Why two grading systems exist — and which one applies to you

If your scan report mentions 'Stage III' or your surgeon has used terms like 'Grade IV cartilage', it is natural to wonder whether those words automatically mean an operation. The short answer is: not necessarily — and understanding why requires knowing that your clinical team may actually be using two different grading systems, designed for two different moments in the care pathway.

The Berndt–Harty system was developed in 1959 as a way of describing what a talar osteochondral lesion (OLT) looks like on imaging — originally plain X-ray, later extended to MRI by Anderson and colleagues in 1989. Its four stages describe whether the affected fragment of bone and cartilage is compressed, partially detached, fully detached but in place, or displaced. This is the system most likely to appear in a radiology report, and it guides the first major fork in the road: conservative management versus surgical referral.

The ICRS system, by contrast, measures cartilage depth — how far the damage penetrates through the cartilage layer toward the underlying bone. It is typically applied during arthroscopy or derived from detailed MRI sequences, rather than from a standard scan report. Its grades run from 0 (normal) to IV (full-thickness loss down to bare bone), with four subtypes at Grade III tracking precisely how close the lesion is to the subchondral plate.

Your consultant may draw on both. What the scan shows (Berndt–Harty stage), what the cartilage depth looks like (ICRS grade), the size of the defect in square centimetres, how long you have had symptoms, and your age and activity level are all factored into the same decision. Grade alone does not determine your treatment — but knowing what each system is measuring helps you follow that conversation clearly. The sections below explain each system and what it tends to mean in practice.

Berndt–Harty stages I–IV: what plain imaging and MRI reveal

Each Berndt–Harty stage describes a point on a spectrum from early bone stress to a loose fragment inside the joint — and each carries a different default management path.

Stage I — subchondral compression. The bone beneath the cartilage has been compressed but no fragment has separated. Plain X-ray often appears normal; an MRI showing bone marrow oedema is usually needed to confirm the diagnosis. Protected weight-bearing and immobilisation — typically for six to twelve weeks — is standard first-line care, and many Stage I lesions settle with this approach alone.

Stage II — partial detachment. The fragment has begun to lift but remains tethered. A conservative trial of similar length is still appropriate at this stage, and surgery is generally considered only if symptoms persist after that window.

The Anderson MRI extension: IIA and IIB

Anderson and colleagues' 1989 modification of the system introduced two clinically important subgroups at Stage II that are visible on MRI but invisible on plain X-ray.

• Stage IIA describes a subchondral cyst beneath an intact cartilage surface. Because the overlying cartilage is undamaged, this pattern is a candidate for retrograde drilling — accessing the cyst from the underside of the bone so the cartilage cap is preserved rather than disturbed.
• Stage IIB describes the same cyst finding but with an open articular surface. The intact cap is gone, which shifts the approach toward anterograde debridement rather than cartilage-sparing drilling.

This distinction matters because two patients with superficially similar imaging can require fundamentally different surgical routes.

Stage III — fully detached, in situ. The fragment is completely free but has not migrated from its crater. This is the main inflection point in the system. Stable fragments in younger patients with intact cartilage may still be offered a supervised conservative trial; however, most cases at this stage proceed to surgical intervention. Characterising it as uniformly operative would overstate the evidence.

Stage IV — displaced fragment. The fragment has moved out of position. Surgical management is appropriate regardless of patient age, activity level, or symptom duration — the mechanical problem cannot resolve without addressing the displaced piece.

ICRS grades 0–IV: cartilage depth at closer assessment

Think of the ICRS scale as a depth ruler applied to the cartilage surface rather than a system for describing the whole lesion. Where Berndt–Harty answers 'has the fragment moved?', ICRS answers 'how far through the cartilage has the damage gone?' That distinction shapes how the two systems are used in combination. Importantly, an ICRS grade is usually assigned during arthroscopy or derived from detailed MRI sequences such as T2 mapping — it will not appear on a standard X-ray report.

Grades 0 and I — normal or superficial. Grade 0 is a structurally normal cartilage surface; Grade I describes superficial softening or fissuring without meaningful depth involvement. Neither typically requires anything beyond load management and targeted rehabilitation.

Grade II — less than 50% of cartilage depth. Damage reaches less than halfway through the cartilage layer but does not threaten the subchondral bone beneath. Depending on symptoms and defect size, this may be observed, managed conservatively, or addressed with minimally invasive measures.

Grade III — more than 50% depth, subchondral bone not yet exposed. This is the threshold at which cartilage repair is most often considered. Four subtypes (IIIa–d) track how close the lesion has come to the calcified cartilage layer and the subchondral plate. Subtypes IIIa and IIIb sit above or at the calcified layer; IIIc has crossed through it; IIId presents as surface blistering overlying a deep lesion. The practical significance of IIIc and IIId is that even though the bone may appear intact on imaging, the damage is close enough to the subchondral plate that surgeons often treat these subtypes similarly to Grade IV.

Grade IV — full-thickness loss, bone exposed. The cartilage is entirely absent across the defect. Intervention at this stage cannot address the cartilage surface alone — the subchondral bone must also be treated, whether through marrow stimulation, a regenerative scaffold, or a more complex reconstruction.

One important limitation: the ICRS scale says nothing about fragment displacement or subchondral cyst morphology — precisely the information that Berndt–Harty and the Anderson MRI modification supply. Both systems together give a more complete picture than either provides alone.

