➢ The poor intrinsic healing potential of tendon and ligamentous tissue has driven the search for biological agents that can improve healing, most notably platelet-rich plasma (PRP) and autologous stem cells.
➢ Current evidence best supports the use of PRP as treatment for chronic, degenerative tendinopathies such as lateral epicondylitis of the elbow or patellar tendinopathy. It does not support the use of PRP to promote tendon or ligament-to-bone healing in rotator cuff repair or anterior cruciate ligament (ACL) reconstruction.
➢ Clinical evidence regarding the use of autologous stem cells as treatment for tendon and ligament abnormalities is currently limited. However, the initial results appear to be promising, particularly following rotator cuff repair.
➢ A major limitation in the evaluation of current data is the lack of standardization in the preparation and composition of PRP, which are often not reported. Future investigators should make an effort to report the method of preparation and final composition of PRP used.
➢ Autologous stem cells were most often harvested from bone marrow aspirate but were occasionally derived from dermal tenocyte cells. Recent studies have explored new sources for stem cells, including adipose tissue, peripheral blood, and human amniotic tissue. We expect that clinical data evaluating these treatments will soon emerge.
Disorders of tendons and ligaments are a common cause of orthopaedic complaints. Many of these disorders are attributable to the low intrinsic healing potential of these structures, which have poor vascular supply and often degenerate with age1. The combination of poor tendon-healing and tissue degeneration can also make surgical repair difficult, with high structural failure rates having been reported even for common procedures such as rotator cuff repair2,3. These findings have prompted a search for biological agents that can augment the host healing response and thereby regenerate healthy, native tissue.
Normal tendon-healing progresses through a sequence of inflammation, repair, and remodeling, resulting in the formation of scar tissue that is histologically distinct from tendon1,4,5. This scar tissue is characterized histologically by hypercellularity and disorganized type-III collagen6 and has been shown to have inferior biomechanical properties compared with native tendon4-6.
Healing of the enthesis, where tendon inserts into bone, is even more complex. The native enthesis is composed of four distinct zones: tendon, unmineralized fibrocartilage, mineralized fibrocartilage, and bone7. This gradual transition plays a crucial role in reducing stress at the interface between the compliant tendon or ligament and stiff bone8. Soft-tissue-to-bone healing, however, results in the formation of a fibrovascular scar without a graded fibrocartilage interface, which has been shown to exhibit inferior biomechanical properties8. This same fibrovascular scar tissue has been shown to form after surgical attachment of tendon to bone5,9.
The poor healing response of tendons and ligaments is thought to be due to the low vascularity of these tissues, which limits the delivery of necessary cells, nutrients, and growth factors1. In vitro and animal studies on the application of cells or growth factors to healing tendon have shown promising results in terms of the regeneration of healthy tissue with improved mechanical properties1,5, leading to attempts to translate these results into clinically meaningful treatments.
Platelets have been shown to contain numerous growth factors associated with the healing response, including insulin-like growth factor (IGF), transforming growth factor-β (TGF-β), and vascular endothelial growth factor (VEGF)10, which have been shown to mediate bone maintenance, cell migration, and angiogenesis, respectively6. Platelet-rich plasma (PRP) is derived from whole blood processed via centrifugation so that it contains a hyperphysiological concentration of autologous platelets. The local concentration of growth factors has been shown to be exponentially increased following PRP injection10, which has fueled recent interest in the use of PRP as treatment for a variety of orthopaedic disorders. One method of PRP preparation can be seen in Video 1.
Stem-cell therapies are also being investigated for their ability to aid in tissue regeneration. Stem cells are unspecialized cells with the ability to self-renew and to differentiate into various tissue types5. Although stem cells can be derived from both embryonic and adult sources, most studies involve the use of adult bone marrow stromal or mesenchymal stem cells (MSCs) because of ethical and regulatory constraints5. Autologous MSCs can be harvested from multiple sources, such as bone marrow1,5, adipose tissue11,12, peripheral blood13, and even subacromial bursa5, making them an attractive option for potential treatments.
