Injured tendons heal slowly because of less blood supply and reduced metabolism compared with many other tissues1,2. The use of growth factors has been proposed as a strategy to improve tendon healing3. Platelet-derived growth factor (PDGF), transforming growth factor-β1 (TGF-β1), vascular-endothelial growth factor (VEGF), epidermal growth factor (EGF), basic fibroblastic growth factor (bFGF), and insulin-like growth factor type-I (IGF-I), among others, have been recognized to play key roles in tendon healing4,5. Although these growth factors potentially could be used therapeutically to accelerate the complex process of tendon healing, it seems unlikely that a single growth factor will give a positive result in vivo. Some authors have suggested that the interaction of various growth factors with a natural balance of anabolic and catabolic functions, present in the right concentration at the right time, would be necessary to optimize the tissue environment and to favor the tendon healing process3,6.
Autologous plasma, generally called platelet-rich plasma (PRP), is being used to treat musculoskeletal injuries. Platelets contain a complex pool of growth factors, interleukins, and other biologically active proteins, such as fibrinogen, fibronectin, and vitronectin, which have prominent roles in the healing process7. Several clinical studies have demonstrated that applications of PRP can improve the healing response after tendon injuries8-16, but there is an intense debate about the effectiveness of these products6,17-19.
The purpose of the present study was to evaluate the effect of plasma rich in growth factors (PRGF), which is one type of leukocyte-poor PRP, on tendon healing. We compared the histological appearance of acutely surgically injured and repaired Achilles tendons that had been treated with PRGF with that of repaired tendons that had been infiltrated with saline solution in sheep. Our hypothesis was that weekly infiltrations of PRGF onto the injured tendons would accelerate the healing response and would improve the histological properties of these tendons.
Experimental Model of Divided Achilles Tendon
The present study was approved by the Bioethical Committee on Animal Research at our institution. We strictly adhered to the guidelines for the use of laboratory animals proposed by the National Institutes of Health. We used twenty-eight skeletally mature female Merino sheep weighing between 45 to 55 kg for our study. A veterinary surgeon performed general and orthopaedic physical examinations on each animal to guarantee that all sheep were free from lameness.
With use of sterile technique, the right hindlimb of the anesthetized sheep was prepared for surgery. A lateral skin incision was made over the Achilles tendon. The peritenon was incised longitudinally (Fig. 1-A). A full-thickness transverse tendon division of the Achilles tendon was performed with use of a scalpel, 5 cm proximal to its insertion into the calcaneus (Fig. 1-B). The tendon was repaired with use of a three-loop pulley pattern with a nonabsorbable synthetic monofilament (polypropylene) (Premilene USP 1; B Braun Aesculap, Melsungen, Germany) (Fig. 1-C). The peritenon was sutured with a continuous absorbable synthetic monofilament (polyglyconate) (Monosyn USP 3-0; B Braun Aesculap). The wound was closed in layers and was covered with a sterile dressing.
The repaired tendon was protected during the postoperative period with use of a custom system that allowed the animal to bear weight on the operatively treated hindlimb without subjecting the Achilles tendon to macromovements20,21. A tarsal transarticular external skeletal fixation system was placed in the medial aspect of the limb just after skin suture and was left in place throughout the experiment. The tarsal joint was fixed at an angle of 140° (Fig. 1-D) and was covered with a protective bandage. An antibiotic (benzylpenicillin [15,000 IU/kg, administered intramuscularly] plus dihydrostreptomycin [15 mg/kg, administered intramuscularly]) (Shotapen LA; Virbac Animal Health, Barcelona, Spain) was given at the time of surgery and again after seventy-two hours. Analgesic buprenorphine 0.02 mg/kg/8 hr, administered intramuscularly (Buprex 0.3 mg; Schering-Plough, Madrid, Spain) was given for five days after surgery.
