Evaluation of clinical and anatomical outcome of staged stenting after acute coiling of ruptured intracranial aneurysms

Abstract

Introduction

Stent-assisted coiling has widened indications and improved stability of endovascular treatment of intracranial aneurysms. However, stent-assisted coiling is usually not used to treat acutely ruptured intracranial aneurysms to avoid antiplatelet therapy. The objective of this study is to evaluate a strategy of staged endovascular treatment of ruptured intracranial aneurysms including coiling at the acute phase with complementary stenting with or without coiling at the subacute phase.

Material and methods

Between 2012 and 2017, we retrospectively identified, in our prospectively maintained database, all patients treated for a ruptured intracranial aneurysm based on this staged stenting strategy. Clinical charts and imaging follow-up were analyzed to assess the procedural safety and feasibility as well as clinical and anatomical outcome.

Results

We identified 23 patients with 23 intracranial aneurysms including 15 (65.2%) women with a mean age of 50 years (range 24–69 years). No rebleeding occurred during the mean delay of 24.3 days between initial coiling and stenting. All procedures were successful and additional coiling was performed in 5/23 procedures (21.7%). Clinical status was unchanged in all patients. At follow-up, the modified Rankin scale was graded 0 in 19/23 (82.6%), 1 in 2/23 (8.7%), and 2 in 2/23 (8.7%) patients, respectively. The rate of complete occlusion rose from 30.4% before the stenting procedure to 52.2% immediately after and 72.7% at follow-up.

Conclusion

This strategy of early staged stenting in selected patients is safe and improves immediate intracranial aneurysm occlusion and long-term stability in this population at high risk of intracranial aneurysm recurrence with coiling alone.

Introduction

Endovascular treatment (EVT) of ruptured intracranial aneurysms (IAs) is now the first option to prevent early rebleeding.¹ One limitation of EVT is related to the risk of aneurysm recurrence at follow-up, specifically for ruptured IAs.² Stent-assisted coiling (SAC) is now established as a safe and effective treatment that significantly decreases the recurrence rate^3,4 as it allows to obtain a better immediate occlusion, works as a scaffold for secondary neck endothelialization, and provides a flow diversion effect.^5,6 However, SAC requires a double antiplatelet treatment (DAT) and therefore, it is not routinely performed to treat ruptured IAs to avoid hemorrhagic complications. For these reasons, some teams including ours may decide to perform EVT of ruptured IAs in two stages in selected patients: first, coiling of the aneurysmal sac in emergency to prevent a rebleeding; second, complementary stenting with or without adjunctive coiling at a subacute stage.⁷ In our department, we identify patients at high risk of recurrence based on clinical and angioarchitectural characteristics and patients in whom complete IA occlusion was not achieved at the acute phase. These patients undergo a stenting of the parent artery, completed with coiling of the residual sac if the occlusion is incomplete, if the remnant is judged to bear a risk of evolution, and rerupture and if its size allows to place additional coils.

The objective of this study is to evaluate the feasibility, safety, and efficacy of this staged EVT of ruptured IAs.

Material and methods

This retrospective study was approved by our local Ethical Committee.

Population

Patients

Between January 2012 and June 2017, we retrospectively identified, in our prospectively maintained database, all patients initially treated by coiling at the acute phase for a ruptured IA and who benefited of a complementary stenting procedure. We have excluded patients treated with flow diverter stents.

Aneurysms

All patients underwent pre-therapeutic digital subtraction angiography (DSA) including AP and lateral views of both carotid and vertebral arteries. Then, 3D rotational angiography was obtained on the artery bearing the target IAs to precisely depict aneurysm morphology. The IA location and size were recorded as well as the neck size.

The IA occlusion after initial coiling was evaluated on immediate post-treatment DSA using a 3-grade scale⁸: (1) complete occlusion (no contrast filling the aneurysm), (2) neck remnant (residual contrast filling the aneurysmal neck), and (3) incomplete occlusion (residual contrast filling the aneurysmal body).

