Mesenchymal Stem Cells in the Treatment of Cartilage Defects of the Knee: A Systematic Review of the Clinical Outcomes

Abstract

Background:

Osteochondral lesions are a common clinical problem and their management has been historically challenging. Mesenchymal stem cells have the potential to differentiate into chondrocytes and thus restore hyaline cartilage to the defect, theoretically improving clincal outcomes in these patients. They can also be harvested with minimal donor site morbidity.

Purpose:

To assess the clinical and functional outcomes of mesenchymal stem cell implantation to treat isolated osteochondral defects of the knee. A secondary purpose is to assess the quality of the current available evidence as well as the radiological and histological outcomes. We also reviewed the cellular preparation and operative techniques for implantation.

Study Design:

Systematic review.

Methods:

A comprehensive literature search of 4 databases was carried out: CINAHL, Embase, MEDLINE, and PubMed. We searched for clinical studies reporting the outcomes on a minimum of 5 patients with at least 12 months of follow-up. Clinical, radiological, and histological outcomes were recorded. We also recorded demographics, stem cell source, culture technique, and operative technique. Methodological quality of each study was assessed using the modified Coleman methodology score, and risk of bias for the randomized controlled studies was assessed using the Cochrane Collaboration tool.

Results:

Seventeen studies were found, encompassing 367 patients. The mean patient age was 35.1 years. Bone marrow was the most common source of stem cells utilized. Mesenchymal stem cell therapy consistently demonstrated good short- to medium-term outcomes in the studies reviewed with no serious adverse events being recorded. There was significant heterogeneity in cell harvesting and preparation as well as in the reporting of outcomes.

Conclusion:

Mesenchymal stem cells demonstrated a clinically relevant improvement in outcomes in patients with osteochondral defects of the knee. More research is needed to establish an optimal treatment protocol, long-term outcomes, and superiority over other therapies.

Registration:

CRD42020179391 (PROSPERO).

Keywords

mesenchymal stem cells osteochondral cartilage defect knee

Osteochondral lesions are common clinical problem encountered in orthopaedic practice, and their etiology remains unknown.³⁸ The incidence is unclear largely because of the number of asymptomatic lesions thought to be present in the population. Estimates of prevalence^2,6 have varied between 5% and 10%; however, some studies place the incidence at around 35% to 60%.^12,17,27,44 Osteochondral lesions present a particular challenge in view of the limited restorative potential of articular cartilage given the avascular nature of the tissue. The condition commonly affects adolescents and young adults, causing pain, swelling, and activity restriction. If left untreated, it can progress to osteoarthritis of the knee,² leading to a large health burden on the patients and health services at large.

With the lack of success in nonoperative management of symptomatic lesions, a plethora of surgical interventions has been utilized, aimed at restoring the articular cartilage. These include microfracture, autologous chondrocyte implantation (ACI), osteochondral autograft, or osteochondral allograft.⁴⁴

Microfracture attempts to restore cartilage by exposing the articular surface to local mesenchymal stem cells (MSCs).¹⁹ Although it is a relatively inexpensive and minimally invasive technique, denser and less stiff fibrocartilage is formed instead of hyaline cartilage. The long-term outcomes have therefore been disputed.⁴⁷ ACI is often utilized for larger lesions (>2 cm²) and has been shown to have good long-term outcomes.^33,43,46 However, this is a 2-stage procedure, and the differentiation of chondrocytes to a fibroblast-like phenotype is a potential drawback. Osteochondral autograft transplantation is a technique where hyaline cartilage from a nonweightbearing articular surface is transplanted to the defect. The procedure has been shown to produce favorable outcomes in osteochondral defects of the patella and femoral condyles.⁴⁸

MSCs as a therapy could overcome these disadvantages. They can be harvested with minimal donor morbidity from multiple sources (bone marrow, blood, synovium, and adipose tissue) and injected into the site of the defect as a single-stage procedure.⁴⁴ MSCs have 2 qualities that make them suitable for cartilage repair. The first is their potential to differentiate into chondrocytes to fill the defect.⁶² The second is their capacity to exert paracrine effects on the surrounding extracellular matrix to mobilize existing chondrocytes.¹⁰ Numerous studies have shown highly promising results in animal models.⁴ However, the evidence for clinical effectiveness is more limited.⁵ MSCs are also currently being explored in treatments for osteoarthritis,^22,31 but human studies focusing on isolated cartilage lesions are fewer in number.

The primary purpose to this review is to assess the clinical outcomes of patients with articular cartilage defects of the knee treated with MSC implantation. Secondary goals are to examine protocols for treatment as well as the histological and radiological outcomes.

Methods

Inclusion Criteria

The articles were screened to check that they met the following criteria: level of evidence 1 to 4, use of mesenchymal or adipose-derived stem cells on the knee, reporting of clinical/functional outcomes, minimum of 5 patients, minimum of 12-month follow-up, and English language. In cases where there were overlapping cohorts between studies, the study with the longer follow-up was included. All animal studies and studies on patients with osteoarthritis of the knee were excluded.

Search Strategy

A comprehensive search strategy was carried out using the NICE Healthcare Databases Advanced Search (National Institute of Health and Clinical Excellence) of 4 databases between January 1, 2010, and February 1, 2020: CINAHL, Embase, MEDLINE, and PubMed. We used the following search strategy: (Osteochondral OR Chond* OR “Cartilage defect”)ti.ab AND Knee AND (“Stem cells” OR MSCs OR ADSCs OR Adipose derived stem cells” OR “Bone Marrow derived stem cells”). This yielded 1240 results across the 4 databases; this was reduced to 914 after duplicates were removed. The titles and abstracts were screened by the first 2 authors (M.J. and R.J.), and 42 potentially relevant articles were found. Of the remaining studies, the full manuscripts were reviewed by the first 2 authors to ascertain whether the inclusion criteria had been met. If there was disagreement between the first 2 authors regarding whether a study should be included, the matter was referred to the most senior author (A.V.) for a decision. After this process, 15 studies were selected for inclusion in the review (see Figure 1). The references of these manuscripts were then independently screened by the first 2 authors, searching for any other relevant studies. Two further studies were found by this method, bringing the final total to 17 studies. The review was prospectively registered with PROSPERO (CRD42020179391).

Data Extraction

Once the studies were selected, the manuscripts were reviewed independently by the first 2 authors. The primary objective was to collect data on the clinical and functional outcomes at the end of the follow-up period. Radiological and histological outcomes were also collected if available. Patient demographics, number of cells injected, stem cell source, operative technique, adverse events, treatment failure, and details of concomitant therapies were also recorded where available. The 17 articles encompassed 367 patients.

Risk-of-Bias Assessment

The randomized controlled trials found in this review were methodically assessed for bias by the first 2 authors independently using the revised Cochrane risk-of-bias tool.⁵⁷ The following 5 domains were assessed for low risk of bias, some concerns of bias, or high risk of bias: (1) risk of bias arising from the randomization process, (2) risk of bias attributed to deviations from intended intervention, (3) risk of bias attributed to missing outcome data, (4) risk of bias in measurement of outcome, and (5) risk of bias in selection of reported results.

