Why microfracture falls short for larger talar defects

A diagnosis of talar osteochondral lesion (OLT) often comes with a follow-up question: why not choose the simpler, less invasive procedure? Microfracture — which involves drilling small holes into the bone beneath the damaged cartilage to stimulate a healing response — has been used for decades and carries a shorter operating time. The problem lies in what it actually produces.

Rather than regenerating the native hyaline cartilage that lines a healthy talar dome, microfracture yields fibrocartilage — a structurally weaker tissue that does not match the mechanical demands placed on the ankle. Research confirms this repair tissue has inferior mechanical properties compared to native hyaline cartilage, and clinical experience shows it tends to break down within two to three years under load.

There is a second, less visible concern: microfracture leaves the subchondral bone plate — the hard layer directly beneath the cartilage — incompletely restored to flush alignment in the majority of cases. One comparative study found flush subchondral restoration in only 3 of 14 patients after marrow stimulation, against 10 of 14 after fixation-based repair (p=0.02). Disruption of this layer may accelerate longer-term joint wear.

For small, low-stress lesions, microfracture has a legitimate historical role, and short-term outcomes in that size range remain reasonable. But the evidence has moved on — and once a lesion crosses certain size and location thresholds, the case for a more durable repair becomes compelling.

The defect size thresholds that shift the recommendation

Three size thresholds help make sense of when the evidence tips in favour of OATS — and they form a logical ladder rather than a set of rigid cut-offs.

The most clinically significant boundary sits at 150 mm². Beyond this area, arthroscopic marrow stimulation carries a significantly high failure rate. In a cohort of 19 patients with large cystic lesions exceeding this threshold, primary OATS produced mean AOFAS scores rising from 66.8 to 88.7 and VAS pain scores falling from 6.95 to 2.6, with 80% of patients rating their result as excellent or good at nearly seven years' follow-up. Notably, the bony cysts had resolved completely on radiographs in every case — an outcome microfracture is poorly equipped to achieve, since it addresses only the cartilage surface rather than the underlying bone void.

A second, softer inflection point sits around 100 mm². Lesions above this area tend to transmit substantially greater shearing forces to adjacent cartilage and are generally more symptomatic — a meaningful shift in the biological environment that may justify stepping up to OATS even before the 150 mm² mark is reached.

For completeness, finite element modelling identifies a biomechanical inflection as early as roughly 36 mm² (approximately 6 × 6 mm), where ankle joint stiffness changes measurably during midstance and push-off. This finding underscores how sensitively the ankle responds to focal defects — though clinically, this lower threshold still falls within the range where microfracture may be appropriate for otherwise straightforward lesions.

The practical guide that emerges: microfracture may remain reasonable for small lesions below roughly 100–150 mm²; OATS is favoured once lesions exceed that range, have a cystic component, or sit in high-stress zones regardless of surface area.

How lesion location on the talar dome affects the decision

Where a lesion sits on the talar dome is a second, independent variable that works alongside size to guide the choice of repair strategy.

The dome is not loaded uniformly. In silico modelling — computer simulations of walking mechanics — shows that certain zones lose stiffness far more sharply when a defect is present. Zone 8 (mid-posterior) registers a −49% stiffness change with a 9 mm defect; zone 7 (posteromedial) drops by −33%; zone 3 (anterolateral) by −24%. In plain terms, the dome bends more under walking load in these regions, and a cartilage gap there carries a disproportionate mechanical consequence. For lesions in these high-stress, medial and posterior zones, restoring native hyaline cartilage — as OATS provides — has a stronger biomechanical rationale than accepting the fibrocartilage produced by marrow stimulation.

Zone 9, the posterolateral corner, is the least mechanically sensitive area on the dome. Lesions confined to this zone may remain appropriate for microfracture, particularly when they also fall below the size thresholds discussed above.

The clinical weight of zone distribution becomes clearer when the epidemiology is brought in: approximately 73% of talar OLTs are medial — 31% centromedial and 28% posteromedial. The majority of patients presenting with an OLT will therefore have a lesion in precisely the zones where OATS carries the greatest biomechanical advantage. Published case series add a further refinement: anteromedial lesions respond slightly better to mosaicplasty than anterolateral ones, a consideration when planning the approach.

What OATS mosaicplasty involves and what recovery looks like

During the procedure, a surgeon removes one or more small cylindrical cores — typically 6–10 mm in diameter — from a low-load region of the knee, usually the peripheral edge of the femoral condyle where cartilage sees little stress during normal walking. Each core contains both the surface cartilage layer and a column of supporting bone beneath it, giving the graft structural depth as well as a durable surface. Those plugs are then press-fitted into precisely sized channels drilled into the talar defect, designed to sit flush with the surrounding dome cartilage.

The transferred tissue is hyaline or hyaline-like cartilage, meaning it closely matches the mechanical properties of native talar cartilage — the material advantage over marrow stimulation that the opening section of this article described.

The donor knee: a trade-off that deserves honest discussion

Harvesting from the knee is the central clinical compromise, and it warrants direct discussion before any decision is made. The harvest site heals, but patients should expect some post-operative knee discomfort alongside their ankle recovery. A 2025 case series of 21 patients with Bristol/Hepple grade ≥3 OLTs, followed over three to four years, reported no graft failures and no severe donor-site complications — a reassuring signal from a relatively small cohort. The authors and the wider literature acknowledge, however, that donor-site morbidity remains an underreported outcome in talar-focused studies, so the full picture is not yet clearly established.

