Assessments of Therapeutic Effects of Platelet-rich Plasma in Knee Osteoarthritis: Possible Role of Inammatory Cytokines

Background: Osteoarthritis (OA) is a multifactorial disease that commonly affects the knee. Tumor necrosis factor-α (TNF-α) is able to regulate inammation in OA. Macrophage migration inhibitory factor (MIF) may be involved in the pathophysiology of arthritis. Platelet-rich plasma (PRP) may reduce pain associated with OA. The current study aimed to assess the possible therapeutic effects of PRP in patients with knee OA of various severities. Methods: A prospective study was performed on 90 patients were included and categorized into mild (30 cases, moderate (30 cases) and severe (30 cases) knee OA. Three intra-articular (I.A) injections of PRP, 2 weeks a part, were received. Pain score and MRI Osteoarthritis Knee Score (MOAKS) were assessed. Serial synovial uid cytokines assays in the form of Tumor necrosis factor-α (TNF-α) and Macrophage migration inhibitory factor (MIF), were performed using commercially available ELISA assay kits. The assays were performed pre-injection (S1), two weeks from the 1 st I.A injection and two weeks from the 2 nd I.A injection (S3) for all included patients. Results: The mean values of pain score and synovial TNF-α and MIF levels were signicantly higher levels (S1, pre-injection) among severe OA when compared with both mild and moderate cases, p (cid:0) 0.05 for all. There were signicantly lower pain score and synovial TNF-α and MIF levels at S3 in mild, moderate and severe knee OA when compared with S1 values, p (cid:0) 0.05 for all. There was signicant improvement in synovitis in both mild and moderate cases (p (cid:0) 0.05 for both). Conclusion: I.A injection of PRP signicantly reduces the synovial uid TNF-α and MIF levels with great therapeutic effects on both synovitis via reducing inammatory cytokines, and bone marrow lesions mainly for mild knee OA and to a


Background
Osteoarthritis (OA) is a disease primarily affecting the knee that is de ned as a progressive loss of joint function and pain from a gradual deterioration of the articular cartilage [1]. Unfortunately, it is a frequent site due to the high use and stress of the knee joint for painful conditions including OA [2].
OA was thought to be primarily a cartilage degenerative disease; however, the latest research has shown that OA is a multifactorial cause, involving numerous causative factors such as trauma, mechanical forces, in ammation, biochemical reactions, and metabolic derangements [3]. The role of in ammation is not well understood and there is an ongoing debate to determine whether the in ammatory reaction triggers changes in the OA or, rather, the in ammation is secondary to changes in the OA [3].
The tumor necrosis factor (TNF) α has been shown to be associated with the progression of osteoarthritis. It has been shown that TNF-α is capable of regulating in ammation in an osteoarthritis rat model by downregulating the signaling pathway of phosphoinositide 3-kinase / protein kinase B (PI3K / AKT) in synovial broblasts [4].
Macrophage migration inhibitory factor (MIF) is a pro-in ammatory cytokine produced by macrophages which may contribute to arthritis pathophysiology by promoting in ammation and angiogenesis [5].
Recently, current approaches to OA treatment have integrated the use of biologics which mediate the in ammatory process [1]. Platelet-rich plasma (PRP) is increasingly being used for its ability to in uence tissue regulation due to growth factors in high platelet levels, thereby reducing OA-related pain [6,7]. PRP is obtained after whole blood centrifugation, yielding a highly concentrated product with platelets. The αgranules within the concentrated platelet solution contain growth factors and proteins that are vital to the coagulation cascade [8]. PRP is thought to in uence degeneration of cartilage by altering the autophagy in chondrocytes. Aging cartilage gradually loses its reversible quiescence and its ability to self-renew [9].
The current work aimed to use clinical and radiological scores to determine the potential therapeutic effects of PRP in patients with various severities of knee OA. In addition, serial biochemical assays of the TNF-α and MIF levels of synovial uid and their associations with both pain and radiological scores were performed among these patients.

