A New Era for Knee Osteoarthritis: Genicular Artery Embolization with Rapidly Resorbable Microspheres Delivers Real Results
194 patients. 12 months of follow-up. A median pain score cut in half — with zero severe side effects. Here's the full breakdown of the landmark 2026 study.
If you or someone you love has been limping through life with knee osteoarthritis — told to just "manage the pain" until things get bad enough for a joint replacement — a study published in the June 2026 issue of Radiology (the official journal of the Radiological Society of North America) deserves your full attention. Researchers at Charité – Universitätsmedizin Berlin have just reported the results of a prospective clinical study on a minimally invasive procedure called Genicular Artery Embolization (GAE) — and the data is, frankly, impressive.
The study specifically examined a newer type of embolic agent: Rapidly Resorbable Gelatin-based Microspheres (RRGMs). These are tiny, engineered particles injected into the blood vessels supplying the diseased knee tissue, designed to temporarily block blood flow to abnormal tissue — and then dissolve within hours. The result? Clinically significant, sustained pain relief for at least 12 months, with a safety record that sets a new benchmark for the field.
This article breaks down what the study found, why it matters, how the procedure works, who it's appropriate for, and what the medical community is still working to understand. We've kept things accessible, but nothing has been dumbed down — because you deserve accurate information, not just reassuring headlines.
What Is Knee Osteoarthritis — and Why Is It So Hard to Treat?
Osteoarthritis of the knee is not simply "wear and tear." That's a description that has stuck around far too long and badly undersells the complexity of what's actually happening inside the joint. OA is now understood as a disease driven by chronic low-grade inflammation, pathological blood vessel growth (neovascularization), and the ingrowth of pain-sensing nerves along those new vessels.
That last point is critical: the same chemical signals that tell blood vessels to grow also attract sensory nerve fibers. When the knee's synovial lining becomes hypervascular — flooded with tiny new blood vessels — those nerves follow. The result is a joint that is exquisitely and persistently sensitized to pain. This is called the angiogenic-neural axis, and it explains why OA pain can be so disproportionate to what imaging actually shows.
This understanding is precisely what makes GAE such a biologically logical treatment. Instead of masking pain with medications or waiting for the joint to deteriorate to the point of replacement, GAE directly targets the hypervascular tissue driving the pain — shutting off the blood supply to the abnormal synovial tissue and, by extension, disrupting the neurovascular environment that perpetuates it.
Despite the scale of the problem — OA is a leading cause of disability worldwide, with a burden projected to continue rising through 2050 — patients have historically had few options between long-term anti-inflammatory use and major surgery. Physical therapy, corticosteroid injections, and hyaluronic acid infiltrations provide temporary relief for many patients, but they don't address the underlying vascular pathology. GAE does.
What Is Genicular Artery Embolization, and How Does It Work?
Genicular Artery Embolization is a minimally invasive, image-guided procedure performed by interventional radiologists. The term "embolization" refers to the deliberate blocking of a blood vessel — a technique that has been used in medicine for decades to treat conditions ranging from uterine fibroids to liver tumors. GAE applies this principle to the abnormal periarticular (around the joint) blood vessels responsible for synovial hypervascularity in the osteoarthritic knee.
The Procedure, Step by Step
In the Fleckenstein study, GAE was performed as follows. Under local anesthesia, the interventional radiologist gained access to the common femoral artery in the groin using a standard antegrade (downward) approach, placing a 4-French vascular sheath. Using a 4-French diagnostic catheter positioned in the distal superficial femoral artery, digital subtraction angiography (DSA) was performed to visualize the blood vessels around the knee. A microcatheter (1.7 or 1.9 French, paired with a 0.014-inch guidewire) was then navigated into each genicular artery supplying the synovium.
A vessel was selected for embolization when a hyperemic blush — a characteristic "stain" or flush of contrast agent indicating abnormally increased blood flow — was identified on DSA. In cases where no blush was visible but at least two of three secondary criteria were met (pain reproduction on contrast injection, vessel location within the symptomatic compartment, or evidence of collateral flow to a hypervascular vessel), that vessel was also treated.
