Therapeutic Advances in Tendinopathy Quantiﬁed Microscopically Using Bonar Score, with a Special Reference to PRP Therapy—A Systematic Review of Experimental Studies

: (1) Background: The Bonar scoring system serves in the microscopic evaluation of tendon pathology. However, it can be easily adapted to investigate decreasing degeneration after treatment and quantify the healing progress. We believe that there is an actual need for a connection between clinical observations and tissue alterations arising during the treatment process, to gain superior functional outcomes. Herein, we perform a systematic review of the Bonar score’s application in the histopathological assessment of therapeutic advances in tendinopathy, with special reference to PRP therapy. (2) Methods: A systematic review was performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The searching strategy was conducted across major databases: PubMed, Cochrane Central, ScienceDirect, SciELO, and Web of Science. The risk-of-bias assessment was made using the Cochrane Collaboration’s Risk of Bias Tool and SYRCLE. (3) Results: The searching strategy produced 807 articles, and after selection, 22 studies were included. We collected 21 animal studies (n = 472) and 1 human study (n = 45). Three types of tendons were taken into account: 14 Achilles tendons, 7 supraspinatus tendons (SST), and in 1 case, Achilles and patellar tendons simultaneously. A variety of therapeutic methods were used—from intra-tendinous substance injections to surgical procedures or mechanical stimuli—but platelet-rich plasma (PRP) therapy dominated among them and was present in six studies. Most of the collected studies included an assessment of the tendons’ histopathology based on the classical Bonar score (with four variables and one observer). The staining protocol was based on the hematoxylin and eosin technique. An evaluation of therapeutic effects showed 15 positive results, 6 negative results, and 1 neutral result of treatments. (4) Conclusions: To understand the tendinopathy phenomenon, a link between histopathology and clinical observations in chronic tendon disorders is required due to the possibility of functional outcome improvements. The Bonar scoring system is well established in tendon pathology assessment and could also be adopted to assess therapeutic results in tendon disorders. Studies that included the PRP application showed Bonar-scoring-system-based evidence of superior tendinous tissue healing related to improved clinical results. composition, tenocyte morphology, and expansion of capillaries.


Introduction
Acute injuries of the tendons are among the most common in the human body [1]. The injured tendon healing process consists of three main phases, which are essential for soft tissue healing. However, in the case of an incomplete healing process, degeneration of the tissue may occur [2]. The tendinopathy process is characterized by the uncontrolled production of both collagenous and non-collagenous extracellular matrix (ECM), apoptosis in the local fibroblast population, and chaotic invasion of newly formed capillaries [2][3][4][5][6][7][8]. These histopathological alterations are reflected in decreased mechanical durability of the tendon and clinically manifest as tendinopathy. The healing of the injured tendon is conducted by a complex network of cytokines and mediators, which dynamically interact during regeneration of the tendinous tissue. This can be used in the therapy of tendon disorders [2,9]. The presence of specific types of growth factors in adequate time is crucial to fully recovering the tissue [2,[9][10][11]. Many agents, such as mesenchymal stem cells (MSCs), growth factors, and hemostatic agents, have been studied to improve the tendon healing process [4,[12][13][14][15].
It is currently believed that five main growth factors play a crucial role in the tendon healing process: insulin-like growth factor-1 (IGF-1), transforming growth factor-β (TGFβ), vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and basic fibroblast growth factor (bFGF). Further, all of them are included in plateletrich plasma (PRP). This is an autologous plasma fraction with a platelet-rich cellular component containing growth factors in alpha granules [16]. The efficacy of platelet concentrates in promoting wound healing and tissue regeneration is at the center of a recent academic debate [17]. Some authors even tried this therapy in osteonecrosis; however, the results are not sufficient to establish this method as a gold standard [18]. Moreover, the regeneration process demands cells to supply the local fibroblast population responsible for extracellular matrix production [2]. Recent studies revealed the positive role of MSCs, which are the progenitors for the fibroblast population, in tendon injury therapy [19][20][21]. Moreover, Vigano et al. suggested that autologous microfragmented adipose tissue reduces the inflammatory and catabolic markers in tenocytes [22]. Other authors isolated the extracellular vesicles from human adipose-derived MSCs to develop new therapeutic approaches in osteoarthritis [23].
The most common and well-established system to quantify the pathological changes in tendinous tissue is the Bonar scale [24,25]. The Bonar scoring system is based on four main variables: tenocyte morphology, ground substance accumulation, the extent of the neovascularization process, and the disruption of the collagen bundle architecture. Each variable is assessed on the scale of 0 to 3 points, where 0 indicates a normal appearance of tendinous tissue and 3 indicates the most severe detectable pathology [24]. The Bonar scoring system serves in the microscopic evaluation of the various regions of tendon pathology, but on the other side, it can be easily used reversely to investigate the decreasing degeneration after effective treatment and quantify the healing progress [24,[26][27][28][29][30]. The link between the histopathological scale and therapeutic advances of various treatment methods in chronic tendon disorders has been evaluated in only a few studies. Despite the numerous reports about improved clinical conditions after tendon treatment based on functional scales, surprisingly, the microscopic state of the tissue was not assessed [31,32]. We believe that there is a real need for a connection between clinical observations and tissue alterations arising during the treatment, to gain superior functional outcomes.
This study was aimed to perform a comprehensive, systematic review of the Bonar score's application in the histopathological assessment of therapeutic advances in tendinopathy, focusing on PRP therapy.

