Abstract
Background
In recent years, mesenchymal stem cells (MSCs) have been widely applied in the clinical treatment of knee osteoarthritis (KOA), demonstrating promising therapeutic efficacy. However, intervention protocols for MSCs have not yet been standardized, and evidence regarding the impact of different injection frequencies on the efficacy and safety of MSC treatment in KOA remains limited.
Objective
This study aims to integrate evidence from conventional and network meta-analyses to evaluate the efficacy and safety of different intervention frequencies (single vs. repeated MSC injections) in the treatment of knee osteoarthritis (KOA).
Methods
A systematic search was conducted in PubMed, Embase, the Cochrane Library, and Web of Science databases up to March 1, 2025. Traditional meta-analysis was performed to assess the efficacy and safety of MSC therapy for KOA, followed by a network meta-analysis (NMA) to evaluate the effectiveness of single versus repeated MSC injections. Outcome measures included WOMAC and VAS scores at 3, 6, and 12 months, as well as the incidence of adverse events (AEs). Traditional meta-analysis and NMA were conducted using Review Manager 5.3 and Stata SE16.0, respectively.
Results
A total of 16 RCTs involving 622 patients were included. Traditional meta-analysis showed that MSC therapy significantly improved pain and knee joint function in KOA patients at different time points (3, 6, and 12 months). The results of the network meta-analysis (NMA) indicated that, compared to single MSC injections, repeated MSC injections provided greater improvements in pain and functional scores at 6 and 12 months, demonstrating superior efficacy. However, repeated MSC injections were also associated with a higher incidence of adverse events.
Conclusions
Both single and repeated MSC injections could improve pain and knee joint function in patients with KOA. Compared to single injections, repeated MSC injections may offer superior therapeutic benefits; however, they are associated with a higher incidence of adverse events. In clinical practice, potential side effects of repeated MSC administration must be carefully considered. Future research should focus on large-scale, multicenter, and long-term randomized controlled trials to further validate the efficacy and safety of MSC therapy for KOA.
Supplementary Information
The online version contains supplementary material available at 10.1186/s13287-025-04638-2.
Keywords: Osteoarthritis, Mesenchymal stem cells, Repeated injections, Network meta-analysis, Efficacy
Introduction
Osteoarthritis (OA) is the most prevalent chronic joint disease worldwide, with 595 million symptomatic and radiographically diagnosed cases reported globally in 2020 [1]. OA has become a major cause of disability among adults over the age of 70, and it is projected that by 2025, the number of individuals with OA-related disability will reach 11.016 million [2]. The knee joint, being the most commonly affected site in OA, exhibits significant age- and sex-related differences in prevalence [3]. Epidemiological data indicate that the prevalence of symptomatic knee OA (KOA) in individuals aged 60 and above is 10% in men and 13% in women, with over 250 million people affected worldwide [4, 5]. Characterized by pain and restricted joint mobility, KOA directly leads to diminished quality of life and impaired physical function [6]. Recent studies suggest a trend toward earlier onset of KOA, potentially associated with changes in lifestyle and the increasing prevalence of metabolic syndrome [7]. Additionally, OA-related healthcare expenditures impose a substantial economic burden, with annual costs in the United States reaching as high as $139.8 billion [8]. As the global population continues to age and obesity rates rise, the disease burden of KOA is expected to intensify [9, 10].
Current clinical interventions for KOA primarily rely on non-surgical and surgical approaches. Non-surgical treatments include lifestyle modifications (such as exercise and weight management), pharmacological pain relief (e.g., nonsteroidal anti-inflammatory drugs), and intra-articular injections (including corticosteroids and hyaluronic acid), aiming to alleviate pain and improve joint function [11, 12]. However, these measures offer only short-term symptomatic relief and fail to reverse the degenerative changes in cartilage, allowing disease progression to continue [13, 14]. For patients in the end-stage of KOA, joint replacement surgery may serve as a standard operation, yet it is limited by factors such as prosthesis longevity, the risk of postoperative complications, patient age, and surgical tolerance [15]. Existing treatment strategies fall short in terms of long-term efficacy and disease control, highlighting the urgent need for innovative therapies that can both manage symptoms and delay disease progression.
Mesenchymal stem cells (MSCs) have emerged as a focal point in regenerative medicine research for KOA due to their unique biological properties [16]. Studies have shown that MSCs modulate the joint microenvironment through multiple mechanisms, including immunoregulation, anti-inflammatory effects, and promotion of regeneration, thereby inhibiting degenerative progression and facilitating cartilage repair [6–9, 14, 17, 18]. Their secretory activity and trilineage differentiation potential (osteogenic, chondrogenic, and adipogenic) further support the capacity to repair damaged knee joint tissues [19, 20]. Compared to knee arthroplasty, intra-articular injection of MSCs offers a minimally invasive alternative [21], with multiple preclinical and clinical trials confirming its safety and effectiveness in symptom relief [22, 23].
However, there is still controversy regarding the standardized treatment protocols for MSCs, including factors such as the source of MSCs and the intervention dosage. Previous meta-analyses have primarily focused on exploring the effects of different sources and intervention doses on the efficacy for KOA patients [24, 25]. Notably, in an animal study, Mahmoud et al. [26] found that multiple MSC injections could effectively improve knee joint lesions induced by osteoarthritis and reduce inflammation, which has sparked widespread attention on the efficacy and safety of single versus multiple MSC injections. Most past studies have involved a single intra-articular MSC injection, assuming that multiple injections may offer potential benefits. Recently, Matas et al. [27] were the first to demonstrate, through a triple-blind randomized controlled trial (RCT), that repeated MSC injections show good safety over a 1-year follow-up period compared to the control group. Therefore, this study first compares the efficacy and safety of different intervention protocols in treating KOA patients through traditional meta-analysis. Given the limited number of direct comparative studies between single and repeated MSC injections, we also use network meta-analysis (NMA) to integrate both direct and indirect evidence, exploring the differences in efficacy and safety at different follow-up time points (3, 6, and 12 months) between single versus repeated injections. This meta-analysis aims to identify the intervention protocol most likely to benefit patients, balancing efficacy and safety risks, and provide evidence-based recommendations for the optimal MSC intervention strategy for KOA.
Materials and methods
This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for conducting a network meta-analysis [28]. The study protocol was prospectively registered in the PROSPERO database (CRD420251025791).
Search strategies
We searched all articles in online databases, including Embase, PubMed, Web of Science, and the Cochrane Library, with a search cutoff date of March 2025. We constructed the search strategy using MeSH terms and free words, mainly focusing on “mesenchymal stem cells” and “knee osteoarthritis”. Detailed search strategies based on Boolean operators can be found in Supplementary Table 1. Two authors independently performed the search process, and any discrepancies in search results were discussed and resolved with a third author. To avoid missing relevant articles, we also conducted a reference search for all included studies.
Eligibility criteria
Population (P): Adults (≥ 18 years old) with radiographically confirmed knee osteoarthritis. Studies involving inflammatory arthritis, post-traumatic cases without degenerative changes, or studies with significant comorbidities were excluded.
Interventions (I) and Comparators (C): Intra-articular injection of mesenchymal stem cells (MSCs), administered as a single dose or multiple doses (≥ 2 injections). The control group represented “conventional therapy”, including hyaluronic acid (HA), corticosteroids, saline, placebo, and acetaminophen. Trials combining MSC therapy with surgery or adjunctive biologics were excluded.
Outcomes (O): The primary outcomes were changes in the Western Ontario and McMaster University Osteoarthritis Index (WOMAC) total score and visual analogue scale (VAS) score at 3-, 6- and 12-months post-intervention. Secondary outcomes included the Knee Injury and Osteoarthritis Outcome Score (KOOS), total adverse events (AE), and serious adverse events (SAE). Studies lacking quantifiable outcome data were excluded.
Study Design (S): Only full-text, peer-reviewed RCTs or controlled clinical trials were considered. Conference abstracts, reviews, meta-analyses, animal studies, and non-controlled studies were excluded.
Study selection
All retrieved records were imported into Endnote 20 (Clarivate Analytics) for automatic removal of duplicates. The remaining records underwent a two-stage screening process: (1) First stage: Two independent authors screened titles and abstracts based on predefined eligibility criteria to exclude clearly irrelevant studies. (2) Full texts of potentially eligible studies were reviewed for further assessment according to the same criteria. Any discrepancies during this stage were resolved through discussion with a third author. The study selection process was illustrated using a PRISMA flow diagram, detailing the number of exclusions and reasons at each stage.
Data extraction
A standard Excel template was created for data extraction, and the entire extraction process was conducted independently by two authors. The following variables were extracted: (1) Study characteristics: First author, publication year, country, study type, sample size per group, patient age, height, weight, BMI, randomization method, and follow-up duration. (2) MSC intervention details: MSC source, injection frequency, dose, and injection site. (3) Baseline data, WOMAC total score and VAS score at 3, 6 and 12 months, KOOS score, and safety outcomes (AE and SAE). For graphical data, the mean ± standard deviation (mean ± SD) was extracted using Origin 2021. Any discrepancies in data extraction were resolved through joint review and discussion with a third author. In cases of unclear outcome measurement data, attempts were made to contact the corresponding author.
Risk of bias assessment
The methodological quality of the included RCTs was assessed using the Cochrane risk of bias tool. Two independent researchers evaluated five domains: selection bias, performance bias, detection bias, attrition bias, and reporting bias. Each domain was rated as “low risk of bias”, “unclear risk of bias”, or “high risk of bias”. In cases where the evaluation results were not consistent, a third researcher participated in discussions to reach a consensus.
Assessment of evidence certainty
The Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework was used to assess the quality of evidence for the NMA results [29–31]. The evidence quality levels are classified into four categories: high, moderate, low, and very low. For direct comparisons (which in this study are primarily based on RCTs, with an initial quality level of high), potential downgrading was considered based on factors such as risk of bias, inconsistency, indirectness, imprecision, and publication bias. For indirect comparisons, the starting quality is determined by the lower quality of direct evidence from the primary contributing first-order loops, and further assessment was made to determine whether intransitivity exists, which may lead to downgrading. The quality of the evidence for the NMA results is then determined by combining both direct and indirect evidence, as well as their relative contribution (information size) to the network estimate. Furthermore, when determining the final grade of the NMA results, careful consideration was given to local inconsistencies within the network and the precision of the combined effect size in the NMA [32].
Statistical analysis
A conventional meta-analysis was performed using Review Manager 5.3 to calculate effect sizes from direct evidence. For categorical data, including AE and SAE, the odds ratio (OR) with a 95% confidence interval (CI) was used. For continuous data, including WOMAC, VAS, and KOOS scores, the standardized mean difference (SMD) and its corresponding 95% CI were calculated. Since the conventional therapy group included various different control groups, the outcomes prior to injection were used as the control group for traditional meta-analysis. Statistical heterogeneity was assessed using the I² statistic, with I² >50% indicating substantial heterogeneity. The choice between fixed-effects and random-effects models was based on clinical and methodological considerations, including the anticipated heterogeneity in study designs, populations, and MSC intervention protocols. A random-effects model was prioritized given the expected clinical variability across studies. To enhance the reliability of the analysis, only studies with ≥ 3 direct comparison pieces of evidence were included in the pooled analysis. To assess the potential impact of publication bias on the results, subgroup analyses were conducted to investigate whether studies with differing methodological quality yielded inconsistent findings. The trim-and-fill method [33] was used to assess the impact of publication bias on the results and the stability of the findings. A P-value < 0.05 was considered statistically significant.
NMA was conducted using the “mvmeta” and “network” packages in Stata/SE 16.0. The evidence network was visualized using a network plot, where nodes represented interventions, with node size reflecting sample size, and the thickness (or numbers) of connecting lines indicating the number of direct comparisons. A global inconsistency test was performed, and a consistency model was adopted when the P-value of the Z-test was > 0.05 [34]. Additionally, the surface under the cumulative ranking (SUCRA) was used to rank the interventions based on outcome measures, providing a direct interpretation of the probability that each intervention was the most effective [35]. Publication bias for each outcome was assessed using funnel plots. Sensitivity analysis was used to assess the robustness of the results.