The RRGMs were then injected slowly and under continuous fluoroscopic visualization until flow stasis was achieved and the hyperemic blush was eliminated on post-embolization DSA. Patients were admitted overnight and discharged following clinical examination and groin ultrasound.
- Access site: Ipsilateral antegrade femoral artery — 100% of cases
- Median vessels embolized per procedure: 4 (IQR: 3–4)
- Median RRGM volume per procedure: 6 mL (IQR: 4.3–8.1 mL)
- Median fluoroscopy time: 15.4 minutes (IQR: 10.4–20.4 min)
- Mean dose-area product: 1020.5 µGy·m²
- Technical success: 100% (239/239 procedures)
What Makes RRGMs Different from Previous Embolic Agents?
This is arguably the most important scientific question the study addresses — because GAE itself isn't new. What IS new is the specific material used to perform it. To understand why RRGMs represent a meaningful advancement, you need to understand the limitations of the two approaches that preceded them.
Imipenem-Cilastatin (IPM-CS): Effective but Complicated
The most widely used embolic agent in GAE has historically been imipenem-cilastatin — an antibiotic that, when mixed with contrast medium, forms crystalline particles ranging from 10 to 70 micrometers. IPM-CS's temporary embolic effect is thought to contribute to its favorable safety profile, and clinical studies have consistently shown meaningful, sustained symptom relief with it. However, it comes with significant practical and ethical problems. It is not universally available in all countries, its use in this context is entirely off-label, its particles are not calibrated in size (introducing procedural variability), and — most importantly — using an antibiotic as an embolic agent raises legitimate concerns around contributing to antimicrobial resistance, an issue the World Health Organization has flagged as a critical global threat.
Permanent Microspheres: Precise but Potentially Problematic
Calibrated permanent microspheres (such as tris-acryl polymer microspheres) were developed as a more standardized alternative. They offer consistent sizing, which improves procedural reproducibility, and clinical studies have shown they perform comparably to IPM-CS in terms of efficacy. The problem, however, is right there in the name: they're permanent. Non-resorbable embolic material that remains indefinitely in the vessels raises concerns about long-term vascular consequences, chronic foreign body inflammatory response, and a significantly higher rate of adverse skin events. In one published study, discoloration rates with permanent particles reached as high as 65% — a striking and clinically meaningful figure.
RRGMs: Bridging the Gap
Rapidly Resorbable Gelatin-based Microspheres (brand name: Nexsphere-F, 100–300 µm; NextBiomedical) were specifically engineered to combine the best attributes of both approaches. They are size-calibrated like permanent microspheres, offering reproducibility and predictability. But unlike permanent microspheres, they are designed to dissolve within hours of injection — making their embolic effect temporary, much like IPM-CS, while avoiding the antibiotic stewardship concerns associated with repurposing an antibiotic as an embolic agent.
In theory, this temporary vessel occlusion is sufficient to disrupt the pathological neurovascular environment — and the 2026 Fleckenstein study provides the first large-scale prospective data suggesting that theory holds in clinical practice.
Embolic Agent Comparison: IPM-CS vs. Permanent Microspheres vs. RRGMs
| Feature | IPM-CS | Permanent Microspheres | RRGMs (Nexsphere-F) |
|---|---|---|---|
| Particle Calibration | Uncalibrated (10–70 µm) | Calibrated | Calibrated (100–300 µm) |
| Resorbability | Temporary (transient) | Permanent | Dissolves within hours |
| Regulatory Status | Off-label use | Approved embolic agent | Purpose-built for GAE |
| Antimicrobial Risk | Yes (antibiotic repurposing) | None | None |
| Skin Discoloration Rate | Up to 18% | Up to 65% | 6.3% (this study) |
| Long-term Vascular Risk | Low (temporary) | Elevated concern | Low (temporary) |
| Clinical Efficacy (Pain) | Meaningful reduction | Comparable | Median NRS 7→3 at 12 mo |
The Study Design: Prospective, Rigorous, and Real-World
The study enrolled consecutive participants at a single center (Charité Berlin) between July 1 and November 30, 2024. All 194 participants had OA-related knee pain that was refractory to at least 3 months of conservative treatment, including anti-inflammatory medications, corticosteroid and/or hyaluronic acid injections, and physiotherapy. Every patient was reviewed by a multidisciplinary board combining orthopedic surgery and interventional radiology before being approved for GAE — a critical quality safeguard that ensured appropriate patient selection.
