- Email: [email protected]
Unruptured intracranial aneurysm as a cause of cerebral ischemia
Clinical Neurology and Neurosurgery 113 (2011) 28–33
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Unruptured intracranial aneurysm as a cause of cerebral ischemia L. Calviere a,∗ , A. Viguier a , N.A. Da Silva Jr. b , C. Cognard b , V. Larrue a a b
Department of Neurology, Toulouse University, Toulouse, France Department of Neuroradiology, Toulouse University, Toulouse, France
a r t i c l e
i n f o
Article history: Received 9 September 2009 Received in revised form 19 August 2010 Accepted 28 August 2010
Keywords: Aneurysm thrombosis Ischemic stroke Unruptured intracranial aneurysm
a b s t r a c t Background: Unruptured intracranial artery aneurysms (IAs) can be revealed by cerebral ischemia. Little is known on the clinical course and outcome of patients with this condition. We report our findings in a consecutive series of 15 such patients. Methods: We retrospectively analyzed patients with ischemic stroke (IS) or transient ischemic attack (TIA), unruptured IA on the symptomatic cerebral artery, and no other potential cause of cerebral ischemia consecutively treated in a tertiary stroke unit. Results: Fifteen patients (ten women, and five men) were identified. Their mean age was 49.7 years (range, 37–80 years). Ten patients presented with IS, and five with TIA. The median diameter of IA was 7.5 mm (range, 2.5–23 mm). Aneurysm thrombosis was found on imaging in 9/10 patient with IS, and 1/5 patients with TIA (p = 0.017). Thirteen patients were given an antiplatelet agent. Mean follow-up until last visit or treatment of aneurysm was 393 days (median 182 days; range, 6–1825 days). There was no ischemic recurrence. Partial or complete recanalization of aneurysm thrombosis occurred in 7/10 patients. Two patients, both with initial aneurysmal thrombosis and on antiplatelet therapy, experienced aneurysm rupture. Conclusion: Unruptured IA is a rare cause of IS/TIA. IS is associated with aneurysm thrombosis. Our findings suggest that aneurysm thrombosis is a dynamic process which is associated with a low rate of ischemic recurrence on antiplatelet therapy but may be followed by subarachnoid hemorrhage. © 2010 Elsevier B.V. All rights reserved.
1. Introduction An unruptured intracranial artery aneurysm (IA) is a rare cause of ischemic stroke (IS) or transient ischemic attack (TIA). Published case series have included small numbers of patients [1–7]. Therefore the clinical and radiological characteristics of patients with this condition, and the risks of ischemic recurrence and aneurysm rupture are not well known. We report a consecutive series of 15 patients with IS or TIA with unexplained etiology except the finding of an unruptured IA on the symptomatic artery. We assessed their clinical and radiological characteristics, the mechanisms of IS/TIA, and the risks of recurrent IS/TIA and secondary subarachnoid hemorrhage. 2. Patients and methods We retrospectively analyzed patients with IS or TIA, an unruptured IA of the symptomatic cerebral artery, and no other potential cause of stroke/TIA consecutively cared for in a tertiary stroke unit from September 1999 to September 2007. IS
∗ Corresponding author at: Service de Neurologie vasculaire CHU ToulouseRangueil, 1 avenue Jean Poulhes, 31403 Toulouse Cedex, France. Tel.: +33 610625533/561322641; fax: +33 561322643. E-mail address: [email protected] (L. Calviere). 0303-8467/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.clineuro.2010.08.016
was defined as a sudden neurological deficit and a corresponding ischemic lesion on MRI. TIA was defined as a sudden neurological deficit of short duration without evidence of brain infarction on MRI. Aneurysms were diagnosed on MRI and MR angiography or CT angiography performed for IS/TIA evaluation. Subarachnoid hemorrhage was excluded by MRI, and in patients with headache by cerebrospinal fluid analysis including chromoprotein analysis. 2.1. Imaging protocol All patients underwent magnetic resonance imaging (MRI) at baseline. MRI was performed on a 1.5 T scanner (Gyroscan NT Intera; Philips, Best, the Netherlands). The examination protocol included T1 and T2 weighed turbo-spin-echo, FLAIR, diffusion-weighed images, and T2 gradient-echo sequences. Analysis of intracranial arteries was performed with 3D timeof-flight (TOF) magnetic resonance angiography (MRA). Cervical arteries were studied with 3D-gadolinium-enhanced MR angiography sequences. Digital subtracted angiography (DSA) was also performed at baseline in 14 patients. 2.2. Analysis of aneurysm thrombosis and recanalization The diagnosis of aneurysm thrombosis/recanalization was made by a combined analysis of brain MRI, MRA, and DSA. The analysis of
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
29
Fig. 1. Patient 1. (A) Initial cerebral MRI, T2 weighed imaging. Brain infarction in the superficial MCA territory (star). Aneurysm of the MCA bifurcation (7 mm in diameter) with heterogeneous signal (arrow). (B) Conventional angiography of the left internal carotid artery performed on the same day. The aneurysm was not injected confirming thrombosis. (C) Conventional angiography of the left internal carotid artery performed 3 months later showing complete recanalization of aneurysm (discontinued arrow).
thrombosis was performed by the same neuroradiologist (DS NA) blinded to the clinical characteristics. Aneurysm thrombosis on baseline MRI was defined as an area of high or intermediate signal intensity within the aneurysm sac on T1 weighed imaging, with variable signal intensity on T2 imaging depending on the age of thrombus, and low intensity on T2 gradient-echo imaging (Figs. 1A and 2). Thrombosis was classified as complete or partial. Thrombosis was considered partial when there was an associated area of signal void within the aneurysm sac on T1 imaging. Initial aneurysm thrombosis was also diagnosed when DSA during follow-up showed a significant increase in size of aneurysm suggesting partial thrombosis at baseline with subsequent dissolution of thrombus (Figs. 1B and C and 3). Secondary aneurysm thrombosis during follow-up was diagnosed when repeat MRI showed thrombosis not visible on baseline MRI, or when DSA showed lack of injection or only partial injection of the aneurysm sac as compared with baseline imaging. Recanalization on MRI was diagnosed when the initial area of high or intermediate signal intensity on T1 weighed imaging was replaced by a signal void on repeat MRI. Recanalization on DSA was
defined as an increase in the size of aneurysm as compared with baseline DSA. (Figs. 1C and 3). 2.3. Etiological investigation of stroke/TIA All patients with IA had an extensive etiological work-up to rule out other potential causes of stroke/TIA. This included brain and cervical MR angiography, carotid duplex, ECG, transesophageal echocardiography, and laboratory investigations (complete blood cell count, C-reactive protein, prothrombin time, activated partial thromboplastin time, serum creatinine, glucose, cholesterol, LDL-cholesterol, and triglycerides levels). We used transcranial Doppler (TCD) continuous monitoring for microembolic signals (MES) detection in the symptomatic middle cerebral artery (MCA) downstream from the aneurysm. Details on the methods for MES diagnosis have been published previously [8,9]. 2.4. Follow-up Patients were followed up in the stroke unit. All patients but one had repeat MRI and MR angiography during follow-up. We
Fig. 2. Patient 15. (A) Cerebral MRI, T2 weighed imaging. Well limited mass on the left MCA bifurcation with a non-circulating heterogeneous signal core corresponding to an aneurysm with complete thrombosis, 23 mm in diameter (arrow). (B) Cerebral MRI, T1 weighed imaging. Well limited mass on left MCA bifurcation, with a non-circulating isosignal core corresponding to an aneurysm with complete thrombosis (discontinued arrow). (C) Cerebral MRI, T2 weighed imaging. Insular infarction downstream from the aneurysm (arrowhead).
