Endovascular treatment of type 1 and type 4 non-saccular aneurysms of cerebral arteries – a single-Centre experience

Abstract

Aim of the study

The aim of this study was to evaluate our results regarding treatment options, complications, and outcomes in patients with non-saccular aneurysms of cerebral arteries belonging to type 1 and type 4 according to Mizutani’s classification.

Methods

A total of 26 aneurysms in 26 patients were treated between 2014 and 2019. There were 13 males (mean age 42.77 ± 11.73 years) and 13 females (mean age 50.84 ± 9.37 years). In 23 cases the onset was haemorrhagic and in three cases non-haemorrhagic. A combination of conventional stents and coils was used in 10 cases, conventional stents and flow diverters in three cases, flow diverters and coils in five cases, and flow diverters only were used in eight cases. Radiological results of treatment were assessed after eight months and clinical after one year.

Results

In 24 patients, aneurysms were occluded at the end of the follow-up period. An iatrogenic dissection and two haemorrhagic complications were registered. In three cases, parent arteries were occluded due to re-growth of the aneurysm, which caused middle cerebral artery infarction in one case. A favourable clinical outcome was registered in 19, patients, and non-favourable in five. Two patients died in the early postoperative period due to extensive damage to the brain parenchyma caused by initial bleeding.

Conclusion

Our results indicate that treatment of type 1 and type 4 non-saccular aneurysms with various combination of stents and flow diverters, with or without coils, is promising, although very challenging and technically demanding.

Introduction

Non-saccular aneurysms are a large group of aneurysms which, according to current classifications, comprises the following types of aneurysms: blood blister aneurysms (BBA), dissecting aneurysms (DA), fusiform aneurysms (FA), and pseudoaneurysms (PA).^1–4 All those aneurysms share several common features which distinguish them from saccular aneurysms.

Saccular aneurysms are always localised on the branching site of an artery. The pathologic substrate of a saccular aneurysm is a localised defect of the internal elastic lamina (IEL), which is followed by more profound changes in the arterial wall, protrusion of other layers of the arterial wall through this defect, and formation of a saccular aneurysm. ⁵ Unlike a localised defect of the wall of an artery bearing a saccular aneurysm, the defects of the walls of arteries bearing non-saccular aneurysms are much more complex and involve larger segments of the artery. ⁶ Non-saccular aneurysms display a very wide variety of forms, from a barely visible irregularity of the arterial wall, which is typical for BBAs (Figure 1), to a huge, bizarrely shaped dilatation of the lumen which involves an artery and two or more of its branches, which is typical for DAs (Figures 2 and 3).

Figure 1.

The figure shows patient 1 (Table 2): right-sided digital subtraction angiography (DSA) with (a) early arterial phase lateral projection and (b) antero-posterior (AP) projection (b) showing irregular narrowing of the supra-ophthalmic segment of internal carotid artery (ICA) and hypoplasia of the right anterior cerebral artery (ACA) (see arrows). The patient was first treated with two Silk stents, but due to rebleeding three days later, two more stents were deployed; (c) AP plain radiograph of the skull after the second treatment shows the flow diverting stents deployed from cavernous segment of right internal carotid artery to right middle cerebral artery (MCA); (d) Follow-up with DSA shows reconstructed lumen of right ICA and flow in right MCA.

Figure 2.

Patient 4 (Table 2): (a) T2 axial shows an oval, extra-axial structure with inhomogeneous signal intensity occupying ambient cistern and compressing left superior colliculus. Conventional angiography (b). Left vertebral subtracted angiography, AP projection (c). 3D reconstruction, lateral view) revealed an irregular aneurysm of left posterior cerebral artery which involves the calcarine artery and parietooccipital artery; (d) spot image, AP projection, status after catheterization of aneurysm and “Y” stenting: arrow head indicates tip of the catheter in the aneurysm, arrow indicates proximal markers of two Atlas stents in posterior cerebral artery, dotted arrows indicate distal markers of the stents in the calcarine artery and parietooccipital artery; last run ((e) left vertebral subtracted angiography, AP projection) showed occluded aneurysm with minor neck remnant and patent posterior cerebral artery and its branches; (f) one-month follow-up, left vertebral subtracted angiography, AP projection shows regrowth of aneurysm and slow flow in the branches of the posterior cerebral artery.

Figure 3.

Patient 4 (Table 2): (a) spot image, lateral projection: test occlusion of the left posterior cerebral artery; arrow indicates 7 × 7 Hyperform balloon inflated in the posterior cerebral artery proximal to aneurysm; (b) vertebral angiogram, AP projection, subtracted showing occlusion of the posterior cerebral artery; (c) left carotid angiogram, AP projection, subtracted shows retrograde filling of the left posterior cerebral artery; (d) MR follow-up, T2 axial shows no infarctions and an occluded aneurysm.