MSCs can differentiate into multiple cell lineages, including osteocytes, chondrocytes, tenocytes, and adipocytes, in response to environmental stimuli5. Therefore, MSCs theoretically have the ability to regenerate complex, non-homogenous tissue types such as the tendon-bone transition at the enthesis, and animal studies have confirmed that treatment with MSCs can promote the formation of a cartilaginous transition zone during tendon-bone healing14,15. Animal studies have shown that tendinopathies that are treated with MSCs demonstrate more normal histological characteristics16, improved biomechanical properties17, and better functional outcomes after healing18. Clinical data remain limited but are emerging rapidly.
The purpose of the present article is to provide a thorough review and critical analysis of currently available clinical data on the use of PRP and autologous stem cells as treatment for various tendon and ligament abnormalities. We hope that this review will help clinicians to make informed treatment decisions, highlight the limitations of currently available data, and identify areas where more study is needed.
Materials and Methods
We conducted a comprehensive search on the U.S. National Library of Medicine’s PubMed database for all clinical trials involving the use of biological agents as treatment for tendon and ligament disorders by searching all combinations of the terms platelet rich plasma, PRP, stem cell, and bone marrow aspirate with tendon, ligament, rotator cuff, lateral epicondylitis, tennis elbow, patellar tendon, jumper’s knee, anterior cruciate ligament, ACL, and Achilles tendon. A total of 2307 citations were identified from initial electronic searches.
The search was limited to articles published in the English language in peer-reviewed journals up to August 1, 2015. Only studies involving human trials were included. Case reports, literature reviews, abstract-only publications, and letters to the editor were excluded, as were in-progress studies. For studies investigating the use of PRP, only controlled trials were included. Because of the limited availability of high-quality clinical data on cellular treatments, uncontrolled studies and case series were included. Two studies were excluded because they included patients with a variety of different tendinopathies without stratifying the results according to treatment site. A total of fifty-six studies are included in this review. An additional three articles were identified and were assessed to be continuations of already included studies; the results of those three studies are also discussed in this review. Articles were classified on the basis of the pathological condition treated, as described below.
Rotator Cuff Tears
The prevalence of rotator cuff tears has been estimated to be 20.7% in the general population, with the prevalence increasing to 50% among those over eighty years of age4. As such, surgical repair of the rotator cuff is a common procedure and has been associated with substantial functional improvement4. Despite good clinical outcomes, retear rates of >80% have been reported3-5,19,20. The high failure rate is thought to be related to the poor tendon-healing response4,5; thus, biological agents that augment healing would have the potential to benefit a large number of patients.
We identified seventeen studies investigating the use of PRP as an adjunct for surgical repair of rotator cuff tears2-4,19-32, including thirteen randomized controlled trials (RCTs); one prospective cohort study; two case-control studies; and one cohort study involving a retrospective control group. The number of patients ranged from twenty to eighty-eight. All patients were managed with arthroscopic rotator cuff repair; the studies varied as to whether concomitant procedures such as acromioplasty and/or biceps tenodesis were always performed, were performed if indicated, or resulted in exclusion of patients from the study. The most common outcome measures were (1) magnetic resonance imaging (MRI) assessment for the evaluation of retears and (2) clinical outcome scores, most commonly visual analog scale (VAS) scores2-4,19,21,22,27,29,30,32, the Disabilities of the Arm, Shoulder and Hand (DASH) score3,21,22,30,32, and the University of California at Los Angeles (UCLA) shoulder-rating scale4,19-22,24,27,31.
Radiographically, PRP failed to show a clear benefit in terms of lowering retear rates. Of the fourteen studies that assessed the integrity of the cuff repair at least six months postoperatively, four demonstrated that PRP treatment was associated with lower rates of retear21,23,26,29, ten demonstrated no difference3,4,19,22,24,25,27,28,30,31, and one demonstrated that PRP treatment was associated with a higher rate of retear20. Charousset et al. reported that, although they found no significant difference in terms of failure rates, the retears tended to be smaller in the PRP group24.
