Effects of focused ultrasound and dry needling on tendon mechanical properties

Abstract

Tendon injuries are extremely common, resulting in mechanically weaker tendons that could lead to tendon rupture. Dry needling (DN) is widely used to manage pain and function after injury. However, DN is invasive and high inter-practitioner variability has led to mixed success rates. Focused ultrasound (fUS) is a non-invasive medical technology that directs ultrasound energy into a well-defined focal volume. fUS can induce thermal ablation or mechanical fractionation, with bioeffect type controlled through ultrasound parameters. Tendons must withstand high physiological loads, thus treatments maintaining tendon mechanical properties while promoting healing are needed. Our objective was to evaluate mechanical effects of DN and 3 fUS parameter sets, chosen to prioritize mechanical fractionation, on Achilles and supraspinatus tendons. Ex vivo rat Achilles and supraspinatus tendons (50 each) were divided into sham, DN, fUS-1, fUS-2, and fUS-3 (n=10/group). Following treatment, tendons were mechanically tested. Elastic modulus of supraspinatus tendons treated with DN (126.64±28.1MPa) was lower than sham (153.02±29.3MPa; p=0.0280). Stiffness and percent relaxation of tendons treated with DN (Achilles: 114.40±31.6N/mm; 49.10±6.1%; supraspinatus: 109.53±30.8N/mm; 50.17±7.6%) were lower (all p<0.0334) than sham (Achilles: 141.34±20.9N/mm; 60.30±7.7%; supraspinatus: 135.14±30.2N/mm; 60.85±15.4%). Modulus of Achilles and supraspinatus tendons treated with fUS-1 (159.88±25.7MPa; 150.12±22.0MPa, respectively) were similar to sham (156.35±23.0MPa; 153.02±29.3MPa, respectively). These results suggest that fUS preserves mechanical properties better than DN, with fUS-1 performing better than fUS-2 and fUS-3. fUS should be studied further to fully understand its mechanical and healing effects to help evaluate fUS as an alternative, non-invasive treatment for tendon injuries.

Introduction

Musculoskeletal injuries affect over 102 million Americans annually, with an associated annual cost estimated at $796 billion (Yelin et al., 2016). Tendinopathy accounts for ~30% of musculoskeletal pain consultations (Kaux et al., 2011) with Achilles and supraspinatus tendons most prone to pathology (Maffulli et al., 2003). Microtears and collagen fiber disorganization accompanying tendinopathy (Andres and Murrell, 2008), result in mechanically weaker tendons (Soslowsky et al., 2000). Tendons are load-bearing structures that must withstand high physiological loads, e.g. 12.5x bodyweight by the Achilles tendon (Komi PV, 1992) and up to 1.4x bodyweight at the glenohumeral joint (Westerhoff et al., 2009). Therefore, tendinopathy treatments should be assessed based on their mechanical effects to ensure treatments do not diminish tendon mechanical properties.

Conservative non-invasive treatments for tendinopathy include rest, physical therapy, and/or nonsteroidal anti-inflammatory medications (Rees et al., 2009). When these options fail to improve patients’ symptoms (Andres and Murrell, 2008), percutaneous and invasive treatments, like dry needling (DN), are considered. Dry needling is widely used for managing pain or range of motion (Dommerholt et al., 2019). The American Physical Therapy Association defines DN as a “skilled intervention that uses a thin filiform needle to penetrate the skin and stimulate underlying myofascial trigger points, muscular or connective tissues for management of neuromusculoskeletal pain and movement impairments” (APTA, 2013). The rationale behind DN is that repetitive needle insertions disrupt the chronic degenerative process, introduce localized bleeding, and facilitate healing by stimulating growth factors (Chiavaras and Jacobson, 2013; Krey et al., 2015). However, repeated needle insertions may disrupt collagen fibers, given that the diameters of collagen fiber (~1 μm-300 μm) (Silver F H, 1992) and needle (514.4 μm-1270 μm) (Stoychev and Finestone, 2020) are not comparable. Disrupting the hierarchical collagen structure may further mechanically weaken the injured tendon, negatively affecting the tendons’ ability to withstand load.

