Safety and effectiveness of Neuroform Atlas stent-assisted coil embolization for ruptured intracranial aneurysms

Population

This study represents a retrospective review that included patients who underwent endovascular treatment for ruptured aneurysms at a single institution. Between January 2018 and September 2022, endovascular coil embolization was performed in 245 patients with ruptured intracranial aneurysms. Radiological records obtained between January 2018 and December 2022 were retrospectively examined. The study included patients aged 20–90 years who underwent endovascular treatment exclusively for ruptured saccular aneurysms. Patients were excluded if they underwent stent monotherapy, had atypical aneurysms (e.g., thrombosed or dissecting aneurysms), or had aneurysms with a maximum diameter greater than 25 mm. Among the 245 patients with ruptured cerebral aneurysms, 191 were treated using simple coiling or the Neuroform Atlas SAC (Fig. 1). Patient data, including age, sex, past medical history (such as hypertension, diabetes mellitus, and dyslipidemia), height, weight, aneurysm size, location of the ruptured aneurysm, follow-up period, periprocedural complications (such as intraprocedural rupture (IPR) and TEE, and late complications (such as retreatment and in-stent stenosis (ISS)), were meticulously collected.

Fig. 1.

Flow diagram of patient selection

Procedures and medications

SAC was performed when the dome-to-neck ratio was above 2 or when it was difficult to create a stable frame using coiling alone. The bare platinum coils used included the target coil (Stryker, Cork, Ireland) and MicroPlex coil (MicroVention, Aliso Viejo, CA, USA). NAS was exclusively used. All catheters were flushed with heparinized normal saline, and sufficient coil packing into the aneurysm sac was achieved before the intravenous (IV) bolus of heparin (3000 IU to 5000 IU). In cases of SAC, pretreatment with antiplatelet agents can be challenging due to the risk of hemorrhage. Consequently, only IV heparin (3000 IU to 5000 IU), an antithrombotic agent, was injected immediately after sufficient coil packing following stenting. Following the heparin injection, the activated clotting time was maintained between 250 and 300 s, as usual. Heparin at a dose of 3000–5000 IU was administered every 6 h for 24 h after the procedure. A loading dose of dual antiplatelet therapy (aspirin 300 mg and clopidogrel 300–450 mg) was administered orally immediately, and the maintenance dose (75 mg clopidogrel daily for 8–12 weeks and 100 mg aspirin daily indefinitely) was started one day after the procedure was completed.

Definition

Treatment-related complications were categorized into early (IPR and TEE) and late (ISS and retreatment) complications. IPR was defined as the identification of an increasing or newly formed hematoma by post-embolization brain computed tomography (CT), the detection of contrast leakage during the procedure, or by post-embolization brain CT. TEE was defined as the identification of a filling defect of contrast around the aneurysmal sac on cerebral digital subtraction angiography (DSA) during the procedure or the observation of a signal change on diffuse-weighted Magnetic Resonance Imaging in symptomatic patients after the procedure. Retreatment was defined as the presence of marked compaction and recanalization of a coiled aneurysm on radiography or cerebral DSA. ISS was defined as a decrease in the vessel diameter around the implanted stent on follow-up angiography as compared to the original diameter on immediate post-stenting angiography.

Follow-up protocol

Our protocol for angiographic follow-up using only catheter-selected angiography is as follows: The first angiographic follow-up was conducted 3–6 months after the operation based on clinical outcomes and immediate angiographic results. The next follow-up was performed 12 months later depending on the angiographic results from the first follow-up. The third follow-up occurred 12 to 24 months later, based on the results of the second follow-up. All subjects were followed up for a period ranging from 12 to over 30 months.

Statistical analysis

Univariate analysis was used to identify the factors affecting ISS. The Chi-squared test or Fisher’s exact test was used to analyze categorical variables, whereas the independent t-test was used to analyze continuous variables. All statistical analyses were performed using SPSS version 23.0.0 (IBM, Armonk, NY, USA). Statistical significance was set at p < 0.05.

