(1) Division of Surgery, Department of Clinical Sciences, School of Veterinary Medicine, Shiraz University, Shiraz, Iran
(2) Department of Pathology, School of Veterinary Medicine, Shiraz University, Shiraz, Iran
* Corresponding author Email: firstname.lastname@example.org
Surgical reconstruction of tendon injuries is challenging. Classic reconstructive techniques and tendon transplantation have some significant limitations and tissue engineering is a newer option. Despite significant development in tissue engineering technologies, the role of tissue engineering in tendon healing is still unclear. There are many tissue-engineered products that are commercially available in the market, but most of them have not passed animal and clinical studies, and the behaviour of host immune response to these types of products has not been investigated. Researchers have also focused on in vitro investigations and because of the differences between ex vivo and in vivo situations, translation of their results to clinical practice is of great concern and generally hard to follow. To increase the impact of tissue engineering in tendon healing, more information concerning the structure of tendons, their injuries, healing and host immune response together with the characteristics of biomaterials is needed to produce a more effective tissue-engineered product with the aim to substitute the classic reconstructive methods with the new tissue engineering approaches. This review was aimed to introduce the most important issues in the relationship between tissue engineering and tendon regenerative medicine with the hope that this information would be valuable for those who have concerns about tendon healing.
Tendon tissue engineering is still in its infancy and translation to clinics is accompanied by several important concerns including graft healing incorporation, host immune reaction and in vivo efficacy of the tissue-engineered graft.
Treatment of tendon injuries is challenging[1,2]. Classic surgical reconstructive methods have significant limitations with unclear outcomes, especially in tendon injuries having-large tendon deficits[3,4]. Tendon transplantation is the only available option when the injured tendons cannot be repaired with classic surgical techniques[2,3]. Natural grafts can be divided into three major groups including autografts, allografts and xenografts[2,5,6]. All of them have their own significant limitations. For autografts, these limitations include availability of the autograft of the same size, shape and physiological characteristics of the normal tissue at the recipient site, donor site morbidity and cosmetic concerns, need for another surgical procedure and the time-consuming nature of the second procedure. For allografts, these limitations are the availability of a healthy graft, risk of infection and transmission of fatal viral diseases (e.g. HIV or hepatitis), inefficiency of the graft to incorporate with the healing of the recipient site, rejection of the graft, re-injury due to the low biomechanical performance of the allografts and ethical concerns[2,6,7]. Limitations of the xenografts are similar to that of allografts, but their rejection rate is higher and their value in regenerative medicine because of insufficient experimental and clinical trials studies is questionable. Also, the presence of many unknown zoonotic diseases in animals, about which we have no information, could be another major concern[6,8,9].
For the above reasons, tissue engineering has been introduced to reduce these limitations and improve the outcome of incorporation of the tissue-engineered grafts and improve the healing processes of injured tendons[5,10,11]. In the last decade, tissue engineering has been improved and much advancement has been achieved[12,13]. Several types of scaffolds with different technologies have been introduced so that nowadays there are many commercially available tissue-engineered products in the market. However, most of these tissue-engineered products have not passed in vivo tests and most of the tissue-engineered researches have mainly focused on in vitro assays[6,12,14]. In addition, in vivo studies regarding the role of tissue-engineered products on regenerative medicine have not focused on mechanistic researches, and they have only observed the qualitative results with poor subjective data[5,6]. Therefore, there are many controversies between the results of the in vivo studies and there are also many differences between the sources, designing and preparation methods of the tissue-engineered grafts, which make their comparison hard to follow[5,7,15,16] Regardless of the efficacy of these tissue-engineered products, there are some great concerns that should be addressed in future studies.
This study was aimed to introduce the role of tissue engineering in tendon reconstructive surgery and regenerative medicine, and has focused on the general guidelines for those investigations that translate basic to clinical researches in the field of tendon tissue engineering.