The smallest diameter of the stented vessel was recorded.

Therapeutic strategy and patient selection

In our institution, the treatment modality for ruptured IAs is established by a multidisciplinary team based on angioarchitectural data from CT angiography or DSA. EVT is considered as the first option for ruptured IAs unless immediate surgery is required, e.g. because of a compressive hematoma or EVT is impossible without the placement of a stent and thus dual antiplatelet treatment. In this latter case, EVT is performed only in cases where no other options are possible. In case of EVT, the primary objective of the treatment is always to obtain a complete occlusion, with a packing as dense as possible most often using the balloon-assisted technique. The strategy of early staged stenting aims to improve short-, mid-, and long-term occlusion of selected IAs associated with a high recurrence risk. Indeed, by reducing the recurrence and retreatment rates as well as the rebleeding risk, we expect a better overall clinical outcome and quality of life in those patients. Our multidisciplinary team identified patients with either an incomplete IA occlusion and/or a risk of IA recurrence after coiling estimated to be higher based on one or several aneurysm characteristics such as an aneurysm diameter ≥5 mm, a neck width ≥4 mm. Complementary stenting was not performed in patients with a poor prognosis based on the clinical status (altered consciousness, severely disabling, or life-threatening ongoing conditions) and imaging evaluation. The staged stenting was performed at least two weeks after IA rupture to avoid vasospasm and potential invasive procedures (e.g. intracranial pressure monitoring) after antiplatelet administration and within eight weeks to minimize the risk of early IA growth or rupture. Residual vasospasm was excluded based on pre-stenting DSA. Additional coiling was performed after stenting in case of grade 3 occlusion or in case of grade 2 occlusion with a large residual neck allowing safe and straightforward catheterization and coiling of the sac according to the operator.

Endovascular procedures and antiplatelet protocol

Our antiplatelet protocol includes a loading dose of 300 mg Clopidogrel and 320 mg aspirin the day before the procedure and the day of the procedure, at least 3 h before general anesthesia is induced. If the bodyweight exceeds 80 kg, loading doses are raised to 450 mg Clopidogrel and 480 mg aspirin. The residual platelet activity is not routinely tested in our institution.

In all patients, EVT was performed under general anesthesia and systemic heparinization. The adequacy of systemic anticoagulation was monitored by repeated measurements of the activated clotting time (ACT). A baseline ACT was obtained prior to the bolus infusion 30–50 IU/kg body weight of heparin, and hourly thereafter. The bolus infusion of heparin was followed by a continuous drip (1000–1500 IU/h), with the purpose of doubling the baseline ACT. At the end of the procedure, systemic heparinization was maintained for 24 h in most patients. All procedures were performed by a senior interventional neuroradiologist.

The antiplatelet treatment is maintained as follows: Clopidogrel 75 mg/day from day 1 after EVT for one month (internal carotid artery) or three months (more distal locations). Aspirin is maintained for at least six months and further treatment is individualized depending on stent tolerance on DSA and the presence of other cardiovascular risk factors.

The number, type, and size of implanted stents as well as the realization of an additional coiling were recorded.

Immediate outcome

Procedural and early post-procedural (within 48 h) complications were recorded. Clinical course was recorded by review of clinical charts following the modified Rankin scale (mRS)⁹ at discharge. Worsening of symptoms or death was evaluated by a vascular neurologist and/or an interventional neuroradiologist.

Aneurysm occlusion was evaluated on immediate post-treatment DSA after stenting procedure using the same 3-grade scale.⁸ Immediate stent tolerance was graded as follows: (0) no stenosis and no thrombosis, (1) stenosis <50%, (2) stenosis 50–70%, (3) stenosis >70%, (4) stent occlusion.

Patient follow-up

We recorded delayed complications based on clinical charts review. The clinical follow-up was obtained and evaluated using mRS⁹ at three months.

