We conducted a prospective, randomized, blinded, controlled trial on thirty-two patients with Achilles tendinosis, including twenty-four with unilateral involvement and eight with bilateral involvement. Inclusion criteria were based on the clinical and sonographic evidence of Achilles tendinosis; symptoms of more than six months; and failure of conservative treatment including rest, analgesia, immobilization in a boot cast, acupuncture, and physiotherapy. Patients were initially assessed by a foot and ankle specialist for clinical signs and symptoms of Achilles tendinosis such as pain, swelling, and focal tenderness. The diagnosis of Achilles tendinosis was further confirmed with an ultrasound scan, which demonstrated noninsertional fusiform thickening of the Achilles tendon with or without areas of neovascularization or interstitial tears. The study was confined to midportion Achilles tendinopathy; patients with insertional tendinosis were not included. Patients with previous tendinous injections, previous surgery, and a bleeding tendency were excluded from the study.
There were twenty male and twelve female patients, and the mean duration of symptoms was 17.2 months (range, seven to thirty-five months). The patients ranged in age from twenty-two to sixty-seven years, and the mean age was 45.2 years. The twenty-four patients with unilateral involvement were randomized into a treatment group (twelve patients) and a control group (twelve patients) (Fig. 1). The eight patients with bilateral involvement (sixteen Achilles tendons) were also equally randomized into treatment and control groups, with eight limbs in each group (Fig. 2). The patients were randomized with use of a sequence of random numbers from a computer-generated sequence. Blinding was carried out at all evaluations. Patients were followed for six months.
To comply with the regulatory guidelines, all patients had undergone serological tests to exclude syphilis, hepatitis B and C, and human immunodeficiency virus.
All patients had undergone physiotherapy assessment and treatment and had been given a standardized program of increased eccentric loading and stretching exercises for six months that had been designed by Fahlström et al.31. Subsequently, an ultrasound assessment was repeated, and all of the patients were reassessed with use of visual analog scale (VAS) and Victorian Institute of Sport Assessment (VISA) scoring systems (described below), which were labeled as baseline. At this point, patients who had shown substantial improvement were excluded from the study. However, the patients who failed to show any sign of improvement were randomized into two groups. Informed written consent was obtained from all of the patients prior to any intervention.
With use of a 4-mm-diameter punch-biopsy needle, a skin sample was obtained from the lateral aspect of the thigh on all patients. To save costs, only samples from the cell therapy group were processed, cultured, and expanded in number to produce a collagen-producing fibroblast population. The skin samples were sent to the GMP (Good Manufacturing Practices) laboratory in cell transport medium (Dulbecco's Modified Eagle Medium [DMEM/F-12; Invitrogen, Carlsbad, California] and gentamicin) with 4° to 10°C refrigeration. A computer tracking system was used to generate bar codes for all samples to avoid any mix-up.
Skin samples were washed in a phosphate-buffered saline solution and underwent mechanical disintegration. Additionally, HEPES (4-[2-hydroxyethyl]-1-piperazineethanesulfonic acid)-containing medium with type-1 collagenase and fetal calf serum were added to digest the connective tissue. At fourteen to twenty-four hours after sampling, phosphate-buffered saline solution was added to the samples and the connective tissue cells were isolated by centrifugation. A fibroblast medium was added to each pellet in a 50-mL tube. The medium containing the connective tissue cells was placed into a tissue culture flask and kept in an incubator (5% CO2 at 37°C). Fetal bovine serum was used as a medium additive for the propagation of the connective tissue cells. In the laboratory, tenocyte-like cells were successfully grown in the culture media (Figs. 3-A and 3-B). These skin-derived fibroblasts showed exponential growth in cell medium culture, and a mean of 17.3 million cells (range, 10 to 28 million cells) was achieved over a four-week period in each preparation. The criteria for successful growth were cell viability (a mean [and standard deviation] of 90% ± 10% were living cells), Cluster of Differentiation 90 (CD90) expression for fibroblast marker (a mean of 80% ± 20% were positive), CD56 expression for muscle marker (a mean of 90% ± 10% were negative), CD34 expression for bone stem-cell marker (a mean of 90% ± 10% were negative), and a cell count of approximately 10 million. The expanded cells showed tenocyte-like behavior in an ex vivo stretch model (stretch-oriented collagen type-I and II production). The cultured skin fibroblasts demonstrated histological appearances, such as elongated nuclei and spindle-shaped cytoplasm, similar to those of tenocytes. Cells were characterized by flow cytometry32,33 and the releasing criteria for tenocyte-like cells, including positive CD90 marker and mean viability by 7-aminoactinomycin D staining of 90% ± 10%.
