Author:Ji Liu, Wenqi Song, Ting Yuan, Zhengliang Xu, Weitao Jia, Changqing Zhang

Abstract

Platelet-rich plasma (PRP) has offered great promisefor the treatment of cartilage degradation, and has been proved to have positiveeffects on the restoration of cartilage lesions. But no comparative work has been done between PRP and hyaluronate acid (HA) concerning their restoring effect on cartilage defect, especially by means of animal experiments and histologic assessments. The purpose  of the study was tocompare the therapeutic effects of P-PRP and HA on osteoarthritis in rabbit knees. Thirtyrabbits were used to establish theanimalmodels by creating acartilage defect of 5 mm in diameter on the condyles of the femurs, and were randomly divided into three groups: theP-PRPgroup, HA group andthe control group. Then each groupwas treated with P-PRP, HA or saline solution, respectively. Six and twelveweekslater the rabbits were sacrificed andthe samples were collected. The plateletnumber, the concentrations ofgrowth factors of P-PRP and whole blood, and the IL-1β concentrationin the joint fluid were investigated, and the histological assessment ofthe cartilage were performedaccording to Mankin’s scoring system. Micro-CT was also used to evaluate the restoration of subchondral bone. The platelet concentration in P-PRP is 6.8 fold of that in the whole blood. The IL-1β level in the P-PRP groupwaslower than in the HA group (p<0.01) and in the control group(p<0.01). The restoration of the defected cartilage as well as the subchondral bone was better in the P-PRP group than in the HA group or the control group (P<0.05). Our data showed that P-PRP is better than HA in promoting the restoration of the cartilage and alleviating the arthritis caused by cartilage damage.

Introduction

Due to its poor blood supply and self-renewal capacity, the normal structure and function of cartilage are difficult to restorewhen it’s injured or degenerated. Thepathological changes involve the degeneration or inflammatory reactionof the cartilage, the subchondral bone and synovium, even the formation of osteophytes. The conventional treatments for cartilage degeneration or osteoarthritis (OA) include intra-articular injection of lubricant or anesthetic medication or arthroplastic surgery. Hyaluronic acid (HA) is a kind of acidic mucopolysaccharide, which has been used as a conventional lubricantfor degenerated joints for decades. It can lower the friction between the articular surfaces, and alleviate the joint pain of the patients,but its effect is not longstanding and in some occasions may cause allergicor inflammatory reaction.Platelet-rich plasma (PRP) isa concentrate of autologous platelets, and platelet concentratestechniques could be classified in 4 families based on their fibrinarchitecture and leukocyte content: pure platelet-rich plasma (P- PRP) and leukocyte- and platelet-rich plasma (L-PRP) are liquidplatelet suspensions; P-PRP gel and L-PRP gel are polymerized fibrin gel, when P-PRP and L-PRP are activated,respectively. Pure platelet-rich fibrin (P-PRF) and leukocyte- and platelet-richfibrin (L-PRF) are solid fibrin materials, and cannot be injectedas liquid solution. PRP has been widely used to promote the recovery of the soft tissue lesions, andhas been proved effective forrestoring the impaired cartilage or preventing the aggravation . However, the exact effectof PRP on the cartilagehas not been compared with that of HA on animal models, especially investigated with respect to histological changes and subchondralbone restoration. In thisstudy, P-PRP and HA were injected into theknee joint cavityof rabbits, and effects of both methods on the restoration of cartilage damagewere assessed and compared.