Methodological Quality Assessment

Two authors (M.J and R.J.) independently reviewed and scored the methodological quality of each study using the modified Coleman methodology score (Table 1). Each author used the tool to assign a score out of 100 for each study. If there was disagreement between the authors, then the senior author (A.V.) was consulted, and disagreement was resolved by consensus. A score of 100 indicates that a study avoids confounding factors and biases.¹⁵

Table 1

Modified Coleman Methodology Score

Part A^a	Score
1. Study size: No. of patients
<30	0
31-50	4
51-100	7
>100	10
2. Mean follow-up, mo
<12	0
12-36	4
37-60	7
>60	10
3. Surgical approach
Different approach used and outcome not reported separately	0
Different approaches used and outcome reported separately	7
Single approach used	10
4. Type of study
Retrospective cohort study	0
Prospective cohort study	10
Randomized controlled trial	15
5. Description of diagnosis
Described without percentage specified	0
Described with percentage specified	5
6. Descriptions of surgical technique
Inadequate (not stated, technique unclear)	0
Fair (technique only stated)	5
Adequate (technique stated, details of surgical procedure given)	10
7. Description of postoperative rehabilitation
Not described	0
Described	5
Part B^b
1. Outcome criteria
Outcome measures clearly defined	2
Timing of outcome clearly stated	2
Use of outcome criteria that has reported reliability	3
General health outcome measure included	3
2. Procedure of assessing outcomes
Participants recruited	5
Investigator independent of surgeon	4
Written assessment	3
Completion of assessments by patients themselves with minimal investigator assistance	3
3. Description of patient selection process
Selection criteria reported and unbiased	5
Recruitment rate reported
<90%	0
>90%	5

Only 1 score is given for each of the 7 sections.

Scores may be given for each option in the 3 sections, if applicable.

Results

Study Characteristics

Seventeen studies met the inclusion criteria (Table 2). The 17 studies included 5 randomized controlled trials, 2 cohort studies, 2 case-control studies, and 8 case series. Of the case series, 1 article reported the preliminary results of a pilot study. The follow-up length varied significantly in the studies. Three studies had 12 months of follow-up; 1 study, 18 months; 5 studies, 2 years; 3 studies, 3 years; 2 studies, 5 years; and 2 studies, 8 and 9 years, respectively. Finally, 1 study followed up patients for 10 years. Given the wide variation in follow-up length, there was a significant interval between the first and last patients being treated in this review. The earliest cases were treated by Teo et al⁵⁸ in 2001; they followed their patients for an average of 10 years. The most recent cases were carried out in 2016 by Shimomura et al⁵³; they had a 2-year follow-up period.

Table 2

Study Details and Follow-up Length^a

Study	MSC Source	Concomitant Treatment	Mean Follow-up, mo
Skowroński⁵⁵	Bone marrow and peripheral blood	None recorded	60
Monckeberg⁴¹	Peripheral blood	Microfracture and platelet-rich plasma	60
Sekiya⁵¹	Synovium	ACL reconstruction or meniscal sutures	36
Koh³²	Adipose tissue	Microfracture	24
Kamei²⁹	Bone marrow	Microfracture	12
Shimomura⁵³	Synovium	None recorded	24
de Windt¹⁴	Bone marrow	None recorded	18
Teo⁵⁸	Bone marrow	5 patellar realignment, 5 high tibial osteotomies, 1 partial meniscectomy, 1 ACL	120
Kyriakidis³⁴	Adipose tissue	None recorded	36
Hashimoto²⁵	Bone marrow	Microfracture	12
Buda⁹	Bone marrow	Platelet-rich plasma	36
Pipino⁴⁷	Adipose tissue	Microfracture (all patients), 6 partial meniscectomies, 1 meniscal repair, 1 ACL reconstruction, 3 high tibial osteotomies	24
Gobbi²¹	Bone marrow	None recorded	100
de Girolamo¹³	Bone marrow	1 tibial tubercle osteotomy	108
Saw⁵⁰	Peripheral blood	None recorded	24
Haleem²³	Bone marrow	None recorded	16
Akgun¹	Synovium	None recorded	24

ACL, anterior cruciate ligament; MSC, mesenchymal stem cell.

Participant Characteristics

A total of 367 patients were enrolled, with an average age of 35.1 years (range, 16-70 years). There were 207 male patients and 157 female patients; 1 study did not report the sex for 3 of its 5 patients.²⁹ Trials included patients with an isolated cartilage defect, with a mean defect size of 3.3 cm² (range, 0.25-27 cm²). Nine studies required participants to have a lesion with a minimum International Cartilage Regeneration and Joint Preservation Society (ICRS) grade ≥3; 1 study required a minimum ICRS grade of 4. Two studies specified full-thickness lesions in their inclusion criteria. Four studies specified a minimum lesion size of 2 cm²; 1 study, 3 cm²; and 1 study, 4 cm². In 8 studies, the patients had a concomitant intervention. One group underwent injection of platelet-rich plasma (PRP) with bone marrow–derived MSCs, and 1 study used PRP and microdrilling. Microdrilling alone was used as an adjunct in 3 studies (Table 2). There were 8 patients across 2 studies who had concomitant high tibial osteotomy. Ten patients underwent ACL reconstruction, and 12 had a partial meniscectomy.

Risk-of-Bias Assessment

All 5 randomized controlled trials were deemed to be of high risk of bias in at least 1 category of the Cochrane risk-of-bias tool. Allocation concealment was not clearly described by Hashimoto et al²⁵ and Saw et al.⁵⁰ Four of the 5 randomized controlled trials were deemed to have high concerns for bias in measurement of their outcomes relating to the lack of blinding.^1,25,32,50 Saw et al⁵⁰ and de Girolamo et al¹³ had missing outcome data. Akgun et al¹ had a significant number of patients in their randomized groups who were not analyzed (Figure 2).

Methodological Quality Assessment

There was a clear methodological deficit in most of the studies reviewed. The median modified Coleman methodological quality score was 68 (interquartile range, 59.8-73.7; overall range, 57-86). Studies that used uncultured MSC preparations generally were judged to be of higher methodological quality (Figure 3). The primary reasons for low scores were the study size, duration of follow-up, and study type.

Intervention Characteristics

The selection process found 8 studies that used bone marrow as the source of stem cells; 3 studies, adipose-derived stem cells; 3 studies, synovial-derived stem cells; 2 studies, peripheral blood; and 1 study, bone marrow and peripheral blood. Eight studies used cultured MSCs. The mean duration of the cultures, where recorded, was 24 days (range, 7-35 days), with 14 days being the most common duration of culture. Four studies that cultured MSCs recorded 2 or 3 passages of cells. Ten studies reported the number of cells injected, which ranged from 5 × 10⁶ to 5 × 10⁷. Ten studies noted the use of scaffolds when implanting the MSCs. Five of those studies used a hyaluronic acid–based scaffold; 3 used a fibrin-based scaffold; and 1 study used a collagen bilayer matrix (see Table 3).