Endovascular obliteration with coils is today the preferred treatment of ruptured saccular aneurysms, but non-saccular aneurysms are impossible to treat with coils only. That is why these aneurysms are today mostly treated with conventional stents and flow diverters (FDs), with or without coils.^7,8

A concern regarding the use of stents in the treatment of ruptured aneurysms is the need for dual antiplatelet therapy (DAPT) in haemorrhagic cases, but the increasing literature on the subject shows promising results regarding safety.^9–12

In 2014, we began treating patients with non-saccular aneurysms with various types of stents. In this retrospective study, we present our experiences from the first 26 patients with non-saccular aneurysms treated with stents only or with stents and coils. The aim of this study was to evaluate the applied techniques, procedure-related complications, clinical outcomes, radiological results, and the significance of Mizutani’s classification of non-saccular aneurysms.

Material and methods

Between 2014 and 2019, non-saccular aneurysms in 26 consecutive patients were treated with flow-diverting and conventional stents, with or without coils. Twenty-three cases were haemorrhagic and three cases were non-haemorrhagic. In case 4 (21 y/m) computerized tomography, performed as a part of the investigation of intermittent headache with slowly progressing intensity, revealed a well-delineated, oval structure localized close to left tentorial edge. Additional imaging (magnetic resonance and conventional angiography) revealed an irregular aneurysm involving the left distal posterior cerebral artery, calcarine artery, and temporo-occipital artery (Figure 2). A 51-year-old female (case 6) presented with acute onset dysphasia. Work-up, computerized tomography and conventional angiography, revealed a dissecting aneurysm of the left middle cerebral artery. Case 16 (50 y/f) presented with repeated, short (up to 15 minutes) episodes of numbness in the left half of the body. Angiography revealed an irregular, fusiform dilatation of the P2 segment of the right posterior cerebral artery.

Clinical data collected for the study were age, sex, and clinical condition at admission and on discharge according to the Reaction Level Scale (RLS). ¹³ Information about the aneurysm, the intervention procedure, and complications in the acute phase were collected for all patients. The acute phase was defined as the time when the patient still was being treated in the neurosurgical department after the bleeding.

We used Mizutani’s classification of non-saccular, non-arteriosclerotic aneurysms and classified aneurysms according to the anatomy of the parent artery, form of the aneurysm, and clinical presentation. ⁶ This classification was chosen because it is based on the pathological substrate directly related to the specific macroscopic appearance and clinical symptoms of an aneurysm. The most important limitation of this classification is that not all aneurysms in his study were histologically examined. According to this classification, type 1 dissecting aneurysms are fusiform in shape and histologically are characterised by widespread disruption of IEL without intimal thickening. They are clinically manifested by haemorrhage or infarction. Aneurysms type 4 occur on the non-branching segment of an artery and have the shape of a shallow bulge without identifiable neck. The wall of this aneurysm consists only of fragile adventitia and connective tissue. Clinically, they are always associated with haemorrhage. ⁶ This macroscopic appearance and clinical picture correspond to the description of a BBA. The inclusion criteria are shown in the Table 1.

Table 1.

Inclusion criteria.

1. Localization: intracranial portions of cranial arteries

2. Origin: on the non-branching segment of an artery

3. False lumen: no

4. Intraluminal thrombotic masses: no

5. Unclear transition between diseased and normal arterial wall

6. Fast growth if not treated.

7. Aggressive clinical picture requiring acute treatment

Radiological follow-up

Follow-up digital subtraction angiography (DSA) or, in some cases, when the patients were reluctant to undergo DSA, magnetic resonance angiography (MRA) was done after eight months according to the clinical protocol (Tables 2 and 3). On follow-up DSA/MRA, the degree of reconstruction of lumen vessel and the degree of occlusion of the saccular component of aneurysms, according to the Raymond–Roy classification, ¹⁴ were assessed.

Table 2.

Patients who underwent two or three interventions.

