Endovascular treatment of middle cerebral artery aneurysms: A single center experience with a focus on thromboembolic complications

Abstract

Background

Endovascular treatment of middle cerebral artery (MCA) aneurysms has traditionally been considered difficult due to complex branching patterns, frequent involvement of parent vessels within the aneurysm neck, and a high incidence of thromboembolic complications.

Methods

The cases of 93 MCA aneurysms treated with endovascular intervention at our institution between 2003 and 2015 were retrospectively reviewed. Demographic, clinical, and radiographic variables were recorded and analyzed.

Results

Immediate complete or near-complete occlusion was achieved in 81 (90%) cases. At the longest follow up of 6 months or greater, 83.3% of the aneurysms were stable-to-improved in the Raymond occlusion classification, 8.3% were found to have minor recanalization not requiring retreatment, and 8.3% required retreatment due to significant recanalization.

Thromboembolic events occurred in 18 (19.3%) of cases, but only 1 resulted in permanent vessel occlusion and only 1 resulted in permanent neurologic impairment. Thrombus was resolved with intra-arterial thrombolysis or mechanical thrombectomy in 17 of the 18 cases. There were only two cases resulting in morbidity (2.15%).

There was no statistically significant correlation between aneurysm location, size, morphology, or use of adjuvant device with radiographic outcome or thromboembolic event.

Conclusions

While the rate of thromboembolic events in our series was 19%, the overall morbidity was only 2%. This highlights the fact that even complex MCA aneurysms can safely and effectively be treated by endovascular means with or without the use of balloon or stent assistance, as long as the interventionalist is astutely aware of the possibility of thrombus formation and acts accordingly with thrombolytic therapy when necessary.

Introduction

The middle cerebral artery (MCA) is the third most common location for ruptured intracranial aneurysms, and accounts for 18–22% of all intracranial aneurysms.¹ Traditionally, MCA aneurysms have been considered difficult to treat endovascularly due to complex branching patterns of the parent vessels, unfavorable dome to neck ratios, and frequent involvement of vessels within the aneurysm neck.^1–4 Therefore, the MCA is one of the few locations of aneurysms for which many institutions still prefer surgical clipping to endovascular treatment.

With the improvement in endovascular techniques such as three-dimensional (3D) computational analysis and adjunct devices such as stents and compliant balloons there has been renewed interest in endovascular treatment for MCA aneurysms in the past decade.⁵ Many studies have demonstrated feasibility and efficacy of endovascular treatment, but they still remain relatively underrepresented in the literature and the results vary widely. Feasibility ranges from 81.7–95.2%.^1–4,6–10 Successful radiographic outcome, defined as complete or near-complete aneurysm occlusion, ranges from 82.4–97%, and significant recanalization requiring retreatment from 6.1–20%.^1–4,6–10 A systematic review published in 2011 evaluating 1076 MCA aneurysms treated by endovascular means defined a 95.2% feasibility rate and found that 82.4% resulted in complete or near-complete occlusion, and 12.7% in incomplete occlusion.¹⁰ It found that at long-term follow up, 81% of treated aneurysms were stable or progressed to more complete occlusion, 9.3% had minor recurrence not requiring retreatment, and 9.6% had significant recurrence requiring retreatment.¹⁰

Of particular concern in the endovascular treatment of MCA aneurysms has been thromboembolic (TE) complications. According to the literature, TE events, including transient thrombi that resolve with thrombolytic therapy and do not result in permanent morbidity, occur at a rate of 8.5–28%.^1–4,7–10 In one early study evaluating aneurysms at all locations, MCA aneurysms were responsible for 52% of all TE complications.¹¹ However, the rate of permanent morbidity in large series since 2005 is 2.6–9%.^1–4,7–10 In the 2011 systematic review, the overall morbidity and mortality was 5.1% for unruptured and 6.0% for ruptured MCA aneurysms.