Table 3

Surgical Techniques and Complications^a

Study	Operative Technique	Cultured (Duration) or Uncultured MSCs	No. of Cells Transplanted	Reported Complications
Skowroński⁵⁵	Mini-arthrotomy via medial parapatellar approach (after arthroscopic assessment of lesion). Subchondral bone drilled and filled with iliac bone graft. Lesion covered with collagen matrix and MSCs injected under membrane	Uncultured	4.5 × 10⁵–2.65 × 10⁶	6 treatment failures: 4 in group treated with bone marrow concentrate and 2 in group treated with peripheral blood–derived stem cells
Monckeberg⁴¹	Arthroscopic: microdrilling and debridement of lesion, followed by injection of 10 mL of peripheral blood–derived stem cells in autologous platelet-rich plasma with clot adherent	Uncultured	Not recorded	2 patients had myalgia and fever from filgrastim
Sekiya⁵¹	Arthroscopic: subsynovial tissue from femur (at suprapatellar pouch) initially harvested and cultured. Cells suspended in Ringer lactate solution and 0.5 mL injected into lesion after arthroscopic debridement	Yes (14 d)	4.7 ± 2.1 × 10⁷	None reported
Koh³²	Arthroscopic debridement, followed by microfracture, followed by injection of cell suspension in fibrin glue. Adipose-derived stem cells harvested from buttock the previous day	Uncultured	4.97 × 10⁶	None reported
Kamei²⁹	Arthroscopic debridement of lesion down to subchondral bone, followed by microfractures. Cells cultured with ferucarbotran. Cell suspension injected into lesion. 1-T magnet used to help accumulate suspension into defect (attracted to ferucarbotran)	Cultured (14 d)	1 × 10⁷	Swelling in 3/5 patients (resolved after 2 wk)
Shimomura⁵³	Initial arthroscopic harvesting for synovial MSCs from anterior part of knee joint and cultured. At implantation, defect exposed by mini-arthrotomy and suspension injected into the defect	Cultured (not recorded)	4 × 10⁵ cells/cm²	None reported
de Windt¹⁴	Mini-arthrotomy and debridement of defect. Cell suspension injected into defect with fibrin glue at 0.9 mL/cm² (cell product/defect)	Cultured (not recorded)	1.5-2 × 10⁶/mL	6 nausea and vomiting, 2 urinary retention, 13 arthralgia, 8 joint swelling, 5 crepitations, 1 sensation of knee giving way, 1 new lesion found on second-look arthroscopy, 10 “miscellaneous”
Teo⁵⁸	Arthrotomy and debridement of lesion. Chondral patch sutured over the defect and bone marrow stem cells injected beneath it	Cultured (35 d)	1.5 × 10⁷	None reported
Kyriakidis³⁴	Arthroscopic evaluation of the lesion, followed by debridement (open or arthroscopic, depending on lesion morphology). Cultured cells then injected with hyaluronic acid matrix	Cultured (not recorded)	Not recorded	None reported
Hashimoto²⁵	Arthroscopic debridement of lesion, followed by microfracture. Prepared MSCs were then injected into site	Cultured (14 d)	1 × 10⁷	No serious adverse events
Buda⁹	60 mL of bone marrow aspirated from iliac crest and treated in operating theater. Lesion arthroscopically debrided and treated with 2 mL of bone marrow aspirate injected with hyaluronic acid scaffold	No	Not recorded	None reported
Pipino⁴⁷	Arthroscopic debridement of lesion, followed by microfracture. Thermogelling with PG/GC system to deliver MSCs to lesion site	Not mentioned	Not recorded	None reported
Gobbi²¹	Arthroscopic assessment of lesion initially. 60 mL of bone marrow aspirated from iliac crest and treated in theater. Lesion approached via mini-arthrotomy. Lesion debrided. Treated aspirate then injected with hyaluronic acid scaffold. Secured with sutures or fibrin glue	Uncultured	Not recorded	None reported
de Girolamo¹³	24 mL of bone marrow harvested from iliac crest and treated in theater. Mini-arthrotomy performed and lesion debrided. Microfracturing performed. Treated bone marrow then introduced into lesion site. Defect covered with collagen matrix, which was secured with fibrin glue	Uncultured	Not recorded	2 cases of implantation failure with 1 patient requiring a total knee arthroplasty during the study period
Saw⁵⁰	Arthroscopic debridement. Microfracture drilling was performed. Cells injected as series of 8 intra-articular injections postoperatively. Hyaluronic acid used as scaffold	Cultured (not recorded)	Not recorded	None reported
Haleem²³	Arthrotomy and debridement of lesion, followed by microfractures. Periosteal flap harvested from distal femur or proximal tibia, used to cover defect, and secured with absorbable sutures. Precultured cells then injected into defect site with platelet-rich fibrin glue	Cultured (14 d)	1.5 × 10⁷	None reported
Akgun¹	Mini-arthrotomy and debridement of defect done to healthy cartilage (subchondral bone not reached). Cells implanted with type I/III collagen membrane and secured with fibrin glue and 3-5 absorbable sutures	Cultured (7 d)	8 × 10⁶	None reported

MSC, mesenchymal stem cell; PG/GC, polyglucosamine and glucosamine carbonate.

Nine studies^§ reported the use of MSC marker sorting before delivery of cells. CD90, CD105, and CD44 were the most commonly used positive markers. The presence of CD14, CD34, and CD105 was used to eliminate other cellular content. Two studies that used peripheral blood as the source of their MSCs used filgrastim to augment the production of MSCs in the blood before harvesting.^41,50 Seven studies carried out the treatment as a 2-stage procedure where the cells were harvested, cultured, and then implanted into the knee. The remaining 10 studies extracted the cells and implanted them in a single stage. Seven studies carried out the debridement of the lesion and implantation of the MSCs via an open approach (arthrotomy or mini-arthrotomy), while the remaining 10 carried out the procedure arthroscopically. Fifteen studies made reference to a postoperative rehabilitation protocol. All these studies included a period of nonweightbearing progressing to a period of partial weightbearing before the patient was allowed to fully bear weight. Seven studies specified 6 weeks before the patient was allowed to fully bear weight; 2 studies, 8 weeks; 1 study, 9 weeks; 1 study, 10 weeks; and 1 study, 12 weeks. The remaining studies did not specify the postoperative weightbearing protocol.

Clinical Outcome Measures

There was variation in the outcome measures utilized in the studies. The Knee injury and Osteoarthritis Outcome Score (KOOS) was the most commonly used measure, in 9 studies. This was followed by the International Knee Documentation Committee (IKDC) score, in 9 studies. The Lysholm scale was reported in 4 studies. The visual analog score (VAS) for pain was used in 7 studies. Five studies reported the Tegner score. One study utilized the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC).

