The Intra-articular Injection of Stem Cells as Therapy for Articular Cartilage Repair - What is the Evidence?
World Small Animal Veterinary Association World Congress Proceedings, 2013
Tania Banks, BVSc, FANZCVSc (Small Animal Surgery)
The School of Veterinary Science, University of Queensland, Queensland, Australia

Articular cartilage is a 1–5 mm thick, smooth and wear-resistant layer of tissue with unique and highly specialised structural, functional and mechanical properties. It has a limited capacity for intrinsic repair,1 and minor injuries/lesions may lead to progressive degeneration.2 Repair tissue is fibrocartilage with type I collagen, unable to withstand stress and deformation pressures. Clinically, this manifests as osteoarthritis (OA), a disease with high prevalence and morbidity.

Surgical therapies for OA include marrow stimulation, e.g., microfracture, osteochondral autograft transfer (OAT), joint replacement surgery, tissue engineering approaches, e.g., autologous chondrocyte implantation (ACI) and matrix-induced autologous chondrocyte implantation (MACI). The implantation of stem cells in combination with biodegradable scaffolds, differentiation factors, cytokines, growth factors, surface modifiers, peptides or genes (with or without the use of bioreactors) is a promising and rapidly developing treatment approach for OA. Further studies are underway to optimise this approach; however, it remains a challenge in tissue engineering pursuits to grow articular cartilage to a physiological and biomechanical standard.

Joint replacement can be very successful, with 95% of canine hip replacements returning to normal function. Initial results with the OAT technique in dogs are promising.3,4 Osteochondral graft(s) harvested from minimal weight-bearing portions are transplanted into defects in higher weight-bearing areas/disease site of the same joint. In 2008, Braun et al. reported good-excellent functional outcomes with the 'MegaOATS' technique in humans. Follow-up magnetic resonance imaging (MRI) at 5.5 years showed vital and congruent grafts.5 In 2007, 421 human OA patients treated with ACI, MACI, OATS or microfracture were reviewed. Microfracture results were worse for larger lesions and no technique had consistently better results, although all had improved clinical outcomes over preoperative values (mean follow-up 1.7 years).6 ACI often results in fibrocartilage, and there is also a need to sacrifice healthy cartilage as the source of donor chondrocytes for in vitro expansion. Hence, stem cells seem an attractive cell type for tissue regeneration/ repair. They are able to self-renew, and to differentiate along multiple lineage pathways. Three types of stem cell have been identified; embryonic, those derived from umbilical cord blood, and adult stem cells.7 In adults, stem cells exist in nearly every tissue, playing pivotal roles in maintaining homeostasis, and as replacement parts for expired/damaged units. Adult mesenchymal stem cells (MSCs) are capable of differentiating into several mesenchymal phenotypes (bone, fat, cartilage, ligament, tendon, muscle etc.),8 can self-renew for relatively long periods of time, are easily accessible, and have low immunogenicity9. Adult stem cells have been derived from bone marrow, adipose tissue, synovium, muscle, periosteum and skin etc.10

Recently, the intraarticular (I/A) injection of autologous adult stem cells for OA has been studied. I/A injection is easier and less invasive than open arthrotomy and direct cellular implantation. However, the published evidence regarding the efficacy of I/A injection of stem cells for OA treatment (outlined in Table 1) is not compelling at this stage. Case numbers are low and follow-up is usually very short. The 'stem cells' used may not be fully characterised according to published guidelines,11,12 and the number of injected stem cells varies greatly. The tissue from which the stem cells were derived, the carrier solution used to inject the cells, and the degree of existing OA, and the studied species are also inconsistent. OA models (mechanically or chemically induced) do not wholly reflect real-life disease - both in the pathophysiology of the disease and the condition of the donor stem cells. In some studies, evaluation methods are solely subjective (qualitative) and do not reliably evaluate the effectiveness of the treatment. Validity is greater with studies with strong objective data (quantitative), minimising the effects of any biases.

As mentioned above, the status of the donor stem cells may affect the outcome of the treatment. There is a decline of both the number of adult MSC numbers with increasing age,13,14 and their chondrogenic differentiation and proliferation potential.15 In certain degenerative diseases such as OA, stem cells are depleted and have reduced proliferative capacity and ability to differentiate;16 however, MSCs in adequate numbers and chondrogenic capacity can be isolated from OA patients regardless of age17,18. This may reflect the variability of clinical OA cases. The type of tissue from which the MSCs are derived influences their chondrogenic potential in vitro. This is greater for synovium-derived MSCs than for bone marrow, periosteum, skeletal muscle or adipose-derived MSCs.19 Bone marrow and synovial MSCs have more chondrogenic potential than adipose or muscle MSCs.20 Additionally, synovium-derived MSCs showed similar expansion ability to bone marrow MSCs.19