While there has been no randomized controlled clinical trial dedicated to endovascular versus open surgical treatment of MCA aneurysms, several systemic reviews and meta-analyses have attempted to compare outcomes retrospectively. Such studies quote a complete occlusion rate in microsurgical clipping of 97% and morbidity of 2.1–4.1%.^5,12 Thus, historically, the endovascular treatment of MCA aneurysms has appeared to be inferior to microsurgical clipping with a lower aneurysm occlusion rate and more associated neurologic morbidity. However, as the technology improves and surgeons become more experienced, the data need to be reevaluated. To this end, we present our own experience at a single institution, with a focus on TE complications.

Methods

The cases of 94 MCA aneurysms treated at our institution from 2002 to March 2017 were retrospectively reviewed. When certain aspects of patient data were missing, such as recent cases without long-term follow up, or detailed clinical information prior to the advent of the electronic medical record at our institution in 2006, these patients were excluded from the corresponding part of the analysis.

Clinical variables recorded include patient demographic information, mode of presentation (ruptured or unruptured, new or recurrent aneurysm), and pre- and postoperative functional status according to the Hunt Hess (HH) and modified Rankin scale (MRS) grading systems. Radiographic variables included aneurysm size, neck size, dome to neck ratio, location, and whether an arterial vessel was incorporated into the neck or dome of the aneurysm. Location was broken down into the following categories: M1 segment not involving a bifurcation, M1 at the origin of the anterior temporal artery, MCA bifurcation, or M2 distal to the bifurcation. Aneurysms at the origin of the anterior temporal artery were kept separate from other M1 non-bifurcation aneurysms in order to keep aneurysms involving a branch point separate from those that do not.

Procedure-related variables included whether intervention was successfully accomplished, the use of adjuvant devices such as balloons or stents, and the use of intra-arterial abciximab, a pharmacologic thrombolytic agent, or mechanical thrombectomy for intraoperative thrombus. Immediate radiographic outcome was recorded using the Raymond grading scale,¹³ and complete or near-complete occlusion were defined as Raymond 1 and Raymond 2, respectively. A complication was defined as an event that occurred which required additional intervention either during the case or after, or resulted in permanent neurologic deficit. Morbidity was defined as a permanent negative impact on a patient’s quality of life which was directly related to the procedure.

Follow-up radiographic data were acquired from either catheter angiography, computed tomography angiography, or magnetic resonance angiography. Patients were included in the analysis if there were follow-up data available at least 6 months after the procedure. If there was more than one follow-up study, the data were obtained from the latest study. Follow-up Raymond classification was compared with immediate Raymond classification and used to describe progression of the treated aneurysms over time. Long-term clinical outcomes were assessed from the electronic medical record notes using the MRS. Patients were only included in the analysis if there was a note from at least 6 months post-procedure.

Endovascular treatment

All endovascular procedures were performed using biplane angiography at a single institution, by five different interventionalists. For all cases, general anesthesia was administered by the department of anesthesia. Common femoral artery access was obtained and diagnostic angiography was performed, including rotational acquisition with 3D reconstruction in most cases. Calculations were made based on the images and working planes established.

Intravenous heparin (usually 50 units/kg) was administered prior to the introduction of the guide catheter for unruptured cases, and after at least one coil was securely deployed within the aneurysm dome in ruptured cases. For cases utilizing stents, patients were loaded with and maintained on aspirin 325 mg and clopidogrel 75 mg, and platelet assays were obtained preoperatively to ensure therapeutic levels. For wide-neck aneurysms treated with coil embolization alone (i.e. no stent), and thus a large surface area of coil mass exposed to arterial blood flow, patients were administered a rectal aspirin 325 mg at the conclusion of the case and maintained on a daily dose until follow up. If stents were used, patients were maintained on aspirin and clopidogrel until follow up.

Endovascular thrombolysis

When TE events were observed, abciximab was administered via a microcatheter adjacent to the thrombus. Usually, 10 mg was infused initially followed by an additional 5 mg if needed after reassessment. In rare cases, patients were maintained on an abciximab or heparin drip overnight in the intensive care unit (ICU). If mechanical thrombectomy was required, this was accomplished using stent retriever devices or aspiration catheters.