Plasma Rich in Growth Factors (PRGF): Preparation and Infiltration on Repaired Area
PRGF was prepared with use of the PRGF-Endoret system (Biotechnology Institute [BTI], Vitoria, Spain). Disposable BTI extraction and fractioning tubes were used in a custom-designed bench-top centrifuge. Each 5-mL extraction tube contained 0.5 mL of sodium citrate solution (3.8%) as anticoagulant. Twenty milliliters of blood were collected from the jugular vein of each animal and were divided into four tubes that were centrifuged at 630 times gravity for eight minutes according to the method reported by Anitua el al.22. Blood for PRGF preparation was collected just prior to surgery and was processed intraoperatively. The upper layer of the centrifuged plasma was removed with use of a pipette and was discarded. The remaining 0.5 mL of plasma fraction that lies just above but not including the interphase zone (“buffy coat”) was retrieved with use of a sterile pipette and was placed in a fractioning tube. This plasmatic fraction is the PRGF according to the manufacturer’s instructions. The plasma separation procedure was done in a laminar flow cabinet in accordance with the PRGF-Endoret system manufacturer instructions. The 0.5 mL of PRGF from each of the four tubes were combined to obtain a total volume of 2 mL of PRGF per animal. The platelets were activated by adding 0.1 mL of 10% calcium chloride just prior to the injection of PRGF in the injured tendon. The time delay between blood collection and PRGF application was less than one hour.
The sheep were randomly divided into four groups of seven animals each. In the sheep in the two experimental groups, 2 mL of PRGF were infiltrated intratendinously into the divided tendon stumps with use of a 23-gauge needle just before suturing of the peritenons. The injections were repeated every week for the following three weeks. The repaired tendons in the other two groups (controls) were injected with 2 mL of saline solution plus 0.1 mL of 10% calcium chloride as described above. The infiltrations during the postoperative period were carried out in properly sedated animals with use of sterile technique under ultrasonographic guidance. The animals in one PRGF-treated group and one saline solution-treated group were killed at four weeks, and the remaining animals were killed at eight weeks.
Histological Procedure
After each animal was killed, both Achilles tendons were harvested and were fixed in 10% neutral buffered formalin. Contralateral (non-operatively treated) tendons were studied as normal controls. Each operatively treated tendon was cut lengthwise on the dorsal midline with use of a scalpel blade. A 1-mm-thick slab of tissue that included the entire repaired area was excised. This slab of tissue also contained intact tendon fibers on either side of the repaired area. The margin adjacent to the intact tendon fibers was yellow-inked for histological identification. Two paraffin blocks were obtained from each operatively treated tendon. From each block, six 4-μm-thick consecutive paraffin sections were cut. Sections were mounted in slides, and three were stained with hematoxylin and eosin and three were stained with Masson trichrome.
The stained sections were digitalized with use of a photomicroscope (Axiophot; Carl-Zeiss, Oberkochen, Germany) with an attached digital camera (DS-5M camera head; Nikon, Tokyo, Japan) coupled with a digital imaging controller (DS-L1 camera control unit, Nikon) that allowed us to observe, focus, and capture histological images. Slides were sampled in a systematic randomized manner by superimposing a dotted transparent template onto each slide. Several dots spanning the whole tissue section were marked over the glass of the slide with use of a permanent marker. The regions of interest were chosen during the histological examination beside a previously marked dot. In Masson trichrome-stained sections, nine histological images per slide were captured with a magnification of ×200. In hematoxylin and eosin-stained sections, nine histological images per slide were captured with a magnification of ×200 and nine histological images per slide were captured with a magnification of ×630. These histological images were transferred to a computer equipped with image-analysis software (Image Pro Plus, version 6.0; MediaCybernetics, Rockville, Maryland) to perform quantitative measurements. In order to blind the evaluation of the histological images, all of the captures were encoded with use of an identification number and were randomly evaluated by three independent pathologists.
Morphology of Fibroblast Nuclei
The morphometry of fibroblast nuclei was studied with the aid of image-analysis software on images captured at a magnification of ×630 and stained with hematoxylin and eosin. The outlines of fibroblast nuclei were traced with use of the cursor of the image software (Fig. 2-A). The following quantitative parameters were calculated from each nuclear profile: area (μm2), perimeter (μm), major axis (μm), minor axis (μm) and roundness [(perimeter2)/4 × pi × area]. The nuclear aspect ratio (defined as the ratio of the minor axis to the major axis) was also calculated, with values approaching zero suggesting a spindle shape and with the value of 1.00 representing a perfect circle23. The nuclear orientation angle was defined as the angle between the major axis of the nucleus and the longitudinal tendon axis, with values of 0° representing a nucleus that is perfectly aligned toward the longitudinal axis. When this value increases, the nucleus becomes more angled until it approaches 90°, where it lies perpendicular to the long axis of the tendon23.
Histological Study of Collagen Fibers
The histological appearance of the collagen fibers was studied on images captured at a magnification of ×200 and stained with Masson trichrome. A 5-point semiquantitative grading scale was designed to evaluate the packing and orientation of collagen fibers (see Appendix). This semiquantitative grading scale was a modification based on the grading scales described in three different publications24-26.
