Abstract
Purpose
Osteoarthritis (OA) is a widespread joint disease characterized by the gradual loss of cartilage. Intra-articular injections, including platelet-rich plasma (PRP), are commonly used for treatment, but the optimal PRP preparation method remains debated. This study aims to perform a network meta-analysis of randomized controlled trials to compare the efficacy of different PRP preparation methods and determine the most effective protocols.
Methods
The literature search was conducted based on PRISMA guidelines. Randomized controlled trials (RCTs) evaluating intra-articular injectables in osteoarthritic knees were included. Data were extracted, and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores were analyzed at 3, 6, and 12–18 months. Clinical outcomes were compared using a frequentist network meta-analysis, and treatment options were ranked using the P-Score. Statistical analysis was performed using R 4.3.2.
Results
Twenty-three RCTs with 1752 patients were included. Treatments included PRP, plasma rich in growth factor (PRGF), leukocyte-poor PRP (LP-PRP), leukocyte-rich PRP (LR-PRP), hyaluronic acid (HA), and saline placebo. Leukocyte-rich PRP with low platelet concentration increase, using both anticoagulant and activator showed the best effects on WOMAC pain and stiffness scores within 6 months (WMD = 26.02; 95% CrI, 0.92–52.46). Leukocyte-poor PRP with high platelet concentration increase, using anticoagulant without activator was most effective for WOMAC function and stiffness at 12–18 months (WMD = 18.94; 95% CrI, 8.34–28.12). Long-term results indicated Leukocyte-poor PRP with low platelet concentration increase, using anticoagulant without activator yielded the best outcomes for cartilage repair and function (WMD = 17.09; 95% CrI, -8.4 to 42.78).
Conclusions
Optimizing OA treatment involves tailoring PRP protocols to disease stage, with low platelet, high leukocyte PRP (RPRP_LPC_Y_Y) recommended for early OA due to its anti-inflammatory effects and high platelet, low leukocyte PRP (PPRP-HPC) preferred for advanced OA to promote tissue repair and regeneration.
Supplementary Information
The online version contains supplementary material available at 10.1186/s13018-025-06026-1.
Keywords: Platelet rich plasma, Hyaluronic acid, Saline, Cartilage, Osteoarthritis, Knee, Meta-analysis, Systematic review
Introduction
Osteoarthritis (OA) affects over 520 million people worldwide, often progressing to total knee arthroplasty (TKA) in advanced stages [38, 60]. Intra-articular injections remain a key conservative treatment, providing pain relief and functional improvement by delivering therapeutic agents directly into the joint [6, 37, 42, 56].
Platelet-rich plasma (PRP) is widely used in early-stage OA due to its anti-inflammatory effects and ability to promote tissue regeneration [4, 29, 53]. Leukocyte-rich PRP (LR-PRP) can enhance acute-phase inflammation, aiding tissue repair, while its platelet concentration influences growth factor release, affecting chondrocyte activity and angiogenesis [8, 13, 35, 36, 52, 54, 58, 59]. However, excessive platelet levels may impair tissue repair27. Activators promote rapid growth factor release [49]and when used with anticoagulants, they may enhance platelet activation11. These variations in PRP composition significantly impact clinical outcomes.
Different PRP preparation methods result in compositional and therapeutic discrepancies. Magalon et al. compared five protocols and found significant differences in platelet and growth factor levels, highlighting the need for standardization [36]. Despite this, no meta-analysis has systematically compared PRP preparation methods in knee OA. The PAW classification system, which considers platelet concentration, activation status, and leukocyte content, provides a framework for standardization [15]. Based on this system, we analyzed four PRP preparation methods to evaluate their short- and long-term effects and determine optimal protocols for different OA stages.
Methods
Study selection
Two independent reviewers conducted the literature search following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [27]. Discrepancies were resolved through reconciliation by a third author. The initial screening involved evaluating titles and abstracts, with potentially relevant studies undergoing a full-text review. Additionally, the references of all included studies were manually screened to identify any additional articles meeting the inclusion criteria.
Search strategy
The search strategy was conducted across PubMed, EMBASE, and the Cochrane Library databases on October 16, 2023. Keywords and controlled vocabulary related to osteoarthritis (e.g., “osteoarthr*” or “degenerative arthritis”) and platelet-rich plasma (PRP) (e.g., “PRP,” “buffy layer,” or “platelet gel”) were used, combined with filters for human studies and clinical trial designs such as randomized controlled trials. Boolean operators and field-specific queries refined the search, ensuring comprehensive retrieval of relevant literature. The search strategy is presented in Appendix 1. The combined strategy yielded 519 records from PubMed, 764 from the Cochrane Library, and 1,108 from EMBASE. No restrictions were placed on publication dates.
Eligibility criteria
The inclusion criteria were: (1) randomized controlled trial comparing intra-articular injections in the knee, (2) published in a peer-reviewed journal, (3) published in English, (4) includes outcome scores such as WOMAC, KOOS, ROM, VAS, KC, MAA, IKDC, and Tegner, and (5) full text of studies available.The exclusion criteria were the following: (1) studies without patient-reported outcome measures (e.g., WOMAC, VAS, IKDC); (2) studies with incomplete or non-convertible outcome data; (3) duplicate publications based on the same cohort; (4) conference abstracts, reviews, or editorials; (5) basic science or preclinical research (e.g., animal studies, in vitro studies).
Data extraction
Data extraction was carried out by two independent reviewers using a standardized data sheet. Information collected included study characteristics (study title, country, diagnostic criteria, treatment arms, patient follow-up duration, Kellgren-Lawrence grade, symptom duration, patient age, percentage of female participants, and BMI). Details of PRP application were also gathered, such as injection method, frequency, volume, and site. For PRP preparation methods, the type of anticoagulant used, whether an activator was added, and the specifics of the preparation system (number of spins, speed, and equipment) were noted. Additionally, the leukocyte content and mean-fold change in platelet concentration were recorded. The analysis encompassed a summary of study characteristics and findings, highlighting variability in PRP preparation and protocols. Meta-analyses were performed where data were homogeneous, using a random-effects model, with heterogeneity assessed via the I² statistic. A network meta-analysis (NMA) allowed for both direct and indirect comparisons of different PRP preparation methods. Sensitivity analyses were conducted to ensure robustness, and publication bias was assessed with funnel plots and Egger’s test.
