Protocol Heterogeneity in Platelet-Rich Plasma Treatment for Knee Osteoarthritis: A Scoping Review of Preparation, Delivery, and Outcomes

1. Introduction

The role of platelets in the therapeutic and healing process has been thought to be secondary to the vast array of growth factors released in the context of tissue injury. The acquisition of platelet-rich plasma (PRP) furthers the concept of growth factor and cytokines action delivered at supraphysiologic concentrations to directly impact the site of injury, stimulating recovery and regeneration.[1] Its effects have been extensively investigated in the pathology of knee osteoarthritis (OA), with the latter driven by cartilage loss and concurrent synovial inflammation, altering the structural and functional balance of the joint.[2,3,4,5] In the intra-articular environment, PRP has been demonstrated to possess anti-inflammatory and immunomodulatory properties, promoting anabolic activities that mitigate the deterioration of the knee joint.[6] This has been attributed to the platelet alpha-granules releasing an abundance of growth factors, cytokines, and other bioactive molecules.[7] Given the prevalence of knee OA and the excess morbidity associated with the disease across the globe, the minimally invasive technique of intra-articular PRP delivery serves to be an attractive and relevant therapy in the management of this condition.[8,9,10]

At present, several commercially available devices are available to produce the appropriate quality and volume of PRP for clinical application. However, there is significant variation in the manner of preparation and delivery, including the time and speed of centrifugation, the final volume of PRP derived, platelet concentration achieved, the technique of injection, and the frequency of PRP delivery.[11,12] Our scoping review seeks to delineate the most commonly adopted parameters in the production and delivery of PRP knee injections associated with positive therapeutic outcomes without significant adverse events.

2. Methods

The scoping review was conducted in accordance with criteria defined by Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA). Databases queried included MEDLINE and Embase and the query was performed on 15 November 2025 with time frame of article search set from database inception to 15 November 2025. The definition of search strategy was determined by two authors (QR and CR) and discussed and approved by all authors with necessary amendments made. Article screening and inclusion in the subsequent review were collectively performed by authors QR and CR and, in the event of disagreement, a third author was consulted for opinion. Data charting and summary were then completed in unison on calibrated templates determined prior to article selection. The data points collected included article publication year, country of study completion, patient number, study population age, gender ratio, Kellgren-Lawrence (KL) grade, injection guidance modality, PRP concentration injected, PRP volume injected, injection frequency, centrifuge setting, follow-up durations, complications, and study conclusions. Study conclusions are specifically categorized as superior when PRP outperformed therapeutic alternatives such as corticosteroids, hyaluronic acid (HA), or radiofrequency ablation; non-inferior when PRP was demonstrated to be equivalent in efficacy to alternative therapies; or inferior when PRP was demonstrated to be inferior to alternative therapeutic alternatives or equivalent to research placebos, most commonly saline. Critical appraisal and detailed statistical analyses were not performed in keeping with the delineated checklist published in PRISMA Extension for Scoping Reviews (PRISMA-ScR).[13] No funding was acquired for the study.

2.1. Search Strategy and Selection Criteria

A comprehensive search of PubMed was performed on 15 November 2025 adopting the following strategy: (“platelet rich plasma” OR “autologous conditioned plasma”) AND (“knee”) AND (“osteoarthritis” OR “degeneration” OR “gonarthrosis”). An identical strategy was carried out via query on Embase and the results are displayed in Figure 1.

2.2. Inclusion Criteria

The following inclusion criteria were selected: (1) studies with titles, articles, and abstracts written in the English language; (2) peer-reviewed randomized controlled trials (RCTs) comparing PRP against non-regenerative medicine alternatives; (3) studies including patients ≥ 18 years of age; and (4) studies exclusively investigating PRP effects on knee osteoarthritis.

2.3. Exclusion Criteria

The following exclusion criteria were adopted: (1) articles not written in English; (2) non-human studies; (3) individuals ≤ 18 years of age; (4) studies with multiple regenerative medicine modalities, in addition to PRP, in its treatment arms; (5) review articles, systematic analyses, or meta-analyses; (6) studies investigating pathologies other than knee osteoarthritis.