30
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
Fig. 3. Patient 6. (A) Cerebral angiography of the right internal carotid artery. Aneurysm of the right MCA bifurcation with defect on the inferior part of the sac and occlusion of the inferior branch corresponding to partial thrombosis. (B) Angiography performed several months later. Increase in size of the aneurysm with disappearance of the defect on the inferior part of the sac corresponding to recanalization. The previously occluded inferior branch was also recanalized.
performed questionnaires by phone in January 2008 to verify new symptomatic events. Duration of clinical follow-up was defined as the time from qualifying IS/TIA to phone call, treatment of aneurysm, or death. Radiological follow-up regarding aneurysm thrombosis was defined as the time from aneurysm thrombosis demonstration (either initial or secondary) to last aneurysm imaging.
3. Results
We found aneurysm thrombosis in 10 patients. Aneurysm thrombosis was present at baseline in 8 patients (Figs. 1A and 2). Secondary aneurysm thrombosis was demonstrated in 2 further patients, 86 days and 94 days after baseline MRI, respectively. Aneurysm thrombosis was partial in 7 patients, and complete in 3 patients. An arterial branch from the aneurysm sac was occluded in 2 patients (Fig. 3). Nine (90%) of the 10 patients with IS and one (20%) of the 5 patients with TIA had baseline or secondary aneurysm thrombosis. The association between aneurysm thrombosis and IS was statistically significant (p = 0.017). All patients with preceding headache had aneurysm thrombosis whereas aneurysm thrombosis was seen in only 3/8 (37.5%) patients without headache (p = 0.026). Seven of the 10 patients with aneurysm thrombosis (70%) had preceding headache, whereas 3/10 presented without headache (p = 0.02).
3.1. Patients (Table 1)
3.3. Monitoring for MES
Fifteen (0.47%) patients fulfilling the inclusion criteria were identified among 3202 patients with IS/TIA treated in our stroke unit during the study period. Their mean age was 49.7 years (range, 37–80 years). There were ten women (mean age, 54 years) and five men (mean age, 47.8 years). Ten patients presented with IS, and five with TIA. All the ischemic lesions were recent (hypersignal on diffusion sequence) and distal to aneurysm setting. Seven patients experienced headache before IS/TIA. The headache preceded the ischemic event by a few hours (4 patients) to several days (3 patients). The onset of headache was progressive. The pain was lateralized to the side of the aneurysm in 5 patients, and diffuse in 2 patients. There was neither photophobia nor phonophobia. The headache persisted several days to weeks after the ischemic event.
TCD continuous monitoring for MES detection was performed in 10 cases. It was not performed in 5 patients for the following reasons: location of the aneurysm on the ACA (2 patients), distal location of the aneurysm on the MCA (2 patients), and insufficient temporal acoustic bone window (1 patient). The median delay between cerebral ischemia and TCD monitoring was 12 days (range, 1–450 days). MES were detected in only 2 patients. Aneurysm thrombosis was seen on initial imaging in both cases.
2.5. Statistical analysis We looked for an association between clinical features and aneurysm thrombosis with a Fisher exact test. The significance level was set at p < 0.05. Statistical analysis was performed with the SPPS 14.0 software.
3.2. Aneurysms and thrombosis (Table 1) Twenty-two aneurysms were disclosed in the 15 patients. Five patients had more than one aneurysm. Seven symptomatic aneurysms were located on the MCA, six on the internal carotid artery, and two on the ACA. The median maximal diameter of the aneurysm was 7.5 mm (range, 2.5–23 mm).
3.4. Treatment Thirteen patients were initially treated with an antiplatelet agent (nine with acetylsalycilic acid, three with clopidogrel, and one with the combination of both). The aneurysm was eventually treated in 7 patients after multidisciplinary assessment (patients 4, 7, 9, 11, 12, and 14). The decision regarding conservative or non-conservative management was based on patient characteristics, and IA morphology and accessibility. Five patients underwent endovascular treatment, and 2 patients surgical clipping. One patient died at day 6 before a decision regarding aneurysm treatment was made. The seven remaining patients were treated conservatively. They were given long term
Table 1 Characteristics of patients and aneurysms. Patient
Presentation
Clinical signs
Preceding headache
Vascular risk factors
Number of aneurysms
Symptomatic aneurysm location
Maximal diameter (mm)
Aneurysm thrombosis
Recanalization
1
45/F
Ischemic stroke
Right hemiparesis, aphasia
No
2
MCA bifurcation
7
Complete initial
Complete
2
45/F
Ischemic stroke
Yes
1
Terminal ICA
16
Partial initial
No
3 4
80/F 47/M
Ischemic stroke Ischemic stroke
Right hemiparesis and hemihypoesthesia aphasia Aphasia Left hemiparesis
Cigarette smoking, oral contraception None
1 1
Terminal ICA Terminal ICA
5.5 10
No Partial initial
No Partial
5 6
52/M 37/M
TIA Ischemic stroke
No Yes
1 1
ACA MCA bifurcation
4 5
Partial initial Partial initial
Complete Complete
7
45/M
Ischemic stroke
1
Terminal ICA
7
Partial secondary
Partial
8
46/F
Ischemic stroke
Transient left leg weakness Left arm paresis, hemianopia, anosognosia, hemineglect Left hemiparesis and hemihypoesthesia Right hemiparesis and hemihypoesthesia
None Cigarette smoking, hypercholesterolemia, alcohol Hypercholesterolemia Cigarette smoking, hypercholesterolemia Cigarette smoking
4
MCA
6
Partial secondary
Partial
9
58/M
TIA
Transient right hemiparesis
No
1
MCA bifurcation
15
No
–
10
37/F
TIA
No
1
MCA bifurcation
2.5
No
–
11
44/F
TIA
No
Cigarette smoking
1
MCA bifurcation
2.5
No
–
12 13
41/F 53/F
TIA Ischemic stroke
No Yes
Cigarette smoking, heredity Hypertension
2 2
ICA/ophthalmic ACA
8 5
No Complete initial
– No
14
58/F
Ischemic stroke
Yes
Obesity, hypercholesterolemia
1
Intracavernous ICA
17
Partial initial
Partial
15
58/F
Ischemic stroke
Transient sensorimotor left sided deficit Four transient spells of right hemiparesis and hemihypoesthesia with aphasia Transient aphasia Left leg paresis, frontal syndrome, Left arm sensory deficit, right third cranial nerve palsy Right arm weakness, aphasia
Cigarette smoking, hypertension, hypercholesterolemia Diabetes, hypercholesterolemia Cigarette smoking
Yes
Hypertension
2
MCA bifurcation
23
Complete initial
No
No Yes
Yes No
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
Age/sex
M: male; F: female; TIA: transient ischemic attack; ACA: anterior cerebral artery; MCA: middle cerebral artery; ICA: internal carotid artery.
