RPS 1509 - The many faces of interventional radiology: a journey through innovation and impact

Author Block: B. Miles, A. Sharma, W. Oza, T. Elwell, J. Moss; Dallas/US
Purpose: Despite multiple FDA safety communications (2010, 2014) and multidisciplinary guideline statements (CHEST 2016; SIR 2020) recommending timely retrieval of inferior vena cava (IVC) filters, retrieval rates remain low. This represents not a knowledge gap but an implementation failure. We leveraged a global, multi-institutional dataset to quantify morbidity and mortality associated with indwelling filters compared to retrieval, entirely in the post-guideline era.
Methods or Background: We queried the TriNetX Global Collaborative Network, encompassing more than 150 health systems and 150 million patient records to identify adults undergoing IVC filter insertion (CPT 37191) between 2018–2025. Patients were stratified into (1) Filter Remains (no retrieval) and (2) Filter Removed (retrieval CPT 37193 within one day to six months). Propensity score matching balanced demographic covariates. Outcomes included severe sepsis, haemorrhage, cerebral infarction, and all-cause mortality. Risk ratios (RR), risk differences, and numbers needed to treat (NNT) were calculated.
Results or Findings: A total of 58,549 patients across 155 health systems were analyzed (46,173 Filter Remains; 12,376 FilterRemoved). Despite clear guidelines, only 21% underwent retrieval. Retained filters were associated with significantly worse outcomes: Severe sepsis:
10.8% vs 2.5% (RR 4.3; NNT=12); Haemorrhage: 2.6% vs 1.1% (RR 2.3; NNT=69); Cerebral infarction: 10.0% vs 4.9% (RR 2.0; NNT=20); Mortality: 30.2% vs 4.2% (RR 7.2; NNT=4).
Conclusion: In a cohort spanning 2018–2025, only one in five filters was retrieved and retained filters were associated with sharply higher rates of infection, stroke, haemorrhage, and death. Absolute differences translate to one death prevented for every 4 filters retrieved. These findings demonstrate a persistent guidelines-to-practice gap and highlight the need for structured retrieval pathways, automated alerts, and institutional accountability to align practice with safety standards.
Limitations: Reasons for leaving IVC filters in place were not available.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: B. Miles, S. Patel, T. Elwell, W. Oza, J. Moss; Dallas/US
Purpose: Management of intermediate-risk pulmonary embolism (PE), characterized by right ventricular (RV) strain without hemodynamic compromise, is complicated by a lack of comparative outcome data and variability in clinical guidelines. This study was conducted as a real-world comparative analysis using TriNetX, a federated EHR network with over 100 million patients across U.S. health systems. We evaluated the mortality and bleeding outcomes of CDT versus anticoagulation alone in intermediate-risk PE at one year.
Methods or Background: This retrospective cohort study used data from the TriNetX Global Collaborative Network. Adult patients with acute PE (ICD-10-CM I26.9) and evidence of RV strain (troponin ≥ 0.04 ng/mL or BNP ≥ 125 pg/mL) were included. Two cohorts were defined: (1) the CDT cohort, identified via CPT codes for mechanical thrombectomy (37184–37186), thrombolytic infusion (37211–37214), and/or alteplase administration (J2997); and (2) the anticoagulation-only group, defined by heparin or DOAC use within 24 hours of admission without CDT procedure codes. Following propensity score matching, 5,523 patients were included in each group. Primary outcomes included mortality and major bleeding (defined using ICD-10 codes I97.4, K92.2, and D62) within one year of treatment.
Results or Findings: One-year mortality was significantly lower in the CDT group (8.88% versus 10.57%), with an adjusted risk ratio (aRR) of 0.84 (95% CI: 0.75–0.90, p < 0.0001). Major bleeding occurred in 8.88% of CDT patients versus 11.25% in the anticoagulation cohort, with an aRR of 0.84 (95% CI: 0.76–0.94, p < 0.0001), indicating that CDT was not associated with increased bleeding risk.
Conclusion: CDT is associated with a mortality advantage without increased bleeding. CDT may be an effective and safer option than systemic anticoagulation for treatment of these patients.
Limitations: This study was retrospective. Prospective studies will be needed to confirm.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: M. M. Taepper¹, A. Dmytriw², N. Ntoulias¹, A. Brehm¹, M-N. Psychogios¹, V. Schulze-Zachau¹; ¹Basel/CH, ²Oxford/UK
Purpose: Vessel perforation is a rare but severe complication of mechanical thrombectomy for acute ischemic stroke. We aimed to identify the procedural steps at which perforations most frequently occur and to compare bleeding duration, functional outcome, and survival across these steps.
