Why shoulder cartilage damage often goes undertreated

Roughly half of younger patients diagnosed with glenohumeral osteoarthritis receive no non-pharmacological care at all — a 2025 population analysis of 21,369 people with glenohumeral OA (mean age 50.3 years) found that figure sits at 54%. For many, the gap between 'take some anti-inflammatories and wait' and 'you need a shoulder replacement' is left unfilled, not because nothing exists, but because awareness of joint-preservation options at this stage remains limited.

This is the clinical window that matters: damage significant enough that watchful waiting provides diminishing returns — Grade III focal cartilage loss, where more than half the cartilage thickness is gone or subchondral bone is exposed — yet still localised enough that total shoulder arthroplasty is disproportionate. ChondroFiller injection is designed for precisely this stage: an ultrasound-guided, outpatient collagen scaffold that recruits the patient's own progenitor cells to support repair of focal defects, before the damage spreads to the point where preservation is no longer possible. Whether a patient falls within that window depends on the grade and pattern of their damage — not symptoms alone.

What Grade II–III glenohumeral damage means for your options

Cartilage grading under the Outerbridge and ICRS systems runs from 0 (intact) through to IV (full-thickness loss with exposed subchondral bone). Grade II–III sits in the middle: at Grade II, the surface is softened and fissured but more than half the cartilage thickness remains; by Grade III, that majority is gone or early bone exposure is visible at the base of the lesion. The damage is real and progressive, but in a focal Grade III defect it remains spatially contained — a discrete crater rather than a swept landscape.

That focal-versus-diffuse distinction is the central eligibility question. Defects of roughly 1.5–6 cm² with structurally sound cartilage at the borders define the candidacy window for ChondroFiller injection, with most published evidence centred on lesions in the 2–3 cm² range. Once damage becomes diffuse — bone-on-bone changes across multiple joint surfaces rather than a single delimited area — the preservation window closes, and shoulder arthroplasty becomes the more appropriate discussion.

The glenohumeral joint carries lower axial load than the hip or knee, but its rotational envelope is considerably wider. That combination means the cartilage rim bordering a focal defect faces varied, multi-directional stress with every arm movement. Intervening while that rim remains structurally sound matters more here than in joints with simpler load paths: once the surrounding cartilage degrades, a focal picture can become diffuse, and the preservation window described above is no longer available.

How ChondroFiller injection works in the shoulder

The treatment is delivered in a single outpatient appointment under ultrasound guidance — no operating theatre, no general anaesthesia, and no need to harvest the patient's own cells beforehand. ChondroFiller is a sterile, acellular Type I collagen liquid that self-sets into a biodegradable hydrogel scaffold within three to five minutes of being placed into the cartilage defect.

Once in position, the scaffold does not repair the defect directly. Instead, it acts as a structured environment that the body's own progenitor cells — drawn from the surrounding synovium and subchondral bone — migrate into, mature within, and use to deposit new cartilage matrix over a period of six to twenty-four months. This process is termed matrix-induced chondrogenesis. An ex vivo osteochondral model provided mechanistic confirmation: ChondroFiller produced a 2.4-fold increase in DNA content within the scaffold by day 14, consistent with active progenitor cell recruitment rather than passive space-filling.

Because the mechanism relies on cell migration rather than an injected cell population, there is no upper age limit for candidacy and no requirement for a prior surgical stage to harvest or culture tissue. Autologous chondrocyte implantation (ACI) — the established surgical approach for larger glenohumeral defects — involves two separate procedures, restricted laboratory availability, and theatre-based delivery. ChondroFiller injection, by contrast, is completed in a single session at the London Cartilage Clinic on Harley Street. The glenohumeral joint is a named manufacturer indication for this CE-marked Class III medical device.

What studies show about ChondroFiller at this stage of damage

The hip cohort represents the closest structural analogue for glenohumeral outcomes. In a prospective series of 26 adults who received ChondroFiller gel for acetabular cartilage defects greater than 2 cm², 17 of 21 patients assessed at 12–60 months achieved good or excellent results; two required total hip replacement. Those numbers give a real-world outcome floor: not all patients do well, and at that end of the spectrum, joint replacement remains the path forward.

What sharpens the hip data for clinical decision-making is the Tönnis grading finding. Patients who carried pre-existing Tönnis 2–3 osteoarthritis — meaning widespread degenerative change rather than focal damage — had uniformly poor results, regardless of how their individual lesion appeared in isolation. Shoulder-specific clinical sources cite this finding directly as applicable to glenohumeral candidacy: if the joint as a whole is substantially degenerated, the collagen scaffold cannot compensate for the absence of a structurally sound surrounding environment. That is the hard boundary the hip data establishes.