Younger patients with acute sports-related injuries tend to achieve the most favourable recovery trajectories in published series. Outcome data specific to ankle OATS beyond ten years are more limited than for equivalent knee procedures, and the realistic timeframe for returning to full activity should be discussed individually at consultation.

Objective recovery tracking can usefully supplement self-reported scores between clinic visits; markerless motion-capture technology — such as the UKCA-registered MAI Motion® system used at MSK Doctors — can quantify gait mechanics at baseline and at follow-up appointments, giving both patient and clinician a measurable picture of functional progress.

Which patients are best placed for mosaicplasty — and what to address first

Candidacy for mosaicplasty narrows when the clinical picture as a whole is considered, not just the scan report. The 2025 Libyan case series — mean patient age 30 years, with sports trauma as the dominant cause — confirms what the broader literature suggests: focal, acute-onset lesions in younger, physically active patients yield the strongest and most consistent results. When the damage is contained and the joint is otherwise healthy, the graft integrates into a mechanically favourable environment.

The single most important pre-operative variable after lesion size and zone is lower-limb alignment. Varus malalignment — where the mechanical axis tips the load inward onto the medial dome — does not necessarily disqualify a patient from mosaicplasty, but leaving it uncorrected substantially undermines the result. A 44-patient study published in 2025 found that postoperative VAS pain scores averaged 4.4 in patients with varus alignment compared with 1.5 in those with neutral or valgus alignment (p<0.001), a near-threefold difference — and crucially, cartilage fill quality on MOCART 2.0 MRI scoring was equivalent between groups. The graft appeared to heal; the pain did not resolve because the mechanical loading remained abnormal. Combining mosaicplasty with a corrective osteotomy to restore alignment is therefore a recognised, standard-of-care step rather than an unusual escalation, and weight-bearing alignment radiographs should be obtained as a routine part of pre-operative planning.

Mosaicplasty addresses focal, contained defects — typically up to 2 cm², with a mosaic arrangement extending the ceiling to approximately 4 cm². Diffuse or advanced osteoarthritis affecting multiple compartments is outside the scope of cartilage restoration, and a different pathway — joint preservation or replacement — applies in those cases.

When the defect is too large for OATS — and what evidence currently supports

For defects that exceed what a single or mosaic autograft can physically fill — typically beyond 4 cm² — fresh osteochondral allograft (OCA) provides the next sequential option rather than a fallback after failure. A systematic review covering 12 studies and 191 patients (mean age 37.5 years) reported 86.6% graft survival at a mean follow-up of 56.8 months, with significant improvements in both AOFAS and VAS scores across all included studies. OCA removes the donor-site trade-off entirely, though tissue availability and processing requirements introduce their own logistical considerations.

What the evidence base actually supports — and where it is heading

The size thresholds and zone priorities described in this article rest on a foundation of retrospective cohorts, biomechanical modelling, and systematic reviews rather than head-to-head randomised controlled trials comparing OATS directly against microfracture in talar OLTs. That distinction matters clinically: the thresholds are well-supported and consistently reproduced across independent lines of evidence, but they remain open to refinement as prospective data accumulate. A convergence of biomechanical, histological, and outcomes evidence pointing to the same 100–150 mm² boundary is a stronger signal than any single study alone — and it is actionable now.

Long-term ankle-specific OATS data beyond ten years are sparser than the equivalent knee literature, and 2026 publications signal that researchers are actively revisiting where microfracture's useful role ends. The picture is clear enough to guide individual decisions today; it is not yet closed.

For patients with a focal talar OLT — particularly those with a lesion above 150 mm², a high-stress zone, or prior marrow-stimulation failure — the evidence consistently points toward osteochondral restoration as the more durable path. Translating that into a specific plan requires imaging review and alignment assessment with a consultant.

1. [1] Effects of Losartan and Fisetin on Microfracture-Mediated Cartilage Repair of Ankle Cartilage in a Rabbit Model. (2024). https://doi.org/10.1177/03635465241285902 https://doi.org/10.1177/03635465241285902
2. [2] Osteochondral Lesion of Talus Treated by Mosaicplasty from the Knee as Donor Site, Orthopedic Surgery Ward. Tobruk Medical Center, L. (2025). https://doi.org/10.54361/ljmr.19.2.25 https://doi.org/10.54361/ljmr.19.2.25
3. [3] Long-term evaluation of Primary Osteochondral Autograft Transfer System (OATS) for large cystic type osteochondral lesion of talus. (2018). https://doi.org/10.1177/2473011418s00440 https://doi.org/10.1177/2473011418s00440
4. [4] A Systematic Review of Fresh Osteochondral Allograft Transplantation for Osteochondral Lesions of the Talus. (2020). https://doi.org/10.1177/2473011420S00382 https://doi.org/10.1177/2473011420S00382
5. [5] Osteochondral Lesion of the Talus: Quality of Life, Lesion Site, and Lesion Size. (2023). https://doi.org/10.1016/j.fcl.2023.07.007 https://doi.org/10.1016/j.fcl.2023.07.007
6. [6] The subchondral bone healing after fixation of an osteochondral talar defect is superior in comparison with microfracture. (2017). https://doi.org/10.1007/s00167-017-4654-z https://doi.org/10.1007/s00167-017-4654-z
7. [7] In silico assessment of talus osteochondral lesion size and location on biomechanical load distribution using tilted talar dome planes. (2025). https://doi.org/10.5152/j.aott.2025.25444 https://doi.org/10.5152/j.aott.2025.25444
8. [8] The effect of talus osteochondral defects of different area size on ankle joint stability: a finite element analysis. (2021). https://doi.org/10.1186/s12891-022-05450-2 https://doi.org/10.1186/s12891-022-05450-2