CORR Synthesis: What Is the Role of Platelet-rich Plasma Injection in the Treatment of Tendon Disorders?

Abstract

In the Beginning… The term platelet-rich plasma (PRP) can be traced back to the 1960s and 1970s, when hematologists used it to describe blood plasma with supraphysiologic concentrations of platelets [2, 3]. PRP was used as a treatment for patients with thrombocytopenia [2]. When used for this indication, it was derived by pooling blood from a single donor or multiple donors and given as an intravenous transfusion. In the 1980s, researchers began to demonstrate the clinical applications of PRP beyond hematologic conditions. Small series from three decades ago explored the use of PRP in patients with leg ulcers and as autologous transfusions in cardiac surgery [20, 29], but these indications have since fallen out of favor. In the 1990s, maxillofacial surgeons used PRP as an alternative to fibrin glue in stabilizing oral implants [6], and to this day still has applications in dentistry [7]. Although orthopaedic surgeons began using PRP sparingly in the 1990s for fracture care, it was not until 1999 that PRP was used in sports medicine. Dr. Allan Mishra used PRP to treat an Achilles tendon injury in Steve Bono, the quarterback of the San Francisco 49ers that year, which sparked the interest of the sports medicine community [43]. However, Dr. Mishra did not publish his work on PRP for elbow tendinosis until 2006 [39]. Since then, clinical use of PRP in treatment of tendon pathologies has rapidly expanded, as has research on the topic. The Argument Physicians have continued to expand the indications for using PRP over the past decade. PRP has been used for operative and nonoperative management of tendinopathy and osteoarthritis and as an agent to promote bone healing [15, 26, 44]. Although PRP is seeing wide clinical use, some have suggested that the supporting evidence—in terms of high-quality clinical research—is lagging behind the pace of adoption of this new treatment [13, 33]. Clinicians have pointed out that the enthusiasm for the potential of biologics, rather than sound science demonstrating whether the potential can be achieved in practice, has driven widespread use [13]. In turn, the industry surrounding biologics has grown to an estimated USD 7.4 billion in the United States in 2017 [9]. A subsequent rise in advertisements aimed at consumers have ignored science and misled patients [28]. On a larger scale, enthusiastic PRP adoption has implications for society as a whole. Some authors have pointed out that continued funding of biologics research on debunked science causes an opportunity cost that impairs the progress of more-promising treatments [37]. These same authors argued that physicians must work with regulatory authorities and medical boards to promote guidelines for biologics and fulfill our obligations as scientists [37]. Physicians have a responsibility to understand the research behind PRP to navigate this ethically complicated set of circumstances, and to provide the best care to patients. Essential Elements We sought to identify all randomized controlled trials (RCTs) published in the last decade about PRP in the treatment of chronic tendinopathic conditions of the Achilles tendon, patellar tendon, rotator cuff, and common extensor origin of the lateral epicondyle. We searched the MEDLINE and Cochrane databases on March 1, 2020 to include articles published from January 1, 2010 to March 1, 2020. We searched for the terms “platelet rich plasma”, “platelet plasma”, “autologous blood”, and “platelet concentrate” to capture the wide terminology used to describe PRP. The following terms were added for each tendon injury: “Achilles tendon”, “rotator cuff”, “lateral epicondylitis”, and “patellar tendon.” This initial query returned 683 results in total, with 153 for Achilles tendinopathy, 107 for patellar tendonitis, 166 for lateral epicondylitis, and 257 for rotator cuff pathology (Fig. 1). Duplicates were removed, and then the records were screened to remove all studies without English full-text articles available. Also, all studies that were not therapeutic human trials were removed. This yielded 87 articles, many of which were excluded because they involved non-PRP biologic substances, used PRP as an adjunct to surgery, or studied acute tendon ruptures. A total of 25 relevant RCTs were included: four on Achilles tendinopathy, three on patellar tendonitis, 11 on lateral epicondylitis, and seven on rotator cuff tendinopathy.Fig. 1: The search and study selection process is shown here.All RCTs were scored by one author (MWL) using the Jadad scale [24] from 0 to 5 (higher scores represent greater methodological rigor). A score of 4 or