Unlike saccular or berry aneurysms, which present more often with subarachnoid hemorrhage, fusiform aneurysms present more often with ischemic stroke or mass effect. The most time-tested treatment of fusiform vertebrobasilar aneurysms consists of flow reduction or flow reversal. Recently, flow diversion has been attempted with mixed results in the posterior circulation. Given the described pathophysiologic processes of fusiform aneurysms that may be altered with modern medical therapies, future investigators may look to medical treatment of these lesions, especially in cases of poor surgical candidates.
Key points
- •
Vertebrobasilar aneurysms present most commonly with brainstem ischemic stroke or compressive symptoms of the brainstem, cerebellum, or cranial nerves.
- •
The natural history of patients with fusiform vertebrobasilar aneurysms presenting with ischemic stroke or compressive symptoms is for the presenting signs and symptoms to steadily progress.
- •
Flow reduction or flow reversal is the most time-tested management strategy for fusiform vertebrobasilar aneurysms presenting with compressive symptoms, but this treatment relies largely on the presence of adequate collateralization through the posterior communicating arteries.
- •
Mixed results have been reported with the use of flow diversion for fusiform vertebrobasilar aneurysms, and their use should be reserved for patients with compressive symptoms and poor collateralization through the posterior communicating arteries.
- •
Patients with vertebrobasilar aneurysms presenting with ischemic stroke and no compressive symptoms are best managed with anticoagulation.
Introduction
Vertebrobasilar fusiform aneurysms are among the most daunting lesions treated by cerebrovascular surgeons. Many names have been linked to these lesions, including giant serpentine aneurysm, giant fusiform aneurysm, S aneurysm, megadolichobasilar artery, dolichoectacic artery, fusiform aneurysm, and transitional aneurysm. Perhaps the most universal definition of these lesions is aneurysms with separate inflow and outflow ostia. The earliest descriptions of fusiform aneurysms of the vertebrobasilar arterial system were most consistent with a dolichoectatic basilar artery. There is no arteriography available for most cases, and most patients presented with cranial neuropathies.
A fusiform aneurysm of the basilar artery was first described by Wells in 1922 on surgical exploration in a patient with paresis of cranial nerves 6 to 8 and obstructive hydrocephalus. Dr Walter Dandy operated on a series of 10 patients with trigeminal neuralgia and described a so-called S aneurysm. Greitz and Lofstedt in 1954 reported 5 cases of ectasia of the basilar artery. The patients of their series were more consistent with compressive or ischemic presentations that are associated with fusiform aneurysms. Reports since this time have drastically expanded our understanding of these lesions. In this article, fusiform vertebrobasilar aneurysms are reviewed, including incidence, presentation, natural history, pathophysiology, and treatment, with a suggested algorithm for treatment based on review of the literature. Dissecting aneurysms, which have a distinct behavior of presenting with subarachnoid hemorrhage (SAH) followed by frequent early rebleeding, are not discussed.
Introduction
Vertebrobasilar fusiform aneurysms are among the most daunting lesions treated by cerebrovascular surgeons. Many names have been linked to these lesions, including giant serpentine aneurysm, giant fusiform aneurysm, S aneurysm, megadolichobasilar artery, dolichoectacic artery, fusiform aneurysm, and transitional aneurysm. Perhaps the most universal definition of these lesions is aneurysms with separate inflow and outflow ostia. The earliest descriptions of fusiform aneurysms of the vertebrobasilar arterial system were most consistent with a dolichoectatic basilar artery. There is no arteriography available for most cases, and most patients presented with cranial neuropathies.
A fusiform aneurysm of the basilar artery was first described by Wells in 1922 on surgical exploration in a patient with paresis of cranial nerves 6 to 8 and obstructive hydrocephalus. Dr Walter Dandy operated on a series of 10 patients with trigeminal neuralgia and described a so-called S aneurysm. Greitz and Lofstedt in 1954 reported 5 cases of ectasia of the basilar artery. The patients of their series were more consistent with compressive or ischemic presentations that are associated with fusiform aneurysms. Reports since this time have drastically expanded our understanding of these lesions. In this article, fusiform vertebrobasilar aneurysms are reviewed, including incidence, presentation, natural history, pathophysiology, and treatment, with a suggested algorithm for treatment based on review of the literature. Dissecting aneurysms, which have a distinct behavior of presenting with subarachnoid hemorrhage (SAH) followed by frequent early rebleeding, are not discussed.