Data analysis
Based on the concentration of leukocytes in PRP (categorized into leukocyte-poor PRP and leukocyte-rich PRP) and the degree of platelet concentration increase (low increase < 3-fold, moderate increase 3-fold to < 5-fold, high increase ≥ 5-fold), and the usage of activators and anticoagulants(Used: Y, Not Used: N, Unknown: X), the preparation methods were classified into 14 categories in Table 1.
Table 1.
PRP Preparation classification by leukocyte, platelet concentration, and additives
| Abbreviation | Full Name | |
|---|---|---|
| 1 | RPRP_HPC_Y_Y | Leukocyte-rich PRP with high platelet concentration increase, using both anticoagulant and activator |
| 2 | RPRP_HPC_Y_N | Leukocyte-rich PRP with high platelet concentration increase, using anticoagulant without activator |
| 3 | RPRP_HPC_X_N | Leukocyte-rich PRP with high platelet concentration increase, using anticoagulant with unknown activator status |
| 4 | RPRP_MPC_Y_Y | Leukocyte-rich PRP with moderate platelet concentration increase, using both anticoagulant and activator |
| 5 | RPRP_MPC_Y_N | Leukocyte-rich PRP with moderate platelet concentration increase, using anticoagulant without activator |
| 6 | RPRP_LPC_Y_Y | Leukocyte-rich PRP with low platelet concentration increase, using both anticoagulant and activator |
| 7 | PPRP_LPC_Y_N | Leukocyte-poor PRP with low platelet concentration increase, using anticoagulant without activator |
| 8 | PPRP_HPC_Y_N | Leukocyte-poor PRP with high platelet concentration increase, using anticoagulant without activator |
| 9 | PPRP_MPC_Y_Y | Leukocyte-poor PRP with moderate platelet concentration increase, using both anticoagulant and activator |
| 10 | PPRP_MPC_Y_N | Leukocyte-poor PRP with moderate platelet concentration increase, using anticoagulant without activator |
| 11 | PPRP_MPC_X_Y | Leukocyte-poor PRP with moderate platelet concentration increase, using anticoagulant and activator |
| 12 | PPRP_LPC_Y_Y | Leukocyte-poor PRP with low platelet concentration increase, using both anticoagulant and activator |
| 13 | PPRP_LPC_N_N | Leukocyte-poor PRP with low platelet concentration increase, without anticoagulant and activator |
| 14 | PPRP_LPC_X_N | Leukocyte-poor PRP with low platelet concentration increase, without anticoagulant and with unknown activator status |
The assessment of clinical outcomes included WOMAC pain sub score, WOMAC function sub scores WOMAC stiffness sub scores, WOMAC total scores. The results were evaluated at 6 and 12–18 months or the closest reported follow-up times within 3 months. WOMAC pain sub scores, WOMAC function sub scores WOMAC stiffness sub scores, WOMAC total scores were normalized on a scale from 0 to 100.
Statistical analysis
All statistical analyses were performed using R(version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria). A frequentist approach to network meta-analysis with a random effects model was performed using the netmeta package version 0.9–6 in R [48]. For continuous outcomes, the relative effect sizes were reported as standardized mean differences (MD), and for dichotomous outcomes, the relative effect sizes were.
reported as odds ratios (OR). The effect sizes were reported with 95% confidence intervals (95% CI).Heterogeneity was quantitatively assessed using the I² statistic, supplemented by Cochran’s Q test to determine the presence of statistically significant differences. An I² > 50% was interpreted as moderate or higher heterogeneity, while an I² > 75% indicated substantial heterogeneity [25]. To explore the potential impact of study-level covariates and assess the stability of our findings, meta-regression analyses were conducted using key variables including mean age, mean body mass index (BMI), proportion of female participants, and total sample size. Given the heterogeneity in follow-up durations across included trials, outcome measures were stratified by three predefined timeframes: short-term (within 3 months), mid-term (6 months), and long-term (12 to 18 months), to reduce the impact of temporal variation on effect estimates. Publication bias was examined through visual inspection of funnel plots and tested quantitatively using Egger’s regression test. If asymmetry was identified or Egger’s test indicated significance (P < 0.05), we further applied trim-and-fill or non-parametric methods to evaluate the stability of the results. In addition, key methodological features of the included studies, such as comparator types (e.g., hyaluronic acid or saline), follow-up duration, and baseline demographics (age, sex, BMI), were carefully documented and considered in subgroup and sensitivity analyses to account for potential sources of bias. The frequentist analogue to the surface under the cumulative ranking (SUCRA) probabilities called the P-score was used to rank the treatments. This method allows each treatment to be ranked on a scale from 0 to 1, where 0 indicates the least effective treatment and 1 indicates the most effective [48].
Results
Search result
The initial search resulted in 2391 papers. Of them, 855 were duplicates and, thus, excluded. Only 59 were RCTs. 37 of these were excluded because they did not match the topic or did not report quantitative data according to our outcomes of interest. T. Ultimately, 22 articles that met the eligibility criteria were included (Fig. 1) [6, 7, 12, 18–20, 23, 26, 30, 33, 34, 40, 42–46, 50, 51, 55, 61].A total of 1,752 patients, including 1,239 females, were enrolled in the study. The mean age of the patients included in the studies ranged from 50 to 75 years. The severity of the disease was primarily concentrated at Kellgren-Lawrence grades of 2 to 3. Details of the study characteristics are presented in Table 2, and network diagrams of available comparisons and studies are illustrated in Appendix 2.