greater was required for study inclusion in the review. On the Science The clinical potential for PRP is rooted in sound science. Platelets, which are acellular, contain alpha granules that house a vast array of cytokines necessary for biologic healing [3]. When hemostasis is disrupted, such as in musculoskeletal injuries, the platelets are activated and the alpha granules are released [3, 6]. By injecting a patient’s own platelets to create a local supraphysiologic concentration, the homeostatic response is upregulated to promote healing [3, 6]. In practice, after centrifuging whole blood there remains a “buffy coat”, which contains leukocytes and PRP [18]. The injection mixture is classified as leukocyte-rich or leukocyte-poor, indicating the relative concentration of these cells [18]. There is a lack of consensus regarding the most effective cellular composition for each treatment indication [34]. It is important that readers are aware of PRP’s wide array of cellular compositions in clinical practice, which may affect the healing response. In a tendon injury, the pathway for healing begins with the inflammatory cascade, followed by angiogenesis, activation of stem cells, and finally regeneration of the healthy tendon [6]. PRP acts on each step of this pathway to promote tissue regeneration by modulating proinflammatory cytokines and cell migration [6, 12, 36]. Initially, PRP increases vessel permeability through the release of catecholamines and expression of vascular endothelial growth factor [3]. Subsequently, PRP promotes neutrophil chemotaxis through the release of chemokines [4, 6]. Angiogenesis and arteriogenesis are then influenced by growth factors released from platelets (vascular endothelial growth factor, platelet-derived growth factor, and transforming growth factor-beta 1) to restore blood perfusion to the injured tissue [10]. Finally, mesenchymal stem cell proliferation and migration is stimulated by chemoattractants released from platelets [6, 51]. These pathways, among others, act synergistically to promote the healing cascade under the influences of PRP. Specific to tendon injury, several studies have demonstrated PRP has a positive effect on tenocytes [5, 25, 58]. Inflamed tenocytes demonstrate robust healing in response to PRP in vitro [5]. PRP also has pleiotropic effects on tenocytes in vitro via cytokine regulation [25]. PRP may also promote a protective environment, as tenocytes were spared from the harming effects of dexamethasone and ciprofloxacin when treated with PRP [58]. These studies have provided a solid foundation and rationale for the implementation of PRP in tendon injury treatment. What We (Think) We Know PRP has been studied in a variety of tendon injuries, most notably injuries of the Achilles tendon, patellar tendon, common extensor tendon of the lateral epicondyle, and rotator cuff tendons. Because PRP is autologous, it is believed to have a favorable safety profile and has been widely studied with many RCTs. However, efficacy is in question, and may vary depending on the indication for which it is used. Achilles Tendinopathy Although treatment for insertional tendinopathy of the Achilles largely is rooted in eccentric calf muscle training [1], PRP has been explored as an adjunct to the current regimen to accelerate healing and shorten the recovery time [11, 16, 17, 31]. Several RCTs demonstrated no improvement for this condition treated with PRP. One small RCT found no difference between a placebo saline injection and a single PRP injection for the treatment of Achilles tendinopathy 24 weeks after injection [17]. Although this is an important finding that questions the efficacy of PRP for Achilles tendinopathy, the time point is relatively short given the chronicity of this condition. Additional investigation of this same patient population was carried out to 1 year, again finding no difference in functional outcomes or tendon structure under ultrasound imaging [16]. The use of ultrasound imaging, although inferior to MRI, provided additional insight into the effects of PRP and strengthens this study. A more recent study likewise found no improvement in Victorian Institute of Sports Assessment – Achilles Questionnaire (VISA-A) scores but noted ultrasonographic improvement in the thickness of the tendon after a single injection of PRP [31]. Although these studies are relatively few in number, the cumulative results indicate that PRP is no better than normal saline for the treatment of Achilles tendinopathy. In contrast, one RCT found mild improvement after PRP injection compared with placebo. This study had three arms: the placebo group received a single saline injection, the treatment group received four PRP injections at 