Genicular Artery Embolization for Primary Knee Osteoarthritis

Abstract

Genicular artery embolization (GAE) is a novel treatment for knee osteoarthritis (OA) with increasing interest and utilization worldwide. The genicular arteries are selectively embolized using microspheres with subsequent cessation of abnormal synovial hyperemia, a process believed to arrest the inflammatory changes and pain associated with OA. Current data indicate that GAE offers sustained, 6-month improvement in Western Ontario and McMaster Universities Osteoarthritis index (WOMAC) and visual analog scale scores in 80 to 85% of patients with mild to moderate OA. As GAE gains traction, future studies are needed to establish its safety and efficacy relative to placebo and other standard-of-care therapies while also characterizing populations for which it is most effective.

Knee osteoarthritis (OA) is a condition that occurs over time due to the progressive loss of cartilage in the joint. The condition is seen most commonly in those older than 50 years and can lead to debilitating joint pain and/or disability. Current treatments for mild to moderate knee OA are centered around strengthening exercises and pain medication. Severe cases are often associated with disability and are usually treated surgically with total knee arthroplasty. Within this spectrum, however, a large subgroup of patients with intermediate-stage knee OA are difficult to manage with limited options as they present refractory to medications but do not exhibit enough joint disease to warrant full joint replacement. Recent studies have shown promise in treating this cohort and addressing this treatment gap with genicular arterial embolization. ¹

The pathophysiology of OA has traditionally been thought to be a degenerative process associated with chronic “wear-and-tear”; however, new evidence suggests that low-grade inflammation is a primary contributor. This inflammatory cascade results in synovial neoangiogenesis with the release of vascular endothelial growth factor (VEGF) and Ang-1 thought to be noteworthy contributors. ² These proangiogenic factors result in vascularization accompanied by innervation of the normally aneural joint space, a process that results in pain and disability of the knee. ^{2 3} It is postulated that arrest of this neovascularity through selective embolization of the abnormal genicular arteries can arrest this process and improve symptoms attributed to OA. The aim of this review is to discuss the relevant anatomy and technique for GAE and also assess the current body of evidence supporting its use as a viable treatment for patients suffering from moderate-to-severe chronic knee OA.

Anatomy

The vessels that support the knee joint consist of five major arteries, specifically the descending genicular, superior medial/lateral genicular, middle genicular, and inferior medial/lateral genicular arteries. ⁴ Together these arteries anastomose around the knee joint with each supplying a major portion of the knee. The popliteal artery is a continuation of the superficial femoral artery and gives rise to the superior medial and superior lateral genicular arteries, both of which supply the anterior portions of the femoral condyle, as well as its adjacent synovial tissue. The lateral and medial condyles are supplied by middle genicular arteries. The lateral and medial inferior genicular arteries also arise from the popliteal artery, specifically below the gastrocnemius muscle. ⁴ These two arteries supply the knee joint and bones of the lower leg.

Technique

GAE is a minimally invasive procedure performed under moderate sedation and pain medication. Prior to beginning, metallic markers ( Fig. 1 ) can be placed at the site of the patient's pain for visualization during fluoroscopy. The procedure can be performed from a radial or femoral approach. The distal superficial femoral artery of the affected limb is catheterized and digital subtraction angiography is then utilized to identify the target genicular arteries.

Fig. 1.

Genicular artery embolization for a 72-year-old with severe chronic left knee pain. Superior femoral arteriogram ( a ) demonstrates dominant superomedial and superolateral genicular artery supply to medial and lateral femorotibial joint space with associated hyperemia (red circles ). Pre- and postembolization of the superomedial ( b, c ), superolateral ( d, e ), and inferomedial ( f, g ) genicular artery was performed until hyperemic “blush” ( b, d, f ) is no longer seen.

Abnormal arteries are described to have a “tumor blush”–like pattern ( Fig. 1 ) when contrast is injected. The metallic marker can be used to guide which arteries are to be embolized. The target genicular arteries are subselected with a microcatheter and confirmed with angiography and/or cone beam CT. Embolization is then performed with the endpoint typically characterized as suppression in the filling of abnormal arteries (i.e., blush). Once this is achieved, the catheter is removed and hemostasis is achieved. In some cases, during the initial embolization process, pain similar to that of the patient's OA can be reproduced, confirming that the appropriate arteries have been selected. Pain is often reassessed 1 to 2 hours after completion of the procedure once medication effects have subsided.

