Introduction
Knee osteoarthritis (KOA) is one of the most common degenerative joint diseases with continuous pain and loss of function [
1] and characterized by progressive loss of articular cartilage, inflammation of synovial membrane, and changes in the bones under the cartilage [
2‐
5]. It was reported that among older adults, the risk of lower limb disability caused by KOA is at least 40% [
6], and KOA is considered as one of top ten causes of disability [
7]. To date, however, there are no treatment methods that can reverse or alter the progression of KOA. Although total knee arthroplasty (TKA) is regarded as the last choice if osteoarthritis progresses to end-stage [
8], there is a significant risk of complications including revision, infection, and unsatisfied function [
8‐
10]. Therefore, in terms of the younger and middle-aged patients of KOA, non-surgical interventions attract more and more attention, including physical therapy, oral nonsteroidal anti-inflammatory drugs (NSAIDs), hyaluronic acid (HA), ozone, and corticosteroids injection [
8,
11].
In the last 10 years, growth factors aroused people’s interest for its properties of repair tissue lesion and maintain normal tissue structure, especially platelet-rich plasma (PRP) injection [
12‐
15]. PRP contains a high concentration of platelets, which are obtained by centrifugation of autologous blood [
16]. Various growth factors and cytokines are released after the degranulation of platelets and to accelerate cartilage matrix synthesis, restrain synovial membrane inflammation, and promote cartilage healing [
17,
18]. Owing to the properties of regenerative effect and anti-inflammatory potential, PRP is widely used in musculoskeletal diseases, such as rotator cuff tear, lateral epicondylitis, patellar tendinopathy, osteoarthritis [
19‐
26].
Lots of articles [
27‐
36] compared the clinical outcomes of intra-articular PRP injection with other conservative treatment methods (including oral NSAIDs, HA, and corticosteroids injection), and there are different results among these comparisons. The American Academy of Orthopaedic Surgeons Clinical Guidelines suggested that HA injection is not recommended for the treatment of KOA, while PRP injection is “not recommend for or against” [
8]. The OA Research Society International (OARSI) Guidelines [
37] provide an “uncertain” recommendation for HA injection in the treatment of KOA, while do not mention the PRP injection. Meanwhile, Campbell [
34] conducted a systematic review of overlapping meta-analyses, suggested that PRP injection may increase the local adverse reactions than HA. However, several meta-analyses [
30‐
33,
35] published in the last 3 years indicated PRP injection does not have more adverse events than HA injection.
Therefore, the purpose of this study was (1) to perform a summary of meta-analyses comparing PRP injection with HA injection for KOA patients, (2) to determine which meta-analysis provides the best available evidence to making proposals for the use of PRP in the treatment of KOA patients, and (3) to highlight gaps in the literature that require future investigation. We hypothesized that PRP injection is more effective in the treatment of KOA patients and with a similar risk of adverse events than HA and placebo.
Methods
Literature search
The PubMed, EMBASE, and Cochrane database was searched to perform a summary of meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement [
38]. It has been registered in PROSPERO. The lasted literature search was conducted on November 12, 2018. The following search terms were used: [platelet-rich plasma OR PRP] AND [knee arthritis OR arthritis OR knee osteoarthritis OR osteoarthritis] AND [meta-analysis OR systematic review]. At the same time, the citations of the included meta-analyses were evaluated to see if there were any suitable literatures for inclusion. When necessary, the corresponding author of the study was contacted for further information.
Inclusion criteria
The inclusion criteria are as follows:
1.Compared the outcomes of intra-articular platelet-rich plasma (IA-PRP) injection with intra-articular hyaluronic acid (IA-HA) or placebo injection
2.The meta-analysis of randomized controlled trials
5.Published after the year 2000
Exclusion criteria
The exclusion criteria are as follows:
1.Did not compare the outcomes of intra-articular platelet-rich plasma (IA-PRP) injection with intra-articular hyaluronic acid (IA-HA) or placebo injection
2.Meta-analysis included non-randomized controlled trials
3.Cadaveric, animal, or biomechanical research
5.Published before the year 2000
6.Network meta-analysis or overlapping meta-analysis
Quality appraisal
Each included study was evaluated with the Quality of Reporting of Meta-analyses (QUOROM) system [
39]. It is divided into 6 headings, and 18 items totally, including searching, validity assessment, data abstraction, trial flow, study characteristics, and so on. The Oxman-Guyatt quality appraisal tool [
40] was also used to assess the quality of meta-analysis. Two trained reviewers assessed the included meta-analysis respectively. And the final decision was made by a third author after which reviewed the article if they have different opinions. Moreover, the bias was noted while it was reported by individual trials in the literature. In addition, three authors used the Jadad decision algorithm [
41] to guide interpretation of discordant reviews respectively, and the results determined which of the included systematic reviews provided the highest quality current evidence to make recommendations for knee osteoarthritis.
Data extraction and statistical analysis
Data were extracted by two trained authors from the included articles. It included the following data: author, the year of publication, level of evidence included in the studies, the searched databases, eligibility criteria, no. of included articles, no. of patients, basic patient information, time of follow-up, adverse events, patient satisfaction. And the following standardized outcome scores were collected: visual analog scale (VAS) pain score, Western Ontario and McMaster Universities Osteoarthritis Index pain (WOMAC) score, International Knee Documentation Committee (IKDC) score, and Lequesne index.