Incidence
The incidence of fusiform aneurysms of the vertebrobasilar system is low. Dolichoectasia of any intracranial artery in the general population is estimated at less than 0.05%. In cases of vertebral angiography, the incidence ranges from 17 of 10,000 (0.17%) for all indications to 132 of 2265 (5.8%) in cases of stroke. Another series of 387 patients undergoing computed tomography (CT) or magnetic resonance (MR) angiography (MRA) for stroke showed 10 patients (2.6%) with vertebrobasilar dolichoectasia. Autopsy series also show a wide range, but remain low in incidence, ranging from 6 of 5762 (0.10%) in a series from 1914 to 1956 at Columbia University to 5 of 7500 (0.07%) in a VA hospital series. One clinical series of treated posterior circulation aneurysms classified 4 of 528 (0.76%) aneurysms as fusiform vertebrobasilar aneurysms.
The discrepancy in incidence in these studies is likely related to a combination of the loose definitions of these lesions and subselection. In addition, the incidence may increase in high-risk populations. Yu and colleagues found a strong correlation with hypertension (64%) and tobacco smoking (74%) with intracranial arterial ectasia. In addition, Mitsias and Levine described a high incidence in Fabry disease, with symptomatic disease of a dilated vertebrobasilar system found in 60% of heterozygotes and 67% of homozygotes. Conversely, vertebrobasilar aneurysms have been reported in children but are rare. In 3000 postmortem brain examinations in children, Housepian and Pool found no case of any intracranial aneurysm.
Presentation
Because of the low incidence of these lesions and the relative paucity of reported cases, initial reports focused primarily on the striking pathologic features and rarely provided adequate clinical details. However, a review of published series and case reports shows some common features. Unlike saccular aneurysms, vertebrobasilar fusiform aneurysms show a significant male predominance, accounting for greater than 70% of the 408 reported cases ( Tables 1 and 2 ). Although affected patients ranged from age 5 years to 87 years, the average reported age at diagnosis was ∼60 years, significantly older than that typically presenting with saccular aneurysms. Although not often reported, commonly encountered comorbidities include hypertension (31%–69%), diabetes mellitus (10%–15%), hyperlipidemia (40%), coronary artery disease (23%–28%), and smoking (50%). Data regarding connective tissue disorders were almost universally absent, an oddity given the mechanisms underlying vessel ectasia and the known importance of family history in the risk of aneurysm formation. This subject was addressed only in the well-studied Mayo Clinic cohort, in whom 4% were found to have a known connective tissue disorder, including Fabry disease and autosomal-dominant polycystic kidney disease.
| Author, Year | Number of Patients | Male (n) (%) | Mass Effect | Ischemic Event | SAH | Notable |
|---|---|---|---|---|---|---|
| Flemming et al, 2005 | 159 | 118 (74) | 35 | 44 | 5 | 63 (incidental) |
| Drake & Peerless, 1997 | 61 | 30 (49) | 35 | 15 | NR | 11 (headache) |
| Coert et al, 2007 | 39 | NR | 8 | 5 | 26 | |
| Milandre et al, 1991 | 23 | 16 (76) | 13 | 9 | 1 | |
| Nishizaki et al, 1986 | 23 | 19 (82) | 6 | 11 | 2 | 2 (incidental); 1 (cerebellar intracranial hemorrhage) |
| Herpers et al, 1983 | 22 | 11 (50) | 9 | 9 | 1 | 2 (incidental); 1 (dementia) |
| Anson et al, 1996 | 20 | 17 (85) | 10 | 5 | 4 | |
| Echiverri et al, 1989 | 13 | 11 (85) | 4 | 9 | 1 | |
| Boeri & Passerini, 1964 | 10 | 7 (70) | 8 | NR | 2 | |
| Kalani et al, 2013 | 7 | 6 (86) | 6 | 1 | 0 | |
| Pessin et al, 1989 | 7 | 5 (71) | 0 | 7 | 0 | |
| Giang et al, 1988 | 6 | 5 (83) | 3 | 2 | 0 | 1 (incidental) |
| Meckel et al, 2013 | 5 | 4 (80) | 3 | 2 | 0 | |
| Nakatomi, 2000 | 4 | 3 (75) | 3 | 1 | 0 | |
| Sluzewski et al, 2001 | 3 | 3 (100) | 1 | 1 | 1 | |
| Wenderoth et al, 2003 | 2 | 2 (100) | 0 | 1 | 1 | |
| Binning et al, 2011 | 1 | 1 (100) | 1 | 0 | 0 | |
| Cohen, 2012 | 1 | 1 (100) | 0 | 1 | 0 | |
| Greenberg, 2007 | 1 | 1 (100) | 0 | 0 | 1 | |
| Islak et al, 2002 | 1 | 0 (0) | 0 | 0 | 1 |
| Author, Year | Years Treated | n | Location | Aneurysm Type | Treatment (%) | Good Outcome by Location (%) a |
|---|---|---|---|---|---|---|