Across other joints, the pattern is consistent. A controlled multicentre knee RCT (n=13 ChondroFiller patients) showed significantly improved IKDC scores at 3, 6, and 12 months, with no adverse events; MRI at week 4 confirmed good defect filling, with progressive cartilage maturation visible through week 52. In the wrist, a prospective study of 25 treated patients demonstrated significantly better cartilage quality at follow-up arthroscopy compared with controls — median Outerbridge scores of 1.5 versus 3.0 (p=0.006) and ICRS scores of 1 versus 3 (p=0.002).

No shoulder-specific randomised controlled trial exists for ChondroFiller in the glenohumeral joint. The evidence base for this application rests on a CE-marked manufacturer indication and inference from the hip and knee data above — the hip cohort being the structurally closest analogue given its ball-and-socket architecture, though the glenohumeral joint's considerably wider rotational demands are a meaningful biomechanical difference that any extrapolation should acknowledge.

Who is suitable and who is not

Suitability turns on two questions: is the joint environment sound enough to support the scaffold, and is the defect still delimited rather than diffuse? The defect window itself — Outerbridge or ICRS Grade III–IV, focal rather than widespread — was set out in the previous section. What matters here is what rules a patient out.

MRI is a clinical prerequisite, not a procedural formality. Accurate defect sizing is necessary before injection planning begins, both to confirm the lesion falls within the treatable range and to verify that surrounding cartilage is intact — the scaffold integrates into that adjacent tissue, so its condition directly affects the outcome.

Age is not a disqualifying factor. Because ChondroFiller relies on progenitor cell migration rather than transplanted or cultured cells, the candidacy criteria are driven by defect pattern and joint status, not the patient's age.

The firm contraindications are:

• Widespread OA with bone-on-bone changes across multiple joint surfaces
• Ligamentous instability or gross mechanical malalignment
• Tönnis 2–3 equivalent diffuse OA — the hip cohort data, applied by shoulder clinicians, established that this degree of background joint degeneration produces consistently poor outcomes regardless of the individual lesion
• Active joint infection
• Known allergy to collagen or murine-derived proteins

Technique precision also matters: the wrist study found fibrous tissue formation occurred only in overfilled defects, not in flush applications. That is why specialist assessment and accurate defect characterisation before injection are clinically important, not advisory.

What to expect from assessment through recovery

Recovery starts the day of the injection, but the scaffold requires time to stabilise. In vitro loading data confirm that ChondroFiller does not immediately protect the opposing joint surface — the collagen matrix needs several weeks to integrate with surrounding tissue before the shoulder can be loaded normally. During that window, heavy lifting and overhead activities are typically avoided; lighter daily tasks are generally resumed sooner, but return to demanding rotational sport is a longer progression measured in months rather than weeks.

The outcome timeline mirrors the hip cohort data: meaningful functional gains were recorded across a 12–60-month follow-up period, reflecting how matrix-induced chondrogenesis works — progenitor cells continue maturing within the scaffold over months, so improvement accumulates rather than arrives all at once.

Technical precision at the point of injection shapes this entire trajectory. The wrist study showed fibrous tissue risk rises with overfilling; flush, accurately sized placement does not carry the same risk. For the 54% of younger glenohumeral OA patients who currently receive no structured treatment, a well-characterised defect and correctly placed scaffold are the prerequisites for any worthwhile outcome. Liquid Cartilage™ is available in the UK at the London Cartilage Clinic on Harley Street, where Professor Paul Y. F. Lee leads the programme; assessment appointments can be requested at londoncartilage.com.

1. [1] Management of glenohumeral osteoarthritis in the younger patient: population-level analysis. (2025). https://doi.org/10.1016/j.ocarto.2025.100705 https://doi.org/10.1016/j.ocarto.2025.100705
2. [2] Arthroscopic utilization of ChondroFiller gel for hip articular cartilage defects: cohort with 12–60-month follow-up. (2021). https://doi.org/10.1093/jhps/hnab002 https://doi.org/10.1093/jhps/hnab002
3. [3] Controlled randomized multicenter study: ChondroFiller liquid vs microfracturing for focal knee cartilage defects. (2016). https://doi.org/10.5348/VNP05-2016-1-OA-1 https://doi.org/10.5348/VNP05-2016-1-OA-1
4. [4] Minced Cartilage Procedure for One-Stage Arthroscopic Repair of Chondral Defects at the Glenohumeral Joint. (2021). https://doi.org/10.1016/j.eats.2021.03.012 https://doi.org/10.1016/j.eats.2021.03.012
5. [5] Cartilage reconstruction using Chondrofiller in intra-articular distal radius fractures. (2025). https://doi.org/10.1186/s42836-025-00333-y https://doi.org/10.1186/s42836-025-00333-y
6. [6] Influence of cartilage defects and collagen gel on integrity of adjacent cartilage: biomechanical in-vitro study. (2024). https://doi.org/10.1007/s00402-024-05530-z https://doi.org/10.1007/s00402-024-05530-z
7. [7] Development of an Ex Vivo Osteochondral Biomimetic Platform for Investigation of Cartilage Regeneration. (2025). https://doi.org/10.3390/ijms262311759 https://doi.org/10.3390/ijms262311759