Statistical analysis

Descriptive statistics were used to provide a summary of clinical and radiographic characteristics such as demographics, aneurysm features and procedural methods. Associations between aneurysm characteristics or procedural methods and radiographic outcome of aneurysm treatment as well as intraoperative TE events were evaluated. A two-sample t-test or Wilcoxon rank-sum test, as appropriate, was used for continuous measures and the Chi-square test or Fisher’s exact test, as appropriate, was used for categorical measures. All statistical tests were two-sided and p < 0.05 was considered statistically significant. All analyses were performed in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

Results

Patient demographics

Of the 94 MCA aneurysms reviewed, 93 underwent successful endovascular treatment. The one aneurysm that could not be treated was a neck recurrence, and the interventionalist was unable to keep coil loops securely within the neck. The average age of the patient population was 57 ± 15 years, ranging from 14 to 83, with 67 (72.04%) females and 26 (27.96%) males. Of the 93 treated aneurysms, 40 (43.01%) were ruptured, 53 (56.98%) were unruptured, and 6 (6.45%) were recurrences of previously clipped or coiled aneurysms (Table 1).

Table 1.

Pre-operative clinical and radiographic characteristics of the 93 treated MCA aneurysm cases.

Characteristics	n (%) or mean (std), n = 93
Clinical
Female	67 (72.04%)
Age	57 (15)
Ruptured aneurysm	40 (43.01%)
Recurrent aneurysm	6 (6.45%)
Pre-op MRS 0–2	63 (80.76%)
Radiographic
Maximum diameter	6.01 (2.73)
Neck size	3.25 (1.20)
Dome/neck ratio	1.87 (0.61)
Dome/neck ratio < 2.0	44 (55.70%)
Vessel in neck	60 (65.22%)
Vessel in dome	3 (3.26%)
Location
Left side	45 (48.39%)
MCA bifurcation	67 (72.04%)
Anterior temporal	11 (11.83%)
M1 non-branch point	11 (11.83%)
M2	4 (4.30%)

MCA: middle cerebral artery; MRS: modified Rankin scale

Aneurysmal features

The average maximum diameter of the treated aneurysms was 6.01 ± 2.73 mm, ranging from 1.4 to 14.8 mm. The 1.4 mm aneurysm was a pseudoaneurysm obtained during surgical resection of a tumor. It is the only pseudoaneurysm in the series, and the only pediatric patient in the series. The average neck size was 3.25 ± 1.20 mm, ranging from 1.5 to 8.7 mm. The average dome to neck ratio was 1.87 ± 0.61, ranging from 0.86 to 3.22. A total of 44 (55.70%) of the aneurysms had a dome to neck ratio of <2 (Table 1).

Of the aneurysms treated, 45 (48.39%) were on the left side, 11 (11.83%) were on the M1 segment of the MCA not involved with a bifurcation, 11 (11.83%) were at the origin of the anterior temporal artery, 67 (72.04%) were at the MCA bifurcation, and 4 (4.30%) were on the M2 segment. Of these, 60 (65.22%) had a vessel arising from the neck, and 3 (3.26%) had a vessel arising from the dome of the aneurysm (Table 1).

Use of adjunctive devices

In the treatment of the 93 MCA aneurysms, an adjunctive device (balloon or stent) was used in 43 (46.24%) cases. Balloon remodeling was utilized in 24 (25.81%) cases, and stents were used in 19 (20.43%) cases (Table 2). Of the stents, 3 were pipeline embolization devices, 12 were low-profile visualized intraluminal support devices (LVIS Jr), and 4 were Neuroform® stents. Of the LVIS cases, four were Y-stent configurations incorporating both branches of the MCA bifurcation. Overall, 18 of the 19 stent-assisted embolization cases were planned preoperatively; 1 was placed in the acute setting due to a coil fracture.

Table 2.

Procedural characteristics, complications, and immediate radiographic and clinical outcomes of treated MCA aneurysm cases.