From grade to treatment: how the decision path actually works

Grade alone does not determine which treatment a patient receives — it sets a threshold, and defect area then does most of the remaining work. The practical algorithm runs in two phases: grade (Berndt–Harty or ICRS) decides whether conservative care can continue or surgery is needed; size in cm² decides which surgery.

Conservative management: grades I and II

As established in the staging discussion above, Berndt–Harty Stages I and II and ICRS Grades 0–II are the non-operative territory for most patients. Surgery is considered only after a supervised conservative programme has been given a fair trial — this section picks up from that threshold.

Small defects: bone marrow stimulation first

Once surgery is indicated, defects smaller than approximately 1.5–2 cm² are typically addressed by arthroscopic bone marrow stimulation — either microfracture or drilling through the subchondral bone to recruit repair cells. Good short-term outcomes are widely reported, though some evidence suggests results may decline over time at the larger end of this size range. Mosaicplasty (osteochondral autograft transfer, or OATS) has shown higher clinical scores than microfracture in longer follow-up, but it requires harvesting donor tissue from elsewhere in the joint, which carries its own morbidity — a trade-off that requires careful discussion.

Mid-range defects: 2–3 cm²

In this band, mosaicplasty remains an option, and single-stage regenerative scaffold approaches become competitive alternatives. Injectable collagen scaffold pathways — which deliver a matrix that recruits the patient's own progenitor cells to support cartilage regeneration — sit within this size range as a category of option distinct from marrow-stimulation techniques and from more complex cell-based surgery.

Larger defects: 3 cm² and above

For defects at or above 3 cm², the SUMMIT trial provided evidence that matrix-associated autologous chondrocyte implantation (MACI) produced better KOOS pain and function scores than microfracture at both two and five years. This finding has shifted practice at the larger end toward cell-based repair, though MACI is a staged surgical procedure with a longer recovery pathway.

Fragment fixation

Where a large fragment is judged biologically viable — intact enough to heal if repositioned — fixation with cortical bone pegs or compression screws offers the option of securing the native tissue rather than replacing it. This approach is reserved for cases where the fragment's biology supports union; it is not a general alternative to the size-based algorithm above.

The consistent theme across all four decision points is that grade and area work together. Neither variable alone predicts the right intervention; the surgeon's planning integrates both, alongside patient age, activity demands, and previous treatment history.

Which scan goes with which grade

A normal X-ray after an ankle injury is not reassurance — it is simply the beginning of the investigation. Plain radiographs miss Stage I lesions entirely: because the cartilage surface and the earliest subchondral compression show no bony fragment, there is nothing for the X-ray to detect. Stage II lesions are frequently underestimated for the same reason. If a patient's pain persists and the X-ray is unremarkable, that is not a contradiction — it is a prompt to look with a different tool.

MRI is the diagnostic standard for OLTs. It resolves cartilage signal changes, bone marrow oedema, and early subchondral haemorrhage — pathology that CT and plain film cannot show. This is why MRI underpins the Anderson-modified staging that surgeons use for clinical decision-making: without it, cystic lesions beneath an intact cartilage cap remain invisible, and the distinction between a conservative and an operative case may be missed.

CT serves a different purpose: pre-operative planning rather than diagnosis. Where MRI answers 'what is damaged and how deeply?', CT answers 'exactly how large is the bony defect, where does it sit, and can a surgeon reach it?' Supplementary CT taken in maximum plantarflexion — a technique described by van Bergen and colleagues — confirms arthroscopic access to posteromedial lesions that can otherwise be obscured.

In practice, most patients heading toward surgery will need both. The two scans answer different questions, and acting on one without the other risks either missing the diagnosis or planning an operation without sufficient anatomical detail.

What the grade doesn't capture — and why specialist assessment matters

Grading systems are tools, not verdicts. Berndt–Harty describes fragment displacement and bony integrity; ICRS describes cartilage depth — and neither was designed to stand alone. A Stage III Berndt–Harty finding says nothing about cartilage thickness; an ICRS Grade III result says nothing about whether a cyst is forming silently beneath an intact cartilage cap. Both gaps carry real clinical weight.

Patient-level factors widen the picture further. A 22-year-old with a high-demand sporting life and no prior treatment sits in a materially different conversation than a 55-year-old with identical imaging, a previous microfracture that failed, and reduced bone stock. The grade is unchanged; the appropriate path is not. Age, activity demands, prior conservative or surgical treatment, and bone quality all alter the algorithm before a surgeon reaches any size threshold.

The evidence base adds honest uncertainty. Studies linking Berndt–Harty or ICRS grade to long-term outcomes differ substantially in how they select patients, define comparators, and measure success — which means that quoting 'success rates by grade' draws more from the literature than it can currently support. Grade is a starting framework, not a prognosis.

What specialist assessment does is bring these threads together: both grading systems, defect area from CT and MRI, fragment viability, bone quality, and the full treatment history. The questions worth arriving with are practical — whether the cartilage cap over a cystic lesion is intact, whether bone stock is sufficient for stimulation-based repair, and whether the lesion is primary or a revision case. Those answers are what convert a stage or grade into a personalised plan.

For patients in London exploring that conversation, Liquid Cartilage™ — the injectable collagen scaffold delivered as an outpatient procedure — is available at the London Cartilage Clinic on Harley Street. Appointments can be booked at londoncartilage.com.

1. [1] Osteochondritis dissecans. https://en.wikipedia.org/?curid=3762029 https://en.wikipedia.org/?curid=3762029