), 1523–1534. 10.1016/j.ultrasmedbio.2014.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Lin, Y. C. , Mao, M. H. , Lin, Y. R. , Lin, H. H. , Lin, C. A. , and Wang, L. A. (2014). “ All-optical switching in GaAs microdisk resonators by a femtosecond pump–probe technique through tapered-fiber coupling,” Opt. Lett. 39(17), 4998–5001. 10.1364/OL.39.004998 [DOI] [PubMed] [Google Scholar]
12. Notarnicola, A. , and Moretti, B. (2012). “ The biological effects of extracorporeal shock wave therapy (ESWT) on tendon tissue,” Muscles Ligaments Tendons J. 2(1), 33–37. [PMC free article] [PubMed] [Google Scholar]
13. Parsons, J. E. , Cain, C. A. , Abrams, G. D. , and Fowlkes, J. B. (2006). “ Pulsed cavitational ultrasound therapy for controlled tissue homogenization,” Ultrasound Med. Biol. 32(1), 115–129. 10.1016/j.ultrasmedbio.2005.09.005 [DOI] [PubMed] [Google Scholar]
14. Simon, J. C. , Sapozhnikov, O. A. , Wang, Y.-N. , Khokhlova, V. A. , Crum, L. A. , and Bailey, M. R. (2015). “ Investigation into the mechanisms of tissue atomization by high-intensity focused ultrasound,” Ultrasound Med. Biol. 41(5), 1372–1385. 10.1016/j.ultrasmedbio.2014.12.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Smallcomb, M. , Elliott, J. , Khandare, S. , Butt, A. A. , Vidt, M. E. , and Simon, J. C. (2021). “ Focused ultrasound mechanical disruption of ex vivo rat tendon,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 68(9), 2981–2986. 10.1109/TUFFC.2021.3075375 [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Smallcomb, M. , and Simon, J. C. (2019). “ Histotripsy in collagenous tendons,” Proc. Mtgs. Acoust. 39(1), 020004. 10.1121/2.0001249 [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Szabo, T. L. (2016). Diagnostic Ultrasound Imaging: Inside Out ( Academic, New York: ). [Google Scholar]
18. Ter Haar, G. (2007). “ Therapeutic applications of ultrasound,” Prog. Biophys. Mol. Biol. 93(1-3), 111–129. 10.1016/j.pbiomolbio.2006.07.005 [DOI] [PubMed] [Google Scholar]
19. Vlaisavljevich, E. , Lin, K. W. , Warnez, M. T. , Singh, R. , Mancia, L. , Putnam, A. J. , Johnsen, E. , Cain, C. , and Xu, Z. (2015b). “ Effects of tissue stiffness, ultrasound frequency, and pressure on histotripsy-induced cavitation bubble behavior,” Phys. Med. Biol. 60(6), 2271–2292. 10.1088/0031-9155/60/6/2271 [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Vlaisavljevich, E. , Maxwell, A. , Warnez, M. , Johnsen, E. , Cain, C. A. , and Xu, Z. (2014). “ Histotripsy-induced cavitation cloud initiation thresholds in tissues of different mechanical properties,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 61(2), 341–352. 10.1109/TUFFC.2014.6722618 [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Vlaisavljevich, E. , Xu, Z. , Arvidson, A. , Jin, L. , Roberts, W. , and Cain, C. (2015a). “ Effects of thermal preconditioning on tissue susceptibility to histotripsy,” Ultrasound Med. Biol. 41(11), 2938–2954. 10.1016/j.ultrasmedbio.2015.07.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Watts, A. E. , Nixon, A. J. , Yeager, A. E. , and Mohammed, H. O. (2012). “ A collagenase gel/physical defect model for controlled induction of superficial digital flexor tendonitis,” Equine Vet. J. 44(5), 576–586. 10.1111/j.2042-3306.2011.00471.x [DOI] [PubMed] [Google Scholar]
23. Yuldashev, P. V. , Karzova, M. M. , Kreider, W. , Rosnitskiy, P. B. , Sapozhnikov, O. A. , and Khokhlova, V. A. (2021). “ ‘HIFU beam’: A simulator for predicting axially symmetric nonlinear acoustic fields generated by focused transducers in a layered medium,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 68(9), 2837–2852. 10.1109/TUFFC.2021.3074611 [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Zhou, Y. , and Lei, Y. (2020). “ Cavitation histotripsy at the surface of soft material using the dual-frequency excitation,” Eng. Res. Express 2(2), 025015. 10.1088/2631-8695/ab8957 [DOI] [Google Scholar]