While nearly all studies demonstrated improvement in terms of pain and functional outcomes following rotator cuff repair, very few studies demonstrated a meaningful difference between the treatment and control groups in terms of clinical outcome. Studies investigating pain and function in the immediate postoperative period also failed to show that PRP had a significant benefit with regard to reducing postoperative pain or accelerating the return of function2,32.
Two studies investigated the use of stem cells as a surgical adjunct during rotator cuff repair. Ellera Gomes et al., in a case series of fourteen patients, used bone-marrow aspirate, harvested from the iliac crest, as an adjunctive treatment during mini-open rotator cuff repair33. Twelve months postoperatively, MRI showed intact repairs in all fourteen patients; this rate of success was noted to be higher than historical averages33. The authors also reported substantial clinical improvement following treatment as measured with use of the UCLA shoulder-rating scale. However, clinical improvements are expected following rotator cuff repair surgery, so the effect of bone-marrow aspirates on clinical outcomes cannot be determined without a control group.
In a larger study of ninety patients with a minimum of ten years of follow-up, Hernigou et al. compared outcomes following arthroscopic rotator cuff repair with and without MSCs34. At six months, the investigators found complete tendon-healing, assessed with MRI, in all forty-five patients (100%) in the treatment group, compared with thirty (67%) of the forty-five patients in the control group. These differences were sustained over the long term: at ten years, the investigators found an intact rotator cuff in thirty-nine patients (87%) in the treatment group, compared with twenty patients (44%) in the control group. Interestingly, when examining the treatment group, the investigators found that the six patients who had a retear of the rotator cuff had had a lower number of MSCs transplanted in comparison with the thirty-nine patients who did not have a retear. The number of transplanted MSCs was found to be the most relevant variable affecting outcome, with the authors recommending injection of at least 30,000 cells per shoulder to obtain optimum results.
Two RCTs investigated the use of PRP as primary treatment for rotator cuff tendinopathy without concomitant surgical treatment35,36. Rha et al. reported that PRP led to better reduction of pain and disability for at least six months in comparison with dry needling35. However, Kesikburun et al. found no significant difference in outcome at any time between the treatment and control groups36.
Finally, we identified one pilot study on the use of autologous MSCs as primary treatment for rotator cuff tears. Centeno et al. found substantial reduction in pain and disability following the injection of autologous MSCs37; however, the study lacked a control group, and more research is needed to determine the efficacy of autologous stem cells as a primary treatment for rotator cuff tendinopathy.
Anterior Cruciate Ligament (ACL) Rupture
Rupture of the ACL is common, and surgical ACL reconstruction is among the most frequently performed orthopaedic procedures. After reconstruction, two biological responses occur: incorporation of the intra-articular ligamentous graft and healing within the bone tunnel38,39. A key limitation associated with ACL reconstruction is the lengthy recovery time associated with graft maturation, during which time patients are unable to return to athletic activity40-42. Another common finding following ACL reconstruction is widening of the bone tunnels, which has been reported to occur in 25% to 100% of cases43. The potential benefits of biological augmentation of ACL reconstruction include accelerating graft maturation, reducing tunnel widening, and reducing donor-site morbidity.
We identified thirteen articles investigating the use of PRP as a surgical adjunct during ACL reconstruction38,39,41-51, including nine RCTs38,42-44,47-51, one retrospective cohort study46, and three case-control studies39,41,45. A variety of different grafts were used, with eight studies involving the use of hamstring tendon as graft source, two involving the use of bone-tendon-bone (BTB) autograft, one involving the use of BTB allograft, one involving the use of tibialis allograft, and one involving the use of different graft types that were selected on the basis of patient characteristics.
The most commonly used outcome measure was graft maturation and incorporation into osseous tunnels as seen on MRI38,41,42,44,45,47,49-51. However, the radiographic parameters that were used to quantify graft maturation and healing varied among studies. Other outcomes included tunnel widening43,47,48,50, instability on clinical examination43,48,49,51, and clinical outcome scores46-48.