Our imaging protocol for follow-up of conventional stenting procedures in hemorrhagic presentation includes a magnetic resonance angiography (MRA) at 6 and 12 months and then yearly as well as a DSA at 6 months, 12 months, and 5 years.

Aneurysm occlusion was evaluated using the same 3-grade scale.⁸

The stent tolerance was evaluated and potential stenoses were graded as follows: (0) no stenosis, (1) stenosis <50%, (2) stenosis 50–70%, (3) stenosis >70%, (4) stent occlusion.

Results

Patients

From January 2012 to June 2017, a ruptured IA was found in 291/331 patients (88%) admitted in our institution for non-traumatic SAH. Those IAs were treated surgically in 52/291 patients (18%) and by EVT in 239/291 patients (82%). Among EVT procedures, we identified 23/279 (8.2%) early staged stenting procedures for 23 IAs. There were 15 (65.2%) women and 8 (34.8%) men with a mean age of 50 years (range 24–69 years).

Patients were classified according to the World Federation of Neurological Surgeons grading scale at presentation.¹⁰ Thirteen patients were grade I, six were grade II, and four were grade IV.

The SAH was classified according to the Fisher classification.¹¹ One was grade I, two were grade II, 19 were grade III, and one was grade IV.

Aneurysms

Mean aneurysm diameter was 7.1 mm (range 4–15 mm) and the mean neck diameter was 3.4 mm (range 2–6 mm). The mean dome-to-neck ratio was 2.08 (range 1.16–3.75). Nine/23 IAs were located on the anterior communicating artery (Acom) (39.1%), 5/23 on the middle cerebral artery (MCA) (21.7%), 4/23 on the posterior communicating artery (Pcom) (17.4%), 1/23, respectively, on the postero-inferior cerebellar artery, anterior choroidal artery, basilar tip, pericallosal artery, and carotid termination (4.3% each). Aneurysm characteristics are detailed in Table 1.

Table 1.

Aneurysm and neck diameters.

Aneurysm	Dome (mm)	Neck (mm)	Dome/neck ratio	Location
1	8.15	5	1.63	Acom
2	8.8	4.7	1.87	Pica
3	5	2.5	2	Acom
4	4.4	2	2.2	Carotid T
5	4.8	2.5	1.92	Ant. choroidal
6	7	5	1.4	Acom
7	8	4	2	Pcom
8	7.7	4	1.93	MCA
9	6.1	3.5	1.74	Pericallosal
10	6	3.4	1.77	Basilar tip
11	4.3	3.7	1.16	Acom
12	9.6	3	3.2	Pcom
13	11.7	6	1.95	MCA
14	4	2	2	Acom
15	15	4	3.75	Pcom
16	6.5	3	2.17	Acom
17	5.7	2.9	1.97	Acom
18	4.9	3	1.63	Pcom
19	12	5.5	2.18	Acom
20	6.6	2.3	2.87	Acom
21	6	3	2	MCA
22	6	3	2	MCA
23	5	2	2.5	MCA

Acom: anterior communicating artery; Ant. choroidal: anterior choroidal artery; Carotid T: carotid termination; MCA: middle cerebral artery; Pcom: posterior communicating artery; Pica: postero-inferior cerebellar artery.

The mean smallest diameter of the stented segment of the parent artery was 2.3 mm (range 1.5–3.5 mm).

After emergency coiling, occlusion was graded 1 in 7/23 (30.4%), 2 in 13/23 (56.5%), and 3 in 3/23 (13%) patients, respectively. Balloon-assisted coiling was performed in 9/23 procedures (39%).

Endovascular procedures

The mean delay between initial coiling and stenting was 24.3 days (range 2–49 days). In one patient, stenting was performed only two days after coiling because she presented more than a week after initial rupture with a severe and symptomatic vasospasm, resolved at the time of stenting. No rebleeding was observed during this delay. All 23 stenting procedures were successful. One stent was implanted in 22 patients whereas a second stent was necessary in one case because of unsatisfying neck coverage with the first device due to proximal migration after deployment. Overall, 24 stents were thus deployed including nine Baby Leo (Balt, Montmorency, France), seven Enterprise (Codman Neurovascular, Raynham, MA, USA), four Lvis Junior (Microvention, Aliso Viejo, CA, USA), two Enterprise2 (Codman Neurovascular, Raynham, MA, USA), one Leo (Balt, Montmorency, France), and one Solitaire (ev3, Irvine, CA, USA) device.