2-week intervals, and the third group received a single injection consisting of corticosteroid, saline, and anesthetic [11]. Although PRP improved pain and functional outcomes compared with placebo, it was not better than corticosteroid. Importantly, although there was improvement in muscle function for all groups, there was no difference among the groups. The improvement in VISA-A at 24 weeks was 11 points for the PRP group, which exceeds the minimum clinically important difference (MCID) of 6.5 [38]. However, this improvement was no different from that observed in the corticosteroid group. Furthermore, this protocol involved four PRP injections in comparison with a single corticosteroid injection. The cost and time investment to perform four PRP injections for a mild benefit—no better than that of a single corticosteroid injection—does not validate PRP as a viable treatment option for this condition. Despite several high quality RCTs, conclusive evidence in support of PRP to treat patients with Achilles tendinopathy is lacking [33]. Patellar Tendonitis Patellar tendonitis tends to occur in young, jumping athletes because of repetitive eccentric contraction of the extensor mechanism. The treatment of this condition includes activity modification and physical therapy focused on eccentric training. PRP has been studied as an alternative to surgery in recalcitrant pathologic conditions. Although an initial RCT showed promise [56], additional RCTs have not confirmed it [19, 54]. In 2013, a small study compared PRP with extracorporeal shock wave therapy and found improvement in Victorian Institute of Sports Assessment – Patella Questionnaire (VISA-P) and VAS scores at 1 year [56]. The differences were small, with a VISA-P of 13.7 at 1 year, which is about the same as the MCID, which has been reported to be 13 [22]. The difference in VAS was 1.7, which is not much greater than the MCID of 1.5 [35]. This study must be interpreted with caution given the small effect size and lack of control group. A subsequent study compared a single PRP injection with dry needling, noting an improvement in VISA-P scores among the PRP group compared with the dry needling group at 12 weeks but no difference at 26 weeks [19]. A more recent large, multicenter RCT to date showed no difference between patients undergoing PRP injections and those with a placebo [54]. Fifty-seven patients with more than 6 months of symptoms were randomized to a single leukocyte-rich PRP, leukocyte-poor PRP, or saline injection in addition to a standard physical therapy regimen. There was no difference among the three groups in VISA-P scores or pain score at any timepoint up to 1 year. This study not only had a larger sample size but also followed patients for a longer duration. Overall, the evidence to date does not support the use of PRP for the treatment of patellar tendonitis. Until or unless a convincing study were to say otherwise, the evidence for use of PRP in patients with patellar tendinitis does not justify the practice. Lateral Epicondylitis Lateral epicondylitis is a chronic condition defined by angiofibroblastic dysplasia of the common extensor tendon of the elbow [30]. The mainstay of treatment is nonsurgical, including activity modification, offload bracing, and corticosteroid injections. PRP has been widely studied as an alternative to corticosteroids. Two large studies performed early in the 2010s in support of PRP sparked interest in its use in lateral epicondylitis [21, 40], which has remained a relatively common indication to this day. The first study followed 100 patients treated with PRP or corticosteroid injection for 2 years, noting improvement in the PRP group in pain and function from 26 weeks through 2 years [21, 46]. Although the results at 2 years showed a VAS improvement of 2.1, which is greater than the MCID of 1.5 [35], no study over the last decade has been able to replicate a result of this magnitude. This study, despite data collection at several time points, found no difference until 26 weeks post-injection. With a natural history of resolution, a difference seen only at long-term follow-up may simply confirm this fact and may be explained by patient selection differences. The second RCT was performed in 2014 at 12 centers during a 5-year period, enrolling 230 patients to receive PRP or dry needling [40]. At 12 weeks, there was no difference in pain scores scores between the two groups; at 24 weeks there was an improvement but its effect size was small, below the MCID of 1.5, and so that difference would probably not be considered important by patients. Other studies since have shown mixed results. When compared with whole blood injections, PRP improved the VAS at 8 weeks, but