Studies assessing the efficacy of DN for tendinopathy report mixed outcomes (Krey et al., 2015; Stoychev and Finestone, 2020). Practitioners often vary in location, direction, dosage, and depth of needle insertions, and do not always use real-time ultrasound guidance. Most studies compare DN as the control group and DN followed by injection of platelet-rich plasma or autologous blood as the intervention group (Bell et al., 2013; Mishra et al., 2014), therefore lacking a true control group. Although DN is widely used in the clinic, limited controlled laboratory studies (Kim et al., 2015; Riggin et al., 2019) have explored its mechanical effects. Understanding the mechanical effects of DN will help determine whether DN diminishes tendon mechanical properties.

In physical therapy, ultrasound is commonly used as an imaging modality to diagnose tendon injuries and/or guide treatment. Therapeutic ultrasound at low intensities can facilitate increase in tensile strength and collagen synthesis in tendons (Enwemeka, 1989; Yeung et al., 2006) and promote soft tissue healing (Speed, 2001). Focused ultrasound (fUS) is a non-invasive medical technology that directs ultrasound energy into a well-defined focal volume without damaging the intervening tissues (Bailey et al., 2003; Dubinsky et al., 2008). Choice of ultrasound parameters and material properties of the target tissue can direct fUS bioeffects toward thermal ablation from absorption of acoustic energy (Bailey et al., 2003; ter Haar, 2007) or mechanical fractionation from creation, oscillation, and collapse of cavitation bubbles (Khokhlova et al., 2014, 2015). fUS can be monitored in real-time using magnetic resonance imaging (for thermal effects) or ultrasound imaging (for cavitation bubbles) (Khokhlova et al., 2014, 2015). fUS has been used clinically to treat uterine fibroids (Kim et al., 2012), liver tumors (Leslie et al., 2012), and pancreatic cancer (Khokhlova and Hwang, 2011). Although these treatments use thermal bioeffects, recent studies have explored the capabilities of fUS to generate mechanical bioeffects by mechanically fractionating tissue within the focal volume while minimizing thermal effects – this approach is termed histotripsy (Khokhlova et al., 2015).

Mechanical bioeffects of fUS are typically induced using a sequence of acoustic pulses to fractionate (liquefy) tissue into subcellular fragments. Varying ultrasonic parameters, like pulse duration, pulse repetition frequency (PRF), and pressure amplitude, can emphasize mechanical or thermal bioeffects (Smallcomb and Simon, 2019). Mechanical bioeffects of fUS can potentially introduce micro-damage to elicit the healing response, similar to DN; however, tendons have been resistant to mechanical fractionation (Vlaisavljevich et al., 2011) and it remains unclear how varying fUS parameters influence the type of bioeffect. If fUS can induce micro-damage similar to DN while facilitating improved collagen alignment (Best et al., 2018; Yeung et al., 2006), this could represent a promising non-invasive alternative for treating tendon injuries.

Our previous studies (Smallcomb et al., 2021; Smallcomb and Simon, 2019) identified a narrow parameter space that emphasized mechanical fractionation over thermal ablation in tendons. The objective of this study was to evaluate and compare the effects of DN and 3 different parameter sets of fUS, chosen to prioritize mechanical fractionation, on the mechanical properties of tendons. Mechanical testing was performed on ex vivo rat Achilles and supraspinatus tendons. We hypothesized that tendons treated with one fUS parameter set, that prioritizes mechanical fractionation, would perform better mechanically than the other fUS parameter sets and DN.

Materials and Methods

Study design and animal use

This study was approved by The Pennsylvania State University Institutional Animal Care and Use Committee (IACUC). Fifty Achilles and 50 supraspinatus tendons (Table 1) were harvested from Sprague–Dawley rats (16M/16F, 8–11 weeks, 176–200 grams). Tendon samples were submerged in phosphate buffered saline (PBS) after dissection to maintain hydration. Tendons were randomly divided into five groups: sham, DN, fUS-1, fUS-2, fUS-3, with 10 Achilles and 10 supraspinatus tendons per group.