Nr/sex/age	Onset	RLS on admission	Aneurysm/s type, location	First intervention/device	Second/third intervention/device	Complication or technical difficulty	RLS at discharge	Antiplatelet regime	GOSE/on FU	Radiological result on FU
1/f/50	H	3a	Type 4, ICA	Stenting of ICA: Silk – 3 × 25, 3 × 20 (TD)	Indication: re-bleeding after first intervention <1 week Intervention: additional stenting: Silk × 2 (2.5 × 20), TD	None	1	Clopidogrel, ASA	3	Conv. angiography: Reconstructed lumen of ICA
2/f/65	H	1	1.Type 4 with saccular component, ICA, 2. Saccular aneurysm, P Comm A	Stenting of ICA and embolization of saccular component: Silk (3.5 × 20) + coils	Embolization of P Comm A: coils	None	1	Prasugrel, ASA	7	Conv. angiography: Occluded aneurysms (RR – 1) and reconstructed lumen of ICA
3/m/47	H	4	1. Type 1, A2, 2. Saccular aneurysm, A Comm A	Stenting of A2 and embolization of cavity: Baby Leo 2 × 25 and coils	Embolization of A Comm A aneurysm: coils	None	3b	Prasugrel, ASA	3	Conv. angiography: Occluded aneurysms (RR – 1) and reconstructed lumen of A2
4/m/21	NH	1	Type 1, PCA	Stenting of PCA end embolization of cavity: Atlas 3 × 24, 3 × 21, "Y" stenting, + coils	Indication: regrowth of aneurysms < 4 monthsIntervention: occlusion of PCA	None	1	Ticagrelor, ASA (discontinued after occlusion of PCA)	8	MR/MRA - occluded aneurysm and parent artery, no infarctions
5/m/45	H	1	Type 4 with saccular component, ICA	Stenting of ICA: Silk: 3.5 × 25, 3.5 × 20, TD	Indication: re-growth after 1st intervention < 1 week: Intervention: additional stenting: Silk: 3.5 × 20, 4 × 15, TD	None	1	Ticagrelor, ASA	6	Conv. angiography: Occluded aneurysm (RR – 1) Reconstructed lumen of ICA
6/f/51	NH	1	Type 1, MCA	Stenting of M2 and embolization of cavity: Atlas (3 × 24) + coils	2nd intervention, re-coiling due to regrowth < 4 months3rd intervention – coil occlusion of MCA due to re-growth < 8 months	None	1	Ticagrelor, ASA (ASA discontinued after occlusion of MCA)	5	MR/MRA: Occluded aneurysm, occluded parent artery, no infarctions
7/f/56	H	3	Type 1, MCA	Stenting of M2 and embolization of cavity: Atlas (3 × 15) + coils	Indication: regrowth and re-bleeding < 1 month. Intervention: coil occlusion of MCA and evacuation of hematoma	None	3	Ticagrelor, ASA (discontinued before craniotomy)	No FU	No FU
8/f/44	H	1	Type 4, ICA	Stenting of ICA: Silk: 3 × 25	Indication: changed form of aneurysm < 2 weeks: Intervention: additional stenting: FRED: 4 × 23, TD	None	1	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of P1/2
9/f/59	H	7	Type 4, ICA	Stenting of ICA: Silk: 3 × 25	Indication: changed form of aneurysm < 2 weeks: Intervention: additional stenting: Silk 3.5 × 20, TD	Failed detachment of Leo stent – successfully withdrawn	2	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of ICA

RLS: Reaction Level Scale; TD: telescopic deployment; H: haemorrhagic; NH: non-haemorrhagic; ICA: internal carotid artery; MCA: middle cerebral artery; PCA: posterior cerebral artery; P Comm A: posterior communicating artery; A Comm A: anterior communicating artery; FU: follow-up.

Table 3.

Patients who underwent one intervention.