The five operating interventional radiologists had between 2 and 25 years of experience, which actually strengthens the external validity of the data: the results weren't achieved only by the most skilled hands in the room. Forty-five participants (23%) underwent bilateral treatment, with the second procedure performed within 4 weeks of the first.
Outcomes were tracked using two validated, patient-reported measures: the Numeric Rating Scale (NRS) for pain (0–10, higher = worse) and the Knee Injury and Osteoarthritis Outcome Score (KOOS), which evaluates five domains — pain, symptoms, activities of daily living, sports/recreation, and quality of life (0–100 for each, higher = better). Assessments occurred at baseline, 6 weeks, 3 months, 6 months, and 12 months. The 6-month check was performed in-person by an orthopedic surgeon.
The Results: What the Data Actually Shows
Pain Reduction
The headline finding is unambiguous. The median NRS pain score dropped from 7 (IQR 6–8) at baseline to 4 (IQR 2–5) at 6 weeks, and continued improving — stabilizing at 3 (IQR 2–5) at both 6 and 12 months. That represents a median reduction of 4 full points on a 10-point scale (95% CI: 3.5–4.0), and it is statistically significant at p < 0.001 at every follow-up timepoint.
Critically, 80% of participants achieved pain improvements exceeding the Minimum Clinically Important Difference (MCID) for NRS scores at 12 months. The MCID is a threshold defined by prior research as the smallest change in a score that patients themselves consider meaningful — in this case, a reduction of at least 2 points on the NRS. Reaching that benchmark in 4 out of 5 patients at the one-year mark is a clinically meaningful finding, not just a statistically significant one.
Functional Outcomes (KOOS)
Every single KOOS subscale showed significant improvement over the study period. The improvements were consistent, sustained, and all reached statistical significance (p < 0.001).
| KOOS Domain | Baseline Median | 12-Month Median | Change | % Achieving MCID at 12 mo |
|---|---|---|---|---|
| Daily Activities | 53 | 71.5 | +18.5 pts | ~69% |
| Sports & Recreation | 15 | 36 | +21 pts | ~80% |
| Symptoms | 51 | 68 | +17 pts | ~60% |
| Pain | 44 | 65 | +21 pts | ~69% |
| Quality of Life | 19 | 40 | +21 pts | ~55% |
The sports and recreation domain deserves particular mention — starting at a median of just 15 out of 100, it more than doubled by 12 months. This reflects how dramatically restricted physically active patients with knee OA can become, and how meaningful any real improvement in that domain is to their daily lives.
Safety Profile
This is where RRGMs particularly shine. Across 239 procedures, no moderate or severe adverse events were recorded. Total. The only adverse events observed were:
- Transient skin discoloration over the embolized area: 15 of 239 procedures (6.3%). All resolved spontaneously within 24 hours, without any sequelae — classified as Grade 1 (mild).
- One superficial groin hematoma (0.4%) — also Grade 1, also resolved without sequelae.
To put the skin discoloration rate in perspective: prior published data reports this complication occurring in 15.6% of GAE cases overall (from a recent meta-analysis of 657 patients). With permanent particles specifically, rates as high as 65% have been published. The 6.3% rate observed with RRGMs is the lowest reported in the field to date — and every single case resolved within a day.
How Does This Compare to Prior GAE Studies?
The authors contextualize their findings against the broader GAE evidence base, and the comparison holds up well. A 2024 systematic review and meta-analysis across 21 studies found significant pain improvement after GAE, with a weighted mean difference in visual analog scale scores of −36.2 points (95% CI: −43.0 to −29.5) and a mean KOOS pain improvement of 20.3 points (95% CI: 13.5–27.2). The present study falls within and at the upper end of that range.