Results
Search results
A total of 527 potential records were initially retrieved from the databases. After automatically removing 182 duplicates using Endnote 20,345 records remained. Subsequently, 304 studies were excluded based on title and abstract screening by two authors. The full texts of the remaining 41 articles were reviewed, and 16 studies [27, 36–50] met the eligibility criteria and were included in both the conventional and network meta-analyses. The detailed study selection process is illustrated in Fig. 1.
Fig. 1.
PRISMA flow diagram of the study selection process
Study characteristics
The 16 studies included were all RCTs published in English, with a time span of within the past 10 years. The studies covered multiple regions globally, including Europe (Spain, Portugal, Latvia), Asia (China, South Korea, India, Iran), the Americas (United States, Mexico, Chile), and Oceania (Australia), reflecting broad regional participation.
A total of 622 patients were involved in the clinical trials. The experimental group included 323 patients, with 261 receiving a single injection of MSCs, and the remaining 62 patients receiving repeated injections of MSCs. Sixteen studies (23 datasets) involved 547 patients. Fourteen studies (18 datasets) described single MSC injections, while four studies (5 datasets) investigated repeated MSC injections. Notably, two studies included both single and repeated MSC injections. Some studies compared the efficacy of multiple MSC dosage cohorts with conventional treatments, and each cohort was labeled as (1), (2), or (3). Furthermore, the sources of MSCs included BMSC (N = 9), AD-MSC (N = 6), and UC-MSC (N = 1). The injection doses ranged from 1 × 107 to 1 × 108 cells. The control group treatments included hyaluronic acid (N = 9), saline (N = 3), placebo (N = 1), acetaminophen (N = 1), and corticosteroids (N = 1). The main characteristics of the included studies are shown in Tables 1 and 2.
Table 1.
Main characteristics of all articles included in the meta-analysis
| Author | Year | Country | Study design | Age (MSC vs. Traditional treatment, Years ± SD, range) | Number of female/male (MSC vs. Traditional treatment) | Height (cm) | Weight (kg) | BMI (kg/m2) (MSC vs. Traditional treatment) | Kellgren-lawrence grade (MSC vs. Traditional treatment) | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| I | II | III | IV | |||||||||
| Bastos et al. [36] | 2020 | Portugal | RCT | 55.7 ± 7.8 vs. 55.9 ± 13.4 | 6/10 vs. 8/9 | NA | NA | 30.6 ± 4.5 vs. 31.0 ± 4.7 | 1 vs. 1 | 7 VS 4 | 5 VS 7 | 3 VS 5 |
| Chen et al. (1) [37] | 2021 | China | RCT | 67.7 ± 6.84 vs. 70.5 ± 8.37 | 14/3 vs. 5/3 | NA | 66.0 ± 6.80 vs. 60.5 ± 10.80 | 27.65 ± 3.03 vs. 25.47 ± 3.49 | 0 vs. 0 | 10 vs. 5 | 7 vs. 3 | 0 vs. 0 |
| Chen et al. (2) [37] | 2021 | China | RCT | 68.6 ± 6.45 vs. 70.5 ± 8.37 | 15/2 vs. 5/3 | NA | 64.2 ± 10.97 vs. 60.5 ± 10.80 | 26.72 ± 4.19 vs. 25.47 ± 3.49 | 0 vs. 0 | 10 vs. 5 | 7 vs. 3 | 0 vs. 0 |
| Chen et al. (3) [37] | 2021 | China | RCT | 64.9 ± 4.91 vs. 70.5 ± 8.37 | 12/3 vs. 5/3 | NA | 62.9 ± 10.47 vs. 60.5 ± 10.80 | 25.66 ± 3.78 vs. 25.47 ± 3.49 | 0 vs. 0 | 12 vs. 5 | 3 vs. 3 | 0 vs. 0 |
| Emadedin et al. [38] | 2018 | Iran | RCT | 51.7 ± 9.2 vs. 54.7 ± 5.3 | 7/12 vs. 9/15 | NA | NA | 30.2 ± 4.4 vs. 31.5 ± 5.4 | NA | 2 vs. 1 | 13 vs. 20 | 4 vs. 3 |
| Freitag et al. (1) [39] | 2019 | Australia | RCT | 54.6 ± 6.3 vs. 51.5 ± 6.1 | 5/5 vs. 3/7 | 1.75 ± 0.1 vs. 1.73 ± 0.1 | 97.1 ± 20.2 vs. 76 ± 14.5 | 31.6 ± 5.9 vs. 25.2 ± 3.4 | NA | NA | NA | NA |
| Freitag et al. (2) [39] | 2019 | Australia | RCT | 54.7 vs. 51.5 ± 6.1 | 6/4 vs. 3/7 | 1.7 ± 0.1 vs. 1.73 ± 0.1 | 88.7 ± 23.5 vs. 76 ± 14.5 | 30.4 ± 5.6 vs. 25.2 ± 3.4 | NA | NA | NA | NA |
| Garay-Mendoza et al. [40] | 2017 | Mexico | RCT | 55.67 ± 12.02 vs. 59.32 ± 10.85 | 23/7 vs. 22/9 | 1.63 ± 0.07 vs. 1.59 ± 0.86 | 78.23 ± 13.00 vs. 80.41 ± 17.52 | 29.48 ± 5.22 vs. 31.61 ± 7.38 | NA | NA | NA | NA |
| Garza et al. [41] | 2020 | USA | RCT | 59.5 ± 11.7 vs. 57.1 ± 9.1 | 6/7 vs. 7/6 | NA | NA | 28.8 ± 4.3 vs. 27.1 ± 2.7 | NA | 4 vs. 4 | 9 vs. 9 | NA |
| Goncars et al. [42] | 2017 | Latvia | RCT | 53.44 ± 15 vs. 58.55 ± 13 | 13/15 vs. 21/10 | NA | NA | NA | 7 sv 9 | 21/19 | NA | NA |
| Gupta et al. [43] | 2016 | India | RCT | 58.10 ± 8.23 vs. 54.90 ± 8.27 | 7/3 vs. 10/0 | 156.85 ± 9.64 vs. 152.25 ± 9.72 | 73.10 ± 15.86 vs. 66.10 ± 7.67 | 29.73 ± 6.09 vs. 28.84 ± 4.91 | NA | 4 vs. 3 | 6 vs. 7 | NA |
| Kuah et al. (1) [44] | 2018 | Australia | RCT | 50.8 ± 7.29 vs. 55.0 ± 10.42 | 2/6 vs. 3/1 | 172.6 ± 10.99 vs. 165.0 ± 7.87 | 82.9 ± 12.44 vs. 69.8 ± 11.55 | 27.7 ± 2.05 vs. 25.5 ± 2.84 | 0 vs. 1 | 2 vs. 0 | 6 vs. 3 | NA |
| Kuah et al. (2) [44] | 2018 | Australia | RCT | 55.0 ± 5.15 vs. 55.0 ± 10.42 | 3/5 vs. 3/1 | 174.4 ± 11.99 vs. 165.0 ± 7.87 | 81.9 ± 14.23 vs. 69.8 ± 11.55 | 26.8 ± 2.98 vs. 25.5 ± 2.84 | 1 vs. 1 | 1 vs. 0 | 6 vs. 3 | NA |
| Lamo-Espinosa et al. (1) [45] | 2016 | Spain | RCT | 65.54 ± 3.58 vs. 59.36 ± 1.90 | 4/6 vs. 3/7 | NA | NA | 27.39 ± 2.19 vs. 28.45 ± 1.68 | 0 vs. 0 | 1 vs. 4 | 2 vs. 2 | 7 vs. 4 |
| Lamo-Espinosa et al. (2) [45] | 2016 | Spain | RCT | 57.84 ± 1.87 vs. 59.36 ± 1.90 | 2/8 vs. 3/7 | NA | NA | 29.14 ± 1.48 vs. 28.45 ± 1.68 | 0 vs. 0 | 3 vs. 4 | 3 vs. 2 | 4 vs. 4 |
| Lamo-Espinosa et al. (1) [46] | 2018 | Spain | RCT | 64.40 ± 10.0 vs. 60.33 ± 0.73 | 4/4 vs. 2/7 | NA | NA | 27.49 ± 7.50 vs. vs 29.00 ± 1.53 | NA | 1 vs. 4 | 2 vs. 2 | 5 vs. 3 |
| Lamo-Espinosa et al. (2) [46] | 2018 | Spain | RCT | 58.21 ± 2.99 vs. 60.33 ± 0.73 | 2/6 vs. 2/7 | NA | NA | 28.48 ± 2.40 vs. 29.00 ± 1.53 | NA | 2 vs. 4 | 3 vs. 2 | 4 vs. 3 |
| Lee et al. [47] | 2019 | South Korea | RCT | 62.2 ± 6.5 vs. 63.2 ± 4.2 | 9/3 vs. 9/3 | 159.4 ± 7.2 vs. 159.8 ± 7.0 | 66.5 ± 11.1 vs. 65.7 ± 12.4 | 25.3 ± 4.9 vs. 25.4 ± 3.0 | NA | 6 vs. 5 | 6 vs. 6 | 0 vs. 1 |
| Lee et al. [48] | 2025 | Korea | RCT | 71.2 ± 5.7 vs. 63.8 ± 8.6 | 8/4 vs. 12/0 | NA | NA | 25.8 ± 3.2 vs. 24.1 ± 3.0 | 0 vs. 0 | 4 vs. 5 | 6 vs. 6 | 2 vs. 1 |
| Lu et al. [49] | 2019 | China | RCT | 55.03 ± 9.19 vs. 59.64 ± 5.97 | 23/3 vs. 23/3 | 161.35 ± 6.43 vs. 162.46 ± 5.66 | 63.46 ± 10.69 vs. 62.81 ± 9.44 | 24.27 ± 3.04 vs. 24.26 ± 2.59 | 2 vs. 4 | 18 vs. 16 | 32 vs. 32 | 0 vs. 0 |
| Matas et al. (1) [27] | 2019 | Chile | RCT | 56.1 ± 6.8 vs. 54.8 ± 4.5 | 6/4 vs. 5/4 | NA | NA | 27.6 ± 2.6 vs. 27.9 ± 3.4 | NA | 5 vs. 7 | 5 vs. 2 | NA |
| Matas et al. (2) [27] | 2019 | Chile | RCT | 56.7 ± 4.1 vs. 54.8 ± 4.5 | 5/5 vs. 5/4 | NA | NA | 27.4 ± 2.6 vs. 27.9 ± 3.4 | NA | 6 vs. 7 | 4 vs. 2 | NA |
| Vega et al. [50] | 2016 | Spain | RCT | 56.6 ± 9.24 vs. 57.3 ± 9.09 | 9/6 vs. 10/5 | NA | NA | NA | NA | 6 vs. 7 | 6 vs. 5 | 3 vs. 3 |
BMI: Body mass index, MSC Mesenchymal stem cell, NA Not applicable, SD Standard deviation, vs. Versus
Table 2.