Radiographic results have been mixed and sometimes contradictory. Two studies indicated that PRP accelerated the process of graft maturation as measured with MRI signal intensity42,47, whereas four studies did not demonstrate a significant difference as measured with the same parameter44,45,49,51. Vogrin et al. reported increased vascularization of the tibial tunnel (assessed with MRI) in the PRP group at the time of the first follow-up (at four to six weeks); however, this difference did not remain significant at later time points50. No study indicated that PRP had a significant effect in reducing tunnel widening43,47,48,50 or improving postoperative stability38,48,49,51.
We identified one study that investigated the use of autologous stem cells for ACL reconstruction40. In that study, Silva et al. prospectively randomized forty-three patients to ACL reconstruction with or without stem cells. In the study group (twenty patients), bone marrow was aspirated from the anterior iliac crest and bone marrow concentrate was injected into both the femoral end of the graft and the femoral tunnel during ACL reconstruction. The investigators found no significant difference between the groups in terms of graft-to-bone healing (assessed with MRI) at the time of the three-month follow-up.
We identified two RCTs investigating the use of PRP for reducing donor-site morbidity following BTB autograft harvest52,53. De Almeida et al. reported a reduced patellar tendon gap area and significantly reduced pain in the immediate postoperative period when patients who had received PRP at the harvest site were compared with those who had not52. Similarly, Cervellin et al. found better healing and improved clinical scores when patients who had been managed with PRP were compared with those who had not at one year of follow-up, although the differences were only significant (p = 0.041) for the Victorian Institute Sport Assessment (VISA) score53.
Lateral epicondylitis, or tennis elbow, is a common condition, affecting 1% to 3% of the population each year54. Symptoms consist of local elbow tenderness and pain with resisted wrist extension at the origin of the extensor carpi radialis brevis (ECRB)55. The cause is thought to be related to overuse, leading to microtears in the origin of the ECRB on the lateral epicondyle54. Histologically, lateral epicondylitis is characterized by a process of angiofibroblastic and myxoid degeneration without acute inflammation, which is thought to be secondary to failure of the natural tendon-repair mechanism56. Because of its high prevalence and poor natural healing response, lateral epicondylitis has become an area of interest for the use of regenerative medicine.
We identified seven studies—all RCTs—investigating the use of PRP for the treatment of chronic lateral epicondylitis55-61 as well as one continuation of an already included study62. The number of patients ranged from twenty-four to 230, with three trials including at least 100 patients55,57,60. The control group varied among trials, with PRP being compared with autologous whole blood injection (two studies57,59), bupivacaine injection (two studies55,61), corticosteroid injection (two studies56,60), and both corticosteroid injection and saline solution injection (one three-armed study58). In all studies, outcomes were measured on the basis of clinical outcome scores, most frequently VAS, Patient-Rated Tennis Elbow Evaluation (PRTEE), and/or DASH scores.
The results of the use of PRP for the treatment of lateral epicondylitis were fairly consistent and appeared to depend heavily on the time after injection. In all three studies in which PRP was compared with corticosteroid injection, corticosteroids outperformed PRP at early time points (approximately four to six weeks after injection)56,58,60. However, PRP was consistently found to outperform both corticosteroids and bupivacaine injection at time points greater than three to six months after treatment55,56,60,61. Gosens et al. found that patients who had been managed with PRP still had substantial benefits at two years after injection, whereas patients who had been managed with corticosteroids had largely returned to baseline scores62.
We identified two studies investigating the use of stem cells for the treatment of lateral epicondylitis; both were uncontrolled pilot studies54,63. Connell et al., in a study of twelve patients with refractory lateral epicondylitis who were managed with the injection of autologous cells derived from skin tenocyte-like cells, reported that clinical scores improved following injection and that ultrasound showed a decrease in the number of tears, an increase in the number of new vessels, and an increase in tendon thickness54. Singh et al., in a study of thirty patients who were managed with stem cells derived from autologous bone marrow aspirate, also reported improvement in clinical scores following injection63. Neither study demonstrated any adverse reaction linked to treatment. RCTs are needed to evaluate the use of stem cells for treating lateral epicondylitis.