| Drake & Peerless, 1997 | 1965–1992 | 61 | Basilar (37) VBJ (10) Vertebral (14) | Fusiform | Flow reduction (18) Flow reversal (64) Trapping (10) Wrap/explore (7) Clip reconstruction (2) | Basilar (73) VBJ (60) Vertebral (64) |
| Coert et al, 2007 | 1991–2005 | 39 | Basilar/VBJ (18) Vertebral/PICA (21) | Fusiform/dolichoectatic | Surgery (26) Embolization (67) Surgery and embolization (8) | Basilar/VBJ (39) Vertebral/PICA (62) |
| Anson et al, 1996 | 1986–1994 | 19 | Basilar (8) VBJ (5) Vertebral (6) | Fusiform (13) Dolichoectatic (6) | Heterogeneous b | Basilar (28) VBJ (80) Vertebral (83) |
| Leibowitz et al, 2003 | 1997–2000 | 10 | Basilar (1) VBJ (5) Vertebral (4) | Fusiform | Endovascular balloon or coil occlusion | Basilar/VBJ (16) c Vertebral (25) d |
| Aymard et al, 1991 | NR | 9 | Basilar (3) VBJ (3) Vertebral (3) | Fusiform | Endovascular balloon occlusion | Basilar (33) VBJ (100) Vertebral (67) |
a mRS ≤2, GOS ≥4, or reported good or normal.
b Combinations of thrombectomy with aneurysmorrhaphy, proximal or distal occlusion, bypasses, anticoagulation, clip reconstruction, trapping.
c Continued aneurysm filling in all cases; 67% died.
d All completely occluded, 100% improved mRS from presentation.
In addition to the small subset of patients in whom vertebrobasilar fusiform aneurysms were incidentally diagnosed, the presenting symptoms in most patients are related to 3 basic mechanisms: mass effect, ischemia, or aneurysmal rupture. Mass effect, which was observed in 43% of patients (see Table 1 ), occurred when the ectatic vessel compressed surrounding tissues, including the brainstem and cerebellum, resulting in numerous cranial nerve palsies or in noncommunicating hydrocephalus, and typically developed over years. In our experience, even although the clinical course is characteristically slowly progressive, it is often punctuated by stuttering episodes of abrupt exacerbation, typically as an early manifestation but also as a late preterminal manifestation. These episodes often correspond to intramural hemorrhage or microdissection evident on brain MR imaging (MRI) studies ( Fig. 1 ).
When discussing cranial neuropathies caused by vertebrobasilar fusiform aneurysms, the most common cranial nerves involved are V to VIII. Nishizaki and colleagues, for example, reported dysfunction affecting the facial nerve in 4 of 6 patients, whereas in Herpers and colleagues’ study, hemifacial spasm accounted for 22% of patients presenting with compressive cause. Similarly, trigeminal neuralgia and abducens nerve palsy were frequently implicated as clinical correlates of brainstem compression. Defects involving other cranial nerves, including cranial nerves IX, X, and XII, were more rarely described. Although varying by aneurysm location, obstructive hydrocephalus was consistently reported as a sequela of brainstem compression. In many of these patients, headache was a principal presenting symptom.
Perhaps the most common presentation of patients with vertebrobasilar fusiform aneurysms involves ischemic stroke symptoms. Accounting for ∼44% of patients (see Table 1 ), this subgroup comprises clinical syndromes ranging from transient ischemic attacks (TIAs) to catastrophic pontine ischemia consistent with a locked-in syndrome. In the Mayo Clinic cohort, Flemming and colleagues further delineated the distribution of infarction in these patients, showing a preference for the pons (50%), lateral medulla (9%), cerebellar hemispheres (7%), midbrain/thalamus (4.5%), and occipital lobes (4.5%). TIAs comprised 25% of clinical presentations in the population of patients with ischemia. In the cohort described by Echiverri and colleagues, ischemic insult most often affected the pons (33%) and thalamus (11%), whereas 55% of patients experienced TIAs alone. Similarly, in a review of 23 dolichoectatic vertebrobasilar aneurysms, Nishizaki and colleagues identified ischemia to the pons (30%) and temporal lobe (9%). Cerebral lacunar infarcts accounted for 17% of patients in this series, whereas TIAs and vertebrobasilar insufficiency occurred in 22% of patients. Pessin and colleagues found that 43% of patients presented with extensive ischemic insult to the pons. The remaining patients in this series initially suffered TIAs (57%) and subsequently progressed to develop formal ischemic lesions.