Additional coiling was performed in 5/23 procedures (21.7%) because of a grade 3 (n = 3) or grade 2 (n = 2) occlusion after initial coiling (Figure 1). No additional coils were placed in 18/23 IAs for several reasons. In 6/18 patients, we observed an improvement of IA occlusion: the initial occlusion had improved from a grade 3 to a grade 2 (n = 1) or a grade 1 (n = 2) or from a grade 2 to a grade 1 (n = 3). In 9/18 patients, the occlusion was stable (grade 2, n = 7) or worsened (grade 1 to grade 2, n = 2) and the risk of evolution or rupture after stenting was judged very low. In 3/18 patients, the occlusion was complete (grade 1) and stable and the stenting was performed to reduce the risk of IA recurrence.

Figure 1.

Ruptured Acom aneurysm. (a) Digital subtraction angiography (DSA) showing an irregular aneurysm at the left A1–A2 junction (arrow). (b) Working view DSA showing the occlusion of the dome and the bleb (arrow) after coil embolization at the acute phase. Note residual opacification of the neck (arrowhead). (c) Unsubtracted view showing complementary stenting (arrow) and coiling (arrowhead) after two weeks. (d) Working view DSA showing complete aneurysm occlusion after the second procedure.

In one patient, IV ReoPro® was administered immediately after occurrence of a clot within the origin of the side branch after placement of a 2.5/18 mm Baby Leo stent in the M1–M2 segments of the MCA. Normal cerebral circulation was recovered without additional maneuvers within 20 min with uneventful course thereafter.

In one patient, a groin hematoma occurred that only required manual compression. There were no cerebral hemorrhagic complications.

Immediate outcome

Clinical status was unchanged in all patients after the stenting procedure.

Twelve/23 patients (52.2%) were asymptomatic (mRS 0) while 11/23 presented sequelae from the previous hemorrhage resulting in a mRS 2 in 6/23 (26.1%) patients and mRS 3 in 5/23 (21.7%) patients.

Based on immediate post-procedural DSA, aneurysm occlusion was graded 1 in 12/23 (52.2%) and 2 in 11/23 (47.8%), respectively. The occlusion was graded 1 in 5/5 (100%) patients with additional coiling. In 18 patients without additional coiling, IA occlusion was graded 1 and 2 in 7/18 (39%) and 11/18 (61%), respectively. Stent tolerance was graded 0 in 23/23 patients.

Patient follow-up

At three months, the mRS was graded 0 in 19/23 (82.6%), 1 in 2/23 (8.7%), and 2 in 2/23 (8.7%) patients, respectively.

One patient treated with an Enterprise stent in the ICA presented a transient ischemic attack at day 14 because of spontaneous cessation of DAT. Imaging including CTA and brain MRI showed no stenosis or thrombosis within the stent and one hyperintense cortical spot on DWI. This patient completely recovered within minutes. Further follow-up was uneventful after re-administration of DAT according to our protocol.

Follow-up imaging was available in 22/23 (95.6%) patients because one patient who had presented IA rupture during her vacation decided to be followed-up in her country. The mean follow-up duration was 20.6 months (range 1–60 months) and the most recent imaging consisted of DSA in 13/22 (59.1%) patients and MRA in 9/22 (40.1%) patients. Six- and 12-month DSA were available in 21/22 (95.5%) and 16/22 (72.7%) patients, respectively.

One patient with a wide-necked 12 mm Acom IA treated with a Baby Leo stent without additional coiling showed a significant recurrence at six months requiring a retreatment with coils that resulted in a grade 2 occlusion that is stable at 20 months.

