Defect size is the primary decision driver

The answer to 'autograft or allograft?' is determined almost entirely by the size and depth of the defect, not by one technique being inherently superior to the other. Two independently published clinical summaries converge on the same practical ladder: OATS — in which cylindrical plugs of the patient's own bone and cartilage are transferred into the lesion — suits focal defects up to approximately 2 cm² using a single plug, or up to around 4 cm² when several plugs are arranged in a mosaic pattern. Once a lesion exceeds that ceiling, fresh osteochondral allograft (OCA), using size-matched donor tissue, becomes the preferred option; donor supply carries no equivalent size restriction.

Two factors can shift the decision toward OCA regardless of lesion area. First, significant subchondral bone loss makes a surface-only autograft repair inadequate — OCA replaces the full depth of damaged cartilage and underlying bone in a single procedure. Second, a history of failed index cartilage surgery — microfracture, ACI, or prior mosaicplasty — typically positions OCA as the appropriate salvage route, because repeated marrow stimulation can compromise the subchondral bone plate and limit the options remaining.

Both techniques target focal, contained osteochondral defects in otherwise mechanically reasonable knees. Neither is an appropriate treatment for diffuse or advanced osteoarthritis.

Anatomically, the medial femoral condyle is the most commonly affected site across both surgical populations — a pattern seen in a systematic review of paediatric knee cartilage lesions, where the medial femoral condyle accounted for the largest single group of defect locations, and reproduced in a dedicated return-to-sport review of OCA in athletes, where 44% of defects fell on the medial femoral condyle.

What OATS involves and who it suits

OATS is a single-stage procedure in which one or more cylindrical plugs — bone and cartilage together — are harvested from a low-load area of the patient's own femoral condyle and press-fitted directly into the prepared defect. Because the tissue comes from the same knee, there is no waiting for donor matching and no risk of disease transmission or immune rejection.

The strongest argument for OATS is its durability within the appropriate size range. A 2024 longitudinal study of 63 patients found that 78% achieved a clinically meaningful improvement on the IKDC score at two years — a threshold sustained by 75% at ten years. A mosaicplasty registry study of comparable size (mean age 27.4 years, mean lesion 2.3 cm²) recorded IKDC scores rising from 46 to 70 over the same period, with only 2 of 63 patients requiring conversion to arthroplasty. These results suggest that early functional gain after OATS tends to reflect the long-term trajectory.

Donor site morbidity

Harvesting plugs from within the knee is not without consequence. In adults, the documented morbidity rate at the harvest site is approximately 7.8%. A 2026 paediatric series found 13.5% of OATS patients required subsequent knee surgery — most commonly chondroplasty for fibrocartilage overgrowth at the donor site — compared with zero in an OCA comparison group. This is a meaningful figure that informs patient selection, particularly in younger or still-growing patients.

OATS is also not suited to every knee configuration. Standard contraindications include:

• Global or diffuse cartilage disease rather than a focal contained lesion
• Varus or valgus malalignment left uncorrected
• Kissing lesions on opposing joint surfaces
• Single-plug arthroscopic cases where the defect exceeds 20 mm

Where any of these factors are present, the clinical question shifts toward allograft or a different restorative pathway.

What OCA involves and when autograft is not enough

Fresh osteochondral allograft transplantation (OCA) steps in where autograft tissue simply cannot go. When a defect is too large for the patient's own knee to safely donate enough plugs, when bone loss runs too deep for a surface repair, or when an earlier cartilage procedure has already failed, a size-matched graft sourced through a tissue bank and transplanted within a defined window to preserve viable chondrocytes becomes the appropriate route. Donor matching is based on joint geometry; the graft is shaped to fit the prepared recipient bed, then secured — typically by press-fit technique — in a single operative session.

Survivorship in real terms

Long-term data from a 2021 cartilage series show 82.6% graft survivorship at five years and 69.6% at ten years, with 68% of grafts still in situ and functional at a mean of 12.9 years — failure defined as arthroplasty conversion, revision, or a Hospital for Special Surgery score below 70. These are stringent endpoints, so the figures represent a realistic rather than optimistic picture.