Thromboembolic events

Choi et al. [4] used previous models of Neuroform and Enterprise stents (Codman, Raynham, MA, USA) for SAC, and the antithrombotic regimens were the same as ours. They reported procedural TEE in the SC versus the SAC (12.4% VS 25.5%) for ruptured aneurysms. In another meta-analysis [24], the occurrence of TEE in the SAC (stent-assisted coiling) group, which used the previous stent model, was 12.3%, which was higher than that in the non-SAC group (8.4%). In our study, TEE occurred in 5 out of the 127 patients who underwent SC (3.9%) and in 4 of the 64 patients who underwent NASAC (6.3%). Thus, using NAS for SAC has allowed us to observe a lower incidence of TEE as compared to the results obtained using a previous model of the stent. This is believed to be due to the thinner strut thickness (55 microns) and a lower metal-to-artery ratio (6.0-12.0%). Therefore, the stent profile could be considered a crucial factor in these occurrences.

Russo et al. [17] initially reported the TEE rate in ruptured aneurysms treated with NASAC. Among the 61 patients, 22 (36.1%) experienced acute stent thrombosis, ranging from minimal aggregation (four patients, 6.5%) to complete occlusion (18 patients, 29.6%). In their study, although IV Aspirin was used during the procedure, the TEE rate was higher than that in our study. This is because the proportion of patients with aneurysms in large-diameter arteries (especially distal ICA, 9.8%) was smaller in their study than in ours. As shown in our results (see Table 3), TEE appears to occur more frequently in blood vessels with small diameters. Additionally, differences related to ethnicity or multicenter studies may be influenced by differences in TTE rates.

In a study by Yi et al. [23] that compared SC and NASAC groups to assess TEE in the treatment of ruptured aneurysms, the results for SC and NASAC were 4.2% and 4.2%, respectively. In the case of SAC, the TEE rate was similar to that of SC, which could be attributed to the effects of antiplatelet premedication. When comparing the two studies, the use of premedication with antiplatelet agents was a distinguishing factor in the procedural process. We did not use antiplatelet premedication as it was considered disadvantageous for premature rupture or IPR. Therefore, we chose not to administer a premedication of antiplatelets in our study. More than half of our cases did not exhibit any symptoms, yet our rate was slightly higher at 6.3% compared to their results.

Ruptured aneurysms that require stent placement are morphologically complex and challenging to treat. A multiple-catheter technique without stents was employed to create a stable framework, necessitating repetitive coil movements. These movements may have contributed to the occurrence of IPR and TEE. The use of stents can reduce the treatment time and risk of intraprocedural rupture due to excessive coil movement during the multiple catheter technique of simple coiling. In addition, it can help increase the coil packing density more easily and efficiently. Considering these factors, although the stent usage rate was slightly higher (6.3%) than that of simple coiling (3.5%), the use of stents for ruptured cerebral aneurysms is reasonable.

Anti-thrombotic medication

TEE, a major concern in neurointervention, is often unpredictable and difficult to control, especially when stents are inserted. Although antithrombotic therapy can be employed to reduce TEE, caution is warranted because of the increased risk of hemorrhagic complications associated with it. Many researchers are exploring various treatment modalities to mitigate hemorrhagic complications and decrease TEE rate [2,4,17,18,22-24].

