Autologous Micro-fragmented Adipose Tissue in Patients With Knee Osteoarthritis: A Systematic Review and Meta-analysis


 Background: Microfragmented adipose tissue (MFAT)-containing mesenchymal stem cells (MSCs) combined with surgery is a promising strategy for the early management of knee osteoarthritis (KOA). This study aimed to explore the efficacy and safety of autologous MFAT-MSCs for the management of knee KOA.Methods: PubMed, Embase, the Cochrane Library, and Web of Science for potentially eligible studies published up to June 2021. The primary outcome was the Knee injury and Osteoarthritis Outcome Score (KOOS). The secondary outcomes were pain assessed by visual analog scale (VAS)/numeric rating scale (NRS), quality of life (QOL) (apart from the KOOS), and adverse events (AEs). The random-effects model was used in all analyses.Results: Eight studies (331 patients) were included. The mean differences in KOOS scores between pre-operation and post-operation (mean, 95%CI) were 22.1 (18.7, 25.3), 19.5 (15.4, 23.6), 23.0 (19.0, 26.9), 30.8 (25.5, 35.8), and 29.9 (24.8, 35.0) for pain, symptoms, ADL, sports/recreation, and QOL, respectively. The mean differences in pain VAS between pre-operation and post-operation were -3.026 (-3.884, -2.202). The mean differences in QOL between pre-operation and post-operation (mean, 95%CI) were -25.10 (-29.95, -20.20), 0.039 (-0.079, 0.170), and 0.33 (-0.99, 1.6) for the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), EQ-5D, and University of California in Los Angeles (UCLA), respectively. The use of MFAT-MSCs was not associated with bruising, bleeding, hematoma, drainage, infection, and swelling but was associated with soreness, pain, and stiffness.Conclusions: MFAT-MSC has potential benefits for KOA while being safe. A long-term follow-up and randomized controlled trials are necessary for confirmation.

Osteoarthritis Outcome Score (KOOS), which includes ve domains: 1) pain, 2) symptoms, 3) activities of daily living (ADL), 4) sports/recreation, 5) and QOL. The continuous variables in the form of mean ± standard deviation (SD) at the last follow-up were extracted. The secondary outcomes were pain assessed by visual analog scale (VAS) or numeric rating scale (NRS), QOL (apart from the KOOS), and AEs. Disagreements in data extraction were solved by discussion until reaching a consensus. If needed, a third investigator was invited to the discussion.

Quality of the evidence
The level of evidence of all articles was assessed independently by two authors (** and **) according to the methodological index for non-randomized studies (MINORS) (21). Discrepancies in the assessment were resolved through discussion until a consensus was reached.

Statistical analysis
The Bayesian hierarchical (random-effects) meta-analysis model was used to analyze the scale score changes in the eligible studies. The meta-analysis was performed using a random-effects model because a high degree of between-study heterogeneity was expected. Posterior distributions were obtained after the input of prior distribution (mean=0, SD=100) to the Bayesian meta-analysis. Since the scale score is a continuous variable, the effect size (µ), standard mean difference (SMD), and 95% credibility interval (95%CrI) (i.e., the Bayesian analog of frequentist con dence intervals) for each dimension in the experimental group were estimated, as well as heterogeneity (τ). Sensitivity analysis was discussed by adjusting half-normal (HN) prior distributions, using the scale parameters 1.0. [3.4]. Bayesmeta and metafor packages in R 4.0.3 were used for data analysis. Patient satisfaction was analyzed using the command "metaprop" in R 4.0.3. Pooled forest plots were presented for all outcomes. Cochrane's Q-test and the I 2 statistic were determined to assess heterogeneity, with Q-test P<0.10 or I 2 > 50% indicating signi cant heterogeneity. Two-sided P-values <0.05 were considered statistically signi cant. The possible publication bias was not assessed using funnel plots and Egger's test because the numbers of included studies were <10 in all analyses, in which case the funnel plots and Egger's test could yield misleading results (22).

Results
Study selection Figure 1 presents the study selection process. The initial search identi ed 105 records, but 54 duplicates and 21 records marked as ineligible by automation tools were excluded before the screening. Then, 29 records were screened, and 14 were excluded. Fifteen reports were sought for retrieval, but three could not be retrieved. Twelve reports were assessed for eligibility; two were excluded because they were different reports about the same studies, and two because the outcomes did not match the preselected ones. Finally, eight studies were included.

AEs
The use of MFAT-MSCs was not associated with bruising, bleeding, hematoma, drainage, infection, and swelling but was associated with soreness, pain, and stiffness ( Figure 5).

Discussion
MFAT-MSCs combined with surgery is a promising strategy for the early management of knee OA, but strong evidence is lacking. This meta-analysis aimed to explore the e cacy and safety of autologous MFAT-MSCs for the management of knee OA. The results suggest that MFAT-MSC has potential bene ts for knee OA while being safe. A long-term follow-up and randomized controlled trials are necessary for con rmation.
This meta-analysis showed that using MFAT-MSCs after surgery could improve the ve subscores of the KOOS, pain scores, and QOL scores. It is consistent with the fact that all wight included studies reported some degree of improvements after the use of MFAT-MSCs in their patients with knee OA (16, [23][24][25][26][27][28][29]. Still, because all included studies have positive results, a publication bias is possible but could not be examined in the present study because of the small number of included studies (22). Nevertheless, other studies that were not eligible to the present meta-analysis support the results of the present study, suggesting that the use of MFAT-MSCs is promising for knee OA (30,31). MFAT-MSCs also showed bene ts in other indications, such as menopausal vaginal atrophy, perianal stula repair, and diabetic foot (27,(32)(33)(34). MFAT-MSCs can reduce in ammation (35), increase the proliferation of cells involved in tissue repair (36), and increase tissue regeneration and repair (30,37). MFAT-MSCs secrete the placental growth factor, hepatocyte growth factor, angiogenin, platelet-derived growth factor, and interleukins-13, -3, -16, and -27, all acting together to enhance tissue repair (37)(38)(39). Still, the exact action mechanisms are being studied (31).
Classical management of knee OA includes corticosteroid injection, but an in ammatory are occurs in 2%-25% of the cases (2). Hyaluronic acid injection carries a risk of ares and granulomatous in ammation (40). In the present meta-analysis, the meta-analysis of AEs suggests soreness, pain, and stiffness after MFAT-MSCs injection. Still, it has to be noted that these AEs were rare and only reported by a small number of studies. In addition, whether the safety pro le of MFAT-MSCs is better than with other therapies remains to be investigated in comparative trials.
Four studies included outcomes that could not be pooled in the present meta-analysis. These studies showed improvements in the Tegner Lysholm Knee score, International Knee Documentation Committee (IKDC) score, and Oxford Knee Score (OKS), but not in the Emory Quality of Life (EQOL) score and delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC) index (23,25,28,29). Of course, different assessments methods have different degrees of subjectivity/objectivity and measure different outcomes. Therefore, it might highlight the need to use multiple assessment tools and to use similar tools among studies to improve the comparability of the results.
This meta-analysis has limitations. The number of included studies was small because the MFAT-MSC strategy is relatively novel. There was no comparator since the treatment is relatively new and randomized controlled trials are lacking. Heterogeneity was high because of the different patient populations, devices, and techniques.

Conclusion
In conclusion, MFAT-MSCs and surgery is probably a simple, sustainable, quick, and minimally invasive strategy for managing knee OA, with bene ts and few AEs. Studies with long-term follow-up and randomized controlled trials with a large number of patients are needed to draw de nitive conclusions and enlarge the indications of MFAT-MSCs.       Forest plot of adverse events (AEs).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.