Table 1:

Study design showing number of tendons in each treatment group

	Achilles tendons (50)					Supraspinatus tendons (50)
	Sham	DN	fUS-1	fUS-2	fUS-3	Sham	DN	fUS-1	fUS-2	fUS-3
Male	6	5	5	5	6	7	5	6	6	5
Female	4	5	5	5	4	3	5	4	4	5

Sample preparation

Achilles tendon samples were prepared by removing the gastrocnemius muscle, plantaris tendon, and soft tissue around the tendon; muscles and soft tissue around the foot were removed to expose the bones. Supraspinatus tendon samples were prepared by removing the supraspinatus muscle and soft tissue around the tendon; muscles and soft tissue around the humerus were removed to expose the bone. The foot (Achilles tendon samples) and humerus (supraspinatus tendon samples) were embedded in a custom holding fixture using epoxy (Loctite, Düsseldorf, Germany) while keeping the tendon submerged in PBS. A second layer of epoxy was applied to the calcaneus or humeral head to prevent failure at growth plate.

Focused ultrasound (fUS)

Focused ultrasound treatment was conducted in a filtered, deionized, and degassed (<6 mg/L oxygen) water tank (Fig. 1). Custom fixtures were used to hold the tendon taut, leaving an acoustic window for the tendon. A single-element 1.5 MHz fUS transducer with f-number of 0.7 (modified H-234, Sonic Concepts, Bothell, WA, USA) was aligned with tendon midbody and treated 1.5-mm deep at one site per tendon. Treatment was monitored in real-time using a P4–2 imaging transducer (Philips/ATL, Bothell, WA, USA) and B-mode ultrasound imaging system (Vantage-128, Verasonics®, Kirkland, WA, USA). Acoustic parameters including pulse durations, PRF, and treatment times were varied (Table 2) with peak pressure amplitudes set at 89 MPa (peak positive pressure) and 26 MPa (peak negative pressure). These parameter sets were chosen based on preliminary work (Khandare et al., 2021; Smallcomb et al., 2021; Smallcomb and Simon, 2019) exploring a range of acoustic parameters producing mechanically-induced fUS disruption based on conventional boiling histotripsy parameters (10 ms pulses at 1 Hz PRF) while minimizing thermal injury. Immediately after fUS treatment, tendons were prepared for mechanical testing.

Figure 1:

Focused ultrasound (fUS) experimental arrangement showing the fUS transducer with coaxially aligned imaging probe targeting the supraspinatus tendon.

Table 2:

Three acoustic parameter sets for focused ultrasound

Acoustic parameters for the three focused ultrasound (1.5 MHz) parameter sets
	Pulse duration (ms)	Pulse repetition frequency (Hz)	Duty cycle (%)	Treatment time (s)
fUS-1	1	10	1	60
fUS-2	10	1	1	15
fUS-3	5	1	0.5	60

Dry needling (DN)

A 30 G acupuncture needle (311.2 μm diameter, Tai-Chi, Suzhou, Jiangsu, China) was inserted and quickly removed from tendon midbody 5 times over 12 s (Fig. 2). Number of needle insertions was scaled down based on rat Achilles tendon thickness (Huang et al., 2004) being ~1/10th of human Achilles tendon thickness (Iwanuma et al., 2011). Frequency of needle insertions was scaled down from clinical methodology based on previous studies in humans reporting 40–50 needle insertions in 2 minutes (Krey et al., 2015). Tendons were then submerged in deionized water for 30 minutes to mimic fUS environmental conditions, with no exposure to ultrasound. Tendons were prepared for mechanical testing immediately following DN treatment.

Figure 2:

Dry needling setup showing 30 G needle that was inserted and quickly removed from the midbody of the Achilles tendon.

Sham

Tendons in sham group were submerged in deionized water for 30 minutes to mimic fUS environment conditions without applying ultrasound. Immediately after sham treatment, tendons were prepared for mechanical testing.

