In this study, we compared the safety, angiography outcome, and clinical outcome of acutely RIAs with incomplete occlusion after SAC and NSC. We first discussed the advantages and indications of using stents for treating acutely RIAs, and then analyzed the reasons for higher complications rate and mortality in the SAC group. Our study had also indicated that the progressive occlusion of stent deployment can promote complete occlusion of acutely RIAs with incomplete occlusion. We were able to detect several risk factors and independent predictors of recurrence.
Clipping and Coiling are 2 accepted treatment strategies for IAs. Clipping has an advantage in treating acutely ruptured wide-necked and complex IAs. It also can improve complete occlusion rate and reduce costs [9,10,11]. Surgery, however, may also be challenging in some of these lesions, since clips may slip, and surgical access may be limited because of the swelling of the brain in the acute setting of a subarachnoid hemorrhage (SAH). Coiling was associated with a reduced mortality rate and improved outcomes for patients with RIAs according to the results of the International Subarachnoid Aneurysm Trial (ISAT) [9, 12]. And Maud et al. found that coiling has better cost-effectiveness ratio, the costs of coiling would be progressively decreased and eventually reversed as time goes on . Other advantages of coiling are less surgical injury, better cognitive outcome, and lower complication rate [10, 13]. In addition, with the rapid technological advances in endovascular treatment, aneurysms with complex geometry can also be effectively treated with endovascular techniques [2, 3, 14,15,16]. Therefore, endovascular coiling is becoming a widely accepted treatment modality of RIAs.
There are many endovascular coiling techniques to treat complex and wide-neck IAs. The most common assisted-coiling techniques are balloon, flow diverter (FD) and stent. The balloon-assisted coiling (BAC) provides a contemporary scaffold effect that prevents coil protrusion and is associated with an improved occlusion rate after endovascular therapy in previous studies [17,18,19]. Wang et al.  conducted a meta-analysis comparing BAC and SAC in IAs, and found that SAC achieved a better complete occlusion rate of aneurysms at 6 months or later after the procedure, with comparable complication and retreatment rates. According to previous studies, FD tends to achieve improved occlusion rates at immediate and follow-up angiography [14, 15]. However, the periprocedural complication rate seems to be high. Zhou et al.  reported that the complication rate for RIAs was 30.6%% after FD. And FD requires prolonged dual anti-platelet therapy (DAPT), limiting utility in the setting of acute SAH. Additionally, temporary SAC is seemed to be an alternative option for RIAs .
Acutely RIAs treated with SAC is still controversial because of concerns about the complexity of the procedure, thrombotic risk, and antiplatelet drugs [22, 23]. These drugs expose patients to an elevated risk of hemorrhagic complications if additional surgical interventions such as decompressive craniectomy, external ventricular drainage (EVD) placement, are required [23, 24]. Moreover, DAPT can increase the risk of all types of bleeding by more than 40% . It is still gradually accepted by neurosurgeons due to its advantages in terms of effectiveness, durability, and facility in treating complex and wide-necked IAs [3, 4, 26]. SAC can also improve aneurysmal complete occlusion rate, reduce recurrence rate, and decrease incidence of rebleeding or retreat rate [27, 28]. Furthermore, SAC brings a comparable complication rate than to NSC [2, 17, 26,27,28,29]. And the clinical outcome of the two groups are also comparable, such as a recent systematic review by Zhang et al. , shown that favorable clinical outcome rate of SAC and NSC group at discharge and at follow-up were 73.4% vs 74.2, 85.6% vs 87.9% respectively.
In the SAC group of this study, the incidence of ischemic and hemorrhagic complication was 21.8% (22/101), 14.8% (15/101), respectively. It showed a higher incidence than previous systematic reviews. Nishido et al.  revealed that the complication rate of stent group was 9.4% (28/299; ischemia: 7.0%; hemorrhage: 2.3%). Bodily et al.  reported that total complications rate was 13.6% (39/288) in the stent group (hemorrhage: 8%; ischemia: 5.6%). In several large sample studies, Ryu et al.  reported that the event rates of stent group in thromboembolism and intra- and postprocedural hemorrhage were 11.2% (122/1090), 5.4% (59/1090), and 3.6% (39/1090), respectively. In the article by Shapiro et al. , the overall procedure complication rate of stent group was 19% (288/1517). In a recent systematic review by Zhang et al. , showed that the overall perioperative complications in the SAC group was 20.2% (101/499).