3. Results

A total of 21 independent articles were identified and incorporated into our review using the databases previously defined (Figure 1).[14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34] The articles selected were published between 2012 and 2025, with studies performed across North America, South America, Europe, and Asia. The total number of patients in each study ranged from 21 to 288, with follow-up durations ranging from 12 weeks to 60 months. Hyaluronic acid was used as the alternative injectate in close to half of the studies (n=10/21),[17,22,23,24,25,26,27,30,32,33] with the remaining being corticosteroid [15,21,28,33], radiofrequency ablation,[14,20] and saline placebo.[16,29,31,34] Captured patient population knee osteoarthritis severity ranged widely from Kellgren-Lawrence grades 0 to IV, and all articles reported at least one measurable clinical end point indicating degree of pain control or physical functionality. Twelve studies described the injection modalities adopted for their respective PRP therapies.[14,16,17,19,20,21,27,28,29,30,33,34] Ultrasound [16,19,20,21,30] and the landmark technique [17,27,28,29,34] were utilized in 5 studies each while fluoroscopy [14,33] was engaged in the remaining 2 studies. Article demographics and injection methodologies are summarized and presented in Table 1 and Table 2.

3.1. PRP Centrifugation Regimen

Eighteen studies defined centrifugation settings to attain the desired concentrations of PRP injectate (Table 2). Nine studies [14,16,21,24,28,29,31,33,34] performed a single spin, and the remaining 9 studies [18,19,20,22,23,25,26,30,32] performed a double spin. The lowest rates of the first and second spins were both 1500 revolutions per minute (rpm), while the longest were 3500 and 4000 rpm, respectively.[19,20,29,31] The shortest spin duration was a 5-minute single spin demonstrated in studies performed by Smith et al. [31] and Tschopp et al. [33], and the longest was performed by Dorio et al.[19] for a total of 22 minutes, split into a first spin of 12 minutes and a second spin of 10 minutes. [19]Smith et al.’s short-spin duration protocol achieved improvement in WOMAC score for the PRP group from 2 weeks following injection until study completion at 12 months (p < 0.001), which was statistically significantly superior to the saline group.[31] The long-spin duration protocol adopted by Dorio et al. yielded no difference in overall VAS pain scores between groups which received PRP, plasma, and saline (p = 0.499).[19] This was no different for all secondary measures involving WOMAC, QoL, PGA, ADL and TUGT scores.

3.2. PRP Injection Concentration

A total of 14 studies analyzed published, defined platelet concentrations in their injectates (Table 2). Concentration of platelets utilized for intra-articular injection ranged from a value close to patient’s baseline at 1.15 times of platelet enrichment to 9.85 times of platelet enrichment, demonstrated via the same study.[27] Out of the remaining 26 patients, platelet enrichment was achieved through at least 3.31 times of baseline concentration, and the study revealed equivalent efficacy between injectates containing PRP and HA after the third infiltration (p = 1.0), at 3 months (p = 0.227), and at 6 months (p = 1). Fossati et al. prepared platelets at a much lower concentration at approximately 1.8 times of baseline values for 116 patients, which is at the low end of the platelet concentration spectrum across all analyzed studies.[24] The study similarly demonstrated little difference between treatment groups PRP, HA, and PRP with HA measured via WOMAC (p = 0.45, 0.64), IKDC (p = 0.12, 1), KOOS (p = 0.15, 0.44), and VAS (p = 0.40, 0.65) at 6 and 12 months, respectively. Other included studies described platelet enrichment at magnitudes ranging from 2 to 5 times of patients’ baselines (Table 2).