Preparation and intervention characteristics of MSC
| Author | Year | MSC (MSC vs. control) | Control | Injection | Number of injections | |||
|---|---|---|---|---|---|---|---|---|
| Harvest location | MSCs type | MSCs source | MSCs No. | |||||
| Bastos et al. [36] | 2020 | Posterior iliac crests | BM-MSC | Auto | 40 × 106 | Corticosteroid | Joint cavity injection | 1 |
| chen et al. (1) [37] | 2021 | Ultrasonic-assisted liposuction | AD-MSC | Auto | 16 × 106 | HA | Joint cavity injection | 1 |
| chen et al. (2) [37] | 2021 | Ultrasonic-assisted liposuction | AD-MSC | Auto | 32 × 106 | HA | Joint cavity injection | 1 |
| chen et al. (3) [37] | 2021 | Ultrasonic-assisted liposuction | AD-MSC | Auto | 64 × 106 | HA | Joint cavity injection | 1 |
| Emadedin et al. [38] | 2018 | Iliac crests | BM-MSC | Auto | 40 × 106 | Saline | Joint cavity injection | 1 |
| Freitag et al. (1) [39] | 2019 | Abdomen | AD-MSC | Allo | 100 × 106 | HA | Joint cavity injection | 1 |
| Freitag et al. (2) [39] | 2019 | Abdomen | AD-MSC | Allo | 100 × 106 | HA | Joint cavity injection | 2 |
| Garay-Mendoza et al. [40] | 2017 | Iliac crests | BM-MSC | Auto | 10mL | Oral acetaminophen 500 mg | Joint cavity injection | 1 |
| Garza et al. [41] | 2020 | NA | AD-MSC | Allo | 3.0 × 107 | Placebo | Joint cavity injection | 1 |
| Goncars et al. [42] | 2017 | Iliac crest | BM-MSC | Auto | 5–10 mL | HA | Joint cavity injection | 1 |
| Gupta et al. [43] | 2016 | NA | BM-MSC | Allo | 25 × 106 | HA | Joint cavity injection | 1 |
| Kuah et al. (1) [44] | 2018 | NA | AD-MSC | Allo | 3.9 × 106 | Placebo | Joint cavity injection | 4 |
| Kuah et al. (2) [44] | 2018 | NA | AD-MSC | Allo | 6.7 × 106 | Placebo | Joint cavity injection | 4 |
| Lamo-Espinosa et al. (1) [45] | 2016 | Iliac crest | BM-MSC | Auto | 10 × 106 | HA | Joint cavity injection | 1 |
| Lamo-Espinosa et al. (2) [45] | 2016 | Iliac crest | BM-MSC | Auto | 100 × 106 | HA | Joint cavity injection | 1 |
| Lamo-Espinosa et al. (1) [46] | 2018 | Iliac crest | BM-MSC | Auto | 10 × 106 | HA | Joint cavity injection | 1 |
| Lamo-Espinosa et al. (2) [46] | 2018 | Iliac crest | BM-MSC | Auto | 100 × 106 | HA | Joint cavity injection | 1 |
| Lee et al. [47] | 2019 | Abdomen | AD-MSC | Auto | 1 × 108 | Saline | Joint cavity injection | 1 |
| Lee et al. [48] | 2025 | Iliac crest | BM-MSC | Auto | 1 × 108 | Saline | Joint cavity injection | 1 |
| Lu et al. [49] | 2019 | Abdomen | AD-MSC | Auto | 5 × 107 | HA | Joint cavity injection | 4 |
| Matas et al. (1) [27] | 2019 | Umbilical cords | UC-MSC | Allo | 20 × 106 | HA | Joint cavity injection | 1 |
| Matas et al. (2) [27] | 2019 | Umbilical cords | UC-MSC | Allo | 20 × 106 | HA | Joint cavity injection | 2 |
| Vega et al. [50] | 2016 | Posterior iliac crests | BM-MSC | Allo | 40 × 106 | HA | Joint cavity injection | 1 |
AD-MSC Adipose-derived mesenchymal stem cell, Allo Allogeneic, Auto Autologous, BM-MSC Bone marrow-derived mesenchymal stem cell, HA Hyaluronic acid, MSC Mesenchymal stem cell, NA Not applicable, UC-MSC Umbilical cord-derived mesenchymal stem cell
Risk of bias assessment
The risk of bias for the included RCTs was assessed using the Cochrane Collaboration tool [51]. Most studies were consistent in registration and reporting, with selective reporting deemed to have a low risk of bias. One study employed a single-blind design and was rated as high risk for performance and detection bias. Two studies were considered to have a high risk of performance bias. Missing data in two studies were not included in the intention-to-treat (ITT) analysis, resulting in a high risk of attrition bias. The majority of studies were assessed as having a low risk of bias in random sequence generation, blinding, incomplete outcome data, and selective reporting. Additionally, if the study received funding from the manufacturer of the experimental group preparation, it was considered to have an unclear risk of bias in other areas. Overall, the methodological quality of the included RCTs was deemed acceptable. The results of risk of bias assessment are shown in Fig. 2A and B.
Fig. 2.
The risk of bias graph of included studies. A Risk of bias graph; B Risk of bias summary
Results of conventional meta-analysis
WOMAC total score
The WOMAC total score is primarily used to assess knee pain, functional limitations, and stiffness [52]. We conducted a conventional meta-analysis to evaluate the improvement in WOMAC total score at different time points across subgroups (single and repeated MSC injections) compared to baseline.
WOMAC total score at 3 months. The single-injection group included 7 studies (11 datasets), involving 136 patients with knee osteoarthritis; the repeated-injection group included 1 study with 10 patients. Pooled analysis showed a significant improvement in WOMAC total score at 3 months after a single MSC injection compared to baseline (SMD = −1.11, 95% CI −1.56 to −0.66, P < 0.00001), with substantial heterogeneity (I² = 64%) (Fig. 3A). Due to the availability of only one dataset in the repeated-injection group, further evaluation was not feasible. Moreover, the pooled results indicated no significant difference in the improvement effect on WOMAC score between single and repeated injections (P = 0.13).
Fig. 3.
Forest plots illustrating the effect of single and repeated MSC injections on WOMAC total score at different time points. A WOMAC total score at 3 months. B WOMAC total score at 6 months. C WOMAC total score at 12 months
WOMAC total score at 6 months. The single-injection group included 12 studies (16 datasets) involving 217 patients with knee osteoarthritis, while the repeated-injection group included 3 studies with a total of 46 patients. Pooled analysis showed that at 6 months after MSC injection, both single and repeated injections led to significant improvements in WOMAC total score compared to baseline (SMD = − 1.51, 95% CI − 2.01 to − 1.00, P < 0.00001), with considerable heterogeneity (I² = 84%) (Fig. 3B). Additionally, the pooled results indicated no significant difference in the improvement effect on WOMAC score between single and repeated injections (P = 0.97).
WOMAC total score at 12 months. The single-injection group included 9 studies (13 datasets) involving 156 patients, while the repeated-injection group included 3 studies involving 46 patients. Pooled analysis showed that at 12 months after MSC injection, both single and repeated injections resulted in significant improvements in WOMAC total score compared to baseline (SMD = −1.93, 95% CI − 2.46 to − 1.39, P < 0.00001), with substantial heterogeneity (I² = 78%) (Fig. 3C). Additionally, the pooled results indicated no significant difference in improvement between single and repeated injections (P = 0.67). The results across different time points suggested that both single and repeated MSC injections may provide sustained improvement in WOMAC total scores compared to pre-injection levels. Notably, in the single-injection group, there was a trend toward reduced improvement at 12 months compared to 6 months, which may reflect a relatively short duration of effect. In contrast, the repeated-injection group showed potential for more durable benefits. However, further confirmation through network meta-analysis is needed.
WOMAC score of function
WOMAC score (Function) at 3 months. The single-injection group included 4 studies (7 datasets) involving 92 patients, while the repeated-injection group included 1 study (2 datasets) involving 16 patients. Although the pooled results showed no significant heterogeneity (I² = 0%), a random-effects model was adopted for the analysis due to variations in MSC sources, intervention details, and baseline characteristics of the patients. Pooled analysis showed that at 3 months after MSC injection, both single and repeated injections resulted in significant improvement in WOMAC Function score compared to baseline (SMD = − 0.94, 95% CI − 1.23 to − 0.66, P < 0.00001) (Fig. 4A). Moreover, the pooled results indicated no significant difference in functional improvement between single and repeated injections (P = 0.99).
Fig. 4.
Forest plots showing the effect of single and repeated MSC injections on WOMAC score (Function) at different time points. A WOMAC score (Function) at 3 months. B WOMAC score (Function) at 6 months. C WOMAC score (Function) at 12 months
WOMAC score (Function) at 6 months. The single-injection group included 5 studies (8 datasets) involving 102 patients, while the repeated-injection group included 2 studies (3 datasets) involving 26 patients. Pooled analysis showed that at 6 months after MSC injection, both single and repeated injections resulted in significant improvement in WOMAC Function score compared to baseline (SMD = − 1.12, 95% CI − 1.41 to − 0.83, P < 0.00001) (I² = 12%) (Fig. 4B). Furthermore, the pooled results revealed no significant difference in the improvement of WOMAC Function score between single and repeated injections (P = 0.38).
WOMAC score (Function) at 12 months. The WOMAC score (Function) at 12 months. The single injection group included 5 studies (8 datasets) with 102 patients, while the repeated injection group included 2 studies (3 datasets) with 26 patients. The pooled analysis based on a random-effects model showed that, compared to pre-treatment levels, the WOMAC score (Function) at 12 months significantly improved after either single or repeated MSC injections (SMD = − 1.19, 95% CI − 1.51 to − 0.86, P < 0.00001) (I² = 0%) (Fig. 4C). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Function) (P = 0.59). These results suggest that both single and repeated MSC injections could significantly improve knee joint function scores at all time points. The single injection group showed an increasing improvement trend as follow-up time extended. However, no statistically significant differences were observed between the two injection protocols.
WOMAC score of pain
WOMAC score (Pain) at 3 months. The WOMAC score (Pain) at 3 months. The single injection group included 5 studies (8 datasets) with 107 patients, while the repeated injection group included 1 study (2 datasets) with 16 patients. Due to the presence of substantial heterogeneity (I² = 59%) and differences in MSC characteristics and patient baseline features, a random-effects model was employed. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Pain) at 3 months significantly improved after both single and repeated MSC injections (SMD = − 1.40, 95% CI − 1.86 to − 0.94, P < 0.00001) (Fig. 5A). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Pain) (P = 0.22).
Fig. 5.
Forest plots illustrating the effect of single and repeated MSC injections on WOMAC score (Pain) at different time points. A WOMAC score (Pain) at 3 months. B WOMAC score (Pain) at 6 months. C WOMAC score (Pain) at 12 months
WOMAC score (Pain) at 6 months. The single injection group included 6 studies (9 datasets) with 117 patients, while the repeated injection group included 2 studies (3 datasets) with 26 patients. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Pain) at 6 months significantly improved after both single and repeated MSC injections (SMD = − 1.33, 95% CI − 1.78 to − 0.87, P < 0.00001) (I² = 65%) (Fig. 5B). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Pain) (P = 0.18).
WOMAC score (Pain) at 12 months. The single injection group included 6 studies (9 datasets) with 117 patients, while the repeated injection group included 2 studies (3 datasets) with 26 patients. In view of the differences in MSC characteristics and patient baseline features among the included studies, a random-effects model was applied. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Pain) at 12 months significantly improved after both single and repeated MSC injections (SMD = − 1.22, 95% CI − 1.54 to − 0.89, P < 0.00001) (I² = 34%) (Fig. 5C). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Pain) (P = 0.60). These results suggest that both single and repeated MSC injections could significantly reduce the WOMAC pain score at all follow-up time points. However, no statistical differences were observed between the single and repeated injection groups. Given the limited number of studies and small sample sizes, these results should be interpreted with caution.
WOMAC score of stiffness
WOMAC score (Stiffness) at 3 months. The single injection group included 4 studies (6 datasets) with 92 patients, while the repeated injection group included 1 study (2 datasets) with 16 patients. Given the heterogeneity in MSC characteristics and patient baseline features among the included studies, a random-effects model was utilized. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Stiffness) at 3 months significantly improved after both single and repeated MSC injections (SMD = − 0.80, 95% CI − 1.08 to − 0.52, P < 0.00001) (I² = 0%) (Fig. 6A). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Stiffness) (P = 0.97).
Fig. 6.
Forest plots showing the effect of single and repeated MSC injections on WOMAC score (Stiffness) at different time points. A WOMAC score (Stiffness) at 3 months. B WOMAC score (Stiffness) at 6 months. C WOMAC score (Stiffness) at 12 months
WOMAC score (Stiffness) at 6 months. The single injection group included 5 studies (8 datasets) with 102 patients, while the repeated injection group included 2 studies (3 datasets) with 26 patients. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Stiffness) at 6 months significantly improved after both single and repeated MSC injections (SMD = − 0.92, 95% CI − 1.18 to − 0.66, P < 0.00001) (I² = 0%) (Fig. 6B). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Stiffness) (P = 0.23).