Fig. 1.
Flow diagram of the study selection procedure
Table 2.
Characteristics of studies included in the analysis
| Study | Level of evidence | K-L Grading | Arms | Patients Follow-up | Age (year) | Female (%) | BMI, kg/m2 | PRP Operation Methods | PRP Preparation Methods | Outcome | measurement timepoint (month) | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Injection Volume (ml) | Anticoagulant | Activator Added | Preparation System (Times) | White Blood Cells | Platelet Mean-fold Change | ||||||||||
| Su et al.,2018 | I | 2–3 | PRP | 25 | 54.16 ± 6.56 | 56.00% | 28.17 ± 1.43 | 6 | Y | Y | 2 | rich | 5.61x | VAS, WOMAC | 1,3,6,12,18 |
| HA | 30 | 53.13 ± 6.41 | 60.00% | 28.69 ± 1.13 | |||||||||||
| Tucker et al.,2021 | I | 2–3 | PRP | 11 | 57.5 ± 1.8 | 66.67% | 30.9 ± 1.5 | 5 | Y | Y | NR | rich | 2.59x ± 0.68 | VAS, WOMAC | 3,6,12 |
| saline | 6 | 57.2 ± 3.9 | 27.27% | 29.1 ± 2.1 | |||||||||||
| Yaradilmis et al.,2020 | I | 2–3 | three doses of LP-PRP | 30 | 58.93 ± 6.25 | 90.00% | 32.53 ± 6.25 | 3 | Y | N | 2 | poor | 4.6x | VAS, WOMAC | 2,6,12 |
| three doses of LR-PRP | 30 | 60.3 ± 7.65 | 86.70% | 31.27 ± 4.08 | 3 | Y | N | 2 | rich | 1.9x | |||||
| HA | 30 | 63 ± 9.17 | 86.7% | 32.4 ± 4.2 | |||||||||||
| Yurtbay et al.,2022 | I | 1–3 | one dose of PRP | 62 | 53.29 ± 12.97 | 33.87% | 31.09 ± 5.52 | 5 | Y | Y | 1 | rich | > 5x | KOOS, VAS | 1,3,6,12,24 |
| three doses of LR-PRP | 63 | 57.38 ± 8.78 | 14.29% | 30.68 ± 4.63 | 5 | Y | Y | 1 | rich | > 5x | |||||
| one dose of NS | 59 | 56.29 ± 10.53 | 18.64% | 30.67 ± 4.51 | |||||||||||
| three doses of NS | 53 | 53.47 ± 11.31 | 33.96% | 29.22 ± 4.79 | |||||||||||
| Lin et al.,2019 | I | NR | PRP | 31 | 61.17 ± 13.08 | 70.97% | 23.98 ± 2.62 | 2 | N | N | 1 | poor | 1.81x ± 0.34 | WOMAC, IKDC | 1,2,6,12 |
| HA | 29 | 62.53 ± 9.9 | 65.52% | 26.26 ± 2.99 | |||||||||||
| saline | 27 | 62.23 ± 11.71 | 62.96% | 24.98 ± 3.12 | |||||||||||
| Spaková et al.,2012 | I | 1–3 | PRP | 60 | 52.80 ± 12.43 | 45.00% | 27.9 ± 4.1 | 3 | Y | N | 3 | rich | 4.5x | WOMAC | 3,6 |
| HA | 60 | 53.20 ± 14.53 | 48.33% | 28.3 ± 4.0 | |||||||||||
| Himanshu et al.,2021 | I | 1–3 | PRP | 64 | 64.4 | 39.06% | 24.90 | 8 | Y | N | 2 | poor | 6.25x | WOMAC, IKDC | 1,3,6,12 |
| HA | 68 | 65.8 | 38.24% | 25.29 | |||||||||||
| Buendía-López et al.,2018 | I | 1–2 | PRP | 33 | 56.15 ± 3.001 | 51.52% | 24.9 ± 0.32 | 5 | NR | Y | 2 | poor | 3.87x | WOMAC, VAS | 6,12 |
| HA | 32 | 56.63 ± 2.9 | 53.13% | 24.9 ± 0.41 | |||||||||||
| Dório et al.,2021 | I | 2–3 | PRP | 20 | 66.4 ± 5.6 | 95.00% | 28.3 ± 4.1 | 1.4-5 | Y | N | 2 | poor | 3x | WOMAC, KOOS | 3,6 |
| saline | 21 | 62.5 ± 8.1 | 90.48% | 27.6 ± 3.8 | |||||||||||
| Duymus et al.,2017 | I | 2–3 | PRP | 33 | 60.4 ± 5.1 | 96.97% | 27.6 ± 4.6 | 5 | NR | N | 1 | rich | 9-13x | WOMAC, VAS | 1,3,6,12 |
| HA | 34 | 60.3 ± 9.1 | 97.06% | 28.4 ± 3.6 | |||||||||||
| Filardo et al.,2015 | I | NR | PRP | 94 | 53.32 ± 13.2 | 36.17% | 26.6 ± 4.0 | 5 | NR | Y | 2 | rich | 4.6x ± 1.4 | IKDC, KOOS | 2,6,12 |
| HA | 89 | 57.55 ± 11.8 | 41.57% | 26.9 ± 4.4 | |||||||||||
| Huang et al.,2019 | I | NR | PRP | 40 | 54.5 ± 1.2 | 79.20% | 25.23 ± 4.15 | 4 | NT | N | 1 | poor | 2x | WOMAC, VAS | 3,6,12 |
| HA | 40 | 54.8 ± 1.1 | 84.20% | 24.51 ± 3.09 | |||||||||||
| Nunes-Tamashiro et al.,2022 | I | 2–3 | PRP | 34 | 67.6 ± 7.4 | 90.90% | 29.22 ± 3.2 | NR | Y | N | 1 | poor | 2.5-5x | WOMAC, VAS | 1,2,6,12 |
| saline | 33 | 68 ± 6.2 | 88.20% | 30.23 ± 4.1 | |||||||||||