Thirteen studies reported the radiological outcomes of their interventions. Magnetic resonance imaging (MRI) was the most common modality used. MOCART (magnetic resonance observation of cartilage repair tissue) scoring was the most commonly used system to evaluate the success of the repair.³⁹ Seven studies carried out second-look arthroscopies on a proportion of their patients, with 6 of those taking biopsy specimens for histological analysis and grading.

KOOS Results

Three studies used the overall KOOS.^14,47,55 The follow-up interval varied in the 3 studies (range, 18-60 months). However, all 3 reported statistically significant improvement in the KOOS at the end of the follow-up period. Seven studies^{1,13,29,32,34,53,58} cited individual components of the KOOS. Of these, Koh et al³² and Akgun et al¹ demonstrated statistically significant improvement in all aspects of the KOOS at the end of their 2-year follow-up. Kamei et al²⁹ indicated a statistically significant improvement in KOOS–Quality of Life after 12 months. Shimomura et al⁵³ documented a statistically significant improvement in all aspects of the KOOS at 2 years in their series of 5 patients. Kyriakidis et al³⁴ also saw statistically significant improvement in all aspects of the KOOS at 3 years. In the Teo et al⁵⁸ study, patients with lesions >4 cm² tended to fair worse on the Symptoms subset of the KOOS, which assesses swelling, locking, range of movement, and stiffness (P = .035). However, they did not find this correlated with increased pain or reduced physical activity. Similarly, they did not find an association with age and outcomes.

IKDC Score

Seven studies utilized the IKDC score.^{9,21,25,29,41,50,58} All studies cited a significant improvement with variable follow-up (range, 1-10 years). The IKDC score varied among the time points. The mean preoperative IKDC score was 30 in the 6 studies that documented it. Three studies noted that the mean IKDC score at 6 months after treatment was 69.5. The mean IKDC score at 12 months after treatment in the 5 studies that cited it was 80.0 (range, 68.8-85.4). Two studies reported the IKDC score at 18 months and 2 years. At these time points, the mean scores were 80.7 and 78.4. Monckeberg et al⁴¹ found a mean IKDC score of 82.2 at 5 years in their 20 patients. Gobbi and Whyte²¹ noted a mean IKDC score of 85 at a mean follow-up of 8 years. All studies indicated a progressive increase in IKDC score after treatment, with no study citing a decline in mean scores after a particular interval. Teo et al⁵⁸ stated that male patients had better outcomes than female patients over the course of the study (P = .044); they also found no significant correlation between patient age and outcome and no difference in outcomes between bone marrow–derived MSCs and ACI (P = .989). They found no significant relationship between body mass index and outcomes. Gobbi and Whyte²¹ indicated no significant difference in scores when comparing age, lesion size, or an associated osteotomy (P = .16, .63, and .57, respectively).

Lysholm Score

Four studies reported statistically significant improvements in the Lysholm score postintervention in their cohorts.^13,23,51,55 Follow-up varied between 1 and 10 years. De Girolamo et al¹³ noted a slight decline in Lysholm score after a peak at 2 years, when the mean value rose to 96.1. It fell to a mean 91.4 at 5 years and 89.1 at 9 years. Skowroński and Rutka⁵⁵ also noted a slight decline in Lysholm scores between 3 and 5 years. The mean preoperative score across all studies was 57.7. The mean score at 6 months was 75.7 in the 3 studies that cited it. At 1 year, the mean was 91.6. Teo et al⁵⁸ found a similar correlation when using the Lysholm to the IKDC score with male patients tending to have better outcomes (P = .002). In the randomized controlled trial of Koh et al,³² age and sex were not significant factors in outcomes. They also stated that lesion size did not influence outcomes (P = .276).

WOMAC Score

Pipino et al⁴⁷ found a statistically significant improvement in all subsections of the WOMAC score when compared at 2 years versus preoperatively in their series of 46 patients treated with adipose-derived MSCs.

Tegner Score

Three studies reported statistically significant improvement in the self-reported Tegner activity scale^1,21,34 after treatment. Sekiya et al⁵¹ and de Girolamo et al¹³ cited no significant improvement in Tegner scores. Teo et al⁵⁸ reported that male patients were more likely to have improved Tegner scores with treatment. Gobbi and Whyte²¹ found no correlation between lesion size or age and Tegner scores after treatment.

VAS Scores

All studies that reported VAS scores showed significant reduction after treatment^{1,13,14,21,32,53,55}; follow-up ranged from 1 to 9 years. Koh et al³² and Gobbi and Whyte²¹ did not find an association between age or lesion size and VAS scores.

Radiological Outcomes

Seven studies demonstrated significant improvements in MOCART scores from pre- to postoperatively.^{1,9,25,29,32,34,53} Kyriakidis et al³⁴ carried out MRI scans on 20 of the 25 patients enrolled in their study and found no statistically significant difference in MOCART scores between 12 and 24 months postoperatively. They did not include findings of preoperative MRI. Saw et al⁵⁰ and Monckeberg et al⁴¹ demonstrated significant improvements in ICRS morphologic scoring system (MSS) scores after treatment at 18 months and 5 years, respectively. Haleem et al²³ did not score their sample, but 3 of 5 patients had complete filling per MRI 12 months after treatment, with the remaining 2 patients having partially filled defects. Koh et al³² found no association between patient age or lesion size and MOCART score.

Histologic Findings

Six studies performed a second look arthropscopy with biopsy on a portion of their cohorts.^{9,14,32,50,51,53} Sekiya et al⁵¹ performed a biopsy on 4 patients; of these, 1 defect was found to contain fibrous cartilage in the superficial and deep zones. Two patients hadfibrous cartilage superficially but hyaline cartilage in the deep zone. Finally, 1 patient had hyaline cartilage in the superficial and deep zones.⁵¹ Similar histological results were seen in the cohort of Buda et al,⁹ who also biopsied 4 patients. Shimomura et al⁵³ also demonstrated hyaline-type cartilage in the deep zones on all 5 cases they biopsied with fibrocartilage in the superficial zone. Koh et al³² reported better ICRS II scores (P = .036) across their group treated with adipose-derived stem cells and microfractures compared with their group treated with microfractures alone. Saw et al⁵⁰ demonstrated significantly better (P = .022) ICRS II scores in their patients treated with adipose-derived MSCs with hyaluronic acid compared with the group treated with hyaluronic acidalone; they biopsied 16 patients in each group. de Windt et al¹⁴ demonstrated a mean ICRS II score of 70 (range, 46-89).

Complications

Six studies made reference to any adverse events or complications.^{13,14,25,29,41,55} Of these, 2 studies reported a total of 8 implantation failures between them.^13,55 Monckeberg et al⁴¹ reported 2 cases of transient myalgia and pyrexia which were then attributed to filgrastim administration. Two studies reported arthralgia and joint swelling which in all cases resolved^14,29 (Table 3).