Nr/sex/age	Onset	RLS on admission	Aneurysm type/location	Intervention/device	Complication or technical difficulty	RLS at discharge	Antiplatelet regime	GOSE on FU	Radiological result on FU
10/m/46	H	1	Type 1, VA	Stenting of VA: Silk (2 × 20)	None	1	Clopidogrel, ASA	7	Conv. angiography: Reconstructed lumen of VA
11/f/59	H	3a	Type 1, VA	Stenting of VA: Silk (4 × 20)	None	3	Clopidogrel, ASA	3	Conv. angiography: Reconstructed lumen of VA
12/f/41	H	3 b	Type 1, VA	Stenting of VA Silk × 2 (2.5 × 25 m 3.5 × 25), TD	None	FO	Clopidogrel, ASA	FO	No FU
13/f/62	H	3a	Type 4 with saccular component, ICA	Stenting of ICA and embolization of saccular component: Leo: 3.5 × 18, 3.5 × 25 (TD) + coils	None	FO	Prasugrel, ASA	FO	No FU
14/m/48	H	1	Type 1, PCA	Stenting of PCA: Silk (2 × 20) + 2 Baby Leo (2 × 25 + 2 × 18)/TD and stabilisation of foreshortened stent	Migration of proximal end of Silk due to foreshortening. System stabilized with 2 Baby Leo stents	1	Prasugrel, ASA	6	Conv. angiography: Reconstructed lumen of PCA
15/m/59	H	5	Type 1, PCA	Stenting of PCA: Silk (2 × 15) + Baby Leo (2 × 18), TD	None	2	Prasugrel, ASA	4	Conv. angiography: Reconstructed lumen of PCA
16/f/5023	NH	1	Type 1, PCA	Stenting of PCA and embolization of cavity: Enterprise: 4.5 × 37 + coils	Baby Leo (2.5 × 25) migrated into the aneurysm after detachment due to foreshortening – left in the aneurysm	1	Prasugrel, ASA	7	6 months conv-ang. – occluded aneurysm and reduced flow in PCA. 10 months conv. ang. occluded aneurysm occluded PCA, collaterals to PCA territory, no infarction on MR
17/f/45	H	1	Type 4 with saccular component, ICA	Stenting of ICA and embolization of saccular component: Silk: 3.5 × 20 + coils	Rupture of aneurysm after detachment of last coil. Bleeding stopped by balloon inflation	2	Prasugrel, ASA	6	Conv. angiography: Occluded aneurysm(RR – 1) Reconstructed lumen of ICA and occluded aneurysm (RR – 1)
18/m/50	H	3	Type 1, PICA	Stenting of PICA and embolization of saccular component: Baby Leo: 2 × 18 + coils	None	2	Prasugrel, ASA	6	Conv. angiography: Reconstructed lumen of PICA and occluded aneurysm (RR – 1)
19/m/29	H	1	Type 1, temporal branch of MCA	Stenting of temporal branch of MCA and embolization of cavity: Atlas: 3 × 24 + coils	Failed detachment of Baby Leo, 2.5 × 25. Left. Increased hematoma, craniotomy and evacuation of hematoma	1	Prasugrel, ASA	6	Conv. angiography: Occluded aneurysm and parent artery. No infarction
20/m/45	H	1	Type 1, VA	Stenting of VA and embolization of cavity: Aclino (4 × 20 HRF) + coils	None	1	Ticagrelor, ASA	5	MR angiography: Reconstructed lumen of VA and occluded aneurysm (RR – 1)
21/f/50	H	5	Type 4, ICA	Stenting of ICA: FRED (3.5 × 24) + Lvis J (3.5 × 33, 3.5 × 28) + Atlas (4 × 21), TD	None	2	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of ICA
22/f/29	H	1	Type 1, PCA	Stenting of PCA and embolization of cavity: Baby Leo stent (2.5 × 25) + coils	None	1	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of PCA and occluded aneurysm (RR – 1)
23/m/58	H	4	Type 4 with saccular component, ICA	Stenting of ICA and embolization of saccular component: Surpass (3 × 25) + coils	None	2	Ticagrelor, ASA	5	Conv. angiography: Occluded aneurysm(RR – 1) and reconstructed lumen of ICA
24/m/45	H7	1	Type 1, VA	Stenting of VA: Silk: 2.5 × 25, 2.5 × 20 V4, TD	Iatrogenic dissection of V2 – solved by Atlas (4.5 × 30, 4.5 × 21) and Leo stent (4.5 × 30)	1	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of VA
25/m/24	H	2	Type 1, P1/2	Stenting of P1/2 and embolization of cavity: Silk (2.5 × 15) + coils	None	1	Ticagrelor, ASA	7	Conv. angiography: Reconstructed lumen of P1/2 and occluded aneurysm (RR – 1)
26/m/30	H	1	Type 1, P1/2	Stenting of P1/2 and embolization of cavity: Silk (2 × 15) + coils	None	1	Ticagrelor, ASA	7	MR angiography: Reconstructed lumen of P1/2 and occluded aneurysm (RR – 1)

RLS: Reaction Level Scale; TD: telescopic deployment; H: haemorrhagic; NH: non-haemorrhagic; ICA: internal carotid artery; MCA: middle cerebral artery; PCA: posterior cerebral artery; P Comm A: posterior communicating artery; A Comm A: anterior communicating artery; FO: fatal outcome; FU: follow-up.

Clinical follow-up

Clinical functional outcome according to the Glasgow Outcome Scale Extended (GOSE) score was assessed after one year. However, due to clinical circumstances, the follow-up varied regarding time point. The functional outcome in six patients (8, 9, 16, 21, 24 and 26) was assessed at the time of angiographic follow-up (i.e. eight months after the treatment). The clinical condition of these patients did not worsen after this examination. A GOSE score equal to or larger than 5 was defined as a favourable functional outcome.

These follow-up protocols were applied in all cases regardless of onset of disease (haemorrhagic and non-haemorrhagic) and treatment option.

Neurointerventional procedures

The treatment modality of the aneurysm was decided by neurosurgeons and neurointerventionalists together, with the perspective of offering the safest alternative. Demographic data and technical details of all interventions and follow-up of all patient are shown on Table 2 (patients who underwent two or three interventions) and Table 3 (patients who underwent one intervention).

The combination of various devices was determined by the complexity of angioarchitecture and the form, size, and localization of aneurysm in each particular case. Low profile stents combined with coil occlusion of aneurysms were used when a satisfactory radiological and clinical result of this combination of devices was anticipated. FDs were used when flow diversion was assessed as being the most important element of the treatment. They were combined with coils in some cases to obliterate the cavity of an aneurysm. Low profile stents were used in combination with FDs to stabilize or correct the position of a FD and/or to augment the flow-diverting effect.