31
32
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
antiplatelet therapy and treatments for control of their vascular risk factors. 3.5. Clinical follow-up Mean clinical follow-up was 393 days (median 182 days; range, 6–1825 days) for the whole group, and 184 days (median 125 days; range, 31–514 days) for patients whose aneurysm was excluded. There was no ischemic recurrence during follow-up. Two patients (patient 5 and patient 15) experienced aneurysm rupture, on day 6 and 182, respectively. Both patients with aneurysm rupture had aneurysm thrombosis on baseline MRI. Aneurysm thrombosis was complete in 1 patient (patient 15), and partial in the other patient (patient 5). The initial aneurysm diameter was 23 mm (patient 15), and 4 mm (patient 5). Both patients were on antiplatelet therapy at the time of rupture. One patient (patient 15) died from acute brain herniation. The other patient received endovascular treatment of the ruptured aneurysm (patient 5). 3.6. Radiological follow-up All patients with initial or secondary aneurysm thrombosis, except the patient who died on day 6, had a repeat imaging in a mean delay of 66 days from the first demonstration of aneurysm thrombosis (median 45 days; range, 14–181 days). Partial or complete recanalization of aneurysm thrombosis occurred in 5/9 (55%) patients. Further imaging in a mean delay of 132 days (median 94 days; range, 26–350 days) showed recanalization in 7/9 (77.7%) patients. Recanalization was partial in 4 patients and complete in 3 patients (Table 1). 4. Discussion We report a series of 15 patients with unruptured IA as the sole potential cause of stroke/TIA. Most patients with IS had evidence of aneurysm thrombosis. No patient experienced ischemic recurrence during follow-up. In contrast secondary aneurysm rupture occurred in 2/15 patients. Unruptured IAs are not exceptional findings in patients with IS. Among 258 consecutive patients with acute IS who underwent CT or MR angiography, 17 (6.6%) were found to have IAs [10]. This prevalence is approximately three times that in the general population [11,12]. Nevertheless, unruptured IAs in patients with IS are usually regarded as incidental findings. Their higher prevalence in this setting may be explained by the fact that hypertension and cigarette smoking are risk factors for both atherosclerosis and subarachnoid hemorrhage [11,13]. Rarely, however, unruptured IAs may have a causal relationship with stroke. In the prospective series reported by Oh et al. the unruptured IA was regarded as the possible cause of stroke in only 1 (0.4%) patient [10]. Using MRA in 118 patients with IS, Qureshi et al. found only two (1.7%) patients with unruptured IA in the same arterial territory as stroke [14]. In another series of 500 patients with TIA, three (0.6%) patients had an IA possibly explaining the ischemic event [5]. Possible mechanisms of stroke related to unruptured IA include aneurysm thrombosis with distal emboli or extension of thrombosis to the parent artery, and compression of the parent artery by a large aneurysm [2,15–18]. Patients were included in the present study if they had an aneurysm on the symptomatic cerebral artery, no evidence of subarachnoid hemorrhage, and no other potential cause of IS/TIA despite an extensive etiological work-up. Our findings confirm the rarity of unruptured IA as a potential cause of IS/TIA because we could retrieve only 15 (0.47%) such cases among 3202 patients included in our stroke database over a period of 8 years.
Distal emboli were the likely mechanism of IS/TIA in most patients from our study. Extension of aneurysm thrombus into the artery lumen was demonstrated in only 2 patients. Aneurysms were of relatively small size and did not exert compression on the parent artery. We found aneurysm thrombosis at baseline or during followup in 10/15 patients. Almost all patients with IS had evidence of aneurysm thrombosis with a statistical significant association. This result suggests that the demonstration of the aneurysm thrombosis is an important step to affirm the causal relationship between IS and unruptured IA. It was one of the criteria proposed by Cohen et al. [3]. In contrast, aneurysm thrombosis was uncommon in patients with TIA. This may suggest that unruptured IA in patients with TIA could be an incidental finding. Alternatively, small or evanescent thrombi presumably associated with TIA may have been missed by MRI. Headache preceded IS/TIA by a few hours to several days in nearly half of cases. Pain onset was not as explosive as in aneurysm rupture but rather progressive. Subarachnoid hemorrhage was excluded by cerebrospinal fluid analysis. All patients with headache had evidence of aneurysm thrombosis on imaging with a significant statistical association, suggesting that headache might be related to the thrombotic process. Thus, we believe that unruptured IA should be considered in patients with IS/TIA and preceding headache. Aneurysm thrombosis showed a trend toward regression over time. Regression was, however, a slow process resulting in complete recanalization in only 3/10 patients after a median follow-up of 94 days. Our findings suggest also an initial phase of thrombus instability may exist because 2 patients without evidence of thrombus at baseline had aneurysm thrombosis on the first repeat imaging, then regression of thrombus on further follow-up imaging. Aneurysm thrombosis is a common finding in giant cerebral aneurysms [16,19]. In the Italian cooperative study on giant intracranial aneurysm including 240 patients, aneurysm thrombosis was noted in 48% of aneurysms with diameters between 20 and 25 mm, and in 76% of larger aneurysms [20]. Despite this high prevalence of aneurysm thrombosis, unruptured giant aneurysms are rarely responsible for cerebral ischemia [16]. In previous studies reporting on unruptured IAs associated with IS/TIA, the majority of aneurysms were of small size [5–7,21–23]. Our study confirms this fact since the median diameter of aneurysm was only 7.5 mm and no patient had giant aneurysm. Factors that may contribute to aneurysm thrombosis include blood stagnation and endothelial injury due to turbulent blood flow within the aneurysm [24,25]. Associated medical conditions that induce increased coagulability may occasionally play a role. Characteristics of blood flow within the aneurysm may explain the dynamics of aneurysmal thrombosis noted in our study with an early phase of thrombus instability followed by a trend toward regression. Endothelial injury may initiate platelet deposition, leukocyte–platelet interaction, and thrombogenesis. Higher intraaneurysmal blood flow velocities in relatively small aneurysms may facilitate both embolization and lysis of thrombus formed within the aneurysm. In contrast slow flow in giant aneurysms may protect against embolization but also preclude thrombolysis. Previously published studies have shown a low risk of ischemic recurrence among patients treated conservatively [6]. In good agreement, we did not observe any ischemic recurrence during follow-up. Furthermore, asymptomatic MES in the MCA downstream from the aneurysm were rarely found in our study (2 of 10 patients). MES are detected in approximately 40% of patients with recently symptomatic carotid artery atherosclerosis. In this setting, MES are a strong risk factor for recurrent stroke [9]. The relative rarity of MES in patients with aneurysm thrombosis may thus indicate a low risk of recurrent stroke. However, the low rate
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
of patients with MES may be also explained by some methodological reasons including the delay between the ischemic event and TCD monitoring, and the relatively short duration of monitoring (30 min). In addition, most of our patients were treated with an antiplatelet agent which may have reduced both the rate of MES and recurrent IS/TIA [9]. Two (13.3%) patients from our series experienced aneurysm rupture during follow-up. Both had initial aneurysm thrombosis and were on antiplatelet therapy at the time of rupture. Rupture occurred early in 1 patient, 6 days after the initial ischemic event, and on day 182 in the other patient. Cases of thrombosed aneurysm causing brain infarction, and followed by subarachnoid hemorrhage have been rarely reported [26]. No aneurysm rupture was seen in two previously published series including nine and 12 patients, respectively [6]. Aneurysm thrombosis may increase the risk of rupture through the recruitment of neutrophils and secretion of various proteolytic enzymes weakening the aneurysm wall [27–30]. The instability of aneurysm thrombosis that we have documented may also have played a role. Possibly, antiplatelet therapy facilitated the rupture [24,25], although the role of antiplatelet therapy as a risk factor for subarachnoid hemorrhage is uncertain [13]. There is no consensus on the management of patients with unruptured IA revealed by cerebral ischemia [5,6,24,25]. Based on previous studies which showed a low risk of recurrent stroke and rupture in patients treated with antiplatelet agents, a conservative approach has been often recommended, at least for patients with small aneurysms [5–7,31]. Our findings on the instability of aneurysm thrombosis and the occurrence of aneurysm rupture in patients with initial aneurysm thrombosis suggest that treatment of unruptured IAs revealed by cerebral ischemic events may be warranted. There are several limitations to our study mainly due to its retrospective design. We could not provide a reliable estimate of the prevalence of unruptured intracranial aneurysms because some patients with stroke/TIA evaluated at our institution during the study period only had transcranial color-coded sonography instead of CT or MR angiography to explore their intracranial vessels. Some transient and minor clinical events during follow-up may have been ignored, thus underestimating the true incidence of recurrent TIA. Indications for antiplatelet therapy and aneurysm treatment were not predefined. In addition, although our case series was relatively large compared with those previously published on the same subject, the sample size was small. Also, the duration of follow-up was relatively short, precluding any conclusion on the risk of ischemic recurrence or aneurysm rupture in the long term. 5. Conclusion Unruptured IA is a rare cause of IS/TIA. IS is associated with aneurysm thrombosis. Our findings suggest that aneurysm thrombosis is a dynamic process associated with a low rate of ischemic recurrence on antiplatelet therapy but that may be followed by subarachnoid hemorrhage. These lesions should be controlled for recanalization and should be considered for treatment whenever they are amenable to endovascular or neurosurgical treatment, with respect to patient considerations. References [1] Antunes JL, Correll JW. Cerebral emboli from intracranial aneurysms. Surg Neurol 1976;6(1):7–10. [2] Fisher M, Davidson RI, Marcus EM. Transient focal cerebral ischemia as a presenting manifestation of unruptured cerebral aneurysms. Ann Neurol 1980;8(4):367–72.