Complications of the stent group in this study was mainly caused by intraoperative thrombosis (15.8%, 16/101) and IOR (11.9%, 12/101). The number one reason of high intraoperative thrombosis events probably was related with the method of antiplatelet administration. Ryu et al.  concluded that the antiplatelet regimen used in this study was more likely to bring perioperative thromboembolic complications to patients, Second, the high intraoperative thrombosis events of stent group might be associated with the experience of the endovascular neurosurgeon. The neurosurgeon with little experience may need more endovascular procedure time to place stents, that may increase the incidence of intraoperative thrombosis. These junior neurosurgeons also need to repeatedly retrieve and release stents to ensure that the stents are placed in the optimal position, which may induce vessel endothelium injury, platelet aggregation, vasospasm, or thrombosis. Although, the intraoperative thrombosis rate of the SAC group in this study (15.8%) was consistent with previous literature (5.3–20%) [2, 30, 32, 33], such high incidence of intraoperative thrombosis events still deserved attention and improvement. Third, the poor Plavix response and inadequate use of DAPT may also contribute to the high thromboembolism complications rate of present study. Lee et al.  found that 42.9% of their patients had a poor response to clopidogrel, and all thromboembolism events occurred in the poor-response group, which implied that a poor response to antiplatelet medication might induce thromboembolism complications. Piotin et al.  and Choi et al.  reported that the SAC group had a significantly higher permanent complication rate than the coiling-alone group when DAPT wasn’t used in patients treated with stenting. Therefore, we concluded that the consequences of DAPT’s inadequate use may be similar to that of poor Plavix response or not using of anti-platelet drugs. Additionally, acute SAH is usually regarded as a hypercoagulable state in which the tendency for thrombosis is high [29, 37]. Patients who undergo SAC after period of vasospasm (day 15–20) might have a different complication rate compared to patients treated early with SAC. We therefore performed a comparative analysis of complications in early (before 72 h) and late (day 3–28) stage of SAC and NSC group, and found that timing of treatment has no significantly influence on complications (Tables 4 and 5). These results showed that the SAC is as safe early as well as late after aneurysm rupture.
IOR is a severe complication of coiling and is associated with high rates of morbidity and mortality. According to the review of the Cerebral Aneurysm Re-rupture After Treatment (CARAT) study, the coiling-related IOR occurred in 5% of patients, and IOR was associated with 64% rate of periprocedural death and disability . Recently, Kocur et al.  summarized previous literature and showed that the incidence of IOR ranges from 1 to 8.7% and the risk of permanent neurologic disability and death related to IOR is up to 63%. There are many risk factors for the IOR, such as ruptured aneurysms, small aneurysms size, device size, race, and lower initial Hunt and Hess Grade [38,39,40,41], even intra-aneurysmal contrast media injection and sudden rise in arterial pressure may also induce IOR . In this study, the IOR occurred in 14.6% of patients, such high IOR rate in the two groups probably contributes to the experience of the endovascular neurosurgeon and the aneurysms status. In the NSC group, the incidence of coil protrusion may higher than SAC group because aneurysm neck lacks stents coverage. Coil protrusion usually requires re-coiling, and a high frequency of repeated coiling may increase the incidence of IOR in the NSC group. It’s worth noting that IOR caused 77.8% (7/9) of all death events in present study. Moreover, the mortality of patients with IOR in the SAC group (41.7%) was higher than that in the NSC group (11.8%). These results revealed that stent deployment may raise the death risk of patients with IOR because tirofiban can affect the physiological hemostasis when aneurysm IOR occurs.