Most tellingly, the MCID achievement rates in this study (55–69% across KOOS subscores) are directly comparable to those reported in the GENESIS trial — a landmark study using 100–300 µm nonresorbable tris-acryl polymer microspheres — which found 56–69% of participants achieved MCID across KOOS subscores at 3-month follow-up. The crucial difference is that the RRGM study follows participants for 12 months, and the favorable results are maintained. This suggests that you do not need permanent particles to achieve comparable efficacy — a significant finding for the field.
Who Is a Candidate for GAE?
Based on the inclusion and exclusion criteria used in this study, GAE with RRGMs appears most appropriate for patients who:
- Have confirmed knee osteoarthritis (KL Grade II, III, or IV) with corresponding pain and functional limitation
- Have failed at least 3 months of conservative treatment including anti-inflammatories, corticosteroid or hyaluronic acid injections, and physiotherapy
- Are not yet ready for, or are not good candidates for, total knee replacement
- Do not have rheumatoid or infectious arthritis
- Have adequate renal function (eGFR ≥ 30 mL/min/1.73 m²)
- Do not have significant peripheral arterial disease or severe axial knee deviation (>15°)
- Do not have therapy-refractory coagulopathy
The information in this article is educational and is not a substitute for individualized medical advice. If you think GAE might be appropriate for your situation, discuss it with your orthopedic surgeon and an interventional radiologist experienced in the procedure. GAE is not universally available yet, and patient selection requires expert assessment.
Acknowledged Limitations: What the Study Doesn't Tell Us Yet
The authors are commendably transparent about the boundaries of their data, and intellectual honesty demands we acknowledge them here too.
No control group: This is an observational study — there was no sham procedure or non-GAE comparison group. That means we cannot definitively separate the true procedural effect from placebo response, natural symptom fluctuation over time, or regression to the mean. Randomized controlled trials are the next essential step.
Single-center design: All procedures were performed at Charité Berlin, which is a highly specialized academic medical center. Outcomes at less-experienced centers or in different healthcare systems may differ, though the multi-operator design (5 operators, 2–25 years of experience) partially mitigates this concern.
No objective functional measures or imaging endpoints: Outcome data relied entirely on patient-reported scores. No validated objective functional tests or post-procedure MRI/ultrasound data were included. The authors note that no such validated parameters currently exist specifically for GAE, making patient-reported outcomes the accepted standard — but this remains a gap in the field.
Missing follow-up data: While follow-up rates were strong (94% at 6 weeks, declining to 79% at 12 months), 21% of participants had missing 12-month data. The authors used a complete-case analysis and censored missing data at each timepoint — an intention-to-treat analysis might yield more conservative estimates.
No data beyond 12 months: The study's observation period ends at one year. We do not yet know whether the benefits are maintained at 2, 3, or 5 years, or what the re-treatment rate will be. This is critical for positioning GAE relative to knee replacement in long-term clinical planning.
🎥 Watch · Explainer Video
See Genicular Artery Embolization in Action
Watch this video for a visual walkthrough of how GAE works — from the arterial anatomy of the knee to the embolization technique and what patients can expect.
Frequently Asked Questions
How long does the GAE procedure take?
Based on the study data, the median fluoroscopy time was approximately 15.4 minutes, though total procedure time (including preparation, anesthesia, and post-procedure assessment) will be longer. Patients were admitted overnight and typically discharged the following day after groin ultrasound confirmed no complications.
Is GAE painful? What is the recovery like?
The procedure is performed under local anesthesia, so patients are awake but the access site is numbed. Some patients experience a sensation of warmth or mild discomfort in the knee during contrast injection. Post-procedure, there may be temporary knee soreness, and the most commonly observed adverse event — transient skin discoloration — resolves within 24 hours. Recovery is generally far quicker than surgical alternatives, with most patients returning to normal daily activity within days.
Does GAE replace the need for knee replacement?
Not necessarily, and this study does not claim so. GAE is positioned as a treatment for patients who have failed conservative management but who are not yet at the stage of — or are not good candidates for — total knee replacement. It may delay the need for surgery significantly, and for some patients may eliminate the need entirely. However, patients with severe structural joint damage (KL Grade IV), severe axial malalignment, or other complicating factors may still ultimately require surgical intervention. This is a conversation to have with both your orthopedic surgeon and interventional radiologist.