| Lana et al.,2016 | I | 1–3 | PRP | 36 | 60.9 ± 7 | 80.60% | 27.42 ± 6.89 | 5 | Y | Y | 2 | rich | 6.5x | WOMAC, VAS | 1,3,6,12 |
| HA | 36 | 60 ± 6.6 | 91.70% | 28.24 ± 8.77 | |||||||||||
| Louis et al.,2018 | I | NR | PRP | 24 | 53.2 ± 11.7 | 41.67% | 25.6 ± 2.9 | 3 | Y | Y | 2 | rich | 3.3x ± 0.7 | WOMAC | 1,3,6 |
| HA | 24 | 48.5 ± 11.5 | 54.17% | 27.0 ± 2.9 | |||||||||||
| Park et al.,2021 | I | 1–3 | PRP | 55 | 60.6 ± 8.2 | 70.91% | 25.5 ± 2.2 | 3 | Y | Y | 1 | rich | 3x | IKDC, VAS, WOMAC | 3,6 |
| HA | 55 | 62.3 ± 9.6 | 85.45% | 25.9 ± 2.8 | |||||||||||
| Patel et al.,2013 | I | NR | PRP | 27 | 53.11 ± 11.55 | 59.26% | 26.28 ± 3.23 | 8 | Y | Y | 1 | poor | 3x | WOMAC, VAS | 3,6 |
| saline | 23 | 53.65 ± 8.17 | 73.91% | 26.21 ± 2.93 | |||||||||||
| Hüseyin et al.,2019 | I | 1–3 | PRP | 30 | 61.30 ± 7.91 | 96.67% | 30.37 ± 4.47 | 4 | Y | Y | 2 | rich | 4-6x | WOMAC, VAS | 1,6 |
| saline | 27 | 60.19 ± 6.80 | 88.89% | 30.70 ± 3.97 | |||||||||||
| Raeissadat et al.,2017 | I | 2–3 | PRGF | 36 | 57.0 ± 7.18 | 80.56% | 28.6 ± 2.82 | 5 | Y | Y | 3 | poor | 0 | WOMAC, VAS | 2,6 |
| HA | 33 | 59.5 ± 7.54 | 81.82% | 27.5 ± 2.9 | |||||||||||
| Raeissadat et al.,2020 | I | 2–3 | PRGF | 50 | 57.08 ± 7.3 | 72.00% | 27.92 ± 2.7 | 5 | Y | Y | 3 | poor | 0 | WOMAC, VAS | 2,6,12 |
| HA | 52 | 58.63 ± 7.09 | 71.15% | 28.65 ± 3.02 | |||||||||||
| Raeissadat et al.,2021 | I | 2–3 | PRP | 52 | 56.09 ± 6.0 | 75.00% | 27.41 ± 2.6 | NR | Y | N | 2 | rich | 4-6x | WOMAC, VAS | 2,6,12 |
| PRGF | 51 | 56.07 ± 6.3 | 72.55% | 27.50 ± 2.1 | NR | Y | Y | 3 | rich | 5x | |||||
| HA | 49 | 57.91 ± 6.7 | 75.51% | 27.46 ± 2.2 | |||||||||||
| Di Martino et al.,2022 | I | 1–3 | LR-PRP | 90 | 55.2 ± 9.8 | 31.11% | 26.1 ± 4.5 | 5 | Y | Y | 2 | rich | 4.6x | IKDC, KOOS | 2,6,12 |
| LP-PRP | 85 | 55.7 ± 10.7 | 41.18% | 27.4 ± 4.1 | 5 | Y | Y | 2 | poor | 4.4x | |||||
According to the Cochrane risk-of-bias tool for assessing the overall risk of bias, 17 studies were rated as low risk, 6 studies 18,20,26,30,50,51 as moderate risk. Among them, 6 studies 18,20,26,30,50,51 did not specify the randomized scheme in the article. The results of the Cochrane risk-of-bias assessment are presented in Appendix 3.
WOMAC pain
In the short-term, RPRP_LPC_Y_Y demonstrated the greatest efficacy compared to HA and Saline (WMD = 26.42; 95% CrI, 4.08 to 47.7), with the highest probability of being the best treatment (SUCRA = 0.98). The model showed minimal heterogeneity (global I² = 0%) and good fit (posterior mean residual deviance: 25.41/27).
For medium-term outcomes, RPRP_LPC_Y_Y remained the top-ranked treatment (SUCRA = 0.89), but its effect was less pronounced (WMD = 20.74; 95% CrI, -10.31 to 48.58). Long-term analysis was conducted in two sub-networks due to connectivity limitations: PPRP_HPC_Y_N had the most significant impact in the HA network (WMD = -1.71; 95% CrI, -4.41 to 0.98, SUCRA = 0.70), while RPRP_LPC_Y_Y was most effective in the Saline network (WMD = 5.61; 95% CrI, -16.93 to 27.88, SUCRA = 0.65).
WOMAC function
Short-term analysis showed that RPRP_LPC_Y_Y was the most effective treatment (WMD = 18.98; 95% CrI, -4.47 to 42.73, SUCRA = 0.92). In the medium term, PPRP_HPC_Y_N demonstrated superior efficacy (WMD = 18.94; 95% CrI, 8.34 to 28.12, SUCRA = 0.94). Long-term results indicated that PPRP_LPC_Y_N was the highest-ranked treatment (WMD = 17.09; 95% CrI, -8.4 to 42.78, SUCRA = 0.86). All models showed good fit with minimal heterogeneity.