WOMAC score (Stiffness) at 12 months. The single injection group includes 5 studies (8 datasets) with 102 patients, while the repeated injection group includes 2 studies (3 datasets) with 26 patients. The pooled analysis showed that, compared to pre-treatment levels, the WOMAC score (Stiffness) at 12 months significantly improved after both single and repeated MSC injections (SMD = − 1.01, 95% CI − 1.30 to − 0.72, P < 0.00001) (Fig. 6C). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the WOMAC score (Stiffness) (I² = 14%) (P = 0.69). In conclusion, compared to pre-injection levels, both single and repeated injections could improve knee joint function, stiffness, and pain at various follow-up time points. There were no significant differences in the effectiveness between the two injection protocols. However, due to study heterogeneity and small sample sizes, the results should be interpreted with caution.
VAS score
VAS score at 3 months. The single injection group includes 6 studies (10 datasets) with 123 patients, while the repeated injection group includes 1 study (2 datasets) with 16 patients. Given the heterogeneity in MSC characteristics and patient baseline features among the included studies, a random-effects model was utilized. The pooled analysis showed that, compared to pre-treatment levels, the VAS score at 3 months significantly improved after both single and repeated MSC injections (SMD = − 1.22, 95% CI − 1.60 to − 0.84, P < 0.00001) (I² = 50%) (Fig. 7A). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the VAS score (P = 0.98).
Fig. 7.
Forest plots showing the effect of single and repeated MSC injections on VAS score at different time points. A VAS score at 3 months. B VAS score at 6 months. C VAS score at 12 months
VAS score at 6 months. The single injection group included 10 studies (14 datasets) with 194 patients, while the repeated injection group included 3 studies (4 datasets) with 52 patients. The pooled analysis showed that, compared to pre-treatment levels, the VAS score at 6 months significantly improved after both single and repeated MSC injections (SMD = − 1.46, 95% CI − 1.82 to − 1.10, P < 0.00001) (I² = 65%) (Fig. 7B). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the VAS score (P = 0.91).
VAS score at 12 months. The single injection group included 7 studies (11 datasets) with 133 patients, while the repeated injection group included 3 studies (4 datasets) with 52 patients. The pooled analysis showed that, compared to pre-treatment levels, the VAS score at 12 months significantly improved after both single and repeated MSC injections (SMD = − 1.57, 95% CI − 1.96 to − 1.18, P < 0.00001) (I² = 60%) (Fig. 7C). Furthermore, the pooled results also indicated no significant difference between the effects of single and multiple injections on the improvement of the VAS score (P = 0.41). Through the pooled analysis of the VAS score, we found that both single and repeated MSC injections may improve patient pain at different time points compared to pre-treatment. However, there were no significant differences in efficacy between the two injection protocols.
KOOS score
The KOOS score is a tool used to evaluate knee joint damage based on the subjective feelings of patient, including five aspects: daily activities, pain, quality of life, sports, and symptom improvement [53]. We conducted a traditional meta-analysis using data from the 12-month follow-up time point to explore the effects of different injection protocols before and after injection on the KOOS score.
KOOS score (activities of daily living) at 12 months. The single injection group included 3 studies, involving 38 patients, while the repeated injection group included 1 study, involving 10 patients. The pooled analysis of the single injection group showed that, compared to pre-treatment levels, the KOOS score (activities of daily living) at 12 months significantly improved (SMD = 1.35, 95% CI 0.77 to 1.93, P < 0.00001; I² = 0%) (Fig. 8A).
Fig. 8.
Forest plots illustrating the effect of single and repeated MSC injections on KOOS score at 12 months. A KOOS score (Activities of daily living) at 12 months. B KOOS score (pain) at 12 months. C KOOS score (Quality of life) at 12 months. D KOOS score (Sport) at 12 months. E KOOS score (symptom improvement) at 12 months
KOOS score (pain) at 12 months. The single injection group includes 4 studies, involving 66 patients, while the repeated injection group includes 1 study, involving 10 patients. The pooled analysis of the single injection group showed that, compared to pre-treatment levels, the KOOS score (pain) at 12 months significantly improved (SMD = 1.35, 95% CI 0.99 to 1.71, P < 0.00001; I² = 0%) (Fig. 8B).
KOOS score (quality of life) at 12 months. The single injection group includes 3 studies, involving 38 patients, while the repeated injection group includes 1 study, involving 10 patients. The pooled analysis of the single injection group showed that, compared to pre-treatment levels, the KOOS score (quality of life) at 12 months significantly improved (SMD = 1.37, 95% CI 0.76 to 1.99, P < 0.0001; I² = 43%) (Fig. 8C).
KOOS score (sport) at 12 months. The single injection group includes 3 studies, involving 38 patients, while the repeated injection group includes 1 study, involving 10 patients. The pooled analysis of the single injection group showed that, compared to pre-treatment levels, the KOOS score (sport) at 12 months significantly improved (SMD = 1.43, 95% CI 0.62 to 2.24, P = 0.0005; I² = 65%) (Fig. 8D).
KOOS score (symptom improvement) at 12 months. The single injection group includes 3 studies, involving 38 patients, while the repeated injection group includes 1 study, involving 10 patients. The pooled analysis of the single injection group showed that, compared to pre-treatment levels, the KOOS score (symptom improvement) at 12 months significantly improved (SMD = 1.04, 95% CI 0.61 to 1.47, P < 0.00001; I² = 0%) (Fig. 8E). These results suggested that, compared to the control group, single MSC injection may have a positive impact on daily activities, pain, quality of life, sports, and symptom improvement in patients with knee osteoarthritis. However, due to the limited number of studies in the repeated MSC injection group, caution should be exercised when interpreting these results further.
WORMS score (articular cartilage)
The WORMS score is a rating of cartilage damage in the knee joint based on MRI imaging [54]. We conducted a pooled analysis of the scores from several studies. The single injection group includes 4 studies (7 datasets) with 90 patients, while the repeated injection group includes 1 study with 10 patients. The results showed that, compared to pre-treatment levels, the WORMS score (articular cartilage) did not significantly improve after a single MSC injection (SMD = 0.11, 95% CI − 0.16 to 0.39, P = 0.42; I² = 0%) (Fig. 9). Due to the limited number of studies and sample sizes, caution should be exercised when interpreting these results further.
Fig. 9.
Forest plots illustrating the effect of single and repeated MSC injections on WORMS score (articular cartilage) at 12 months
Results of network meta-analysis
WOMAC total score
WOMAC total score at 6 months. In the network meta-analysis, a total of 17 datasets were included, involving 487 patients. The network evidence plot for the three intervention protocols is shown in Fig. 10A. The network forest plot illustrates the results of three pairwise comparisons, showing no statistical difference in treatment effects between the single and repeated MSC injection groups (Fig. 10B). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC total score with a probability of P = 73.8 (Fig. 10C). This was followed by the single MSC injection (P = 48.1) and conventional therapy (P = 28.2).
Fig. 10.
Network meta-analysis of single and repeated MSC injections on WOMAC total score at 6 and 12 months. A Evidence network diagram of WOMAC total score at 6 months. B Forest plot showing the comparative evidence for single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for WOMAC total score at 6 months after single and repeated MSC injections. D Evidence network diagram of WOMAC total score at 12 months. E Forest plot showing the comparative evidence for single and repeated MSC treatments. F SUCRA curve and area under the curve (%) for WOMAC total score at 12 months after single and repeated MSC injections
WOMAC total score at 12 months. In the network meta-analysis, a total of 14 datasets were included, involving 358 patients. The network evidence plot for the three intervention protocols is shown in Fig. 10D. The network forest plot showed that, compared to single MSC injection, repeated MSC injection may effectively reduce the WOMAC total score at 12 months (SMD = −1.47, 95% CI − 2.86 to − 0.07; very low certainty) (Fig. 10E). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC total score with a probability of P = 98.8 (Fig. 10F). Single MSC injection ranked second with a probability of P = 48.1. These results suggested that repeated MSC injection may be the best option for improving the WOMAC total score at 12 months.
WOMAC score of function
WOMAC score (Function) at 6 months. A total of 10 datasets involving 246 patients were included. The main evidence for all pairwise comparison protocols regarding WOMAC score (Function) is shown in Fig. 11A. Among these, repeated MSC injection significantly reduced the WOMAC score (Function) at 6 months compared to single MSC injection (SMD = − 1.24, 95% CI − 2.33 to − 0.15; low certainty) (Fig. 11B). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Function) with a probability of P = 99.7 (Fig. 11C). This was followed by single MSC injection (P = 36.0) and conventional therapy (P = 14.9). These results suggest that repeated MSC injection may be the best option for improving the WOMAC score (Function) at 6 months.
Fig. 11.
Network meta-analysis of single and repeated MSC injections on WOMAC score (Function) at 6 and 12 months. A Evidence network diagram of WOMAC score (Function) at 6 months. B Forest plot showing the comparative evidence for single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for WOMAC score (Function) at 6 months after single and repeated MSC injections. D Evidence network diagram of WOMAC score (Function) at 12 months. E Forest plot showing the comparative evidence for single and repeated MSC treatments. F SUCRA curve and area under the curve (%) for WOMAC score (Function) at 12 months after single and repeated MSC injections
WOMAC score (Function) at 12 months. The data is consistent with that at 6 months, including a total of 10 datasets involving 246 patients. The network evidence plot for the three intervention protocols is shown in Fig. 11D. The network forest plot showed that, compared to the conventional therapy group, repeated MSC injection may effectively reduce the WOMAC score (Function) at 12 months (SMD = − 0.30, 95% CI − 0.59 to − 0.01; low certainty) (Fig. 11E). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Function) with a probability of P = 93.5 (Fig. 11F). Single MSC injection ranked second with a probability of P = 54.1. These results suggest that repeated MSC injection may be the best option for improving the WOMAC score (Function) at both 6 and 12 months.
WOMAC score of pain
WOMAC score (Pain) at 6 months. A total of 11 datasets involving 276 patients were included. The main evidence for all pairwise comparison protocols regarding WOMAC score (Pain) is shown in Fig. 12A. The network forest plot illustrates the results of the three pairwise comparisons, showing no statistical difference in treatment effects between the single and repeated MSC injection groups (Fig. 12B). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Pain) with a probability of P = 75.1 (Fig. 12C). This was followed by single MSC injection (P = 56.8) and conventional therapy (P = 18.1). These results suggest that repeated MSC injection may be the best option for improving the WOMAC score (Pain) at 6 months.
Fig. 12.
Network meta-analysis of single and repeated MSC injections on WOMAC score (Pain) at 6 and 12 months. A Evidence network diagram of WOMAC score (Pain) at 6 months. B Forest plot showing the comparative evidence for single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for WOMAC score (Pain) at 6 months after single and repeated MSC injections. D Evidence network diagram of WOMAC score (Pain) at 12 months. E Forest plot showing the comparative evidence for single and repeated MSC treatments. F SUCRA curve and area under the curve (%) for WOMAC score (Pain) at 12 months after single and repeated MSC injections
WOMAC score (Pain) at 12 months. In the network meta-analysis, a total of 11 datasets involving 276 patients were included. The network evidence plot for the three intervention protocols is shown in Fig. 12D. The network forest plot illustrates the results of the three pairwise comparisons, showing no statistical difference in treatment effects between the single and repeated MSC injection groups (Fig. 12E). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Pain) with a probability of P = 89.6 (Fig. 12F). This was followed by conventional therapy (P = 39.6) and single MSC injection (P = 20.7). These results suggest that repeated MSC injection may be the best option for improving the WOMAC score (Pain) at both 6 and 12 months.