Tendon mechanical testing

Tendon width and thickness were measured from tendon midbody using digital Vernier caliper, and cross-sectional area was calculated assuming an elliptical cross-section. Evenly spaced Verhoeff stain lines, four lines on Achilles tendon (Fig. 3a) and three lines on supraspinatus tendon (Fig. 3b), were marked perpendicular to the tendon long axis for optical strain analysis. Proximal end of the tendon was glued (Loctite, Düsseldorf, Germany) between two pieces of fine grit sandpaper and fixed in custom metal grips (cf. Fig. 3a, b). The custom grips and testing fixtures were then mounted on an MTS 858 Mini Bionix mechanical testing system (MTS Systems Corp., Eden Prairie, MN, USA) using a 50 lbs load cell (MTS Systems Corp., Eden Prairie, MN, USA).

Figure 3:

(a) The foot was embedded in a custom holding fixture using epoxy and 4 Verhoeff stain lines were marked on the Achilles tendon; (b) the humerus was embedded in a custom holding fixture using epoxy and 3 Verhoeff stain lines were marked on the supraspinatus tendon.

Tendons were tested mechanically following procedures described previously (Huang et al., 2004; Khandare et al., 2021). Briefly, tendon samples were preloaded to 0.01 N, then preconditioned for 10 cycles between 0.1 N and 0.5 N. After returning to 0.1 N following the 10th cycle, 300 s hold was applied to allow the tissue to equilibrate. Next, stress relaxation test was performed, in which tendons were exposed to 5 % strain at a rate of 2.5 mm/s and held for 600 s to allow the tendon to reach equilibrium. Samples were then returned to zero load and a load-to-failure test in axial tension was performed at a constant rate of 0.015 mm/s (~0.3 % strain/s). Tendons were sprayed with PBS at regular intervals. The experiment was recorded using a Panasonic Lumix DC-GH5 camera at 30 frames/s.

Stress was calculated as force divided by initial cross-sectional area and Lagrangian strain (vertical) was calculated from Verhoeff stain line displacement using ImageJ software (1.52a, National Institutes of Health, Bethesda, Maryland, USA). Measures of viscoelastic properties, including peak stress and equilibrium stress, were determined from the stress-relaxation curve, and percent relaxation was calculated as percent change from peak stress to equilibrium stress. Stiffness and elastic modulus were calculated from the linear portion of load-to-failure curve, as the slope of load-displacement curve and stress-strain curve, respectively. Maximum load and ultimate tensile stress (UTS) were determined as maximum force and maximum stress, respectively, at tendon failure. A custom MATLAB® (The MathWorks, Inc., Natick, MA) program was used to determine the transition point between the toe region and the linear region on the stress-strain curve, and transition point stress and strain were calculated.

Statistical analysis

Normal distribution of the data was tested using the Shapiro–Wilk test. Outliers were removed based on Z-score of ±2 (i.e. 2 standard deviations above or below group mean). Parametric statistical analysis was performed using one-way ANCOVA, with sex as a covariate, to compare group means. Two-way ANCOVA was used along with Tukey’s Honestly Significant Difference (HSD) test for multiple comparisons to determine the interaction between treatment and tendon (Achilles and supraspinatus). All analyses were performed in SAS software (v9.4, SAS Institute, Inc., Cary, NC, USA), with significance set at p<0.05.

Results

For modulus, UTS, and maximum load, main effects of treatments (p=0.0203, p=0.0008, and p=0.0009, respectively) and tendons (p=0.0072, p<0.0001, and p<0.0001, respectively) were observed. For stiffness and transition point strain, significant interaction between treatments and tendons (p=0.0437 and p=0.0079) and main effect of treatments (p=0.0125 and p=0.0001) were observed. For percent relaxation, main effects of treatments (p=0.0001) were observed. For transition point stress, main effects of tendons (p=0.0106) were observed.

