Achilles tendon rupture affects one or more structures. This injury is commonly caused by acute trauma, such as tendon laceration, resulting in a typical tarsus plantigrade stance in dogs with complete rupture (3). Achilles tendon rupture can induce irreversible lameness if left untreated. Therefore, surgical repair is generally recommended (8). Surgical methods for repairing ruptured Achilles tendons include suturing, immobilization, augmentation, and lengthening.

Various augmentations for tendon repair using autologous tissue had been attempted using fascia lata (32), flexor digitorum longus (30), peroneus brevis (30), gracilis (23), and plantaris tendon (19) grafts, which have been described in the literature. The advantages of autologous tissue implantation include a lower risk of implant-associated complications, such as wound infection, foreign body reactions, postoperative seromas, foreign body reactions, and soft tissue erosion (11). There are many reports using fascia lata as an autograft in humans (1), as well as in the veterinary literature (32). A fascia lata graft has been reported to induce lameness of the donor limb, including seroma, hematoma, pain, and wound dehiscence (4). In addition, the fascia lata has a smaller surface area than the thoracolumbar fascia (TF) (17). The TF has been used as an autograft material, which can be harvested easily, and no complications of the donor site have been reported (10).

Augmentation using synthetic materials has also been applied to repair tendons, such as polypropylene mesh (PM) (13,15) and titanium plates (9). PM is widely used for soft tissue reconstruction, including tendon repair augmentation. The advantages of using a PM include the following: no additional surgery (hence no donor-site morbidity), unlimited availability of sizes, less adhesion formation, high tensile strength, and provision of a scaffold for fibroblast infiltration (13,15). A distinct disadvantage of synthetic grafts is the additional cost and possible risk of foreign body reaction after implantation. However, PM is known to have little to no tissue-reactive material (2).

Immobilization has been commonly applied using splints (9), casts (25), transarticular external skeletal fixation (32), and calcaneotibial screws (9) with surgical intervention. Immobilization is limited by isometric muscle contraction (22). Moreover, at least 6 weeks of immobilization is required because of the relatively poor blood supply and slow healing of the tendon (6,8). The gap at the site of the ruptured tendon is also filled with scar tissue. This might decrease the tensile strength, stiffness, and function compared to the intact tendon. Thus, the tendon is susceptible to recurrence of rupture (15,16). However, early rehabilitation and the possibility of weight-bearing are effective in accelerating recovery (21). Therefore, a strong tendon repair method is necessary, allowing early rehabilitation and tendon healing (27,29). Thus, many authors have recommended tendon augmentation for repair (3).

The purpose of this study was to compare the augmentation methods using TF and PM using biomechanical and histological assessments.

Animals and study design

Twenty-four male New Zealand white (NZW) rabbits (3.20 ± 0.15 kg) were randomly divided into three groups, and they were all considered as skeletally normal and healthy by physical examination and complete blood count (CELL-DYM 3700, Abbott laboratory, USA). The Institutional Animal Care and Use Committee of Kyungpook National University approved all the experimental protocols used (2020-0095). All NZW rabbits were reared in individual standard rabbit cages and received a standard rabbit diet with free access to food or water. The animals were divided into the following three groups:

1. Control group: Kessler sutured for completely transected Achilles tendon.

2. PM group: Kessler sutured with the PM padding for completely transected Achilles tendon.

3. TF group: Kessler sutured with TF padding for completely transected Achilles tendon.

Each NZW rabbit served as its own control, and the right lateral gastrocnemius tendon (LGT) was used as the normal control.

Preparation for augmentation

The TF was obtained from the thoracolumbar region of the NZW rabbit (Fig. 1A) immediately before exposing the Achilles tendon under anesthesia. The NZW rabbit was placed in sternal recumbency. The surgical site for the fascia was clipped and aseptically prepared. A 5-cm incision was made at the surgical site, and blunt dissection was performed to access the fascia. The sufficient fascia was then excised and trimmed to 10 mm × 25 mm (Fig. 1A). The incision site was closed routinely. The PM (Prolene^® mesh, Ethicon, Inc., New Jersey, USA) was trimmed to 10 × 25 mm (Fig. 1B).