The embolic agent of choice varies in literature with operator preference. Imipenem/cilastatin was initially selected due to its crystalline structure and embolic effects when suspended in a contrast medium. ^{5 6 7} However, this material is not yet FDA approved for this application and may induce a reaction in those with allergies to the antibiotics in the compound. More commonly, 75- to 100-μm particles have been utilized, although studies have not found any difference in efficacy compared with imipenem/cilastatin spheres. ^{6 7} Of note, patients who had embolization performed with Embozene particles reported greater mean decrease in VAS and WOMAC scores compared with patients who had embolization with imipenem/cilastatin at 1 month. ⁸ However, at 6 months, both VAS and WOMAC scores were comparable. ⁸

Clinical Outcomes

Early Studies

GAE was initially performed for the management of spontaneous recurrent hemarthrosis after total knee arthroplasty. ^{9 10} More recent publications have started to look at the procedure as an option for patients with mild to moderate OA. One of the first prospective studies evaluating this technique in the treatment of moderate to severe knee OA was performed in Japan by Okuno et al in 2015, detailing 11 patients who underwent GAE utilizing imipenem/cilastatin and 3 utilizing 75-μm particles (Embozene; Varian Medical). ^{4 5} Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain scores were used and improvement was sustained in most cases to final follow-up examination which on average was 12 ± 5 months. ⁴

In 2017, Okuno et al continued their work by conducting another prospective study, this time evaluating 72 patients (95 knees) who were treated with transcatheter arterial embolization from 2012 to 2016. ⁶ This study also included the original cohort seen in the 2015 trial. Outcomes were set to be evaluated at 1, 4, and 6 months with additional outcomes being evaluated every 6 months after until a maximum of 4 years. ⁶ Mean WOMAC scores significantly decreased from baseline until 24 months after treatment. ⁶ In addition, the clinical success rates at 6 months and 3 years after embolization were 86.3 and 79.8%, respectively. ⁶ MR imaging was also performed on 29 patients (35 knees) at baseline and the 2-year mark to assess treatment response. The Whole-Organ Magnetic Resonance Imaging Score (WORMS) at baseline and 2 years found no significant difference in regard to pathological changes of the knee joint. The WORMS score, however, did find statistically significant improvement in synovitis when compared with baseline. ⁶

Genicular Artery Embolization for Moderate versus Severe Osteoarthritis

In 2019, Lee et al performed a retrospective study in Korea which assessed clinical outcomes of mild to moderate knee OA versus severe OA. ⁷ GAE was performed in a total of 41 patients (71 knees), 30 of which were bilateral cases. Fifty-nine cases were considered to be mild-moderate, while only 12 patients were considered severe OA cases. All patients in this study were embolized with imipenem/cilastatin. This study utilized the visual analog scale (VAS) to assess for improvements in knee pain. In the mild-moderate OA subgroup, VAS scores were significantly decreased at the endpoints of 1 day, 1 week, 1 month, 3 months, and 6 months. ⁷ Improvements in VAS in this subgroup were maintained until a mean of 10 ± 3 months. ⁷ The severe OA subgroup had statistically significant improvements in VAS at the endpoints of 1 day, 1 week, and 1 month only. ⁷ Additionally, the severe OA subgroup VAS scores returned to initial severity within a 3-month period. ⁷

In the United States, Bagla et al published a prospective pilot study that tracked 20 patients between 2017 and 2018 at two different U.S. sites. ⁴ This study measured VAS and WOMAC at baseline as well as after 1-, 3-, and 6-month endpoints. All 20 patients were embolized with 75- to 100-μm Embozene particles. Ninety-five percent of the cohort saw improvements in VAS after 1 month, while all patients had WOMAC improvements during this time period. These improvements were sustained until the 6-month mark with 85 and 82% improvement in VAS and WOMAC, respectively. ⁴

Genicular Artery Embolization for Early-Stage Osteoarthritis

In 2020, Landers et al undertook a 24-month pilot study evaluating GAE in early-stage knee OA. ¹¹ Ten patients were selected under the parameters of Kellgen-Lawrence grade 1–2 knee OA on X-ray and at least 6 months of moderate knee pain refractory to conservative treatment. ¹¹ This study differed from both Lee et al and Okuno et al as polyvinyl alcohol (PVA) was used as an embolic in addition to imipenem/cilastatin. PVA agent typically consists of a 1-mL solution containing 90 to 180 µm embolic material with particle sizes on average of 10 to 70 µm. ¹⁰ Out of the 10 total patients, 70% saw improvement after 1 and 6 months, 60% after 12 months, and 30% after 24 months posttreatment. ¹¹

