Recent years have seen the flourishing of a completely new approach for the treatment of cartilage lesions. What we have seen is the transition from a traditional approach focusing on the concept of “repair” to the revolutionary idea of “regeneration” [1-3]. This fascinating perspective is one of the hottest topics of the current pre-clinical and clinical cartilage research: although orthopaedic practitioners have been always managing chondropathy and osteoarthritis, the most recent discoveries in the field of growth factors (GFs) have led to widespread enthusiasm about the application of innovative biological strategies for treating these conditions. A new figure is emerging: the “orthobiologist”, i.e. an orthopaedic surgeon specializing in biological and bio-engineered treatments, both conservative and surgical.

In this particular field, the role played by blood derivatives, and in particular Platelet-rich Plasma (PRP), is preeminent. Platelet-derived GFs contained in PRP are the most exploited way to administer a biological stimulus to several different damaged tissues, such as cartilage, tendons and muscle that might benefit from this particular approach [4]. Being able to treat patients with a product derived directly from their own blood is an attractive proposition due to the theoretical reduced risks of intolerance and side effects than those commonly ascribed to traditional commercial drugs.

Concerning the application of this biological treatment, cartilage is one of the most targeted tissues [5]. In fact, the incidence of this kind of lesions is increasing in relation to the ever-growing interest in sport which, beyond some unquestionable beneficial effects on health, is also responsible for both traumatic and degenerative lesions of musculo-skeletal tissues [6, 7]. Young people, the most sport-active population, are often affected by these lesions. The treatment of cartilage pathology in these patients is often conservative, due to the limits of the current surgical treatments for cartilage lesions and the lack of indication for invasive joint metal resurfacing [3]. The constant research for innovative solutions has been one reason for the booming interest in biological approaches.

The biological rational behind this kind of treatment is the topical administration of several important molecules normally involved in joint homeostasis, healing mechanism and tissue regeneration. First of all platelet-derived GFs, which are a group of polypeptides playing important roles in the regulation of growth and development of several tissues, including cartilage. Platelets contain storage pools of GFs [4, 8, 9] such as: platelet-derived growth factor (PDGF); transforming growth factor (TGF-β); platelet-derived epidermal growth factor (PDEGF); vascular endothelial growth factor (VEGF); insulin-like growth factor 1 (IGF-1); fibroblastic growth factor (FGF); epidermal growth factor (EGF) etc.

Alpha granules are also a source of cytokines, chemokines and many other proteins [4] involved in stimulating chemotaxis, cell proliferation and maturation, modulating inflammatory molecules and attracting leukocytes [4]. Besides alpha granules, platelets also contain dense granules, which store ADP, ATP, calcium ions, histamine, serotonin and dopamine, that also play a complex role in tissue modulation and regeneration [10]. Finally, platelets contain lisosomal granules which can secrete acid hydrolases, cathepsin D and E, elastases and lisozyme [11], and most likely other not yet well characterized molecules, the role of which in tissue healing should not be underestimated. Several in vitro and in vivo animal studies have showed the potential beneficial effect of PRP in promoting cellular anabolism and tissue regeneration [8] and set the rational for the application of platelet concentrates in humans.

The exact definition of PRP is not clear and will probably remain so for years to come. The reason lies in the fact that blood-derived products present so many variables to be considered when making comparisons that it is not possible to establish a univocal definition of PRP. It is more correct to acknowledge that there are several PRP formulations, whose comparison in terms of potentials and limits is very difficult to test due to the large inter-product variability. A good starting point might be to clarify what the essential features are to define a blood product as PRP: it is generally regarded as a blood derivate with a higher concentration of platelets compared to basal level. It has been proposed that platelet concentration should be at least 200% of the peripheral blood PLT count [12, 13] even if in the literature PRP concentrations have been reported to range widely, up to 8 times that of basal levels [12]. Therefore, just concerning the definition of PRP, it is possible to understand the first variable involved in the comparison of different types of PRP [14]: the platelet count that can vary in such a significant way that the possible correlation between platelet concentration and clinical outcome could be a critical aspect. That is not a negligible issue, because if the biological rational of this kind of therapy is the supplementation of GFs, the number of platelets is a crucial point in determining the total amount of GFs administered. However, with regards to the scientific evidence available up to now, good clinical outcome can be obtained even with lower platelet concentration. Some studies also report correlations between outcome and platelet count [13, 15] but further investigations are needed to fully explore this topic. Furthermore, the plasma itself contains important molecules involved in the healing mechanism of connective tissues, also contributing to the platelet stimulus to tissue regeneration.