Unlike giant saccular aneurysms, aneurysmal rupture is, perhaps, the least common diagnostic phenotype of vertebrobasilar fusiform aneurysms. Invariably developing sudden headache with or without focal neurologic signs, only ∼13% of all patients (see Table 1 ) presented with corresponding SAH. Most of these patients were described in a study on the treatment of vertebrobasilar aneurysms at Stanford University. In this report, only 39% of 26 patients with ruptured fusiform aneurysms had good neurologic status (Hunt and Hess grade 1–2). Conversely, 61% of patients presented with a pretreatment Hunt and Hess grade of 3 to 4. In all, aneurysmal rupture comprised 67% of their study population. In contrast, only 23% of patients in Anson and colleague’ study and 3% of described cases in Flemming and colleagues’ study presented with SAH. All 5 patients with SAH in Flemming and colleagues’ study were of a poor clinical grade.
Natural history
The natural history of vertebrobasilar fusiform aneurysms is dependent on the presenting signs and symptoms. Flemming and colleagues reported the most complete longitudinal evaluation of fusiform vertebrobasilar aneurysms over a 12-year period, yielding 159 cases with 719 patient-years of follow-up. Dissecting vertebrobasilar aneurysms were excluded from this analysis. These lesions were defined as having a vessel caliber 1.5 times normal without an identifiable neck as previously defined by Huber. These investigators proposed a radiographic classification of vertebrobasilar fusiform aneurysms, including fusiform = aneurysmal dilation of the involved segment; dolichoectactic = uniform dilation of the involved segment; transitional = uniform dilation of the involved segment with superimposed aneurysmal dilation.
In the classification scheme by Flemming and colleagues, 57% of patients had a dolichoectatic artery, 25% had a transitional aneurysm, and 18% had a fusiform aneurysm. Of patients in this series, 20.8% had an indeterminate type. The transitional-type or fusiform-type aneurysms were more likely to be symptomatic than dolichoectactic aneurysms. In this series, predictors for mortality after adjustment for age were transitional type (hazard ratio [HR] = 3.56), fusiform type (HR = 7.7), and basilar involvement (HR = 8.77). Conversely, dolichoectatic aneurysms appeared more benign. Of patients presenting asymptomatically, 66% had a dolichoectatic-type aneurysm. Asymptomatic patients did not often become symptomatic, developing stroke caused by the aneurysm in 7.8% of cases, mass effect caused by aneurysm in 3.1% of cases, and hemorrhage in 1.6%. In their description of nonatheromatous fusiform cerebral aneurysms, Mizutani and colleagues reported the natural history of 6 vertebrobasilar dolichoectatic lesions (Mizutani type 2). Similar to the findings by Flemming and colleagues, all dolichoectatic vertebrobasilar aneurysms reported by Mizutani and colleagues were incidentally discovered with no significant neurologic sequelae in 2 to 5 years of follow-up. Both Mizutani and colleagues and Flemming and colleagues reported no cases of transformation of a dolichoectactic aneurysm to a fusiform aneurysm.
Ischemic Stroke
Flemming and colleagues found that the ischemic stroke risk of these lesions after their diagnosis was higher than the hemorrhagic risk. The rate of any cerebral infarction after diagnosis at 1-year, 5-year, and 10-year follow-up was 6.1, 17.3, and 25.4%, respectively. The rate of cerebral infarction related to the aneurysm at 1-year, 5-year, and 10-year follow-up was 2.7%, 11.3%, and 15.9%, respectively. For patients presenting with stroke, the risk of a recurrent ischemic stroke was 6.7%/y at a median time of 1.73 years. Risk factors for stroke in the Mayo Clinic cohort included patients who were initially symptomatic, previous history of ischemia caused by the aneurysm, and a transitional type of aneurysm with HRs of 16.2, 3.88, and 3.3, respectively. The median survival in their cohort was 7.8 years, with death caused by stroke occurring in more than one-third of cases (the most common cause). Recurrent stroke was less common in anticoagulated patients and more common in patients treated with antiplatelets. Recurrent stroke was not affected by thrombus presence, enlargement over time, size, hypertension, hyperlipidemia, or gender.