Study design and participants
The present prospective study was conducted on 90 patients with knee OA, recruited from the orthopedic outpatient clinics, Qena University Hospital-South Valley University-Egypt, during the period from February, 1st 2019 to January, 31st 2020. Based on radiological assessments, the included patients were categorized into three groups (mild, moderate and severe OA) according to the OA severity. Each group included 30 patients.
The exclusion criteria is as follows: polyarticular disease; knee arthroscopy in the previous year; HA or steroid IA penetration in the preceding 3 months; history of infectious disease and autoimmune disorders such as diabetes, rheumatoid arthritis, hematologic diseases (coagulopathy), serious cardiovascular diseases, infections or immunodepression; anticoagulant therapy or an anti-aggregating agent; use of non-steroidal anti-in ammatory drugs 2 weeks prior to blood sampling; and < 10 g / dL of hemoglobin [10].

Clinical assessments
Detailed medical history and thorough clinical examination from every included subject involving age, gender, weight, height, calculation of body mass index (BMI) (kg/m 2 ), were recorded. As regards the BMI, it is classi ed into: underweight (if BMI 18.5), normal weight (18.5-24.9), overweight (25-29.9), obesity class 1 (30-34.9), obesity class 2 (35-39.9), extreme obesity class 3 ( 40) [11]. Local examination of both knee joints and assessment of knee joint pain for the included patients was done using Visual Analog Scale (VAS) score [12]. OA diagnosis was based on visual identi cation of the presence of eburning on the articular surfaces of the distal femur, proximal tibia, or patella, right or left. Eburnation is a sclerotic, ivory-like subchondral bone reaction that takes place from bone-on-bone contact at sites exposed to advanced cartilage erosion [ 13.14]. The VAS is a common tool that uses a 10 cm scale for pain intensity measurement, where 0 = no pain and 10 = unable to move. Pain score assessment was performed before beginning PRP therapy and two weeks after the second PRP IA injection.
Radiological assessments of severity of knee OA MRI Osteoarthritis Knee Score (MOAKS) was used to assess the severity of knee OA [15]. In brief, for assessment of both bone marrow abnormality or patella-femoral cartilage volume (grade I (mild if the lesion involves 33% of subregional volume); grade II (Moderate: if the lesion involves 33%-66% of subregional volume), and grade III (severe: if the lesion involves 66% of subregional volume).Regarding synovitis: grade 1 (mild or small -uid continuous in the retropatellar space); grade 2 (Moderate or medium -slight convexity of the suprapatellar bursa), and grade 3 (severe or large -evidence of capsular distension ). Meniscal desintegrity score of 0-3 applied for amount of extrusion in 4 locations: medial meniscus (medial and anterior extrusion) and lateral meniscus (medial and anterior extrusion).

Biochemical workup
A. PRP preparation: The PRP needed for IA injection has been prepared under complete aseptic conditions. In order to avoid the effect of food intake on puri ed PRP, on the day of injection the patients were instructed to fast for 4 hours before blood collection. Approximately 20 mL of venous blood were drawn from the antecubital vein using an aseptic technique in an effort to avoid irritation and trauma to the platelets. The blood was obtained as an anticoagulant in ten extraction tubes (2 mL each), containing sodium citrate. The tubes were then centrifuged at 2100 rpm at room temperature for 8 minutes to separate the blood in each tube into the plasma, the buffy coat and the residual red blood cells ( RBCs) [10]. Using a pipette, the PRP situated just above the selectively precipitated RBCs but not including the buffy coat was carefully aspirated from each tube. In each affected joint, 10-mL PRP samples (from the 10 blood-collecting tubes) were obtained for each patient, and 5 mL were used for IA injection [11].
B. Serial assays of synovial uid pro-in ammatory cytokines (TNF-α and MIF): TNF-α and MIF were measured in the synovial uid samples, using commercially available ELISA assay kits supplied by Chongqing biospes, China, Catalog No: BYEK3327-48T, and BYEK3015, respectively, using microplate ELISA reader (EMR 500, USA), according to manufacture protocol. To extract cell debris, synovial uid samples were immediately centrifuged at 3000 rpm for 15 min, and the supernatant was aliquoted into 1-mL cryotubes and frozen at − 80 °C until use. The serial assays were performed to the synovial uid samples three times. First sample (S1) was pre-injection; the second sample (S2) was two weeks from the rst injection, and the third sample (S3) was two weeks from the second injection.