Classification of tendon injuries
To design a suitable treatment strategy for tendon reconstruction, it is important to know the nature of the tendon injury and its correlation with the goals of tissue engineering. Tendon injuries are varied between degenerative tendinopathies and tendon defects[2,17] and could be grouped under three classifications. The first classification is the sharp ruptured/transected tendon injuries. These types of tendon injuries happen due to extrinsic traumatic forces such as a gunshot or vehicular trauma and/or during orthopaedic procedures when the tendon should be incised to extend the surgical approaches or when the bone fixative materials are implanted to stabilize fractured bones. In this type, the tendon ends are sharply transected and direct suturing is possible. In the second type, classified as the blunt tendon ruptures, the injuries have resulted from high stress forces such as those resulting from sport injuries. Degenerative tendinopathic conditions are other subtypes of these injuries. These types of injuries are divided into three subgroups including stress, strain and rupture injuries. Based on the nature of the injuries from partial to complete tear of all the collagen fibres, these types face different challenges. In complete blunt tendon ruptures, the edges of the ruptured tendon are not uniform and are fuzzy; thus, it is necessary to debride these tendon edges and facilitate suture placement. Accordingly, some parts of the ruptured tendon may have been lost and direct suturing is of limited value in this condition[2,3,9]. Usually, the defect area is not large and healing response is not the major concern in these types of injuries[4,20]. The last classification belongs to those injuries that have resulted from massive trauma, burning and severe degenerative changes, which result in the formation of large defects in the injured tendons. In these types of injuries, healing response of the injured tendon is of great concern[3,4,9].
Process of tendon healing
Tendon healing could be classically divided into three phases including inflammatory or exudative, fibroplasia or proliferation and remodelling phases. There is some overlapping between these phases and there are other transitional stages in each phase, which have never been clearly defined to date. Moreover, the quality and period of all these phases could be varied. Classic tendon healing could be expected to happen in those tendon injuries with low degrees of tissue loss, but characterization of tendon healing could differ based on the severity of tendon injury and the treatment modality. For example, the healing characteristics of tendon defects treated with autografts have some differences when compared with those treated with allografts. Perhaps these differences are larger when the tissue-engineered grafts are used to repair the defect site[12,17,23,24].
Inflammatory phase of tendon healing
Inflammation is a major start key of tendon healing, without which effective healing is not expected.
Immediately after tendon injury, the lag phase starts. The vascularity of the injured tendon is impaired so that ischaemia commences in the injured area. The tenocytes die and deliver the cytokines and proteases. This could be due to lysis of the cell membranes. These cytokines increase the permeability of the un-severed vascular structures of the injured area. In addition, injuries may result in vascular rupture so that the blood supply enters the injured area. By exposure of the platelets to the injured collagen fibres of the injured area, they aggregate with fibrin strands to form blood clot. Platelets then deliver cytokines, growth factors (e.g. platelet-derived growth factor) and inflammatory mediators to initiate the inflammatory stage of the wound healing in the injured area. By these mechanisms, some reversible changes take place in the endothelial lining of small venules and capillaries and the inflammatory cells including neutrophils, and then macrophages, lymphocytes and plasma cells enter the injured area. The fibrin clot acts as both a chemotactic medium and scaffold for the inflammatory cells so that they can migrate on the fibrin strands throughout the injured area.
Infiltration and debriding
At this stage (Fig. 1A), the inflammatory cells infiltrate the injured area and start to degrade the necrotic tissues and fibrin clot by phagocytosis and enzymatic lyses[23,25]. Neutrophils enter faster than other inflammatory cells (Fig. 1A) and play a major role in the presence of bacterial infection, for phagocytic activity and enzymatic lyses of debris, but do not play a crucial role in the following processes of wound healing. Macrophages enter the injured area later than neutrophils and have significant roles not only in phagocytosis and proteolysis of the necrotic tissue, fibrin, degraded collagen and elastin fibrils and foreign body material but they also deliver matrix metallo proteinases (MMPs), cytokines, growth factors and angiogenic mediators in the injured area and have a crucial role in further healing processes (Fig. 3A)[17,23]. MMP-9 and MMP-13 degrade the necrotic tissues and are useful in the inflammatory stage, and MMP-2, MMP-3 and MMP-14 change the inflammatory phase to fibroplasia and collaborate in tendon remodelling at later stages. Cytokines and other chemo-attractant mediators of macrophages chemotactically attract the mesenchymal cells, tenoblasts and endothelial cells into the injured area and by delivering growth factors, induce cell differentiation and proliferation[24,25]. Therefore, macrophages have a major role during different phases of tendon healing[26,27]. Growth factors such as vascular endothelial growth factors (VEGF), basic fibroblast growth factors, platelet-derived growth factors and tissue growth factor-β have significant roles in cell differentiation, cell proliferation and tissue maturation and collaborate in all phases of tendon healing with different mechanisms.
Histopathological changes in tendon healing. (A) Inflammatory stage is characterised by the presence of numerous inflammatory cells in the injured area. (B) Early to mid-fibroplasia stage of tendon healing. The tissue is hypercellular, but insufficient amount of collagen has been produced by the tenoblasts. The healing tissue is highly amorphous. (C) Late fibroplasia. The collagen mass density has increased. The blood vessels have large calibre and are mature but the tissue is amorphous. (D) Early remodelling. The tissue is well aligned compared to that of (C). The cellularity has decreased and most of the blood vessels have been degenerated. (E) Maturation to consolidation stage of remodelling. The collagen fibres are highly mature and the cellularity has decreased compared to (E). (F) Normal tendon (H&E, Scale bar A–F: 120 μm).