Taking this retreatment into account, occlusion at latest follow-up, when compared to immediate occlusion, was stable in 15/22 (68.2%), improved in 6/22 (27.3%), and worsened in 1/22 (4.5%) patients, respectively, resulting in 16/22 (73%) grade 1 occlusions and 6/22 (27%) grade 2 occlusions. The occlusion was graded 1 and 2 in 4/5 (80%) and 1/5 (20%) patients with additional coiling, respectively. In 17 patients without additional coiling, IA occlusion was graded 1 and 2 in 12/17 (71%) and 5/17 (29%), respectively. The IA occlusion immediately post-treatment, at 6-month DSA, 12-month DSA, and latest follow-up imaging is summarized in Table 2.

Table 3.

Stent tolerance.

	Delay
Stent tolerance	Immediate (n = 23)	6-month DSA (n = 21)	12-month DSA (n = 16)	Latest follow-up (n = 21)
No stenosis	95.7% (n = 22/23)	90.5% (n = 19/21)	93.7% (n = 15/16)	100% (n = 21/21)
Stenosis <50%	4.3% (n = 1/23)	9.5% (n = 2/21)	6.3% (n = 1/16)	0% (n = 0/21)
Stenosis 50–70%	0% (n = 0/23)	0% (n = 0/21)	0% (n = 0/16)	0% (n = 0/21)
Stenosis >70%	0% (n = 0/23)	0% (n = 0/21)	0% (n = 0/16)	0% (n = 0/21)
Stent occlusion	0% (n = 0/23)	0% (n = 0/21)	0% (n = 0/16)	0% (n = 0/21)

DSA: digital subtraction angiography.

Table 2.

Aneurysm occlusion

	Delay
	immediate (n=23)	6 months DSA (n=21)	12 months DSA (n=16)	Latest follow-up (n=22)
Occlusion
Complete occlusion	52.2% (n=12/23)	66.7% (n=14/21)	75% (n=12/16)	72.7% (n=16/22)
Neck remnant	47.8% (n=11/23)	28.6% (n=6/21)	25% (n=4/16)	27.3% (n=6/22)
Incomplete occlusion	0% (n=0/23)	4.8% (n=1/21)	0% (n=0/16)	0% (n=0/22)

DSA: digital subtraction angiography.

Regarding stent tolerance, only one stenosis was detected. One patient treated with a Leo (Balt, Montmorency, France) stent in the ICA developed a transient grade 1 stenosis at 6- and 12-month follow-up. The stent tolerance immediately post-treatment, at 6-month DSA, 12-month DSA, and latest follow-up imaging is summarized in Table 3.

Discussion

Our results show that a strategy of staged EVT of selected ruptured IAs may be safe, effective with a high rate of satisfying and stable occlusion at follow-up.

Population and IA characteristics

Demographical data as well as angioarchitectural characteristics of IAs of our population are similar to those of previous series of ruptured IAs¹ including the first large series reporting results of a similar strategy of early staged stenting.⁷ However, in our population, MCA and Acom IA were the most frequent locations whereas ICA aneurysms were more represented in their series. Our study protocol also introduces a selection bias regarding clinical outcome as we did not include patients with a poor neurological status within the first two weeks after SAH.