Discussion

To the best of our knowledge, this is the largest review assessing the outcomes of MSC therapy for osteochondral lesions. The key finding is that all studies showed a statistically significant improvement in functional outcomes in their patients. This provides evidence of the clinical efficacy of stem cell use in cartilage regeneration. The findings are consistent regardless of the MSC source (synovium, bone marrow, adipose tissue, or peripheral blood). Moreover, no serious adverse events were indicated in any of the studies reviewed, although there was a small number of implantation failures (n = 8). The efficacy of MSCs in treating cartilage damage has been demonstrated in numerous studies on osteoarthritic knees. As early as 2008, Centeno et al¹¹ showed in a case report that intra-articular injection of autologous MSCs regenerated cartilage. This has since been followed up with a number of larger-scale studies.^60,61 Our review has demonstrated that these results can be replicated in knees with isolated cartilage defects as well as osteoarthritis.

Another key matter relates to whether MSCs are superior to other surgical treatments available for cartilage repair. Seven studies in this review compared the outcomes of MSCs against a control group.^{1,13,21,25,32,50,58} Of these, 3 used microfracture^13,25,32 as the control group; the MSC-treated cohort also underwent microfracture treatments before MSC implantation. In all 3 studies, the MSC group had significantly better functional and radiological outcomes than the group treated with microfracture alone. In the de Girolamo et al¹³ study, the improvement in Lysholm scores was maintained by the end of the 100-month follow-up, although the scores did decline at 100 months as compared with 12 months in the group where microfracture and matrix chondrogenesis (autologous matrix-induced chondrogenesis) were augmented with stem cells. This decline was witnessed in another study⁵⁵ and may need to be investigated further. Koh et al³² found better KOOS pain scores in their MSC cohort as compared with microfracture, but improvements in other elements of the KOOS in MSCs compared with microfracture did not cross the statistically significant threshold. In the Koh et al and Hashimoto et al²⁵ studies, the MSC group demonstrated better radiological outcomes at 12 and 24 months, respectively. These results are expected given the theoretical benefits of stem cell therapy. As mentioned earlier, microfracture leads pluripotent stem cells, which differentiate to fill the defect with fibrocartilage as opposed to more biochemically advantageous hyaline cartilage.⁵ Of note, in all 3 studies,^13,25,32 the superiority was shown within 12 months of follow-up. As patients with osteochondral defects carry significant risk of developing osteoarthritis in the future, theoretically there is a long-term cost-benefit advantage to MSCs over microfracture alone. However, no study reviewed had a long-enough follow-up period to demonstrate this.

Two studies^1,42,58 compared MSCs with ACI. Teo et al⁵⁸ had 36 patients in their case series and followed their cohorts over 10 years. They previously published the outcomes for the same cohort of patients at 2 years.⁴² They reported functional outcomes alone and did not assess radiological outcomes. They did take a biopsy specimen in 1 of the 36 patients treated with MSCs at 2 years but did not provide an objective reporting of the tissue quality.^42,58 They found no difference in functional outcomes between the 2 treatment groups, and both therapies were demonstrated to be effective in improving pain and patient function. ACI has been shown to lead to the formation of hyaline cartilage,⁴⁰ which provides an explanation for the similar outcomes as compared with MSCs in this study. However, the study may have also been underpowered to detect a significant difference because of the small number of patients. Akgun et al,¹ in contrast, found significantly better scores in all subscales of the KOOS in patients treated with MSCs versus those treated with ACI after 24 months. Their study was the only one to compare radiological outcomes using the MOCART scoring system. They found better defect filling and less marrow edema in the MSC group. Unfortunately, the number of patients in the study was relatively small (7 in each group); however, this raises the question of whether these superior radiological outcomes would translate to better patient outcomes in the long term.

As mentioned, the MOCART scoring system was the most common scoring system used to judge radiological outcome. This system contains 9 components (Table 4). Three studies quoted the results for the individual components of the scoring system.^32,34,51,53 Sekiya et al⁵¹ and Shimomura et al⁵³ found all their defects completely filled 24 months after MSC implantation. In the Shimomura et al study, all patients had subchondral edema present at 6 months, which disappeared at final follow-up of 12 months. Kyriakidis et al³⁴ scanned 20 of their 25 patients and found 4 cases of incomplete filling at 24 months and 2 cases of subchondral bone edema. Of the studies that compared MSCs with microfracture, de Girolamo et al¹³ and Koh et al³² found better defect filling and integration in those treated with MSCs than those treated with microfracture alone. As with functional outcomes, all studies reported improvement in radiological appearance after treatment.

Table 4

MOCART Scoring System Assessment Criteria^{39

a}

Degree of defect filling: hypertrophy, incomplete, exposed subchondral bone

Integration of border zone: complete vs incomplete

Repair tissue

Surface: intact or damaged

Structure: homogeneous vs inhomogeneous

Signal intensity of repair tissue: isointense, moderately hyperintense, markedly hyperintense

Subchondral

Lamina: intact or nonintact

Bone: intact, edema, granulation tissue/cyst/sclerosis

Adhesions: present vs not present

Synovitis: present vs not present

MOCART, magnetic resonance observation of cartilage repair tissue.

Six studies described biopsy results of a proportion of their cohorts. Koh et al³² found increased type II collagen and proteoglycan staining in the group treated with MSCs and microfracture as compared with microfracture alone. The other 5 studies demonstrated similarly positive histologic findings in their samples.^9,14,42,53 The studies also reported an improvement in subchondral bone edema with MSCs. Other studies in the literature have explained this to be due to the anti-inflammatory effect of MSCs.²⁸

Bone marrow–derived MSCs were the most common source utilized in the review,^{9,14,21,23,25,29,55,58} followed by adipose-derived stem cells,^32,34,47 peripheral blood,^41,50,55 and then synovial-derived stem cells.^1,51,53 Evidence from in vitro studies^49,54 and animal studies⁴⁵ has suggested that synovial-derived stem cells have better chondrogenic potential than bone- or adipose-derived stem cells as well as less hypertrophy.¹⁸ Also, a greater number of cells can also be obtained from harvesting synovium.⁵³ However, they require arthroscopy to harvest, necessitating a 2-stage procedure. Bone marrow–derived stem cells can be harvested and implanted as a single stage.^9,13,21 Animal and in vivo studies have shown more ability to harvest a sufficient number of cells with good proliferation capacity from bone marrow.²⁴ They also result in less donor-site morbidity than synovial-derived stem cells. Peripheral blood–derived stem cells offer a low cost and easy-to-extract source of stem cells,^41,50 although doing so requires the administration of filgrastim to stimulate cellular proliferation. This carries potential side effects, such as myalgia, fever, and headaches, which patients in the Monckeberg et al⁴¹ series experienced. Currently, no optimal source for MSCs has been established.