A 2023 nationwide database study of 6,391 OCA patients recorded a cumulative five-year salvage surgery rate of just 1.71%. Two factors independently reduced that risk: age between 20 and 29 years (adjusted odds ratio 0.49) and concomitant or prior bony realignment — high tibial or distal femoral osteotomy — which cut the risk further (adjusted odds ratio 0.24). Correcting malalignment before or alongside OCA is therefore not merely an optional adjunct.

The upper limit

For defects reaching 9–10 cm², a 2025 systematic review of salvage OCA reported failure rates approaching 39% in that bracket. At that scale, the procedure should be framed honestly as one that preserves function and defers joint replacement rather than permanently restoring the joint.

BMAC augmentation

A prospective RCT found that adding bone marrow aspirate concentrate at the time of OCA reduced reoperation rates from 35.3% to 5.3% at two years (p=0.02). Patient-reported scores — IKDC and KOOS JR — showed no significant difference over the same period. That gap between fewer reoperations and unchanged symptom scores at two-year follow-up means the clinical weight of this finding is still unclear, and BMAC remains a selectively used adjunct pending longer follow-up data.

Recovery timelines and return to sport

OCA takes a mean of four months longer to reach full activity than OATS — a clinically relevant gap for patients whose work or sport has a specific return date in mind. Both procedures ultimately achieve high return-to-sport rates, but the path differs in length and structure.

OCA rehabilitation milestones

After OCA, the typical phased progression runs:

• Protected weight-bearing for the first six to eight weeks, to allow graft integration
• Low-impact activity — swimming, cycling, walking on level ground — from around four to six months
• Higher-impact and return to sport from nine to twelve months

A 2026 rehabilitation review noted wide protocol variation between centres, so precise week-by-week timelines differ. These milestones reflect a consensus range rather than a universal schedule. In athlete cohorts, OCA achieves a return-to-sport rate of approximately 88% overall, with a 2024 Mills comparison confirming the four-month lag behind OATS; a separate systematic review of 471 athletes reported 72% returning to sport at a weighted mean of 11.1 months post-surgery, with 84% playing at the same or higher level.

OATS recovery

OATS patients follow a similarly phased rehabilitation but typically reach each milestone earlier. Recovery from the donor harvest site — in addition to the repaired defect — contributes to the overall timeline and should be factored into expectations from the outset.

A note on extrapolating athlete data

Most return-to-sport figures come from athlete-specific cohorts. Recreational patients or those returning to physically demanding work rather than competitive sport may find the timelines translate only approximately to their situation. The longer OCA timeline reflects the biology of integrating a larger, bone-inclusive graft — it is a trade-off for treating a defect that autograft could not safely address, not a shortcoming of the technique.

Other factors that shape the decision

Defect size sets the initial framework, but surgeons weigh several additional variables before settling on a technique.

Malalignment — varus or valgus — redistributes load across the joint in a way that accelerates graft deterioration regardless of surgical precision. It is a formal contraindication to OATS used in isolation and substantially raises OCA failure risk. The 2023 nationwide database study (6,391 patients) found that concomitant or prior bony realignment — high tibial osteotomy (HTO) or distal femoral osteotomy (DFO) — independently reduced failure risk by roughly three-quarters (adjusted odds ratio 0.24). Correcting alignment before or at the time of grafting is therefore a meaningful part of the surgical plan, not an optional refinement.

Age and disease extent bear on the outcome in a related way. Younger patients with corrected alignment carry the best long-term OCA survivorship. Older patients, or those with diffuse rather than focal disease, fall outside the repair envelope — for them, the appropriate conversation shifts towards joint replacement rather than preservation.

Prior surgery also changes the calculus. Failed marrow-stimulation raises failure risk for any subsequent cartilage repair, in part through disruption to the subchondral bone plate. In that setting, OCA is more robust as a salvage option than repeat autograft transfer. Kissing lesions — opposing defects on the femur and tibia — and global chondral disease are contraindications to both procedures in isolation.