WOMAC stiffness
RPRP_LPC_Y_Y was the most effective short-term intervention (WMD = 26.02; 95% CrI, 0.92 to 52.46, SUCRA = 0.98). In the medium term, PPRP_HPC_Y_N ranked highest (WMD = 3.52; 95% CrI, 1.22 to 5.96, SUCRA = 0.89). Long-term analysis was conducted in two sub-networks: PPRP_HPC_X_N was the top treatment in the HA network (WMD = -0.8; 95% CrI, -1.98 to 0.39, SUCRA = 0.77), while PPRP_MPC_Y_N ranked highest in the Saline network (WMD = 1.62; 95% CrI, -5.35 to 8.54, SUCRA = 0.73).
These results indicate that PRP preparations with high platelet and leukocyte concentrations, particularly RPRP_LPC_Y_Y and PPRP_HPC_Y_N, provide superior short- and medium-term benefits, while different PRP formulations maintain efficacy in the long term. The full network meta-analysis results are available in Appendix 4.
Additional analysis
All the models had a good fit, and the potential scale reduction factor was very close to 1. No significant inconsistency was detected between other direct and indirect comparisons (P > 0.05) (Fig. 2).Network meta-regression showed no significant interactions between female, BMI, patient age, sample size, and treatment effect(Appendix 5).
Fig. 2.
The forest plot illustrates the differences in WOMAC Pain, WOMAC Function, and WOMAC Stiffness scores in comparison to HA and Saline, along with the SCURA values
Discussion
The NMA evaluates the relative clinical efficacy of different PRP preparation protocols for osteoarthritis in comparison to standard treatments such as HA and saline.
The results revealed demonstrate that the effects of different PRP preparation methods on WOMAC Pain, WOMAC Function, and WOMAC Stiffness scores vary across in short-term, medium-term and long-term periods of OA treatment.
In the short-term follow-up, the RPRP_LPC_Y_Y preparation obtained best results in all WOMAC Pain, Function, and Stiffness scores, indicating the rich leukocyte and low platelet combination PRP with an immediate clinical efficacy in OA treatment. The inflammatory environment in OA pathology will activate local immune response and promote pro-inflammatory factors, which may further trigger and exacerbate the inflammatory reaction. These cytokines take an important role OA progress, as well as the development of pain in OA patient. For instance, TNF-α as a well-known inflammatory cytokine will stimulate nociceptors in OA patients when elevated in the synovium or joint fluid, which finally causes the sensation of OA-related pain [5]. Moreover, IL-1Ra as an anti-inflammatory mediator in OA, was found to have a higher expression in PRP with rich leukocyte than that with a lower leukocyte level [57]. Hence, as the increased leukocytes of RPRP_LPC_Y_Y preparation were injected into the joint, they might play a similar role with NSAIDs drugs in directly suppressing local inflammatory responses [10]. Early intervention to inflammatory environment can further reduce the production of pro-inflammatory cytokines and proteolytic enzymes associating with cellular catabolism and apoptosis [28, 62]. That may ultimately alleviate pain sensitization and tissue fibrosis, as well as prevent joint adhesions [47]. Moreover, well controlled pain allows OA patients to move their knee joints, resulting in joint stiffness relief and function improvement. Therefore, inhibition of inflammatory environment by leukocyte-rich, low-concentration platelet PRP alleviated pains and joint dysfunction of OA patients in an early point and then obtained remarkable WOMAC scores in the short-term follow-up.
PPRP_HPC preparations manifested the most favorable outcomes in the long-term follow-up suggesting that the combined therapy of leukocyte-poor and high-platelet PRP holds significant clinical efficacy in the treatment of osteoarthritis. The low-leukocyte and high-platelet PRP combination enhanced the secretion of growth factors which may play a crucial role in cartilage repair and functional recovery in osteoarthritis [35]. For instance, PDGF and TGF-βcontained in the PRP can stimulate chondrocytes proliferation and differentiation contributing to the repair of cartilage and other joint tissues [31, 36]. The enhanced secretion of IGF-1 can promotes collagen production and extracellular matrix production promoting tendon and cartilage healing [41]. These growth factors also enhance the blood supply around the joint, which may further accelerate cartilage regeneration and ultimately improve joint function [21, 22]. Beyond these effects, high-platelet PRP induces the temporal and spatial release of bioactive molecules that not only stimulate resident chondrocytes but also attract mesenchymal stem cells to the injured site, fostering a regenerative microenvironment [1, 2]. The sustained delivery of factors such as VEGF and IGF-1 contributes to subchondral vascular remodeling and matrix biosynthesis [2]while the modulation of the local immune response—characterized by downregulation of catabolic cytokines like IL-1β and TNF-α and upregulation of anti-inflammatory mediators like IL-10—helps to mitigate chronic inflammation and create favorable conditions for tissue repair [1, 2]. In certain patient populations, individual variability in regenerative capacity and immune status may influence response to PRP therapy, making the optimization of leukocyte and platelet profiles particularly important for enhancing clinical outcomes [3]. It is also important to note that the efficacy of PRP may be affected by concurrent medication use; agents such as NSAIDs or corticosteroids have been shown to impair platelet activation and suppress growth factor release, thereby potentially attenuating the therapeutic effect of the injection [24]. While cartilage repair is not an instant process, PPRP_HPC preparations presented no significant advantages in the early stage of functional scores. PPRP_HPC gradually showed benefits in supporting tissue regeneration over time which is reflected in better long-term WOMAC scores [22]. Thus, in the mid-term follow-up (6–12 months), dual roles of different leukocyte and platelet concentrations might cause confused statuses of inflammation control and tissue regeneration in the OA joint and lead to varied clinical outcomes.