3.3. PRP Injection Volume

Fifteen studies defined the volume of PRP injected during each treatment with volumes ranging from 1.4 mL to 8 mL (Table 2). Dorio et al. prepared PRP injectate concentrations at 1 x 10⁶ platelets/mm³, attaining PRP volumes ranging from 1.4 mL to 5 mL for each injection. The team compared efficacy of PRP against plasma and saline injections and were unable to elicit statistically significant advantage of PRP over the 2 alternatives, as measured using VAS and WOMAC.[19] The PRP group did demonstrate higher incidence of adverse events (p = 0.025) described as pain of mild or moderate intensity with a mean duration of 2.1 days. Elksniņš-Finogejevs et al. prepared higher volumes of PRP injectate at 8 mL per treatment, evaluating its efficacy against 40 mg triamcinolone acetonide and 5 mL of 2% lidocaine.[21] VAS score at 1 year demonstrated a significantly greater reduction in pain in the PRP group compared to the corticosteroid group (-3.1 ± 2.0 vs − 0.8 ± 1.8, p = 0.0002). Functional improvement of the joints was similar between the 2 groups (p > 0.05). It was noted that 15 patients in the PRP group developed a mild synovitis, which self-resolved without adverse events. Patel et al. similarly performed intra-articular injections with higher PRP volume of 8 mL.[29] The study evaluated efficacy of a single PRP injection versus 2 sequential injections and a third group of placebo saline injection. At every timepoint of follow-up, defined as 6 weeks, 3 months, and 6 months, pain, stiffness, and physical function were all superior in the first 2 groups using PRP injections compared to placebo (p < 0.001 for all categories). Out of the remaining articles, 6 and 4 papers described PRP injection volumes of 5 mL and 3 mL respectively, making them the most frequently injected volumes in our study selection (Table 2).

3.4. PRP Injection Frequency

The most popular therapeutic regimen was 3 PRP injections, with one performed each week (Table 2). Nine articles described this defined regimen, with 3 additional articles reporting 3 injections within the duration of one month and 3 injections spaced 2 weeks and 15 days apart, respectively.[24,25,27] In total, 12 articles described 3 consecutive doses of PRP as part of their therapeutic regimen. Eleven articles demonstrated superiority to saline injection and at least equivalent, if not improved, efficacy in pain control and function compared to alternative therapies, defined as either corticosteroids or HA. Only Bennell et al. failed to demonstrate superiority of PRP injections compared to saline, revealing no difference in pain or measured tibial cartilage volume as part of the primary outcome.[16]

The next most frequent injection regimen was a one-off single PRP injection to the joint described in 7 of the 21 articles. Out of the studies adopting low frequency single PRP injections, 2 failed to demonstrate PRP superiority compared to alternative therapy. Elawamy et al. showed that at 6 (p = 0.01) and 12 months (p = 0.04) that VAS and ISK were both superior in the pulsed radiofrequency ablation group.[20] Similarly Tschopp et al. did not find significant determination of pain and function being attributed to the content of the injectate, whether it is corticosteroid, HA, or PRP.[33]

Patel et al. included 2 groups, with the first being a single injection and the second being 2 injections of PRP spaced 3 weeks apart, incorporating both low frequency and intermediate frequency injection groups.[29] The team subsequently revealed no significant difference in pain score (p = 0.410) during follow-up at 6 months. Cerza et al. adopted the highest frequency of PRP injections, performing weekly injections for a total of 4 times.[17] At 4, 12, and 24 weeks, WOMAC scores were significantly lower in the PRP group compared to the HA group (p< 0.001 for each period).

3.5. Adverse Events

Significant major adverse events were not reported in any of the studies, and the most comprehensive capture of complications were carried out by Patel et al. where the group recorded in detail the incidence of syncope, dizziness, nausea, gastritis, sweating, tachycardia, pain, and stiffness.[29] Many of these were transient, lasting merely 30 minutes, and none led to any clinically-relevant consequences. The study further delineated correlation between total platelet injected and the incidence of event occurrence, noting a positive relationship (p = 0.02). Multiple other studies similarly demonstrated more minor events occurring in groups with PRP intervention, which self-resolved without long-term complications.[16,18,19,21,23,27,30,32]

