Platelet-Rich Plasma for Treating Chronic Noncancer Pain: A Systematic Review and Meta-analysis of Randomized Controlled Trials

Abstract

Introduction

Chronic noncancer pain represents a significant global health challenge, contributing to disability, lost productivity, diminished quality of life, and substantial socioeconomic burden. Platelet-rich plasma (PRP) has emerged as a promising therapeutic option for managing chronic pain. However, a comprehensive assessment of its efficacy and the evidence supporting its use remains limited. This study aimed to systematically evaluate the analgesic effectiveness of PRP compared with placebo or active drug treatments across a wide range of chronic noncancer pain conditions using a rigorous meta-analytic approach. The goal is to provide evidence-based insights to inform clinical decision-making and improve patient outcomes.

Methods

Following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, a comprehensive literature search was conducted in the PubMed, Embase, MEDLINE, and Cochrane Library databases to identify randomized controlled trials (RCTs). Studies were screened according to predefined inclusion and exclusion criteria. A random-effects model was applied to account for heterogeneity among studies. The primary outcome, pain scores in patients with chronic noncancer pain, was assessed using the standardized mean difference (SMD). The risk of bias of the included studies was evaluated using the Revised Cochrane Risk-of-Bias Tool (RoB 2). The quality of evidence was rated by the Grade of Recommendations Assessment, Development, and Evaluation (GRADE) approach.

Results

A total of 691 RCTs were screened, and 56 studies (comprising 103 comparisons and 7142 patients) were eligible for analysis. PRP was associated with a statistically significant reduction in pain scores compared with both active drug treatments and placebo (SMD = −0.37, 95% confidence interval (CI) −0.59 to −0.15, p = 0.001). No significant differences were observed in pain scores for follow-up periods shorter than 3 months (SMD = 0.12, 95% CI −0.16 to 0.40, p > 0.05). A statistically significant and moderate reduction in pain score was found for follow-up durations of at least 3 months (SMD = −0.69, 95% CI −0.98 to −0.40, p < 0.001). Meta-analyses of subgroups revealed statistically significant and moderate pain reduction in favor of PRP versus active drug treatments for osteoarthritic knee pain (SMD = −0.59, 95% CI −1.01 to −0.17, p = 0.009) and rotator cuff tendinopathy/tear (SMD = −0.60, 95% CI −1.01 to −0.19, p = 0.01), but no significant differences for plantar fasciitis (SMD = 0.03, 95% CI −0.98 to 1.04, p > 0.05). PRP was associated with moderate pain reduction when compared with corticosteroid (SMD = −0.53, 95% CI −0.98 to −0.08, p = 0.02) and hyaluronic acid injection (SMD = −0.55, 95% CI −0.89 to −0.21, p = 0.004).

Conclusions

PRP injections appear to effectively reduce pain in various chronic noncancer pain conditions and show superior analgesic efficacy compared with corticosteroid and hyaluronic acid injections. These findings suggest that PRP may be a preferred treatment option for managing chronic noncancer pain, offering a more sustainable alternative for long-term pain relief.

Systematic Review Registration

PROSPERO CRD42023441115.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40122-025-00751-5.

Key Summary Points

Why carry out this study?

Chronic noncancer pain is a significant global health issue, contributing to disability, lost productivity, and a considerable socioeconomic burden. Platelet-rich plasma (PRP) has emerged as a promising therapeutic option for managing various chronic pain conditions. However, there has been no comprehensive and systematic assessment of the impact of PRP on chronic noncancer pain and its efficacy compared with placebo or active drug treatments.

Given the prevalence and burden of conditions such as osteoarthritic knee pain and rotator cuff tendinopathy, understanding the effectiveness of PRP in these conditions is crucial. This study aims to fill this gap by systematically evaluating the analgesic effectiveness of PRP in a wide range of chronic noncancer pain conditions, providing evidence-informed insights to guide clinical decision-making and improve patient outcomes.

What was learned from the study?

PRP injections demonstrate significant efficacy in reducing pain across various chronic noncancer pain conditions. Specifically, PRP was found to be effective in conditions such as osteoarthritic knee pain and rotator cuff tendinopathy/tear. In these cases, PRP not only provided statistically significant pain reduction compared with placebo and active drug treatments but also showed superior analgesic efficacy over corticosteroid and hyaluronic acid injections.

This finding holds substantial implications for improving patients’ quality of life and offers a promising, long-term solution for chronic pain management. By providing a more sustainable alternative for pain relief, PRP could be a valuable addition to community and public health strategies aimed at managing chronic pain conditions.

