Abstract
Background
The introduction of flow diverters (FDs) has revolutionized the treatment of complex intracranial aneurysms; subsequent surface modifications have extended their indications to ruptured, bifurcation, and distal aneurysms. The aim of this study is to assess the real-world feasibility and safety of HPC surface-modified devices in the treatment of ruptured and unruptured aneurysms.
Methods
This independent, multicenter, prospective observational study evaluated the outcomes of patients treated with implanted Phenox p64 or p48 MW-HPC-FDs between 2020 and 2022. The sub-analysis of the procedures, complications (clinical and technical), and additional treatment requirements assessed the devices’ feasibility and safety in the peri-procedural period.
Results
One hundred and forty aneurysms (35% ruptured and 65% unruptured) were treated in 140 patients (37 females and 103 males) using 153 HPC-FDs (111 p64 and 42 p48): 100 aneurysms were saccular (71%), 22 dissecting (16%), 13 fusiform (9%), and five blister-like (3.6%). The successful deployment rate was 97.9%. The overall peri-procedural mortality rate was 9%, including three device-related deaths (2.2%); the overall morbidity rate was 12.1% (6.4% severe, 3.6% mild, and 2.1% asymptomatic adverse events); 4.4% of the clinical adverse events were certainly device related, and 2.2% were probably device related.
Conclusions
The use of p64 and p48 HPC-FDs is highly feasible and acceptably safe, although further data are needed to assess the impact of the coating on safety in emergency and elective procedures.
Introduction
Flow diverters (FDs) have transformed neurovascular interventions by offering a viable treatment option for complex intracranial aneurysms.1,2 Originally intended for large or giant, wide-necked aneurysms of the internal carotid artery (ICA), recent advances have extended their use to ruptured, bifurcation, and distal ones. 3
However, the first-generation FDs were associated with a higher rate of technical and clinical complications than standard treatments such as coiling,4,5 and their optimal deployment required significant expertise.1,2 Furthermore, the antiplatelet therapy necessary to prevent thromboembolic complications added the further limitations of prolonged bleeding in case of complications (especially in patients at high risk for concurrent disease or those who had experienced a recent intracranial hemorrhage), and the crucial need for long-term patient compliance in order to avoid the risk of subsequent thromboembolic events.
To overcome these limitations, various technological improvements have been made, including the introduction of devices with modified surfaces and lower profiles6,7 resulting in reduced platelet activation, adhesion and clot formation. These devices currently include the Pipeline Flex Embolisation Device with Shield Technology (SPED; Medtronic, Dublin, Ireland), the Derivo Embolisation Device (DED; Acandis GmbH & Co, KG, Pforzheim, Germany), and the p64 and p48 microwire (MW)-hydrophilic polymer-coated (HPC) flow modulation devices (HPC-FD; Phenox, Bochum, Germany). Indeed, animal studies and in vitro tests have confirmed that these surface modifications reduce platelet aggregation.8,9 Furthermore, retrospective studies have demonstrated that the use of HPC-FD p64 in dual antiplatelet therapy (DAPT) results in low rates of procedural complications. 10 However, there is still a lack of real-world feasibility, safety and efficacy data.
Our article describes the preliminary results of the analysis of data from the independent Italian registry RUAID (Ruptured and Unruptured Aneurysm Italian Dataset), regarding the 30-day peri-procedural feasibility and safety of p64 and p48 HPC-FDs in the treatment of cerebral aneurysms.
Material and methods
Ethical aspects
This study was carried out as part of the ONIRIC (Outcomes in Neurointerventional Radiology Indications and Complications) research project in accordance with the principles laid down in the Declaration of Helsinki, approved by our Ethics Committee (Lombardy Region, section Fondazione IRCCS Istituto Neurologico Carlo Besta), the STROBE cohort reporting guidelines 11 and supported by the Italian Ministry of Health (RC 2024).
Study design
Between January 2020 and December 2022 all the consecutive patients carrying an intracranial aneurysm who underwent HPC-FD implantation, were enrolled in this independent, multicenter, prospective observational post-marketing study, involving 18 Italian Neurointerventional Centers expert in the use of cerebral FD (minimum experience: ≥10 cases per year). The intracranial aneurysm could have been saccular, fusiform or dissecting, ruptured (RIA), unruptured (UIA), or even recanalized.