Procedure and timing of IA injections of PRP
The patient was put in 20 degree exion in supine position with the knee. Under aseptic conditions, a 21gage needle was used to inject 5 mL of PRP into the suprapatellar knee joint pouch, using a superolateral method. Local anesthetics were not used. Patients were instructed after the injection to refrain from physical exercise for at least 24 hours but no restriction was speci ed regarding activities of daily living. Three IA-PRP injections were administered at 2-week intervals. The same physician who was engaged in selecting and testing participants did the injections [10].
Statistical analysis IBM SPSS Statistics v.22 was used for analysis of the data. Mean ± SD was used for expression of quantitative data, while, number and percentage were used for qualitative data. Kolmogro-Smirnov and Shapiro-Wilk normality tests were used. Mann-Whitney U test was used when comparing between two quantitative data and one way-ANOVA was for comparison between more than two quantitative data. Post Hoc test was used for multiple comparisons between different variables. For qualitative variables, Chi-square (χ2) and Fisher's Exact tests were used. Pearson's correlation coe cient (r): test was used for correlating data. The level of signi cance was considered P value 0.05.

Effect of I.A injection of PRP on pain score among patients with knee osteoarthritis
There were signi cantly lower mean pain score values when assessed two weeks from the second I.A injection in patients with mild, moderate or severe knee OA (1.6 ± 0.5, 2.4 ± 1.9, and 5.1 ± 1.9, respectively) vs. the pre-injection pain score (3.1 ± 0.9, 5.9 ± 0.9, and 8.7 ± 0.7, respectively), p 0.05 for all (Table.3). There were no signi cant improvements in both patello-femoral cartilage volume and meniscal desintegrity in various severities of knee OA, p 0.05 for all (Fig. 2C, Table.4).
Therapeutic outcomes of the IA injection of the PRP in patients with various severities of knee OA Comparing the therapeutic e ciency of I.A injection of PRP among patients with knee OA, there was improvement in all patients with mild knee OA (30 cases, 100%), while 15 cases (50%) with knee OA showed improvement and only 6 cases (20%) of patients with severe knee OA showed improvement regarding pain score (Table.5).

