Studies retrieved
A summary of the studies (n=23) and their key features/findings is provided in Table I. According to the subjects, studies in the literature were able to be classified as in vivo or in vitro studies. In vivo studies had been performed on rats [6 papers (10-15)], rabbits [5 papers (6,16-19)], horses [one paper (20)], ponies [2 papers (21,22)] and dogs [one paper (23)], an on normal tendons [9 papers (6,12,14,16,17,19,21-23)] and diseased tendons [6 papers (10,11,13,18,20,24)]. The experimental model for diseased tendons was either a ruptured tendon, in the form of a complete and sutured rupture or a partial rupture [2 papers (10,13)] or a collagenase-induced tendinopathy [4 papers (11,15,18,20)]. Two studies were performed on human tissue, namely one on human Achilles tendon, using microdialysis of the peritendinous environment, and one on rotator cuff tendinopathy with surgical cure. In general, one limb was subjected to ESWT application and the contralateral limb was used as a control.
In vitro studies had been performed on cultured tendon cells from either normal or diseased tendons. A total of three papers reported results on cultured normal tendon cells and three papers compared cultured normal cells with tendinopathic cells. It is noteworthy that in vitro studies were not able to be directly translated to in vivo conditions, although they shed light on important mechanisms of action.
Finding the optimal dose. Information was structured according to whether the experimental model was in vivo or in vitro, and the focus was on cell viability and ECM alterations.
In vivo, for normal rabbit Achilles tendon, Rompe et al (6) determined three main EFD levels with different effects. Doses of up to 0.28 mJ/mm2 did not damage the tendon and adjacent structures, or the alterations were transitory and reversible within four weeks. However, a dose of 0.60 mJ/mm2 produced marked damage to the tendon and paratenon, with fibrinoid necrosis within the tendon and inflammatory and reparative peritendinous reactions that were still present at 4 weeks (6). In the normal rabbit quadriceps tendon, a lower EFD (0.35 mJ/mm2) left no structural alteration at 10 days, a medium EFD (0.5 and 0.9 mJ/mm2) produced edema in the paratenon at 10 days that resolved at 28 days, and a higher EFD (1.2 mJ/mm2) induced marked modifications in cells and ECM at 10 days. Researchers opined that the quadriceps tendon is more ‘resistant’ to ESWT than the Achilles tendon (16). In rats, normal Achilles tendon exposed to 0.15 mJ/mm2 and 1,000 or 1,500 pulses exhibited no structural alteration at 3 weeks. A higher dose of 0.20 mJ/mm2 and 2,000 pulses produced collagen disorganization, inflammatory mononuclear reaction and an increased number of capillaries (12).
In vitro, in normal Achilles rat tenocyte cultures, doses of 0.36 mJ/mm2 and 50, 100 and 250 pulses did not alter normal cell viability, whereas the same EFD with a higher number of pulses and an EFD of 0.68 mJ/mm2, irrespective of the number of pulses, reduced the proportion of viable cells. The cell viability suppression was found to be dose-dependent (25). Normal human tendon cell cultures exhibited no modification when exposed to 0.08, 0.14 and 0.17 mJ/mm2, either at 500 or 1,000 pulses. Furthermore, certain energy levels were found to augment cell viability: 0.14 mJ/mm2 and 1,000 pulses was found to increase the number of viable cells at 24 and 48 h (26), and a dose of 0.17 mJ/mm2 and 1,000 pulses promoted cell viability at 7 and 10 days after treatment, as measured by the increased content of DNA (27). For both tendinopathic and normal human cell cultures, 0.17 mJ/mm2 and 250 or 500 pulses did not alter cell viability, whereas a higher number of pulses (1,000 and 2,000) led to cell death. Furthermore, a regimen with 500 pulses was indicated to promote cell proliferation (28). The variation among the reported doses may be explained by the variability of the model on which they were applied, as cultures were harvested from different human tendons.
As for the ECM, in an in vivo study on normal ponies' tendons, 0.14 mJ/mm2 and 600 pulses produced a rapid stimulation of matrix turnover (3 h) and an increase of enzymatic cleavage of collagen. In the long term (6 weeks), damaged collagen was decreased, presumably replaced by mature collagen (22).