Characteristics	n (%), n = 93
Immediate radiologic outcome (n = 90)
Raymond 1	56 (62.22%)
Raymond 2	25 (27.78%)
Raymond 1 or 2	81 (90.00%)
Raymond 3a	8 (8.89%)
Raymond 3b	1 (1.11%)
Adjuvant device
Balloon	24 (25.81%)
Stent	19 (20.43%)
Balloon or stent	43 (46.24%)
Complications
Transient intra-procedural thrombus	14 (15.05%)
Permanent intra-procedural thrombus	1 (1.08%)
Intraoperative rupture	1 (1.08%)
Post-procedural thrombus requiring takeback	3 (3.23%)
Vessel dissection	1 (1.08%)
Coil migration/fracture	2 (2.15%)
Rebleed	1 (1.08%)
Intervention for thrombus
Intra-arterial reopro	16 (17.20%)
Mechanical thrombectomy	2 (2.15%)
Clinical
New permanent neurologic deficit	2 (2.15%)
Immediate post-op MRS 0–2	59 (80.82%)

MCA: middle cerebral artery; MRS: modified Rankin scale

Initial angiographic results

Of the treated aneurysms 81 (90%) achieved a complete or near-complete (Raymond 1 or 2) occlusion at the conclusion of the procedure. A total of 56 (62.22%) of the outcomes were classified as Raymond 1, 25 (27.78%) as Raymond 2, 8 (8.89%) as Raymond 3a, and 1 (1.11%) as Raymond 3b (Table 2). The three cases utilizing a pipeline embolization device were excluded from the immediate outcomes analysis because the aneurysm dome is not expected to be immediately occluded in these cases. In the one case classified as Raymond 3b, the aneurysm was multi-lobulated and the presumed rupture point was completely coiled, leaving the other, more difficult segment intentionally open with a plan to treat with a stent in the future.

Procedure-related complications

TE events were the most common complication, as expected. There were 15 immediate intra-procedural TE events, of which 14 (15.05%) resolved and 1 (1.08%) resulted in permanent occlusion of a small M2 branch. There were three delayed, post-procedural TE events in two patients (one patient had bilateral MCA aneurysms treated simultaneously) within the same day of treatment. In both cases, a new neurologic deficit was quickly noted in the neurologic ICU, the patient was rushed to the angiography suite, a thrombus was found and was completely resolved with abciximab. Abciximab was used in 16 (17.20%) cases, and mechanical thrombectomy was required in 2 (2.15%) cases (Table 2).

Other, non-TE complications were rare. There was one (1.08%) intraoperative rupture, one (1.08%) arterial dissection, two (2.15%) cases of coil migration or fracture, and one (1.08%) delayed rebleed after treatment (Table 2). In the case of the rupture, a balloon was quickly inflated, coils were used to pack the rupture site and the patient was taken for craniotomy and clipping with a good neurologic outcome. In one case from 2005, device failure led to a coil fracture, and subsequent retrieval led to an arterial dissection which required a stent to be placed acutely, which developed thrombus and required abciximab to resolve. In the case of the rebleed, craniotomy was performed that night for clot evacuation and aneurysm clipping.

Clinical outcomes

Of the patient population, 63 (80.76%) had a pre-operative MRS of 0–2, and 59 (80.82%) had a postoperative MRS of 0–2 (Table 2). Only one of the isolated TE events resulted in permanent neurologic deficit, including the one that could not be resolved. There were only two cases where the postoperative MRS was worse than the pre-operative MRS: the case involving the coil fracture, dissection, and stent placement, and the postoperative rebleed. In the case of the intraoperative rupture, there was no significant hematoma and the patient awoke from anesthesia neurologically intact.

Of the 55 patients with recorded follow-up clinical information, 49 (89.09%) had an MRS of 0–2 (Table 3). There were no cases of delayed worsening outcome. There were only two cases of permanent neurologic deficit resulting from the procedures in this series, and thus the associated morbidity was 2.15%.

Table 3.

Follow-up clinical and radiographic characteristics of treated MCA aneurysm cases.