We found variable use of scaffolds in our studies. While some studies have demonstrated positive outcomes without the use of scaffolds,^47,51-53 their use is increasingly common, with the goal being to provide an ideal 3-dimensional microenvironment for MSC proliferation.²⁰ In an in vitro study, Kim et al³⁰ found that fibrin glue sustains the functional survival and paracrine function of MSCs in addition to its use to fix the prosthesis in place.^14,32,58 Other common scaffolds that we found in use were hyaluronic acid,^9,25,34,50 PRP,^23,41 and collagen matrix.^13,55 PRP has the advantage of being biodegradable, and it contains chondrogenesis-inducing growth factors.⁷ The ability of MSCs to adhere to and differentiate within the scaffold has been shown in a number of studies,^36,37,56 and the use of scaffolds has been shown to have favorable outcomes in the treatment of osteoarthritis.⁵⁹ We could not find any human study specifically investigating the efficacy of scaffolds. There have been concerns about the potential for foreign body interactions with the use of scaffolds⁵³; however, this did not manifest in any of the patients in this review.

Three studies described results concerning injection of freshly isolated bone marrow MSCs.^9,13,21 This is believed to be advantageous compared with culture-derived MSCs as the heterogeneous composition of the aspirate may enhance tissue regeneration. This process would in addition avoid issues relating to a lack of standardization of protocols for stem cell preparation. There is also a risk for bacterial growth within the culture.²⁶ Cultured bone marrow–derived MSCs have been advocated because the heterogeneous nature of bone marrow in particular means MSCs potentially comprise only 0.001% of the cellular content of the aspirate. Cultures are therefore a means to generate an adequate amount of MSCs and permit the use of cell surface markers for positive selection.¹⁶ However, the use of cultures requires a 2-stage procedure: the initial harvesting and the implantation. If comparable or even superior results can be achieved with injection of freshly aspirated stem cells, then this presents a clear economic advantage for the use of MSCs over ACI, which is also a 2-stage culture-based procedure.

Dulbecco’s modified Eagle medium was the most commonly used culture medium in MSC preparation.^23,29,53,58 This medium contains between 10% and 15% fetal bovine serum (FBS). Sekiya et al⁵¹ and Hashimoto et al²⁵ used human autologous serum in their cultures. Serum supplementation is vital, as it provides nutrients, attachment, and growth factors to the MSCs. However, FBS has been associated with high lot-to-lot variability and a risk of transmission of infectious agents.⁸ There have also been cases of immunological reactions with the formation of anti-FBS antibodies in the host,⁸ all of which potentially affect clinical outcomes for patients. The studies in our review did not note any such side effects. However, there has been a drive to search for alternative preparations to FBS.⁸

Neither of the 2 studies that investigated the link between age and outcomes found a significant correlation between functional outcomes and age.^21,58 Lesion size also did not appear to influence outcome,^21,32,58 which suggests that MSCs are a flexible treatment with the potential to delay the onset of arthritis in older patients with larger lesion sizes. There was an association between male sex and functional outcomes in 1 study⁵⁸; the reasons behind this are not entirely clear and perhaps warrant investigation in future studies.

There were no major adverse events in the studies reviewed. The current clinical data available on the use of MSCs in a number of pathologies largely support the safety of their use. Tumorigenicity remains an important theoretical risk factor. There has been literature indicating that MSCs may be the cell of origin of Ewing sarcoma.³⁵ In 2009, Amariglio et al³ published a case of a glioma originating from a neural stem cell transplantation. The mean follow-up of our studies was 3.6 years (range, 1-10 years); therefore, patients may not have been monitored long enough for tumorigenesis to manifest.

This review has a number of limitations. First, there was significant heterogeneity in the way that the studies published their outcomes. This prevented any meta-analysis of the data and limited the review to a narrative synthesis of the findings. Second, a number of patients had treatments in addition to the MSCs, ranging from high tibial osteotomies to PRP injection. It is difficult to assess to what extent the positive outcomes were due to the MSC treatment alone, as it was statistically assessed in only 1 study.²¹ Patients with osteochondral injuries often have additional injuries or issues with lower limb malalignment that require treatment,³⁸ making this a challenging limitation to overcome. In addition, there was significant variation among treatment protocols, ranging from source and preparation of cells, operative technique, and rehabilitation protocol. This again makes a meta-analysis challenging. Third, the interventions were carried out over a 15-year period across the 17 studies, which likely contributed to the variation in the preparation and implantation techniques. Skowroński and Rutka⁵⁵ completed one of the earliest cases in 2002. They reported 6 cases of implantation failure of their 46 cases, in contrast to only 2 other cases across the remaining 321 patients in this review. This, for example, may be a reflection of more refined techniques being used in later studies. Finally, the median Coleman methodology score was 68 (range, 57-86), and each randomized controlled trial was deemed to be of high risk of bias in at least 1 domain of the Cochrane tool. As such, there were significant methodological flaws in the studies, which need to be taken into account when reflecting on the results. The variable methodological quality may also mean that adverse events or complications were underreported.

Recommendations for Future Studies

We first suggest that studies ensure that the details of the cartilage defect are noted, such as size, location, and ICRS grade. The diagnostic history should also be noted, as well as patient factors such as age, body mass index, and any concomitant treatment given.

There should be a standardized reporting criterion for how the MSCs are prepared. The volume of aspirate, sight of injection, and cell count should be reported. For noncultured preparations, method of centrifugation and treatment of the aspirate should be reported. For cultured preparations, the duration of culture, culture medium, the cellular markers used for sorting, and the number of passages should be stated.

The method of delivery should be accurately recorded, as should the use of scaffolds. This is important, as interest in MSCs is growing; however, there is no clear established optimal protocol for preparation or dose of cells for injection.

Future trials should aim to adopt a consistent and well-established protocol for assessment and reporting of outcomes, as the large degree of heterogeneity seen in this review made a meta-analysis unfeasible.

There is still a paucity of large randomized controlled trials assessing the superiority of MSCs against other established therapies. In addition, long-term follow-up is needed to establish the long-term cost-benefit effect and safety of MSCs.

Conclusion

The interest in the use of MSCs in the treatment of osteochondral defects is certainly growing. To date, clinical studies have shown promising short- to medium-term outcomes with good safety profiles. More research is needed to establish the optimal cell source, preparation method, and dose of cells. More research is also needed to establish any superiority over ACI and the long-term outcomes of these patients.

Footnotes

§

References 1, 13, 23, 25, 29, 32, 34, 53, .

Submitted June 3, 2020; accepted September 9, 2020.

The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

An online CME course associated with this article is available for 1 AMA PRA Category 1 Credit™ at . In accordance with the standards of the Accreditation Council for Continuing Medical Education (ACCME), it is the policy of The American Orthopaedic Society for Sports Medicine that authors, editors, and planners disclose to the learners all financial relationships during the past 12 months with any commercial interest (A ‘commercial interest’ is any entity producing, marketing, re-selling, or distributing health care goods or services consumed by, or used on, patients). Any and all disclosures are provided in the online journal CME area which is provided to all participants before they actually take the CME activity. In accordance with AOSSM policy, authors, editors, and planners’ participation in this educational activity will be predicated upon timely submission and review of AOSSM disclosure. Noncompliance will result in an author/editor or planner to be stricken from participating in this CME activity.