WOMAC score of stiffness
WOMAC score (Stiffness) at 6 months. A total of 10 datasets involving 246 patients were included. The main evidence for all pairwise comparison protocols regarding WOMAC score (Stiffness) is shown in Fig. 13A. The network forest plot illustrates the results of the three pairwise comparisons, showing no statistical difference in treatment effects between the single and repeated MSC injection groups (Fig. 13B). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Stiffness) with a probability of P = 62.9 (Fig. 13C). This was followed by conventional therapy (P = 47.8) and single MSC injection (P = 39.3).
Fig. 13.
Network meta-analysis of single and repeated MSC injections on WOMAC score (Stiffness) at 6 and 12 months. A Evidence network diagram of WOMAC score (Stiffness) at 6 months. B Forest plot showing the comparative evidence for single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for WOMAC score (Stiffness) at 6 months after single and repeated MSC injections. D Evidence network diagram of WOMAC score (Stiffness) at 12 months. E Forest plot showing the comparative evidence for single and repeated MSC treatments. F SUCRA curve and area under the curve (%) for WOMAC score (Stiffness) at 12 months after single and repeated MSC injections
WOMAC score (Stiffness) at 12 months. In the network meta-analysis, a total of 10 datasets involving 246 patients were included. The network evidence plot for the three intervention protocols is shown in Fig. 13D. The network forest plot showed that, compared to the conventional therapy group, repeated MSC injection may effectively reduce the WOMAC score (Stiffness) at 12 months (SMD = − 0.32, 95% CI − 0.62 to − 0.02; low certainty) (Fig. 13E). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the WOMAC score (Stiffness) with a probability of P = 85.7 (Fig. 13F). This was followed by single MSC injection (P = 59.9) and conventional therapy (P = 4.40). These results suggest that repeated MSC injection may be the best option for improving the WOMAC score (Stiffness) at both 6 and 12 months.
VAS score
VAS score at 6 months. A total of 17 datasets involving 482 patients were included. The main evidence for all pairwise comparison protocols regarding the VAS score is shown in Fig. 14A. Among these, repeated MSC injection significantly reduced the VAS score at 6 months compared to conventional therapy (SMD = − 0.77, 95% CI − 1.26 to − 0.28; low certainty) (Fig. 14B). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the VAS score with a probability of P = 94.7 (Fig. 14C). This was followed by single MSC injection (P = 54.5) and conventional therapy (P = 0.8). These results suggest that repeated MSC injection may be the best option for improving the VAS score at 6 months.
Fig. 14.
Network meta-analysis of single and repeated MSC injections on VAS score at 6 and 12 months. A Evidence network diagram of VAS score at 6 months. B Forest plot showing the comparative evidence for single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for VAS score at 6 months after single and repeated MSC injections. D Evidence network diagram of VAS score at 12 months. E Forest plot showing the comparative evidence for single and repeated MSC treatments. F SUCRA curve and area under the curve (%) for VAS score at 12 months after single and repeated MSC injections
VAS score at 12 months. The network meta-analysis included a total of 14 datasets involving 360 patients. The network evidence plot for the three intervention protocols is shown in Fig. 14D. The network forest plot showed that, compared to the conventional therapy group, repeated MSC injection may effectively reduce the VAS score at 12 months (SMD = − 0.90, 95% CI − 1.57 to − 0.23; low certainty) (Fig. 14E). Based on SCURA ranking results, the repeated MSC injection group was ranked as the best treatment for reducing the VAS score with a probability of P = 95.4 (Fig. 14F). Single MSC injection ranked second with a probability of P = 53.7. These results suggest that repeated MSC injection may be the best option for improving the VAS score at both 6 and 12 months.
AE and SAE
In the traditional meta-analysis of related adverse events, the single injection group includes 9 studies (12 datasets) involving 308 patients, while the repeated injection group includes 4 studies (5 datasets) involving 115 patients. The pooled analysis showed that, compared to conventional therapy, both single and repeated MSC injections significantly increased the incidence of adverse events (OR = 5.72, 95% CI 2.52 to 13.01, P < 0.0001; low certainty) (I² = 48%) (Fig. 15A). Furthermore, the pooled results also indicated no significant difference in the incidence of adverse events between the different injection protocols (P = 0.95). For SAE, the single injection group includes 3 studies (4 datasets) involving 89 patients, while the repeated injection group includes 2 studies involving 72 patients. The pooled analysis showed that, compared to conventional therapy, neither single nor repeated MSC injections reduced the incidence of serious adverse events (OR = 2.13, 95% CI 0.56 to 8.01, P = 0.26) (Fig. 15B). Additionally, the pooled results indicated no significant difference in the incidence of serious adverse events between the different injection protocols (P = 0.44).
Fig. 15.
Forest plot showing the comparative evidence for adverse events following single and repeated MSC injections in knee osteoarthritis. A Adverse events. B Serious adverse events
In the network meta-analysis, a total of 15 datasets involving 404 patients were included. The network evidence plot shows the pairwise comparisons for the three intervention methods (Fig. 16A). The network forest plot indicates that, compared to the conventional therapy group, repeated MSC injections may increase the incidence of adverse events (OR = 6.86, 95% CI 2.62 to 17.97; very low certainty) (Fig. 16B). Based on SCURA ranking results, the conventional therapy group was ranked as the best treatment for reducing adverse events with a probability of P = 97.3 (Fig. 16C). This was followed by single MSC injection (P = 27.9) and repeated MSC injection (P = 24.8). These results suggest that, compared to conventional therapy, repeated MSC injections may increase the occurrence of adverse events in patients with knee osteoarthritis.
Fig. 16.
Network meta-analysis of adverse events following single and repeated MSC injections. A Evidence network diagram of adverse events. B Forest plot showing the comparative evidence for adverse events following single and repeated MSC treatments. C SUCRA curve and area under the curve (%) for adverse events following single and repeated MSC injections
Evaluation of evidence certainty using the GRADE method
In the comparison between single vs. repeated MSC injections, the GRADE assessment indicated that the evidence quality for repeated MSC injections in reducing the WOMAC total score at 12 months was “very low,” and for reducing the WOMAC Function score at 6 months, the evidence quality was “low”. In the comparison between repeated MSC injections and conventional therapy, the evidence quality for repeated MSC injections in reducing the WOMAC Function score at 12 months, WOMAC Function score at 12 months, VAS score at 6 months, and VAS score at 12 months was all assessed as “low”, while the evidence certainty for increasing the incidence of adverse events was considered “very low”. A summary of the findings from the NMA and the assessment of evidence certainty can be found in Table 3.
Table 3.
Summary of findings
| Outcome | Comparison | Direct estimate | Indirect estimate | NMA estimate | Comments | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| SMD (95% CI) | Certainty in the evidence |
SMD (95% CI) | Certainty in the evidence |
SMD (95% CI) | Certainty in the Evidence |
|||||
| WOMAC total score at 6 months | Single-dose MSC vs. conventional therapy | − 0.26 (− 0.91 to 0.39) | ⊕ very lowa, b,c | − 0.43 (− 1.20 to 0.33) | ⊕ very lowa, d | − 0.54 (− 2.19 to 1.11) | ⊕ very low | Single or repeated MSC injections may not reduce the WOMAC total score at 6 months, but the evidence is highly uncertain. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.53 (− 0.95 to − 0.11) | ⊕⊕ lowa, d | − 0.24 (− 0.91 to 0.44) | ⊕ very lowa | − 0.26 (− 0.80 to 0.27) | ⊕ very low | ||||
| Repeated vs. single-dose MSC | 0.09 (− 1.18 to 1.36) | ⊕ very lowa, b,d | − 0.35 (− 0.83 to 0.12) | ⊕ very lowa | − 0.43 (− 2.07 to 1.22) | ⊕ very low | ||||
| WOMAC total score at 12 months | Single-dose MSC vs. conventional therapy | − 0.42 (− 0.92 to 0.07) | ⊕ very lowa, b,c | − 0.12 (− 0.91 to 0.67) | ⊕ very lowa, d | − 0.79 (− 2.17 to 0.58) | ⊕ very low | Compared to a single MSC injection, repeated injections may reduce the WOMAC total score at 12 months, but the evidence is highly uncertain. | ||
| Repeated-dose MSC vs. conventional therapy | − 1.34 (− 2.63 to − 0.06) | ⊕ very lowa, b,d | − 1.06 (− 1.76 to − 0.37) | ⊕ very lowda | − 0.31 (− 0.84 to 0.22) | ⊕ very low | ||||
| Repeated vs. single-dose MSC | − 0.74 (− 1.48 to 0.00) | ⊕ very lowa, b,d | − 0.45 (− 0.96 to 0.06) | ⊕ very lowa | − 1.47 (− 2.86 to -0.07) | ⊕ very low | ||||
| WOMAC score (Function) at 6 months | Single-dose MSC vs. conventional therapy | − 0.12 (− 0.47 to 0.2) | ⊕⊕⊕ Moderatea | 0.94 (− 0.17 to 2.05) | ⊕ very lowa, d | − 0.39 (− 1.42 to 0.64) | ⊕⊕⊕ Moderate | Compared to a single MSC injection, repeated injections may reduce the WOMAC Function score at 6 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | 0.06 (− 1.53 to 1.64) | ⊕ very lowa, b,d | − 1.05 (− 2.03 to − 0.08) | ⊕⊕ lowa | − 0.10 (− 0.44 to 0.25) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 0.96 (− 1.89 to − 0.03) | ⊕⊕ lowa, d | 0.17 (− 0.48 to 0.83) | ⊕ very lowa | − 1.24 (− 2.33 to − 0.15) | ⊕⊕ low | ||||
| WOMAC score (Function) at 12 months | Single-dose MSC vs. conventional therapy | − 0.27 (− 0.55 to 0.00) | ⊕⊕⊕ Moderatea | 1.42 (0.28 to 2.55) | ⊕ very lowa, d | 0.34 (− 0.85 to 1.27) | ⊕⊕⊕ Moderate | Compared to conventional therapy, repeated injections may reduce the WOMAC Function score at 12 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | 0.33 (− 1.01 to 1.68) | ⊕ very lowa, b,d | − 1.46 (− 2.47 to − 0.45) | ⊕⊕ lowa | − 0.30 (− 0.59 to − 0.01) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 1.26 (− 2.23 to − 0.29) | ⊕⊕ lowa, d | 0.49 (− 0.16 to 1.13) | ⊕ very lowa | − 0.61 (− 1.57 to 0.35) | ⊕⊕ low | ||||
| WOMAC score (Pain) at 6 months | Single-dose MSC vs. conventional therapy | − 0.32 (− 0.81 to 0.18) | ⊕⊕ lowa, b | − 0.63 (− 1.67 to 0.40) | ⊕⊕ lowa, d | − 0.54 (− 2.19 to 1.11) | ⊕⊕ low | Single or repeated MSC injections may not reduce the WOMAC Pain score at 6 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.47 (− 1.03 to 0.08) | ⊕⊕ lowb, d | − 0.07 (− 0.99 to 0.85) | ⊕⊕ lowa | -0.26 (-0.80 to 0.27) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | 0.19 (− 0.69 to 1.07) | ⊕⊕ lowa, d | − 0.20 (− 0.81 to 0.42) | ⊕⊕ lowa | − 0.43 (− 2.07 to 1.22) | ⊕⊕ low | ||||
| WOMAC score (Pain) at 12 months | Single-dose MSC vs. conventional therapy | 0.12 (− 0.43 to 0.66) | ⊕⊕ lowa, b | 0.56 (− 0.57 to 1.70) | ⊕⊕ lowa, d | − 0.38 (− 2.41 to 1.64) | ⊕⊕ low | Single or repeated MSC injections may not reduce the WOMAC Pain score at 12 months, but the evidence is highly uncertain. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.69 (− 1.74 to 0.36) | ⊕⊕ lowb, d | − 1.18 (− 2.19 to − 0.17) | ⊕⊕ lowa | 0.16 (− 0.51 to 0.83) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 1.26 (− 2.23 to − 0.30) | ⊕⊕ lowa, d | − 0.67 (− 1.31 to − 0.03) | ⊕⊕ lowa | − 1.35 (− 3.40 to 0.71) | ⊕⊕ low | ||||