Materials and Methods

The preparation of P-PRP

Afterthe rabbits had spent one week in the cages, blood samples were extracted to prepareP-PRP according to Landesburg’s protocol as reported previously. Briefly, afterthe rabbits were anesthetized with intravenous injections of pentobarbital, 9 mlwhole blood was extracted from the central auricular artery with a 10 ml syringepre-filled with 1 ml 2.5% sodium citrate as anticoagulant. Analiquot of 20μL of each bloodsample was drawn out for cell counting, and the blood was centrifuged at 200 g for 10 min into three layers, the plasma, the plateletsand the red blood cells.Then the plasma and the platelets are extracted to undergo another centrifugation at 200 g for10 min, and most of the supernatant plasmais discarded. The plasma above the buffycoat layer was carefully collectedwith a pipette. The white cells were not collected in order to prepare the P-PRP. Another 20μl of P-PRP was used for cell counting. To activate the P-PRP, 1/10 volume of CaCl2 was added into the liquid P-PRP, which would turn into clots with lucid serum containing abundant platelet derivedgrowth factors. The serum was carefully collectedand wasthen kept in -80°C for further use. To compare the concentration of the growth factors in P-PRP and in the wholeblood, enzyme linked immunosorbent assay (ELISA)was performed.

Animal models andtreatment

Thirty-three adult rabbits (6–8 months old) were involved inthe study. All procedures and handling to the animalswas approved by the Animal Research Committeeof Shanghai JiaotongUniversity School of Medicine. Sixty knees in thirty rabbits wereused for the establishment of the cartilage defect model. Briefly, after the animals were anesthetized with intravenous injectionof pentobarbital, a lateral para-patellar skin incision was made. The knee jointwas exposed when the joint capsule was sliced open and the patellarwas extracted laterally.In direct vision, afull-thickness cartilage and subchondral bone defect that was 5 mm in diameter and 3 mm in depth was made in the patellar groove using a 5 mm sterilestainless drill. Then the rabbits were evenly dividedin a randomized manner into three groups:the P-PRP group, the HA group and the control group. And the rabbits of the three groups received 0.3 ml P-PRP, 0.3 ml low-molecular weight HA (mol wt 70,000–120,000, Sigma) solution,and 0.3 ml 0.9% salinerespectively, which was injectedinto both knee joints once a week for 3 weeks consecutively, with the first injection administered 3 days postoperatively. The remaining 3 rabbits were used as a positive control for the micro-CTscanning and were not given surgery or injection.

Histological assessment

Sixand twelve weeks after the first injection,the rabbits were sacrificed with overdose of pentobarbital (120 mg/kg) and the bilateral distal parts of femurswere dissected. For the thirtyrabbits which received surgery and injections, the samples were fixed in 4% paraformaldehyde for 48 hours,and then was dehydrated, embedded, cut and stainedwith H&E and Safranin-O, to observe the changes to the structure and ECM of the cartilage. The sections were examinedand scored in a blindmanner by three dependent trained investigators, accordingto Mankin’s score for osteoarthritis showed in table 1.

Micro-CT scanning

Twelve weeks after the operation, Micro-CT scanning (Sky- scan1172, Bruker-microct, Belgium)was used to observe the micro-structural restoration of the subchondral bone in the defectarea. After fixed in 4% formaldehyde for 48 hours, the samples were held immobile with the femoralaxis perpendicular to the scanning plane. The samples were scannedthrough a 180urotation angle with a rotationstep of 0.6uat 18 mm resolution. The region of interest (ROI) was cylinder-shaped in this study, which was extracted from the samplewith a delineation of a 5 mm circularcovering the original defect area on the first slice and included 56 deeper slices (adding up to 1 mm approximately). A global threshold was used for all samples to identify bone/non-bone tissue, which was set at 68–255. Then the 3-D images were established. The quality of the regenerated bone in ROI was assessed and presented with the average bone mineral density (BMD) and bonevolume/total volume (BV/TV). Thesamples from the 3 rabbits of the positive control were also scanned,to get the BMD and the BV/TVvalues calculated from the same ROI as in the other 3 groups.

Detection of the joint fluid IL-1βlevel

Before therabbits were sacrificed 12w postoperatively, the joint cavity was flushed with 0.3 mL normal saline for 3 times, and the flushing solutionfrom the joint cavity was collected fordetection of IL-1 concentration via ELISA method.

Statistical analysis

Data was collected from the triplicateor quadruplicate samples and were presented as mean standarddeviation (SD). The statistical process was carried out with the SPSS 11.5software and the statistically significant values were defined as p<0.05 orp<0.01 based on one-wayanalysis of variance(ANOVA).