Achilles tendon

Although precautions were taken to avoid breakage at the growth plate, six (2 sham, 2 DN, 1 fUS-2, 1 fUS-3) out of fifty Achilles tendon samples broke at the growth plate and they were removed from data analysis; 1 sample from sham group was excluded due to tissue slippage in the fixture. Forty-three tendon samples were successfully tested: 7 sham (4M/3F); 8 DN (3M/5F); 10 fUS-1 (5M/5F); 9 fUS-2 (4M/5F); and 9 fUS-3 (5M/4F) and 5 measures were removed as outliers. All tendons failed at the insertion site with a small section of tendon still attached to the calcaneus. Modulus of DN (141.97±27.1 MPa), fUS-1 (159.88±25.7 MPa), fUS-2 (149.88±23.9 MPa), fUS-3 (147.22±19.3 MPa) were not different from sham (156.35±23.0 MPa) (Fig. 4a). Stiffness of sham (141.34±20.9 N/mm) was higher than DN (114.40±31.6 N/mm; p=0.0225), fUS-1 (106.41±15.6 N/mm; p=0.0023), fUS-2 (113.47±18.1 N/mm; p=0.0129), and fUS-3 (111.17±17.1 N/mm; p=0.0074) (Fig. 4b). Percent relaxation of sham (60.30±7.7 %) was higher than DN (49.10±6.1 %; p=0.0021), fUS-1 (44.78±5.9 %; p<0.0001), fUS-2 (47.41±6.7 %; p=0.0004), and fUS-3 (46.22±6.9 %; p=0.0002) (Fig. 4c). UTS of sham was lower than fUS-2; UTS of DN was lower than fUS-2 and fUS-3 (Table 3). Maximum load of sham was lower than fUS-2 and fUS-3; maximum load of DN was lower than fUS-2 (Table 3). Transition point stress of sham was lower than DN and fUS-1 (Table 3). Transition point stress of fUS-1 was higher than fUS-3 (Table 3). Transition point strain of DN was higher than sham, fUS-1, fUS-2, and fUS-3 (Table 3).

Figure 4:

(a) Elastic modulus, (b) stiffness, and (c) percent relaxation of Achilles tendons, and (d) elastic modulus, (e) stiffness, and (f) percent relaxation of supraspinatus tendons exposed to sham (7 Achilles; 10 supraspinatus), dry needling (DN) (8 Achilles; 10 supraspinatus), and 3 parameter sets of focused ultrasound: fUS-1 (10 Achilles; 10 supraspinatus), fUS-2 (9 Achilles; 10 supraspinatus), and fUS-3 (9 Achilles; 10 supraspinatus). * indicates (n-1) samples due to removal of outliers.

Table 3:

Ultimate tensile strength, maximum load, transition point stress, and transition point strain of Achilles tendons and supraspinatus tendons exposed to sham, dry needling (DN), and 3 parameter sets of focused ultrasound (fUS-1, fUS-2, and fUS-3)

Achilles tendons
	Ultimate tensile strength (MPa)	Maximum load (N)	Transition point stress (MPa)	Transition point strain
Sham	15.56±3.2	26.64±4.1	1.05±0.6	0.012±0.002* (p=0.0012)
DN	15.31±2.7	28.51±4.8	1.73±0.6* (p=0.0205)	0.016±0.002
fUS-1	16.43±2.4	29.50±5.9	1.84±0.6* (p=0.0055)	0.013±0.001* (p=0.0087)
fUS-2	18.41±3.0* (p=0.0403) # (p=0.0212)	33.26±5.8* (p=0.0141)	1.48±0.6	0.012±0.002* (p=0.0005)
fUS-3	18.14±2.2# (p=0.0352)	31.14±4.7* (p=0.0383)	1.29±0.4 ! (p=0.0344)	0.011±0.002* (p<0.0001)
Supraspinatus tendons
	Ultimate tensile strength (MPa)	Max load (N)	Transition point stress	Transition point strain
Sham	8.88±3.7	11.13±4.8	1.12±0.8	0.012±0.002* (p=0.0428)
DN	12.99±2.3* (p=0.0159)	17.20±3.4* (p=0.0071)	1.05±0.5	0.014±0.003
fUS-1	14.89±4.1* (p=0.0006)	18.97±5.9* (p=0.0007)	1.12±0.6	0.013±0.000
fUS-2	12.84±4.5* (p=0.0194)	16.48±5.8* (p=0.0161)	1.12±0.4	0.013±0.001
fUS-3	14.44±3.4* (p=0.0015)	17.72±3.7* (p=0.0037)	1.40±0.5	0.014±0.002

Data reported as mean ± standard deviation.

^*indicates the value is significantly different from sham.