| WOMAC score (Stiffness) at 6 months | Single-dose MSC vs. conventional therapy | 0.05 (− 0.23 to 0.33) | ⊕⊕ lowa, d | 0.12 (− 0.93 to 1.17) | ⊕⊕ lowa, d | 0.19 (− 0.75 to 1.12) | ⊕⊕ low | Single or repeated MSC injections may not reduce the WOMAC Stiffness score at 12 months, but the evidence is highly uncertain. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.35 (− 0.90 to 0.20) | ⊕⊕ lowa, d | − 0.41 (− 1.34 to 0.53) | ⊕⊕ lowa | 0.03 (− 0.28 to 0.34) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 0.47 (− 1.36 to 0.42) | ⊕⊕ lowa, d | − 0.37 (− 0.99 to 0.24) | ⊕⊕ lowa | − 0.17 (− 1.10 to 0.76) | ⊕⊕ low | ||||
| WOMAC score (Stiffness) at 12 months | Single-dose MSC vs. conventional therapy | − 0.12 (− 0.47 to 0.22) | ⊕⊕ lowa, d | 0.55 (− 0.51 to 1.62) | ⊕ very lowa, d | 0.32 (− 0.62 to 1.26) | ⊕⊕ low | Compared to conventional therapy, repeated injections may reduce the WOMAC Stiffness score at 12 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.08 (− 0.99 to 0.82) | ⊕ very lowa, b,d | − 0.95 (− 1.90 to 0.00) | ⊕⊕ lowa | − 0.32 (− 0.62 to − 0.02) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 0.69 (− 1.60 to 0.21) | ⊕⊕ lowa, d | 0.18 (− 0.44 to 0.81) | ⊕ very lowa | − 0.38 (− 1.34 to 0.58) | ⊕⊕ low | ||||
| VAS score at 6 months | Single-dose MSC vs. conventional therapy | − 0.87 (− 1.35 to − 0.38) | ⊕ very lowa, b,c | − 0.50 (− 1.47 to 0.47) | ⊕⊕ lowd | − 1.17 (− 3.09 to 0.74) | ⊕⊕ low | Compared to conventional therapy, repeated injections may reduce the VAS score at 6 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.69 (− 1.50 to 0.12) | ⊕⊕ lowb, d | − 0.89 (− 1.80 to 0.01) | ⊕ very lowa | − 0.77 (− 1.26 to − 0.28) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 0.16 (− 1.03 to 0.72) | ⊕⊕⊕ Moderated | 0.09 (− 0.37 to 0.55) | ⊕ very lowa | − 1.32 (− 3.25 to 0.61) | ⊕⊕⊕ Moderate | ||||
| VAS score at 12 months | Single-dose MSC vs. conventional therapy | − 0.91 (− 1.45 to − 0.36) | ⊕ very lowa, b,c | 1.13 (− 0.01 to − 2.27) | ⊕⊕ lowd | − 0.24 (− 2.44 to 1.97) | ⊕⊕ low | Compared to conventional therapy, repeated injections may reduce the VAS score at 12 months, but the evidence certainty is low. | ||
| Repeated-dose MSC vs. conventional therapy | − 0.70 (− 1.87 to 0.47) | ⊕⊕ lowb, d | -2.32 (-3.42 to -1.23) | ⊕ very lowa | − 0.90 (− 1.57 to − 0.23) | ⊕⊕ low | ||||
| Repeated vs. single-dose MSC | − 1.78 (− 2.83 to − 0.73) | ⊕⊕⊕ Moderated | 0.08 (− 0.41 to 0.57) | ⊕ very lowa | − 1.64 (− 3.93 to 0.65) | ⊕⊕⊕ Moderate | ||||
| Adverse events | Single-dose MSC vs. conventional therapy | 5.95 (2.96 to 11.97) | ⊕⊕ lowa, c | 0.10 (− 1.93 to 2.13) | ⊕ very lowa, d | 0.26 (0.02 to 2.88) | ⊕⊕ low | Compared to conventional therapy, repeated injections may increase adverse events, but the evidence is highly uncertain. | ||
| Repeated-dose MSC vs. conventional therapy | 7.07 (0.93 to 53.95) | ⊕ very lowa, b,d | 2.71 (0.80 to 4.61) | ⊕ very lowa, d | 6.86 (2.62 to 17.97) | ⊕ very low | ||||
| Repeated vs. single-dose MSC | 4.32 (0.98 to 19.16) | ⊕ very lowa, b,d | − 0.32 (− 1.49 to 0.86) | ⊕ very lowa, d | 1.22 (0.08 to 19.17) | ⊕ very low | ||||
CI Confidence interval, MSC Mesenchymal stem cell, NMA Network meta-analysis, OR Odds ratio, SMD Standardized mean difference. ⊕⊕⊕⊕: high certainty; ⊕⊕⊕: moderate certainty; ⊕⊕⊕: low certainty; ⊕: very low certainty; (a) Certainty in the evidence downgraded by one level due to significant risk of bias; (b) Certainty in the evidence downgraded by one level due to inconsistency; (c) Certainty in the evidence downgraded by one level due to significant publication bias; (d) Certainty in the evidence downgraded by one level due to imprecision
Sensitivity analysis
Additionally, sensitivity analysis was conducted to evaluate the robustness of the results from the traditional meta-analysis. Figure 17A shows the sensitivity analysis of the WOMAC scores at different time points (3 months, 6 months, and 12 months). When any single study was excluded, the pooled results from the remaining studies were consistent with the original results, demonstrating the robustness of the findings. Similarly, sensitivity analyses of the WOMAC score (Function), WOMAC score (Pain), and WOMAC score (Stiffness) at different time points (3 months, 6 months, and 12 months) also showed the robustness of the pooled results (Fig. 17B). Moreover, consistent conclusions were also observed in the pooled results for the VAS scores and adverse events at different time points (3 months, 6 months, and 12 months). Overall, the meta-analysis results for various outcome measures were stable (Figures S1 and S2). To further evaluate the robustness of the results, subgroup analyses were performed to assess the potential impact of studies with a high risk of bias on the pooled estimates. The results of the subgroup analyses showed no significant differences between the high-risk and low-risk bias subgroups in terms of VAS score at different time points, WOMAC total score, WOMAC pain score, WOMAC stiffness score, WOMAC function score, and adverse events (Figures S3–S9), further supporting the stability of the meta-analysis findings.
Fig. 17.
Sensitivity analysis of different outcome measures in traditional meta-analysis. A WOMAC total score at 3, 9, and 12 months. B WOMAC score (Function), WOMAC score (Pain), and WOMAC score (Stiffness) at different time points (3, 6, and 12 months)
Publication bias
Funnel plots and Egger’s test were used to assess publication bias. In the traditional meta-analysis, the funnel plots for most outcome measures were symmetrical, and Egger’s test showed no significant evidence of publication bias (Figures S10-S14) (Table 4). For outcomes with a P-value < 0.05 in Egger’s test, the trim-and-fill method was further employed to evaluate the potential impact of publication bias on the results (Figure S15). The analysis revealed that no studies were imputed for WOMAC total score at 6 months, WOMAC total score at 12 months, VAS score at 6 months, and VAS score at 12 months, and the pooled estimates remained unchanged before and after trimming. Therefore, there was no indication of publication bias according to the trim-and-fill method (no new studies added), suggesting that these results were robust. In the network meta-analysis, the funnel plots for all outcome measures were also symmetrical (Figures S16-S21). Overall, the analysis results of the included studies were robust.
Table 4.
Analysis of publication bias for different outcome measures in traditional meta-analysis
| Outcome | Egger’s test (p value) | t value | Pooling model |
|---|---|---|---|
| WOMAC total Score at 3 months | 0.192 | − 1.400 | Random |
| WOMAC total Score at 6 months | 0.000 | − 7.240 | Random |
| WOMAC total Score at 12 months | 0.000 | − 8.720 | Random |
| WOMAC score (Function) at 3 months | 0.392 | 0.910 | Random |
| WOMAC score (Function) at 6 months | 0.098 | − 1.850 | Random |
| WOMAC score (Function) at 12 months | 0.054 | − 2.220 | Random |
| WOMAC score (Pain) at 3 months | 0.558 | − 0.610 | Random |
| WOMAC score (Pain) at 6 months | 0.101 | − 1.820 | Random |
| WOMAC score (Pain) at 12 months | 0.053 | − 2.230 | Random |
| WOMAC score (Stiffness) at 3 months | 0.222 | 1.340 | Random |
| WOMAC score (Stiffness) at 6 months | 0.291 | − 1.120 | Random |
| WOMAC score (Stiffness) at 12 months | 0.326 | − 1.040 | Random |
| VAS score at 3 months | 0.455 | − 0.780 | Random |
| VAS score at 6 months | 0.003 | − 3.470 | Random |
| VAS score at 12 months | 0.003 | − 3.640 | Random |
VAS Visual analogue scale, WOMAC Western Ontario and McMaster University Osteoarthritis Index
Discussion
Main findings
KOA is a progressive disease that can ultimately lead to disability, with its incidence steadily increasing across various populations [55]. Aside from joint replacement surgery for advanced KOA, clinical management typically relies on non-surgical approaches to slow disease progression, including intra-articular injections of HA, PRP, and MSC. Given their multilineage differentiation potential, immunomodulatory and anti-inflammatory properties, and ability to enhance angiogenesis, MSCs have emerged as a promising therapeutic strategy to compensate for the limited regenerative capacity of cartilage in KOA [56, 57]. However, standardized protocols for MSC preparation and intervention are still lacking, which may lead to variability in therapeutic efficacy and safety among KOA patients. While Recent studies have investigated the effects of different MSC sources and dosages [25, 58], our focus is on comparing the efficacy and safety between different injection frequencies (single vs. repeated injections), aiming to provide evidence for the development of standardized intra-articular MSC intervention protocols.
To the best of our knowledge, this is the first study to compare the efficacy and safety of single versus repeated MSC injections for KOA patients using both traditional and network meta-analysis. A total of 16 randomized controlled trials (RCTs) involving 23 treatment arms and 622 patients were included in this study. Conventional meta-analysis revealed that, compared to baseline, both single and repeated MSC injections significantly improved pain (VAS score) and knee function scores (WOMAC and KOOS), including domains such as function, mobility, stiffness, quality of life, and symptoms at 3, 6, and 12 months. However, no significant differences in efficacy were observed between the single and repeated injection groups in direct comparisons. In the network meta-analysis, we found that repeated MSC injections demonstrated significant advantages over conventional treatments in reducing VAS pain scores at 6 months, WOMAC Function and Stiffness scores at 12 months, and VAS pain scores at 12 months. Among these indicators, SUCRA rankings showed that repeated MSC injections provided the most favorable therapeutic outcomes. Furthermore, compared to single injections, repeated MSC injections significantly improved WOMAC Function scores at 6 months and total WOMAC scores at 12 months. These findings suggest that, in clinical practice, repeated MSC injections may be more effective in alleviating pain and improving knee joint function in the early to mid-stages of KOA, thereby offering superior therapeutic benefits.
Does a higher dose of MSCs lead to better therapeutic outcomes?
Mechanistically, MSCs and their secreted products (such as exosomes and cytokines) exert chondroprotective and regenerative effects through multiple pathways, including promoting chondrocyte differentiation, reducing apoptosis, enhancing metabolic activity, and modulating immune responses [59, 60]. Therefore, it is anticipated that administering a higher quantity of MSCs may yield enhanced therapeutic benefits. Essentially, strategies involving repeated or high-dose MSC injections can be considered equivalent to the administration of a large quantity of MSCs. In terms of injection frequency, Matas et al. [27] demonstrated that repeated MSC injections (administered every six months) significantly improved knee function and reduced pain compared to a single injection. Another study also found that two intra-articular MSC injections provided better joint stability than a single injection in KOA patients [39]. These findings align with our results, as the network meta-analysis (NMA) showed that repeated MSC injections significantly improved knee function scores at both 6 and 12 months compared to single injections.