Characteristics	n (%), n = 60
Follow-up data availability
>6 months	60 (64.51%)
>1 year	39 (41.93%)
Raymond score at longest follow up (>6 mo)
Raymond 1	43 (71.67%)
Raymond 2	16 (26.67%)
Raymond 1 or 2	59 (98.33%)
Raymond 3a	0
Raymond 3b	1 (1.67%)
Radiographic change
Stable or improved	50 (83.33%)
Minor recanalization, not requiring  retreatment	5 (8.33%)
Recanalization requiring retreatment	5 (8.33%)
Clinical
MRS 0–2 at longest follow up  (>6 mo), n = 55	49 (89.09%)

MCA: middle cerebral artery; mo: months; MRS: modified Rankin scale

Radiographic follow up

Of the 93 treated aneurysms, 60 (64.51%) had radiographic follow up of at least 6 months, and 39 (41.93%) at least 1 year (Table 3). There are several reasons for this follow-up rate. Some patients treated in the acute setting undergo treatment at our institution and then transfer follow-up care to other institutions for insurance reasons or in order to be closer to home. Some were simply lost to follow up. There were two recorded deaths, unrelated to the procedure, both of which presented as HH grade 5. Furthermore, eight of the patients were treated too recently to have long enough follow-up data at the time of writing.

Of the patients with at least 6 months follow-up data, 59 (98.33%) were categorized as Raymond 1 or 2. There were 43 (71.67%) as Raymond 1, 16 (26.67%) as Raymond 2, none as 3a and 1 (1.67%) as 3b. The one Raymond 3b aneurysm was treated with a pipeline embolization device, and underwent placement of a second pipeline embolization device. A total of 50 (83.33%) of the aneurysms were stable-to-improved radiographically, 5 (8.33%) had minor recanalization which did not require retreatment, and 5 (8.33%) had significant recanalization requiring retreatment (Table 3). Of those that required retreatment, two had immediate post-op Raymond 1 occlusion, two had Raymond 2, and one had Raymond 3b (pipeline).

Associations with good radiologic outcome

There were no variables that had statistically significant association with good immediate radiologic outcome (Table 4) or with stable-to-improved radiologic outcome at follow up (Table 5). However, there are some trends that are worth noting. Of the nine aneurysms that had a poor radiologic outcome (Raymond 3a or 3b), six were MCA bifurcation aneurysms with an M2 vessel arising from the neck. In aneurysms for which balloon remodeling or stenting was utilized, 35 out of 40 (87.50%) resulted in a good immediate outcome. Likewise, at longest follow up, 70% of the aneurysms that had minor or major recanalization were MCA bifurcation aneurysms and had a vessel arising from the neck, and 85.71% of aneurysms that were stable-to-improved radiologically utilized balloon or stent assistance.

Table 4.

Associations with good immediate radiologic outcome (Raymond 1 or 2). None reach statistical significance.

Characteristics	No, n = 9	Yes, n = 81	p-value
Ruptured (n = 40)	3 (7.50%)	37 (92.50%)	0.7258
MCA bifurcation (n = 66)	6 (9.09%)	60 (90.91%)	0.6957
Anterior temporal (n = 11)	1 (9.09%)	10 (90.91%)	1.0000
M1 non-branch (n = 9)	2 (22.22%)	7 (77.78%)	0.2207
M2 (n = 4)	0 (0%)	4 (100%)	1.0000
Maximal diameter	6.13 (3.79)	6.02 (2.58)	0.6593
Neck size	3.20 (1.34)	3.20 (1.13)	0.9947
Dome/neck	1.92 (0.70)	1.88 (0.61)	0.8640
Dome/neck <2 (n = 43)	4 (9.30%)	39 (90.70%)	1.0000
Vessel in neck (n = 59)	6 (10.17%)	53 (89.83%)	1.0000
Balloon (n = 24)	3 (12.50%)	21 (87.50%)	0.6957
Stent (n = 16)	2 (12.50%)	14 (87.50%)	0.6586
Balloon or stent (n = 40)	5 (12.50%)	35 (87.50%)	0.5035

MCA: middle cerebral artery

Table 5.