33
[3] Cohen MM, Hemalatha CP, D’Addario RT, Goldman HW. Embolization from a fusiform middle cerebral artery aneurysm. Stroke 1980;11(2):158– 61. [4] Fukuoka S, Suematsu K, Nakamura J, Matsuzaki T, Satoh S, Hashimoto I. Transient ischemic attacks caused by unruptured intracranial aneurysm. Surg Neurol 1982;17(6):464–7. [5] Przelomski MM, Fisher M, Davidson RI, Jones HR, Marcus EM. Unruptured intracranial aneurysm and transient focal cerebral ischemia: a follow-up study. Neurology 1986;36(4):584–7. [6] Qureshi AI, Mohammad Y, Yahia AM, Luft AR, Sharma M, Tamargo RJ, et al. Ischemic events associated with unruptured intracranial aneurysms: multicenter clinical study and review of the literature. Neurosurgery 2000;46(2):282–9 [discussion 289–90]. [7] Nanda A, Vannemreddy PS. Cerebral ischemia as a presenting feature of intracranial aneurysms: a negative prognostic indicator in the management of aneurysms. Neurosurgery 2006;58(5):831–7 [discussion 831–37]. [8] Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral embolic signals in patients with carotid stenosis. Correlation with appearance of plaque ulceration on angiography. Stroke 1995;26(5):813–5. [9] Markus HS, Droste DW, Kaps M, Larrue V, Lees KR, Siebler M, et al. Dual antiplatelet therapy with clopidogrel and aspirin in symptomatic carotid stenosis evaluated using Doppler embolic signal detection: the Clopidogrel and Aspirin for Reduction of Emboli in Symptomatic Carotid Stenosis (CARESS) trial. Circulation 2005;111(17):2233–40. [10] Oh YS, Lee SJ, Shon YM, Yang DW, Kim BS, Cho AH. Incidental unruptured intracranial aneurysms in patients with acute ischemic stroke. Cerebrovasc Dis 2008;26(6):650–3. [11] Rinkel GJ, Djibuti M, Algra A, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms: a systematic review. Stroke 1998;29(1): 251–6. [12] Vernooij MW, Ikram MA, Tanghe HL, Vincent AJ, Hofman A, Krestin GP, et al. Incidental findings on brain MRI in the general population. N Engl J Med 2007;357(18):1821–8. [13] Rinkel GJ. Natural history, epidemiology and screening of unruptured intracranial aneurysms. J Neuroradiol 2008;35(2):99–103. [14] Qureshi AI, Isa A, Cinnamon J, Fountain J, Ottenlips JR, Braimah J, et al. Magnetic resonance angiography in patients with brain infarction. J Neuroimaging 1998;8(2):65–70. [15] Brownlee RD, Tranmer BI, Sevick RJ, Karmy G, Curry BJ. Spontaneous thrombosis of an unruptured anterior communicating artery aneurysm. An unusual cause of ischemic stroke. Stroke 1995;26(10):1945–9. [16] Whittle IR, Dorsch NW, Besser M. Spontaneous thrombosis in giant intracranial aneurysms. J Neurol Neurosurg Psychiatry 1982;45(11):1040–7. [17] Hoffman WF, Wilson CB, Townsend JJ. Recurrent transient ischemic attacks secondary to an embolizing saccular middle cerebral artery aneurysm. Case report. J Neurosurg 1979;51(1):103–6. [18] Schaller B, Lyrer P. Focal neurological deficits following spontaneous thrombosis of unruptured giant aneurysms. Eur Neurol 2002;47(3):175–82. [19] Schunk H. Spontaneous thrombosis of intracranial aneurysms. Am J Roentgenol Radium Ther Nucl Med 1964;91:1327–38. [20] Rosta L, Battaglia R, Pasqualin A, Beltramello A. Italian cooperative study on giant intracranial aneurysms. 2. Radiological data. Acta Neurochir Suppl (Wien) 1988;42:53–9. [21] Friedman JA, Piepgras DG, Pichelmann MA, Hansen KK, Brown Jr RD, Wiebers DO. Small cerebral aneurysms presenting with symptoms other than rupture. Neurology 2001;57(7):1212–6. [22] Cohen JE, Itshayek E, Gomori JM, Grigoriadis S, Raphaeli G, Spektor S, et al. Spontaneous thrombosis of cerebral aneurysms presenting with ischemic stroke. J Neurol Sci 2007;254(1–2):95–8. [23] McLaughlin N, Bojanowski MW. Unruptured cerebral aneurysms presenting with ischemic events. Can J Neurol Sci 2008;35(5):588–92. [24] Sutherland GR, King ME, Peerless SJ, Vezina WC, Brown GW, Chamberlain MJ. Platelet interaction within giant intracranial aneurysms. J Neurosurg 1982;56(1):53–61. [25] Ohno K, Suzuki R, Masaoka H, Matsushima Y, Monma S, Inaba Y. Unruptured aneurysms in patients with transient ischemic attack or reversible ischemic neurological deficit. Report of eight cases. Clin Neurol Neurosurg 1989;91(3):229–33. [26] Kato M, Kaku Y, Okumura A, Iwama T, Sakai N. Thrombosed unruptured cerebral aneurysm causing brain infarction followed by subarachnoid hemorrhage – case report. Neurol Med Chir (Tokyo) 2005;45(9):472–5. [27] Bruno G, Todor R, Lewis I, Chyatte D. Vascular extracellular matrix remodeling in cerebral aneurysms. J Neurosurg 1998;89(3):431–40. [28] Fontaine V, Jacob MP, Houard X, Rossignol P, Plissonnier D, Angles-Cano E, et al. Involvement of the mural thrombus as a site of protease release and activation in human aortic aneurysms. Am J Pathol 2002;161(5):1701–10. [29] Carrell TW, Burnand KG, Booth NA, Humphries J, Smith A. Intraluminal thrombus enhances proteolysis in abdominal aortic aneurysms. Vascular 2006;14(1):9–16. [30] Jin D, Sheng J, Yang X, Gao B. Matrix metalloproteinases and tissue inhibitors of metalloproteinases expression in human cerebral ruptured and unruptured aneurysm. Surg Neurol 2007;68(Suppl. 2):S11–6 [discussion S16]. [31] Nagashima M, Nemoto M, Hadeishi H, Suzuki A, Yasui N. Unruptured aneurysms associated with ischaemic cerebrovascular diseases. Surgical indication. Acta Neurochir (Wien) 1993;124(2–4):71–8.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Unruptured intracranial aneurysm as a cause of cerebral ischemia L. Calviere a,∗ , A. Viguier a , N.A. Da Silva Jr. b , C. Cognard b , V. Larrue a a b
Department of Neurology, Toulouse University, Toulouse, France Department of Neuroradiology, Toulouse University, Toulouse, France
a r t i c l e
i n f o
Article history: Received 9 September 2009 Received in revised form 19 August 2010 Accepted 28 August 2010
Keywords: Aneurysm thrombosis Ischemic stroke Unruptured intracranial aneurysm