Corroborating the results of the published papers, it may be concluded that there is a variability of optimal doses between in vivo and in vitro studies, among different tendons in the same species and among the same tendons in various species. There is a clear dose-related effect of ESWT, both in terms of cell viability and ECM.
Neovascularization. It is widely accepted that neovascularization is important for healing, as it removes by-products and increases the oxygen content and the number of inflammatory and reparatory cells.
In vivo, in normal rabbit tendons, significantly increased neo-vessels were observed at 4 weeks after one session of 0.12 mJ/mm2, 500 pulses, which persisted up to 12 weeks. Angiogenesis-related markers [endothelial nitric oxide synthase (eNOS), vascular endothelial growth factor (VEGF) and proliferating cell nuclear antigen (PCNA)] were significantly increased at one week. eNOS and VEGF displayed higher levels up to 8 weeks and decreased up to 12 weeks, whereas PCNA continued to have increased levels at 12 weeks (17). In normal dog Achilles tendon-bone junction, application of one session of 0.18 mJ/mm2, 1,000 pulses, was followed by an increase of new capillaries at 4 weeks (17-fold) and 8 weeks (16-fold), most of them located adjacent to the peritendineum. New capillaries were also noted in the adjacent tendon tissues. Muscularized vessels, in a proportionate number to capillaries, were found at 4 weeks and persisted at 8 weeks, mostly in the peritendineum and the adjacent tendon. Both capillaries and muscularized vessels are indicative of neovascularization. No alteration of normal cancellous bone architecture was reported (23). By contrast, a study on normal insertional patellar tendon (rabbit) indicated no alteration in the number of capillaries at 6 weeks after one session of different ESWT regimens (1,000 or 2,000 pulses, 0.18, 0.27 or 0.36 mJ/mm2) (19).
Achilles tendinopathy (rabbit) displayed new capillaries in the peritendinous tissue together with larger blood vessels, including differentiated arterioles and venules, at 4 weeks after 2-weekly sessions of 0.29 mJ/mm2, 1,500 pulses (18). A similar result was reported for the tendinosis of superficial digital flexor tendon (horse) after 3 ESWT sessions, each one 3 weeks apart, with 0.14 mJ/mm2, 1,500 pulses. At five weeks after treatment completion, there were significantly more capillaries in the treated tendons (20). In the study model of a cut Achillean tendon (rat), one session of 500 pulses at 0.19 mJ/mm2 produced a significant increase in the number of capillaries after 20 days (13). In humans with grade III rotator cuff tendinopathy (Riley classification), one session of 4,000 pulses at 0.3 mJ/mm2 increased the vascular volume area with evidence of nodular neo-angiogenesis in deeper tissue layers. For grade IV tendinopathy, evidence of new vessel formation was not significant, mostly due to the small number of cases (24).
In normal human cell cultures, ESWT induced a strong and significant release of TGF-β and VEGF, with a maximum at day 2 and persistence of higher levels at day 7. VEGF is stimulated by the release of certain cytokines (IL-6 and IL-10) and is an important promotor of neo-angiogenesis, contributing to the healing process (27).
Cellularity. In the regeneration process, the cornerstone becomes the activated tenocyte that behaves like a modified form of fibroblast or fibrocyte.
In the in vivo tendinopathic model, ESWT induced acceleration of the healing process, increasing the number of blast-like forms at 4 weeks. At 16 weeks, the healing process was complete, with mature tenocytes in a parallel array (18). Increased expression of PCNA, a protein associated with the S-phase of a dividing cell and a marker of tenocyte proliferation, was documented up to 12 weeks. The presence of immature forms, blast-like, recruited from the peritendon into the lesion, was evident at one week; at 4 weeks, the mature forms (spindle-shaped tenocytes) were arranged into tendon bundles (11).
Tenocyte metabolism is accelerated with intense production of growth factors (TGF-β1 and IGF-1) involved in ECM biosynthesis. Tenocytes displayed an increased number of nuclei and were actively producing ECM (20). Within 6 weeks after ESWT application, TGF-β1 and IGF-1 were significantly increased compared to the control, contributing to tendon healing (11).