Methods or Background: We conducted a retrospective cohort study of patients with vessel perforation during thrombectomy between January 2015 and September 2022 across 32 centers. Outcomes included bleeding duration, defined as angiographic extravasation; 90-day functional outcome, assessed by the modified Rankin Scale (mRS); and overall survival, evaluated by Kaplan–Meier analysis. Outcomes were stratified by the procedural step of perforation.
Results or Findings: Among 447 patients (mean age 73 years, 59% female), perforations occurred during retraction of a stent retriever or aspiration catheter in 45% (n=201), probing beyond the occlusion in 26% (n=117), access to the occlusion in 13% (n=58), device deployment in 9% (n=42), percutaneous transluminal angioplasty (PTA) or stent placement in 5% (n=24), and contrast injection or thrombolysis in 1% (n=5). Perforations during PTA/stent placement were associated with longer bleeding duration (1.82 [1.14 – 2.90 95% CI], p=0.01) and worse 90-day mRS (1.10 [0.33-1.80 95%CI], p=0.01) compared with all other steps. Survival also differed significantly by perforation step (p=0.01), with the lowest survival following PTA/stent-related perforations.
Conclusion: Vessel perforations most often occurred during device retraction and probing beyond the occlusion. Although less frequent, perforations during PTA or stent placement carried the poorest prognosis, likely due to prolonged bleeding.
Limitations: Retrospective design and incomplete anticoagulation data.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethical commission approval and patient consent were not required according to current local legislation as all data were anonymized before analysis and the project involved assessing safety and quality of routine patient management in the participating institutions.

Author Block: B. S. Tan, A. J. W. Tan, S. Arjunana, W. Lee, Z. Tan, A. Patel, K. H. Tay, T. H. Heng, K. D. Zhuang; Singapore/SG
Purpose: To compare the efficacy of drug-coated balloon angioplasty (DCBA) (Ranger, Boston Scientific; Marlborough, Massachusetts, USA) to conventional balloon angioplasty (CBA) of all stenoses in the peripheral circuit of dysfunctional dialysis arteriovenous fistula (AVF). We present the preliminary analysis of a randomized controlled trial (RCT) comparing DCBA against CBA of our first 73 patients.
Methods or Background: This prospective single-center RCT recruited participants with stenoses in the peripheral circuit of failing mature AVF. Participants with central vein stenoses or AVF thrombosis were excluded. After successful angioplasty of all AVF stenoses in the peripheral circuit (lesion preparation achieving < 30% residual stenosis), 94 participants (72 male, 76.6%; mean age 66.8 ± 9.5 years) were randomized to DCBA or CBA in a 1:1 ratio, stratified by the number of AVF stenoses (single versus multiple). The primary outcome measure was clinically driven 6-month access circuit primary patency (ACPP). Participant follow-up for patency was up to 12 months post recruitment.
Results or Findings: The outcomes of 73 study participants (DCBA: 35, CBA: 38) (59 males, mean age: 66.2 years old) who had completed a minimum of six months follow-up were analyzed. The types of AVFs were radiocephalic (64.4 %), brachiocephalic (24.7%) and brachiobasilic (11%). In 43 patients, a single lesion was treated, while two or more lesions were treated in 30 patients. ACPP at six months was not significantly different between both groups (DCBA: 79.4%, CBA: 77.8%, p= 0.82).
Conclusion: In this initial report of our first 73 patients, the treatment of all peripheral AVF stenoses with DCBA did not yield a statistically significant difference in ACPP at six months when compared to CBA. Further analysis of the full cohort with longer follow-up may yield more definitive results.
Limitations: Preliminary results
Funding for this study: Boston Scientific Foundation.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approved

Author Block: A-I. Nica¹, T. J. Vogl¹, F. Wacker², R. Klöckner³, C. Booz¹, L. S. Alizadeh¹; ¹Frankfurt/DE, ²Hannover/DE, ³Lübeck/DE
Purpose: To evaluate the safety profile of Selective Internal Radiation Therapy (SIRT) in the treatment of primary and secondary liver malignancies, with a specific focus on the incidence and management of periprocedural complications.