Limitations
This study has several important limitations. First, the presence of clinical and methodological heterogeneity among the included studies — such as variations in patient characteristics, PRP preparation protocols, injection frequencies, control interventions (HA or saline), and outcome assessment tools — reduces direct comparability and may impact the reliability of pooled estimates. To explore potential sources of heterogeneity, meta-regression analyses were performed based on study-level variables, including mean age, BMI, sex distribution, and total sample size. Second, the lack of standardized classification criteria for PRP preparations across studies undermines the consistency and scientific reproducibility of comparisons. Third, although publication bias was assessed using funnel plots and Egger’s test, the presence of asymmetry in some comparisons may still influence the robustness of findings. Fourth, the scarcity of long-term follow-up data, particularly for trials using HA or saline controls, limits the ability to draw definitive conclusions about sustained treatment efficacy. Lastly, six studies were rated as having a moderate risk of bias, which could influence the overall quality and credibility of the findings. These limitations should be considered when interpreting the results and highlight key areas for methodological improvement in future research.
Conclusions
This network meta-analysis supports a tailored approach to PRP treatment based on osteoarthritis severity and therapeutic objectives. For early-stage OA, where symptoms are primarily pain and stiffness, low platelet concentration, high leukocyte PRP (e.g., RPRP_LPC_Y_Y) appears more effective, potentially due to its anti-inflammatory properties. In contrast, for advanced OA with significant cartilage damage, high platelet concentration, low leukocyte PRP (e.g., PPRP_HPC) may offer greater benefit by providing a higher concentration of growth factors to support tissue repair and regeneration. While our findings provide evidence-based guidance for protocol selection, clinical implementation should also consider feasibility factors such as preparation complexity, standardization, cost, and patient-specific characteristics. Further high-quality randomized trials with standardized PRP classification and longer follow-up durations are needed to strengthen the evidence base and optimize individualized treatment strategies.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Author contributions
Dongsheng Yu contributed to the conception and design of the study, conducted the data analysis, and wrote the manuscript. Jiani Zhao assisted in the interpretation of the data and provided critical revisions to the manuscript. Kun Zhao, as the corresponding author, supervised the overall study, ensured the integrity of the research, and contributed to the final manuscript revisions. All authors have read and approved the final manuscript.
Funding
This study was supported by the Natural Science Foundation of China (Young Scientists Fund, Grant No.82402959).
Data availability
No datasets were generated or analysed during the current study.
Declarations
Human ethics and consent to participate
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Jiani Zhao Equal contribution as first authors.
References
- 1.Andia I, Maffulli N. A contemporary view of platelet-rich plasma therapies: moving toward refined clinical protocols and precise indications. Regen Med. 2018;13(6):717–28. 10.2217/rme-2018-0042. [DOI] [PubMed] [Google Scholar]
- 2.Andia I, Maffulli N. Blood-Derived products for tissue repair/regeneration. Int J Mol Sci. 2019;20(18):4581. 10.3390/ijms20184581. Published 2019 Sep 17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Andia I, Maffulli N. Some patients (and some of us) respond better to some biological therapies: the as yet unsolved conundrum. J Orthop Traumatol. 2018;19(1):1. 10.1186/s10195-018-0505-z. Published 2018 Aug 20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Anitua E, Sánchez M, Orive G, Padilla S. A biological therapy to osteoarthritis treatment using platelet-rich plasma. Expert Opin Biol Ther. 2013;13(8):1161–72. 10.1517/14712598.2013.801450. [DOI] [PubMed] [Google Scholar]
- 5.Assi R, Quintiens J, Monteagudo S, et al. Innovation in targeted intra-articular therapies for osteoarthritis. Drugs. 2023;83:649–63. 10.1007/s40265-023-01863-y. [DOI] [PubMed] [Google Scholar]
- 6.Bannuru RR, Schmid CH, Kent DM, et al. Comparative effectiveness of Pharmacologic interventions for knee osteoarthritis: a systematic review and network meta-analysis. Ann Intern Med. 2015;162(1):46–54. 10.7326/m14-1231. [DOI] [PubMed] [Google Scholar]
- 7.Bansal H, Leon J, Pont JL, et al. Platelet-rich plasma (PRP) in osteoarthritis (OA) knee: correct dose critical for long-term clinical efficacy. Sci Rep. 2021;11:3971. 10.1038/s41598-021-83025-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Behera P, Dhillon M, Aggarwal S, Marwaha N, Prakash M. Leukocyte-poor platelet-rich plasma versus bupivacaine for recalcitrant lateral epicondylar tendinopathy. J Orthop Surg (Hong Kong). 2015;23(1):6–10. [DOI] [PubMed] [Google Scholar]
- 9.Boswell SG, Schnabel LV, Mohammed HO, et al. Increasing platelet concentrations in leukocyte-reduced platelet-rich plasma decrease collagen gene synthesis in tendons. Am J Sports Med. 2014;42(1):42–9. 10.1177/0363546513490656. [DOI] [PubMed] [Google Scholar]
- 10.Buendía-López D, Medina-Quirós M, Fernández-Villacañas Marín MÁ. Clinical and radiographic comparison of a single LP-PRP injection, a single hyaluronic acid injection and daily NSAID administration with a 52-week follow-up: a randomized controlled trial. J Orthop Traumatol. 2018;19(1):3. 10.1186/s10195-018-0501-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Byrne AM, Bouchier-Hayes DJ, Harmey JH. Angiogenic and cell survival functions of vascular endothelial growth factor (VEGF). J Cell Mol Med. 2005;9(4):777–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.DeLong JM, Russell RP, Mazzocca AD. Platelet-rich plasma: the PAW classification system. Arthroscopy. 2012;28(7):998–1009. 10.1016/j.arthro.2012.04.148. [DOI] [PubMed] [Google Scholar]