4. Study Results

Seventeen of the 21 studies demonstrated at least equivalence in efficacy comparing PRP against alternative non-placebo therapy, defined as saline or a sham procedure with no intended therapeutic relevance aside from the placebo effect. Six of the 17 clearly demonstrated a significant therapeutic benefit and are therefore classified as superior (Table 2). In the superior group, Anwar et al. ran the study with the largest group of participants (n = 200), comparing a single PRP injection to genicular nerve ablation under fluoroscopic guidance.[14] Both groups showed a steady decrease in VAS from 2 weeks to 6 months following therapies. However, at 12 and 24 months, the recurrence of pain was significantly quicker in the RFA group, attaining VAS of 4.73 ± 2.63 and 6.06 ± 2.01 compared to PRP’s 2.99 ± 1.78 and 4.05 ± 1.82, respectively, over the same period, both with p values < 0.05.[14]

Eleven studies were considered non-inferior when pitted against alternative therapies of corticosteroids, HA, radiofrequency ablation, and local anesthetic. Chu et al. analyzed a total of 630 participants with 308 in the PRP group and 322 in the sham saline group.[18] The respective injectates were applied for a total of 3 times spaced 1 week apart. Although WOMAC scores were improved in both groups at 3 months, only the PRP group demonstrated sustained benefits at 48 months (p < 0.001). This result was similar when the authors looked at IKDC scores where the improvement remained significantly robust only in the PRP group up to 60 months (p < 0.001). The study also picked up on objective measurables such as TNF-alpha and IL-1B levels in synovial fluid and tibiofemoral cartilage volumes, with both demonstrating superiority in the PRP group (both p < 0.001).[18]

Of the remaining 4 studies considered therapeutically inferior, 3 demonstrated equivalent efficacy to saline while 1 was seen to be inferior to radiofrequency ablation (Table 2). The RESTORE trial randomized a total of 288 subjects to receive either PRP or saline injections for a total of 3 times spaced 1 week apart. At 12 months, despite pain score improvement favoring the PRP group, the difference failed to achieve statistical significance (-2.1, SD 2.7 PRP vs -1.8, SD 2.5 saline).[16] There was also no demonstrable difference between tibial cartilage volume change between the groups. The authors further delineated 12 secondary end points, including but not limited to knee effusion measures, areas of cartilage thinning, meniscus morphology, cartilage defect, and perceived global change. Only perceived global change at 2 (p = 0.02) and 12 months (p = 0.05) statistically favored the PRP group.

5. Discussion

Our scoping review showed the extensive variability present in the preparation and conduction of PRP injections in the management of knee osteoarthritis, ranging from centrifuge settings, PRP injectate volume and concentration, therapeutic frequency of injection, and the image-guidance modality, when adopted, during injection. This trend is perceived despite our extensive efforts in attempting to standardize the articles selected, excluding papers that compared multiple regenerative medicine treatment modalities to acquire discernable clinical outcomes when PRP was pitted against current standards of treatment or placebo. This review is therefore timely in delineating basic practice standards for PRP injection of the knee in attaining therapeutic efficacy while avoiding clinically significant complications.

Therapeutic efficacy of intra-articular knee injections while adopting various image-guidance techniques were explored in multiple studies. Lundstrom et al. explored ultrasound-guided intra-articular knee injection of HA compared to the traditional landmark technique (mostly suprapatellar bursa approach), measuring efficacy via subsequent arthroplasty rates and repeat HA and corticosteroid injections.[35] The authors revealed that knee arthroplasty rates were significantly reduced in the US-guided cohort compared to the pure landmark technique (33.2% vs 45.8%; p< 0.001). Furthermore, significantly fewer patients in the US-guided cohort received subsequent corticosteroid injection compared to the landmark-guided cohort (27.4% vs 34.0%; p< 0.002), suggesting greater accuracy in needle placement when US is adopted for injections.[35] This was corroborated in other studies where image-guidance, defined as US, fluoroscopy, or arthrography, substantially improved injection accuracy (98% accuracy) compared to landmark approaches using superolateral patellar and superomedial patellar techniques (87% and 82%, respectively).[36,37,38] Given repeated demonstration of superiority in efficacy of image-guidance in intra-articular knee injections, it would be prudent to adopt a suitable mode of image-guidance in PRP injections of the knee.