Endovascular procedures

We observed a high rate of technical success and no periprocedural clinical complications. These results are in line with the previously published series of Feng et al.⁷ reporting the experience of a staged stenting strategy with or without adjunctive coiling in 47 patients with ruptured IAs. This may be explained by the fact that stenting procedures were performed by trained interventional neuroradiologists at distance of the acute phase. Indeed, it is known that EVT at the acute phase of SAH bears a higher procedural risk.¹² Moreover, procedures were shorter as coiling was already at least partially performed. As it has been showed for EVT with flow diverter stents,¹³ one might expect that the thromboembolic risk correlates with the duration of the procedure. Also, the strategy of early staged stenting may have led to shorter procedure times for the emergency coiling. Indeed, although the aim of EVT remains basically to obtain complete occlusion of the IAs, taking this early staged stenting strategy into account, we are more prone to accept a neck remnant if we consider that the IAs are protected against rebleeding. As it has been reported in two series of staged treatments of ruptured IAs with complementary placement of conventional stents⁷ or various approaches,¹⁴ we observed no early rebleeding in our study. Brinjikji et al.¹⁵ observed one early rebleeding (3.22%) in a population of 31 patients managed with a staged treatment with flow diverter stents. For these reasons, we believe that this approach is safe and effective to prevent early rebleeding. However, our objective is always to secure as much as possible the IAs during the initial coiling.

In our experience, we plan stenting after the vasospasm period as one may fear poor stent tolerance and thromboembolic complications in this setting. However, our experience and recent literature seem to show that the placement of a stent—or a stentriever—in a cerebral artery may contribute to heal the vasospasm.^16,17 Nevertheless, when stenting is not considered as the only option to treat the IAs at the acute phase, we prefer to avoid the vasospasm period as the stent could be undersized and result in stenosis or migration after resolution of the spasm.

The antiplatelet treatment did not result in significant hemorrhagic complications as we only observed a groin hematoma that was controlled by compression without further clinical consequences. Moreover, the two-week delay between SAH and early staged stenting, allowed to avoid hemorrhagic events related to invasive procedures, made necessary by the clinical or neurological condition such as ventricular drainage or craniectomy.

We did not observe acute rebleeding after EVT neither in patients managed with the early staged stenting strategy nor in patients without complementary stenting. This may be explained by the fact that our strategy is not based on a first EVT aimed to obtain a loose packing followed by a staged stenting to improve a poor occlusion. Indeed, as we always try to obtain an occlusion as complete as possible and a packing as dense as possible at the acute phase, only a few patients have been selected for early staged stenting among all EVT for ruptured IAs in the study period (23/279, 8.2%).

Clinical outcome

The absence of procedural complications resulted in an unchanged clinical status in all patients after stenting and no permanent morbidity at follow-up. The clinical outcome—91.3% of mRS 0–1 at three months—in our series compares favorably with previously published series of ruptured IAs¹⁸ and is similar to the series of Feng et al.⁷ The main explanation is probably that waiting at least two weeks after rupture to perform stenting may have selected patients with a better prognosis. Indeed, the patients who died or who showed a neurological degradation within these two weeks did not benefit from the stenting procedure. We believe that this strategy allows to combine advantages of coiling at the acute phase of SAH on clinical outcome with those of SAC on the long-term stability of IA occlusion.

Anatomical outcome

We observed an improvement in the IA occlusion after early staged stenting. Indeed, the rate of complete occlusion rose from 30.4% before the procedure to 52.2% immediately after. Similarly, the rate of incomplete occlusion shifted from 13 to 0% after the procedure. Again, these results are in line with those presented by Feng et al.⁷ who reported complete occlusion rates rising from 14/47 (29.8%) before complementary treatment to 24/47 (51.1%) after. Although this improvement in IA occlusion is related to the additional coiling rather than the stent itself, a denser packing was made possible by the neck coverage provided by the stent.

As it has been previously reported for unruptured IAs treated by SAC,^3–5 we observed a long-term stability or improvement of IA occlusion at follow-up with only one retreatment. Besides the effect of the stent, the adjunctive coiling allowing to obtain a better initial packing may also play a role in this low recurrence rate. Indeed, the recurrence risk is related to the initial occlusion grade.^2,18–20 However, additional coiling was performed in only 5/23 IAs in our series and although the rate of grade 1 occlusion was higher in patients with additional coiling (80%, n = 4/5) than in those without additional coiling (71%, n = 12/17), it worsened in patients with additional coiling (100% after staged stenting to 80% at follow-up) while it improved in patients without additional coiling (39–71%). As the number of patients is very small, it seems difficult to draw conclusions from these data about the effect of additional coiling.