In vitro, in normal cell cultures (rats), ESWT at an optimal dose promoted tenocyte proliferation as indicated by upregulation of PCNA as early as 6 h. At 48 h, there was no more stimulatory effect. At 24 h, there was increased expression of TGF-β1 as a mark of increased ECM production and increased production of NO by fibroblasts. NO production was increased at 24 h by certain values of EFD (50 and 100 pulses and 0.36 mJ/mm2), while a higher number of pulses (250 and 500) did not affect NO production (12). Normal human cell cultures did not display any alteration of NO levels after ESWT. Thus, the role of NO in promoting tendon healing remains to be fully elucidated (27).
One session of ESWT increased the proliferation of either healthy or ruptured tendon-derived human cultured cells, with a more prominent effect on the ruptured tendon-derived cells (a ratio of 1.75). ESWT induced a migratory phenotype in both types of culture, particularly evident for cells derived from ruptured tendons, triggering cell mobility (29).
Tissue regeneration requires maintenance of the cell phenotype. Phenotyping drift, a natural tendency in cultured rabbit and avian tenocytes, has also been demonstrated in human Achilles tendon cultures, as a tendency to shift from an elongated shape toward an ovoid shape. The phenomenon may be responsible for altered protein synthesis in vivo, increasing the ratio of type III:I collagen and inducing tendon pathologies (30). In normal human tendon cultures, one session of ESWT was proved to prevent the phenotyping drift, maintaining the normal tenocyte secretory profile of type I collagen (26).
The cellular compartment of the human tendon includes, besides tenocytes, a population of stem cells, human tendon-derived stem/progenitor cells (hTSPCs), characterized by several phenotypes and a potential for multilineage differentiation (osteoblasts, adipocytes, chondrocytes, tenocytes) (31,32). In cultures of both human normal and diseased tendon cells, ESWT was able to enhance hTSPC differentiation toward a tenocytic-like lineage, visible at day 21(33).
ECM. On normal tendons (ponies), ESWT application was followed by a rapid increase of degraded collagen at 3 h and a decrease at 6 weeks. The levels of glycosaminoglycans (GAG) at 3 h were unchanged and decreased at 6 weeks. However, no changes in hyaluronic acid, DNA and total collagen levels were noted either at 3 h or at 6 weeks. The results pointed to an anabolic response. The degraded collagen increase at 3 h may reflect an accelerated collagen turnover (slightly improbable for such a short duration) or direct physical damage. Overall, the anabolic response decreased at 6 weeks (21). On normal cultured human cells, ESWT increased the collagen content, predominantly type I (26).
Tendon regeneration is based on collagen synthesis, initially in the form of type III, followed by a maturation process and transformation in type I. Type I collagen appears like well organized, banded fibrils and provides high tensile strength. Type III collagen takes a woven pattern. Hydroxyproline is an aminoacid component of collagens accounting for ~13% and represents a reliable index for newly formed collagen. Pyridinoline is a crosslink residue of collagen and reflects its rate of degradation.
In the cut and sutured tendon model (rats), ESWT increased the hydroxyproline content, i.e., collagen synthesis, from 2- to 3-fold at days 3 and 9 compared to the control (10). The healing process under ESWT was followed by minimal or absent adhesions that did not disturb the configuration of the tendon.
In the model of collagenase-induced tendinopathy (rats), the optimal dose of ESWT reversed the increased concentrations of DNA, GAG and hydroxyproline to normal. Fiber bundles displayed a regular arrangement at 12 weeks. Higher doses elicited inhibitory effects on biochemical characteristics and an irregular array of fiber bundles (11). In the same model of tendinopathy (rabbits), the ESWT-treated tendons displayed higher levels of hydroxyproline and pyridinoline at 4 and 16 weeks, as an indication of intensive collagen synthesis and maturation (18).
At the nanostructural level, ESWT increased the collagen fibril diameter and the fibrillary adhesion force, accelerating healing (15).
In vitro, normal human cultured tissue displayed increased expression of scleraxis at 24 h, which decreased later, and an increased expression of type I collagen at day 4 that returned to normal levels at day 7(27). Human tendinopathic cultures were characterized by higher gene expression of type I collagen and scleraxis, a possible indication of the repair process; ESWT induced a reduction of their levels toward normal. The process may be interpreted as an impairment effect of ESWT on cell differentiation in favor of cell proliferation and migration to ensure regeneration (29).