[c2] 1. Allen, L. M. , and Silverman, R. K. (2000). “ Prenatal ultrasound evaluation of fetal diastematomyelia: Two cases of type I split cord malformation,” Ultrasound Obstet. Gynecol. 15(1), 78–82. 10.1046/j.1469-0705.2000.00019.x [DOI] [PubMed] [Google Scholar]

[c1] 2.APTA (2013). Description of dry needling in clinical practice: An educational resource paper.

[c3] 3. Canney, M. S. , Khokhlova, V. A. , Bessonova, O. V. , Bailey, M. R. , and Crum, L. A. (2010). “ Shock-induced heating and millisecond boiling in gels and tissue due to high intensity focused ultrasound,” Ultrasound Med. Biol. 36(2), 250–267. 10.1016/j.ultrasmedbio.2009.09.010 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c4] 4. Duck, F. A. (2013). Physical Properties of Tissues: A Comprehensive Reference Book ( Academic, New York: ). [Google Scholar]

[c5] 5. Groth, K. , Berezhanskyy, T. , Aneja, M. K. , Geiger, J. , Schweizer, M. , Maucksch, L. , Pasewald, T. , Brill, T. , Tigani, B. , Weber, E. , Rudolph, C. , and Hasenpusch, G. (2017). “ Tendon healing induced by chemically modified MRNAS,” Eur. Cells Mater. 33, 294–307. 10.22203/eCM.v033a22 [DOI] [PubMed] [Google Scholar]

[c7] 6. Kaux, J. F. , Forthomme, B. , le Goff, C. , Crielaard, J. M. , and Croisier, J. L. (2011). “ Current opinions on tendinopathy,” J. Sport Sci. Med. 10(2), 238–253. [PMC free article] [PubMed] [Google Scholar]

[c8] 7. Khandare, S. , Smallcomb, M. , Klein, B. , Geary, C. , Simon, J. C. , and Vidt, M. E. (2021). “ Comparison between dry needling and focused ultrasound on the mechanical properties of the rat Achilles tendon: A pilot study,” J. Biomech. 120, 110384. 10.1016/j.jbiomech.2021.110384 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c9] 8. Khokhlova, T. D. , Wang, Y.-N. , Simon, J. C. , Cunitz, B. W. , Starr, F. , Paun, M. , Crum, L. A. , Bailey, M. R. , and Khokhlova, V. A. (2014). “ Ultrasound-guided tissue fractionation by high intensity focused ultrasound in an in vivo porcine liver model,” Proc. Natl. Acad. Sci. U.S.A. 111(22), 8161–8166. 10.1073/pnas.1318355111 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c24] 9. Lafon, C. , Zderic, V. , Noble, M. L. , Yuen, J. C. , Kaczkowski, P. J. , Sapozhnikov, O. A. , Chavirier, F. , Crum, L. A. , and Vaezy, S. (2005). “ Gel phantom for use in high-intensity focused ultrasound dosimetry,” Ultrasound Med. Biol. 31(10), 1383–1389. 10.1016/j.ultrasmedbio.2005.06.004 [DOI] [PubMed] [Google Scholar]

[c10] 10. Li, T. , Chen, H. , Khokhlova, T. , Wang, Y.-N. , Kreider, W. , He, X. , and Hwang, J. H. (2014). “ Passive cavitation detection during pulsed HIFU exposures of ex vivo tissues and in vivo mouse pancreatic tumors,” Ultrasound Med. Biol. 40(7), 1523–1534. 10.1016/j.ultrasmedbio.2014.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c11] 11. Lin, Y. C. , Mao, M. H. , Lin, Y. R. , Lin, H. H. , Lin, C. A. , and Wang, L. A. (2014). “ All-optical switching in GaAs microdisk resonators by a femtosecond pump–probe technique through tapered-fiber coupling,” Opt. Lett. 39(17), 4998–5001. 10.1364/OL.39.004998 [DOI] [PubMed] [Google Scholar]

[c12] 12. Notarnicola, A. , and Moretti, B. (2012). “ The biological effects of extracorporeal shock wave therapy (ESWT) on tendon tissue,” Muscles Ligaments Tendons J. 2(1), 33–37. [PMC free article] [PubMed] [Google Scholar]