The risk of recurrent stroke was also seen in the series of dolichoectatic vertebrobasilar arteries reported by Milandre and colleagues. Of 21 patients presenting with stroke or compressive signs and symptoms, 8 (38%) had an ischemic event after presentation, and this was more common in patients initially presenting with stroke. Ince and colleagues evaluated all patients with stroke presenting to the Mayo Clinic from 1985 to 1989 and found that dolichoectasia of any distribution had a relative risk of 2.4 for recurrent stroke compared with patients without dolichoectasia. The 2 patient groups had no difference in age, gender, or medical comorbidities. Eighty-three percent of these patients at least had some involvement of the vertebrobasilar system.
In a report of 23 cases with dolichoectatic basilar arteries defined as the basilar bifurcation at least 21 mm above the dorsum sellae, Nishizaki and colleagues managed 19 cases conservatively. These investigators did not discern between dolichoectatic-type or fusiform-type aneurysms. Two of the 7 patients presenting with pontine infarction died of medical complications. The other 5 patients were treated conservatively, with stable signs and symptoms at follow-up. The natural history after the ictus described in this smaller series is seemingly more benign than that described by Flemming and colleagues.
Compressive Symptoms
In the series by Flemming and colleagues, 35 patients (22% of the cohort) presented with symptoms caused by compression. Compressive symptoms included brainstem compression, hydrocephalus, or cranial neuropathies. Of these 35 patients, 77% had mild or no disability at presentation. At 1 year, only 46% had mild or no disability, 25% moderate, and 25% severe disability or were dead. At 5 years, only 18% remained with mild or no disability, and 50% of the patients had severe disability or death. In addition, 12 patients in the cohort (7.5%) who did not initially present with compressive symptoms developed compressive symptoms.
Compressive symptoms correlate with growth of the aneurysm. Fifty-two of 159 patients in the Mayo Clinic cohort had serial imaging. Twenty-five patients (48%) had radiographic documentation of aneurysm growth with cross-sectional enlargement of 1.3 mm per year. Lesion growth was statistically associated with symptomatic compression at initial diagnosis, a transitional or fusiform type of aneurysm, and a larger diameter at diagnosis (15 vs 8 mm). These investigators also found more intramural hemorrhage in the enlarging aneurysms by T1 MRI compared with aneurysms that did not enlarge, but this was not statistically significant. The 5-year mortality of growing aneurysms was 56.5% versus 3.7% for aneurysms that were not growing.
Nishizaki and colleagues followed 7 patients with compressive symptoms. These symptoms included 4 brainstem compression symptoms, 1 vertigo, 1 dysarthria and cerebellar compression, and 1 with cranial nerve 9–12 palsy. The patient with cerebellar compression and vertigo improved with medical therapy. The remaining patients remained stable in their signs and symptoms. Mizutani and colleagues reported the natural history of 8 Mizutani type 3 vertebrobasilar aneurysms (consistent with a fusiform or translational aneurysm defined by Flemming and colleagues), with follow-up ranging from 1 to 5 years. All patients initially presented with brainstem compression. Six aneurysms (75%) grew over the follow-up period. This finding is consistent with a 71% growth rate of fusiform and transitional aneurysms seen by Flemming and colleagues. One patient (18%) died of progressive brainstem compression.
Hemorrhage
In the 12-year Mayo Clinic experience, the annual rupture risk was 0.9% for all fusiform vertebrobasilar aneurysms. The annual risk was 2.3% for fusiform and transitional aneurysms but only 0.4% for dolichoectatic aneurysms. This risk was increased with aneurysm enlargement (4 of 6 ruptures had documented growth). Flemming and colleagues also quoted the risk of rupture being increased with aneurysm diameter, because all cases that ruptured with imaging were greater than 10 mm, whereas only 34% of unruptured cases were greater than 10 mm. The exception to this theory was very large aneurysms (30–40 mm), which always had thrombus and never presented with rupture. Of the 8 patients with fusiform basilar aneurysms followed by Mizutani and colleagues for 1 to 5 years, 3 patients (38%) had fatal SAH.
Fusiform cerebral aneurysms have a not insignificant hemorrhage rate of 2.3% per year. The natural history of unruptured symptomatic lesions is for the presenting symptoms to continue to worsen, whether this is compressive or ischemic in nature. However, recurrent ischemic events may be reduced with anticoagulation more than with antiplatelet therapy. The natural history for asymptomatic dolichoectatic lesions, conversely, is relatively benign.
Related posts:
General Technical Considerations for the Endovascular Management of Cerebral Aneurysms
Pitfalls and Complications Management in the Endovascular Treatment of Aneurysms
Endovascular Treatment of Basilar Aneurysms
Cerebral Vasospasm
Endovascular Management and Treatment of Acute Ischemic Stroke
Endovascular Management of Intracranial Atherosclerosis
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