Additionally, some studies even reported that ruptured aneurysm was an independent predictor of complications [35, 37, 42]. Ryu et al.  found that the risk of complications rate were higher in the RIAs than in the UIAs group. And Bechan et al.  reported that the complication rate of SAC with early adverse events in RIAs was 10 times higher than that in UIAs. And, we found that previous studies included both RIAs and UIAs in reporting the overall complication rate [22, 30, 31]. Therefore, all included aneurysms of this study were RIAs may also be one of the reasons why the complication rate is higher than that in previous literature. The age, sex, underlying disease, past medical history, initial GOS score, Hunt-Hess grade and Fisher grade may also affect the complication rate and clinical outcome of patients with acutely RIAs, but it is comparable between the two groups in this study. There was a significantly wider aneurysm neck and smaller dome-to-neck ratio in SAC group. However, this difference is inevitable, and the original intention of using stents is to treat complex and wide-necked IAs. Many studies reported differences or significant differences in aneurysm neck and dome-to-neck ratio between the two groups [2, 3, 28, 29, 43].
There was no statistical difference in mortality and rebleeding rate between the two groups in this study, and are consistent with previous studies [2, 23, 27,28,29, 43]. The mRS score 0–2 at discharge and 6-month follow-up of SAC vs NSC was 81.9% vs. 90.6% (P = 0.081), 83.3% vs. 89.9% (P = 0.244) respectively in this study. Zhang et al.  reported similar clinical outcome to our study. We also found that the favorable clinical outcome rates of recent studies were improved than before [12, 19, 27]. The improvement probably attributed to advances in interventional techniques and physicians’ experience in managing SAH patients. In summary, stent placement is associated with a higher rate of complications, mortality and rebleeding rate, but the favorable clinical outcome was still similar to the NSC group.
The use of stent usually brings future superior angiography outcome for patients, even in incomplete occlusion aneurysms [1, 2, 5, 22, 27]. This result has also been confirmed by multivariable analysis of this study. Coil compaction and aneurysm growth are two methods of aneurysm recurrence [44, 45]. Stent placement may increase the capacity of coils to resist the impact of blood flow. Stents can also promote thrombosis within the aneurysmal sac by a scaffold effect, which reduces the recurrence rate [1, 5, 6, 35]. Flow-diversion function and facilitation of endothelialization of stents may also induce changes in intra-aneurysmal hemodynamics and promote delayed occlusion and obtain complete occlusion in the long term [6, 46]. However, our study still has a higher recurrence rate and lower occlusion rate than previous studies both in the SAC and NSC group [1, 5,6,7]. Furthermore, in several previous studies, progressive complete occlusion of stent placement was achieved 76.5–88.3% in follow-up imaging studies at 6 months [1, 5], which is also higher than 65.6% reported in this study. The first reason is that previous studies mainly described the UIAs, while all included aneurysms in this study were RIAs. RIAs are usually regarded as a predictor of recurrence [2, 35]. Second, the overall statistics of this study show higher recurrence rate and lower occlusion rate compared to literature data probably related to the experience of the endovascular neurosurgeon. Additionally, Bechan et al.  reported that the DAPT maybe increase the risk of aneurysm recurrence. Nevertheless, our study found that despite the use of DAPT, the SAC group’s recurrence rate was still lower than in the NSC group, which suggested that the progressive occlusion of stent deployment may not be entirely compensated by the progressive regrowth of aneurysms and the use of DAPT.
Although we conducted some objective and reliable results, and still with some limitations. The major limitation of this study is that this is a retrospective study with no randomization; the patient’s selection of SAC group didn’t establish uniform inclusion and exclusion criteria, so selection bias was inevitable. Additionally, aneurysm characteristics between the SAC group and the NSC group were the slight differences, especially the differences in aneurysm neck and dome-to-neck ratio. Finally, the BAC was not thoroughly considered in deciding the treatment plan because the BAC technique is still developing in our institute and we didn’t have much experience, so the complications of SAC group may overestimate by this study due to patients who are suitable for BAC are selected into the SAC group. Another limitation is that this study had no long-term follow-up results and small sample size.