Results

The assessment of the P-PRP

Themean platelet numberin the peripheral whole blood and platelet rich plasma was 25.56±10.54×10⁴/μL and174.33±35.89×10⁴/μL,respectively. The platelet concentration in PRP is 6.8 fold of that in the wholeblood (Fig. 1). And the

concentrations of PDGF, TGF-b1 and bFGFare significantly higher than those in the whole blood (Table.2).

Gross observation

After the surgery, there was slight swelling and redness with increased skin temperatureon the joints. Six weeks afterthe surgery, defects in the control group were filled with disrupted fibrous tissue, with little contactwith surrounding cartilage, And not much changestook place but only thickerfibrous coverage could be observed after another six weeks. In the HA group, the defectsdidn’t repair much better except the surface of the regenerative tissue was glossier. Thedefects treated with PRP werefilled with regenerated glossy white tissue, which appeared

integrated with normal tissue,though there was a slightconcavity and tiny fissures inthe center.

Histological assessment

The regenerative tissue in the controlgroup was thin and rough, even the subchondral bone was revealed partially withfissures. The Safranin-O staining was pale, indicating that the ECM wasreduced. In the HA group, the surface of the cartilage was smoother, with a fewregions of slightly restored transparent cartilaginous matrix, but the chondrocyteswere still reduced and dispersed. In the PRP group, notable proliferation of chondrocytesformed in clusters can be found in the H&E stained slices, which were embeddedin thick ECM as was shown in the safranin-O staining. Also capillary infiltrationcould be observed in the control group, which was not very obvious in the PRP group(Fig. 2). The Mankin’s score of the three groups were shown in Fig. 3. The scoreof the PRP group was lower than the other two groups (p<0.01).

Micro-CT assessment

Micro-CT scanning andanalysis was used to assess theregeneration of the subchondral bone in the defectarea. The calcified regenerated subchondral bone was indicated in orange in the 3D images,and tissue with lower densitysuch as fibroustissue was in deep blue. The average BMD of the PRP group was the highest among the three groups (P<0.05). There was no significant differencebetween the average BMD values of the HA group and the controlgroup (Fig. 4).

The IL-1 concentrationin joint fluid

The IL-1 concentrationin the joint fluid was highest in the controlgroup and lowest inthe PRP group, as is shownin Fig. 5.

Discussion

In this study, the effect of P-PRPon repairing the cartilage defect was evaluated and compared with HA. P-PRP and HA wasrespectivelyinjected into rabbits’knee joints wherecartilage defect model was made. The repairing effectof the two methods were compared via histological investigation, micro-CT analysis, and IL-1 test of the joint fluid. Our findingsdemonstrated that better histological outcomesand restoration of subchondral bone wereobtained with administration of P-PRP. We also found that P-PRP couldalleviate the inflammatory reaction in the joint cavitymore than HA, as was indicated with the changesof IL-1 level in the joint fluid.

Asmentioned above, the platelet concentrates family could be classified into four categories: P-PRP,L-PRP, P-PRF and L-PRF. In our study, we chose the P-PRP which containspoor white cells, inthat it is the restoringeffectof the abundantgrowth factors on the cartilage defect that we intend to observe.The effects of the cell contentsin PRP were not thetopic of investigation of this study.

HAis a kind of acid mucopolysaccharide produced by the synovium and can lubricatethe joint and protect the cartilage. In normal conditions, there is a 2 mm-thick coating of HA-protein compound on the surfaceof articular cartilage, conjugated with the collagen inthe surface area, which can maintain the balanceof the substance exchange between the joint fluid and the cartilage,and at the same time prevent the proteoglycan from being lost. So this may explain why intra-articular complement of HA can relievethe degeneration of the cartilage.