Metalloproteases and cytokines. Cultures from normal tendon subjected to one session of ESWT exhibited increased levels of specific interleukins (IL-β1, IL-6 and IL-10), with a maximum at day one for IL-β1 and IL-6 and at day 2 for IL-10, and persistence of higher values at day 7. In this context, IL-β1 induced IL-6 release, which, in turn, promoted IL-10 increase, an anti-inflammatory cytokine to limit the inflammation process (27).
It was postulated that excreted MMPs and cytokines, particularly interleukins, from diseased tenocytes are breaking the ECM and induce tendon damage. Decreasing levels of these substances may reflect healing. Increased excretion of IL-6 and MMP-1, MMP-2 and MMP-13 was found in diseased cultured tenocytes; ESWT application reduced the higher levels of MMP-1, MMP-13 and IL-6 at 72 h, possibly contributing to a metabolic normalization. In normal cultured human tenocytes, only IL-1 levels were increased after ESWT, an event that necessitates further research (28).
Analysis of the microdialysate from the peritendinous space of human normal and tendinopathic patellar and Achilles tendons revealed that tendinopathy was associated with high levels of IL-6 and IL-8. ESWT application was followed by an increase of these two interleukins, in both normal and diseased tendons within the interval from 1 to 4 h. IL-β1 and IL-2 variations were not significant pre- and post-treatment. Pro-MMP-9, a latent form, increased after ESWT, without any change in the active form levels. The lack of significance of MMP activity changes raised the possibility of grouping individuals as responders and non-responders, with a proportion of responders of 60% in the general population (defined as exhibiting a minimum 5-fold increase in any MMP level at any point post-ESWT) (34).
The large spectrum of results of the two papers on IL and MMP levels deserves further investigation.
Lubricin production. Lubricin, a lubricating molecule, found in diarthrodial joints, is produced also by tenocytes enhancing the gliding of the tendons. ESWT increased lubricin expression and production in a dose-dependent manner in normal rat knee tendons and the effect was visible in the short term (4 days) (14).
Biomechanical properties of tissue after ESWT application. In the model of collagenase-induced tendinopathy, ESWT was followed by restoration of mechanical load-to-failure and stiffness of healing tendons, with a higher ultimate tensile load that increased from 4 to 16 weeks (11,18). The mean force required to rupture the tendon was higher 20 days after one session of ESWT (13).
ESWT, a method derived from urinary lithotripsy, represents a new therapeutic approach for tendon pathology, with encouraging results on pain and function. The biological mechanisms that sustain its clinical results have been extensively studied both in vitro and in vivo.
Research from the last 20 years indicated that there is a certain optimal dose for the maximum therapeutic effect for each studied species and for each studied tendon. Tendon- and species-specificity is a trait of the therapy. Energies that exceed the optimal value reduce cell viability and disorganize the ECM in a dose-dependent manner, with deleterious consequences on tendon structure.
Scientific papers agreed on several structural alterations that promote tendon healing. ESWT induces neovascularization in the tendinopathic tissue, with extensive capillary formation from the peritendinous structures. The inflammatory reaction that followed the collagenase-induced tendinopathy was accelerated by ESWT.
The therapy acts on both cellular and extracellular compartments of the tendon structure. Tenocyte proliferation and activation as blast-like forms with enhanced protein synthesis together with accelerated mobility reconstruct the normal structure. Accelerated collagen turnover, with type III collagen synthesis and subsequent type I collagen maturation, was documented in the ECM. As a result, biomechanical properties of the treated tendons improved significantly in comparison with non-treated tendinopathic structures. The mechanisms are outlined in Fig. 2.
An important clinical effect on plantar fasciitis, a structure that may be assimilated to a tendon, is worth mentioning. Biological studies on plantar fascia are currently lacking; the analgesic effect of ESWT is comparable to that of autologous blood-derived products in the medium-term (6 months). Among the forms of ESWT, the success rate is higher with medium- and high-intensity ESWT (35,36).
Future research will focus on the role of cytokines and metalloproteases that mediate, at a cellular level, the degenerative process and its evolution under ESWT application.