In addition, we also recorded the following characteristics of each included systematic review: the rationale for repeating the systematic review, the number of previous systematic reviews actually cited compared with maximum number that could possibly have been cited in each study, the search methodology of each included study, the demographic data and characteristics of the review, the heterogeneity and subgroup analyses of primary studies in the systematic review, and the conclusions of the review regarding whether IA-HA was more clinically effective in terms of pain relief, functional scores, and side effects.
Discussion
This summary of meta-analyses suggested that based on the best currently available evidence, intra-articular platelet-rich plasma (PRP) injection provides more superior pain relief, efficacious function improvement, and similar risk of adverse events when compared with HA injection and placebo in the treatment of KOA patients. However, we were not able to confirm the effect of other aspects of PRP on the treatment of KOA patients, including numbers of PRP injections (1 or ≥ 2), injection intervals (weekly or monthly), PRP spinning techniques (single or double), mean platelet concentration (> or < 5 × baseline), PRP category (LP-PRP or LR-PRP), and use an activator or not.
In the past few years, more and more researchers noticed the potential of PRP in the treatment of musculoskeletal diseases, such as rotator cuff tear, lateral epicondylitis, patellar tendinopathy, osteoarthritis, and Achilles tendon repair [
19‐
26]. Given the properties of regenerative effect and anti-inflammatory potential in PRP, a number of researches [
19,
23,
24,
27‐
35] explored the curative effect of intra-articular PRP injection in the treatment of patients with osteoarthritis, especially with knee osteoarthritis. However, in the current clinical guidelines of orthopedic surgeons, the use of PRP injection for KOA patients is uncertain [
8,
37]. Few guidelines recommend PRP injection to treat KOA. This may be related to the different results reported in current high-quality evidence-based medical articles. Of the meta-analyses published in the last 3 years, only one article [
31] considered PRP to have similar efficacy to HA, and other articles [
30,
32,
33] suggested that PRP injection is more effective than HA in KOA patients.
Three meta-analyses [
30,
32,
33] evaluated adverse events after PRP injection and HA or placebo in the treatment of KOA patients, including pain, stiffness, syncope, dizziness, headache, nausea, or infection. Shen et al. [
33] indicated no severe complications were recorded and all adverse events were self-resolved in days. All of these articles suggested no statistical difference in adverse events between PRP injection and HA or placebo. Only one meta-analysis [
30] compared the pooled effect sizes of primary outcomes with the minimum clinically important differences (MCID), which determinate whether significant outcomes have clinically meaningful implications [
55,
56]. It demonstrated that compared with HA, PRP injection has better pain relief and function improvement in the 12 months follow-up (the CI of WOMAC pain and function scores was greater than the MCID), with no statistical difference in 6 months follow-up.
Riboh et al. [
57] performed a network meta-analysis, which compared the clinical outcomes and adverse events between
LP-PRP,
LR-PRP, HA, and placebo. It included 6 RCTs (Level I) and 3 prospective comparative studies (Level II) and illustrated the effect of different leukocyte concentrations on PRP injection. This article suggested that
LP-PRP has better functional outcome scores compared with HA and placebo in the treatment of KOA, with no difference between
LR-PRP and HA. It also found no significant difference between PRP, HA, and placebo in adverse events and indicated leukocyte concentration may not directly relate to adverse events in PRP injection. In addition, unfortunately, we rarely found other scholars that compare the effects of different preparation methods, concentrations, and frequency of injection on the efficacy of PRP in the treatment of KOA. This is perhaps the focus of our future research.
The strengths of this summary of meta-analyses are based on the best currently available evidence to evaluate the clinical outcomes of PRP injection in the treatment of KOA patients. Three authors used these different appraisal tools [
39‐
41] to assess the quality of each included meta-analysis, and each meta-analysis was Level I evidence.
Limitations
There are also several limitations in this study. First, all included meta-analyses only evaluate the clinical outcomes of PRP injection at 6 months and 12 months follow-up, none of them was a median or long follow-up. Second, only one paper [
30] performed a subgroup analyses of the different details of PRP injection, such as times of PRP injection, PRP spinning techniques, mean platelet concentration, PRP category, activation or not, and risk of bias. Therefore, it is not clear about the effect of the different details in PRP injection. Third, none of the included meta-analysis conducted a subgroup analysis about the OA grade (including Kellgren-Lawrence grade and Ahlback grade), so we do not understand which grade of OA can get more benefits from intra-articular PRP injection. Otherwise, heterogeneity is inevitable among the patients included in these meta-analyses, such as the age of patient, duration of knee pain before injection, sex, BMI, and so on. Finally, although a total of 1677 patients were included in this summary of meta-analysis, the included meta-analysis had included several primary articles which only contained a smaller sample size, and it may be a potential source of bias.
Therefore, more rigorous randomized controlled trials, which focus on a very specific question, such as which PRP spinning techniques, or which mean platelet concentration of PRP, or which frequency of injection of PRP can provide better clinical outcomes or which grade of OA can get more benefits from intra-articular PRP injection, are also needed to perform. Meanwhile, the articles with med-long-term follow-up are also needed to conduct and assess the curative effect of PRP injection.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.