All subjects had 10 mL of venous blood aspirated from the antecubital fossa. The blood was injected into a tube containing citrate, after which it was centrifuged (2000 rev/min for fifteen minutes in an LC6 centrifuge; Sarstedt, Montreal, Quebec, Canada) to obtain the plasma supernatant. The control group subsequently had their sample discarded. All patients were asked to lie prone on the examination table with their feet dangling over the edge of the bed. Hypoechoic areas and intratendinous tears were identified on ultrasound prior to injection. All patients then received a local anesthetic with a bolus injection of 5 mL of bupivacaine (0.25% Marcaine; AstraZeneca, London, United Kingdom) on the ventral surface of the Achilles tendon in its midsection with use of a 21-gauge needle. In the treatment group, the needle was repositioned into the tendon substance so that an additional injection could be made. A combination of cells with autologous platelet-rich plasma (labeled the implant stage) was injected into the site of tendinous and fibril discontinuity of the Achilles tendons of the cell group with use of a double-barreled syringe technique containing equal amounts (2 mL) of cell suspension in one chamber and autologous centrifuged plasma in the other.
Following the procedure, patients were asked to remain in the prone position for five minutes without plantar flexing the foot. Exercise and heavy lifting were not allowed for the ensuing forty-eight hours. The patients were examined by the physiotherapist on the day of treatment and were given a leaflet with instructions to begin a standardized eccentric-loading physiotherapy program after forty-eight hours; they returned for follow-up evaluations at one, three, and six months.
At the first visit (labeled the screening stage), all patients were assessed for pain and functional disability parameters with use of validated outcome measures. Subjective and objective assessments of the Achilles tendon were achieved by gauging functional status with use of the VISA questionnaire34, a validated index of Achilles tendinopathy that includes questions regarding symptoms, functional tests, and ability to play sports. All patients were asked to complete the VISA questionnaire before undergoing an ultrasound scan or injection at the screening, baseline, and implant stages and at the six-week, three-month, and six-month follow-up evaluations. Six questions in the questionnaire regarding the level of health were scored independently with use of a VAS, ranging from 0 to 10 (with 0 being the worst and 10 being the best level of health). The VISA score ranged from a maximum of 100 for a physically fit and asymptomatic person to a minimum of 0 for a symptomatic individual who was not physically fit.
Ultrasound assessment was performed at the time of recruitment (the screening stage); six weeks following physiotherapy, at the time of skin biopsy (baseline); at the time that the cells were implanted; and at six weeks, three months, and six months after the cells were implanted (Figs. 4-A and 4-B). Patients were scanned in the prone position with the feet hanging over the edge of the table. The Achilles tendons were scanned in both longitudinal and transverse dimensions on gray-scale and color Doppler imaging with minimal pressure applied35 and were assessed for thickness, echotexture, interstitial tears, and neovascularity36-38.
The trial was registered with the National Research Ethics Service (National Patient Safety Agency) in the United Kingdom, with a registration number 08/H0724/27.
Statistical Analysis
In this prospective, randomized, controlled, blinded trial, cell treatment was compared with physiotherapy treatment (the control). Five repeated measurements, including the baseline primary outcome measure with the VISA score and the secondary outcome measure with the VAS score, were carried out. Both VISA and VAS scores were tested for normal distribution, which was not fulfilled across all repeated measurements; therefore, nonparametric significance testing with use of the Mann-Whitney U test for between-group comparison was applied. In total, five between-group comparisons were performed. To correct for multiple comparisons, the Bonferroni procedure was applied and p values of <0.001 were considered to indicate significant differences.
Source of Funding
The study was funded by the Austrian biotech company Innovacell.
Twenty-four patients with unilateral involvement were randomized into two groups, with twelve patients in the cell therapy group and twelve patients in the control group. With use of the Mann-Whitney U Test, significant differences in the VISA scores were observed between groups at the second visit (p = 0.02), at three months (p = 0.007), and at six months (p < 0.001). The median VISA values increased from 35.00 (before treatment) to 79.50 (six months after treatment) in the treatment group, while the values marginally increased from 30.50 (before treatment) to 34.00 (six months after treatment) in the control group. The between-group VAS scores were significantly different at the second visit and at six months (p = 0.014 and p < 0.001, respectively). The median VAS values decreased in the treatment group from 3.00 (before treatment) to 1.00 (six months after treatment), whereas the median values slightly decreased in the control group from 5.00 (before treatment) to 4.00 (after treatment). For the patients with unilateral involvement, the distribution of VISA and VAS scores before treatment and twenty-six weeks after treatment in the cell therapy and control (physiotherapy) groups is illustrated in Figures 5 and 6. No adverse effects were observed in the cell group.