Extracranial aneurysms or fistula-AVM related aneurysms were excluded. RIAs treated within 30 days of a subarachnoid hemorrhage (SAH) were considered emergencies; aneurysms treated after 30 days were considered unruptured. 12
The collected baseline data were age, sex and ethnicity, medical history (co-morbidities and family history of SAH), clinical data (symptoms and modified Rankin score, mRS), aneurysm data (side, location, segment, diameter, neck dimension, morphology, thrombosis, previous treatment), and anti-platelet medication. The collected procedural data were vessels measurement modality, FD type (p64 or p48) and size, deployment technique, additional coiling, pre- and post-procedural imaging, complications (clinical and technical), and any additional treatment required. Upon discharge, the mRS and adverse events were recorded.
The patients were followed up after 3, 6, and 12 months in accordance with the standard practice at each center. Follow-up imaging data included all techniques used to visualize the aneurysm sac and parent vessel (digital subtraction angiography, DSA; computed tomography angiography, CTA; magnetic resonance angiography, MRA). Follow-up clinical data included changes in clinical status, from baseline to discharge (mRS), and pharmacological treatment, clinical complications, and any additional notes.
Devices: p64 and p48 Mw-HPC flow diverters
HPC-FDs are self-expanding, low-porosity stents with a <12-nm glycan-based multilayer hydrophilic polymer coating simulating features of the glycocalyx (thus reducing thrombogenicity) whose endovascular implantation allows the reconstruction of extra- and intra-cranial arteries. They respectively consist of 64 and 48 platinum-filled microwires made of nitinol micro-tubing. The thinness of the coating does not affect the mechanical properties of the strands or the devices.
The devices are inserted into the cranio-cervical vasculature via a microcatheter (internal diameter 0.021 inches) and, before being fully deployed, can be retracted to allow their repositioning or removal. 8 Both devices have a central, independently movable nitinol-wire with an atraumatic distal tip, and the p48 is specifically designed to treat small-diameter (1.75–3 mm) distal vessels. 3
Study endpoints
The study assessed the peri-procedural (0–30 days) feasibility and safety of the FDs. Feasibility was determined by evaluating the rates of successful and sub-optimal deployment (incorrect positioning or inadequate vessel apposition) at the end of the procedure; any additional maneuver required to optimize device opening was recorded.
Safety was assessed by evaluating clinical complication and mortality rates during the procedure and over the following 0–30 days. Adverse events were classified as serious (SAEs: events leading to permanent neurological deficits), mild (MAEs: events leading to temporary neurological deficits), or asymptomatic (AEs: without clinical significance); furtherly they were classified as certainly or probably device/procedure-related or unrelated.
The level of patient independence (changes in the mRS) was recorded at the time of discharge and during follow-up.
Data collection and analysis
The study data were recorded and stored in a protected website database allowing only encrypted access. Each center, using a unique password, was able to upload pseudo-anonymized data and images of their patients, which were centrally reanalyzed by two blinded interventional neuroradiologists with extensive experience in neurointerventional procedures (MP, AV). Each case was independently evaluated by both, who separately completed a data form; any disagreement in their analyses were resolved by means of discussion and consensus. Technical failures and misdeployments were excluded from the imaging follow-up analysis but included in the clinical follow-up analysis.
Statistical analysis
Categorical variables are descriptively presented as counts and were compared using Chi-squared tests with contingency tables; subgroup analyses were made of ruptured and unruptured aneurysms. All the analyses were made using Jamovi software for Windows. 13
Results
Patient population
One hundred and forty aneurysms were treated in 140 patients (103 females, 73.6%, and 37 males, 26.4%) using a total of 153 HPC-FDs (111 p64 and 42 p48; a second FD was added in 13 cases) (Table 4); 116 procedures (82.9%) were ex novo and 24 (17.1%) were retreatments after previous coiling (Tables 1 and 2).
Population data.
Baseline aneurysm data.
Sixty-seven patients (47.9%) were aged 50–65 years, 49 (35.0%) > 65 years, 23 (16.4%) 35–50 years, and just one (0.7%) 18–35 years. The most prevalent comorbidity was hypertension (46 patients, 32.9%), followed by cardiovascular diseases (14, 10.0%), diabetes (2, 1.4%), and other diseases (21, 25.0%); 57 patients (40.7%) had no comorbidity (Table 1).