a b s t r a c t Background: Unruptured intracranial artery aneurysms (IAs) can be revealed by cerebral ischemia. Little is known on the clinical course and outcome of patients with this condition. We report our findings in a consecutive series of 15 such patients. Methods: We retrospectively analyzed patients with ischemic stroke (IS) or transient ischemic attack (TIA), unruptured IA on the symptomatic cerebral artery, and no other potential cause of cerebral ischemia consecutively treated in a tertiary stroke unit. Results: Fifteen patients (ten women, and five men) were identified. Their mean age was 49.7 years (range, 37–80 years). Ten patients presented with IS, and five with TIA. The median diameter of IA was 7.5 mm (range, 2.5–23 mm). Aneurysm thrombosis was found on imaging in 9/10 patient with IS, and 1/5 patients with TIA (p = 0.017). Thirteen patients were given an antiplatelet agent. Mean follow-up until last visit or treatment of aneurysm was 393 days (median 182 days; range, 6–1825 days). There was no ischemic recurrence. Partial or complete recanalization of aneurysm thrombosis occurred in 7/10 patients. Two patients, both with initial aneurysmal thrombosis and on antiplatelet therapy, experienced aneurysm rupture. Conclusion: Unruptured IA is a rare cause of IS/TIA. IS is associated with aneurysm thrombosis. Our findings suggest that aneurysm thrombosis is a dynamic process which is associated with a low rate of ischemic recurrence on antiplatelet therapy but may be followed by subarachnoid hemorrhage. © 2010 Elsevier B.V. All rights reserved.
1. Introduction An unruptured intracranial artery aneurysm (IA) is a rare cause of ischemic stroke (IS) or transient ischemic attack (TIA). Published case series have included small numbers of patients [1–7]. Therefore the clinical and radiological characteristics of patients with this condition, and the risks of ischemic recurrence and aneurysm rupture are not well known. We report a consecutive series of 15 patients with IS or TIA with unexplained etiology except the finding of an unruptured IA on the symptomatic artery. We assessed their clinical and radiological characteristics, the mechanisms of IS/TIA, and the risks of recurrent IS/TIA and secondary subarachnoid hemorrhage. 2. Patients and methods We retrospectively analyzed patients with IS or TIA, an unruptured IA of the symptomatic cerebral artery, and no other potential cause of stroke/TIA consecutively cared for in a tertiary stroke unit from September 1999 to September 2007. IS
∗ Corresponding author at: Service de Neurologie vasculaire CHU ToulouseRangueil, 1 avenue Jean Poulhes, 31403 Toulouse Cedex, France. Tel.: +33 610625533/561322641; fax: +33 561322643. E-mail address: [email protected] (L. Calviere). 0303-8467/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.clineuro.2010.08.016
was defined as a sudden neurological deficit and a corresponding ischemic lesion on MRI. TIA was defined as a sudden neurological deficit of short duration without evidence of brain infarction on MRI. Aneurysms were diagnosed on MRI and MR angiography or CT angiography performed for IS/TIA evaluation. Subarachnoid hemorrhage was excluded by MRI, and in patients with headache by cerebrospinal fluid analysis including chromoprotein analysis. 2.1. Imaging protocol All patients underwent magnetic resonance imaging (MRI) at baseline. MRI was performed on a 1.5 T scanner (Gyroscan NT Intera; Philips, Best, the Netherlands). The examination protocol included T1 and T2 weighed turbo-spin-echo, FLAIR, diffusion-weighed images, and T2 gradient-echo sequences. Analysis of intracranial arteries was performed with 3D timeof-flight (TOF) magnetic resonance angiography (MRA). Cervical arteries were studied with 3D-gadolinium-enhanced MR angiography sequences. Digital subtracted angiography (DSA) was also performed at baseline in 14 patients. 2.2. Analysis of aneurysm thrombosis and recanalization The diagnosis of aneurysm thrombosis/recanalization was made by a combined analysis of brain MRI, MRA, and DSA. The analysis of
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
29
Fig. 1. Patient 1. (A) Initial cerebral MRI, T2 weighed imaging. Brain infarction in the superficial MCA territory (star). Aneurysm of the MCA bifurcation (7 mm in diameter) with heterogeneous signal (arrow). (B) Conventional angiography of the left internal carotid artery performed on the same day. The aneurysm was not injected confirming thrombosis. (C) Conventional angiography of the left internal carotid artery performed 3 months later showing complete recanalization of aneurysm (discontinued arrow).
thrombosis was performed by the same neuroradiologist (DS NA) blinded to the clinical characteristics. Aneurysm thrombosis on baseline MRI was defined as an area of high or intermediate signal intensity within the aneurysm sac on T1 weighed imaging, with variable signal intensity on T2 imaging depending on the age of thrombus, and low intensity on T2 gradient-echo imaging (Figs. 1A and 2). Thrombosis was classified as complete or partial. Thrombosis was considered partial when there was an associated area of signal void within the aneurysm sac on T1 imaging. Initial aneurysm thrombosis was also diagnosed when DSA during follow-up showed a significant increase in size of aneurysm suggesting partial thrombosis at baseline with subsequent dissolution of thrombus (Figs. 1B and C and 3). Secondary aneurysm thrombosis during follow-up was diagnosed when repeat MRI showed thrombosis not visible on baseline MRI, or when DSA showed lack of injection or only partial injection of the aneurysm sac as compared with baseline imaging. Recanalization on MRI was diagnosed when the initial area of high or intermediate signal intensity on T1 weighed imaging was replaced by a signal void on repeat MRI. Recanalization on DSA was
defined as an increase in the size of aneurysm as compared with baseline DSA. (Figs. 1C and 3). 2.3. Etiological investigation of stroke/TIA All patients with IA had an extensive etiological work-up to rule out other potential causes of stroke/TIA. This included brain and cervical MR angiography, carotid duplex, ECG, transesophageal echocardiography, and laboratory investigations (complete blood cell count, C-reactive protein, prothrombin time, activated partial thromboplastin time, serum creatinine, glucose, cholesterol, LDL-cholesterol, and triglycerides levels). We used transcranial Doppler (TCD) continuous monitoring for microembolic signals (MES) detection in the symptomatic middle cerebral artery (MCA) downstream from the aneurysm. Details on the methods for MES diagnosis have been published previously [8,9]. 2.4. Follow-up Patients were followed up in the stroke unit. All patients but one had repeat MRI and MR angiography during follow-up. We
Fig. 2. Patient 15. (A) Cerebral MRI, T2 weighed imaging. Well limited mass on the left MCA bifurcation with a non-circulating heterogeneous signal core corresponding to an aneurysm with complete thrombosis, 23 mm in diameter (arrow). (B) Cerebral MRI, T1 weighed imaging. Well limited mass on left MCA bifurcation, with a non-circulating isosignal core corresponding to an aneurysm with complete thrombosis (discontinued arrow). (C) Cerebral MRI, T2 weighed imaging. Insular infarction downstream from the aneurysm (arrowhead).