The eight patients with bilateral involvement were analyzed separately. The eight patients (sixteen Achilles tendons) were divided equally into treatment and control groups, with eight Achilles tendons in each group. With the number of patients studied, there were no significant differences in VISA and VAS scores between treated and control groups, with the exception of the VAS score at six months (p = 0.038).
Although the literature has recently reflected a great interest in platelet-rich plasma as a treatment modality in chronic tendinopathy, a recent study by de Vos et al.39 found no advantage after platelet-rich plasma injection in the Achilles tendon compared with placebo. However, this treatment continues to be advocated in sports medicine circles.
Over the past few years, the application of mesenchymal stem-cell therapy in promoting Achilles tendon-healing with use of animal models has shown promising early results40,41. One of the major disadvantages of therapy with mesenchymal stem cells, however, is their potential to differentiate into undesirable tissue types such as bone, cartilage, or muscle tissues. Tendon-derived progenitor cells have also been utilized for the treatment of tendon disorders42,43. Nonetheless, compromising the donor site for this purpose remains a great obstacle in any future development of this technique.
Both animal and human studies have produced encouraging early results in the utility of skin-derived fibroblasts in promoting tendon-healing44,45. Additionally, skin cells are widely available and relatively easy to harvest. In a stretch model, skin fibroblasts can express the extracellular matrix glycoprotein tenascin and develop gap junctions, enabling collagen fibers to align in accordance with tension lines46. Moreover, skin fibroblasts produce type-I and III collagen fibers along the tenocytes when used with plasma gel and subjected to mechanical stimulation-stretch by careful training of the tendon involved. In our study, plasma scaffold was used to apply linear periodic stretch to these cells, which led to increased type-I and III collagen expression.
In a pilot study of twelve patients, Connell et al.47 demonstrated that the use of skin-derived fibroblasts in the treatment of refractory lateral epicondylitis is both safe and effective. Our study is a further extension to these findings. Although no fully objective examination data supporting the improvement were reported, our preliminary short-term results showed improvement in the pain (VAS) and function (VISA) scores in the cell therapy group of patients with unilateral involvement. However, no significant differences were observed in the VISA or the VAS scores of the patients with bilateral involvement. This could be attributed to the small sample size of the patients with bilateral involvement and to the fact that patients with bilateral Achilles tendinopathy might have found it difficult to score each side independently.
It is presumed that the combination of cell preparation and plasma together forms a gel-like structure, which can be used to fill up interstitial tears in the tendon tissue. Additionally, the gel helps to retain injected cells within the tendon tissue. A second effect of the plasma gel is the transfer of the mechanical force through to the cell surfaces. This transfer of the mechanical force appears to stimulate the cells to produce and secrete collagen fibers that help to repair the damaged tendon. We believe that by placing collagen-producing cells into the site of a tendon injury and/or tears, these cells will lay down the so-called building blocks for repair, i.e., type-III collagen. Furthermore, the addition of plasma also utilizes any potential positive effects of growth factors present in the plasma.
One limitation of this study is the small number of patients included in the trial. The pretreatment factors such as sensitivity to physiotherapy might have affected the subsequent differences in the intergroup functional scores of the patients with unilateral involvement. The patients with bilateral involvement had to be analyzed separately, and the results demonstrated no significant difference between the treatment and control groups. The patients with bilateral involvement could not have been considered independent because of the likelihood of systemic effect. The exact numbers of cells or injections that are actually needed to achieve the optimum regeneration of the tendon remain uncertain. We were not able to perform histopathological correlation, as no surgical specimens were available. Further studies are recommended to test the effectiveness of this procedure in patients with insertional tendinosis. Furthermore, the long-term efficacy and safety of this novel technique needs to be determined through larger series with a longer duration of follow-up. We remain uncertain as to whether needling of the tendon during the procedure could have contributed to the healing response. Moreover, because plasma was used as a carrier medium to allow cells to be embedded in a set gel, the possible effect of growth factors on tendon-healing could not be excluded. Further studies are needed to clarify this potential synergistic effect.
Many challenges in the field of cell therapy exist and should be recognized. These include transportation of the cells, storage requirements, and costly cell preparation. In addition, cells need to be manufactured in a dedicated facility that must meet stringent regulatory guidelines. Regardless, cell therapy remains an exciting potential treatment for Achilles and other tendon disorders.
In conclusion, the use of skin-derived fibroblasts in the treatment of Achilles tendinopathy is safe. Although this novel technique may have a role in ameliorating a patient's symptoms and physical performance, larger studies with longer-term follow-up periods are needed to achieve maximum healing potential and determine the long-term effectiveness.