Given that conceptually, the clinical benefits of PRP are derived from growth factors, cytokines, and bioactive platelet molecules stored in platelet alpha granules, one would extrapolate that the concentration of platelets in the injectate would be determinant of therapeutic efficacy.[39] In vitro, it had been postulated that PRP concentration is directly associated with proliferation of certain mesenchymal cell lines, including mesenchymal stem cells, chondrocytes, osteoblasts, and fibroblasts.[40,41,42,43,44,45] Given that the normal human range of platelet concentration is between 150,000 to 450,000 platelets/µL, it was previously suggested that a 4-fold increase of platelets from this baseline could yield more predictable benefits.[46] Bensa et al. performed a sub analysis on platelet concentrations in their meta-analysis, defining low PRP concentration as < 800,000 platelets/µL and high concentration as 1 x 10⁶ ± 2 x 10⁵ platelets/µL.[47] The authors found that when compared against placebo in osteoarthritic knee injections, the high platelet concentration articles demonstrated statistical superiority in both VAS and WOMAC scores at 6 and 12 months. Bansal et al. seemingly corroborated this suggestion of high platelet count in their therapeutic injectate, their protocol attaining a mean platelet concentration of 14.38 ± 1.76 x 10⁵ platelets/µL. The authors further recommended an injection volume of 8 mL, suggesting that this regimen was crucial in achieving sustained chondroprotective effects up to 12 months out from initial treatment. Despite the definitions of concentration cut-offs being arbitrary, there may be value in attempting to achieve a higher platelet count in preparation of PRP injectates.

The frequencies of intra-articular PRP injections adopted in our selection of studies have been highly variable. Tao et al. determined in their meta-analysis that a triple-dose regimen of PRP injections is superior to a single dose in the treatment of osteoarthritic knee pain.[48] This was seen in a significantly lower VAS at 12 months (p < 0.0001) in studies with triple-dose regiments, and the frequency of adverse events registered were not significantly higher in the triple-dose regimen studies (p = 0.23). On the contrary, comparing double and single-dose regimens, neither the VAS (p = 0.50) nor the rate of adverse events (p = 0.28) were different between the single and double-dose regimens. Functionality was also shown to be improved measured using WOMAC and IKDC scores when adopting a multiple-injection regimen (p < 0.01).[49] Interestingly, the same study did not demonstrate an improvement in VAS between the single and multiple dosing groups (p = 0.48). Clinical practice guideline published by American Academy of Orthopedic Surgeons hinted at three intra-articular PRP injection being the standard that appeared to achieve more favorable clinical results.[50]

The safety profile of PRP knee injections has been relatively consistent in our review, with the most frequent complaint being a transient self-limiting localized inflammatory reaction (Table 2). A major clinical concern of intra-articular interventions is the deterioration of cartilage volume with successive therapy over the long run, while PRP in relation to current mainstream treatments of corticosteroids and HA appeared to be protective with respect to morphological cartilage parameters measured via quantitative MRI.[51] Overall, safety attributes of PRP given its autologous nature would continue to be one if its most reassuring features.

6. Limitations

Our study is constrained by its very nature of being a scoping review, and therefore no in-depth statistical analysis was performed. In addition, the immense variability in the manners in which the RCTs were performed made it exceptionally challenging to derive measured recommendations that could be standardized across the board. The individual RCT conclusions drawn, describing clinical outcomes of PRP against placebo, could not be interpreted meaningfully against that of a current standard of care, such as a corticosteroid injectate. These further compounds the difficulty making recommendations that could be implemented across guidelines.

7. Conclusions

PRP as a therapeutic modality in the treatment of knee osteoarthritis has proven itself to be safe and largely efficacious when compared against current standards of treatment involving corticosteroids and HA. While much more work in the future is needed to statistically ascertain the parameters for PRP delivery, a strategy of platelet concentration to 4 times of baseline followed by a total injection frequency of 3 sessions performed under reliable image-guidance would be a prudent one to adopt in the treatment of knee osteoarthritis.