Introduction

Chronic noncancer pain is a major health issue that contributes to disability affecting 20% of the world population, loss of work, and impaired quality of life, as well as significant socioeconomic burden [1]. It is also an important reason for long-term opioid use [2, 3]. Given the concerns with the opioid epidemic, alternative analgesic strategies to reduce long-term opioid use is essential [2]. Management of chronic noncancer pain requires a biopsychosocial and multifactorial treatment approach, including the use of pharmacological drugs and active lifestyle treatment (e.g., exercise, weight loss). Platelet-rich plasma (PRP) is a derivative of autologous whole blood with concentrations of platelets that are 3–5 times higher than typical levels after centrifugation [4]. It contains an abundance of cytokines, growth factors, and various bioactive molecules that promotes tissue regeneration, enhances angiogenesis, modulates the immune system, and produces anti-inflammatory effect [5–8]. Injection of PRP into target tissues has been gaining popularity for managing chronic pain, especially musculoskeletal pain conditions [9].

PRP therapy is considered as a safe, effective, and viable treatment option [10]. It has been employed in clinical trials to address certain chronic pain conditions, including knee osteoarthritis and rotator cuff injuries [11, 12]. However, evidence from systematic reviews and meta-analyses evaluating the analgesic efficacy of PRP on pain intensity in patients with chronic noncancer pain is lacking. Current systematic review and meta-analyses have studied its effect on specific indications such as osteoarthritis of the knee and lateral epicondylitis, rather than across different chronic pain conditions [13, 14]. The analgesic effect of PRP in a number of other chronic noncancer pain conditions is not clear. Additionally, there is a relative scarcity of clinical evidence regarding its effectiveness compared with other common active injectates such as corticosteroid and hyaluronic acid. It is not known whether different preparations of PRP would affect analgesic outcomes. Given its emerging role as an analgesic option, it is important to perform a comprehensive systematic review on its pain-relieving effect for all types of chronic noncancer pain conditions.

We performed a systematic review and meta-analysis to evaluate the analgesic effectiveness of PRP across a broad range of chronic noncancer pain conditions, involving a comprehensive search for randomized controlled trials (RCTs). Additional aims included comparing PRP’s efficacy against different active drugs and assessing its analgesic effects across various chronic pain types. Through this study, we offer an extensive and quantitative assessment of PRP’s role in managing chronic noncancer pain. This is crucial for establishing its clinical value as a potential treatment alternative for individuals experiencing chronic noncancer pain.

Methods

Our study followed the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, which is specifically designed for reporting systematic reviews of randomized controlled trials [15]. We registered our review protocol in PROSPERO (CRD42023441115).

Eligibility Criteria

This study was a systematic review that utilized data exclusively from published literature, without conducting any new trials or experiments. As such, ethical approval was not required, as the study relied solely on previously conducted research and did not involve any new research by the authors that included human participants or animals. The inclusion criteria were as follows:

Participants

Participants with various chronic noncancer pain conditions, such as osteoarthritic knee pain, plantar fasciitis, rotator cuff tendinopathy/tears, and other pain conditions.

Intervention

Intervention involving PRP as a treatment method, with pain score as a research outcome.

Comparator

Comparison with placebo or active drug treatments.

Outcome

The primary outcome focused on assessing the analgesic efficacy of PRP injections in reducing pain scores among patients with chronic noncancer pain. Pain was measured using validated scales such as the visual analog scale (VAS), numeric rating scale (NRS), or other measurements. Secondary outcomes included evaluations of physical and/or emotional functioning. Eligible instruments for assessing physical and/or emotional functioning varied depending on the condition being studied and included the following: the American Orthopaedic Foot and Ankle Society (AOFAS) score for plantar fasciitis, the American Shoulder and Elbow Surgeons (ASES) score and the Western Ontario Rotator Cuff (WORC) index for rotator cuff pain, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) for osteoarthritic knee pain, and other condition-specific measurement scales.

Study Types

This review included RCTs without geographical limitations but was restricted to studies published in the English language.

Exclusion Criteria

The exclusion criteria consisted of (1) cancer-related pain, (2) before–after comparison, (3) oral medication and surgery, (4) being incomplete or lacking results, (5) nonrandomized controlled trials, qualitative studies or observational, commentaries or case reports, and (6) preclinical animal laboratory studies.

Comprehensive Strategy for Literature Search

An extensive search of the literature was carried out using PubMed, Embase, MEDLINE, and the Cochrane Library to locate all peer-reviewed randomized controlled trials comparing PRP with placebo or an active drug treatment for patients suffering from chronic noncancer pain up until May 2023. The keyword search was restricted to English publications. The title or abstract of the paper must have included at least one keyword by containing either “platelet-rich plasma” or “platelet rich plasma” or “PRP”, and “pain”, and either the words “RCT”, “randomized controlled trial”, or “randomised controlled trial” [16]. Two investigators performed the systematic search for literature, selection of studies, and collection of relevant data independently. They screened all articles by examining titles, abstracts, and full texts. Discrepancies were discussed between the two investigators.