Patellar tendinopathy, also known as jumper’s knee, is a painful knee condition often seen in athletes participating in high-impact, jumping sports64,65. The condition is thought to be due to overuse leading to micro-ruptures of the patellar tendon, activating a poor tendon-healing response66. The end result is a chronic degenerative process characterized histologically by microtears of the collagen fibers, incomplete healing, and neovascularization67, without an underlying inflammatory component66. These findings have led physicians and researchers to move away from anti-inflammatory treatments for patellar tendinitis and to search for biological agents that promote a native healing response.
We identified three studies investigating the use of PRP as treatment for patellar tendinopathy68-70. Filardo et al., in a nonrandomized prospective cohort study of thirty-one patients, found that PRP treatment did not lead to significant improvement in terms of pain; however, patients who were managed with PRP and physiotherapy had greater improvement in the level of sports activity in comparison with those who were managed with physiotherapy alone68. Dragoo et al., in a small RCT of twenty-three patients, found that patients who were managed with PRP had better improvement on the VISA questionnaire at twelve weeks in comparison with those who were managed with dry needling69. This benefit was no longer apparent at the time of the latest follow-up (twenty-six weeks). Finally, in a larger RCT of forty-six patients, Vetrano et al. reported that patients who were managed with PRP had better improvement in terms of clinical scores at six months and one year in comparison with those who received extracorporeal shock-wave therapy70.
We identified two studies that investigated the use of stem cells for the treatment of patellar tendinopathy71,72. Pascual-Garrido et al., in an uncontrolled pilot study of eight patients, investigated the use of autologous bone marrow aspirate and reported substantial, lasting symptomatic improvement without adverse events71. In the RCT by Clarke et al., which included sixty knees (forty-six patients), patients who were managed with injection of skin-derived tenocyte cells (a type of collagen-producing stem cell derived from dermal fibroblasts) suspended in autologous plasma were compared with patients who were managed with injection with autologous plasma alone72. The authors reported faster and greater improvement in VISA scores in the patients who were managed with tenocyte cells, with the improvement persisting at the time of the latest follow-up (six months).
Achilles Tendon Disorders
Disorders of the Achilles tendon generally fall into two main categories: Achilles tendon ruptures, which are acute injuries, and Achilles tendinopathy, which is a chronic, degenerative process. For both conditions, the poor regenerative capability of the Achilles tendon makes treatment difficult73.
We identified two studies investigating the use of PRP for the treatment of acute Achilles tendon rupture74,75. Schepull et al. evaluated PRP as an adjunct to surgical repair of Achilles tendon rupture75, and Kaniki et al. investigated the use of PRP for the nonoperative treatment of Achilles tendon rupture74. Neither study demonstrated that PRP provided any beneficial effect.
De Vos et al. performed an RCT to investigate the use of PRP as treatment for chronic Achilles tendinopathy without surgery76, and two later articles provided additional follow-up data on the same study group73,77. The investigators reported that PRP had no significant benefit in comparison with the injection of saline solution with regard to clinical outcome scores76,77 or tendon appearance on ultrasound73,77.
Many common orthopaedic complaints can be attributed to the poor healing response of tendons and ligaments. Therefore, biological agents that can enhance this poor native healing response have the potential to benefit many patients. In the present article, we reviewed the currently available clinical data regarding the use of PRP and stem cells for the treatment of orthopaedic soft-tissue abnormalities.
PRP has the ability to bring high concentrations of growth factors to poorly vascular tendon or ligament tissue, with many of these growth factors having been shown to improve tendon-healing in in vitro or animal studies. However, translating these laboratory results into a clinically meaningful treatment has so far met with mixed success.