by an amount that is unlikely to be perceptible to patients (0.9 out of 10, well below any reasonable MCID) [48]. This trial was repeated on a larger group with 1-year follow up, which showed no difference in pain scores throughout the study [47]. Two studies compared PRP with corticosteroid and placebo, and they found no differences [32, 45]. Ultrasonography was also performed, showing improvement in tendon characteristics in the corticosteroid group only [32]. More recent studies have continued to show minimal or no improvement with PRP. A direct comparison between a single injection of normal saline, leukocyte-rich PRP, and leukocyte-poor PRP found no difference in clinical outcomes up to 8 weeks [57]. One study randomized patients to receive two ultrasound-guided PRP or saline injections and demonstrated no difference in pain scores up to 1 year [41]. Although this study lacked functional outcomes and study size was relatively small, the methodology was stringent. Functional outcomes, as evaluated by the DASH score were no different at 6 months between PRP and saline injections [52]. Most recently, a larger study of 119 patients found no clinically important difference after PRP treatment as measured by VAS and DASH [35]. This particular study used an MCID of 1.5 for the VAS and noted that although no difference was found, the 95% confidence intervals did not exclude the MCID. The authors indicate that if this study missed a difference that exists, the effect size is likely rather small. Overall, the evidence is inconclusive with regard to the use of PRP to treat lateral epicondylitis. Although early work supported PRP for lateral epicondylitis, the effect sizes and study quality of those papers are of concern. The most recent studies indicate PRP may not be an effective treatment for lateral epicondylitis. As an alternative to corticosteroid injections, PRP remains a markedly more expensive treatment modality and one that lacks strong supporting evidence. Clinicians must interpret the initial supportive results with caution. Rotator Cuff Tendonitis Rotator cuff disease affects many people with a wide range of pathology, ranging from inflammation to complete tears. Although tendonitis and partial tears are largely managed nonoperatively with physical therapy and corticosteroid injections, full-thickness tears often result in surgical repair. PRP is being explored as a subacromial injection for nonoperative treatment. PRP has been compared with saline injection and dry needling in this area [27, 49, 53]. In one study, 40 patients with complete or partial rotator cuff tears on MRI were randomized to receive PRP or saline and showed no difference in VAS, Western Ontario Rotator Cuff Index, or Shoulder Pain and Disability Index (SPADI) at 1 year follow-up [27]. The issue with this study was the inclusion of both complete and partial tears, which makes it difficult to draw a definitive conclusion. In contrast, partial-thickness tears specifically were treated in three studies. Eighty patients were randomized to normal saline or PRP for MRI-confirmed partial supraspinatus tears without any clinical improvement in pain or function up to a year [53]. To strengthen this study, MRI scans were obtained at 7 months without any appreciable tendon healing in the PRP group. A contrasting study found improvement in SPADI and ROM for up to 6 months when PRP was compared with dry needling [49], but the effect size in this study was small, and probably not clinically important. The difference in SPADI at 6 months was approximately 11.8, which is smaller than the MCID, which ranges from 8 to 13 [50] (we believe it is prudent to use larger MCIDs when considering expensive or invasive treatments). Thus, the effect size should be questioned. Other trials have compared subacromial PRP injections with other treatments, such as physiotherapy or corticosteroid injections [8, 23, 42, 55]. When compared with a supervised exercise regimen, PRP was inferior [42]. Another more recent study found no difference in DASH scores through 6-month follow-up [8]. The authors of this study argued that PRP might be considered as an alternative when corticosteroid use is contraindicated because of patient factors. We believe there is little basis for this viewpoint. Given the additional cost and the need for an autologous blood draw for PRP delivery, a clinically meaningful difference in treatment effect size is needed to justify the use of PRP over corticosteroids. One study did note improvement with PRP over corticosteroid at 12 weeks post-injection, but this difference did not persist at 6 months [55]. These differences were small in magnitude and effect size again is of concern because the