In a study by Russo et al. [17], 500 mg of IV acetylsalicylic acid (ASA) was administered following stent insertion. In fact, IV aspirin was used in the stent of the previous model of NAS. However, since the introduction of the NAS in 2018, IV ASA imports have been discontinued in Korea. Although there is no definitive evidence on the effectiveness of IV aspirin in preventing TEE, theoretically, preventative effects can be expected. A study by Yi et al. [23] loaded Dual Antiplatelet Therapy (DAPT) immediately before stent insertion. Theoretically, when stent placement is anticipated in ruptured cerebral aneurysms, premedication with loading doses of aspirin and clopidogrel can be considered [20]. Clopidogrel, when administered with a loading dose of 600 mg, takes approximately 2 hours to achieve adequate antiplatelet effects, whereas a loading dose of 300 mg requires 4 hours [8]. Therefore, the procedure timing should be carefully planned considering these factors. Although one report suggests that the selective use of antiplatelet agents in ruptured cerebral aneurysms reduces thromboembolic events without increasing bleeding complications [7], further research is needed to assess the overall benefits and risks of routine platelet inhibition beyond specific cases. In a report by Kim et al., all patients received IV tirofiban instead of premedication with loading doses of aspirin or clopidogrel before stent-assisted coiling [10]. In a study by Limaye et al., IV tirofiban was administered as a maintenance dose immediately before SAC and flow diversion device deployment occurred intraoperatively [13]. Multiple studies suggest that tirofiban could be safe and beneficial [10,12-14]. The effectiveness of tirofiban has been reported in the literature with excellent recanalization rates and favorable outcomes [10,12-14]. However, although many reports on the effectiveness of tirofiban injection have been published, the Korean Ministry of Food and Drug Safety has approved the use of tirofiban for cardiovascular disease but has not yet approved its use for cerebrovascular disease despite many research results. Therefore, caution should be exercised in its use.

In our study, an IV Heparin bolus was injected without additional premedication with the antiplatelet regimen in all cases. Some physicians did not use a separate prethrombotic regimen for SAC. However, this approach appears to increase the risk of TEE. Some authors are concerned about the risk of rebleeding and the need for additional surgery with the use of antiplatelet premedication [1,2,11,21]. Choi et al. suggested that SAC without antiplatelet premedication is a viable option for patients with acutely ruptured aneurysms [4]. Although direct comparisons are challenging due to the variability of protocols among physicians, further research is needed to determine whether these differences have contributed to the disparity in thromboembolic event rates.

Retreatment

In our study, the retreatment rates were 19/72 (26.4%) in the SC group and 10/45 (22.2%) in the NASAC group. In a study by Yang et al. on the insertion of a previous model of stents, the retreatment rate was significantly higher in the SC group as compared to the SAC group (13/73, 17.8% vs. 3/53, 5.7%) [22]. According to the meta-analysis, there was a significantly lower rate of retreatment at follow-up in the SAC group as compared to that in the SC group (14/290 vs. 74/447; 4.8% vs. 16.6%) [24]. Muto et al. showed that the retreatment rate for patients who underwent SAC for ruptured aneurysms was 6/40 (15%) [15]. Our study using the NAS showed a higher retreatment rate as compared to other studies using previous models of the stents, which could be attributed to the lower metal coverage of the Atlas. We believe that this might be explained by the change in blood flow due to stent placement and that the higher metal coverage might be more likely to induce higher packing density. In a study by YI et al. [23], the NASAC group exhibited a lower rate of recanalization and retreatment than the SC group after propensity score matching (11.3% vs. 25.4%, and 7.0% vs. 16.9%, respectively). However, the retreatment rate in the NASAC Group was higher in our study. Our study showed a higher retreatment rate than other studies, even when using the same NAS, for the following reasons: We have stricter retreatment indications and follow-up protocols than other institutions. In addition, our protocol for angiographic follow-up involved only catheter-selected angiography. However, the definitions and indications for retreatment vary among practitioners.

Limitations

The present study had a few limitations. First, it was a retrospective chart review conducted at a single institution. These situations might have increased the risk of selection bias. Second, the sample size enrolled in this study was too small to reach statistically significant conclusions. Third, it is difficult to apply the study results to cases of cerebral aneurysms with atypical aneurysms (e.g., giant, thrombosed, or dissecting aneurysms). Despite these limitations, our study could provide guidance for planning treatment strategies.