A combination of conventional stents and coils was used in 10 cases. In case 4, two Atlas stents were deployed in ‘Y’ fashion (Figure 2). Conventional stents and FDs were used in three cases, FDs and coils in five cases, and FDs only in eight cases. Coils were used for embolization of saccular components of type 4 aneurysms and for embolization of fusiform or irregular cavities of type 1 aneurysms (Tables 2 and 3).

In two cases concomitant saccular aneurysm were embolized in separate session (cases 2 and 3). In four cases, FD stents were deployed in two interventions (cases 1, 5, 8, and 9). The first stent/s was/were deployed immediately after the rupture of the aneurysm and the second one/s shortly after the first intervention due to re-rupture, re-growth, or change of form of the aneurysm.

In two cases, parent vessels were occluded at the second or third session due to re-growth of aneurysm despite successful stenting and coiling (cases 4 and 6). Occlusion of parent artery did not cause ischaemic lesions in either case (Figures 2 and 3).

The neurointerventional devices and methods used depended on the nature of the aneurysm and the preferences of the neurointerventionalist. The following CS and FD were used: Baby Leo, Leo, Silk (FD) (BALT EXTRUSION SAS 10 rue Croix Vigneron 95,160 Montmorency, France), Enterprise (Codman, 1302 Wrights Lane East West Chester, PA 19,380 USA), Neuroform Atlas and Surpass (FD) (Stryker, 47,900 Bayside Parkway, Fremont, CA, USA), FRED, LVIS Jr. (Microvention, Aliso Viejo, CA, USA), and Aclino HRF (Acandis, Theodor-Fahrner-Straße 6,75,177 Pforzheim, Germany).

Dual antiplatelet regime (DAPR) and anticoagulation

The three patients with non-haemorrhagic onset received a loading dose of antiplatelet drugs a day prior to the intervention consisting of 300 mg aspirin and 180 mg ticagrelor (Astra Zeneca AB. SE-151 85 Södertälje, Sweden). The patients with haemorrhagic onset received abciximab (0.25 mg per kilogram bodyweight) intravenously immediately before deployment of the stent. All patients received an intravenous bolus of 5000 IU of heparin after placement of a guiding catheter in the cervical artery. The heparin was added during the intervention to maintain the activated clotting time between 2 and 2.5 times compared with the baseline value. The post-procedural antiplatelet regime consisted of aspirin, 75 mg daily, continued for 12 months or longer. Along with aspirin, clopidogrel (Actavis Pharma Inc, 400 Interpace Parkway, Parsippany, NJ, USA) 75 mg daily, or prasugrel (Eli Lilly Sweden AB, Solna, Sweden), 10 mg daily, or ticagrelor 180 mg daily, divided in two doses, were prescribed for six months.

During the study period (five years), we updated our DAP regime to achieve the safest and, at the same time, the most effective antiplatelet effect. Clopidogrel was used at first, but due to unreliable therapeutic effects, it was replaced with prasugrel. ¹⁵ Prasugrel is a very potent inhibitor of thrombocyte aggregation, and haemorrhagic complications caused by this drug are well documented. ¹⁶ Although we have never had serious haemorrhagic complications related to the use of prasugrel, we decided to replaced it with ticagrelor. Ticagrelor is an antiplatelet drug with significant advantages compared with thienopyridines (clopidogrel or prasugrel). The pharmacokinetic properties of this agent are predictable, rapid onset (30 min) and rapid offset (3–5 days). Additionally, unlike clopidogrel or prasugrel, this drug does not require metabolic transformation, and receptor blockade generated by this drug is reversible. ¹⁷

Ethical issues

We confirm that the study was conducted in accordance with Helsinki Declaration as revised in 2013.

Results

Angiographic results

Angiographic results and clinical conditions at the end of follow-up period are summarised in Tables 2 and 3. The explanation of the abbreviations below Table 2 also applies to the Table 3.

In 19 cases, parent arteries were reconstructed, and aneurysms in 24 patients were closed on follow-up angiography.

Occlusion of parent arteries performed at a second or third session due to re-growth of aneurysms despite primary treatment with stent and coils did not cause ischaemic lesions (cases 4 and 6).

Spontaneous occlusion of a parent artery, which occurred in two cases during the follow-up period, was asymptomatic owing to well-developed collaterals that supplied the territories of the occluded arteries (cases 16 and 19).

Clinical results

Clinical results at hospital discharge

The clinical picture in 13 patients corresponding to RLS 1 and in one patient corresponding to RLS 3 was unchanged at discharge. Clinical improvement in eight patients and worsening of symptoms in two patients were registered during hospitalisation.

Clinical results at end of follow-up

The clinical condition of the patient with a type 1 aneurysm of the left middle cerebral artery, which was occluded in the acute phase due to fast re-growth of the aneurysm, was poor, and the GOSE score was not determined (case 7). Nineteen patients had a GOSE score equal to or more than 5, which was considered a favourable outcome, and in four patients the GOSE was 4 or less which was considered non-favourable outcome. In other words, the favourable outcome was registered in 79% of patients who survived.