30
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
Fig. 3. Patient 6. (A) Cerebral angiography of the right internal carotid artery. Aneurysm of the right MCA bifurcation with defect on the inferior part of the sac and occlusion of the inferior branch corresponding to partial thrombosis. (B) Angiography performed several months later. Increase in size of the aneurysm with disappearance of the defect on the inferior part of the sac corresponding to recanalization. The previously occluded inferior branch was also recanalized.
performed questionnaires by phone in January 2008 to verify new symptomatic events. Duration of clinical follow-up was defined as the time from qualifying IS/TIA to phone call, treatment of aneurysm, or death. Radiological follow-up regarding aneurysm thrombosis was defined as the time from aneurysm thrombosis demonstration (either initial or secondary) to last aneurysm imaging.
3. Results
We found aneurysm thrombosis in 10 patients. Aneurysm thrombosis was present at baseline in 8 patients (Figs. 1A and 2). Secondary aneurysm thrombosis was demonstrated in 2 further patients, 86 days and 94 days after baseline MRI, respectively. Aneurysm thrombosis was partial in 7 patients, and complete in 3 patients. An arterial branch from the aneurysm sac was occluded in 2 patients (Fig. 3). Nine (90%) of the 10 patients with IS and one (20%) of the 5 patients with TIA had baseline or secondary aneurysm thrombosis. The association between aneurysm thrombosis and IS was statistically significant (p = 0.017). All patients with preceding headache had aneurysm thrombosis whereas aneurysm thrombosis was seen in only 3/8 (37.5%) patients without headache (p = 0.026). Seven of the 10 patients with aneurysm thrombosis (70%) had preceding headache, whereas 3/10 presented without headache (p = 0.02).
3.1. Patients (Table 1)
3.3. Monitoring for MES
Fifteen (0.47%) patients fulfilling the inclusion criteria were identified among 3202 patients with IS/TIA treated in our stroke unit during the study period. Their mean age was 49.7 years (range, 37–80 years). There were ten women (mean age, 54 years) and five men (mean age, 47.8 years). Ten patients presented with IS, and five with TIA. All the ischemic lesions were recent (hypersignal on diffusion sequence) and distal to aneurysm setting. Seven patients experienced headache before IS/TIA. The headache preceded the ischemic event by a few hours (4 patients) to several days (3 patients). The onset of headache was progressive. The pain was lateralized to the side of the aneurysm in 5 patients, and diffuse in 2 patients. There was neither photophobia nor phonophobia. The headache persisted several days to weeks after the ischemic event.
TCD continuous monitoring for MES detection was performed in 10 cases. It was not performed in 5 patients for the following reasons: location of the aneurysm on the ACA (2 patients), distal location of the aneurysm on the MCA (2 patients), and insufficient temporal acoustic bone window (1 patient). The median delay between cerebral ischemia and TCD monitoring was 12 days (range, 1–450 days). MES were detected in only 2 patients. Aneurysm thrombosis was seen on initial imaging in both cases.
2.5. Statistical analysis We looked for an association between clinical features and aneurysm thrombosis with a Fisher exact test. The significance level was set at p < 0.05. Statistical analysis was performed with the SPPS 14.0 software.
3.2. Aneurysms and thrombosis (Table 1) Twenty-two aneurysms were disclosed in the 15 patients. Five patients had more than one aneurysm. Seven symptomatic aneurysms were located on the MCA, six on the internal carotid artery, and two on the ACA. The median maximal diameter of the aneurysm was 7.5 mm (range, 2.5–23 mm).
3.4. Treatment Thirteen patients were initially treated with an antiplatelet agent (nine with acetylsalycilic acid, three with clopidogrel, and one with the combination of both). The aneurysm was eventually treated in 7 patients after multidisciplinary assessment (patients 4, 7, 9, 11, 12, and 14). The decision regarding conservative or non-conservative management was based on patient characteristics, and IA morphology and accessibility. Five patients underwent endovascular treatment, and 2 patients surgical clipping. One patient died at day 6 before a decision regarding aneurysm treatment was made. The seven remaining patients were treated conservatively. They were given long term
Table 1 Characteristics of patients and aneurysms. Patient
Presentation
Clinical signs
Preceding headache
Vascular risk factors
Number of aneurysms
Symptomatic aneurysm location
Maximal diameter (mm)
Aneurysm thrombosis
Recanalization
1
45/F
Ischemic stroke
Right hemiparesis, aphasia
No
2
MCA bifurcation
7
Complete initial
Complete
2
45/F
Ischemic stroke
Yes
1
Terminal ICA
16
Partial initial
No
3 4
80/F 47/M
Ischemic stroke Ischemic stroke
Right hemiparesis and hemihypoesthesia aphasia Aphasia Left hemiparesis
Cigarette smoking, oral contraception None
1 1
Terminal ICA Terminal ICA
5.5 10
No Partial initial
No Partial
5 6
52/M 37/M
TIA Ischemic stroke
No Yes
1 1
ACA MCA bifurcation
4 5
Partial initial Partial initial
Complete Complete
7
45/M
Ischemic stroke
1
Terminal ICA
7
Partial secondary
Partial
8
46/F
Ischemic stroke
Transient left leg weakness Left arm paresis, hemianopia, anosognosia, hemineglect Left hemiparesis and hemihypoesthesia Right hemiparesis and hemihypoesthesia
None Cigarette smoking, hypercholesterolemia, alcohol Hypercholesterolemia Cigarette smoking, hypercholesterolemia Cigarette smoking
4
MCA
6
Partial secondary
Partial
9
58/M
TIA
Transient right hemiparesis
No
1
MCA bifurcation
15
No
–
10
37/F
TIA
No
1
MCA bifurcation
2.5
No
–
11
44/F
TIA
No
Cigarette smoking
1
MCA bifurcation
2.5
No
–
12 13
41/F 53/F
TIA Ischemic stroke
No Yes
Cigarette smoking, heredity Hypertension
2 2
ICA/ophthalmic ACA
8 5
No Complete initial
– No
14
58/F
Ischemic stroke
Yes
Obesity, hypercholesterolemia
1
Intracavernous ICA
17
Partial initial
Partial
15
58/F
Ischemic stroke
Transient sensorimotor left sided deficit Four transient spells of right hemiparesis and hemihypoesthesia with aphasia Transient aphasia Left leg paresis, frontal syndrome, Left arm sensory deficit, right third cranial nerve palsy Right arm weakness, aphasia
Cigarette smoking, hypertension, hypercholesterolemia Diabetes, hypercholesterolemia Cigarette smoking
Yes
Hypertension
2
MCA bifurcation
23
Complete initial
No
No Yes
Yes No
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
Age/sex
M: male; F: female; TIA: transient ischemic attack; ACA: anterior cerebral artery; MCA: middle cerebral artery; ICA: internal carotid artery.