The platelet count is strictly linked to the procedures employed: there are two main preparation methods used in clinical practice [14]. The first one is by using a laboratory centrifuge and the second one a density gradient cell separator. In the first case several variables, such as number of centrifugations, their speed and timing, might influence the final product in terms of concentration of different cellular types. Furthermore, this preparation method is more technician-dependent and therefore its reproducibility is biased. Concerning cell separators, they are closed-circuit devices that allow PRP preparation without excessive manipulation of the blood. A large and constantly increasing number of these devices are available on the market, each with its own features and specifications, so it is impossible to obtain the same products and again comparison is very difficult.

Furthermore, we have to consider the overall cell types contained in PRP: even in this case the preparation method is a key factor. In fact, some PRP preparations, besides plasma and platelets, also contain leukocytes and residual blood cells. With regards to this, some disagreement has emerged among authors. In fact, whereas the antimicrobial effect of leukocyte-rich PRP has been underlined, some authors highlight the fact that leukocytes can release matrix metalloproteinases and reactive oxygen species capable of damaging articular tissues and determining a catabolic effect. Concerning the presence of residual blood cells, it has been reported in some studies that erythrocytes have a pro-inflammatory effect in the joint [4].

Storage procedures are also a hot topic: in fact, freeze-thawing is one method but it is thought to impair platelet function and lifespan, and alter the GFs’ release pattern in a negative way, besides favoring the accumulation of pyrogenic cytokines and increasing the risk of bacterial proliferation. For these reasons some authors prefer fresh administration of PRP immediately after its preparation (thus requiring blood harvesting for each injection, in case of multiple treatments) [16].

Another issue is the activation method. Several options are available even in this case: from no activation, counting on the in-vivo activating effect of endogenous collagen, to the use of chemical agents or biomaterials or even physical agents. Calcium chloride and thrombin are the most commonly used substances, whereas batroxobin is less often employed. These molecules contribute to PRP clot formation and increase overall GFs release even if their effect on single GF kinetic is controversial and needs further studies. In particular, thrombin, besides being a strong activator of platelets, has its own biological properties that might influence the interpretation of the effect of PRP [17].

Beyond the intrinsic differences among PRP preparations due to all the aforementioned factors, applicative variables should also be taken into account. Individual authors apply their own therapeutic protocol: the amount of PRP used for each injection, the number of the injections and the time intervals can vary widely, so study comparison is even harder than just comparing products differences. Standardization has not yet been proposed on this issue, thus the different protocols of each author introduce further confounding factors. Furthermore, authors often report their experience with PRP applied to treat different stages of disease, ranging from chondropathy and early osteoarthritis (OA) to moderate/severe OA. This is worthy of attention since PRP can produce different effects according to the particular disease phase [18] and it will be an important issue to define exactly the treatment target [19] in order to determine clearly which specific stage of cartilage pathology can benefit the most from this treatment. In the meantime, this is a further variability factor when considering the results of a trial and when comparing trials, where the lack of homogeneity in patient selection can be considered another bias.

In the light of all the issues considered, there follows a summary of the factors playing key roles in study comparison [14]:

• PRP preparation method;
• PRP formulation and cellular content, including concentration rates and cell counts;
• PRP storage modalities;
• PRP activation methods;
• PRP therapeutic protocols (amount, number of injections, intervals etc.);
• Disease phase treated.

The scenario is therefore very complex and clinicians, up till now, have made their decisions without being guided by real scientific evidence. The lack of commonly accepted guidelines and the variety of biological and procedural differences partially explain the limits of the available literature and justify the need to perform high quality trials to clarify the high number of still open questions [20].

Cartilage pathology has been treated both surgically (Table 1) and conservatively (Table 2) by the application of platelet-derived GFs. The most widely studied joint is without doubt the knee, and the majority of studies report results of the intra-articular administration of PRP.

Just a few months ago the scientific literature on clinical PRP application consisted mainly of reviews or experts’ opinions rather than trials [42]. The debate has always been fervent but, without the support of robust clinical trials, it was inconclusive. Besides the large amount of in vitro and animal studies, only a few papers discuss clinical results and their average quality was poor, being mostly case series. In the last year, and especially in very recent months, a step forward in the literature has begun, especially with regards to conservative therapy, and hopefully this quality improvement of the new data will help to answer some of the open questions on PRP.