In terms of MSC injection dosage, the commonly used dose is approximately 5 × 10⁷ cells [61]. The studies included in our analysis involved injection doses ranging from 1 × 10⁷ to 1 × 10⁸ cells. Preliminary clinical trials have explored the impact of higher MSC doses on patient outcomes. Chahal et al. [62] observed that, over a 12-month follow-up, patients in the high-dose MSC group showed significant reductions in cartilage catabolic biomarkers and MRI-assessed synovitis. Another study with a two-year follow-up reported sustained clinical improvement in the high-dose group [63]. Similarly, Lamo-Espinosa et al. [45] found that patients receiving high-dose MSCs exhibited reduced joint damage. A recent meta-analysis comparing low (0–25 × 10⁶), medium (25–50 × 10⁶), and high (> 50 × 10⁶) cell doses demonstrated that high-dose MSC treatment yielded the most favorable outcomes, particularly in improving early pain and functional impairment [25]. However, a small-sample clinical trial by Pers et al. [64] reported improvements in pain and function with low-dose MSCs, while the high-dose group did not show statistically significant benefits. Due to the non-randomized design and limited sample size, these findings should be interpreted with caution. Overall, current evidence suggests that high-dose MSC therapy may offer superior clinical benefits for KOA patients. Nevertheless, large-scale and long-term randomized controlled trials are still needed to confirm whether patients can consistently benefit from intra-articular injections of large quantities of MSCs.
Safety evaluation
Beyond clinical efficacy, safety assessment is a critical factor in evaluating the clinical applicability of MSC therapies. In terms of post-injection safety, our meta-analysis revealed that both single and repeated MSC injections were associated with a higher incidence of AEs compared to conventional treatments. Network meta-analysis further demonstrated that repeated MSC injections significantly increased the AE rate relative to standard therapy. According to the data, AEs were reported in all treatment arms involving MSCs: 91 cases in the single-injection group and 53 cases in the repeated-injection group. Most AEs were mild and localized, primarily involving post-injection site pain. Importantly, neither intervention showed a significantly increased rate of SAEs. The single-injection group reported six SAEs, while the repeated-injection group reported one. Notably, apart from one case of prepatellar bursitis [44] and two cases of prolonged pain lasting up to four weeks [39], the remaining SAEs were not deemed related to MSC administration. In a separate study, Pers et al. [64] reported a case of unstable angina in the MSC treatment group. Current data suggest that SAEs are more frequently associated with high-dose MSC treatments (> 5 × 10⁷ cells). Therefore, while repeated MSC injections may enhance clinical outcomes, they also appear to increase the risk of injection-related AEs. Striking a balance between therapeutic efficacy and safety remains a critical issue that warrants further investigation.
Strengths and limitations
Our work represents the first comprehensive meta-analysis comparing the efficacy and safety of single and repeated MSC injections using both conventional and network methodologies, providing valuable insights for the clinical standardization of MSC intervention protocols. Firstly, in contrast to previous meta-analyses that focused on different MSC sources, treatment methods, and intervention doses, we comprehensively compared the efficacy and safety data of single versus multiple MSC injections for KOA, summarizing evidence of effectiveness for different injection frequencies. This is crucial for guiding clinical interventions. Secondly, we refined the measurement indicators based on different follow-up time points (3, 6, and 12 months). This time-point division provides further insight into the duration of benefit from single/repeated MSC injections for KOA patients, which was overlooked in previous meta-analyses. Moreover, while prior meta-analyses pointed to a dose-dependent increase in AEs [25], our study found that repeated MSC injections improved KOA outcomes without showing higher rates of AEs or SAEs compared to single injections. This result reflects the risk-benefit balance of multiple MSC injections and provides novel insights for personalized clinical interventions in KOA. For example, repeated MSC injections may benefit moderate to severe KOA patients with limited consideration for AE risks, whereas single injections may be more suitable for early KOA patients seeking short-term symptom relief. Of course, from a cost-effectiveness perspective, repeated injections may lead to higher patient costs, and future clinical applications should fully weigh these factors.
Therefore, this study integrated both traditional and network meta-analyses, overcoming the limitation of insufficient direct comparative evidence between single and repeated MSC injections. It provided a comprehensive comparison of the efficacy and safety of the two intervention strategies in the treatment of KOA. The analysis included a full range of outcome indicators commonly used to assess KOA and evaluated clinical efficacy at 3, 6, and 12 months, offering a thorough assessment of early- to mid-term outcomes for both single and repeated MSC injections. However, several limitations should be acknowledged. First, in order to obtain a sufficient number of comparison arms, the conventional treatment group included a variety of commonly used clinical medications and placebos, which may have introduced heterogeneity and affected the reliability of the pooled results. Second, most of the included RCTs were single-center studies with small sample sizes, lacking large-scale, multicenter trial data. Moreover, there was variability in the source and dosage of MSCs across studies, potentially contributing to significant heterogeneity in the outcomes. Therefore, future research should focus on conducting large-scale, multicenter clinical trials with extended follow-up periods. In addition, further efforts are needed to standardize MSC preparation and intervention protocols to validate the clinical efficacy and safety of MSC-based treatments for KOA.
Conclusion
In this study, both single and repeated MSC injections improved knee joint function and reduced pain in KOA patients across different follow-up time points (3, 6 and 12 months) compared to baseline. Notably, repeated MSC injections demonstrated advantages over single injections and conventional therapies in alleviating early-stage pain and enhancing mid-term knee function. However, given the higher incidence of adverse events observed in the repeated injection group, further large-scale randomized controlled trials with extended follow-up durations are necessary to validate the efficacy and safety of different injection frequencies.
Supplementary Information
Acknowledgements
Not applicable.
Abbreviations
- AD-MSC
Adipose-derived mesenchymal stem cells
- AE
Adverse event
- BMSC
Bone marrow mesenchymal stem cells
- BMI
Body mass index
- CI
Confidence interval
- HA
Hyaluronic acid
- ITT
Intention-to-treat
- I²
I-squared (Heterogeneity Statistic)
- KOA
Knee osteoarthritis
- KOOS
Knee Injury and Osteoarthritis Outcome Score
- MD
Mean difference
- MeSH
Medical Subject Headings
- MSC
Mesenchymal stem cell
- NMA
Network meta-analysis
- OA
Osteoarthritis
- OR
Odds ratio
- PRISMA
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
- RCT
Randomized controlled trial
- SAE
Serious adverse event
- SD
Standard deviation
- SUCRA
Surface under the cumulative ranking
- UC-MSC
Umbilical cord mesenchymal stem cells
- VAS
Visual analogue scale
- WOMAC
Western Ontario and McMaster Universities Osteoarthritis Index
- WORMS
Whole-Organ Magnetic Resonance Imaging Score
Author contributions
Li Deng, Chongyu Zhao and Ying Zhang conceived the study and wrote the manuscript. Li Li and Liming Zhou carried out the data collection and data analysis. Xiaoyu Mu and Haomin Sun assessed the quality of the studies. All authors reviewed the results and approved the final version of the manuscript.
Generative AI in scientific writing
The authors declare that they have not use AI-generated work in this manuscript” in this section.
Funding
Not applicable.
Data availability
The supplementary tables and figures presented in the study are included in the Supplementary Material, further inquiries can be directed to the corresponding author.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Li Deng and Chongyu Zhao are co-authors
Contributor Information
Li Li, Email: 148420420@qq.com.
Ying Zhang, Email: Zying20242024@163.com.
References
- 1.Global regional, national burden of osteoarthritis. 1990–2020 and projections to 2050: a systematic analysis for the global burden of disease study 2021. Lancet Rheumatol. 2023;5(9):e508–22. [DOI] [PMC free article] [PubMed]
- 2.Hao Z, Wang Y, Wang L, Feng Q, Li H, Chen T, et al. Burden evaluation and prediction of osteoarthritis and site-specific osteoarthritis coupled with attributable risk factors in China from 1990 to 2030. Clin Rheumatol. 2024;43(6):2061–77. [DOI] [PubMed] [Google Scholar]
- 3.Singh A, Das S, Chopra A, Danda D, Paul BJ, March L, et al. Burden of osteoarthritis in India and its states, 1990–2019: findings from the global burden of disease study 2019. Osteoarthritis Cartilage. 2022;30(8):1070–8. [DOI] [PubMed] [Google Scholar]
- 4.Katz JN, Arant KR, Loeser RF. Diagnosis and treatment of hip and knee osteoarthritis: a review. JAMA. 2021;325(6):568–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Doyle EC, Wragg NM, Wilson SL. Intraarticular injection of bone marrow-derived mesenchymal stem cells enhances regeneration in knee osteoarthritis. Knee Surg Sports Traumatol Arthrosc. 2020;28(12):3827–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cao P, Li Y, Tang Y, Ding C, Hunter DJ. Pharmacotherapy for knee osteoarthritis: current and emerging therapies. Expert Opin Pharmacother. 2020;21(7):797–809. [DOI] [PubMed] [Google Scholar]
- 7.Martel-Pelletier J, Maheu E, Pelletier JP, Alekseeva L, Mkinsi O, Branco J, et al. A new decision tree for diagnosis of osteoarthritis in primary care: international consensus of experts. Aging Clin Exp Res. 2019;31(1):19–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Murphy LB, Cisternas MG, Pasta DJ, Helmick CG, Yelin EH. Medical expenditures and earnings losses among US adults with arthritis in 2013. Arthritis Care Res (Hoboken). 2018;70(6):869–76. [DOI] [PubMed] [Google Scholar]
- 9.Aratikatla A, Maffulli N, Gupta M, Potti IA, Potty AG, Gupta A. Wharton’s jelly and osteoarthritis of the knee. Br Med Bull. 2024;149(1):13–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sharma L. Osteoarthritis of the knee. N Engl J Med. 2021;384(1):51–9. [DOI] [PubMed] [Google Scholar]
- 11.Billesberger LM, Fisher KM, Qadri YJ, Boortz-Marx RL. Procedural treatments for knee osteoarthritis: A review of current injectable therapies. Pain Res Manag. 2020;2020:3873098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rodríguez-Merchán EC. Intraarticular injections of mesenchymal stem cells in knee osteoarthritis: A review of their current molecular mechanisms of action and their efficacy. Int J Mol Sci. 2022. 10.3390/ijms232314953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Makris EA, Gomoll AH, Malizos KN, Hu JC, Athanasiou KA. Repair and tissue engineering techniques for articular cartilage. Nat Rev Rheumatol. 2015;11(1):21–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lubis AM, Lubis VK. Adult bone marrow stem cells in cartilage therapy. Acta Med Indones. 2012;44(1):62–8. [PubMed] [Google Scholar]
- 15.DeFrance MJ, Scuderi GR. Are 20% of patients actually dissatisfied following total knee arthroplasty?? A systematic review of the literature. J Arthroplasty. 2023;38(3):594–9. [DOI] [PubMed] [Google Scholar]
- 16.Pers YM, Ruiz M, Noël D, Jorgensen C. Mesenchymal stem cells for the management of inflammation in osteoarthritis: state of the art and perspectives. Osteoarthritis Cartilage. 2015;23(11):2027–35. [DOI] [PubMed] [Google Scholar]