A major limitation in the ability to draw conclusions regarding the efficacy of PRP is the lack of standardization among studies. The final platelet and leukocyte concentrations of PRP, the volume of PRP used, and the application protocol have varied considerably. To improve standardization, Mishra et al. proposed a method for categorizing PRP on the basis of platelet concentration, leukocyte concentration, and activation status (Fig. 1)78. Unfortunately, a majority of identified studies did not include enough information to classify the PRP used, making comparison difficult.
For example, of the seventeen studies investigating PRP as an adjunct for surgical rotator cuff repair, only seven provided the platelet concentration and thirteen provided the leukocyte concentration of the PRP used. The platelet concentration ranged from 1.7 to 8 times the baseline concentration, with a median value of 3.5 times the baseline concentration4,21,22,26,29-31. Of the four studies in which the platelet concentration was ≤3.5, only one showed improved radiographic outcomes in association with PRP treatment. In comparison, two of the three studies in which the platelet concentration was >3.5 showed improvement in radiographic outcomes. This finding may suggest that higher platelet concentrations are more likely to produce beneficial results, but comparative studies are needed to confirm this hypothesis.
The leukocyte concentration of the PRP used also varied between studies. With regard to rotator cuff repair, nine studies involved the use of leukocyte-poor PRP (LP-PRP)4,19,20,21-23,29-31 and four involved the use of leukocyte-rich PRP (LR-PRP)2,5,24,26. In contrast, studies investigating lateral epicondylitis were more likely to involve the use of LR-PRP (three studies55,59,60) than LP-PRP (one study61). While few authors outlined their reasoning underlying the choice of PRP used, animal studies have shown that LR-PRP causes a greater early inflammatory response than LP-PRP79. This inflammatory response probably underlies why investigators were more likely to choose LR-PRP when treating lateral epicondylitis, as prolotherapies (injection of irritants to promote an inflammatory response) have been shown to be beneficial for patients with lateral epicondylitis80.
Accounting for differences in PRP composition may explain the mixed, sometimes contradictory, results that have been reported. Future investigators should make an effort to report on the composition of PRP used, including, at a minimum, enough information to classify the PRP according to the system described by Mishra et al.78. Comparative studies evaluating different types of PRP are also needed.
Despite these limitations, some conclusions can be drawn from the available data. Evidence is strongest for the use of PRP as a treatment for lateral epicondylitis of the elbow. While more limited, the initial results following the use of PRP to treat patellar tendinopathy also have been promising. It is noteworthy that both conditions arise from a similar pathophysiological process: both are initiated by overuse leading to microtears, leading to an incomplete healing response and chronic, degenerative process54,67. In contrast, the results were mixed at best following the use of PRP as a surgical adjunct during rotator cuff repair or ACL reconstruction, settings in which healing of the tendon-bone or ligament-bone interface is key.
Although there have been only limited clinical trials investigating the use of stem cells for the treatment of orthopaedic soft-tissue abnormalities, the initial results have been promising. The results have been particularly promising for rotator cuff repair, with the large study by Hernigou et al.34 demonstrating excellent results after long-term follow-up. As with PRP, it is important for future authors to comment on the preparation and composition of stem-cell treatments, especially the number of cells injected, as initial studies have shown that this variable may correlate with outcomes34.
In summary, a lack of standardization among studies severely limits the ability to draw conclusions from the currently available data on biological treatments for tendon and ligament abnormalities. Evidence is strongest for PRP as a treatment for lateral epicondylitis. Current evidence does not suggest that PRP provides a substantial benefit when used for the treatment of disorders of the rotator cuff or ACL. There are currently not enough data to strongly recommend for or against autologous stem-cell therapies, although the initial results appear promising. Our conclusions regarding the strength of recommendation for the use of PRP or stem cells as treatment for different soft-tissue abnormalities are summarized in Table I.
Investigation performed at the New York University Center for Musculoskeletal Care, New York, NY
Disclosure: No external funds were received in support of this study. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article.
- Copyright © 2016 by The Journal of Bone and Joint Surgery, Incorporated