[c13] 13. Parsons, J. E. , Cain, C. A. , Abrams, G. D. , and Fowlkes, J. B. (2006). “ Pulsed cavitational ultrasound therapy for controlled tissue homogenization,” Ultrasound Med. Biol. 32(1), 115–129. 10.1016/j.ultrasmedbio.2005.09.005 [DOI] [PubMed] [Google Scholar]

[c14] 14. Simon, J. C. , Sapozhnikov, O. A. , Wang, Y.-N. , Khokhlova, V. A. , Crum, L. A. , and Bailey, M. R. (2015). “ Investigation into the mechanisms of tissue atomization by high-intensity focused ultrasound,” Ultrasound Med. Biol. 41(5), 1372–1385. 10.1016/j.ultrasmedbio.2014.12.022 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c16] 15. Smallcomb, M. , Elliott, J. , Khandare, S. , Butt, A. A. , Vidt, M. E. , and Simon, J. C. (2021). “ Focused ultrasound mechanical disruption of ex vivo rat tendon,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 68(9), 2981–2986. 10.1109/TUFFC.2021.3075375 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c15] 16. Smallcomb, M. , and Simon, J. C. (2019). “ Histotripsy in collagenous tendons,” Proc. Mtgs. Acoust. 39(1), 020004. 10.1121/2.0001249 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c17] 17. Szabo, T. L. (2016). Diagnostic Ultrasound Imaging: Inside Out ( Academic, New York: ). [Google Scholar]

[c6] 18. Ter Haar, G. (2007). “ Therapeutic applications of ultrasound,” Prog. Biophys. Mol. Biol. 93(1-3), 111–129. 10.1016/j.pbiomolbio.2006.07.005 [DOI] [PubMed] [Google Scholar]

[c20] 19. Vlaisavljevich, E. , Lin, K. W. , Warnez, M. T. , Singh, R. , Mancia, L. , Putnam, A. J. , Johnsen, E. , Cain, C. , and Xu, Z. (2015b). “ Effects of tissue stiffness, ultrasound frequency, and pressure on histotripsy-induced cavitation bubble behavior,” Phys. Med. Biol. 60(6), 2271–2292. 10.1088/0031-9155/60/6/2271 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c18] 20. Vlaisavljevich, E. , Maxwell, A. , Warnez, M. , Johnsen, E. , Cain, C. A. , and Xu, Z. (2014). “ Histotripsy-induced cavitation cloud initiation thresholds in tissues of different mechanical properties,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 61(2), 341–352. 10.1109/TUFFC.2014.6722618 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c19] 21. Vlaisavljevich, E. , Xu, Z. , Arvidson, A. , Jin, L. , Roberts, W. , and Cain, C. (2015a). “ Effects of thermal preconditioning on tissue susceptibility to histotripsy,” Ultrasound Med. Biol. 41(11), 2938–2954. 10.1016/j.ultrasmedbio.2015.07.016 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c21] 22. Watts, A. E. , Nixon, A. J. , Yeager, A. E. , and Mohammed, H. O. (2012). “ A collagenase gel/physical defect model for controlled induction of superficial digital flexor tendonitis,” Equine Vet. J. 44(5), 576–586. 10.1111/j.2042-3306.2011.00471.x [DOI] [PubMed] [Google Scholar]

[c22] 23. Yuldashev, P. V. , Karzova, M. M. , Kreider, W. , Rosnitskiy, P. B. , Sapozhnikov, O. A. , and Khokhlova, V. A. (2021). “ ‘HIFU beam’: A simulator for predicting axially symmetric nonlinear acoustic fields generated by focused transducers in a layered medium,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 68(9), 2837–2852. 10.1109/TUFFC.2021.3074611 [DOI] [PMC free article] [PubMed] [Google Scholar]

[c23] 24. Zhou, Y. , and Lei, Y. (2020). “ Cavitation histotripsy at the surface of soft material using the dual-frequency excitation,” Eng. Res. Express 2(2), 025015. 10.1088/2631-8695/ab8957 [DOI] [Google Scholar]

PERMALINK

Dual-frequency boiling histotripsy in an ex vivo bovine tendinopathy model

Molly Smallcomb

Julianna C Simon

Abstract

I. INTRODUCTION