- 13.Duymus TM, Mutlu S, Dernek B, et al. Choice of intra-articular injection in treatment of knee osteoarthritis: platelet-rich plasma, hyaluronic acid or Ozone options. Knee Surg Sports Traumatol Arthrosc. 2017;25(2):485–92. 10.1007/s00167-016-4110-5. [DOI] [PubMed] [Google Scholar]
- 14.Dório M, Pereira RMR, Luz AGB, et al. Efficacy of platelet-rich plasma and plasma for symptomatic treatment of knee osteoarthritis: a double-blinded placebo-controlled randomized clinical trial. BMC Musculoskelet Disord. 2021;22(1):822. 10.1186/s12891-021-04706-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Elik H, Doğu B, Yılmaz F, et al. The efficiency of platelet-rich plasma treatment in patients with knee osteoarthritis. J Back Musculoskelet Rehabil. 2020;33(1):127–38. 10.3233/BMR-181374. [DOI] [PubMed] [Google Scholar]
- 16.Everts PA, Lana JF, Alexander RW, et al. Profound properties of protein-rich, platelet-rich plasma matrices as novel, multi-purpose biological platforms in tissue repair, regeneration, and wound healing. Int J Mol Sci. 2024;25:7914. 10.3390/ijms25147914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Everts PA, Lana JF, Onishi K, et al. Angiogenesis and tissue repair depend on platelet dosing and bioformulation strategies following orthobiological platelet-rich plasma procedures: a narrative review. Biomedicines. 2023;11:1922. 10.3390/biomedicines11071922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Filardo G, Kon E, Di Martino A, et al. Platelet-rich plasma vs hyaluronic acid to treat knee degenerative pathology: study design and preliminary results of a randomized controlled trial. BMC Musculoskelet Disord. 2012;13:229. 10.1186/1471-2474-13-229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gupta A, Jeyaraman M, Maffulli N. Common medications which should be stopped prior to Platelet-Rich plasma injection. Biomedicines. 2022;10(9):2134. 10.3390/biomedicines10092134. Published 2022 Aug 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–60. 10.1136/bmj.327.7414.557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Huang Y, Liu X, Xu X, et al. Intra-articular injections of platelet-rich plasma, hyaluronic acid or corticosteroids for knee osteoarthritis: a prospective randomized controlled study. Orthopade. 2019;48(3):239–47. 10.1007/s00132-018-03659-5. [DOI] [PubMed] [Google Scholar]
- 22.Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17(1):1–12. 10.1016/0197-2456(95)00134-4. [DOI] [PubMed] [Google Scholar]
- 23.Jayaram P, Liu C, Dawson B, et al. Leukocyte-dependent effects of platelet-rich plasma on cartilage loss and thermal hyperalgesia in a mouse model of post-traumatic osteoarthritis. Osteoarthritis Cartilage. 2020;28(10):1385–93. 10.1016/j.joca.2020.07.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kabiri A, Esfandiari E, Esmaeili A, et al. Platelet-rich plasma application in chondrogenesis. Adv Biomed Res. 2014;3:138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lana JF, Weglein A, Sampson SE, et al. Randomized controlled trial comparing hyaluronic acid, platelet-rich plasma and the combination of both in the treatment of mild and moderate osteoarthritis of the knee. J Stem Cells Regen Med. 2016;12(2):69–78. 10.46582/jsrm.1202011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Laudy ABM, Bakker EWP, Rekers M, et al. Efficacy of platelet-rich plasma injections in osteoarthritis of the knee: a systematic review and meta-analysis. Br J Sports Med. 2015;49:657–72. [DOI] [PubMed] [Google Scholar]
- 27.Lin KY, Yang CC, Hsu CJ, Yeh ML, Renn JH. Intra-articular injection of platelet-rich plasma is superior to hyaluronic acid or saline solution in the treatment of mild to moderate knee osteoarthritis: a randomized, double-blind, triple-parallel, placebo-controlled clinical trial. Arthroscopy. 2019;35(1):106–17. 10.1016/j.arthro.2018.06.035. [DOI] [PubMed] [Google Scholar]
- 28.Louis ML, Magalon J, Jouve E, et al. Growth factors levels determine efficacy of platelet-rich plasma injection in knee osteoarthritis: a randomized double-blind noninferiority trial compared with viscosupplementation. Arthroscopy. 2018;34(5):1530–e15402. 10.1016/j.arthro.2017.11.035. [DOI] [PubMed] [Google Scholar]
- 29.Lynch SE, Nixon JC, Colvin RB, Antoniades HN. Role of platelet-derived growth factor in wound healing: synergistic effects with other growth factors. Proc Natl Acad Sci U S A. 1987;84(21):7696–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Magalon J, Bausset O, Serratrice N, et al. Characterization and comparison of 5 platelet-rich plasma preparations in a single-donor model. Arthroscopy. 2014;30(5):629–38. 10.1016/j.arthro.2014.02.020. [DOI] [PubMed] [Google Scholar]
- 31.McCabe PS, Maricar N, Parkes MJ, Felson DT, O’Neill TW. The efficacy of intra-articular steroids in hip osteoarthritis: a systematic review. Osteoarthritis Cartilage. 2016;24(9):1509–17. 10.1016/j.joca.2016.04.018. [DOI] [PubMed] [Google Scholar]
- 32.Nelson AE. Osteoarthritis year in review 2017: clinical. Osteoarthritis Cartilage. 2018;26(3):319–25. 10.1016/j.joca.2017.11.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Nunes-Tamashiro JC, Natour J, Ramuth FM, et al. Intra-articular injection with platelet-rich plasma compared to triamcinolone Hexacetonide or saline solution in knee osteoarthritis: a double-blinded randomized controlled trial with one year follow-up. Clin Rehabil. 2022;36(7):900–15. 10.1177/02692155221090407. [DOI] [PubMed] [Google Scholar]