31
32
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
antiplatelet therapy and treatments for control of their vascular risk factors. 3.5. Clinical follow-up Mean clinical follow-up was 393 days (median 182 days; range, 6–1825 days) for the whole group, and 184 days (median 125 days; range, 31–514 days) for patients whose aneurysm was excluded. There was no ischemic recurrence during follow-up. Two patients (patient 5 and patient 15) experienced aneurysm rupture, on day 6 and 182, respectively. Both patients with aneurysm rupture had aneurysm thrombosis on baseline MRI. Aneurysm thrombosis was complete in 1 patient (patient 15), and partial in the other patient (patient 5). The initial aneurysm diameter was 23 mm (patient 15), and 4 mm (patient 5). Both patients were on antiplatelet therapy at the time of rupture. One patient (patient 15) died from acute brain herniation. The other patient received endovascular treatment of the ruptured aneurysm (patient 5). 3.6. Radiological follow-up All patients with initial or secondary aneurysm thrombosis, except the patient who died on day 6, had a repeat imaging in a mean delay of 66 days from the first demonstration of aneurysm thrombosis (median 45 days; range, 14–181 days). Partial or complete recanalization of aneurysm thrombosis occurred in 5/9 (55%) patients. Further imaging in a mean delay of 132 days (median 94 days; range, 26–350 days) showed recanalization in 7/9 (77.7%) patients. Recanalization was partial in 4 patients and complete in 3 patients (Table 1). 4. Discussion We report a series of 15 patients with unruptured IA as the sole potential cause of stroke/TIA. Most patients with IS had evidence of aneurysm thrombosis. No patient experienced ischemic recurrence during follow-up. In contrast secondary aneurysm rupture occurred in 2/15 patients. Unruptured IAs are not exceptional findings in patients with IS. Among 258 consecutive patients with acute IS who underwent CT or MR angiography, 17 (6.6%) were found to have IAs [10]. This prevalence is approximately three times that in the general population [11,12]. Nevertheless, unruptured IAs in patients with IS are usually regarded as incidental findings. Their higher prevalence in this setting may be explained by the fact that hypertension and cigarette smoking are risk factors for both atherosclerosis and subarachnoid hemorrhage [11,13]. Rarely, however, unruptured IAs may have a causal relationship with stroke. In the prospective series reported by Oh et al. the unruptured IA was regarded as the possible cause of stroke in only 1 (0.4%) patient [10]. Using MRA in 118 patients with IS, Qureshi et al. found only two (1.7%) patients with unruptured IA in the same arterial territory as stroke [14]. In another series of 500 patients with TIA, three (0.6%) patients had an IA possibly explaining the ischemic event [5]. Possible mechanisms of stroke related to unruptured IA include aneurysm thrombosis with distal emboli or extension of thrombosis to the parent artery, and compression of the parent artery by a large aneurysm [2,15–18]. Patients were included in the present study if they had an aneurysm on the symptomatic cerebral artery, no evidence of subarachnoid hemorrhage, and no other potential cause of IS/TIA despite an extensive etiological work-up. Our findings confirm the rarity of unruptured IA as a potential cause of IS/TIA because we could retrieve only 15 (0.47%) such cases among 3202 patients included in our stroke database over a period of 8 years.
Distal emboli were the likely mechanism of IS/TIA in most patients from our study. Extension of aneurysm thrombus into the artery lumen was demonstrated in only 2 patients. Aneurysms were of relatively small size and did not exert compression on the parent artery. We found aneurysm thrombosis at baseline or during followup in 10/15 patients. Almost all patients with IS had evidence of aneurysm thrombosis with a statistical significant association. This result suggests that the demonstration of the aneurysm thrombosis is an important step to affirm the causal relationship between IS and unruptured IA. It was one of the criteria proposed by Cohen et al. [3]. In contrast, aneurysm thrombosis was uncommon in patients with TIA. This may suggest that unruptured IA in patients with TIA could be an incidental finding. Alternatively, small or evanescent thrombi presumably associated with TIA may have been missed by MRI. Headache preceded IS/TIA by a few hours to several days in nearly half of cases. Pain onset was not as explosive as in aneurysm rupture but rather progressive. Subarachnoid hemorrhage was excluded by cerebrospinal fluid analysis. All patients with headache had evidence of aneurysm thrombosis on imaging with a significant statistical association, suggesting that headache might be related to the thrombotic process. Thus, we believe that unruptured IA should be considered in patients with IS/TIA and preceding headache. Aneurysm thrombosis showed a trend toward regression over time. Regression was, however, a slow process resulting in complete recanalization in only 3/10 patients after a median follow-up of 94 days. Our findings suggest also an initial phase of thrombus instability may exist because 2 patients without evidence of thrombus at baseline had aneurysm thrombosis on the first repeat imaging, then regression of thrombus on further follow-up imaging. Aneurysm thrombosis is a common finding in giant cerebral aneurysms [16,19]. In the Italian cooperative study on giant intracranial aneurysm including 240 patients, aneurysm thrombosis was noted in 48% of aneurysms with diameters between 20 and 25 mm, and in 76% of larger aneurysms [20]. Despite this high prevalence of aneurysm thrombosis, unruptured giant aneurysms are rarely responsible for cerebral ischemia [16]. In previous studies reporting on unruptured IAs associated with IS/TIA, the majority of aneurysms were of small size [5–7,21–23]. Our study confirms this fact since the median diameter of aneurysm was only 7.5 mm and no patient had giant aneurysm. Factors that may contribute to aneurysm thrombosis include blood stagnation and endothelial injury due to turbulent blood flow within the aneurysm [24,25]. Associated medical conditions that induce increased coagulability may occasionally play a role. Characteristics of blood flow within the aneurysm may explain the dynamics of aneurysmal thrombosis noted in our study with an early phase of thrombus instability followed by a trend toward regression. Endothelial injury may initiate platelet deposition, leukocyte–platelet interaction, and thrombogenesis. Higher intraaneurysmal blood flow velocities in relatively small aneurysms may facilitate both embolization and lysis of thrombus formed within the aneurysm. In contrast slow flow in giant aneurysms may protect against embolization but also preclude thrombolysis. Previously published studies have shown a low risk of ischemic recurrence among patients treated conservatively [6]. In good agreement, we did not observe any ischemic recurrence during follow-up. Furthermore, asymptomatic MES in the MCA downstream from the aneurysm were rarely found in our study (2 of 10 patients). MES are detected in approximately 40% of patients with recently symptomatic carotid artery atherosclerosis. In this setting, MES are a strong risk factor for recurrent stroke [9]. The relative rarity of MES in patients with aneurysm thrombosis may thus indicate a low risk of recurrent stroke. However, the low rate