If the microspheres dissolve in hours, why does the pain relief last 12 months?
This is one of the most scientifically interesting questions raised by the study. The prevailing hypothesis is that even temporary disruption of the pathological blood supply to the synovium is sufficient to interrupt the neurovascular cycle driving the pain. When the aberrant vessels are occluded — even briefly — the inflammatory and nociceptive nerve network that has grown alongside them is disrupted. The synovial tissue may partially "reset," reducing the chronic inflammatory state. The exact biological mechanisms are still being investigated, and this will likely be a major focus of future research.
Can GAE be performed on both knees?
Yes. In this study, 23% of participants (45 of 194) underwent bilateral treatment, with the second procedure performed within 4 weeks of the first. For bilateral cases, the follow-up period began after the completion of the second treatment.
Is GAE covered by insurance?
Coverage varies significantly by country, insurer, and whether GAE is available in your healthcare system. As a newer procedure still accumulating evidence from clinical trials, it may be considered investigational or experimental by some payers. Patients should contact their insurance provider and confirm the availability of the procedure at institutions in their region. This is a rapidly evolving area as evidence builds.
What is the difference between GAE and genicular nerve blocks?
Genicular nerve blocks (and radiofrequency ablation of the genicular nerves) target the pain signals at the nerve level — they don't treat the underlying abnormal vascularity. GAE, by contrast, addresses the root of the problem at the vascular level, targeting the hypervascular tissue that drives both the inflammation and the nerve growth responsible for the pain. They are mechanistically distinct, though both are minimally invasive. Some patients may have tried nerve blocks before considering GAE.
The Bottom Line
The 2026 Fleckenstein study is the most comprehensive prospective clinical data published to date on GAE using rapidly resorbable gelatin-based microspheres. With 194 patients, a 100% technical success rate, a median pain score halved at 12 months, no moderate or severe adverse events, and 55–80% of participants achieving clinically meaningful improvements across all outcome domains — RRGMs have made a compelling case as the embolic agent of choice for GAE.
They are calibrated for precision. They dissolve to minimize long-term risk. They are purpose-built for this exact application. And they work — not just statistically, but in ways patients actually feel in their daily lives.
The field still needs randomized controlled trials, longer-term follow-up data, and objective functional measures. But for patients who have run out of conservative options and aren't ready for a total knee replacement, GAE with RRGMs is a development worth taking seriously — and discussing with an expert.
📚 Source & Key References
- Fleckenstein FN, David D, Garducci P, et al. Genicular Artery Embolization Using Rapidly Resorbable Gelatin-based Microspheres for Osteoarthritis-related Knee Pain. Radiology 2026;319(3):e253312.
- Chlorogiannis DD, et al. Knee pain improvement after genicular artery embolization for the management of knee osteoarthritis: updated systematic review and meta-analysis of 21 studies. Radiologie (Heidelb) 2024;64(Suppl 1):32–46.
- Little MW, et al. Genicular Artery embolisation in Patients with Osteoarthritis of the Knee (GENESIS) Using Permanent Microspheres: Long-Term Results. Cardiovasc Intervent Radiol 2024;47(10):1432.
- Bagla S, et al. Genicular Artery Embolization for the Treatment of Knee Pain Secondary to Osteoarthritis. J Vasc Interv Radiol 2020;31(7):1096–1102.
- Okuno Y, et al. Midterm Clinical Outcomes and MR Imaging Changes after Transcatheter Arterial Embolization for Knee Osteoarthritis. J Vasc Interv Radiol 2017;28(7):995–1002.
- Han K, et al. Resorbable Microspheres versus Trisacryl Gelatin Microspheres for Uterine Artery Embolization: A Randomized Controlled Trial. Radiology 2024;312(3):e231525.
- GBD 2021 Osteoarthritis Collaborators. Global, regional, and national burden of osteoarthritis, 1990–2020 and projections to 2050. Lancet Rheumatol 2023;5(9):e508–e522.
.png)