Forty-nine patients (35.0%) underwent emergency treatment for SAH in RIA; the remaining 91 (65.0%) were UIA patients, 28 of whom (30.7%) were symptomatic. Symptoms of UIA patients were headache in 13 cases (46.4%) and neurological deficits in 15 (53.6%: cranial nerves, 7; visual decline, 6; hyposthenia and vertigo, 2) (Table 2). Ninety/91 UIA patients had mRS ≤2 (76 = 0; 13 = 1; 1 = 2); the remaining case (SAH recanalized 2 months post-coiling) had mRS=4.
Before their SAH, 47 of the 49 RIA patients (95.9%) had mRS ≤2 (46 = 0; 1 = 1) and two (4.1%) had mRS >2 (2 = 3).
Baseline aneurysm data
The aneurysms were saccular in 100 cases (71.4%), fusiform in 13 (9.3%), dissecting in 22 (15.7%), and blister-like in five (3.6%). Most (75, 53.5%) were 5–10 mm in diameter, 33 (23.5%) were 10–20 mm, and 32 (22.9%) were <5 mm, of which 23 were RIAs according to the UCAS classification. 13 Sixteen aneurysms (11.4%) were partially thrombosed. The aneurysm neck was large (dome to neck ratio >50%) in 93 cases (66.4%). One hundred and seven aneurysms (76.4%) were in the anterior circulation, while 33 (23.6%) in the posterior; 50.0% right and 42.9% left sided (Tables 2 and 3).
Aneurysms; RIA versus UIA.
Pharmacological therapy.
The involved artery was ICA in 58 cases (41.4%), BA in 26 (18.6%), MCA in 19 (13.6%), PCA in 19 (13.6%), and ACA in 18 (12.9%). The most frequent location was ICA para-ophthalmic segment (23 cases, 16.4%).
Anti-platelet regimen
Sixty-nine patients (49.3%) received single antiplatelet therapy (SAPT) with an intra-procedural loading dose (33 continued SAPT post-procedure; 36 switched to DAPT). The other 71 (50.7%) received DAPT starting 3–7 days before the procedure depending on the standard practice of the individual centers: most (67/71, 94.4%) clopidogrel 75 mg and aspirin 100 mg, but four (5.6%) ticlopidine 250 mg and aspirin 100 mg (Tables 4 and 5).
Pharmacological therapy RIA versus UIA.
Table 4 shows anti-platelet treatment sub-groups: 45/49 RIA patients (91.8%) initially received SAPT, with a loading (37/49) or intra-procedural dose (8/49); 4/49 DAPT (8.2%), 1/4 was previously on DAPT, 3/4 were treated few days post-rupture. Eighteen of the patients initially receiving SAPT (40.0%) were switched to DAPT immediately after the procedure, and 27 (60%) continued SAPT, although one of these was switched to DAPT 4-day post-procedure for severe vasospasm.
Sixty-seven of the UIA patients (73.9%) received DAPT as the standard practice of their center, and 24 (26.1%) received SAPT before the procedure with a loading (22) or intra-procedural dose (2), five of whom continued SAPT after the procedure. Only nineteen patients (13.6%) underwent platelet response tests before the procedure.
Endovascular procedures
All 140 procedures were performed under general anesthesia, using a femoral triaxial access in 90 cases (64.3%), a femoral biaxial in 43 (30.5%), a radial biaxial in six (4.3%), and a radial triaxial in one (0.7%) (Table 6).
Procedure data.
Feasibility and technical complications
As there were four technical failures in deploying p64 FDs during three procedures (2.1%), 149 FDs were implanted in 137 patients. Two of the failures were related to triaxial system instability due to severe carotid tortuosity: in one case, the FD was replaced by stent-assisted coiling; in the other, after two failed attempts at implantation, the procedure was halted and rescheduled. During the third procedure, two unsuccessful attempts at deploying a p64 FD were followed by the release of a p48 FD.
FD placement was considered optimal in 135 cases (88.2%), and sub-optimal in 14 (9.2%): these latter cases were due to inadequate wall apposition (7), partial twisting (3), or proximal or distal fish mouthing (4). Six cases of sub-optimal opening required further endovascular maneuvers (percutaneous transluminal angioplasty [PTA] in four, and a second FD in two), and three were managed pharmacologically to prevent thromboembolism; the remaining five required no further action as there was no flow impairment (illustrative case in Figure 1).
Illustrative case. DSA images during placement of a p64-HPC for treatment of an unruptured p-com aneurysm of the right ICA. DSA left images are unsubtracted, right images are subtracted. Top row DSA LL views, bottom bottom row DSA AP views. (Figure B) Final DSA images showing caudal shift of the p64, compared to Figure A (arrow). DSA left images are unsubtracted, right images are subtracted. Top row DSA LL views, bottom row DSA AP views.