References

Akgun

Unlu

Erdal

, et al. Matrix-induced autologous mesenchymal stem cell implantation versus matrix-induced autologous chondrocyte implantation in the treatment of chondral defects of the knee: a 2-year randomized study. Arch Orthop Trauma Surg. 2015;135(2):251-263.

Alford

Cole

. Cartilage restoration, part 1: basic science, historical perspective, patient evaluation, and treatment options. Am J Sports Med. 2005;33(2):295-306.

Amariglio

Hirshberg

Scheithauer

, et al. Donor-derived brain tumor following neural stem cell transplantation in an ataxia telangiectasia patient. PLoS Med. 2009;6(2):e1000029.

Anderson

Irrgang

Kocher

Mann

Harrast

. The International Knee Documentation Committee subjective knee evaluation form: normative data. Am J Sports Med. 2006;34(1):128-135.

Anderson

Little

Toth

, et al. Stem cell therapies for knee cartilage repair: the current status of preclinical and clinical studies. Am J Sports Med. 2014;42(9):2253-2261.

Årøen

Løken

Heir

, et al. Articular cartilage lesions in 993 consecutive knee arthroscopies. Am J Sports Med. 2004;32(1):211-215.

Awad

Hussein

Helwa

, et al. Meta-analysis and evidence base for the efficacy of autologous bone marrow mesenchymal stem cells in knee cartilage repair: methodological guidelines and quality assessment. Stem Cells Int. 2019;2019:3826054.

Bieback

Hecker

Kocaömer

, et al. Human alternatives to fetal bovine serum for the expansion of mesenchymal stromal cells from bone marrow. Stem Cells. 2009;27(9):2331-2341.

Buda

Vannini

Cavallo

, et al. One-step arthroscopic technique for the treatment of osteochondral lesions of the knee with bone-marrow-derived cells: three years results. Musculoskelet Surg. 2013;97(2):145-151.

10.

Caplan

Dennis

. Mesenchymal stem cells as trophic mediators. J Cell Biochem. 2006;98(5):1076-1084.

11.

Centeno

Busse

Kisiday

Keohan

Freeman

Karli

. Increased knee cartilage volume in degenerative joint disease using percutaneously implanted, autologous mesenchymal stem cells. Pain Physician. 2008;11(3):343-353.

12.

Curl

Krome

Gordon

Rushing

Smith

Poehling

. Cartilage injuries: a review of 31,516 knee arthroscopies. Arthroscopy. 1997;13(4):456-460.

13.

de Girolamo

Schönhuber

Viganò

, et al. Autologous matrix-induced chondrogenesis (AMIC) and AMIC enhanced by autologous concentrated bone marrow aspirate (BMAC) allow for stable clinical and functional improvements at up to 9 years follow-up: results from a randomized controlled study. J Clin Med. 2019;8(3):392.

14.

de Windt

Vonk

Slaper-Cortenbach

ICM

Nizak

van Rijen

MHP

Saris

DBF

. Allogeneic MSCs and recycled autologous chondrons mixed in a one-stage cartilage cell transplantion: a first-in-man trial in 35 patients. Stem Cells. 2017;35(8):1984-1993.

15.

Del Buono

Volpin

Maffulli

. Minimally invasive versus open surgery for acute Achilles tendon rupture: a systematic review. Br Med Bull. 2013;109(1):45-54.

16.

Dominici

Le Blanc

Mueller

, et al. Minimal criteria for defining multipotent mesenchymal stromal cells: the International Society for Cellular Therapy position statement. Cytotherapy. 2006;8(4):315-317.

17.

Farr

Cole

Dhawan

Kercher

Sherman

. Clinical cartilage restoration: evolution and overview. Clin Orthop Relat Res. 2011;469(10):2696-2705.

18.

Fernandes

Kimura

Pinheiro

CCG

, et al. Human synovial MSC good manufacturing practices for articular cartilage regeneration. Tissue Eng Part C Methods. 2018;24(12):709-716.

19.

Frehner

Benthien

. Microfracture: state of the art in cartilage surgery? Cartilage. 2017;9(4):339-345.

20.

Girão

Semitela

Ramalho

Completo

Marques

PAAP

. Mimicking nature: fabrication of 3D anisotropic electrospun polycaprolactone scaffolds for cartilage tissue engineering applications. Compos Part B Eng. 2018;154:99-107.

21.

Gobbi

Whyte

. Long-term clinical outcomes of one-stage cartilage repair in the knee with hyaluronic acid–based scaffold embedded with mesenchymal stem cells sourced from bone marrow aspirate concentrate. Am J Sports Med. 2019;47(7):1621-1628.

22.

C-W

Park

Y-B

Kim

Lee

H-J

. Intra-articular mesenchymal stem cells in osteoarthritis of the knee: a systematic review of clinical outcomes and evidence of cartilage repair. Arthroscopy. 2019;35(1):277-288.e2.

23.

Haleem

El Singergy

Sabry

, et al. The clinical use of human culture–expanded autologous bone marrow mesenchymal stem cells transplanted on platelet-rich fibrin glue in the treatment of articular cartilage defects: a pilot study and preliminary results. Cartilage. 2010;1(4):253-261.

24.

Han

Zhang

Han

Chang

Ding

. Mesenchymal stem cells for regenerative medicine. Cells. 2019;8(8):886.

25.

Hashimoto

Nishida

Takahashi

, et al. Transplantation of autologous bone marrow–derived mesenchymal stem cells under arthroscopic surgery with microfracture versus microfracture alone for articular cartilage lesions in the knee: a multicenter prospective randomized control clinical trial. Regen Ther. 2019;11:106-113.

26.

Hause

Eteshola

. Rationale for using direct bone marrow aspirate as a proliferant for regenerative injection therapy (prolotherapy). Open Stem Cell J. 2013;4:7-14.

27.

Hjelle

Solheim

Strand

Muri

Brittberg

. Articular cartilage defects in 1,000 knee arthroscopies. Arthroscopy. 2002;18(7):730-734.

28.

Iyer

Rojas

. Anti-inflammatory effects of mesenchymal stem cells: novel concept for future therapies. Expert Opin Biol Ther. 2008;8(5):569-581.

29.

Kamei

Ochi

Adachi

, et al. The safety and efficacy of magnetic targeting using autologous mesenchymal stem cells for cartilage repair. Knee Surg Sports Traumatol Arthrosc. 2018;26(12):3626-3635.

30.

Kim

Lee

Yoon

Kim

. Fibrin glue improves the therapeutic effect of MSCs by sustaining survival and paracrine function. Tissue Eng Part A. 2013;19(21-22):2373-2381.

31.

Kim

C-W

Park

Y-B

Nam

Lee

J-E

Lee

H-J

. Intra-articular injection of mesenchymal stem cells for clinical outcomes and cartilage repair in osteoarthritis of the knee: a meta-analysis of randomized controlled trials. Arch Orthop Trauma Surg. 2019;139(7):971-980.

32.