Associations with stable or improved radiologic characterization at longest follow up >6 months. None reached statistical significance.

Characteristics	No, n = 10	Yes, n = 50	p-value
Ruptured (n = 21)	4 (19.05%)	17 (80.95%)	0.7287
MCA bifurcation (n = 41)	7 (17.07%)	34 (82.93%)	1.0000
Anterior temporal (n = 9)	1 (11.11%)	8 (88.89%)	1.0000
M1 non-branch (n = 7)	2 (28.57%)	5 (71.43%)	0.3299
M2 (n = 3)	0 (0%)	3 (100%)	1.0000
Maximal diameter	5.91 (1.92)	5.60 (2.36)	0.7034
Neck size	3.43 (1.44)	3.17 (1.28)	0.5703
Dome/neck	1.87 (0.72)	1.81 (0.56)	0.7757
Dome/neck < 2 (n = 33)	6 (18.18%)	27 (81.82%)	1.0000
Vessel in neck (n = 40)	7 (17.50%)	33 (82.50%)	1.0000
Balloon (n = 15)	3 (20.00%)	12 (80.00%)	0.6998
Stent (n = 13)	1 (7.69%)	12 (92.31%)	0.4361
Balloon or stent (n = 28)	4 (14.29%)	24 (85.71%)	0.7375

MCA: middle cerebral artery

Associations with thromboembolic events

There were no variables that had statistically significant associations with intraoperative TE events (Table 6). It is worth noting, however, that TE events only occurred in aneurysms that were at a branch point (MCA bifurcation or anterior temporal). Furthermore, 13 of the 15 (86.67%) intraoperative events occurred in aneurysms that contained a vessel arising from the neck.

Table 6.

Associations with intra-procedural thromboembolic event. None reached statistical significance.

Characteristics	No event n = 78	Thromboembolic event n = 15	p-value
Ruptured (n = 40)	34 (85.00%)	6 (15.00%)	0.5496
MCA bifurcation (n = 67)	56 (83.58%)	11 (16.42%)	1.0000
Anterior temporal (n = 11)	7 (63.64%)	4 (36.36%)	0.0579
M1 non-branch (n = 11)	11 (100%)	0 (0%)	0.2062
M2 (n = 4)	4 (100%)	0 (0%)	1.0000
Maximal diameter	5.95 (2.69)	6.33 (3.02)	0.6382
Neck size	3.22 (1.23)	3.38 (1.11)	0.6917
Dome/neck ratio	1.86 (0.61)	1.96 (0.65)	0.6017
Dome/neck ratio <2 (n = 44)	37 (84.09%)	7 (15.91%)	0.8417
Vessel in neck (n = 60)	47 (78.33%)	13 (21.67%)	0.1264
Balloon (n = 24)	18 (75.00%)	6 (25.00%)	0.1137
Stent (n = 19)	16 (84.21%)	3 (15.79%)	1.0000
Balloon or stent (n = 43)	34 (79.07%)	9 (20.93%)	0.1417

MCA: middle cerebral artery

Discussion

Radiologic outcome

In this study, we review the cases of 94 MCA aneurysms that underwent an endovascular procedure at our institution from 2002 to March 2017. Successful endovascular treatment was feasible in 93 of them (99%). Immediate complete or near-complete occlusion was achieved in 81 (90%). At longest follow up of 6 months or greater, 83.3% of the aneurysms were stable-to-improved in the Raymond occlusion classification, 8.3% were found to have minor recanalization not requiring retreatment, and 8.3% required retreatment due to significant recanalization. These outcomes are similar to previously published series.^1-4,6-10

The endovascular treatment of MCA aneurysms is relatively underrepresented in the literature. Overall, six published series evaluating a total of 795 MCA aneurysms from 2005 to 2010 describe initial rates of complete occlusion of 46.1–77.2%, near-complete occlusion 19.5–50%, and incomplete occlusion 2.6–3.4%.^2–4,7–9 Overall recurrence rates range from 8–27.2%.^2–4,7–9 A meta-analysis from 2011 evaluated 1076 aneurysms from 12 published series and found rates of 82.4% complete or near-complete occlusion, 12.7% incomplete occlusion, and 4.8% failed treatment.¹⁰ The rate of recanalization requiring treatment was 9.6%, and minor recanalization not requiring treatment was 9.3%.¹⁰ Thus, our results of 90% immediate complete or near-complete occlusion and 8.3% major recanalization requiring retreatment are similar to slightly superior.