The present authors critically reviewed the available literature. For what concerns the current understanding on the potential and feasibility of applying PRP in the management of cartilage pathology, first of all, looking at the surgical application it is not possible to draw definite conclusions about the efficacy of this approach. It is a consequence of this particular kind of treatment: it is very difficult to identify how much PRP might contribute to determine the clinical outcome with respect to the surgical treatment performed alone. Comparative studies aimed at assessing the specific role of PRP are needed. Furthermore, in many cases PRP is administered together with other biological augmentation methods, such as mesenchymal stem cells [24-26] or bio-engineered scaffolds [22, 23], so it is even more difficult to determine the contribution of PRP. The studies available are just case series treating disparate conditions in biomechanically very different joints (knee and ankle). Maximum follow-up evaluation is 24 months, so further studies are needed to determine the persistence of the good clinical outcome of these particular procedures. In the near future PRP will be more and more widely used in cartilage regenerative techniques but, despite being safe, according to the present evidence there is still no recommendation for using PRP in such procedures.

Considering PRP as conservative management, things look a bit different, especially when treating the knee region. In fact, whereas a few studies have been published both on hip and ankle cartilage lesions, several trials focus on knee treatment. As a consequence, few indications can be proposed for hip and talar osteochodral lesions. In the first case we have just two case series [34, 35] confirming the safety of PRP and a promising clinical outcome but also a gradual worsening over time: in moderate or severe OA a beneficial effect has not been observed or expired so fastly that patients then required a prosthetic solution. Therefore, at present there are no studies to support an indication for PRP as first line treatment in hip degenerative pathology. For what regards talar osteochondral lesions, only one comparative quasi-randomized trial has been published [41]: despite presenting statistically superior results for the PRP group, the low number of patients treated and the short follow-up evaluation do not allow, even in this case, the use of PRP to be endorsed.

Going back to the knee, it is possible to have more detailed indications. A lot certainly still needs to be clarified, but the considerable amount of studies published allows us to draw some conclusions. The first consideration regards safety of the procedure, which was confirmed by all trials that reported only minor adverse events with just some differences linked to the particular PRP formulation used: in fact, leukocyte-rich PRP seems to determine increased pain and swelling reaction when compared to leukocyte-poor PRP. With regards to the clinical outcome it can be said that this kind of conservative approach is time dependent, since a gradual worsening occurs over time. Filardo et al. [28, 29] were the first to calculate the average duration of the PRP effect, which is estimated at about 9 months. Analyzing efficacy, the encouraging results reported in case series cannot be the backbone to support the clinical application of PRP since each study faces an important bias related essentially to the lack of a control group. On the other hand, comparative trials found that PRP responsiveness might be linked to the stage of disease, where better results, superior to HA formulations, are obtained in young patients affected by early degree of cartilage degeneration, whereas moderate or severe OA show less favorable outcome without difference compared to viscosupplementation.

All the randomized controlled trials available compared the efficacy of PRP versus HA. Actually it should be pointed out that each trial employed a different type of PRP. Two out of three studies [36, 37] revealed better results in the PRP group at 6 months of follow-up. The third one led by Filardo [38], which had the longest follow-up evaluation of patients treated (1 year), reports no overall difference between PRP and HA in terms of clinical outcome; just a tendency towards better results in patients affected by a lower degree of cartilage degeneration was found. Therefore, in the case of more advanced signs of OA, PRP does not seem to be superior to viscosupplementation, a conclusion that was also reached in the study by Sanchez et al. Surprisingly, Cerza et al. reported significantly better results for ACP even in grade III OA: worthy of consideration is the fact that they used a different blood-derived products with evaluation limited up to 6 months of follow-up.

All the randomized trials deal with the application of PRP in patients affected by very different stages of disease, from chondropathy to severe OA. Therefore no conclusions can be drawn about the possibility of applying this approach to a specific phase of cartilage degenerative pathology: sub-group analysis does not allow, in any of the trials published, sufficient statistical strength to provide a real clinical indication. What emerges can be considered just a “suggestion” to avoid the indiscriminate use of PRP, which seems to offer a clinical benefit but cannot yet be considered a first line treatment for this pathology. Moreover, its use should be limited to those patients who can take most advantage from this approach, i.e. young patients with less articular degeneration, or those not responsive to other more traditional treatments.

A lot has still to be understood about PRP for the treatment of cartilage lesions. The only aspect that seems to have enough scientific evidence is the safety of this procedure, which is the first prerequisite for endorsing this approach for clinical practice. However, at present there is no conclusive indication for the use of PRP in this kind of pathology, due to the fact that only a few high quality, randomized trials have been published, mainly about knee pathology. A clear superiority with respect to other traditional treatments has not been fully proven: the great enthusiasm about biological treatments, initially justified by encouraging preliminary results, now needs real support. For the moment PRP cannot be considered as first line treatment for cartilage pathology, and its application should be reserved to patients who, based on the current scientific evidence, can obtain the best results from this approach. Further studies are needed to clarify some fundamental aspects such as the best PRP formulation, the best protocols of administration, and also the patients’ and lesions’ characteristics correlating with the clinical outcome.

The authors confirm that this article content has no conflict of interest.