In a routine tendon injury, fibroblasts are the dominant cells about five days post-injury and this gradually changes the inflammatory phase to fibroplasia or proliferative stage. This stage can be divided into three sub-stages including early fibroplasia or stage of fibrous response, mid-fibroplasia or granulation tissue stage and late fibroplasia or amorphous collagenous stage[18,28].
In this stage, the peritendinous fibroblasts migrate from the injured tendon Sheath to the injured area. This is the extrinsic mechanism of tendon healing[17,25]. Intra-tendinous tenoblasts also infiltrate the injured area from the uninjured parts of the severed tendon. This is the intrinsic mechanism of tendon healing. The function of tenocytes may vary based on their origin. Cells from the tendon sheath produce less collagen and glycosaminoglycans than epitenon and endotenon cells. However, fibroblasts from the flexor tendon sheath proliferate more rapidly. Intrinsic healing results in a better biomechanical performance and fewer complications; in particular, a normal gliding mechanism within the tendon sheath is preserved. In extrinsic healing, scar tissue results in adhesion formation, which disrupts tendon gliding.
The undifferentiated mesenchymal cells also migrate from the peritendinous tissues and fill the injured area. These cells are activated and differentiated into fibroblasts and endothelial cells by the local growth factors. The final part of this stage is characterized by the proliferation of fibroblasts and endothelial cells.
Granulation tissue stage
Tendon injury is associated with vascular damage, diminished blood supply and hypoxia, and the main consequences of hypoxia are further cell degeneration and necrosis. Hypoxia is the key element which activates macrophages to release angiogenic factors and initiate angiogenesis in the injured area and VEGF has a major role in this regard. The endothelial cells aggregate in the granulation tissue, proliferate and regenerate blood vessels (Fig. 1B)[24,25]. The newly formed vessels are primarily obstructed, but some of them connect to each other, re-canalize and some of them are able to be connected to the main circulation, but many of them die because of insufficient nutrition and hypoxia. At this time, there is a reduction in the population of the neutrophils and the acute inflammation changes to chronic inflammation, in which macrophages, lymphocytes and plasma cells are the major inflammatory cells[22,27,28].
Proliferating fibroblasts produce fibronectin, collagen, elastin and polysulphated glycosaminoglycans such as hyaluronic acid, dermatan sulphate, chondroitin sulphate, heparan sulphate, heparin and glucosamine[23,30]. It has been shown that glycosaminoglycans have important roles in tendon healing, and they act as a scaffold for collagen deposition and collaborate in collagen fibril formation and differentiation[23,25,28]. Hyaluronic acid has also been shown to have a significant role in subsiding the inflammatory stage of wound healing. Glycosaminoglycans and collagen type III, which are deposited by immature and mature fibroblast, form the building block of the initial matrix architecture of the newly regenerated tissue. At this stage, the healing tissue has low echogenicity and homogenicity at ultra-sonographical examination, uniformly small-sized unimodal collagen fibrils at the ultra-structural level, minimum ultimate strength, yield strength, maximum stress and modulus of elasticity at biomechanical testing (Figs 1B, 2A and 3B).
Scanning electron microscopy of different stages of tendon healing. (A) At early fibroplasia, a low density of highly immature collagen is seen. (B) At late fibroplasias, the collagen density has increased. (C) At the early remodelling stage, the collagen fibrils tend to assemble to collagen fibres. (D) At the maturation stage of remodelling, the collagen fibres are completely differentiated from aggregation of the matured collagen fibrils. (E) At the consolidation stage of the remodelling phase, the collagen fibres are denser and more aligned. (F) Scanning morphology of the intact collagen bundle with its highly dense and align collagen fibres.
Transmission electron microscopy of different stages of tendon healing. (A) Inflammatory stage: a macrophage is seen with its characteristic cytoplasm and enzymatic granules. The collagen fibrils of the implant are lysed and new fibrils are produced in the tissue. (B) Early fibroplasia. The density of the collagen fibrils is low and they are randomly distributed in different directions. (C) Late fibroplasia: the density of the collagen fibrils has increased and they are more aligned. They are all unimodal and have small diameter. (D) Early remodelling: the diameter and density of the collagen fibrils have increased and they are highly aligned so that they have been only sectioned transversely in one ultra-thin section. (E) Mid-remodelling stage: the collagen fibrils are bimodal so that some of them are in the range of 32–64 nm and others in the range of 65–103 nm. (F) Normal uninjured tendon. The collagen fibrils are highly aligned, they are multimodal and their diameters vary from 32 to 270 nm (scale bar A: 1.6 μm; B:256 nm; C–F: 192 nm).