Figure 1. Materials for Achilles tendon repair augmentation. (A) Thoracolumbar fascia (left) and harvested thoracolumbar fascia (right). (B) Polypropylene mesh.

Surgical procedure

For premedication, 10 mg/kg enrofloxacin (Baytril 50^®, Bayer, Leverkusen, Germany) was injected intramuscularly. For analgesia, 4 mg/kg tramadol (Maritrol^®; Jeil Pharmaceutical Co., Seoul, Korea) was injected intramuscularly. NZW rabbits were anesthetized by intramuscular injection of a combination of 35 mg/kg ketamine (Yuhan Ketamine 50 inj^®, Yuhan Co., Seoul, Korea) and 5 mg/kg xylazine (Rompun^®, Bayer, Leverkusen, Germany). The left hindlimb of each rabbit was used in the experiment. Hair around the incision site was clipped, and surgical asepsis was performed with 70% isopropyl alcohol and 1% chlorhexidine. A 4-cm longitudinal incision was made over the Achilles tendon, proximal to the gastrocnemius muscle from the tuberosity of the calcaneal bone, through the skin and subcutaneous tissue. After exposing the Achilles tendon, the LGT was isolated from the superficial digital and medial gastrocnemius tendons by blunt and sharp dissection. The isolated LGT was then completely transected 2 cm above the tuberosity of the calcaneal bone. The completely transected Achilles tendon was repaired using a uniform Kessler suture in the control group using a 3-0 monofilament non-absorbable suture (Blue nylon, Ailee Co., Busan, Korea) (Fig. 2A). In the PM or TF group, completely transected Achilles tendons were padded with PM or TF, followed by uniform repair with Kessler suture using 3-0 monofilament non-absorbable sutures (Fig. 2B).

Figure 2. Suture method for completely ruptured Achilles tendon. (A) Control group, Kessler suture only. (B) TF and PM group, with TF and PM padding.

Following intradermal suture using 4-0 multifilament absorbable sutures (Vicryl, Ethicon, New Jersey, USA), the operated limb was immobilized by casting. The casts (PP-band; Keumjeong Chemical Company, Hwaseong, Korea) had a 180° angle at the ankle. The casts were well padded using gauze, cotton roll bandage, elastic bandage, and casting tape. The casts were maintained for 8 weeks and removed when the surgical site was reopened to obtain the tendon. The replacement of casts were performed at 4 week, and minor reddish skins were identified in few individuals, but there was no severe or major skin injury. For post-operative management, 10 mg/kg of enrofloxacin and 4 mg/kg of tramadol were injected intramuscularly twice daily for 7 days.

Sample collection

Eight weeks after surgery, NZW rabbits were anesthetized in the same manner as the first surgical procedure, followed by opening of the operated site and contralateral hindlimb. The process of isolating LGT was performed in the same way as previously described. The LGT was transected at the musculotendinous junction and immediately above the calcaneal tuberosity. Samples were divided into four groups: intact, control, PM, and TF. Tendon specimens used for biomechanical assessment were stored in a refrigerator (10°C) until tests were performed (a total of four groups and 20 LGT samples, individual packages in empty 15-mL Falcon tubes). For histological assessment, tendon specimens were stored at room temperature until tests were performed (a total of four groups and twelve LGT tissue samples, individual packages in 15-mL Falcon tubes soaked in 10% neutral buffered formalin).

Biomechanical assessment

Five NZW rabbit LGT samples were collected from each group (a total of four groups and 20 LGT samples, individual packages in empty 15-mL Falcon tubes). The tensile strengths of the individual LGT were measured using a computerized testing machine (SV-H1000, Japan Instrumentation System Co., Ltd., Tokyo, Japan) in Newtons.