Data Extraction and Synthesis

Two investigators extracted the required information from the clinical trials that met the selection criteria: first author, publication year, pain outcome, other outcomes, sample size, types of medication, pain conditions, and duration. In cases where the selected publications featured data from multiple groups, only suitable groups were extracted and used for the following meta-analysis. Change scores from baseline were used for the meta-analysis because they account for individual baseline differences, providing a more standardized measure of treatment effect and directly reflecting the amount of improvement or deterioration over time. However, when change scores were not available, post follow-up pain scores were also considered. Including post follow-up scores allowed for the incorporation of a broader range of studies, enhancing the comprehensiveness and robustness of the analysis.

Evaluating the Risk of Bias in Clinical Trials

In this study, the risk of bias in randomized controlled trials was assessed using the Revised Cochrane Risk-of-Bias Tool (RoB 2) [17]. This tool evaluates five domains of potential bias: the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. For each domain, a series of signalling questions were answered to guide judgment, categorizing the risk of bias as “low,” “some concerns,” or “high.” These judgments were then synthesized to provide an overall risk of bias rating for each study, ensuring a thorough and systematic evaluation of methodological quality.

Rating of Quality of Evidence

The quality of evidence for each meta-analysis was evaluated utilizing the Grade of Recommendations Assessment, Development, and Evaluation (GRADE) method [18]. This method classified the evidence level into four categories: high, moderate, low, and very low, according to five key factors: (i) risk of bias evaluating the potential for bias in the study design, methodology, and execution; (ii) inconsistency examining the variability or heterogeneity in the results of different studies; (iii) indirectness assessing whether the evidence is applicable to the specific research question or target population; (iv) imprecision evaluated the uncertainty in the effect estimates, assessed by evaluating both the optimal information size (OIS) and the width of the confidence intervals (CIs) around the effect estimates. The OIS was determined to ensure that the sample size was sufficient to achieve 80% power for detecting a clinically significant difference; and (v) publication bias considering the possibility that studies yielding positive outcomes have a higher probability of being published compared with those presenting negative or inconclusive findings [18].

Data Synthesis and Statistical Analysis

We evaluated the effect size using the standardized mean difference (SMD) across various pain scale scores. This was then classified on the basis of Cohen’s benchmark: negligible (d < 0.2), small (0.2 ≤ d < 0.5), moderate (0.5 ≤ d < 0.8), and large (d ≥ 0.8) [19]. Hedges’s g was used to express the SMD by applying a correction factor J to Cohen’s d, thereby adjusting for small sample sizes [20].

$$J=1-\frac{3}{4\left(n_1,+,n_2\right)-9}.$$

Given the extensive diversity in subjects, populations, dosage, administration methods, and pain conditions across various clinical trials, a random effects (RE) model was employed using the restricted maximum likelihood (REML) method supplemented with Knapp–Hartung adjustment [16]. The presence of heterogeneity and inconsistency was evaluated using the I², tau² (τ²) and Q statistic (with a p-value of < 0.05 indicating the presence of heterogeneity). I² is a statistic that quantifies the proportion of total variation across studies in a meta-analysis that is due to heterogeneity rather than chance, with more than 50% indicating substantial heterogeneity [21]. Tau² (τ²) is a measure of between-study variance in a random-effects meta-analysis, representing the absolute magnitude of this variance [22]. When τ² = 0, it indicates no between-study variance, meaning that all the studies are estimating the same effect size and there is no heterogeneity. Conversely, when τ² > 0, it signifies some degree of between-study variance. The larger the τ² value, the greater the heterogeneity among the study estimates [22]. Multivariate meta-regression was utilized to assess the impact of both the type of control groups and the length of follow-up periods on treatment effect sizes. To assess publication bias, a funnel plot combined with Egger’s regression statistical test was employed [23]. The analyses were conducted using the metafor and netmeta packages from R version 4.4.0 (2024, URL https://www.r-project.org). The result was defined as statistically significant if a two-sided p-value was less than 0.05.

Results

Study Selection

We identified 691 articles for further selection (Fig. 1). There were 396 papers examined for study eligibility after removing duplicates. Upon screening the title/abstract, 230 studies that did not meet the inclusion criteria were removed. Additionally, 110 studies were removed after reviewing the full-text articles owing to the following reasons: (i) not meet the inclusion criterion (n = 101), (ii) not a randomized controlled trial (n = 7), and (iii) not in English (n = 2). Fifty-six RCTs were selected for systematic review and meta-analysis [5, 6, 10, 11, 24–75]. Some articles included multiple comparisons.

Fig. 1.