Amorphous collagenous stage
Fibroblasts (tenoblasts) mainly start to produce large amounts of type III collagen fibrils; however, the collagen/glycosaminoglycans ratio, which is called collagen density, is low at this stage[23,25]. The metabolic activity of the tenoblasts is high, the proportion of nucleus/cytoplasm is elevated and the transverse diameter of the newly regenerated collagen fibrils is still small, ranging from 32 to 64 nm (Fig. 3C). Such a granulation tissue is hyper-cellular, disorganized and highly vascular. The collagen fibres are haphazardly distributed and there is no correlation in the direction of tenoblasts, fibrous connective tissue and blood vessels at this stage. Therefore, glycosaminoglycan and collagen content of the healing tissue gradually increase up to three weeks post-injury, but then they start to decline and reach their steady state at about five weeks post-injury (Fig. 2B)[23,30].
The remodelling or maturation phase of tendon healing can be divided into three different sub-stages, including early remodelling or alignment stage, mid-remodelling or maturation stage and late remodelling or consolidation stage[1,17].
At this stage, the inflammation has subsided and the amorphous tissue has filled the injured area so that the continuity of the injured tendon is established. This provides an opportunity for the patient to use their injured limb so that the weight-bearing capacity and physical activity of the affected limb increases and weightbearing forces can be transmitted between the bone–tendon–muscle complex[23,26]. These stress forces align the collagen fibres and blood vessels along the stress line (Fig. 3D). Accordingly, cellularity and transverse diameter of the injured area decrease (Fig. 1D). At this stage, the density of the collagen fibrils increases and their alignment improve so that they are mainly oriented uni-directionally along the longitudinal axis of the tendon (Fig. 3D)[22,23]. Accordingly, the ultra-sonographical homogenicity increases and some improvement in the biomechanical properties of the healing tissue is seen.
Due to the decrease in the cellularity and hydration of the tissue and the presence of a large amount of aligned collagen fibrils, the metabolism of the new tissue decreases and the blood vessels degenerate and resorb (Fig. 2D)[21,31]. Most of the newly regenerated blood vessels disappear and only a few blood vessels, having large calibres, can be seen. The weightbearing capacity and physical activity of the patient increases and the immature, but aligned collagen fibrils start to aggregate and differentiate into larger and more mature collagen fibrils (Fig. 3E)[24,26]. The polysulphated glycosaminoglycans (e.g. chondroitin sulphate, dermatan sulphate and keratin sulphate) collaborate in the maturation and differentiation of collagen fibrils. This increases the transverse diameter of the highly aligned collagen fibrils so that their diameter increases from 32–64 to 65–103 nm (Fig. 3E). However, their diameter is still much smaller than that of the normal tendon (e.g. 250 nm) (Fig. 3F)[25,31]. Upon maturation of the collagen fibrils, type III collagen of the healing tissue decreases and is replaced with type I collagen[21,30]. At this time, the tenoblasts mature completely and they transform into metabolically inactive tenocytes that are histologically characterized by longitudinal cigar-shaped nuclei with a much higher proportion of nucleus/cytoplasm than tenoblasts. The above changes accelerate the ultra-sonographical echogenicity and biomechanical properties of the healing tissue so that in a normal uncomplicated healing more than 50% of the normal contralateral tendon’s characteristics should be achieved at this stage.
This is the slowest stage of tendon healing and could continue for years or even to the end of life. At this stage, the matured collagen fibrils aggregate together and produce larger collagen fibrils (>103 nm). There is also improvement in the quality of cross-linking of these collagen fibrils. Moreover, the collagen fibrils are aggregated and covered by endotenon, epitenon and paratenon so that the collagen fibres, fibre bundles and fascicles are formed (Fig. 2D, E)[23,26]. When the fascicles are formed, it can be suggested that tendon healing is in its final stage. By this time, the biomechanical characteristics of the new tendon are almost comparable to those of the normal tendon, but in a large injury it may never reach its normal value. At this stage, the histopathological, ultra-sonographical and ultra-structural morphologies of the repaired tendon are also inferior to that of the intact tendon, and the regenerated tissue is still hypercellular, the diameter of the collagen fibrils is far behind that of the normal contralateral tendon, and the hierarchical organization of the tendon from collagen fibril level to fascicle is not properly developed yet (Fig. 1E vs. 1F)[18,28].