Place of early staged stenting for EVT of ruptured IAs

The use of intracranial stents for EVT of IAs has dramatically widened its indications to wide-neck and fusiform IAs and improved its long-term anatomical results.^3,5 However, SAC is not the first option to treat ruptured IAs at the acute phase. Indeed, the mandatory antiplatelet therapy may increase the risk and importance of hemorrhagic complications of invasive procedures that may be necessary within the first weeks of SAH such as ventricular drainage, hematoma clearance, or decompressive craniectomy.²¹ Moreover, in case of incomplete occlusion of the IAs, rebleeding may occur more frequently under antiplatelet therapy and be associated with a poorer outcome. On the other hand, if antiplatelet therapy is inadequate, the use of intracranial stents may be associated with a higher risk of TE complications.²²

For these reasons, in our department, patients treated for a ruptured IA by coiling at the acute phase with either an incomplete occlusion or a high risk of recurrence benefit from a complementary staged stent-assisted treatment at the subacute phase.

This work aims to evaluate this therapeutic management intended to reduce the recurrence rate of ruptured IAs treated by EVT. This strategy is chosen for only few patients in whom either IA occlusion is considered insufficient despite all efforts performed to optimize this occlusion or in whom the recurrence risk is considered high because of angioarchitectural and clinical characteristics. Overall, only 8.2% of ruptured IAs in our institution have been managed according to this strategy. As stated above, this therapeutic strategy does not modify the way we perform EVT at the acute phase. Specifically, we often perform balloon-assisted coiling for several reasons. First, it allows the temporary occlusion of the parent artery in case of procedural aneurysmal rupture. Second, it improves the stability of the microcatheter and allows to optimize the coil density. The objective of secondary stenting is not to allow or facilitate the EVT of wide-necked IAs. Stenting at the acute phase is an alternative option and might provide similar angiographic results. However, for the reasons detailed above, we only perform acute stenting when no other option is possible. Moreover, in this situation, the objective of stenting is different. It is performed to cover or reduce the neck and to allow coiling that would be otherwise impossible. Flow disruption with the Woven EndoBridge (WEB) represents an alternative strategy as it allows to treat ruptured wide-necked bifurcation IAs by EVT as DAT is not necessary.²³ However, flow disruption may be difficult or impossible to perform due to some anatomical limitations such as the angulation between the neck and the parent artery or the aneurysm shape and size. In our experience, an aneurysm remnant after flow disruption may be more difficult to occlude than after coiling. Finally, although the rerupture rate is reported to be low with the WEB device,^23,24 it remains an uncertainty regarding this point. Therefore, although the WEB device is a useful option in our armamentarium for the EVT of ruptured wide-necked bifurcation IAs, we consider coiling as a first choice at the acute phase unless the angioarchitecture does not allow to stabilize coils in the IAs and preserve the parent arteries.

Finally, although we observed no complications in this small retrospective series, the risk of a second EVT to improve or stabilize the occlusion of already at least partially occluded IAs might be questionable. Larger prospective series with long-term follow-up may help to address this point.

Limitations

Our study is first limited by its retrospective design. The single center design is also associated to a relatively small sample size and a difficulty to generalize our results to a wider practice. Finally, we have not compared the recurrence rate with ruptured IAs treated by EVT without early staged stenting during the same period. However, the interpretation of such a comparison would be difficult. Indeed, IAs in this study have been selected because of their higher risk of recurrence so that both populations would not be similar regarding angioarchitectural characteristics.

Conclusion

Early staged stenting with or without adjunctive coiling at the subacute phase of aneurysmal SAH is safe and may be proposed in selected patients. It improves immediate IA occlusion and long-term stability in this population at high risk of IA recurrence with coiling alone. Longer follow-up and larger prospective series are necessary to confirm these results.

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 retrospective study was approved by our local Ethical Committee under reference P2017/351.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Mine Benjamin https://orcid.org/0000-0002-9727-9196