Flowchart illustrating the process of identifying and selecting relevant studies. RCT randomized controlled trial

Study Quality and Characteristics

A total of 56 RCTs were included (in total, 103 comparisons), consisting of 7142 patients, with the patient sample sizes ranging from 19 to 610 patients (Supplementary Table S1). There were 37 comparisons for osteoarthritic knee pain, 10 comparisons for plantar fasciitis, 16 comparisons for rotator cuff tendinopathy/tears, and the remaining for other types of pain conditions (Supplementary Table S1). PRP was compared with placebo in 29 comparisons, corticosteroid in 34 comparisons, hyaluronic acid (HA) in 30 comparisons, and other active drug treatments in 10 comparisons (Supplementary Table S1). There were 41 comparisons with less than 3 months and 62 comparisons with at least 3 months of follow-up. The RoB 2 tool was used to assess the potential for bias in RCTs. The evaluation showed a low risk of bias in 5 studies, some concerns in 33 studies, and a high risk of bias in 18 studies (Supplementary Table S1). Regarding the blinding in the included studies, 14 studies were triple blinded (the patient, treating physician, and outcome assessor) (Supplementary Table S2). Seventeen studies were double-blinded, and 13 studies were single-blinded. In 12 studies, either no blinding was implemented or it was not clearly stated whether blinding was applied. The quality of evidence for each meta-analysis was assessed using the GRADE method, which rated the evidence quality on a scale ranging from very low to moderate (Supplementary Table S3).

Meta-analysis for Assessing Analgesic Effect in Chronic Noncancer Pain Conditions

The meta-analysis of all 103 comparisons (7142 patients) showed a statistically significant and small decrease in pain intensity across all chronic pain conditions in favor of PRP compared with placebo and active drug treatments (SMD = −0.37, 95% confidence interval [CI] −0.59 to −0.15, p = 0.001) (Supplementary Fig. S1). This pooled result had a moderate level of GRADE rating (Table 1), with significant heterogeneity and inconsistency (τ² = 1.06, I² = 94%, p < 0.001). Owing to the large effect size observed in the study by Cai et al. [29], and its 95% CI not overlapping with other investigations or the pooled results, a sensitivity analysis was carried out to determine the impact of this one study on the overall findings. Upon removing this RCT, a small reduction remained favoring PRP (SMD = −0.31, 95% CI −0.50 to −0.13, p = 0.001). Consequently, the study by Cai et al. was included in subsequent analysis because it did not statistically significantly affect the overall effect size.

Table 1.

Grade of Recommendations Assessment, Development, and Evaluation (GRADE) level of evidence

Meta-analysis	Number of comparisons	Treatment SMD (95% CI)	GRADE level of evidence
All clinical studies	103	−0.37 (−0.59 to −0.15)	⊕⊕⊕○
PRP versus active treatment	74	−0.20 (−0.42 to 0.02)	⊕⊕○○
PRP versus placebo	29	−0.83 (−1.37 to −0.29)	⊕⊕⊕○
< 3 months	41	0.12 (−0.16 to 0.40)	⊕⊕○○
≥ 3 months	62	−0.69 (−0.98 to −0.40)	⊕⊕⊕○
≥ 3 months, PRP versus active treatment	44	−0.59 (−0.84 to −0.34)	⊕⊕⊕○
Osteoarthritic knee	17	−0.59 (−1.01 to −0.17)	⊕⊕⊕○
Plantar fasciitis	7	0.03 (−0.98 to 1.04)	⊕○○○
Rotator cuff tendinopathy/tears	7	−0.60 (−1.01 to −0.19)	⊕⊕○○
PRP versus corticosteroid	20	−0.53 (−0.98 to −0.08)	⊕⊕⊕○
PRP versus HA	18	−0.55 (−0.89 to −0.21)	⊕⊕⊕○
Leukocyte-poor	7	−0.61 (−0.97 to −0.25)	⊕⊕○○
Leukocyte-rich	5	−0.09 (−1.36 to 1.18)	⊕○○○

PRP platelet-rich plasma, SMD standardized mean difference, CI confidence interval, ⊕⊕⊕⊕ high—the true effect is close to the estimated one; ⊕⊕⊕○ moderate—the true effect is likely to be close to the estimated one, but there is a possibility that it is substantially different; ⊕⊕○○ low—the true effect may be substantially different from the estimated one; ⊕○○○ very low—the true effect is likely to be substantially different from the estimated one

Seventy-four comparisons (involving 4706 patients) evaluated the analgesic effect of PRP against active drug treatments, and 29 comparisons (involving 2436 patients) compared PRP with placebo. No significant difference was found between PRP and active drug treatment groups (SMD = −0.20, 95% CI −0.42 to 0.02, p = 0.069) (Fig. 2a), accompanied by a low GRADE rating (Table 1) and significant heterogeneity (τ² = 0.77, I² = 92%, p < 0.05). In the meta-analysis of 29 comparisons comparing PRP with placebo, a statistically significant, large reduction favoring PRP was observed (SMD = −0.83, 95% CI −1.37 to −0.29, p = 0.004) (Fig. 2b), with a moderate GRADE rating (Table 1) and significant heterogeneity (τ² = 1.78, I² = 97%, p < 0.05).