Although it has been shown that some cells do directly incorporate into the local tissues after I/A injection,21,22 the dominant mechanism of action is thought to result from an antiinflammatory effect and stimulation of host reparative cells,23 lasting around 6 months depending on the study (see Table 1). It is well established that MSCs secrete a variety of cytokines and growth factors (paracrine and autocrine activities). Thus, MSCs have 'trophic' effects (from the Greek troph, to nourish); acting to suppress the local immune system, inhibit scarring and cell death, enhance angiogenesis, and stimulate mitosis and differentiation of tissue-intrinsic reparative or stem cells.8 MSCs could also provide antioxidants, free radical scavengers, and chaperone/stress proteins at an ischaemic site, which may all assist cell repair. Based on current published evidence, the I/A injection of the stromal vascular fraction (SVF) for canine OA treatment, which contains adult MSCs in low numbers (compared to studies using in vitro MSC expansion) warrants further investigation. As Veterinary Scientists, our recommendations to clients must be founded on an evidence-based approach of expected outcome, encompassing all treatment options.

Table 1. Papers reporting I/A stem cells for OA treatment


Model & method

Findings & conclusion

Toghraie et al. 201224

Rabbits. Mechanical OA model. 10 rabbits: 1 x 106 AD-MSCs in 1ml medium, 10 control: 1 ml medium 12 wks postop

I/A administration of AD-MSCs can improve OA lesions (radiographs and histology), for studied period (2 months after injection)

Horie et al. 201223

Rats mechanical OA model (hemimeniscectomy) 5 rats for histology, 4 rats for RT-PCR. 2 x 106 human or rat BM-MSCs injected in saline immediately postop (control knee saline only)

2, 4, 8 wks histology, PCR. MSCs enhanced meniscus regeneration meniscus, increased col-II, inhibited OA. MSCs expressed high levels of Ihh, PTHLH, BMP2, rapid reduction in MSC numbers postinjection.

Emadedin et al. 20129

Humans (n=6) knee OA that required joint replacement surgery
20–24 x 106 BM-MSCs passaged-2 in physiological serum

Size of oedematous subchondral patches on MRI reduced, thickness cartilage increased in 50%. Pain, walking, knee flexion improved up to 6 months, then pain was slightly increased, walking decreased. Recommend a second injection 6 months post first.

Guercio et al. 201225

Dogs (n=4) shoulder OA on radiographs, 3–5 x 106 AD-MSCs in PRP or HA.

OA of elbows improved with time (subjective only) follow-up one month.

Davatchi et al. 201126

Humans (n=4)
8–9 x 106 BM-MSCs injected

Mild improvement in radiographs, high improvements in subjective parameters Results in 4/6 patients encouraging but not excellent

Toghraie et al. 201127

Rabbits. Mechanical OA model. 1 x 106 AD-MSCs in 1 ml medium 12 weeks postop

Rabbits that received MSCs less cartilage degeneration, osteophyte formation, subchondral sclerosis than controls for 2 months

Horie et al. 200928

Rat (n=27) mechanical OA
hemimeniscectomy, immediate injection 5x106 synovium-MSCs in saline (control saline only)

2, 4, 8, 12 wks, cells adhered to lesion, differentiated into meniscal cells, promoted meniscal regeneration and coll-II (TEM, gross, histology, PCR)

Mokbel et al. 201121

Donkey (n=27). Chemical OA for 3 wks (group 1), 6 wks (group 2) or 9 wks (group 3), then I/A injection of 1.8–2.3 x 106 (adherent, CD34- CD29+) cells/ml GFP-transfected BM-MSCs + 3 ml HA (R carpus) or 3 ml HA alone (L carpus)

Improved lameness, radiographic and histology scores versus controls. GFP cells detected on surface and interior. Cell-treated slight difference or improvement of the articular cartilage status, minimal effect for severe OA. Injection of MSCs early is ideal.

Frisbie et al. 200929

Horses (n=24). n=8 placebo, n=8 SVF, n=8
10.5 x 106 BM-MSCs in saline day 14. histo, synovial fluid, clinical, histo and biochemistry, radiography to 70 days

No adverse events. No significant treatment effects demonstrated, except for improved synovial fluid PGE2. Cannot recommend the use of stem cells for the treatment of OA represented in model.

Centeno et al. 200830

One human patient (n=1) meniscal pain and damage
45.6 x 106 expanded BM-MSCs in PBS + HA + 2 I/A platelet lysate injections at weekly intervals + dexamethasone I/A 2 weeks postop.

MRI images pre and 3 months postop. MRI evidence of increased meniscus volume, Safe procedure without any complications

Koga et al. 200820

Rabbits (n=36) osteochondral defect, immediate 1 x 107 synovium-MSCs.

1 day, 12, 24 wks postop. placed cells on defect for 10 min, >60% cells attached to defect, histo scores better than controls

Black et al. 200831

Elbow chronic OA (14 dogs) (10 months–11years) demonstrated on radiographs.