- 34.Olesen JL, Heinemeier KM, Haddad F, et al. Expression of insulin-like growth factor I, insulin-like growth factor binding proteins, and collagen mRNA in mechanically loaded plantaris tendon. J Appl Physiol (1985). 2006;101(1):183–8. [DOI] [PubMed] [Google Scholar]
- 35.Park YB, Kim JH, Ha CW, Lee DH. Clinical efficacy of platelet-rich plasma injection and its association with growth factors in the treatment of mild to moderate knee osteoarthritis: a randomized double-blind controlled clinical trial as compared with hyaluronic acid. Am J Sports Med. 2021;49(2):487–96. 10.1177/0363546520986867. [DOI] [PubMed] [Google Scholar]
- 36.Patel S, Dhillon MS, Aggarwal S, et al. Treatment with platelet-rich plasma is more effective than placebo for knee osteoarthritis: a prospective, double-blind, randomized trial. Am J Sports Med. 2013;41(2):356–64. 10.1177/0363546512471299. [DOI] [PubMed] [Google Scholar]
- 37.Raeissadat SA, Gharooee Ahangar A, Rayegani SM, et al. Platelet-rich plasma-derived growth factor vs hyaluronic acid injection in individuals with knee osteoarthritis: a one-year randomized clinical trial. J Pain Res. 2020;13:1699–711. 10.2147/JPR.S210715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Raeissadat SA, Ghazi Hosseini P, Bahrami MH, et al. The comparison effects of intra-articular injection of platelet-rich plasma, plasma rich in growth factor, hyaluronic acid, and Ozone in knee osteoarthritis: a one-year randomized clinical trial. BMC Musculoskelet Disord. 2021;22(1):134. 10.1186/s12891-021-04017-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Raeissadat SA, Rayegani SM, Ahangar AG, et al. Efficacy of intra-articular injection of a newly developed plasma rich in growth factor (PRGF) versus hyaluronic acid on pain and function of patients with knee osteoarthritis: a single-blinded randomized clinical trial. Clin Med Insights Arthritis Musculoskelet Disord. 2017;10:1179544117733452. 10.1177/1179544117733452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Remst DG, Davidson EB, Kraan PM. Unravelling osteoarthritis-related synovial fibrosis: a step closer to solving joint stiffness. Rheumatology (Oxford). 2015;54(11):1954–63. 10.1093/rheumatology/kev228. [DOI] [PubMed] [Google Scholar]
- 41.Rucker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Med Res Methodol. 2015;15:58. 10.1186/s12874-015-0060-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Smith OJ, Talaat S, Tomouk T, Jell G, Mosahebi A. An evaluation of the effect of activation methods on the release of growth factors from platelet-rich plasma. Plast Reconstr Surg. 2022;149(2):404–11. 10.1097/PRS.0000000000008772. [DOI] [PubMed] [Google Scholar]
- 43.Spaková T, Rosocha J, Lacko M, et al. Treatment of knee joint osteoarthritis with autologous platelet-rich plasma in comparison with hyaluronic acid. Am J Phys Med Rehabil. 2012;91(5):411–7. 10.1097/PHM.0b013e3182aab72. [DOI] [PubMed] [Google Scholar]
- 44.Su K, Bai Y, Wang J, et al. Comparison of hyaluronic acid and PRP intra-articular injection with combined intra-articular and intraosseous PRP injections to treat patients with knee osteoarthritis. Clin Rheumatol. 2018;37(5):1341–50. 10.1007/s10067-018-3985-6. [DOI] [PubMed] [Google Scholar]
- 45.Sundman EA, Cole BJ, Fortier LA. Growth factor and catabolic cytokine concentrations are influenced by the cellular composition of platelet-rich plasma. Am J Sports Med. 2011;39(10):2135–40. [DOI] [PubMed] [Google Scholar]
- 46.Sundman EA, Cole BJ, Karas V, et al. The anti-inflammatory and matrix restorative mechanisms of platelet-rich plasma in osteoarthritis. Am J Sports Med. 2014;42:35–41. [DOI] [PubMed] [Google Scholar]
- 47.Tokunaga A, Oya T, Ishii Y, et al. PDGF receptor beta is a potent regulator of mesenchymal stromal cell function. J Bone Min Res. 2008;23(9):1519–28. [DOI] [PubMed] [Google Scholar]
- 48.Tucker JD, Goetz LL, Duncan MB, et al. Randomized, placebo-controlled analysis of the knee synovial environment following platelet-rich plasma treatment for knee osteoarthritis. PM R. 2021;13(7):707–17. 10.1002/pmrj.12561. [DOI] [PubMed] [Google Scholar]
- 49.van Middelkoop M, Arden NK, Atchia I, et al. The OA trial bank: meta-analysis of individual patient data from knee and hip osteoarthritis trials show that patients with severe pain exhibit greater benefit from intra-articular glucocorticoids. Osteoarthritis Cartilage. 2016;24(7):1143–52. 10.1016/j.joca.2016.01.983. [DOI] [PubMed] [Google Scholar]
- 50.Vasconcelos DP, Jabangwe C, Lamghari M, Alves CJ. The neuroimmune interplay in joint pain: the role of macrophages. Front Immunol. 2022;13:812962. 10.3389/fimmu.2022.812962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Wasterlain AS, Braun HJ, Dragoo JL. Contents and formulations of platelet-rich plasma. Oper Tech Orthop. 2012;22(1):33–42. [Google Scholar]
- 52.Werner S, Grose R. Regulation of wound healing by growth factors and cytokines. Physiol Rev. 2003;83(3):835–70. [DOI] [PubMed] [Google Scholar]
- 53.White AG, Birnbaum HG, Janagap C, Buteau S, Schein J. Direct and indirect costs of pain therapy for osteoarthritis in an insured population in the united States. J Occup Environ Med. 2008;50(9):998–1005. 10.1097/JOM.0b013e3181715111. [DOI] [PubMed] [Google Scholar]
- 54.Yaradilmis YU, Demirkale I, Tagral SA, et al. Comparison of two platelet-rich plasma formulations with viscosupplementation in treatment of moderate-grade gonarthrosis: a prospective randomized controlled study. J Orthop. 2020;20:240–6. 10.1016/j.jor.2020.01.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Yoshida R, Cheng M, Murray MM. Increasing platelet concentration in platelet-rich plasma inhibits anterior cruciate ligament cell function in three-dimensional culture. J Orthop Res. 2014;32(2):291–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.