Koh

Y-G

Kwon

O-R

Kim

Y-S

Choi

Y-J

Tak

D-H

. Adipose-derived mesenchymal stem cells with microfracture versus microfracture alone: 2-year follow-up of a prospective randomized trial. Arthroscopy. 2016;32(1):97-109.

33.

Kraeutler

Belk

Purcell

McCarty

. Microfracture versus autologous chondrocyte implantation for articular cartilage lesions in the knee: a systematic review of 5-year outcomes. Am J Sports Med. 2017;46(4):995-999.

34.

Kyriakidis

Iosifidis

Michalopoulos

Melas

Stavropoulos-Giokas

Verdonk

. Good mid-term outcomes after adipose-derived culture-expanded mesenchymal stem cells implantation in knee focal cartilage defects. Knee Surg Sports Traumatol Arthrosc. 2020;28(2):502-508.

35.

Lin

Wang

Lozano

. Mesenchymal stem cells and the origin of Ewing’s sarcoma. Sarcoma. 2011;2011:276463.

36.

Lundquist

Dziegiel

Ågren

. Bioactivity and stability of endogenous fibrogenic factors in platelet-rich fibrin. Wound Repair Regen. 2008;16(3):356-363.

37.

Mackay

Beck

Murphy

Barry

Chichester

Pittenger

. Chondrogenic differentiation of cultured human mesenchymal stem cells from marrow. Tissue Eng. 1998;4(4):415-428.

38.

Mall

Harris

Cole

. Clinical evaluation and preoperative planning of articular cartilage lesions of the knee. J Am Acad Orthop Surg. 2015;23(10):633-640.

39.

Marlovits

Striessnig

Resinger

, et al. Definition of pertinent parameters for the evaluation of articular cartilage repair tissue with high-resolution magnetic resonance imaging. Eur J Radiol. 2004;52(3):310-319.

40.

Mistry

Connock

Pink

, et al. Autologous chondrocyte implantation in the knee: systematic review and economic evaluation. Health Technol Assess. 2017;21(6):1-294.

41.

Monckeberg

Rafols

Apablaza

Gerhard

Rosales

. Intra-articular administration of peripheral blood stem cells with platelet-rich plasma regenerated articular cartilage and improved clinical outcomes for knee chondral lesions. Knee. 2019;26(4):824-831.

42.

Nejadnik

Hui

Feng Choong

Tai

B-C

Lee

. Autologous bone marrow–derived mesenchymal stem cells versus autologous chondrocyte implantation: an observational cohort study. Am J Sports Med. 2010;38(6):1110-1116.

43.

Oussedik

Tsitskaris

Parker

. Treatment of articular cartilage lesions of the knee by microfracture or autologous chondrocyte implantation: a systematic review. Arthroscopy. 2015;31(4):732-744.

44.

Paschos

Sennett

. Update on mesenchymal stem cell therapies for cartilage disorders. World J Orthop. 2017;8(12):853-860.

45.

Pei

Boyce

Kish

. Repair of full-thickness femoral condyle cartilage defects using allogeneic synovial cell-engineered tissue constructs. Osteoarthritis Cartilage. 2009;17(6):714-722.

46.

Peterson

Vasiliadis

Brittberg

Lindahl

. Autologous chondrocyte implantation: a long-term follow-up. Am J Sports Med. 2010;38(6):1117-1124.

47.

Pipino

Risitano

Alviano

, et al. Microfractures and hydrogel scaffolds in the treatment of osteochondral knee defects: a clinical and histological evaluation. J Clin Orthop Trauma. 2019;10(1):67-75.

48.

Redondo

Beer

Yanke

. Cartilage restoration: microfracture and osteochondral autograft transplantation. J Knee Surg. 2018;31(3):231-238.

49.

Sakaguchi

Sekiya

Yagishita

Muneta

. Comparison of human stem cells derived from various mesenchymal tissues: superiority of synovium as a cell source. Arthritis Rheum. 2005;52(8):2521-2529.

50.

Saw

K-Y

Anz

Siew-Yoke Jee

, et al. Articular cartilage regeneration with autologous peripheral blood stem cells versus hyaluronic acid: a randomized controlled trial. Arthroscopy. 2013;29(4):684-694.

51.

Sekiya

Muneta

Horie

Koga

. Arthroscopic transplantation of synovial stem cells improves clinical outcomes in knees with cartilage defects. Clin Orthop Relat Res. 2015;473(7):2316-2326.

52.

Shapiro

Kazmerchak

Heckman

Zubair

O’Connor

. A prospective, single-blind, placebo-controlled trial of bone marrow aspirate concentrate for knee osteoarthritis. Am J Sports Med. 2016;45(1):82-90.

53.

Shimomura

Yasui

Koizumi

, et al. First-in-human pilot study of implantation of a scaffold-free tissue-engineered construct generated from autologous synovial mesenchymal stem cells for repair of knee chondral lesions. Am J Sports Med. 2018;46(10):2384-2393.

54.

Shirasawa

Sekiya

Sakaguchi

Yagishita

Ichinose

Muneta

. In vitro chondrogenesis of human synovium-derived mesenchymal stem cells: optimal condition and comparison with bone marrow–derived cells. J Cell Biochem. 2006;97(1):84-97.

55.

Skowroński

Rutka

. Osteochondral lesions of the knee reconstructed with mesenchymal stem cells—results. Ortop Traumatol Rehabil. 2013;15(3):195-204.

56.

Smith

Sigal

Grande

. Immunology and cartilage regeneration. Immunol Res. 2015;63(1):181-186.

57.

Sterne

JAC

Savović

Page

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898.

58.

Teo

AQA

Wong

Shen

, et al. Equivalent 10-year outcomes after implantation of autologous bone marrow–derived mesenchymal stem cells versus autologous chondrocyte implantation for chondral defects of the knee. Am J Sports Med. 2019;47(12):2881-2887.

59.

Wakitani

Imoto

Yamamoto

Saito

Murata

Yoneda

. Human autologous culture expanded bone marrow mesenchymal cell transplantation for repair of cartilage defects in osteoarthritic knees. Osteoarthritis Cartilage. 2002;10(3):199-206.

60.

Wakitani

Okabe

Horibe

, et al. Safety of autologous bone marrow–derived mesenchymal stem cell transplantation for cartilage repair in 41 patients with 45 joints followed for up to 11 years and 5 months. J Tissue Eng Regen Med. 2011;5(2):146-150.

61.

Wong

Lee

KBL

Tai

Law

Lee

Hui

JHP

. Injectable cultured bone marrow–derived mesenchymal stem cells in varus knees with cartilage defects undergoing high tibial osteotomy: a prospective, randomized controlled clinical trial with 2 years’ follow-up. Arthroscopy. 2013;29(12):2020-2028.

62.

Yoo

Barthel

Nishimura

, et al. The chondrogenic potential of human bone-marrow-derived mesenchymal progenitor cells. J Bone Joint Surg Am. 1998;80(12):1745-1757.