Each LGT was fixed vertically in the stainless steel clamps of the machine without any wrinkles. The completely transected region was located at the center of the stainless clamps. They were pulled at a constant speed (300 mm/min) and at a 5-mm distance. In this study, the peak tensile loads are documented in Newtons.

Histological process

Previously, three NZW rabbit LGT tissue samples from each group were obtained and stored (a total of four groups and twelve LGT tissue samples, individual packages in 15-mL Falcon tubes soaked in 10% neutral buffered formalin). Individual LGT tissue samples were trimmed longitudinally and the completely transected region was located in the central region. For 24 h, all trimmed LGT tissue parts were re-fixed in 10% neutral buffered formalin.

The paraffin blocks contained all three LGT tissues from the same group. These blocks were prepared in all 12 samples obtained (a total of four paraffin blocks) using an automated tissue processor (Shandon Citadel 2000, Thermo Scientific, Waltham, USA) and embedding center (Shandon Histostar, Thermo Scientific, Waltham, USA). Five serial paraffin sections were prepared using an automated microtome (RM2255, Leica Biosystems, Nussloch, Germany) equipped with a 3- to 4-μm-thick tungsten bladder (TC-65, Leica Biosystems, Nussloch, Germany) in each paraffin block.

Representative sections were stained with hematoxylin and eosin for general histopathology, Alcian blue (AB) staining for proteoglycans, and Masson trichrome (MT) for collagen fibers (fibroplasias). Histological profiles of individual longitudinally trimmed NZW rabbit LGT tissues were observed under a light microscope (Model Eclipse 80i, Nikon, Tokyo, Japan). This light microscope was equipped with a histological camera system (ProgResTM C5, Jenoptik Optical Systems GmbH, Jena, Germany) and a computer-assisted automated image analyzer (iSolution FL ver 9.1, IMT i-solution Inc., British Columbia, Canada).

Histomorphometric analysis

To identify histopathological changes in more detail, a modified semi-quantitative grading score from 0 to 3 was applied to the transected regions of each Achilles tendon tissue. The cell density, nuclear roundness, fiber arrangement, and fiber structure were scored. According to this grading system, perfectly normal tendons scored 0; mild and moderate prevalences scored 1 and 2, respectively; and severely abnormal tendons scored 3 (7).

Additionally, the mean number of infiltrated inflammatory cells (cells/mm²) was measured using a computer-assisted image analysis program and histological camera systems, as general histomorphometric analysis. The tissues occupied over 20% of AB and MT staining were regarded as positive compared to the background. The mean AB-positive proteoglycan-occupied regions (%), and MT-positive collagen fiber-occupied regions (%) were measured using an automated image analyzer and histological camera system. Histomorphometric analysis was performed on the central zone of the completely transected regions of each Achilles tendon tissue.

Statistical analysis

All values for tensile strength, semi-quantitative score, and quantitative histomorphometry analysis were presented as the mean ± SD of five NZW rabbit LGTs. Multiple comparison tests were performed for the different dose groups. Variance homogeneity was examined using Levene’s test. If Levene’s test indicated no significant deviations from variance homogeneity, a one-way analysis of variance test was used to analyze the obtained data, followed by Tukey’s honest significant difference test to determine which pairs of group comparisons had significant differences. If significant deviations from variance homogeneity were shown in the Levene’s test, a non-parametric comparison test, the Kruskal–Wallis H test, was performed. When a significant difference was shown in the Kruskal–Wallis H test, the Mann–Whitney U test was performed to determine the specific pairs of group comparisons that were significantly different. Statistical analyses were performed using SPSS for Windows (Release 14.0, IBM SPSS Inc., USA). A p < 0.05 or p < 0.01 was considered statistically significant.

Tensile strength

A noticeable and significant (p < 0.01) decrease in tensile strength was observed in the control group compared to the intact group. Significant (p < 0.01) increases in tensile strength were observed in the completely transected regions in the TF and PM groups compared to those in the control group. There was no significant difference between the PM and TF groups; however, the PM group had an increased average tensile strength compared to the TF group (Fig. 3).