Fig. 2.

The analgesic efficacy of PRP on chronic noncancer pain for a clinical trials versus active drug treatments and b clinical trials versus placebo. PRP platelet-rich plasma, CI confidence interval

When examining the analgesic effect of PRP for a follow-up period of less than 3 months across 41 comparisons (2554 patients), no significant differences were found compared with placebo and active drug treatments (SMD = 0.12, 95% CI −0.16 to 0.40, p > 0.05) (Fig. 3a). This was associated with a low GRADE rating (Table 1) and significant heterogeneity (τ² = 0.65, I² = 90%, p < 0.001). The meta-analysis of 62 comparisons (4588 patients) with at least 3 months of follow-up duration showed a statistically significant and moderate pain reduction in favor of PRP when compared with placebo and active drug treatments (SMD = −0.69, 95% CI −0.98 to −0.40, p < 0.001) (Fig. 3b). This had a moderate GRADE rating (Table 1) and significant heterogeneity (τ² = 1.08, I² = 95%, p < 0.001).

Fig. 3.

The analgesic efficacy of PRP on chronic noncancer pain for a clinical trials with less than 3 months follow-up when compared with placebo/active drug treatments, b clinical trials with at least 3 months when compared with placebo/active drug treatments, and c clinical trials with at least 3 months versus active drug treatments. CI confidence interval, PRP platelet-rich plasma, RE random effects, SD standard deviation, SMD standardized mean difference

Examining Covariate Impact on Effect Size through Meta-regression

A meta-regression analysis was conducted to evaluate the analgesic effect of PRP compared with various comparators. Statistically significant differences were found between RCTs comparing PRP with other active analgesics and those comparing PRP with placebo, as well as between clinical trials with follow-up durations of less than 3 months and those with follow-up durations of 3 months or more (p-values < 0.05) (Table 2).

Table 2.

Meta-regression comparing the analgesic effects for different comparators

Studies I	N	Studies II	N	Estimate	Standard error	p-Value
Placebo	29	Active treatment	74	−0.58	0.22	0.010
≥ 3 months	62	< 3 months	41	−0.79	0.20	< 0.001

N, number of comparisons. The results presented are from a single meta-regression model

Meta-analysis for RCTs with Follow-Up Duration of at Least 3 Months

Since PRP was used mainly for managing chronic noncancer pain, we performed a meta-analysis on RCTs with a follow-up duration of at least 3 months to evaluate its analgesic efficacy over a longer time period. Our primary objective was to assess the effect of PRP in comparison with active drug treatments. Therefore, we selected only the 38 randomized controlled trials (44 comparisons with 2834 patients) that had a follow-up duration of at least 3 months and compared PRP with active drug treatments for all subsequent analyses (Supplementary Table S4). The meta-analysis showed a significant and moderate pain reduction favoring PRP when compared with active drug treatments (SMD = −0.59, 95% CI −0.84 to −0.34, p < 0.001) (Fig. 3c), accompanied by a moderate level of GRADE rating (Table 1) and significant heterogeneity (τ² = 0.54, I² = 89%, p < 0.001).

Meta-analysis in Different Conditions of Chronic Pain

The effects of PRP versus active drug treatments in different chronic pain conditions with at least 3 months follow-up were examined: 17 comparisons for osteoarthritic knee pain (1291 patients), 7 comparisons for plantar fasciitis (449 patients), and 7 comparisons for rotator cuff tendinopathy/tear (501 patients). Meta-analysis was only performed when there were at least five randomized controlled trials available. There were insufficient studies (n < 5) to perform a meta-analysis for the following chronic pain conditions: temporomandibular joint disorders (n = 3), lateral epicondylitis (n = 3), chronic low back pain (n = 3), and others (n = 5).

A statistically significant, moderate pain decrease in osteoarthritic knee was observed in patients with PRP when compared with active drug treatments (SMD = −0.59, 95% CI −1.01 to −0.17, p = 0.009) (Fig. 4a), having a moderate level of GRADE rating (Table 1) and significant heterogeneity (τ² = 0.57, I² = 91%, p < 0.001). No statistically significant differences were identified in pain scores between treatment groups for plantar fasciitis (SMD = 0.03, 95% CI −0.98 to 1.04, p > 0.05) (Fig. 4b), accompanied by a very low level of GRADE rating (Table 1) and significant heterogeneity (τ² = 0.99, I² = 94%, p < 0.001). In the case of rotator cuff tendinopathy/tear, a significant, moderate pain decrease was noted with PRP compared with active drug treatments (SMD = −0.60, 95% CI −1.01 to −0.19, p = 0.01) (Fig. 4c), along with a low level of GRADE rating (Table 1) and significant heterogeneity (τ² = 0.14, I² = 70%, p = 0.002).