The authors have referenced some of their own studies in this review. These referenced studies have been conducted in accordance with the Declaration of Helsinki (1964) and the protocols of these studies have been approved by the relevant ethics committees related to the institution in which they were performed. Animal care was in accordance with the institution guidelines.
Limitations of tendon healing based on the type of tendon injuries
Tendon healing is a complicated and targeted process but it has some limitations[20,32]. Generally, the major limitation of tendon healing is development of peritendinous adhesions[18,22,29]. Incidental injuries or surgical operations may result in disruption of the paratenon or tenosynovium. Also, if this structure is preserved, it usually lyses by the chemical activity of the MMPs and other degrading enzymes which are secreted during the inflammatory phase of tendon healing. Lack of effective mechanisms to guide proliferation and direction of the fibroblasts and collagen fibres at the fibroplasia stage results in proliferation of fibroblasts in a haphazard fashion[20,29,32]. This produces a strong granulation tissue in the periphery of the tendon proper and between the epitenon and paratenon and paratenon and the surrounding fascia that inhibits the movement of the healing tendon in its normal physiological space. By inhibiting the physiological movement of the healing tendon, stress is not transmitted into the healing tissue and this impairs the alignment and maturation stages of tendon healing; thus, the healed tendon would not have its normal physiological role and function. This impairs the function of patients so that its outcome is not acceptable and repeated surgery is required.
Actually all types of tendon injuries are subjected to the development of peritendinous adhesions, but when tendon injury is more severe and the size of the defect area is larger, development of peritendinous adhesion is more aggressive and the outcome is much poor[2,29]. In such cases with significant tissue loss, there is also another limitation that would have a major role in the functional impairment. When the tendon defect is larger than that to be directly repaired by primary surgical repair, the amount of peritendinous adhesion substantially increases because there is no scaffold for the healing cells to guide them to proliferate along the stress line of the tendon. Accordingly, these cells proliferate in different directions. In such large tendon defects, the healing capacity decreases because the fibroblasts migrate in the peritendinous fascia and hinder muscle insertion, resulting in muscle fibrosis. Therefore, the population of the migrated fibroblasts in the defect area is reduced, which is followed by a reduction in the amount of collagen production. Continuity of the defect area in such a tendon injury may not be established[5,17]. Therefore, large tendon deficits have more significant limitations, including the development of peritendinous adhesion, muscle atrophy, muscle fibrosis and improper healing response. These limitations should be considered when a proper tissue engineering approach is to be designed.
Role of tissue engineering in tendon regenerative medicine
In the last decade, tissue engineering has seen much advancement and several manufacturing technologies and treatment modalities have been introduced to reduce limitations of tendon healing and to improve the healing response[5,12,33]. Basically, tissue engineering consists of three different parts, including scaffolds, stem cells and healing promotive factors[5,34]. A major advancement in tendon tissue engineering is related to the scaffolds. The first step in tendon regenerative medicine is to design a suitable environment for cell migration, proliferation and a navigator for tissue alignment, remodelling and maturation. Therefore, there are several factors that have an impact on the effectiveness of the scaffold in this regard including the molecule (s) from which the scaffold is manufactured (basic material of the scaffold), architecture of the scaffold, diameter and orientation of the fibres, their biological characteristics and the amount of free spaces and pore size[5,16]. There are also a numbers of other issues that should be considered in manufacturing a scaffold; for example, a suitable scaffold for tendon tissue engineering should be cytocompatible in vitro and biocompatible and biodegradable in vivo[5,16,33,36,37]. Unfortunately, most of the exogenous-based biomaterials for tendon repair have serious limitations, such as lower capacity for inducing cell proliferation and differentiation (tenoinductivity), poor biocompatibility and remodelling potentials (tenoconductivity)[16,33,38]. To date, no manufactured scaffold has passed all the above issues both in vitro and in vivo and this is the greatest concern. Here, we briefly discussed each of the above characteristics.
Basic material of the scaffold
Several materials have been used so far to produce tissue-engineered scaffolds; however, few of them have been effective in tendon tissue engineering and regenerative medicine[2,40]. Generally, they can be divided into three major groups including biological (natural), synthetic and hybrid materials[12,41,42]. Biological materials such as collagen, elastin, gelatin, chitosan, albumin, alginate, fibrin and chondroitin sulphate have been shown to be effective in tendon healing[36,40,43,44]. Actually, these materials are biocompatible and biodegradable. Their toxicity is low and has some beneficial biological role after implantation in the injured area.