Sub-optimal FD opening led to thromboembolic complications in four cases (three clinical adverse events and one asymptomatic case): one case of FD twisting led to death due to acute parent artery occlusion (PAO) despite multiple attempts at PTA; a second case of twisting led to death due to post-procedural stent occlusion; and a case of FD shortening with proximal fish mouthing leading to acute PAO, successfully treated with tirofiban but followed by post-discharge cerebral ischemia despite DAPT. In the asymptomatic case, the ICA occluded after several unsuccessful attempts to untwist/retrieve the device, but the situation was rescued by optimal contralateral compensation.
Technical complication p64 versus p48
The 14 cases of sub-optimal deployment involved 13 p64 FDs and 1 p48 FDs. As p64 FDs were used in 99 patients, the technical complication rate was 15.15% (13/99), whereas regarding p48 FDs was 7.3% (1/41) (Table 6).
Safety
Mortality
Overall mortality was 9.3% (13/140 patients) (Tables 7 and 8).
Morbi-mortality relation with the device.
Morbi-mortality outcomes SAPT versus DAPT in RIA and UIA cases.
There were three FD-related deaths (3/137, 2.2%) due to peri-procedural PAO, all in emergency-treated RIA patients. Two were related to sub-optimal FD opening (described above); the third involved a patient in their 80s with a Fisher 4 SAH and small A2–A3 bifurcation aneurysm receiving SAPT whose p48 FD occluded 24-hour post-procedure despite mechanical thrombectomy and a second anti-platelet drug (illustrative case in Figure 2). FD-unrelated mortality was 7.1% (10/140): 8 RIA patients (7 because of complications due to SAH, including vasospasm, 1 because of pulmonary complications related to prolonged bedrest) and 2 UIA patients (1 because of multi-organ insufficiency and 1 because of septic complications due to prolonged bedrest from a prior different RIA-SAH occurred 2-month before resulting mRS 4).
Illustrative case. The basal CT shows severe SAH (A). The CTA shows a small dissecting right A2 ruptured aneurysm (arrow in B), confirmed by DSA (arrow in C), treated by p48-HPC FD (arrows in D) under single aggregation. After clinical deterioration the perfusional CT demonstrated hypoperfusion in the right parasagittal territory (in E), due to parent artery occlusion at the FD level (white arrow in F), partially recanalized after mechanical thrombolisis (white arrow in G). Unfortunately the patient did not survive.
Morbidity
Overall, 30-day morbidity was 10% (14/140): nine serious (6.4%) and five mild adverse events (3.6%). There were also three clinically irrelevant adverse events (2.1%).
Serious adverse events (SAEs)
The 9 clinically relevant adverse events requiring hospitalization/prolonging hospitalization/life-threatening and leading to disability/incapacity after 30 days 14 occurred in 6 RIA and 3 UIA patients, with 3/9 certainly and 2/9 probably device-related. The first certainly device-related SAE was an intra-stent thrombosis occurring in DAPT after 24 hours, resistant to rescue maneuvers (tirofiban, stent retrieval, deployment of a closed-stent); the patient remained hemiplegic due to a deep MCA stroke (RIA, discharge mRS 4). The second case was caused by intra-stent platelet aggregation 10-day after the procedure with stent sub-occlusion (UIA, discharge mRS 3). The third ischemic thromboembolic lesions occurred in the left hemisphere causing mild aphasia and right hemiparesis the day after the procedure in a DAPT patient. DSA revealed an intra-stent thrombosis that was not present at the end of the procedure (UIA, discharge mRS 2).
The 2 probably device-related SAEs were a thromboembolic event in the territory of the ACA probably due to a transient thromboembolism in a SAPT patient (RIA, discharge mRS 4), and a case of left hemiparesis occurring after right hemisphere multi-embolic ischemia in a DAPT patient (UIA, discharge mRS 3).
The 4 SAEs unrelated to the device or procedure were 2 cases of ischemia due to vasospasm in two RIA patients (discharge mRS respectively 4 and 5), one case of pulmonary complications (RIA, discharge mRS 3), and one case of thromboembolic complications during coiling of a mirror aneurysm (RIA, discharge mRS 3).