Fibroblast Density
The fibroblast density was determined with use of a counting frame that was superimposed onto each histological image; these images were captured at a magnification of ×630 and were stained with hematoxylin and eosin (Fig. 2-B). The number of fibroblasts per area was calculated according to the formula: fibroblasts/mm2 = N/AF, where N was the number of fibroblasts counted within the counting frame, and AF was the area of the counting frames (in our study, AF = 0.004 mm2).
Vascular Density
Vascular density was quantified with use of histological images that were captured at a magnification of ×200 and stained with hematoxylin and eosin. The number of blood vessels in each image area was divided by the area of the histological image captured at a magnification of ×200 (in our study, 0.142 mm2).
Evaluation of the Inflammatory Cell Infiltration
The infiltration of inflammatory cells was analyzed with use of images captured at a magnification of ×200 and stained with hematoxylin and eosin. As the different inflammatory cell subtypes were not quantified, the inflammatory cell infiltration was defined on the basis of the density of the main inflammatory cell subtype observed in the histological sections (lymphocytes). The density of inflammatory cell infiltration was characterized with use of a 5-point semiquantitative grading scale. A value of 0 was assigned if inflammatory process was absent, a value of 1 was assigned when there was slight inflammatory cell infiltration, a value of 2 was assigned when there was moderate inflammatory cell infiltration, a value of 3 was assigned when there was strong inflammatory cell infiltration, and a value of 4 was assigned where there was severe inflammatory cell infiltration homogeneously spread in the entire field of view.
Statistical Analysis
Descriptive statistics for each study group (including the mean, standard deviation, and frequencies for categorical data) were calculated. The nonparametric Kruskal-Wallis test and the Mann-Whitney U test were used to analyze the significance between groups for quantitative parameters (morphology of fibroblast nuclei, fibroblast density, and vascular density). Categorical data (packing of collagen fibers, orientation of collagen fibers, and inflammatory cell infiltration) were analyzed with use of the chi-square test to determine if the infiltration with PRGF or saline solution yielded significant differences in the frequency distribution between study groups. Values were presented as mean and the standard deviation (SD). The level of significance was set at p < 0.05.
Source of Funding
This study was supported by the García-Cugat Foundation for Biomedical Research and the Ministry of Health of the Spanish Government. One of the authors (J.A.F.-S.) was the recipient of a grant from the Ministry of Education and Science of Spanish Government (AP2006-00150). The authors do not have any financial interest or other relationship with any commercial company related to this study.
Platelet-rich plasma is being used with increasing frequency in sports medicine to try to accelerate healing and to allow an earlier return to sports18,19. Despite this interest, there have been only two clinical studies that have evaluated the effectiveness of PRP in the treatment of ruptures of the Achilles tendon in humans: a case-controlled study that showed a faster functional recovery in association with the use of PRP9, and a randomized controlled trial that showed no effect16. However, several animal studies have shown promising results in association with the use of PRP for the treatment of tendon lesions27-35.
The ratio of the minor axis to the major axis of the nucleus (nuclear aspect ratio) and the angle between the major axis of the nucleus and the longitudinal tendon axis (nuclear orientation angle) provide information about fibroblast nuclei shape and alignment along the longitudinal tendon axis, respectively. In our study, measurement of the nuclear aspect ratio demonstrated that the PRGF-treated tendons had a more elongated nucleus than did the saline solution-treated tendons, suggesting a more mature histological appearance in the PRGF-treated tendons. We observed that fibroblasts in the PRGF-treated groups were more parallel to the longitudinal tendon axis than those in the saline solution groups, as demonstrated by a lower value of the nuclear orientation angle.
Early inflammatory and proliferative phases are defined by a dramatic increase in the fibroblast population. Approximately three or four weeks after the traumatic event, the fibroblast density reaches its peak and then decreases progressively during the remodeling and maturation phases36,37. The neovascularization is prominent during early stages of tendon healing but then progressively disappears27. In our study, the PRGF-treated tendons had lower fibroblast density and a faster decrease in neovascularization compared with the tendons that had been infiltrated with saline solution, suggesting that PRGF was associated with histological changes consistent with accelerated early healing.