Mature tendons are composed of more than 90% type I collagen. This molecule has an excellent physical property and in vivo activity, and its production is not expensive. It can be easily formed into any shape and architecture, and is equipped with most of the healing promotive factors. Elastin is also present in tendons in a much less proportion (about 1%) and its major application in tissue engineering is to produce vascular scaffolds[12,17]. Chitosan is a natural polysaccharide obtained from insects. Nowadays, this molecule is the focus of many research programmes, and it has been shown that this molecule has excellent biological activity. All these materials are biodegradable and biocompatible.
There are also some non-biodegradable biological materials such as silk and carbon fibres. The usage of carbon fibre did not continue because of its high toxic effect and serious inflammatory reactions. However, investigations into silk are still in progress, but most of the studies in this regard are in vitro investigations that have low value in translational medicine[12,34]. Synthetic materials such as polycaprolactone (absorbable), polydioxanone (absorbable), polygalactin 910 (absorbable) and nylon (non-absorbable) are other options with invaluable results[42,45,46]. Many researches have focused on the in vitro characteristics of such materials and those who investigated their in vivo efficacy have not suggested their clinical application and claim that they induce exaggerated inflammatory reactions, are highly cytotoxic and have poor outcome because of their high rates of rejection after implantation[2,12,41,46]. Their major application is in vascular tissue engineering and body wall defects, and most of them are produced by petroleum material[12,41,46]. In fact, their first application in medical sciences was in the surgical field as suture materials. Their outcome was excellent compared to biological materials when their usage was limited to surgical sutures. Probably, their merit as a surgical suture can be attributed to their excellent biomechanical properties. Also, these materials as surgical suture do not have considerable toxicity because the amount of foreign material is considerably less than the scaffolds constructed from these materials. Therefore, due to the excellent biomechanical and physical characteristics of these materials, they were never deleted from the field of tissue engineering, but their proportion in tissue-engineered scaffolds has greatly reduced. Therefore, they have been combined with biological materials to decrease their limitations. However, in vivo studies regarding their efficacy in tendon regenerative medicine are rare[12,46].
Dimensions of the scaffolds
Regardless of the different technologies introduced, most of the tissue-engineered scaffolds designed for tendon tissue engineering are bidimensional[14,48]. These types of scaffolds are commercially available in the market and are produced as films, patches or membranes (Fig. 4A)[2,12,48]. Their major application in clinical practice is to inhibit the development of peritendinous adhesion during tendon healing (Fig. 4B). For this purpose, the injured area is surgically repaired by tension sutures and these scaffolds are wrapped around the injured area of the tendon with a minimum gap (Fig. 4C)[9,48]. In tendon injuries with significant tissue loss, as discussed in this article, the defect area should be repaired by grafts. In this regard, tridimensional scaffolds are more suitable (Fig. 4G, I and J)[5,50]. Unfortunately, tissue engineering is not capable of producing these types of scaffolds, and if produced, because of lack of in vivo studies, their effectiveness in tendon regenerative medicine is unclear[12,49].
(A) Bidimensional scaffold has been produced by tissue engineering technology. (B,C) This scaffold has been wrapped around the injured area of tendon to reduce development of peritendinous adhesion. (D,E) Bidimensional scaffold has been used for augmentation of small tendon defects. (F) Scanning electron micrograph from the surface of the synthetic tissue engineered scaffold. (G) Synthetic tridimensional tissue-engineered scaffold for tendon and ligament repair. (H) Synthetic tridimensional porous scaffold for bone and cartilage repair (I) Collagen-based tridimensional tissue-engineered scaffold has been used in repairing a large tendon defect. (J) Polydioxanone bidimensional scaffold has been wrapped around the tissue-engineered collagen-based bioimplant and has been implanted in a large tendon defect.
Architecture of the scaffolds
Several technologies have been introduced in this regard; however, tissue engineering is still in progress and it is expected that newer designs with more enhanced architectures would be manufactured in the near future[5,49]. The classic tissue-engineered products are the allo- or xeno-geneic-based grafts that are processed only by acellularization technologies[15,16,51,52]. Thus, their architecture could not be re-designed in a more effective manner after implantation in the injured tendons. Despite acellularization of such grafts, their rejection rate is high and therefore, they cannot be considered as tissue grafts[5,7].
Newer approaches such as gel systems, porous systems and electrospinning have been introduced[2,45,53,54]. Each of these technologies has its own purpose, but the gel system is used to produce a tridimensional environment for cell culturing purposes[5,40,53]. The in vivo application of the gel system has significant limitations, including a low biomechanical performance when used as a graft in the injured area[33,50,55]. Also, the absorption rate is fast so that few days after implantation, the architecture is completely depleted by the inflammatory cells and mediators[2,12]. Moreover, the orientation of their polymerized fibres is randomized and is not suitable for the purposes of tendon tissue engineering[2,50,56].