Mild adverse events (MAEs)
The 5 (3.6%) clinically relevant but transient adverse events 14 occurred in 3 UIA and 2 RIA patients; 4 were considered device-related (three certainly and one probably), and 1 device-unrelated. The first certainly device-related MAE was an intra-procedural carotid occlusion due to stent twisting that led to transient aphasia and visual deficit recovered in few hours, thanks to good collateral supply (RIA, discharge mRS 0); the second was a case of mild motor hemi-syndrome with right homonymous inferior quadrantopsia due to a distal thromboembolism in M2, that resolved after mechanical thrombectomy (UIA, discharge mRS 1). In the third one, on DAPT, PTA was necessary to open correctly the FD, unfortunately the day after he developed SAH and ICH (UIA, discharge mRS 1).
The probably device-related MAE was a case of transient right arm hyposthenia that occurred the day after the procedure (UIA patient on DAPT) resolved within a week.
The device- and procedure-unrelated MAE was a minor stroke due to an uncontrolled hypotensive episode during the hospitalization (RIA, discharge mRS 2).
Clinically irrelevant adverse events (AEs)
The 3 procedure-related but clinically irrelevant events, 14 occurred in three RIA patients (2.1%): 2 inguinal hematomas and 1 femoral pseudoaneurysm.
Adverse event rates p64 versus p48 FDs
The 9 device-related adverse events were: 3 deaths, 3 SAEs, and 3 MAEs. The deaths occurred in 2 patients with p64 and in 1 with p48. The clinically relevant events occurred in 2 patients with p64 and 4 with p48. As the p64 was used in 99 patients and the p48 in 41, the adverse event rates were respectively 4.0% and 12.2% (Table 7).
Discussion
RUAID is the first nationwide registry recording the real-world outcomes of the p64 and p48 HPC-FD treatment of ruptured and unruptured cerebral aneurysms. Its heterogeneous population sample of 140 consecutive patients treated at 18 centers includes 49 RIA who underwent emergency treatment. The anti-platelet therapeutic protocols used vary depending on the best clinical practice of each center, with SAPT being used for 49% of the procedures and DAPT for 51%.
The revolutionary impact of the FD treatment in brain aneurysms over the last 15 years is comparable with that of the use of detachable coils in the 1990s.15,16 FDs promote the formation of a thrombus within the aneurysmal sac and stimulate endothelial cell growth along struts over the aneurysmal neck, thus allowing the aneurysm to be excluded from the circulation. 17 However, although their high-mesh density promotes occlusion, it also increases the risk of platelet activation and thromboembolic complications.
FDs were originally approved for the treatment of unruptured large, giant or wide-necked aneurysms of the ICA, but the recent introduction of surface-modified, lower-profiled devices has allowed them to be used to treat bifurcation, distal, and even acutely ruptured aneurysms. 3
Feasibility
Our data indicate an overall device placement success rate of 88.2%; a deployment failure rate of 2.6%, and a sub-optimal opening rate of 9.2%. The intra-procedural problems included inadequate wall apposition, device twisting and fish mouthing, successfully resolved by balloon angioplasty or positioning of a second FD in 5% of cases.
There were differences in the viability of the p64 and p48, with the former being responsible for all deployment failures, and the latter giving rise to fewer technical complications (15.1% vs. 7.3%).
These findings are similar to those of a recent meta-analysis 18 indicating a 4% intra-procedural technical event rate, with balloon-assisted angioplasty required in 2% of cases, although another meta-analysis of outcomes in patients treated with other surface-modified FDs (DEDs and SPEDs) reported a 99.6% successful deployment rate, with balloon angioplasty or other techniques being necessary in only 1% of cases. 4 The Italian FRED registry found that deployment was impossible in 2/163 cases (1.2%), and that misdeployment occurred in a further four cases (2.5%). 19 The Italian Derivo Registry has indicated a sub-optimal opening rate of 10.2% but, as these mainly involved early cases, they were probably due to initial inexperience. 15 To avoid or correct suboptimal FD deployment several strategies have been suggested in the literature, some of them utilized in the series described. 20
However, it is worth noting that our comparable multicenter real-world data refer to a wide range of ruptured and unruptured aneurysms of different nature, size, and location, and provide specific information concerning the apposition and deliverability of HPC-FDs. 20
Safety
We recorded a device-related mortality rate of 2.2%, an overall (certainly and probably) device-related morbidity rate of 6.6% (SAEs 3.7% and MAEs 2.9%), and a certainly device-related event rate (SAEs and MAEs) of 4.4%. All these events were due to thromboembolic and non-hemorrhagic cerebral complications, even considering the RIA subgroup, which remained on antiplatelet therapy during the vasospasm period. The mortality rate is higher than the 1% reported by Li et al. in their meta-analysis of Acandis’ DEDs and Medtronic's SPEDs, with no significant difference between the two FDs; however, all of their aneurysms were unruptured and treated electively in combination with DAPT. 18 The meta-analysis of p64- and p48-treated aneurysms by Suarez et al. indicated mortality rates of respectively 0.49% and 2%, 21 although most of these aneurysms were also unruptured and only a minority of patients received SAPT. A systematic review of complications associated with uncoated FDs by Zhou et al. reported a mortality rate of 2.8%, 22 and Brinjikji et al. found an even higher rate of 4% during the early years of FDs. 23
Our findings compare well with previously published data and suggest that surface-modified FDs lower overall mortality rates.