^#indicates the value is significantly different from DN.

^!indicates the value is significantly different from fUS-1.

Supraspinatus tendon

Fifty tendon samples were successfully tested: 10 sham (7M/3F); 10 DN (5M/5F); 10 fUS-1 (6M/4F); 10 fUS-2 (6M/4F); and 10 fUS-3 (5M/5F) and 4 measures were removed as outliers. All tendons failed at the insertion site with a small section of tendon still attached to the humerus. Modulus of sham (153.02±29.3 MPa) was higher than DN (126.64±28.1 MPa; p=0.0280), fUS-2 (129.30±30.8 MPa; p=0.0476), and fUS-3 (129.18±15.8 MPa; p=0.0450); modulus of fUS-1 (150.12±22.0 MPa) was higher than DN (p=0.0459) (Fig. 4d). Stiffness of sham (135.14±30.2 N/mm) was higher than DN (109.53±30.8 N/mm; p=0.0334); stiffness of fUS-1 (145.98±33.4 N/mm) was higher than DN (p=0.0045) and fUS-2 (112.10±19.1 N/mm; p=0.0136) (Fig. 4e). Percent relaxation of sham (60.85±15.4 %) was higher than DN (50.17±7.6 %; p=0.0247) and fUS-2 (48.90±6.1 %; p=0.0069) (Fig. 4f). UTS of sham was lower than DN, fUS-1, fUS-2, and fUS-3 (Table 3). Maximum load of sham was lower than DN, fUS-1, fUS-2, and fUS-3 (Table 3). Transition point stress of DN, fUS-2, fUS-3 were not different from sham (Table 3). Transition point strain of DN was higher than sham (Table 3).

Discussion

This study examined and compared the effects of DN and 3 fUS parameter sets on the mechanical properties of ex vivo rat Achilles and supraspinatus tendons. Results suggest that both DN and fUS interventions caused a change in tendon mechanical properties. However, tendon mechanical properties from fUS-1 parameter set were closer to those of sham group compared to DN, fUS-2, and fUS-3, especially in supraspinatus tendons.

The modulus, stiffness, and percent relaxation of supraspinatus tendons reduced after DN exposure. These results are consistent with previous studies examining the mechanical effects of DN. Riggin et al. (2019) observed that mild and moderate (3 and 9 needle insertions, respectively) DN application on in vivo rat supraspinatus tendons caused a decline in tendon modulus and stiffness, 1-week post intervention, with the moderate DN group failing to recover properties 6-weeks post intervention. Kim et al. (2015) also observed a decrease in modulus and ultimate tensile strength of rat Achilles tendons 1-week and 4-weeks post-DN intervention (9 needle insertions). Histological observations indicated a loss of collagen alignment, indicating tendon degeneration. Kim et al. (2015) highlight the need for a standardized needling protocol because there exists a lack of uniformity in needling technique (peppering or rotating technique), needle size (18G – 25G), use of ultrasound guidance, and frequency of needle insertion (Chiavaras and Jacobson, 2013; Krey et al., 2015; Stoychev and Finestone, 2020). Moreover, the risk of tendon rupture as a complication of DN increases with severity of a preexisting tendon tear (Chiavaras and Jacobson, 2013). Results from the current study suggest that transition point strain of Achilles and supraspinatus tendons in the DN group was higher than sham, indicative of an elongated toe-region. This suggests that DN may cause collagen disorganization and require more microstructural changes (collagen un-crimping) before transitioning to the linear region of the stress-strain curve. Lee et al., (2017) elucidated mechanisms of tendon damage in rat tail tendon fascicles and observed that the mechanism that causes tendon damage is responsible for decreased linear modulus and elongated toe-region. Reduced modulus, stiffness, percent relaxation, and transition point strain in the DN group suggests that DN diminishes tendon mechanical properties reducing its ability to withstand load. It is also possible that repeated needle insertions weaken the tendon, potentially predisposing the tissue to rupture.