Search Strategy
A systematic review of the collected literature was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines  [33]. To identify all of the essential studies that reported relevant information and data concerning therapy in tendon disorders quantified using the Bonar scale, an extensive search of the major and significant electronic databases (PubMed, Cochrane Central, ScienceDirect, SciELO, and Web of Science) was performed. A systematic investigation was conducted in February 2021, using combinations of the key terms "Bonar tendon" OR "Bonar tendinopathy" OR "Bonar score", with no limits regarding the year of publication. Moreover, an additional intensive search through the references of all identified studies was conducted.

Eligibility Assessment
The search in databases was done by two authors (J.Z., M.G.), independently. Next, three independent reviewers (J.Z., D.S., M.G.) screened all the papers identified for a title, abstract, and full text concerning the application of the Bonar score in the evaluation of tendinopathy treatment. Preclinical and clinical human and animal studies in English were evaluated and analyzed in this systematic review. Non-English-language studies, case studies, reviews, letters to editors, conference abstracts, or studies containing incomplete or irrelevant data were not eligible for inclusion. We also excluded papers without clearly quantified tendon alterations using the Bonar score. Moreover, studies with no specific drugs or treatment methods were not taken into consideration. Additionally, we excluded Bonar score application in structures other than tendons, such as menisci. The senior author and expert in evidence-based medicine (M.G.) made the final decision if there was disagreement among the authors.

Data Extraction
Three independent reviewers (J.Z., M.G., D.S.) extracted the initially screened and relevant data, including the year of the study, country, type of the study, number of subjects, region of tendon, control group inclusion, applied Bonar score and its modifications, number of investigators, area of tendon investigation, staining methods, mean Bonar scores in the treated and control groups, and the applied therapeutic methods and their effects.

Risk-of-Bias Assessment
The risk-of-bias assessment was done using the two different scoring tools. The Cochrane Collaboration's Risk of Bias Tool was employed in this study for quality appraisal of the included human study paper [25]. Risk of bias was assigned to the following domains as 'low', 'high', or 'unclear': sequence generation/allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias), and other sources of bias. Moreover, we used the SYRCLE (Systematic Review Centre for Laboratory animal Experimentation) Risk of Bias Tool in this study for quality appraisal of the included animal studies. The quality of papers was assessed independently by two reviewers, with agreement.

Results
The results of the literature search are presented in the PRISMA flow diagram (Figure 1). The general characteristics and demographic data are presented in Table 1.

Results
The results of the literature search are presented in the PRISMA flow diagram ( Figure  1). The general characteristics and demographic data are presented in Table 1.

Study Selection
This systematic review includes a total of 22 papers with a number of studied subjects of n = 517. We collected 21 animal studies (n = 472) and 1 human study (n = 45), the oldest of which was from 2007 and the most recent from 2020. The included studies originated from various countries, and the most common was Turkey (10 papers). Three types of tendons were taken into account: 14 Achilles tendons, 7 supraspinatus tendons (SST), and in 1 case, Achilles and patellar tendons simultaneously. The control group was included in all studies. There was a variety of therapeutic methods, from intra-tendinous substance injections to surgical procedure or mechanical stimuli, but PRP therapy dominated among them and was present in six studies. In Figures 2 and 3 we present the risk-of-bias assessment using graphs [54,55]. of which was from 2007 and the most recent from 2020. The included studies originated from various countries, and the most common was Turkey (10 papers). Three types of tendons were taken into account: 14 Achilles tendons, 7 supraspinatus tendons (SST), and in 1 case, Achilles and patellar tendons simultaneously. The control group was included in all studies. There was a variety of therapeutic methods, from intra-tendinous substance injections to surgical procedure or mechanical stimuli, but PRP therapy dominated among them and was present in six studies. In Figures 2 and 3 we present the risk-of-bias assessment using graphs [54,55].

Methodology of Microscopic Investigation
Regarding the number of investigators, the most common version was one observer. In three cases, there were two investigators [16,37,48] (Table 2).