With those selection variables mapped, a broader point of humility about what the evidence can actually prove is worth absorbing. A network meta-analysis of 19 randomised controlled trials found no statistically significant difference in patient-reported outcome measures between microfracture, ACI, AMIC, and osteochondral autograft transfer at any timepoint. The implication is not that technique choice is irrelevant — the case for OCA in large or bone-involved defects remains well-supported — but that within the range where more than one approach is technically feasible, precise patient matching matters more than any assumed technique hierarchy. Getting selection right is where the clinical work lies.

Getting assessed at Lincolnshire Knee

Deciding between OATS and OCA cannot be done from symptoms alone. The key variables — defect dimensions, cartilage depth, subchondral bone quality, and limb alignment — all require dedicated imaging and biomechanical evaluation before any surgical conversation is meaningful.

MRI cartilage analysis sits at the centre of that workup. Modern sequences can characterise defect area and depth, identify subchondral bone involvement, and detect changes in the surrounding cartilage that plain X-ray misses entirely. T2 mapping adds a compositional dimension — quantifying cartilage quality rather than just its shape — which matters where the visible defect boundary does not tell the full story. A functional biomechanical assessment can then reveal malalignment patterns that would compromise any graft's long-term survivorship, as discussed in the previous section.

Lincolnshire Knee, part of the MSK Doctors group, uses onMRI™ AI-driven MRI analysis (including cartilage segmentation and T2 mapping) and MAI Motion® biomechanical assessment to build that picture before any procedure is discussed. The approach is staged: assessment first establishes whether non-surgical management remains appropriate, and surgery is considered only where the clinical picture supports it.

Patients can be seen without a GP referral at clinics in Sleaford (NG34) and Grantham (NG31). Book an assessment at lincolnshireknee.co.uk.

1. [1] Microfractures, AMIC, osteochondral autograft transplantation and ACI for knee chondral defects: network meta-analysis of RCTs. (2024). https://doi.org/10.1530/EOR-23-0089 https://doi.org/10.1530/EOR-23-0089
2. [2] Arthroscopic Osteochondral Autograft Transplantation (OAT) in Patients with Focal Osteochondral/Chondral Lesions of the Knee Mid-Term Clinical Outcome. (2025). https://doi.org/10.5704/MOJ.2507.004 https://doi.org/10.5704/MOJ.2507.004
3. [3] Osteochondral Allograft Cartilage Transplantation for a Full-Thickness Femoral Condyle Chondral Lesion. (2019). https://doi.org/10.2106/jbjs.st.18.00041 https://doi.org/10.2106/jbjs.st.18.00041
4. [4] Mosaicplasty/Osteochondral Autograft Transfer Remains a Durable Solution for Symptomatic Chondral Defects of the Knee: Two to Ten-Year Follow-up Analysis. (2024). https://doi.org/10.1177/2325967124s00003 https://doi.org/10.1177/2325967124s00003
5. [5] Osteochondral Allograft Transplantation as a Salvage Procedure After Failed Index Cartilage Surgery of the Knee: A Systematic Review. (2025). https://doi.org/10.1177/03635465241238466 https://doi.org/10.1177/03635465241238466
6. [6] Rehabilitation and Return to Play Protocols Following Osteochondral Autograft and Allograft Transplantation for Knee Chondral Lesions in Athletic Populations. (2026). https://doi.org/10.1007/s12178-026-10033-y https://doi.org/10.1007/s12178-026-10033-y
7. [7] Defining clinically relevant outcome thresholds for pain and function after osteochondral autograft transplantation of the knee. (2024). https://doi.org/10.1002/ksa.12422 https://doi.org/10.1002/ksa.12422
8. [8] Young Age and Concomitant or Prior Bony Realignment Procedures are Associated with Decreased Risk of Failure of OCA in the Knee. (2023). https://doi.org/10.1177/19476035231178374 https://doi.org/10.1177/19476035231178374
9. [9] OATS Harvesting from the Femoral Condyle is not Benign: Knee Outcomes after OATS vs OCA. (2026). https://doi.org/10.1177/2325967126s00127 https://doi.org/10.1177/2325967126s00127