- 17.Zha K, Li X, Yang Z, Tian G, Sun Z, Sui X, et al. Heterogeneity of mesenchymal stem cells in cartilage regeneration: from characterization to application. NPJ Regen Med. 2021;6(1):14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Song Y, Zhang J, Xu H, Lin Z, Chang H, Liu W, et al. Mesenchymal stem cells in knee osteoarthritis treatment: A systematic review and meta-analysis. J Orthop Translat. 2020;24:121–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mamidi MK, Das AK, Zakaria Z, Bhonde R. Mesenchymal stromal cells for cartilage repair in osteoarthritis. Osteoarthritis Cartilage. 2016;24(8):1307–16. [DOI] [PubMed] [Google Scholar]
- 20.Caplan AI. Adult mesenchymal stem cells for tissue engineering versus regenerative medicine. J Cell Physiol. 2007;213(2):341–7. [DOI] [PubMed] [Google Scholar]
- 21.Wei P, Bao R. Intra-articular mesenchymal stem cell injection for knee osteoarthritis: mechanisms and clinical evidence. Int J Mol Sci. 2022. 10.3390/ijms24010059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Khalifeh Soltani S, Forogh B, Ahmadbeigi N, Hadizadeh Kharazi H, Fallahzadeh K, Kashani L, et al. Safety and efficacy of allogenic placental mesenchymal stem cells for treating knee osteoarthritis: a pilot study. Cytotherapy. 2019;21(1):54–63. [DOI] [PubMed] [Google Scholar]
- 23.Soler R, Orozco L, Munar A, Huguet M, López R, Vives J, et al. Final results of a phase I-II trial using ex vivo expanded autologous mesenchymal stromal cells for the treatment of osteoarthritis of the knee confirming safety and suggesting cartilage regeneration. Knee. 2016;23(4):647–54. [DOI] [PubMed] [Google Scholar]
- 24.Han X, Yang B, Zou F, Sun J. Clinical therapeutic efficacy of mesenchymal stem cells derived from adipose or bone marrow for knee osteoarthritis: a meta-analysis of randomized controlled trials. J Comp Eff Res. 2020;9(5):361–74. [DOI] [PubMed] [Google Scholar]
- 25.Huang Z, Zhang S, Cao M, Lin Z, Kong L, Wu X, et al. What is the optimal dose of adipose-derived mesenchymal stem cells treatment for knee osteoarthritis? A conventional and network meta-analysis of randomized controlled trials. Stem Cell Res Ther. 2023;14(1):245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mahmoud EE, Adachi N, Mawas AS, Deie M, Ochi M. Multiple intra-articular injections of allogeneic bone marrow-derived stem cells potentially improve knee lesions resulting from surgically induced osteoarthritis: an animal study. Bone Joint J. 2019;101–b(7):824–31. [DOI] [PubMed]
- 27.Matas J, Orrego M, Amenabar D, Infante C, Tapia-Limonchi R, Cadiz MI, et al. Umbilical cord-derived mesenchymal stromal cells (MSCs) for knee osteoarthritis: repeated MSC dosing is superior to a single MSC dose and to hyaluronic acid in a controlled randomized phase I/II trial. Stem Cells Transl Med. 2019;8(3):215–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Int J Surg. 2021;88:105906. [DOI] [PubMed] [Google Scholar]
- 29.Brignardello-Petersen R, Izcovich A, Rochwerg B, Florez ID, Hazlewood G, Alhazanni W, et al. Grade approach to drawing conclusions from a network meta-analysis using a partially contextualised framework. BMJ. 2020;371:m3907. [DOI] [PubMed] [Google Scholar]
- 30.Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. J Clin Epidemiol. 2018;93:36–44. [DOI] [PubMed] [Google Scholar]
- 31.Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading quality of evidence and strength of recommendations. BMJ. 2004;328(7454):1490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Puhan MA, Schünemann HJ, Murad MH, Li T, Brignardello-Petersen R, Singh JA, et al. A grade working group approach for rating the quality of treatment effect estimates from network meta-analysis. BMJ. 2014;349:g5630. [DOI] [PubMed] [Google Scholar]
- 33.Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics. 2000;56(2):455–63. [DOI] [PubMed] [Google Scholar]
- 34.Dias S, Welton NJ, Caldwell DM, Ades AE. Checking consistency in mixed treatment comparison meta-analysis. Stat Med. 2010;29(7–8):932–44. [DOI] [PubMed] [Google Scholar]
- 35.Rücker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Med Res Methodol. 2015;15:58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Bastos R, Mathias M, Andrade R, Amaral R, Schott V, Balduino A, et al. Intra-articular injection of culture-expanded mesenchymal stem cells with or without addition of platelet-rich plasma is effective in decreasing pain and symptoms in knee osteoarthritis: a controlled, double-blind clinical trial. Knee Surg Sports Traumatol Arthrosc. 2020;28(6):1989–99. [DOI] [PubMed] [Google Scholar]
- 37.Chen CF, Hu CC, Wu CT, Wu HH, Chang CS, Hung YP, et al. Treatment of knee osteoarthritis with intra-articular injection of allogeneic adipose-derived stem cells (ADSCs) ELIXCYTE®: a phase I/II, randomized, active-control, single-blind, multiple-center clinical trial. Stem Cell Res Ther. 2021;12(1):562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Emadedin M, Labibzadeh N, Liastani MG, Karimi A, Jaroughi N, Bolurieh T, et al. Intra-articular implantation of autologous bone marrow-derived mesenchymal stromal cells to treat knee osteoarthritis: a randomized, triple-blind, placebo-controlled phase 1/2 clinical trial. Cytotherapy. 2018;20(10):1238–46. [DOI] [PubMed] [Google Scholar]
- 39.Freitag J, Bates D, Wickham J, Shah K, Huguenin L, Tenen A, et al. Adipose-derived mesenchymal stem cell therapy in the treatment of knee osteoarthritis: a randomized controlled trial. Regen Med. 2019;14(3):213–30. [DOI] [PubMed] [Google Scholar]
- 40.Garay-Mendoza D, Villarreal-Martínez L, Garza-Bedolla A, Pérez-Garza DM, Acosta-Olivo C, Vilchez-Cavazos F, et al. The effect of intra-articular injection of autologous bone marrow stem cells on pain and knee function in patients with osteoarthritis. Int J Rheum Dis. 2018;21(1):140–7. [DOI] [PubMed] [Google Scholar]
- 41.Garza JR, Campbell RE, Tjoumakaris FP, Freedman KB, Miller LS, Santa Maria D, et al. Clinical efficacy of intra-articular mesenchymal stromal cells for the treatment of knee osteoarthritis: a double-blinded prospective randomized controlled clinical trial. Am J Sports Med. 2020;48(3):588–98. [DOI] [PubMed] [Google Scholar]
- 42.Goncars V, Jakobsons E, Blums K, Briede I, Patetko L, Erglis K, et al. The comparison of knee osteoarthritis treatment with single-dose bone marrow-derived mononuclear cells vs. hyaluronic acid injections. Medicina (B Aires). 2017;53(2):101–8. [DOI] [PubMed] [Google Scholar]
- 43.Gupta PK, Chullikana A, Rengasamy M, Shetty N, Pandey V, Agarwal V, et al. Efficacy and safety of adult human bone marrow-derived, cultured, pooled, allogeneic mesenchymal stromal cells (Stempeucel®): preclinical and clinical trial in osteoarthritis of the knee joint. Arthritis Res Ther. 2016;18(1):301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kuah D, Sivell S, Longworth T, James K, Guermazi A, Cicuttini F, et al. Safety, tolerability and efficacy of intra-articular progenza in knee osteoarthritis: a randomized double-blind placebo-controlled single ascending dose study. J Transl Med. 2018;16(1):49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Lamo-Espinosa JM, Mora G, Blanco JF, Granero-Moltó F, Nuñez-Córdoba JM, Sánchez-Echenique C, et al. Intra-articular injection of two different doses of autologous bone marrow mesenchymal stem cells versus hyaluronic acid in the treatment of knee osteoarthritis: multicenter randomized controlled clinical trial (phase I/II). J Transl Med. 2016;14(1):246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Lamo-Espinosa JM, Mora G, Blanco JF, Granero-Moltó F, Núñez-Córdoba JM, López-Elío S, et al. Intra-articular injection of two different doses of autologous bone marrow mesenchymal stem cells versus hyaluronic acid in the treatment of knee osteoarthritis: long-term follow up of a multicenter randomized controlled clinical trial (phase I/II). J Transl Med. 2018;16(1):213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Lee WS, Kim HJ, Kim KI, Kim GB, Jin W. Intra-articular injection of autologous adipose tissue-derived mesenchymal stem cells for the treatment of knee osteoarthritis: a phase IIb, randomized, placebo-controlled clinical trial. Stem Cells Transl Med. 2019;8(6):504–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lee BW, Lee JJ, Jung JY, Ju JH. Intra-articular injection of human bone marrow-derived mesenchymal stem cells in knee osteoarthritis: a randomized, double-blind, controlled trial. Cell Transplant. 2025;34:9636897241303275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Lu L, Dai C, Zhang Z, Du H, Li S, Ye P, et al. Treatment of knee osteoarthritis with intra-articular injection of autologous adipose-derived mesenchymal progenitor cells: a prospective, randomized, double-blind, active-controlled, phase IIb clinical trial. Stem Cell Res Ther. 2019;10(1):143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Vega A, Martín-Ferrero MA, Del Canto F, Alberca M, García V, Munar A, et al. Treatment of knee osteoarthritis with allogeneic bone marrow mesenchymal stem cells: a randomized controlled trial. Transplantation. 2015;99(8):1681–90. [DOI] [PubMed] [Google Scholar]
- 51.Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Walker LC, Clement ND, Deehan DJ. Predicting the outcome of total knee arthroplasty using the WOMAC score: a review of the literature. J Knee Surg. 2019;32(8):736–41. [DOI] [PubMed] [Google Scholar]
- 53.Roos EM, Lohmander LS. The knee injury and osteoarthritis outcome score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 2003;1:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Felson DT, Lynch J, Guermazi A, Roemer FW, Niu J, McAlindon T, et al. Comparison of BLOKS and WORMS scoring systems part II. Longitudinal assessment of knee MRIs for osteoarthritis and suggested approach based on their performance: data from the osteoarthritis initiative. Osteoarthritis Cartilage. 2010;18(11):1402–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Gelber AC. Knee osteoarthritis. Ann Intern Med. 2024;177(9):Itc129–44. [DOI] [PubMed] [Google Scholar]
- 56.Jones IA, Togashi R, Wilson ML, Heckmann N, Vangsness CT Jr. Intra-articular treatment options for knee osteoarthritis. Nat Rev Rheumatol. 2019;15(2):77–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Mazor M, Lespessailles E, Coursier R, Daniellou R, Best TM, Toumi H. Mesenchymal stem-cell potential in cartilage repair: an update. J Cell Mol Med. 2014;18(12):2340–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Chen X, Zheng J, Yin L, Li Y, Liu H. Transplantation of three mesenchymal stem cells for knee osteoarthritis, which cell and type are more beneficial? A systematic review and network meta-analysis. J Orthop Surg Res. 2024;19(1):366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Barry F. Msc therapy for osteoarthritis: an unfinished story. J Orthop Res. 2019;37(6):1229–35. [DOI] [PubMed] [Google Scholar]
- 60.Dickinson SC, Sutton CA, Brady K, Salerno A, Katopodi T, Williams RL, et al. The Wnt5a receptor, receptor tyrosine kinase-like orphan receptor 2, is a predictive cell surface marker of human mesenchymal stem cells with an enhanced capacity for chondrogenic differentiation. Stem Cells. 2017;35(11):2280–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Xiang XN, Zhu SY, He HC, Yu X, Xu Y, He CQ. Mesenchymal stromal cell-based therapy for cartilage regeneration in knee osteoarthritis. Stem Cell Res Ther. 2022;13(1):14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Chahal J, Gómez-Aristizábal A, Shestopaloff K, Bhatt S, Chaboureau A, Fazio A, et al. Bone marrow mesenchymal stromal cell treatment in patients with osteoarthritis results in overall improvement in pain and symptoms and reduces synovial inflammation. Stem Cells Transl Med. 2019;8(8):746–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Jo CH, Chai JW, Jeong EC, Oh S, Shin JS, Shim H, et al. Intra-articular injection of mesenchymal stem cells for the treatment of osteoarthritis of the knee: a 2-year follow-up study. Am J Sports Med. 2017;45(12):2774–83. [DOI] [PubMed] [Google Scholar]
- 64.Pers YM, Rackwitz L, Ferreira R, Pullig O, Delfour C, Barry F, et al. Adipose mesenchymal stromal cell-based therapy for severe osteoarthritis of the knee: a phase i dose-escalation trial. Stem Cells Transl Med. 2016;5(7):847–56. [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
The supplementary tables and figures presented in the study are included in the Supplementary Material, further inquiries can be directed to the corresponding author.