Methodology of Microscopic Investigation
Regarding the number of investigators, the most common version was one observer. In three cases, there were two investigators [16,37,48] (Table 2).  [51] 4 main n/a n/a n/a Kokubu et al. [52] 4 main n/a n/a Toluidine Blue, Alizarin Red Dallaudière et al.
The majority of studies did not present the exact area of microscopic investigation. On the other hand, two authors evaluated the total area of the specimen [45,48], and two others chose the most severely degenerated region of tendon specimens [16,37].

Therapeutic Effects
An evaluation of therapeutic effects showed 15 positive results [1,[38][39][40][42][43][44][46][47][48][49][50][51][52][53], 6 negative results [34][35][36][37]41,45], and 1 neutral result of treatments [16] (Table 3).  The studies with positive therapeutic effects represented the following methods: curcumin application, autologous conditioned serum application, low-molecular-weight heparin and rivaroxaban injection, vitamin C and hyaluronic acid injection, TGF-β1 used with the scaffold, subacromial bursa excision, erythropoietin application, interleukin-1 receptor antagonist application, adipose-derived stem cell injection, bevacizumab application, and, in five studies, PRP application. The mean Bonar score in these studies was 0.5-8.7 points in the study group and 1.5-10 points in control subjects. The papers that revealed negative therapeutic effects used such methods as isotretinoin application, tranexamic acid injection, collagenase (low dose) injection, ankaferd blood stopper ® application, mechanical loading, and high-intensity training application. The mean value of the Bonar score in the treated groups with negative results was 2.75-11.8 points, while in the control group, it varied between 0.9 and 9.167 points. One paper revealed a neutral effect of PRP injection, with a mean Bonar score in the study group of 8.3 points compared to 8.9 points in the control group.