In 2021, Little et al published a prospective pilot study evaluating the role of GAE in patients with mild to moderate knee OA. ¹² Thirty-two patients resistant to conservative treatments underwent GAE using 100 to 300 μm permanent microspheres. Knee MRI was performed at baseline and 12 months. ¹² A semiquantitative analysis was performed utilizing WORMS, which showed a significant improvement in synovitis in all patients at 1 year. However, WORMS also demonstrated significant worsening of osteophytes and bone attrition. ¹² Patient symptoms were evaluated using Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire and a VAS at baseline, 6 weeks, 3 months, and 1 year. KOOS scale indices showed statistically significant improvements in all measures except activities of daily living. ¹² Mean VAS at baseline was 60, 32 at 6 weeks, 36 at 3 months, and 45 at 12 months. ¹²

Genicular Artery Embolization—A Randomized Control Trial

In 2020, Bagla et al published one of the first randomized, sham-controlled studies of GAE. ¹³ Twenty-one patients, 14 of who received a GAE procedure and 7 of who had a sham procedure performed, were blinded and randomized. In this study, patients were allowed to crossover to the treatment arm if no improvement was seen after 1 month. VAS and WOMAC scores after 1 month showed statistically significant improvements compared with the sham procedure. All seven sham patients ultimately crossed over to the treatment arm after 1 month. ¹³

Genicular Artery Embolization Research Consensus Panel

In 2021, a research consensus panel (RCP) consisting of a multidisciplinary group of experts convened to summarize the current existing knowledge of the management of knee OA, while also prioritizing the specific areas of further research. ¹⁴ One of the many important questions that came from this RCP was what specific types of studies are currently needed to further validate GAE as an accepted treatment for knee OA. ¹⁴ Consensus on large sham-controlled randomized trials as the gold standard was met, largely due in part to the history of previous OA studies showing very high placebo effects, some as high as 40%. ¹⁴ Issues with institutional review board approval as well as the practical and ethical ramifications of creating a sham trial were acknowledged. ¹⁴ Trials comparing GAE to current standard of care were determined to be a simpler option; however, deriving true experimental benefit would be more difficult to appreciate. ¹⁴ Optimal patient cohorts for these trials were also discussed with the RCP concluding a presurgical cohort, both ineffectively managed by current standard of care and ineligible for surgery, had the greatest potential to demonstrate the benefits of percutaneous therapies. ¹⁴

Complications

No severe complications as a result of the GAE procedure have been reported in the current literature to date. The earliest studies in Japan done by Okuno et al had only minor adverse events such as resolving subcutaneous hemorrhages at puncture sites as well as transient color changes of the skin in patients treated with Embozene. ^{5 6} Skin discoloration with particle size less than 100 μm has been previously described ⁴ ; this discoloration is thought to be due to distal embolization into small, cutaneous arterioles. ⁴ Padia et al also reported self-resolving focal skin ulcerations in several subjects who received permanent particles such as Embozene. ¹⁵ Studies from Australia and Korea had similar results, with no major adverse events and only transient subcutaneous hematomas at the injection sites in addition to skin redness and self-resolving fevers in rare cases. ^{7 10} MR imaging in some of the studies, performed as far back as 2 years from initial procedure, also showed no evidence of abnormality in the bone, menisci, marrow, and ligaments. ^{6 10}

Conclusions

GAE is a minimally invasive procedure that targets abnormal genicular arteries to embolize and disrupt the abnormal neovascularity of the synovium and knee joint thought to cause pain and dysfunction with knee OA. The current body of published literature on GAE is increasing steadily with reproducible findings supporting its safety and efficacy. Although no current randomized control trial exists comparing GAE to current management techniques for knee OA, current studies have reproduced long-term improvements in pain and quality-of-life metrics. However, most studies to date are limited by their single-arm design and small sample sizes. Additionally, many of these studies have been performed in a handful of regions geographically, thereby limiting generalizability. A recently conducted multidisciplinary RCP determined that a large, well-designed sham-controlled study is needed to ascertain the true benefit of GAE given a high rate of placebo effect seen in OA treatments. Additional studies will also be needed to compare GAE to other standard-of-care treatments with long-term follow-up. Upon confirmation provided by such studies, GAE may offer yet another treatment modality for patients suffering from mild to moderate knee pain secondary to OA.