In comparison with our finding of 6.6% of certainly and probably device-related SAEs and MAEs, Suarez et al. reported that the complication rates associated with p64 and p48 were respectively 4% and 3%; however, only 7.2% of their aneurysms were ruptured, which is considerably lower than our 35%. Furthermore, Li et al. reported an overall pooled morbidity rate of 6.0%, mainly due to ischemic events. 18
The difference in the adverse event rates associated with the p48 (12.2%) and p64 (4%) is possibly because the higher percentage of aneurysms located beyond the circle of Willis and the consequently smaller caliber of their parent vessels.
However, the relatively high rate of complications may also be explained by the generally higher percentage of ruptured aneurysms in comparison with other cohorts. Indeed, SAH can contribute to a hyper-thrombotic state that may be exacerbated by the type of anti-platelet therapy. 21 This is supported by the finding of a 28.3% (95% CI, 12.4–52.5) overall risk of complications (thromboembolic 23.9%; hemorrhagic complications 9.4%) in a recent systematic review of the safety and efficacy of coated FDs in RIA patients receiving SAPT. 24 Advances in FD technology have greatly improved the feasibility and safety of endovascular procedures, but we are still a long way from developing a device that requires less intensive or even no anti-platelet therapy. 25 A recently published multicenter case series of ruptured aneurysm treated with P64/48 under single prasugrel regimen showed only 4.7% 26 flow diverter-dependent complications. Differently from our cases in SAPT, all their patients were tested as good responders before the procedure and some of them were pretreated with eptifibatide (Integrilin; GlaxoSmithKline, Munich, Germany), cangrelor (Kengrexal; Chiesi Farmaceutici, Parma, Italy), or tirofiban (Aggrastat; Carrevio, Vancouver, Canada) IV before switching to prasugrel. Additionally, prasugrel is still not widely available in Italy for cerebral endovascular procedures.
The limitations of our study include the multicenter nature and the heterogeneity of the study population. Differences in the anti-platelet regimens and in the experience of the investigators in deploying the p64 or p48 may have increased the complication rate, affecting the results. However, it is important to note that our dataset reflects the real-world use of FDs in Italy.
Conclusions
Our preliminary real-world data concerning a heterogeneous population of patients with ruptured and unruptured brain aneurysms receiving SAPT or DAPT demonstrate that the p64 and p48 have a generally good feasibility and safety profile, especially in the case of unruptured aneurysms. The p48 seems to be easier to deploy than the p64, but seems to be associated with higher rate of complications. Further studies are required to evaluate the safety of the devices in acute settings and in combination with SAPT.
Footnotes
Author’s Note:
Elisa Ciceri, Matteo Milani, Giuseppe Ganci and Valentina Caldiera are also affiliated with Foundation IRCCS Carlo Besta Neurological Institute, Interventional Neuroradiology Milan, Lombardy, IT. Chiara Riccietti is also affiliated with AUSL di Bologna, Bologna, Emilia-Romagna, IT.
Consent to participate
All participants provided written informed consent prior to enrolment in the study.
Data availability
The data that support the findings of this study are available from the corresponding author, Milani Matteo, upon reasonable request.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical approval
This study was carried out as part of the ONIRIC (Outcomes in Neurointerventional Radiology Indications and Complications) research project in accordance with the principles laid down in the Declaration of Helsinki, approved by our Ethics Committee (Lombardy Region, section Fondazione IRCCS Istituto Neurologico Carlo Besta).
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Italian Ministry of Health (RC 2024).
ORCID iDs
Simone Comelli https://orcid.org/0000-0002-7175-6433