Discussion
In this systematic review, we investigated papers where the Bonar histopathological score, designed primarily to evaluate excess tissue degeneration, was used in the assessment of therapeutic advances in tendinopathy. The Bonar scoring system, well known and established in the quantification of tendinous tissue degeneration, revealed a new attribute in all concerned studies-a function in the evaluation of tendon treatment effectiveness. We carefully selected 22 papers that mainly focused on PRP application in chronic tendon disorders, among which animal studies prevailed. However, the results of this systematic review showed that the Bonar score could also be easily applied in human in vivo studies to assess the shift of degeneration in healed tendinous tissue. Yet, there are some issues that have to be established and determined before widespread clinical application. The studies from the investigated papers were mainly focused on the Achilles, SST, and patellar tendons; however, the results of Bonar system use in the context of treatment results in other localizations, such as the posterior tibialis tendon (PTT), the long head of the biceps tendon (LHBT), and forearm tendons, are unknown. Further, a control group was presented in each paper and based mainly on Achilles tendon specimens. This is an important issue that is obligatory in the evaluation of various therapeutic methods in chronic tendon disorders. Future human studies should also have well-standardized control groups to easier define the therapeutic effect. Finally, the treatment methods used in the selected studies varied, but PRP application predominated. Positive effects of therapeutic methods prevailed; however, in one paper, a neutral effect of PRP application was noted. We still need a standardization of PRP therapy to predict a clinical effect, but on the other hand, the presented studies revealed no side effects of this treatment method.
In recent decades, increasing concern with platelet-rich plasma injections in tendon disorders has been noted. PRP therapy is a relatively inexpensive and easy-to-prepare procedure. PRP is widely used in different fields of medicine; however, the evidence base for the clinical use of PRP is still in its infancy [51]. This therapeutic method intends to augment the natural tendon healing process, supplying the tissue with specific, highly concentrated growth factors. PRP has been recommended as a treatment option in refractory tendinopathies [30]. Its injections into the patellar tendon, Achilles tendon, and tendons of digit flexors after injury have shown positive clinical results and led to superior regeneration [56]. The application of PRP in elbow tendinopathy showed promising results both radiologically and clinically [57,58]. On the contrary, some authors revealed no functional improvement among patients who underwent PRP therapy [59]. In such situations, histopathological assessment could be helpful to augment the functional evaluation of therapeutic results.
The authors of five of the six PRP studies evaluated in this systematic review showed that PRP had a positive effect on both clinical presentation and histopathology, including collagen architecture, ECM composition, tenocyte morphology, and expansion of capillaries.
Moreover, Kim et al. revealed that PRP therapy decreased inflammation and reduced the apoptosis rate in tendinous tissue [42]. On the other hand, a study by Fukawa et al. showed that there was no statistically significant improvement after Achilles tendon PRP treatment [16]. PRP application as a therapeutic method in chronic tendon disorders was the most widely used in the selected group of papers. However, there were other methods with a negative or positive effect on tendinous tissue. Güleç [1,39,40,53]. Moreover, Yoon et al. observed superior biomechanical and histological outcomes in animals treated simultaneously with scaffold and TGF-β [44]. Kokubu et al. transplanted ASCs, as a progenitor population to fibroblasts, into the injury site and observed improved tendon repair and the prevention of ectopic ossification [52]. On the other hand, Pingel et al. and Scott et al. presented a negative influence on tendon regeneration of excessive stress and inferior histopathological outcomes measured using the Bonar system [35,45]. Further, Isotretinoin, Tranexamic acid, and a low dose of Collagenase showed a negative effect on the tendon healing process, leading to more advanced degeneration than that in control groups [34,36,37].
Regarding the applied evaluation method, most of the authors used the four-variable Bonar score. However, three authors introduced a fifth variable representing tenocyte proliferation and counted the number of tenocytes. This alteration was also used in the quantification of pathology in tendinous tissue with the Bonar scale; nevertheless, this fifth variable should be reconsidered in further human studies [60,61]. Unfortunately, the number of microscopic investigators in presented studies was often not available or it was a single observer. In the studies of degeneration supported with the Bonar scale, two observers or at least one experienced investigator ensure the objective assessment of tissue [25]. Still, it is not clear how it should be set for human studies. However, the use of two observers is the best solution to avoid bias. The majority of authors were unable to set a certain area of microscopic investigation. This problem was also met in studies of tendon pathology, and authors usually randomly selected the evaluated area of the slide.
Regarding additional staining, the most frequent were Alcian Blue and Masson Trichrome methods, which are used for better visualization of ECM alterations. These additional stainings facilitate microscopic assessment of the tendon structure; however, none of the authors proved the necessity of these upgrades.
The role that the neovascularization process plays in tendon degeneration and healing is ambiguous; thus, this variable in the Bonar scoring system should be treated with caution [62][63][64]. Kokubu et al. showed that ASCs induced the neovascularization process in the early phase of Achilles tendon healing [52]. Moreover, Dolkart et al. revealed similar scores for the neovascularization variable in both the PRP-treated group and the control group, despite the significant improvement in the overall Bonar score of subjects who underwent PRP treatment [46]. Furthermore, Oztermeli et al. noticed no statistically significant effect on vascularity in the early phase after local or systematic administration of EPO [49]. EPO is a strong proangiogenic agent, while angiogenesis is crucial in tendon healing. Authors in the late period observed statistically significant differences between locally injected EPO (0 points in Bonar) and both systematically injected EPO and the control group (both 1.33 points in Bonar). They suggested that local EPO administration contributes to vascularization in the late period. However, the vascularization score corresponded to randomly scattered, inconspicuous capillaries. It could be concluded that the EPO application provided a faster reorganization of the angiogenesis process in tendinous tissue. New vessel formation is a characteristic feature of the formation phase of the tendon regeneration process which lasts approximately from the 7-8th day until the third week. On the contrary, the invasion of new vessels is known as a typical sign of tendinopathy and often treated using sclerosing drugs [64]. In turn, Fearon et al. considered a complete lack of vascularity in tendinous tissue a pathology, which in their modified Bonar score was rated equally as an abundant neovascularization process [25]. Summarizing, the role of neovascularization in tendinopathy should be reconsidered in the Bonar score as well as avascular regions.
This systematic review was limited by a number of factors. First, the methodology of the included studies, the number of investigators, and the observed specimen area were highly differentiated. Second, the applied therapeutic methods had various influences on tendon histopathology-positive, negative, or neutral-and depending on the alterations, authors scored the tissue using the Bonar system. As the application of this scoring system was not yet well established in the evaluation of tissue regeneration, the results may be biased. Third, in some studies, there was a lack of demographic data, and the study design and methodology may have a significant impact on the interpretation of the results. We supported two well-established tools to minimalize the risk of bias; however, we realize that there are various scales used in the assessment of paper quality. In the majority of the included studies, we observed an overall low risk of bias; however, some studies presented limitations with unclear risk of bias, such as blinding, random housing, and random outcome assessment. Moreover, in one study we observed high risk of bias due to a lack of clarity in the presentation of results and figures.

Conclusions
The Bonar scoring system is well established in tendon pathology assessment and could also be introduced to assess therapeutic results in tendon disorders. To understand the tendinopathy phenomenon, a link between the histopathology and clinical presentation in chronic tendon disorders is required as it may improve functional outcomes. Studies that included PRP application showed evidence of superior tendinous tissue healing, measured using the Bonar score, and improved clinical results. Finally, there are some issues, such as the neovascularization variable in the scoring system, the number of investigators, the area of tissue investigation, specific staining methods, and control groups, that should be reconsidered and set before widespread clinical application. Further randomized clinical studies are needed to confirm these promising results.