Porous scaffolds are used to produce both bi- and tridimensional scaffolds with different purposes (Fig. 4H). Their major application is in bone and cartilage regenerative medicine. By this technology, several pores were introduced in the implant with the aim of filling the newly regenerated tissue, after implantation. Due to the placement and randomized distribution of these pores at different sides of the scaffold, the orientation of the regenerative tissue is not aligned; this irregularity is not desirable in tendon tissue engineering as proper alignment of the newly regenerated tissue is a must (Fig. 4H). In these types of scaffolds, alignment of the newly regenerated tissue is not a major concern because of the nature of the tissue that should be reconstructed. Electrospinning is another but highly expensive and timeconsuming technological approach[2,42,54]. By this technology, it is possible to align the polymerised fibres of the scaffold (Fig. 4F)[2,44]. Also, the diameter of the polymerised fibres can be designed to produce different ranges of fibres from nanometric to micrometric scales.
Other physical characteristics of the scaffolds
Regardless of the above factors that are necessary for consideration, there are some other issues that should be addressed. For example, the diameter and orientation of the polymerized fibres of the scaffold are important (Fig. 4F). Both these characteristics can be controlled and ordered when the new design is approached[40,44]. The tendon-engineered scaffolds should have fibre diameters varying between nano-scale and micro-scale[57,58]. It has been shown that controlling the scaffold fibre diameter is critical in the design of scaffolds for functional and guided connective tissue repair, and provides new insights into the role of matrix parameters in guiding soft tissue healing[57,58].
A tendon’s normal architecture is composed of collagen fibrils with nanometric diameters and collagen fibres and fibre bundles with different micrometric diameters. Therefore, a suitable scaffold for tendon tissue engineering should consist of both micro and nanoscale moderately to highly aligned fibres to be effective in guiding both the nano and microstructure of the newly regenerated tissue after implantation[14,40,44]. Pore size is another important factor. Tendon is not similar to cartilage or bone architecturally and for this reason the size of the pores should be smaller than the scaffolds that are constructed for cartilage and bone tissue engineering. In addition, the number of pores should be fewer. Their orientation is also the most important factor. Actually, in a well-designed tridimensional tendon scaffold, if there are six sides on the scaffold, only two sides should be porous in nature. Their position should be on the proximal and distal aspects of the implant just between the severed tendon edges. The other four sides at the periphery should be non-porous to reduce the amount of invasion of the peritendinous fibroblasts into the implant from the periphery.
Water uptake and water delivery of the scaffold is another important characteristic that should be addressed. A well-designed implant should quickly absorb liquids but deliver them slowly. These characteristics are mainly dependent on the above mentioned factors. With these characteristics, the implant is able to absorb the cellular structures and healing mediators in its architecture and maintain them for a long time. These characteristics are also used for drug delivery. The specific growth promotive factors could be assembled within the scaffold and delivered in a suitable time after implantation of the graft with the aim of increasing the efficiency of the treatment modality.
Biological characteristics of the scaffolds
Biological characteristics are a major concern in the field of tendon tissue engineering[10,33]. At least five possible biological responses have been suggested after implantation of extra cellular matrixes including (i) ECM non-incorporating responses: (a) encapsulation; (b) rejection; and (ii) ECM-incorporating responses: (a) resorption; (b) integration with progressive degradation; (c) adoption and adaptation[7,8,37,44].
Several attempts have been made to improve the biological behaviour of the scaffolds with the aim of improving their biocompatibility, biodegradability and bioefficacy and decreasing the rejection rate[7,39,44,51,59]. Acellularization or decellularization is the first step in this issue, especially in those biologically based tissue-engineered grafts obtained from allo- or xenografts[7,9,33,51]. Selecting a proper material is another approach[34,52,60]. For example, collagen has been shown to have excellent biodegradability and biocompatibility[2,51,61], on the other hand, synthetic materials have low biocompatibility[12,33,44]; thus, the amount of biologically based materials should be increased and the percentage of synthetic materials should be decreased when designing a hybrid scaffold. Sterilization is another factor that should be addressed. By removing all cellular and microbiological structures, the immune response could be reduced in order to increase the chance of graft incorporation in the healing process[15,33,34].
