What happens at two years — and why it matters

For many patients, the first twelve months after microfracture are genuinely encouraging — pain settles, activity returns, and the scores look much the same as those produced by newer scaffold approaches. The problem is what happens next.

Between eighteen and thirty-six months, a consistent pattern emerges in the clinical data: microfracture results plateau or begin to retreat, while outcomes following a ChondroFiller injection tend to hold or continue improving. A 92-patient multicentre randomised trial illustrates the early illusion clearly — IKDC gains at twelve months were statistically equivalent across groups, yet arthroscopic assessment at the same timepoint already showed more fully regenerated cartilage in the non-microfracture arm. Functional parity at one year had masked a divergence already forming in the tissue.

This is not a statistical quirk. The two approaches deposit fundamentally different material into the defect, interact differently with the underlying bone, and trigger different biological cascades. The sections that follow explain each of those mechanisms and trace how they produce the outcomes the evidence consistently records.

Why microfracture produces the wrong kind of cartilage

Microfracture works by drilling small holes through the base of a cartilage defect into the bone beneath, releasing marrow blood and stem cells that clot within the lesion. That clot eventually matures — but not into the tissue it replaces. The repair product is fibrocartilage: dominated by Type I collagen and lacking the dense proteoglycan architecture that gives native hyaline cartilage (Type II collagen) its load-bearing stiffness. A 2019 review of marrow stimulation studies confirmed this as the consistent finding — mechanically inferior collagen I/II fibrocartilage is the principal outcome, not a subset result.

The cellular explanation for why fibrocartilage eventually fails under load came into sharper focus with a 2024 rat study. Fibrocartilage chondrocytes derived from bone marrow stimulation behaved normally under physiological conditions — indistinguishable from native cells in standard culture. But apply low-dose oxidative stress, the kind generated by repetitive joint loading, and fibrocartilage chondrocytes showed a sharp drop in viability and a rise in apoptosis. Native hyaline chondrocytes were unaffected at the same dose. This provides a plausible cellular mechanism for the clinical pattern described above: the tissue survives early rehabilitation but erodes under the ordinary demands of daily activity.

Microfracture also carries a secondary risk specific to the technique itself. Breaching the subchondral plate — the calcified boundary between cartilage and bone — can trigger long-term bone remodelling into the repair zone, or cyst formation, both of which undermine whatever fibrocartilage fill has formed. The technique retains a legitimate historical role for small lesions in carefully selected patients, but these compounding structural limitations explain why it is no longer regarded as the preferred approach for active adults with focal defects.

How ChondroFiller injection takes a different route

Unlike microfracture, which requires drilling into bone under general or regional anaesthesia in an operating theatre, ChondroFiller injection is delivered as an outpatient procedure guided by real-time ultrasound — no surgical incision, no theatre booking. For a patient weighing up options, that difference in access and recovery context is often the first thing that matters.

The biological route taken is equally distinct. ChondroFiller (Meidrix Biomedicals GmbH) is a CE-marked, cell-free Type I collagen hydrogel — an acellular scaffold that, once placed within the defect, acts as a three-dimensional framework the body's own repair cells migrate into. Rather than flooding the lesion with marrow blood and relying on the resulting clot to differentiate, the scaffold supports matrix-induced chondrogenesis: the patient's own bone marrow stromal cells (BMSCs) move into the collagen matrix and begin to lay down repair tissue. There is no breach of the subchondral plate at any point. That distinction matters structurally — the subchondral bone remains intact, sidestepping the risk of bony intrusion or cyst formation that can undermine microfracture repairs over time.

The scaffold mechanism also appears to translate across joints. A 2025 prospective study by Demmer (n=25) examined ChondroFiller applied to chondral defects in the wrist following intra-articular distal radius fractures. At follow-up arthroscopy, treated patients had significantly better cartilage quality than controls — median Outerbridge score 1.5 versus 3.0 (P=0.006), ICRS 1 versus 3 (P=0.002). The finding is not large enough to anchor outcome expectations, but it does suggest the scaffold's recruitment mechanism is not knee-specific.

What the outcome data actually shows over five years

The trajectory follows a recognisable arc. Both approaches tend to produce similar functional scores at twelve months — then the paths diverge.

Microfracture scores in published series typically peak between twelve and eighteen months, then plateau or retreat as the fibrocartilage fill yields under cyclical load. The clearest long-term benchmark is a five-year RCT comparing costal chondrocyte autologous chondrocyte implantation (CCP-ACI) against microfracture. By year five, MOCART scores — a validated MRI measure of repair tissue quality, where 100 represents structurally normal cartilage — stood at 26.7 for microfracture versus 62.3 for ACI (P<0.0001). Lysholm scores, a 0–100 knee function scale, were 64.9 versus 84.5 (P=0.023); KOOS, a broader knee outcome index, reached 303.0 against 390.9 (P=0.017). Worth stating upfront: this five-year dataset compares ACI with microfracture, not ChondroFiller injection. Its relevance is mechanistic — both ACI and ChondroFiller injection avoid fibrocartilage formation and preserve the subchondral plate, so the five-year arc provides the most instructive available proxy for the endpoint the tissue-type difference is heading towards.