The collagen structure is closely related to the biomechanical properties of tendons36. Packing and orientation of collagen bundles are major features in the collagen structure, and they have been evaluated in several studies24-26. We observed an improvement in both the packing and the orientation of collagen bundles in PRGF-treated tendons, showing a more mature collagen structure compared with saline solution-treated tendons.
Platelets modulate inflammation by secreting chemokines that control chemotaxis of leukocytes and macrophages in injured tissues17. The features of the inflammatory response may determine the success of tendon repair17. In our study, PRGF-treated tendons showed lower inflammatory cell infiltration than did saline solution-treated tendons at both four and eight weeks. No research has been conducted to investigate the exact effect of PRP on the inflammatory response after acute tendon injury. Further investigations are needed to clarify the role of PRP in the modulation of the inflammatory response during the healing process.
Platelet-rich plasma is a general term that encompasses many different autologous plasma products38, even though they are obtained with use of different protocols that may result in different concentrations of platelets, leukocytes, and growth factors and even though these products differ both qualitatively and quantitatively and show different biologic effects38,39. The PRGF-Endoret system produces a volume of autologous plasma having a platelet concentration above baseline, free of leukocytes and erythrocytes, elaborated by a one-step centrifugation process and using sodium citrate as anticoagulant and calcium chloride as platelet activator39.
Schepull et al.16 and de Vos et al.11,40 reported that the use of PRP did not demonstrate any effect on the healing of human Achilles tendon in clinical studies, but the methodology that they followed for obtaining the autologous platelet concentrate was different from the PRGF methodology that we used in our experimental study. Schepull et al.16 used a PRP preparation obtained by double centrifugation, which yielded a product with an extremely high concentration of platelets (tenfold above baseline). Previous studies demonstrated that high concentrations of some growth factors, such as TGF-β, could be deleterious for tendon healing as TGF-β drives fibrogenesis, potentially stimulates the development of scar tissue, and is associated with the onset of fibrosis22. de Vos et al.11,40 used a PRP product obtained by means of a methodology that collects a high concentration of leukocytes. PRGF is a product free from leukocytes, avoiding the proinflammatory effects of the proteases and cytokines contained in white blood cells41.
While some studies have involved only a single application of PRP11,16,41, we used weekly applications of PRGF during the following three weeks after surgery on the basis of a previous study in sheep by Anitua et al.22. While PRGF was injected into intact tendons without any induced injury in the study by Anitua et al.22, a complete division of the Achilles tendon was surgically induced in the present study. Such intense damage triggers a strong proliferative healing response in the injured area36,37. Anitua et al. reported that the injection of PRGF in intact tendons increased cell density and promoted neovascularization after four weeks of weekly administration in their experimental model22. Our findings suggested that when PRGF was injected into injured tendons, the histological response was different. We observed a decrease in fibroblast density in PRGF-treated tendons, which may be explained by an acceleration in the healing process induced by the PRGF injections. As the healing process advances, a progressive decrease in cell density occurs in the repaired area and the maturation phase of the healing process begins36,37. Thus, PRGF-treated tendons showed a histologically more advanced stage of tendon healing than did saline solution controls at the same postoperative time.
The present study had several limitations. One limitation was that we did not perform any biomechanical testing of the Achilles tendon to assess whether the histological differences that were observed between the PRGF and saline solution groups were mechanically relevant. The main objective of the present study was to histologically evaluate the effect of PRGF on tendon healing. In light of these promising results, future studies are warranted to clarify whether histological improvements in tendon healing are related to superior biomechanical properties. Another limitation of the present study is related to the formulation of PRGF in sheep. The composition of PRGF in humans recently was described by Anitua et al.42, but we are aware of no studies on the concentration of platelets, leukocytes, erythrocytes, or growth factors in the PRGF of sheep. Another concern is that multiple weekly injections were performed in this study and that these injections may elicit an inflammatory response because of the disruption of the peritenon. We considered this unavoidable damage to be the same in both the PRGF groups and the saline solution groups. Hence, differences observed between experimental groups should be attributed to the injected product itself.
In conclusion, we found that PRGF was associated with histological changes consistent with accelerated early healing process in tendons after acute injury and repair of Achilles tendons in sheep. PRGF-treated tendons showed better orientation and a more elongated shape of fibroblast nuclei, suggesting a more advanced stage of healing compared with saline solution-treated tendons. Histological examination of the repaired areas showed a more mature organization of collagen bundles, less inflammatory cell infiltration, faster vascular regression, and lower fibroblast density in PRGF-treated tendons than in saline solution-treated controls.
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.