Fig. 4.

The analgesic efficacy of PRP when compared with active drug treatments for a osteoarthritic knee pain, b plantar fasciitis, and c rotator cuff tendinopathy/tears. CI confidence interval, PRP platelet-rich plasma, RE random effects, SD standard deviation, SMD standardized mean difference

Meta-analysis: Analgesic Effectiveness of PRP versus Different Active Drug Treatments

Meta-analyses for comparisons with at least 3 months follow-up versus other active drug treatments were conducted when there were enough clinical studies (n ≥ 3). Among the 20 comparisons (1168 patients) available for PRP versus corticosteroid injection, a statistically significant, moderate reduction favored PRP (SMD = −0.53, 95% CI −0.98 to −0.08, p = 0.02) (Fig. 5a), having a moderate level of GRADE rating (Table 1) and significant heterogeneity (τ² = 0.76, I² = 91%, p < 0.05).

Fig. 5.

The analgesic efficacy of PRP on chronic noncancer pain for clinical trials versus a corticosteroid and b HA. CI, confidence interval; PRP, platelet-rich plasma; HA, hyaluronic acid

Furthermore, 18 comparisons (1238 patients) were made to study the analgesic effect of PRP compared with HA. The pooled result revealed a statistically significant, moderate reduction favoring PRP (SMD = −0.55, 95% CI −0.89 to −0.21, p = 0.004) (Fig. 5b), accompanied by a moderate GRADE rating (Table 1) and significant heterogeneity (τ² = 0.37, I² = 85%, p < 0.05).

Meta-analysis for Leukocyte-Poor and Leukocyte-Rich PRP

For the seven comparisons (684 patients) with follow-up duration of at least 3 months and comparing leukocyte-poor PRP with active drug treatments, a statistically significant, moderate reduction favored leukocyte-poor PRP (SMD = −0.61, 95% CI −0.97 to −0.25, p = 0.006) (Fig. 6a). It had a low level of GRADE rating (Table 1) and statistically significant heterogeneity (τ² = 0.09, I² = 66%, p = 0.006).

Fig. 6.

The analgesic efficacy of PRP on chronic noncancer pain for a leukocyte-poor concentration versus active drug treatments and b leukocyte-rich concentration versus active drug treatments. CI confidence interval, PRP platelet-rich plasma, SD standard deviation

In the five comparisons (334 patients) examining the analgesic effect of leukocyte-rich PRP versus active drug treatments, no statistically significant difference was observed (SMD = −0.09, 95% CI −1.36 to 1.18, p > 0.05) (Fig. 6b). The level of GRADE rating was very low (Table 1). Significant heterogeneity was observed (τ² = 0.91, I² = 93%, p < 0.001).

Effects on Functional Scores

We also evaluated AOFAS for plantar fasciitis, ASES and WORC for rotator cuff tendinopathy, and WOMAC for osteoarthritic knee pain as secondary functional outcomes. Five comparisons (339 patients) reported AOFAS for plantar fasciitis, and the result had a statistically significant increase with a small effect size for those given PRP (SMD = 0.45, 95% CI 0.10–0.79, p = 0.02) (Supplementary Fig. S2a), with no significant heterogeneity detected (τ² = 0, I² = 0%, p > 0.05). There were five comparisons (412 patients) for ASES and four comparisons (277 patients) for WORC for rotator cuff tendinopathy. However, the difference in pooled results did not reach the level of statistical significance for ASES (SMD = 0.83, 95% CI −0.63 to 2.28, p > 0.05) or WORC scores (SMD = −0.25, 95% CI −1.09 to 0.60, p > 0.05) (Supplementary Fig. S2b, c). Twelve comparisons (939 patients) investigated WOMAC for osteoarthritic knee pain. The pooled result showed a statistically significant, large reduction in WOMAC scores favoring PRP (SMD = −2.11, 95% CI −3.73 to −0.49, p = 0.02). There was significant heterogeneity (τ² = 6.10, I² = 99%, p < 0.001) (Supplementary Fig. S2d).

Evaluation of Potential Publication Bias

We evaluated the publication bias by plotting the funnel plot and calculating the p-value from Egger’s regression statistical test for all 56 RCTs included in the study. There was no apparent publication bias observed, which was confirmed by visual inspection of the funnel plot (Fig. 7). Similarly, the Egger’s regression test did not show any statistically significant publication bias (p = 0.93).

Fig. 7.