That the gap opens before five years is evident even in earlier data. In a 92-patient multicentre RCT, BiCRI (a biphasic cartilage repair implant) and microfracture produced statistically equivalent IKDC improvement at twelve months — 25.56 versus 27.51. Arthroscopic assessment already showed more fully regenerated cartilage in the BiCRI arm. Twelve-month functional equivalence was not outcome equivalence.

ChondroFiller injection data sits at a shorter horizon, and that asymmetry deserves plain acknowledgement. Published knee series report average IKDC scores rising from approximately 48 before treatment to approximately 80 at three years — a 30-point gain that corresponds to a shift from moderate activity limitation to near-normal daily function — with patient satisfaction exceeding 80% in most cohorts. Simeonov 2024 (n=17, mean age 31) confirmed statistically significant IKDC and Lysholm gains at three, six, and twelve months; the improvement curve flattened between six and twelve months rather than retreating. These are prospective series, not large randomised trials. The data is consistent and clinically meaningful; a large RCT with follow-up beyond three years is still pending.

Which patients see the biggest difference

The gap in outcomes is not uniform across patients — two modifiers shift the risk calculus in ways worth understanding before choosing a pathway.

Age is the clearest. Microfracture depends entirely on the patient's own bone marrow stromal cells, and BMSC stemness — the capacity for chondrogenic differentiation — declines significantly with age. Published data confirm this clinically: microfracture outcomes in older patients are significantly worse than in younger cohorts, a pattern that correlates directly with reduced BMSC repair capacity. For a patient in their mid-fifties, the biological substrate microfracture relies on is already diminished; for a patient in their early thirties — the mean age in Simeonov's ChondroFiller knee series — the repair environment is meaningfully different.

Defect characteristics matter too, though not in a straightforward head-to-head way. Microfracture was historically indicated for lesions under 2 cm²; ChondroFiller injection is indicated for focal defects up to 3 cm², extendable to 6 cm², in joints that are otherwise structurally healthy. That last qualifier is the important one: ChondroFiller injection is not appropriate for diffuse or advanced osteoarthritis, which means the two populations are not equivalent. Any direct outcome comparison carries that methodological caveat.

For patients who have already had microfracture, the subchondral plate disruption it creates can narrow the options for subsequent cell-based procedures — a practical planning consideration rather than a barrier, but one worth raising early with a specialist.

Scaffold-augmented variants of microfracture — incorporating BMP delivery or ECM scaffolds to steer repair toward hyaline-like tissue — are under investigation, but none yet has head-to-head data against ChondroFiller injection at the two-to-three-year horizon where the outcome divergence becomes clinically relevant.

Getting an assessment that looks past the 12-month picture

The practical question this evidence ultimately raises is not which approach produces better twelve-month scores — both often do — but whether the repair tissue holds once the joint returns to normal load. For patients with a focal cartilage defect in an otherwise structurally healthy joint, that shifts the decision criteria: repair durability, not just early symptom relief, should drive technique selection.

Meeting that standard of decision-making requires more than tracking pain scores over time. Specialist cartilage MRI, defect mapping, and assessment of the subchondral plate provide the structural detail needed to match a lesion to the repair pathway most likely to sustain its benefit. It is also the right setting in which to weigh what the current evidence confirms at two to three years against what larger, longer-term trials have yet to establish — a distinction that belongs in any honest pre-treatment discussion. For patients seeking that level of assessment in London, a consultation can be arranged at londoncartilage.com on Harley Street.

1. [1] IMPLANTATION OF CHONDROFILLER LIQUID® AS A SCAFFOLD MATERIAL FOR THE TREATMENT OF CHONDRAL LESIONS OF THE KNEE JOINT. (2024). https://doi.org/10.5272/jimab.2024304.5936 https://doi.org/10.5272/jimab.2024304.5936
2. [2] Comparison of the bioactivity and apoptotic responses of hyaline cartilage chondrocytes and fibrocartilage chondrocytes obtained by bone marrow stimulation in rats. (2024). https://doi.org/10.5152/j.aott.2024.22172 https://doi.org/10.5152/j.aott.2024.22172
3. [3] Articular fibrocartilage - Why does hyaline cartilage fail to repair?. (2019). https://doi.org/10.1016/j.addr.2018.12.015 https://doi.org/10.1016/j.addr.2018.12.015
4. [4] Costal Chondrocyte–Derived Pellet-Type Autologous Chondrocyte Implantation Versus Microfracture: 5-Year Follow-up of a Prospective Randomized Trial. (2024). https://doi.org/10.1177/03635465231222797 https://doi.org/10.1177/03635465231222797
5. [5] SPI1 facilitates microfracture-mediated cartilage regeneration in the elderly by enhancing bone marrow stromal cells stemness. (2025). https://doi.org/10.1177/20417314241311073 https://doi.org/10.1177/20417314241311073