Methods or Background: We conducted a retrospective analysis of data from the national quality assurance registry maintained by the German Society of Minimally Invasive Therapies and Interventional Radiology (DeGIR), covering the period from 2018 to 2023. The study aimed to identify adult patients (aged >18 years) who underwent SIRT for liver tumors. We evaluated the incidence, types, and management of complications, classifying them as either 'procedure-related complications' (PRCs), occurring during or within 24 hours of the intervention, or 'post-procedure complications' (PPCs), defined as those arising more than 24 hours after the procedure.
Results or Findings: The study included 4432 patients (1563 females, 35.3%; 2869 males, 64.7%) with a mean age of 67.4 ± 11.4 years. Most SIRT-related complications were PRCs (36 cases, 0.8%), mainly due to drug side effects (5, 0.1%) and arterial bleeding (3, 0.1%). Nearly half (45.7%) were Category 2/B, requiring prolonged monitoring. One death (Category 6/F, 2.9%) occurred within 24 hours post-intervention. In 50% of cases (18), complications were managed by the interventional radiologist, 27.8% (10) by other specialties, 16.7% (6) required no treatment, and 5.6% (2) were unspecified. PPCs occurred in 4 patients (0.09%), for a total of 40 patients (0.9%) with periprocedural complications.
Conclusion: Our findings suggest that SIRT is a safe procedure for patients with liver tumors, as it demonstrates a low incidence of complications, with many being effectively managed within the same medical specialty.
Limitations: The main limitation of this study is its retrospective design.
Funding for this study: This study received no funding.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study received approval from the Ethics Committee.

Author Block: M. Adinolfi Falcone, L. Rinvenuto, F. Arrigoni, L. Zugaro, A. Barile, E. Di Cesare; L'Aquila/IT
Purpose: Osteoid Osteoma (OO) is a benign but extremely painful bone lesion. Magnetic Resonance-guided Focused Ultrasound Surgery (MRgFUS) is an ablative technique in Interventional Radiology that combines high-intensity focused ultrasound (HIFU) with MR imaging. The aim of this study was to evaluate the long-term outcomes and effectiveness of MRgFUS in the treatment of OO compared with other therapeutic options.
Methods or Background: A retrospective analysis was performed on all MRgFUS procedures for OO conducted at our center over a 13-year period. 67 patients treated were included, all meeting the same inclusion criteria: radiological findings of OO, symptomatic lesions whit VAS > 7, and an adequate acoustic window for ultrasound beam penetration. Demographic data, morphological features, imaging findings and clinical outcomes wcre collected. All variables were statistically analyzed, and results were evaluated in terms of pain relief and nidus characteristics. Standard follow-up included clinical ooutcomes assessment at 7-10 days, 6 months, 1 year, 2 years, and 4 years; MRI at 6 and 12 months; and CT scan at 12 months.
Results or Findings: The procedure was successful in 61 out of 67 patients, with an overall success rate of 91%. All responders showed MRI resolution of bone edema and reduction of perilesional reactive changes. No complications were recorded. Treatment failure was significantly correlated with thick cortical bone overlying the nidus (p=0.014) and with a short nidus-to-ultrasound source distance (p= 0.002).
Conclusion: MRgFUS is a safe, minimally invasive, and effective therapy for OO, associated with absence of complications, rapid recovery, and a low recurrence rate. This study represent the largest reported single-center series with the longest follow-up (13 years), confirming the excellent safety profile and long-term efficacy of MRgFUS in the treatment of Osteoid Osteoma.
Limitations: No limitations were identified.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: J. LI, M. Lu; Chengdu/CN
Purpose: To investigate the effect of pectoral nerve block on pain management in microwave ablation (MWA) for multiple benign breast nodules.
Methods or Background: A total of 94 patients undergoing MWA for benign breast nodules were divided into three groups based on the analgesic method used: local anesthesia group (Group A, n = 31), pectoral nerve block group (Group B, n = 30), and local anesthesia combined with pectoral nerve block group (Group C, n = 33). Pain levels were recorded using the Numerical Rating Scale (NRS) during the procedure and at 2, 6, 12, 24, and 48 hours postoperatively. Postoperative analgesic requirements and complications were also documented.
Results or Findings: The consumption of lidocaine and ropivacaine during the procedure differed significantly among the groups (p < 0.05), with Group B showing the highest usage and Group C the lowest. Group C exhibited lower intraoperative and postoperative NRS scores compared to Groups A and B (P < 0.05), while no significant difference was observed between Groups A and B (P > 0.05). At 2 and 6 hours postoperatively, Group A had lower NRS scores than Group B (P < 0.05), but no significant differences were noted at 12, 24, and 48 hours (P > 0.05). The postoperative analgesic requirement was significantly lower in Groups B and C compared to Group A (P < 0.05). No serious complications occurred after the procedure.