Discussion
Knee osteoarthritis (OA) in subjects over 50 years of age is a major cause of pain and impairment with a serious effect on physical activity and quality of life [16]. The present study evaluates the possible therapeutic effects of I.A injection of PRP in patients with knee OA using pain score, MOAKS and biochemical serial assays of synovial uid cytokines levels in the form of TNF-α and MIF .
Obesity and overweight have long been recognized as important risk factors for OA, in particular knee OA [17]. The current study revealed female predominance with female: male ratio 9:1. Also, there was higher frequency for patients categorized as obesity class 1 followed by obesity class 2, while those categorized as overweight have the least frequency among patients with knee OA. These were corresponds with the ndings of previous studies [18][19][20][21][22][23]. The cause of female predominance may be multifactorial and these involve structural variations, prior trauma, genetic and hormonal disorders, as men had signi cantly higher total tibial and patella cartilage volumes than women and women had a signi cantly higher prevalence of baseline patellar cartilage defects. Over time women showed more volume loss in the knee cartilage compared to their male counterparts [24]. The key but not only mechanism by which obesity will lead to knee OA is possibly increased mechanical loading on the joint. Knee overloading may cause synovial joint breakdown and failure of ligament and other structural supports [24].
While platelet-rich plasma (PRP) has been approved as an agent for knee osteoarthritis therapy, different studies of PRP in the OA knee have found more consistently positive results compared to hyaluronic acid, other intraarticular injections and placebo than in other musculoskeletal tissues [25,26]. Although the causes of osteoarthritis in the knee are not fully understood, laboratory and clinical evidence suggest that in ammatory cytokines can contribute to its pathogenesis [27,28]. The present study revealed signi cantly higher synovial uid in ammatory cytokines (TNF-α and MIF) levels (S1, pre-injection) among patients with severe OA when compared with both mild and moderate cases with signi cantly lower synovial TNF-α and MIF levels two weeks after the second I.A injection of PRP (S3) in patients with mild or moderate or severe knee OA when compared with both the pre-injection synovial uid levels and two weeks from the 1st injection (S2) levels. These indicate that PRP treatment for patients with knee osteoarthritis had bene cial effects in regulating in ammatory factors. In ammatory cytokines have been documented to form a complex regulatory signal network in femoral head osteonecrosis which is mediated by different intracellular kinase signaling pathways to regulate the recruitment, stimulation and activation of autoimmune cells [29]. In accordance, Zhang et al [30] reported that MIF levels in synovial uid were independently associated with the severity of self-reported pain in OA patients but not in serum.
Huang et al [25] reported signi cantly down-regulated plasma concentrations of cytokines involving TNFα after PRP treatment in patients with knee OA.
The current study reported signi cantly positive correlations between the synovial uid TNF-α and MIF levels among the included patients with knee OA. MIF is a potent pro-in ammatory cytokine that can cause the release of many in ammatory cytokines such as interferon (IFN)-γ, interleukin (IL)-1β, 6, 8 and TNF-α by initiating an in ammatory cascade [31].
Regarding the assessments of the therapeutic effects of I.A injection of PRP on pain score and MOAKS in the included patients with knee OA, there were signi cantly lower mean pain score values when assessed two weeks from the second I.A injection in patients with knee OA, with signi cant improvement effects on synovitis and bone marrow lesions mainly for mild knee OA and to a lesser extent for moderate cases. Unfortunately, the included treated patients didn't show any signi cant improvements in the patellafemoral cartilage volume or meniscal disintegrity. In accordance, Burchard et al [32] suggested that intraarticular injection of PRP may improve symptoms of osteoarthritis and decrease pain in patients with knee joint osteoarthritis, independent of the level of cartilage damage quanti ed by the whole organ MRI scoring method. Furthermore, a study by Laudy et al [33] demonstrated that in patients with knee osteoarthritis, PRP injections resulted in decreased pain, improved function and global assessment, and changes in joint imaging.
Favorable outcome was noticed in all included patients with mild knee (100%), while it occurred in 50% of patients with moderate severity, and (20%) of patients with severe knee OA showed improvement regarding pain score. In line with our ndings, Taniguchi et al [10] reported that intra-articular PRP injection is likely to be a safe treatment option for Japanese patients with mild to moderate osteoarthritis of the knee and may have pain relief for up to 6 months. While PRP therapy has controversially discussed restorative effects on the cartilage, anti-in ammatory effects, down-regulation of cytokine levels and joint homeostasis could explain favorable effects in patients with severe osteoarthritis [34,35].
Increasing evidence shows that the over-expressed in ammatory cytokines in the in amed joints play an important pathophysiological role in generating and maintaining OA-induced pain by acting on nociceptive nerve cells [36]. Among the included patients with knee OA, there were signi cantly positive correlations between synovial uid TNF-α and MIF levels with the pain score. In consistent with our ndings, Stannus et al [37] reported that TNF-α was positively linked with total knee worsening and pain.
In conclusion, the current study con rms the presence of in ammatory process involved in the pathogenesis of knee OA including involvement of some in ammatory cytokines in the form of TNF-α and MIF. Additionally, the present study con rms that among the therapeutic mechanisms of using I.A injection of PRP in patients with knee OA is the anti-in ammatory effect, where PRP may lower the synovial uid cytokines (TNF-α and MIF) with subsequent improvement of pain and synovitis. This favorable outcome is prominently achieved in patients having mild and to a lesser extent moderate OA, thus could be established as safe adjuvant biologic therapy for such patients.

Study Limitation
Lack of long term follow up of the included patients, which could be approached in future research and relatively small sample size, were the main limitations of the study. Availability of data and materials

Abbreviations
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. P1= S1 vs. S2; P2= S1 vs. S3; P3= S2 vs. S1.
N.B: TNF-α: tumor necrosis factor-α; PRP: platelet rich plasma; OA: osteoarthritis; SD: standard deviation; S1: rst synovial uid sample assay (before IA injection of PRP); S2: second synovial uid sample assay (after 2 weeks of the 1 st IA injection of PRP); S3: third synovial uid sample assay (after 2 weeks of the 2 nd IA injection of PRP).     MRI analysis of knee joint before and after IA injection of PRP two weeks after the second injection, MRI signals showed that there were signi cant improvement in the synovitis (A) and bone marrow lesions (B) with no improvement in cartilage loss(C) compared with pre injection condition.