An early study published in 2005 found that larger aneurysm size, larger neck, ruptured aneurysms, and incomplete initial occlusion were associated with a higher rate of recurrence.³ These results were supported by another study in 2009.⁸ While these findings were not statistically significant in our series, there was a trend toward worse radiologic outcome for more complex aneurysms. Specifically, of the nine aneurysms with immediate Raymond classification of 3a or 3b, seven (77.8%) were at branch points (MCA bifurcation or anterior temporal artery), and six (66.7%) had a vessel arising from the neck. Likewise, of the 10 aneurysms with minor or major recanalization, 8 (80%) were at a branch point and 7 (70%) had a vessel arising from the neck. This supports the intuitive assumption that complex aneurysms such as those at branch points or involving a branch vessel in the neck are more difficult to treat successfully.

However, despite these trends, the rates of occlusion of the more complex MCA aneurysms were not significantly different than others. In fact, immediate Raymond outcome of 1 or 2 was present in 90.9% of MCA bifurcation aneurysms, 90.9% of anterior temporal branch point aneurysms, and 89.8% of aneurysms with a vessel arising from the neck, which is very similar to the overall rate of 90%. Furthermore, a stable-to-improved Raymond score at the longest follow up was present in 82.9% of MCA bifurcation aneurysms, 88.9% of anterior temporal branch point aneurysms, and 82.5% of aneurysms with a vessel arising from the neck, which is similar to the overall rate of 83.3%. These results might be biased by the fact that a large portion of the aneurysms in the series meet one of these three criteria, but it still may suggest that more complex appearing aneurysms can be treated with acceptable outcomes.

Another interesting finding was that the use of an adjunctive device such as a balloon catheter or stent was not associated with better immediate or follow-up radiologic outcome. Stenting alone trended toward better outcome at follow up, with 92.3% of the cases utilizing a stent resulting in stable-to-improved Raymond classification, but this did not reach statistical significance (most likely because there were only 13 total stenting cases). These results are most likely due to the fact that balloons or stents are used in cases that tend to be more difficult, and thus a bias is created. This finding was described by Vendrell et al. in 2009, where they found that balloon remodeling was associated with more recurrences in the long term (42.6% versus 16.4%, p = .016), but that this is due to the fact that balloon remodeling was only used for aneurysms with ‘complex anatomy.’⁹ As the technology improves and newer stenting technology such as pipeline and low-profile intraluminal support devices are utilized more frequently, their role in complex aneurysms will likely be clarified.

Thromboembolic events and morbidity

TEs were the most common complication in our series, as has been described previously.^1–4,7–10 However, of the 18 total TE events (19.3%), only 1 resulted in a permanent occlusion. The other 17 were completely resolved with intra-arterial thrombolytic therapy or mechanical thrombectomy (2 cases). Furthermore, of these 18 events, only one of them resulted in neurologic deficit or functional impairment. There were only two cases resulting in morbidity (2.15%), both from 2005, which included a coil fracture that required acute stenting which transiently thrombosed and resulted in an infarct, and a rebleed the night after coiling which required craniotomy. This morbidity of 2.15% is superior to the rate quoted in previous studies.^1–10

The rate of TE events in previous studies ranged from 8.5–19.6%.^2–4,7–9 New permanent morbidity ranged from 2.6–10%.^2–4,7–9 The 2011 meta-analysis described an overall morbidity/mortality of 5.1% for unruptured and 6.0% for ruptured aneurysms, and specifically a morbidity due to TE complication of 3.2%.¹⁰ The difference in our results likely reflects improving technology over the last decade in pharmacologic thrombolytic therapy, compliant balloons, intermediate and microcatheters, and an overall hypersensitivity to the issue of TE events as it has been more frequently described over the years.