Focused ultrasound has the potential to overcome certain limitations of DN while retaining its key features. Specifically, fUS is non-invasive and can be used under ultrasound guidance for controlled targeting to create micro-damage. Moduli of Achilles and supraspinatus tendons exposed to fUS-1 were similar to sham. Stiffness and percent relaxation of supraspinatus tendons exposed to fUS-1 were also similar to sham suggesting that fUS-1 preserves tendon mechanical properties better than DN, fUS-2, and fUS-3. Vlaisavljevich et al., (2011) explored the effects of high-intensity fUS on tissues with varying mechanical properties and observed that the threshold to initiate a cavitation bubble cloud increases with increased mechanical strength of the target tissue. It is possible that increased toughness makes tendons more resistant to bubble formation and collapse, thus less susceptible to tissue erosion. Previous studies (Enwemeka, 1989; Yeung et al., 2006) show that application of lower intensities of therapeutic ultrasound induce mild hyperthermia and facilitate increases in tensile strength and collagen alignment. Muratore et al. (2008) also showed the ability of fUS to ablate ex vivo bovine Achilles tendons using lower intensities and shorter pulse durations.

Our previous studies (Khandare et al., 2021; Smallcomb et al., 2021; Smallcomb and Simon, 2019) evaluated histological effects of a range of histotripsy parameters to produce mechanical damage while minimizing thermal injury. Smallcomb et al., (2021) showed that mechanically-induced fUS disruption is achievable in tendons in the form of localized fiber separation and/or fiber pattern disruption within a small range of acoustic parameters. Smallcomb et al., (2021) also demonstrated that there is a narrow parameter space between undetectable mechanical damage and thermal damage in which inconsistent mechanical disruption is achieved. Our current study used 3 fUS parameter sets within that range which emphasize mechanical fractionation over thermal ablation. Outcomes indicate that fUS-1 parameter set preserves tendon mechanical properties better than DN and other 2 fUS parameter sets. Given that certain acoustic parameters can preserve tendon mechanical properties and also induce controlled microdamage in tendons (Smallcomb et al., 2021), this study suggests that fUS should be studied further to determine how this modality can be leveraged as a therapy for tendon injuries. More work is also needed to investigate the healing response induced by fUS in tendons.

While a significant effect of treatment on mechanical properties was expected, the effect of tendon on modulus, UTS, maximum load, and percent relaxation was unanticipated. Differences in mechanical parameters of Achilles and supraspinatus tendons suggest that the specific tendon should be considered when assessing mechanical properties. It is possible that differences in mechanical properties observed here are driven by the different roles of these tendons in vivo. For example, Achilles tendon withstands body weight and facilitates locomotion, whereas supraspinatus provides mobility and stability at the glenohumeral joint. The differences in mechanical properties may indicate that treatments should be tailored to the specific tendon, although more work is needed to determine how treatments should be modified.

There are limitations of this study. This work was performed on ex vivo rat tendons and no biological mechanisms or measures of collagen alignment were included. While measures were taken to scale down needle size (18–25 G in humans versus 30 G in rat) and frequency of needle insertions (30–50 times in humans versus 5 times in rat) from humans to rats, which were comparable to previous studies, maintenance of relative sizes among needles and species are challenging. This study was performed on healthy tendons as a preliminary step to determine mechanical effects of DN and fUS. Although we cannot extrapolate whether DN is detrimental to injured tendons, these results provide important control data for future experiments on pathologic tendons. It is also likely that fUS parameters will need additional tuning to induce similar bioeffects in injured tendons. Further work is needed to determine which characteristics of generated bubbles elicit the level of microdamage needed to induce healing in human tendons, as they are approximately 10 times larger than rat tendons. Despite these limitations, this study’s use of an established animal model allows for controlled and repeatable procedures for DN and fUS to evaluate their mechanical effects.

This study evaluated the effects of DN, and 3 fUS parameter sets on the mechanical properties of Achilles and supraspinatus tendons. Outcomes suggest that fUS preserves tendon mechanical properties better than DN, with fUS-1 performing better than fUS-2 and fUS-3, especially in supraspinatus tendons. This study demonstrates promise in guiding development of fUS parameter sets to produce micro-damage similar to DN. Understanding mechanical and healing effects of fUS will help evaluate fUS for clinical translation as an alternative, non-invasive treatment for tendon injuries.