Two patients (cases 12 and 13) died due to diffuse parenchymal brain damage caused by the initial subarachnoid bleeding.

Complications/technical difficulties

In three cases (11.5%), we registered complications (two haemorrhagic complications and one iatrogenic dissection) and in the same number of cases technical difficulties caused by challenging anatomy of aneurysms and parent arteries. A detailed description of these cases is given below.

Complications

Rupture of the aneurysm in case 17 occurred after detachment of the last coil. Bleeding was stopped by inflation of the remodelling balloon. In the case 19, because of failed detachment, the stent (Baby Leo) was withdrawn from the type 1 aneurysm of the temporal branch of the middle cerebral artery, which led to additional dissection of this artery and re-bleeding. Another stent was deployed, and a fusiform lumen of the dissection embolised with coils. The haematoma was evacuated. The follow-up angiography showed stabile occlusion of the aneurysm and occlusion of the parent artery, which occurred spontaneously during the postoperative course. The patient recovered during the follow-up period (GOSE - 6) due to a rich network of collaterals supplying the territory of the dissected artery. In the case 24, iatrogenic dissection of the V2 segment was solved by deployment of three low profile stents.

Technical difficulties

In case 9, a Leo stent was successfully retrieved after failed detachment.

In case 14, the proximal end of a FD Silk migrated into the aneurysm due to foreshortening of the stent after detachment. The stent was repositioned with two Baby Leo stents, which stabilised the stent construct and had an augmented flow-diverting effect. This led to reconstruction of the artery and collapse of the aneurysm (Figure 5).

Figure 5.

Patient 14 (Table 3): (a) spot image, lateral projection: status after catheterisation of the posterior cerebral artery through the stent. Spots correspond to markers of the microcatheter; (b) spot image, anteroposterior projection: status after stabilisation of Silk FD stent with two Baby Leo stents. Telescopic deployment; (c) vertebral angiography, anteroposterior projection, subtracted: status after stenting – reduced, but still present flow in the aneurysm; (d) DSA follow-up; vertebral angiography, anteroposterior projection, subtracted: reconstructed lumen of the left posterior cerebral artery and obliterated aneurysm.

In case 16, the stent deployed from P2 to P4 migrated into the lumen of the aneurysm due to foreshortening. Another stent was then successfully deployed, and the aneurysm embolised with coils. The last postoperative angiography showed an occluded aneurysm and preserved flow in the posterior cerebral artery. Follow-up angiography revealed that the posterior cerebral artery distal to the aneurysm became occluded during the follow-up period. This occlusion did not cause any infarction because the collaterals effectively supplied the territory of the occluded posterior cerebral artery.

Two complications and two technical difficulties occurred in Type 1 group and one complication and one technical difficulty in Type 4 group.

Discussion

Importance of classification of non-saccular aneurysms

The current classifications of non-saccular aneurysms are inconsistent and vary from source to source.^1–4 According to these classifications, all non-saccular aneurysms are classified into dissecting, fusiform, and pseudoaneurysms or false aneurysms. Terms which are also widely used are blood blister-like aneurysm, and transitional aneurysms. These classifications are confusing and misleading since the term ‘‘dissecting’’ is related to pathological changes in the vessel wall, the terms ‘fusiform’ and ‘blood blister-like’ are related to the shape of the aneurysm, ‘transitional’ to localisation, while the terms ‘pseudoaneurysm’ or ‘false’ aneurysm indicate that these aneurysms differ from the ‘true’ ones. DAs and BBAs share numerous common features, and according to several authors, the BBAs represent a sub-type of DAs.^6,18–20

• The form of BBAs varies from broad, shallow bulging without identifiable neck to a barely visible wall irregularity (Figure 1). In some cases, they even have a lobular component, which can cause diagnostic confusion (Figures 5 and 6). ²¹ In these cases, they resemble multilobulated dissecting aneurysms (compare Figure 4 with Figure 7).

• Unlike regular saccular aneurysms, both types of these non-saccular aneurysms are localised on the non-branching segment of an artery.^19,20,22,23

• The shape and size of these aneurysms change rapidly over short periods of time. They tend to grow and involve larger segments of the parent artery if only partially treated.^19,20,22,23

Figure 6.

Patient 23 (Table 3): (a) right carotid angiography, lateral projection, non-subtracted: blood blister-like aneurysm (type 4), note irregular lumen of internal carotid artery (arrow) with broad based saccular component; (b) spot image, lateral projection shows status after deployment of Surpass 3 × 25 FD stent and catheterisation of the aneurysm – spot corresponds to the tip of microcatheter (arrow); (c) right internal carotid angiography, anteroposterior projection, subtracted shows stagnation of the flow in the aneurysm; (d) spot image, anteroposterior projection shows status after stenting of the internal carotid artery and coiling of the aneurysm; (e) right internal carotid angiography, anteroposterior projection, last run shows occluded aneurysm and normal flow in the internal carotid artery; (f) right internal carotid angiography, anteroposterior projection, DSA follow-up: occluded aneurysm and reconstructed lumen of the internal carotid artery.