duration of improvement is short. This improvement at 12 weeks is therefore unlikely to be clinically important. The authors of another study comparing PRP with corticosteroids concluded that PRP is a safe alternative [23]. However, the data did not support their conclusion, as there was no clinical difference in outcomes between PRP and corticosteroid. There is minimal evidence to support the use of PRP as an injection therapy for nonoperative management of partial-thickness tears. Although most studies show no difference over placebo or corticosteroid, the few that show minor difference are of minimal effect and below the MCID. PRP should not be used for nonsurgical treatment of rotator cuff tendinopathy. Knowledge Gaps and Unsupported Practices There are a number of knowledge gaps of which the reader of research about PRP should be aware. Perhaps the most important of these is the wide variation in the content and quality of each injection. Each study refers to their injection as PRP, but we know there is a wide variability of cellular contents that are being described as PRP. Although describing the injections as leukocyte-rich and leukocyte-poor are an effort to classify the injection, this refers only to the cellular components and does not describe the platelet composition. We believe that not knowing the contents of each PRP injection is the single biggest knowledge gap. Additionally, the ideal protocol for PRP delivery is not well understood. RCTs have employed different approaches to the number of injections, timing between injections, location of the injections, and the use of image guidance. Variations in delivery and cellular heterogeneity is of concern when interpreting future studies and considering how to implement PRP in practice. The widespread clinical uses of PRP for tendon injuries all represent unsupported practices. The best evidence shows no clinical benefit when PRP is used to treat Achilles tendinopathy, patellar tendonitis, and rotator cuff pathology. Even for lateral epicondylitis, which is the best-supported indication, the effect sizes are small and the studies are of inconsistent quality. PRP is an expensive, invasive treatment, often with costly image-guidance needed for delivery without any definitive scientific support. Barriers and How to Overcome Them There remain several barriers regarding the clinical application of PRP. As mentioned previously, the scientific community has not yet agreed on the appropriate composition of biologic elements within PRP. Guidelines are needed to direct researchers on how to measure and report PRP contents. Further studies of PRP for specific disorders must focus on precise protocols and injection preparation methods to address this barrier. Another important barrier is an ethical dilemma. PRP, although no more difficult to deliver than a corticosteroid injection, is more costly. Although some insurers cover PRP despite the lack of supporting evidence, patients are also willing to pay out of pocket. Patients are desperate for a nonsurgical treatment and physicians are promoting PRP as a solution. We believe this is unethical. Until there is clear evidence, physicians have an obligation to their patients to rely on sound science when making treatment decisions. These barriers can be overcome with the help of strong leadership from medical and surgical societies, as well as national agencies. Guidelines are needed to promote quality research, determine a minimum standard for biologic contents, and outline appropriate treatment strategies supported by science. Furthermore, medical societies must lead against false claims by physicians regarding the potential of biologics. Some efforts have been made on this front. For example, the American Academy of Orthopaedic Surgeons (AAOS) published a consensus recommendation for biologics, specifically developing a list of minimum reporting standards for clinical studies evaluating PRP [14]. This is a step in the right direction, but further efforts by national leaders within AAOS and the American Orthopaedic Society for Sports Medicine are needed to guide the medical community. 5-Year Forecast Although laboratory science studies show some promise of PRP for the treatment of tendon disorders, clinical research does not support its use. Despite this, PRP has been used broadly. More encouragingly, though, the pace of inquiry on this topic is high, and we believe that within the next 5 years, it will become evident that most of the conditions now treated with PRP should not be. Although we believe PRP will continue to have a role in the treatment of very few orthopaedic conditions, it will be viewed as a historic footnote in the treatment of most tendon disorders.