Increasing the number of specified molecules in the architecture of the scaffold could reduce the limitation of each molecule[15,59]. The best-suited scaffold for tendon tissue engineering should have the following biological behaviour: it should not acutely or chronically be rejected by the host immune response and should not be encapsulated by fibrous connective tissue. But it is imperative to mildly initiate the immune reaction and modulate inflammation because this can increase the healing rate[2,33]. This incorporating behaviour should not be accompanied by acute resorption of the scaffold. The best biological behaviour of a suitable scaffold is to be accepted as part of a new tendon. However, to date, this behaviour has not been shown for soft tissue scaffolds, but by designing suitable modalities, it may be possible to preserve some parts of the scaffold in a manner that is acceptable as part of the new tendon[33,35]. Integration with progressive degradation is also well accepted because the graft incorporates in tendon healing and collaborates in different stages of the healing process.
Optimization of the scaffolds as the final step
Other aspects of tissue engineering have been developed together with the development of scaffolds including stem cells and healing promotive factors[13,62]. A scaffold provides a suitable substrate for cell attachment, cell proliferation, differentiated function and cell migration. Scaffold matrices can be used to achieve drug delivery with high loading and efficiency to specific sites. Several investigations equipped tissue-engineered scaffolds with different types of stem cells and healing promotive factors to increase the efficiency of tissue-engineered grafts[13,33,43]. The transplanted cellular structures have been shown to significantly improve the healing quality of the repaired tissue. There are also many researches that have shown the efficacy of different healing promotive factors[33,60]. By equipping the tissue-engineered scaffolds with different types of healing promotive factors, the healing response can be controlled, there by making it more efficient in increasing the quality of the repaired tissue. Glycosaminoglycans have been the main focus in this regard. Hyaluronic acid is one of these agents. This glycosaminoglycan has been shown to modulate the inflammatory phase of tendon healing and increases the rate of healing process[18,23]. It also increases the diameter of the newly regenerated collagen fibrils and improves the alignment of the newly regenerated tendon, in vivo. Growth factors are another option[17,33,60]; for example, basic fibroblast growth factor has been shown to have an impact on the cell proliferation, maturation, collagen production and remodelling phase of tendon healing[1,19,24].
Tarantula cubensis is another novel agent that has been shown to decrease the necrotic tissues in the injured area, modulate the inflammatory reaction and reduce the amount of peritendinous adhesions[22,28]. Platelet-rich plasma is another healing promotive agent that initiates cell proliferation and tissue maturation[20,23,32]. These beneficial effects have been suggested to be due to the growth factors that are delivered from the platelets[62,64]. It seems that by equipping the tissue-engineered scaffolds with cells and healing promotive agents, the tissue-engineered graft could be optimized to open new insights in the field of tendon reconstructive surgery and regenerative medicine.
As it has been stated in this article, most of the researches in the area of tendon tissue engineering have not focused on the in vivo conditions and for this reason the real efficacy of such tissue-engineered products and their biological behaviour is unclear. Such concern particularly increases when these types of tissue-engineered products are introduced as tissue substitutes in clinics without passing the scientific requirements and approval. In addition, the significance of in vitro studies in translational medicine is also questionable. For example, it is not clear as to how the cultured stem cells on tissue-engineered scaffolds would be effective when the host-scaffold interaction and immune response to implantation of the scaffold is not clear and not defined. It is not clear how these cells remain alive in the inflammatory phase of tendon healing when large amounts of degrading enzymes are delivered by the inflammatory cells. Therefore, future investigations should consider the in vivo tests and the in vivo researches should focus on the mechanistic approaches and not just on the observations of the final outcome. This includes designing several observational time points after implantation of tissue-engineered grafts using different observational methods to comprehensively discuss about the mechanism of the host-immune response to the graft implantation and host toleration to the foreign body exposure. For observation of the final outcome, it is also suggested to test the repaired tissue with different methodologies including both physical and chemical characteristics of the repaired tissue together with the morphological analysis at different levels (e.g. histopathology, ultra-structure)[65,66]
Despite significant advancement in tissue engineering technologies and producing many different tissue-engineered products, tendon tissue engineering is still in its infancy and translation of the present investigations to clinics is accompanied by several important concerns including graft healing incorporation, host immune reaction and in vivo efficacy of the tissue-engineered graft. There are several blind points in this regard that should be addressed. Tissue engineering is going to be applicable in the near future; however, we should understand what the right purpose is before designing the new approaches. Knowledge about the nature of tendon injuries, healing process and tissue engineering is needed when tissue engineering is selected as the alternative approach. In addition, the researcher should have enough knowledge of the normal structural hierarchy, functional, biochemical and mechanical performance of the specific normal tendon to be able to simulate an appropriate applicable scaffold.
All authors contributed to the conception, design, and preparation of the manuscript, as well as read and approved the final manuscript.
All authors abide by the Association for Medical Ethics (AME) ethical rules of disclosure.