The fact that we found a rate of TE events of 19.3% but only one led to permanent occlusion and only one led to permanent morbidity (1%) is important, as it suggests that these events can be safely dealt with if the interventionalist is vigilant. The use of pharmacologic thrombolytic therapy has long been described as a safe and effective method of completely resolving intraoperative TE events.^1,14–16 The development of a thrombus during an endovascular intervention that is quickly and efficiently resolved can be likened to a bleeding vessel encountered during craniotomy that is quickly and efficiently coagulated without causing harm to the patient.

This point is particularly important because the rate of TE complications has been quoted as a major factor in the argument of surgical versus endovascular treatment of MCA aneurysms. The overall morbidity of craniotomy for clipping of MCA aneurysms in the literature ranges from 2.1–4.1%.^5,12 A recent meta-analysis comparing the two treatment modalities describes rates of TE complications of 1.8% for clipping versus 10.7% for coiling, but no significant difference in clinical outcomes was found.¹⁷ Another describes rates of ‘unfavorable outcomes’ of 2.1% for clipping and 4.9–6.5% for coiling.⁵ Our reported morbidity of 2.15% is comparable with the reported morbidity of open surgical clipping, and thus suggests that the two modalities can achieve similar clinical outcomes. Granted, the reported long-term occlusion rates of 95.5–97% remain superior to endovascular rates, but this will become less relevant as the techniques continue to improve and the use of stent-assisted embolization of recurrences becomes more established.^5,12

In our series, there were no factors that were significantly associated with development of TE events. However, these events only occurred in branch point aneurysms (MCA bifurcation or anterior temporal), and 13 of the 15 intraoperative events (86.7%) occurred during treatment of aneurysms that contained a vessel arising from the neck. In the literature, it has been shown that TE events are more common in ruptured aneurysms,² but were not associated with aneurysm location (i.e. bifurcation versus non-bifurcation).⁸ Furthermore, in our series the cases that used balloons or stents had a higher incidence of TE events (21% versus 16%), but this did not reach statistical significance. Despite the mixed results, it is logical that more complex aneurysms take longer time to treat and thus the interventionalist should be particularly vigilant in watching for the development of thrombus.

The high potential for TE complications in the endovascular treatment of MCA aneurysms has led to such statements as the following in the literature: “We believe that the complex branching anatomy of the MCA reduces the safety of the procedure and leads to unintentional branch occlusion,” and “We prefer not to use stents in the treatment of MCA aneurysms because we have found that it makes the procedure cumbersome and increases the level of risk.”² Our own results and review of the literature suggest the contrary: that endovascular treatment of even complex MCA aneurysms with or without adjuvant devices is safe and complications can be avoided with vigilance and quick action in the event of a developing thrombus. At our institution, we often utilize a delayed angiogram 15–20 minutes after embolization to ensure no delayed formation of thrombus.

Limitations

Our study is retrospective in nature and thus any conclusions are limited and must be interpreted accordingly. We are also limited by number of total patients (94) and length of follow up. As discussed, many patients were lost to follow up due to insurance or geographic reasons, and some recent cases were included that were too early to have follow up.

Conclusions

We report from the cases of 93 treated MCA aneurysms at our institution a good immediate radiographic outcome (Raymond 1 or 2) of 90%, stable-to-improved outcome at longest follow up of 83.3%, with 8.3% requiring retreatment. There was a rate of TE events of 19%, but only a permanent morbidity of 2.15%. We found no statistically significant correlation between aneurysm location, size, morphology, or use of adjuvant device with radiographic outcome or TE event. Therefore, we stress the fact that even complex MCA aneurysms can safely and effectively be treated by endovascular means with or without the use of balloon or stent assistance, as long as the interventionalist is astutely aware of the possibility of thrombus formation and acts accordingly with thrombolytic therapy when necessary.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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