Funnel plot for evaluating publication bias

Discussion

To date, no systematic review and meta-analysis has been conducted to evaluate the analgesic effectiveness of PRP compared with placebo or active drug treatments across all chronic noncancer pain conditions. Our meta-analysis showed that PRP was associated with statistically significant and small reduction in pain intensity compared with active drug treatments and placebo. No significant differences were observed for studies with a follow-up period of shorter than 3 months. On the other hand, when only RCTs of at least 3 months follow-up duration were analyzed, PRP resulted in moderate pain reduction compared with active drug treatments. PRP showed particular effectiveness in reducing pain for specific conditions, such as osteoarthritic knee pain and rotator cuff tendinopathy or tears, but there was no significant difference for plantar fasciitis. Additionally, PRP injections provided larger long-term pain relief compared with corticosteroid and hyaluronic acid injections. Among PRP formulations, leukocyte-poor PRP was associated with a moderate reduction in pain compared with active drug treatments, whereas leukocyte-rich PRP did not show similar benefits.

PRP is an increasingly used treatment modality for managing chronic noncancer pain, and the number of clinical studies has increased rapidly. Prior systematic reviews and meta-analyses have studied PRP in specific pain conditions. Narrative reviews have evaluated the effects of PRP across different chronic pain conditions, and clinical recommendations have been suggested on the basis of qualitative review of available evidence [4, 76, 77]. Therefore, the analgesic effect and role for PRP in chronic noncancer pain remained unclear. In this research, we carried out an extensive systematic review and meta-analysis that quantitatively evaluated the analgesic effect of PRP across different chronic noncancer pain conditions. As the aim of chronic pain management is to provide sustained pain relief, we further evaluated clinical studies that had a follow-up duration of 3 months or longer. This is because the clinical value of treatment intervention would probably be limited if the pain-relieving duration was less than 3 months. Our analysis revealed no significant differences between PRP and placebo/active drug treatments for follow-up periods shorter than 3 months. However, we observed a statistically significant moderate reduction in pain for follow-up durations of at least 3 months. This study provides a large dataset from which to draw conclusions on the longer-term analgesic effectiveness of PRP in treating chronic noncancer pain. As far as we are aware, this study is the first to provide quantitative evidence demonstrating an overall positive analgesic impact in treating chronic noncancer pain.

The analgesic effect of PRP in comparison with other active drug treatments is unclear. This is particularly important for corticosteroid, which has traditionally been the standard treatment drug for injections in various chronic pain conditions. Unlike corticosteroid, which alleviates pain through suppression of inflammation, PRP has the additional effect of increasing tissue healing and may provide longer-lasting pain relief [78]. The application of steroids may also be contraindicated in patients with osteoporosis, pregnancy, and infection [77]. PRP may be a safer therapeutic option and less likely to be limited by contraindications. Two meta-analyses comparing PRP versus corticosteroid for plantar fasciitis and lateral epicondylitis found that PRP provided greater pain relief at a longer follow-up (6–12 months), but provided no advantage in the short term [79, 80]. PRP has also been compared with hyaluronic acid. A meta-analysis that studied osteoarthritic knee pain showed that PRP resulted in reduced pain intensity compared with corticosteroids and hyaluronic acid at 3-, 6- and 12-month follow-up [81]. Our results suggests that PRP may be the preferred treatment option for chronic noncancer pain management, especially when compared with corticosteroids and hyaluronic acid.

We evaluated the analgesic efficacy of PRP by comparing with active drug treatments in specific chronic pain conditions where duration of follow-up was at least 12 weeks. Osteoarthritis of the knee was the most commonly studied pain condition. Our meta-analysis showed positive analgesic effect with PRP for treating patients suffering from knee osteoarthritis. In addition to pain score reduction, subgroup meta-analyses also showed better WOMAC scores in patients treated with PRP. This indicates that PRP not only enhanced pain intensity but also improved functional outcomes for patients experiencing knee pain due to osteoarthritis. This is in agreement with findings from other clinical studies [13]. PRP was also associated with moderate pain decrease for treating chronic pain from rotator cuff tendinopathy/tear, but there was no statistical difference in functional scores (ASES and WORC). One meta-analysis that compared PRP versus normal saline or dry needling/rehabilitation program found no difference at 1 and 3 months, but lower pain scores with PRP at 12 months after injection [82]. Another meta-analysis found a positive analgesic effect with PRP injection at over 24 weeks follow-up for patients with rotator cuff tendinopathy [83]. When comparing PRP injections with active drug treatments for pain management in plantar fasciitis, our meta-analyses did not reveal a statistically significant difference. In another meta-analysis, PRP reduced pain more than corticosteroid at 3 months after injection, but no significant differences were observed at 6 months [84]. These results suggest that PRP is not superior to active drug treatments in providing long-term pain reduction for plantar fasciitis. It is not unexpected that the analgesic effect of PRP appeared to vary between different pain conditions. However, it is noteworthy that PRP was not inferior to active drug treatments for any chronic pain conditions.