Conclusion: The combination of local anesthesia and pectoral nerve block effectively reduces pain during and after MWA for multiple benign breast nodules and decreases the need for postoperative analgesia.
Limitations: No
Funding for this study: This study is funding by Sichuan Provincial Science and Technology Support Program, 2024YFHZ0140
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Institutional Review Board and Ethics Committee of Sichuan Cancer Hospital. All patients provided signed informed consent before the examination. All methods were performed in accordance with the relevant guidelines and regulations.

Author Block: F. Lazzarotto, G. Falcone, A. L. Annese, D. A. Campanacci, F. Fanelli; Florence/IT
Purpose: BEST (Bleomycin Electrosclerotherapy) has been proposed as a novel approach for treating venous and linfatic malformations (VMs). This study reports a single-center initial experience in treating extracranial symptomatic slow-flow malformations.
Methods or Background: A retrospective observational study was conducted from March 2024 to Sep 2025, including 24 patients (28 lesions) with symptomatic VMs. Most had undergone previous unsuccessful treatments.  Electrosclerotherapy was performed using reversible electroporation combined with bleomycin. Clinical records, magnetic resonance imaging (MRI) data, previous treatment history, intervention details, complications, and clinical outcomes were analyzed. A postoperative questionnaire assessed both functional and aesthetic effects.
Results or Findings: Of the 24 patients (mean age 23,8 ± 8,4 years; 8 females), 7 had undergone an average of 2.5 previous invasive treatments. A total of 34 electrosclerotherapy sessions were performed on 28 lesions. The median dose of bleomycin administered was 2,6 mg. Clinical and MRI follow-up schedules were established. The median MRI-derived lesion volume decreased in 73% of the patients . After a median follow-up of 6 months (range 3-9 mos), 14 patients reported complete symptom resolution, while 8 showed significant improvement, 2 no improvement. 6 minor complications occurred, including transient skin discoloration and swelling.
Conclusion: BEST is a safe and effective therapeutic option for LM and VMs, including those resistant to prior treatments.
Limitations: Despite promising results, the small sample size and limited FUP warrant further investigation to confirm these findings.
Funding for this study: No funding received for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: K. D. Martins De Mattos¹, R. Muglia², F. S. Carbone², M. Bertuletti², L. Dulcetta², P. Marra², S. Sironi¹; ¹Milan/IT, ²Bergamo/IT
Purpose: Microwave thermal ablation (MWTA) is an established therapy for HCC in cirrhotic patients, with a low rate of local recurrence. However, in some cases, despite optimal antenna placement and strict adherence to the ablation volume charts, residual disease may persist after treatment. We aimed to evaluate the immediate ablation volume after MWTA in cirrhotic livers, and compare it with the volumes estimated by the manufacturer chart.
Methods or Background: This single-center retrospective study was conducted at a tertiary referral center, including cirrhotic patients BCLC 0/A, treated for HCC by MWTA (Emprint™HP, Medtronic, USA) between November 1, 2022, and August 31, 2025.
HCCs were categorized by size (small <25mm; large 25All MWTAs were performed under CT-hepatic arteriography guidance; we calculated immediate post-ablation necrotic volumes, and compared them to expected ablation volumes, following the manufacturer chart. Volumes were calculated via multiplanar reconstruction (MPR) using the ellipsoid formula (D1×D2×D3×π/6).
Results or Findings: We analyzed 43 HCCs in 29 patients, treated at 100W/150W (36:7). Twelve nodules were perivascular (11 small - 1 large), 31 subcapsular (22 small - 9 large). No nodules were far from liver capsule or veins.
The mean achieved volume for perivascular nodules was 12.37cm³, 42.4% lower than the expected 21.48cm³ (p<0.001). For subcapsular nodules the mean achieved volume was 15.25cm³, 36.3% lower than the expected 23.96cm³ (p<0.001). This significant underestimation corresponded to a Bland-Altman systemic bias of -9.04 cm³. Only 2/43 outlying cases (subcapsular nodules) were identified.
Conclusion: Our findings suggest a systematic underestimation of achieved ablation volumes compared to expected volumes in cirrhotic livers. Increasing ablation volumes compared to manufacturer chart might improve MWTA efficacy.