Figure 4.

Patient 14 (Table 3): (a) 3D reconstruction shows anterior aspect of a multilobulated dissecting aneurysm (type 1) of the left posterior cerebral artery; (b) vertebral angiogram, lateral projection, nonsubtracted shows bizarre form of the aneurysm; (c) spot image, lateral projection, shows status after deployment of Silk FD stent; (d) spot image, lateral projection shows foreshortening of the stent after detachment.

Figure 7.

Patient 5 (Table 2): (a) right internal carotid angiography, AP projection, subtracted: blood blister-like aneurysm (type 4) of internal carotid artery with multilobular saccular component; (b) spot image, AP projection shows status after deployment of two Silk FD stents in telescopic fashion; (c) right internal carotid angiography, AP projection, subtracted, follow-up < 1 week showing re-growth of the aneurysm; (d) spot image, AP projection shows status after deployment of additional two Silk FD stents in telescopic fashion, arrow indicates stagnation of the contrast in the aneurysm; (e) right internal carotid angiography, AP projection, subtracted, late phase shows stagnation of the flow in the saccular component of the aneurysm; (f) DSA follow-up. Right internal carotid angiography, AP projection, subtracted shows reconstructed internal carotid artery. Note the flow in A1 and lateral lenticulostriate arteries. No flow in the posterior communicating artery.

In many recently published papers on the treatment of non-saccular aneurysms, the authors have used the current classifications which are based on macroscopic appearance and localisation of these aneurysms. Anatomical forms, onset, course of the disease, and outcome vary greatly among non-saccular aneurysms. That is why the majority of these publications are very comprehensive, with large amounts of data and not sufficiently clear and informative conclusions. Kiyofuji et al. analysed 13 studies in their meta-analysis of treatment outcomes of posterior circulation non-saccular aneurysms treated with FDs. Since the authors of these studies published results of treatment of all morphological types of non-saccular aneurysms, the only possible conclusion was consequently: ‘Treatment outcomes of non-saccular aneurysms of the posterior circulation are highly variable and are highly dependent on patient selection’. ²⁴ Boghal et al. classified the non-saccular aneurysms they had treated as fusiform, dolichoectatic, and transitional, neglecting the fact that those aneurysms belong to different histological types and therefore required different therapeutic approaches, which was, however, specified in the conclusion. ⁴ Shapiro et al. in a similarly designed literature review analysed the clinical course, treatment, and outcomes of non-saccular aneurysms and dolichoectasias, although the relation between dolichoectasias and non-saccular aneurysms is very unclear. ²⁵ Nasr et al. ²⁶ in their metanalysis analysed 14 studies with a total of 827 patients. Although the study was very comprehensive, the authors concluded that ‘Further studies on the natural history of vertebrobasilar dolichoectatic and fusiform aneurysms with more complete follow-up are needed to better understand the risk factors for progression of these aneurysms.’ These cited studies are just a few of many similarly designed studies on clinical aspects of non-saccular aneurysms. The most important disadvantage of all these studies is the lack of a coherent morphological classification based on histological changes in the wall of diseased artery.

Treatment strategy

The treatment of complex, especially distal, aneurysms with stents and coils entails very serious risks. Decisions regarding treatment strategy were taken by a multidisciplinary team after meticulous analysis of vascular anatomy, morphology of aneurysms, and all treatment options. The patients and/or their relatives were fully informed about the risks of each therapeutic option, and the final decision on the intervention was taken after the facts related to the intervention had been understood and accepted by patients and/or their relatives.

Clinical course and the principles of the endovascular treatment of type 1 and type 4 aneurysms are also very similar. In our sample, 23 cases were haemorrhagic. In the remaining three cases, the neurological deficits were caused by the fast-growing aneurysms, and the rupture was prevented by prompt diagnosis and successful treatment.

The main aims of the treatment were reconstruction of the lumen of the parent artery, prevention of re-bleeding, and protection of the ostia of the side/terminal branches. We have achieved these three aims by deployment of stents or FDs in the diseased segment of the artery and, in some cases, by filling the aneurysmal cavity with coils. Mesh density and, consequently flow-diverting ability, can be augmented by telescopic deployment of stents. ²⁷ We have used this feature of FDs in 6 cases in which multiple FDs created flow diversion enough for reconstruction of the lumen and obliteration of aneurysmal cavity. In two cases the same result was achieved with singe FDs. Single or multiple FDs which were deployed in telescopic fashion covered side branches proximally and distally to circle of Willis: lenticulostriate arteries, ophthalmic artery, anterior choroidal artery, anterior and posterior communicating arteries as well as perforators originating from basilar artery and posterior cerebral artery. Despite very high mesh density in the stented vessels, we have not registered any ischaemic lesions in the territories supplied by these arteries, which is in line with previously published data (Figure 7). ²⁸