PRP is usually categorized into leukocyte-poor PRP and leukocyte-rich PRP on the basis of the neutrophil level [85]. The analgesic effect of leukocyte-poor in comparison with leukocyte-rich PRP is unclear, as there is a lack of head-to-head studies comparing them. In our study, we separately assessed the pain-relieving effects of leukocyte-poor and leukocyte-rich PRP versus active drug treatments in clinical studies with at least 3 months of follow-up. Only seven comparisons were available for leukocyte-poor PRP, and five for leukocyte-rich PRP. We observed a moderate effect size for pain reduction with leukocyte-poor PRP, while there was no statistically significant difference for leukocyte-rich PRP. A meta-analysis for osteoarthritis knee pain found improved WOMAC scores with leukocyte-poor PRP when compared with hyaluronic acid or placebo, but there was no difference with leukocyte-rich PRP [86]. A possible biological explanation could be the difference in levels of pro-inflammatory and anti-inflammatory mediators. Leukocyte-rich PRP is associated with significantly upregulated pro-inflammatory cytokines, including tumor necrosis factor (TNF)-α, interleukin (IL)-6, and IL-1β, while leukocyte-poor PRP raises the expression levels of anti-inflammatory cytokines IL-10 and IL-4 [87]. While pro-inflammatory effect is required for healing, it can also exacerbate pain [85], thereby reducing analgesic efficacy. Nevertheless, the amount of clinical evidence available is limited and not sufficient to draw definitive conclusions.

Limitations

There are some limitations to our meta-analysis. First, substantial heterogeneity and inconsistency existed among a number of analyses. This is because different clinical trials had different chronic pain conditions and methodological heterogeneity as well as differences in PRP preparation and study duration [2, 88]. These factors could impact the validity of the findings, and the results should therefore be interpreted with this in mind. There were not enough clinical studies available to conduct meta-analyses for other conditions of chronic pain such as Achilles tendinopathy and osteoarthritis of the hip. The criteria to only include RCTs having a follow-up duration of no less than 3 months and the requirement for at least three clinical comparisons may have limited the amount of evidence available for quantitative analysis. However, this was necessary to provide clinically important information since the purpose of PRP injection was to achieve long-lasting pain reduction. Another potential limitation was that the analgesic effect versus oral analgesics was not evaluated. The key clinical question would be the effect of PRP versus other injectates, because this would provide information to guide the choice of injectate used for interventional pain procedures. Moreover, side effects of PRP were not studied, because the incidence was generally very low and uncommonly reported. Another limitation is that this study focuses solely on comparing PRP with pharmacological interventions for chronic noncancer pain, and did not take into account active lifestyle treatments, such as exercise and weight management, which have been shown to reduce pain and improve function [33, 89]. There was variability in the quality of blinding across the included studies, which may impact the reliability of the outcomes. The majority of the studies were triple- or double-blinded. However, 13 studies were only single-blinded. Furthermore, there was a lack of clear blinding information in 12 studies, which may raise questions about the methodological rigor employed. Finally, we only included peer-reviewed papers to ensure the credibility, accuracy, and quality of the information being used in our research. However, by not incorporating nonpublished clinical trials, we may have unintentionally introduced potential publication bias.

Conclusions

PRP resulted in a small reduction in pain intensity for chronic noncancer pain when compared with active drug treatments and placebo. Subgroup analyses highlighted significant pain relief in specific conditions, such as osteoarthritic knee pain and rotator cuff tendinopathy or tears. PRP appears to be a promising treatment option, particularly when compared with corticosteroids and hyaluronic acid, as it may provide more durable and sustained pain relief. Overall, the evidence supports the effectiveness of PRP in managing chronic pain over a reasonably long duration, making it a valuable therapeutic choice for certain conditions in clinical practice.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Fengfeng Wang and Fei Meng were responsible for the literature screening and data extraction. Fengfeng Wang was responsible for statistical analysis and writing up the article. Stanley Sau Ching Wong was responsible for planning and guidance on this paper. Fengfeng Wang, Fei Meng, Timmy Chi Wing Chan, and Stanley Sau Ching Wong contributed to the article and approved the submitted version. Fengfeng Wang, Fei Meng, Timmy Chi Wing Chan, and Stanley Sau Ching Wong have provided their consent for the publication of this meta-analysis. Fengfeng Wang, Fei Meng, Timmy Chi Wing Chan, and Stanley Sau Ching Wong agree with the content, and there are no conflicts regarding the submission and dissemination of the findings.

Funding

This study, including the Rapid Service Fee, was supported by the Department of Anaesthesiology, School of Clinical Medicine, Li Ka Shing Faculty of Medicine, The University of Hong Kong.

Data Availability

The data supporting this study are available from the corresponding author upon reasonable request.

Declarations

Conflict of Interest

The authors (Fengfeng Wang, Fei Meng, Timmy Chi Wing Chan, and Stanley Sau Ching Wong) declare that there are no conflicts of interests.

Ethical Approval

This article was based on previously conducted research and did not involve any new research by the authors that included human participants or animals.