Figure 3. Tensile strengths of each group. **p < 0.01 and *p < 0.05 as compared with intact, ^##p < 0.01 as compared with control.

Histological finding

Histopathological assessment

Severe abnormal cell density, nuclear roundness, fiber arrangement, and fiber structure were observed in the control group compared with those in the intact group. In this study, increased inflammatory cell infiltration, accumulation of AB-positive proteoglycans and decrease in MT-positive collagen fibers at histopathological levels were demonstrated in the control group as compared to those of the intact group (Figs. 4, 5).

Figure 4. The representative general histopathological appearance of intact or repaired completely transected NZW rabbit Achilles tendon. Scale bars = 100 μm (A), intact group (B), control group (C), PM group (D), TF group.

Figure 5. Semi-quantitative grading score system on HE-stained completely transected Achilles tendon tissue, **p < 0.01 as compared with intact, ^##p < 0.01 as compared with control. (A) Cell density, (B) nuclear roundness, (C) fiber arrangement, (D) fiber structure.

However, obvious decreases in the abnormalities in cell density, nuclear roundness, fiber arrangement, and fiber structure were observed in the PM and TF groups compared with those in the control group. Decreases in the mean number of infiltrated inflammatory cells and AB-positive proteoglycan-occupied regions with increases in MT-positive collagen fiber-occupied regions were demonstrated in the transected regions in the TF and PM groups as compared to those in the control group (Fig. 4).

Semi-quantitative grading score

Significant (p < 0.01) increases in abnormalities in cell density, nuclear roundness, fiber arrangement, and fiber structure in the semi-quantitative grading score system were observed in the completely transected regions in the control group compared to those in the intact group.

In addition, obvious decreases in abnormalities in cell density, nuclear roundness, fiber arrangement, and fiber structure were observed in the completely transected regions in the PM (p < 0.01) as compared to those of the control group. There was no significant difference between the PM and TF groups, but the PM group showed fewer abnormalities than the TF group (Fig. 5).

Quantitative histomorphometry analysis

Significant (p < 0.01) increases in the mean number of infiltrated inflammatory cells, and AB-positive proteoglycan-occupied regions and decreases in MT-positive collagen fiber-occupied regions were observed in the completely transected regions in the control group as compared to those in the intact group.

Significant decreases in the mean number of infiltrated inflammatory cells and AB-positive proteoglycan-occupied regions with increases in MT-positive collagen fiber-occupied regions were observed in the completely transected regions in the PM (p < 0.01) group as compared to those of the control group. There was no significant difference between the PM and TF groups; however, the PM had a decrease in the mean number of infiltrated inflammatory cells and AB-positive proteoglycan-occupied regions with increases in MT-positive collagen fiber-occupied regions as compared to those of the TF group (Fig. 6).

Figure 6. Quantitative histomorphometry analysis, **p < 0.01 as compared with intact, ^##p < 0.01, ^#p < 0.05 as compared with control. (A) Mean number of inflammatory, (B) mean proteoglycan-occupied regions, (C) mean collagen fiber-occupied regions.

NZW rabbits were used as the experimental models in this study. Rabbits are commonly used in Achilles tendon models because of their availability and relatively large tendons for surgical procedures (5). Rabbits both have similarities and some differences in the Achilles tendon anatomy compared to dogs. The Achilles tendon of a rabbit has three structures: medial gastrocnemius tendon, LGT, and superficial digital flexor tendon (33).

Sole suturing on completely ruptured tendons provides less tension for repair than does augmented tendons (24). Therefore, tendon augmentation is routinely performed for neglected, ruptured tendons. Moreover, tendon apposition can be broken due to muscle contraction before an adequate return of strength to the tendon has occurred, even if the limb is immobilized (3,22).