Parent artery occlusion is a method of choice in the treatment of non-saccular aneurysms in which any other therapeutic alternative is excluded because of possible severe complications or anticipated technical difficulties. ²⁹ Occlusion of the parent artery is technically very feasible, but sudden flow arrest in a large artery may cause disastrous ischaemic lesions. That is why the balloon test occlusion plays the key role in mapping of the collateral circulation before a decision on parent artery occlusion has been made. ³⁰ We have performed the balloon test occlusion in only one case (Case 4, Figure 3(a)). Since the patient tolerated the test, the posterior cerebral artery was successfully occluded in the same session without any neurological deficits occurring (Case 4, Figure 3(b) to (d)).

Technical difficulties and complications

The anatomy of type 1 and type 4 dissecting aneurysms posts a serious challenge for the operator. According to several authors, the extremely fragile walls of these aneurysms and parent vessels carry a risk of per-operative rupture, which additionally complicates the surgery.^31,32 Intracranial stents are very complex devices from a technical point of view. Despite remarkable improvements in stent technology during recent years, complications occur during deployment or detachment of stents. Per-operative in-stent-thrombosis after deployment of FDs is well known complication of stenting with FDs.^7,33 Unclear transition between diseased and normal segments of the parent artery and a very wide communication between aneurysmal cavity and lumen of the artery make choice of stent difficult and deployment of the stent demanding and very tricky which sometimes results in the migration of a stent. ³³ The partial migration of the stent in one case and the complete migration of the stent into the aneurysmal cavity in another were successfully solved using sophisticated endovascular techniques (see results). More complications and technical difficulties occurred in the Type 1 than in the Type 4 group. But, total number of complications and technical difficulties is too small to draw any meaningful conclusion.

Clinical outcomes

We assessed clinical results as favourable in 19 (79%) and non-favourable in 5 (21%) patients. No perioperative deaths were registered. The clinical outcome was favourable even in the cases in which complications or technical difficulties have occurred. This can be explained by the use of appropriate technical solutions to these complications and difficulties. All patients with non-favourable outcomes belonged to the group of patients with haemorrhagic onset who were in poor clinical condition (RLS ≥3) at the admission. In a multicentre study on endovascularly treated blood blister-like aneurysms published by Fang et al., ³² the percent of patients with favourable clinical outcome was 91.1% while the percent of perioperative deaths was 4.7%. Kim et al. ³⁴ registered favourable outcome (modified Ranking score ≤2) in 82.4% of patients with blood blister-like aneurysm which were treated endovascularly. In a group of patients with dissecting aneurysms of middle cerebral artery treated by endovascular means, Zhao et al. ³⁵ registered favourable outcome in 93% cases.

Several authors have pointed out the technical difficulties and complexity of open surgery in the treatment of non-saccular aneurysm and the advantages of endovascular techniques in comparison with open surgery. Ren et al. ³⁶ reported 54.5% favourable outcome after surgical treatment and 76% favourable outcome after endovascular treatment of blood blister-like aneurysms. Balik et al. ³⁷ concluded in their review of surgical treatments of dissecting posterior circulation aneurysms that endovascular treatment is the first choice of treatment of these aneurysms because they are unstable and tend to rebleed. However, they claim that certain subsets of these aneurysms are not amenable to the endovascular approach but still require surgical strategy. Gonzales et al. presented their results of the treatment of blood blister-like aneurysms in 12 patients (9 endovascularly, 2 surgically and 1 conservatively) and reviewed 314 patients from the literature. They concluded that endovascular treatment offers lower morbidity and mortality compared with surgical approaches. ³¹

The nature of non-saccular aneurysm, their tendency to re-growth, and/or to re-bleed regardless of chosen treatment option is the factor which complicates the treatment and has a negative impact on the course and outcome of this disease. ³¹

Conclusion

There are many surgical and endovascular options regarding treatment of non-saccular aneurysms. Published data clearly indicate that endovascular treatment of this type of aneurysms is superior compared with open surgery. Although the group of patients that we have treated is relatively small, the angiographic results of endovascular treatment of type 1 and type 4 non-saccular aneurysms and the clinical outcomes were satisfactory, especially considering the fact that the majority of cases were challenging from the technical point of view.

Unlike the authors who have used current morphological classifications in their studies, we used Mizutani’s classification, which is based on the relation between histological changes, gross morphology of non-saccular aneurysms, and the clinical picture. Although this classification is relatively old and not perfect, it helped us to do a proper selection of patients with aneurysms which demanded similar treatment strategies. Further research on the relation between pathological changes in the walls of non-saccular aneurysms and their gross appearance and new classifications based on the results of that research would have essential importance for future clinical studies on the treatment of non-saccular aneurysms.