Limitations: Retrospective-design
Funding for this study: No funding was provided for this study
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: A. Taheri Amin¹, L. M. Wilms², K. Jannusch², F. Ziayee², P. Minko²; ¹Berlin/DE, ²Düsseldorf/DE
Purpose: Assess the safety, procedural aspects, and clinical outcomes of genicular artery embolization (GAE) with resorbable gelatin microspheres (RGM) in patients suffering from knee osteoarthritis (OA) or ongoing pain following total knee replacement (TKR)
Methods or Background: 45 patients were prospectively observed, 35 with knee OA (Kellgren–Lawrence grades 1–4) and 10 with post–TKR pain. GAE was performed using100–300 µm RGM, with embolic volume and treated vessels recorded. Clinical outcomes were measured using KOOS and NRS at baseline, 6 weeks, 3 and 6 months. The HADS questionnaire was used to screen for psychological comorbidities. Adverse events were monitored.
Results or Findings: At 6 months, the mean NRS decreased by 35% and the KOOS pain subscale improved by 55% (p < 0.001), with no significant differences observed between OA grades and post-TKR . A median of 3 vessels (range 2–6) were embolized, with an average embolic volume of 6.5 mL (range 2.1–18.0 mL) per procedure. Patients with HADS scores > 8 demonstrated smaller improvements in both KOOS and NRS, with up to 2.2 points less improvement per unit increase in HADS (p < 0.05). Total embolic volume was significantly higher in advanced OA and post-TKR cases, rising by an average of 2.7 mL with each OA grade (p < 0.0001). No major complications occurred.
Conclusion: GAE with RGM at doses exceeding 2 mL is safe and provides short-term clinical benefit in all OA grades as well as in post-TKR pain. Patients with advanced OA or post-TKR tend to require higher embolic volumes. Elevated HADS scores are linked to reduced clinical response.
Limitations: This study is limited by its small sample size, short-term follow-up, lack of a control group, and reliance on patient-reported outcomes without systematic imaging.
Funding for this study: No funding was recieved
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: All procedures performed were in accordance with the ethical
standards of the institutional research committee Board (Study number: 2021-1777_5) and with the principles of the 1964 Declaration of Helsinki and its later amendments.

Author Block: A. Taheri Amin¹, E. Kemmer², A-J. Hübner², K. Jannusch², D. Weiß², L. M. Wilms², F. Ziayee², P. Minko²; ¹Berlin/DE, ²Duesseldorf/DE
Purpose: To evaluate the impact of osteoarthritis (OA), synovitis and evoked pain on embolic volume in genicular artery embolization (GAE).
Methods or Background: Radiographs, MRI and DSA images of patients undergoing GAE were retrospectively reviewed. OA was graded (Kellgren-Lawrence) and categorized by predominant location (medial/lateral). Synovitis was graded on unenhanced MRIs using the MRI Osteoarthritis Knee Score. Evoked pain during superselective angiography and embolic volume were systematically recorded. Embolic volume was correlated with OA grade, predominant OA location, synovitis grade and evoked pain. Clinical outcome was assessed using the numeric rating scale at six weeks and months.
Results or Findings: A total of 1,739 vessels were embolized in 450 patients (median: 4 ± 1 vessels per session). Pre-interventional radiographs were available in all patients, with additional MRI scans in 214 patients. Embolic volume correlated moderately with OA-grade (Spearman’s ρ=0,51; p<0.001) and weakly with synovitis grade (ρ=0,20; p<0.01). In patients with predominantly medial OA, embolic volume applied in the medial compartment was significantly higher than in the lateral compartment (2.7±1.5 vs 1.5±1.2 mL; p<0.001), and vice versa for patients with predominantly lateral OA (2.2±1.1 vs 1.6±1.0 mL; p<0.001). Embolic volume was significantly higher in vessels with evoked pain than without (1.4±0.9 vs 0.8±0.3 mL; p<0.001). Pain scores improved significantly at all timepoints (p<0.001).
Conclusion: OA grade, predominant OA location, synovitis grade, and evoked pain influence embolic volume in GAE, illustrating that—similar to the multifactorial genesis of OA—embolization endpoints are likewise shaped by multiple variables.
Limitations: ChatGPT:

The study is limited by its retrospective design, reliance on NRS without long-term follow-up, and MRI data available in only half of patients.
Funding for this study: This work was supported by the local Research Committee of the Medical Faculty of
Heinrich-Heine-University Düsseldorf (grant number: 9772841).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee (Study number: 2021-1777_5) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards