Acceptability of Genicular Artery Embolization for the Treatment of Painful Knee Osteoarthritis: A Single-Center Experience

To the Editor,

Knee osteoarthritis (KOA) affects more than 600 million people worldwide [1]. Inflammatory synovitis is strongly associated with painful KOA, and while targeting inflammation systemically with nonsteroidal anti-inflammatories (NSAIDs) is effective, pain relief from NSAIDs comes with considerable renal, cardiovascular, and gastrointestinal risks [2]. Studies suggest that genicular artery embolization (GAE), which targets inflammatory synovium, is effective in ameliorating KOA pain for up to 24 months [3]. Since normal synovium is metabolically active and crucial to maintaining healthy articular cartilage, selectively targeting only pathologic synovium with GAE is a logical approach. However, since GAE for KOA requires an intraprocedural angiogram to identify areas of hypervascular synovium, groups less likely to choose procedural interventions, (e.g., women and racially and ethnically diverse populations), may be less willing to accept GAE [4, 5]. Differential acceptability of GAE between groups could inadvertently lead to widening disparities in knee pain outcomes. Therefore, this study was undertaken to better understand which KOA patients would be willing to accept GAE.

Patients aged 50–79 years with KOA (ICD-10 codes: M17.0, M17.1, and M17.9) were identified at a single musculoskeletal hospital. To identify a homogenous group of representative KOA patients who may be willing to consider GAE, inclusions included ages 50–79, body mass index (BMI) ≤ 35 kg/m², imaging evidence of moderate KOA (Kellgren and Lawrence Grades 1–3 or KOA on magnetic resonance imaging), and KOA pain refractory to conventional therapy for at least 6 months. Exclusions included mild knee pain, (Knee Injury and Osteoarthritis Outcomes Score (KOOS) pain subscale ≤ 70), systemic rheumatic disease, crystalline arthritis, prior surgery to the index knee, osteonecrosis, contralateral TKA in the past 6 months, and inability to understand and/or read English and potential arthroplasty candidate (Kellgren and Lawrence Grade 4). Patients were invited to participate in a one-time virtual study visit during which they viewed a 20 min PowerPoint presentation administered by a trained assistant, explaining that GAE was a new procedure to treat knee pain in people with KOA and inflamed joint tissue. A hypothetical 6-month sham-controlled RCT was then described, in which 50% of participants with KOA would be randomized to an angiogram-only (sham) arm, and 50% to an angiogram with GAE arm. Finally, the same hypothetical sham-controlled RCT was described, with the additional feature that participants in the angiogram-only arm could opt to receive GAE at trial completion if their knee pain had not resolved. Participants then ranked acceptability of GAE in three scenarios, assuming that they had MRI evidence of inflamed synovial tissue in their knee: (1) routine clinical care, (2) sham-controlled RCT, and (3) sham-controlled RCT with guaranteed access to GAE for those randomized to the control group. Willingness to receive GAE was scored on a 5-point Likert scale. Demographics and patient-reported outcomes measures (PROMs) were recorded. This study was approved by our Institutional Review Board.

Descriptive statistical analysis was performed comparing patients’ characteristics between the three willingness scenarios described, using percentages for categorical variables and mean (SD) or median (IQR) for continuous variables. SAS 9.4 and R software were used to compare differences using Mann–Whitney tests or Chi-squared tests, as appropriate.

One hundred participants were enrolled between 08/2020 and 03/2024. Overall, 84% would be willing (42% definitely willing/42% probably willing) to have GAE as a routine clinical intervention and 53% would be willing (21% definitely willing/32% probably willing) to participate in a sham/controlled RCT. This percentage increased to 61% if GAE was offered to the sham/control group after completion of a 6-month RCT, suggesting guaranteed access to GAE may improve RCT recruitment. Those definitely/probably willing to have GAE in routine care had clinically meaningfully and statistically significantly worse KOOS Quality of Life subscale scores, (37.0 vs. 48.4; p-value = 0.03), (see Supplementary Table 1). There were no clinically or statistically significant differences in age, gender, race, ethnicity, BMI, or other PROMs between those willing or not willing to have GAE in routine clinical care or in either of the two RCT scenarios (see Supplementary Tables 2 and 3).

The most common themes among participants willing to have GAE in routine clinical care were to avoid knee arthroplasty (N = 27), pain relief (N = 21), and believing GAE would be a good option after exhausting available therapies (N = 18).

Among those not willing to have GAE, the most common theme was fear of adverse events and/or a preference for more conservative therapies (N = 9). All participants expressed a desire to know more about GAE, both from their own doctor and from additional studies on larger numbers of people with KOA (N = 16).

The high proportion of subjects with moderate knee pain willing to undergo GAE underscores the lack of satisfaction with existing therapies and the unmet therapeutic need for better treatments for painful KOA. The only PROM associated with willingness to accept GAE in clinical care was KOOS Quality of Life subscale, suggesting that it is how pain affects daily life that primarily drives treatment decision-making. Women were not less likely to accept GAE, which is important, as despite higher levels of KOA, women have lower rates of knee replacements; GAE may, therefore, help reduce gender disparities in KOA treatment. Whether GAE will prevent, rather than delay, a knee replacement is unknown. Long-term follow-up of patients undergoing GAE will be critical to answer this important patient-centered question. Overall, similar acceptability across different demographics and ranges of PROMs suggests that GAE could benefit a wide variety of patients with painful KOA.

Supplementary Material

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s00270-025-03978-z.