Subjective data (lameness, pain, range of motion) only - all improved for 3–6 months.

Black et al. 200732

Hip chronic OA (21 dogs, 1–11 years) demonstrated on radiographs

3 month follow-up
Subjective data all improved.

Lee et al. 200733

Pigs (n=27) mechanical OA partial-thickness articular cartilage defect, 7x106 BM-MSCs in HA + 2 more HA injections weekly intervals (saline or HA for controls).

Improved cartilage healing histology and gross at 6 and 12 wks.

Murphy et al. 200322

Caprine BM-MSCs CrCL transection, hemimeniscectomy + exercise - 6 wks preinjection of 10 x 106 BM-MSCs in HA, express GFP, (HA controls).

Terminate 6 weeks post injection (test n=6, control n=3), terminate 20 weeks postinjection (n=9) control (n=6)

Regeneration of meniscus, retarded OA. Host-derived reparative cells likely stimulated. Some prelabelled MSCs observed in the new meniscus, but too few to account for the regeneration. GFP- positive cells were detected primarily at the surface and also in the center of neomeniscal tissue 6 weeks.

Im et al. 200134

Rabbits (n=16, n=13 control) osteochondral defect, immediate injection 1 x106 BM-MSCs in medium, or medium for controls

At 14 weeks, improved histology scores, improved immunostaining collagen II and PCR col-II in regenerated cartilage. Repair can be enhanced by I/A cultured MSCs

MSC = mesenchymal stem cell; BM-MSC = bone-marrow MSC; AD-MSC = adipose MSC; HA = hyaluronic acid; GFP = green fluorescent protein; CrCL = cranial cruciate ligament; Col-I = collagen type I; Col-II = collagen type II; Ihh - Indian hedgehog; BMP2 = bone morphogenic protein 2; PTHLH = parathyroid hormone-like hormone; PCR = polymerase chain reaction; PRP = platelet rich plasma.


1.  Noguchi T, et al. Clin Orthop Relat Res. 1994;302:251–258.

2.  Guilak F, et al. Biorheology. 2004;41:389–399.

3.  Cook JL, et al. Vet Surg. 2008;37:311–321.

4.  Fitzpatrick N, et al. Vet Surg. 2009;38:173–184.

5.  Braun S, et al. Arth Res Ther. 2008;10(3).

6.  Magnussen R, et al. Clin Orthop Relat Res. 2008;466:952–962.

7.  Lubis AMT, Lubis VK. Acta Med Indones. 2012;44:62–68)

8.  Caplan D. J Cell Biochem. 2006;98:1076–1084.

9.  Emadedin M, et al. Archives Iran Med. 2012;15:422–428.

10. Danisovic L, Zamborsky R, Bohmer D. Experiment Biol Med. 2012;237:10–17.

11. Dominici M, et al. Cytotherapy. 2006;8:315–317.

12. Horwitz E, et al. Cytotherapy. 2005;7:393–395.

13. Caplan A. J Cell Physiol. 2007;213:341–347.

14. Noth U, et al. Nat Clin Pract Rheumatol. 2008;4:371–380.

15. Stolzing A, et al. Mechan Ageing Dev. 2008;129:163–173.

16. Murphy J, et al. Arthritis Rheum. 2002;46:704–713.

17. Kafienah W, et al. Arthritis Rheum. 2007;56:177–187.

18. Scarstuhl A, et al. Stem Cells. 2007;25:3244–3251.

19. Sakaguchi Y, et al. Arthritis Rheum. 2005;52:2521–2529.

20. Koga H, et al. Arth Res Ther. 2008;10.

21. Mokbel AN, et al. BMC Musculoskelet Disord. 2011;12:259.

22. Murphy J, et al. Arthritis Rheum. 2003;48:3464–3474.

23. Horie M, et al. Osteoarth Cart. 2012;20:1197–1207.

24. Toghraie F, et al. Archives Iranian Med. 2012;15:495–499.

25. Guercio A. Cell Biol Internat. 2012;36:189–194.

26. Davatchi F, et al. Int J Rheum Dis. 2011;14:211–215.

27. Toghraie F, et al. Knee. 2011;18:71–75

28. Horie M, et al. Stem Cells. 2009;27:878–887.

29. Frisbie. J Orthop Res. 2009;27(12):1675–1680.

30. Centeno C, et al. Med Hypoth. 2008;71:900–908.

31. Black L, et al. Vet Ther. 2008;9:192–200.

32. Black L, et al. Vet Ther. 2007;8:272–284.

33. Lee K, et al. Stem Cells. 2007;25:2964–2971.

34. Im G, et al. J Bone Joint Surg. 2001;83B:289–294.


Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Tania Banks, BVSc, FANZCVSc (Small Animal Surgery)
The School of Veterinary Science
University of Queensland
Queensland, Australia

MAIN : Joint Surgery : Stem Cells for Articular Cartilage Repair
Powered By VIN