Repair of the Achilles tendon is challenging, and tensile strength is critical for tendon healing. Furthermore, tensile strength affects the biomechanical properties of the Achilles tendon (34), and it is difficult for ruptured tendons to recover their previous tensile strength (14,18,34,36). This can be demonstrated in this study by comparing the intact tendon and the Kessler-sutured tendon. For these reasons, augmentation using TF and PM padding with the Kessler suture technique were used to repair the completely transected NZW rabbit Achilles tendons in this study. The results showed that the PM and TF groups had better tensile strengths than the control group, and the PM group had superior tensile strength than that of the TF group. Therefore, when repairing the Achilles tendon, the augmented tendon with PM or TF has a lower risk of re-rupture, with improved biomechanical properties than the sutured tendon alone.

Tendon healing has three overlapping phases: inflammatory, proliferative, and remodeling. During the inflammatory phase, erythrocytes and inflammatory cells enter the injury site. Until 24 h, monocytes and macrophages predominate and phagocytose necrotic material (26). Vasoactive and chemotactic factors are released with increased vascular permeability, initiation of angiogenesis, stimulation of tenocyte proliferation, and recruitment of more inflammatory cells (31). Tenocytes gradually migrate to the wound and collagen synthesis is initiated. Glycosaminoglycan (GAG) and water content increased. A few days later, the proliferative phase begins and collagen, mainly type III, synthesis peaks in this phase and lasts for a few weeks (26,31). Collagen type I becomes predominant, and a high GAG concentration remains. Approximately 6 weeks later, the remodeling phase starts with decreased cellularity and decreased collagen and GAG synthesis (26). Through their GAG side chains, collagen fibrils are bound to proteoglycans to interconnect the fibrils in a parallel alignment and ensure the gliding of collagen fibrils during movement (35). This phase can be divided into consolidation and maturation stages (31). The consolidation stage begins at 6 weeks and continues for up to 10 weeks. During the consolidation stage, the repaired tissue changes from cellular to fibrous (31). Tenocyte metabolism remains high during this period, inducing tenocytes and collagen fibers to align in the direction of stress (20). After 10 weeks, the maturation stage begins, with a gradual change of fibrous tissue to scar-like tendon tissue over the course of 1 year (20). During the subsequent remodeling phase, and tenocytes and collagen fibers are aligned in the direction of stress (12).

Samples were collected 8 weeks after surgery, during which the consolidation stage of the remodeling phase took place. During this period, the ruptured tendon presented decreased cellularity, slightly round nucleus, slightly fragmented fiber structure, slightly loose/wavy fiber arrangement, low number of infiltrated inflammatory cells, low proteoglycan-occupied regions, and high collagen fiber-occupied regions, compared to the former healing phase. The results of the semi-quantitative grading score showed lower cell density, a slightly round nucleus, a slightly loose/wavy fiber arrangement, and a slightly fragmented fiber structure in the PM and TF groups compared to the control group. There is a significantly decreased mean number of infiltrated inflammatory cells and AB-positive proteoglycan-occupied regions, and increased MT-positive collagen fiber-occupied regions, compared to the control group. This indicates that the Kessler suture with PM and TF padding when repairing the completely transected Achilles tendon enhances healing compared with the sole Kessler suture.

Using fascia as an augmentation that wraps around the tendon can inhibit the blood supply to the healing area (32). PM provides scaffolding for fibroblasts and collagen, which can be formed to resemble normal, orderly, and histologically normal intact tendons (15). Therefore, this would also cause a difference in healing between the PM and TF. Advantages of autografts include no additional cost and minimal immune reaction (28). Owing to its low cost, additional surgery is necessary, which results in prolonged anesthesia and surgery time. However, a PM does not require additional surgery.

In conclusion, both of PM and TF provided potent tensile strength and supported healing with the evidence of histological examinations. This means that augmentation with PM is useful for repairing a completely ruptured Achilles tendon, without additional surgery for graft material harvesting.

To the best of our knowledge, no biomechanical and histologic studies have compared PM augmentation and TF augmentation for repairing a completely ruptured Achilles tendon.

The authors have no conflicting interests.