Thehistological findings showed that an increase of chondrocytes was observed after P-PRP was injectedinto the joint cavity,which was demonstrated via HE staining. And it was companied by increase of production of the cartilageextracellular matrix (ECM), confirmed via safranin-O staining. It is reasonableto relate this effect to the high concentration of the growthfactors released from P-PRP, which were producedby activated platelets and can promote the restoration of the injuredtissues. Moreover, the relation and interaction between the factorsis the key point of the P-PRP potency, and can work far more efficientlythan one single component. Also, it’s been reported that P-PRP canstimulate the articular chondrocyte proliferation and matrix biosynthesis. Among the growth factors from the platelets, IGF can stimulate the chondrocyte proliferation as well as the matrix synthesis; TGF-b canalso induce the transformation of the chondrocytes and the depositof the collagen. These bioactive growth factors accounted for the superiority of P-PRP over HA.

Besides,the micro-CT scanning showedthe reconstructive effect of P-PRP on the subchondral bone, which was not found inthe HA group or the control group. The subchondral area is quite crucial in cartilagereconstruction as it supports the overlyingneocartilage tissue. In our study, the tests of BMD and BV/TV were run concerning the originally defected area in all the groups. These results indicated both the quantityand the qualityof the neo-subchondral bone, which were highest in the P-PRP group. This proved that P-PRP couldeffectively promote the maturation as well as the calcium depositing of the subchondral bone. While in the HA group or the control group, less bone mass was formed. This difference could also be attributedto the growth factors in P-PRP. There has been studies demonstrating beneficial effectsof P- PRP on bone defect healing and bone formation,both experi- mentally and clinically.

Moreover, it was also verified in this experiment that P-PRP showed explicit alleviation on the osteoarthriticchanges, as was supported by the detection of IL-1β concentration in the joint fluid. Inflammatory factors such asIL-1β areindicators of the existence and severity of osteoarthritis. And IL-1β will promote the production of NO and accelerate the degradationof the cartilage by up-regulating the matrix metalloproteinase (MMP-1,13) and down-regulatingthe synthesis of the ECM. So the concen- tration of IL-1β in the synovial fluid and in the serumwould be obviously up-regulated, while there is barely any in normal joints. Theanti-osteoarthritic effect of P-PRP may be the result from the relief of the cartilagedamage, which at the same time reduced the local inflammation as well as the stimulationto the synovium, causing a drop in the IL-1β secretion. On the other hand, P-PRP has a direct influence onthe synovium, as it can lower the NF-kB activity, and suppress the expression ofCOX-2 and CXCR4, which is the important regulatory factor in the inflammatory reactions.Meanwhile, PRP can up-regulate the expression of HGF, IL-4 and TNF-a, while HGFand TNF-a can block the expression of NF-kB to inhibit the inflammation. All thesebio-active features account for the superiority of PRP over HA, which acts as lubricantbetween the fricative articular surfaces and complement to the cartilage ECM, butno efficacy to the restoration of lost chondrocytes.

As one of the novel biological treatments for cartilage lesions, PRP has been exploited widelyin recent years, especially in clinical application. In 2010, Kon et al. publisheda prospective studyon 115 knees of 91 patients, which were treated with 3 injections of 5 mL PRP (1 every 3 weeks). The satisfactory rate was eighty percentwithin the first 6months, whereas a tendency to worsen was reported at 12 months of follow-up. Theauthors confirmed the time-dependency of intra-articular therapy with platelet-derived GFs and estimated the median durationof the PRP effect to be 9 months.Another multi-center study compared the efficacy of PRP and HA intra-articular injections for the treatment of knee cartilage degenerative lesions and osteoar- thritis, which revealed that better results were achievedin younger and more active patientswith a low degree of cartilage degeneration. And this finding agreed with the results of our experiments. The cartilagedefect, or a fresh wound to the cartilage was treated with PRP within one week, which was actually an early stage of cartilage lesion, and revealed better response to PRP than HA.

Conclusion

From the resultsof the study, it can be drawn that P-PRP can effectively restore the defected cartilage,and alleviate the inflammatory reaction of the joint. Moreover, P-PRP is preparedtotally from the autologous whole blood and is therefore safe and of low cost, enabling it as a possible substitute to HA in treating degeneration or damage of articular cartilage.

  参考文献（略）