L. Calviere et al. / Clinical Neurology and Neurosurgery 113 (2011) 28–33
of patients with MES may be also explained by some methodological reasons including the delay between the ischemic event and TCD monitoring, and the relatively short duration of monitoring (30 min). In addition, most of our patients were treated with an antiplatelet agent which may have reduced both the rate of MES and recurrent IS/TIA [9]. Two (13.3%) patients from our series experienced aneurysm rupture during follow-up. Both had initial aneurysm thrombosis and were on antiplatelet therapy at the time of rupture. Rupture occurred early in 1 patient, 6 days after the initial ischemic event, and on day 182 in the other patient. Cases of thrombosed aneurysm causing brain infarction, and followed by subarachnoid hemorrhage have been rarely reported [26]. No aneurysm rupture was seen in two previously published series including nine and 12 patients, respectively [6]. Aneurysm thrombosis may increase the risk of rupture through the recruitment of neutrophils and secretion of various proteolytic enzymes weakening the aneurysm wall [27–30]. The instability of aneurysm thrombosis that we have documented may also have played a role. Possibly, antiplatelet therapy facilitated the rupture [24,25], although the role of antiplatelet therapy as a risk factor for subarachnoid hemorrhage is uncertain [13]. There is no consensus on the management of patients with unruptured IA revealed by cerebral ischemia [5,6,24,25]. Based on previous studies which showed a low risk of recurrent stroke and rupture in patients treated with antiplatelet agents, a conservative approach has been often recommended, at least for patients with small aneurysms [5–7,31]. Our findings on the instability of aneurysm thrombosis and the occurrence of aneurysm rupture in patients with initial aneurysm thrombosis suggest that treatment of unruptured IAs revealed by cerebral ischemic events may be warranted. There are several limitations to our study mainly due to its retrospective design. We could not provide a reliable estimate of the prevalence of unruptured intracranial aneurysms because some patients with stroke/TIA evaluated at our institution during the study period only had transcranial color-coded sonography instead of CT or MR angiography to explore their intracranial vessels. Some transient and minor clinical events during follow-up may have been ignored, thus underestimating the true incidence of recurrent TIA. Indications for antiplatelet therapy and aneurysm treatment were not predefined. In addition, although our case series was relatively large compared with those previously published on the same subject, the sample size was small. Also, the duration of follow-up was relatively short, precluding any conclusion on the risk of ischemic recurrence or aneurysm rupture in the long term. 5. Conclusion Unruptured IA is a rare cause of IS/TIA. IS is associated with aneurysm thrombosis. Our findings suggest that aneurysm thrombosis is a dynamic process associated with a low rate of ischemic recurrence on antiplatelet therapy but that may be followed by subarachnoid hemorrhage. These lesions should be controlled for recanalization and should be considered for treatment whenever they are amenable to endovascular or neurosurgical treatment, with respect to patient considerations. References [1] Antunes JL, Correll JW. Cerebral emboli from intracranial aneurysms. Surg Neurol 1976;6(1):7–10. [2] Fisher M, Davidson RI, Marcus EM. Transient focal cerebral ischemia as a presenting manifestation of unruptured cerebral aneurysms. Ann Neurol 1980;8(4):367–72.
33
[3] Cohen MM, Hemalatha CP, D’Addario RT, Goldman HW. Embolization from a fusiform middle cerebral artery aneurysm. Stroke 1980;11(2):158– 61. [4] Fukuoka S, Suematsu K, Nakamura J, Matsuzaki T, Satoh S, Hashimoto I. Transient ischemic attacks caused by unruptured intracranial aneurysm. Surg Neurol 1982;17(6):464–7. [5] Przelomski MM, Fisher M, Davidson RI, Jones HR, Marcus EM. Unruptured intracranial aneurysm and transient focal cerebral ischemia: a follow-up study. Neurology 1986;36(4):584–7. [6] Qureshi AI, Mohammad Y, Yahia AM, Luft AR, Sharma M, Tamargo RJ, et al. Ischemic events associated with unruptured intracranial aneurysms: multicenter clinical study and review of the literature. Neurosurgery 2000;46(2):282–9 [discussion 289–90]. [7] Nanda A, Vannemreddy PS. Cerebral ischemia as a presenting feature of intracranial aneurysms: a negative prognostic indicator in the management of aneurysms. Neurosurgery 2006;58(5):831–7 [discussion 831–37]. [8] Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral embolic signals in patients with carotid stenosis. Correlation with appearance of plaque ulceration on angiography. Stroke 1995;26(5):813–5. [9] Markus HS, Droste DW, Kaps M, Larrue V, Lees KR, Siebler M, et al. Dual antiplatelet therapy with clopidogrel and aspirin in symptomatic carotid stenosis evaluated using Doppler embolic signal detection: the Clopidogrel and Aspirin for Reduction of Emboli in Symptomatic Carotid Stenosis (CARESS) trial. Circulation 2005;111(17):2233–40. [10] Oh YS, Lee SJ, Shon YM, Yang DW, Kim BS, Cho AH. Incidental unruptured intracranial aneurysms in patients with acute ischemic stroke. Cerebrovasc Dis 2008;26(6):650–3. [11] Rinkel GJ, Djibuti M, Algra A, van Gijn J. Prevalence and risk of rupture of intracranial aneurysms: a systematic review. Stroke 1998;29(1): 251–6. [12] Vernooij MW, Ikram MA, Tanghe HL, Vincent AJ, Hofman A, Krestin GP, et al. Incidental findings on brain MRI in the general population. N Engl J Med 2007;357(18):1821–8. [13] Rinkel GJ. Natural history, epidemiology and screening of unruptured intracranial aneurysms. J Neuroradiol 2008;35(2):99–103. [14] Qureshi AI, Isa A, Cinnamon J, Fountain J, Ottenlips JR, Braimah J, et al. Magnetic resonance angiography in patients with brain infarction. J Neuroimaging 1998;8(2):65–70. [15] Brownlee RD, Tranmer BI, Sevick RJ, Karmy G, Curry BJ. Spontaneous thrombosis of an unruptured anterior communicating artery aneurysm. An unusual cause of ischemic stroke. Stroke 1995;26(10):1945–9. [16] Whittle IR, Dorsch NW, Besser M. Spontaneous thrombosis in giant intracranial aneurysms. J Neurol Neurosurg Psychiatry 1982;45(11):1040–7. [17] Hoffman WF, Wilson CB, Townsend JJ. Recurrent transient ischemic attacks secondary to an embolizing saccular middle cerebral artery aneurysm. Case report. J Neurosurg 1979;51(1):103–6. [18] Schaller B, Lyrer P. Focal neurological deficits following spontaneous thrombosis of unruptured giant aneurysms. Eur Neurol 2002;47(3):175–82. [19] Schunk H. Spontaneous thrombosis of intracranial aneurysms. Am J Roentgenol Radium Ther Nucl Med 1964;91:1327–38. [20] Rosta L, Battaglia R, Pasqualin A, Beltramello A. Italian cooperative study on giant intracranial aneurysms. 2. Radiological data. Acta Neurochir Suppl (Wien) 1988;42:53–9. [21] Friedman JA, Piepgras DG, Pichelmann MA, Hansen KK, Brown Jr RD, Wiebers DO. Small cerebral aneurysms presenting with symptoms other than rupture. Neurology 2001;57(7):1212–6. [22] Cohen JE, Itshayek E, Gomori JM, Grigoriadis S, Raphaeli G, Spektor S, et al. Spontaneous thrombosis of cerebral aneurysms presenting with ischemic stroke. J Neurol Sci 2007;254(1–2):95–8. [23] McLaughlin N, Bojanowski MW. Unruptured cerebral aneurysms presenting with ischemic events. Can J Neurol Sci 2008;35(5):588–92. [24] Sutherland GR, King ME, Peerless SJ, Vezina WC, Brown GW, Chamberlain MJ. Platelet interaction within giant intracranial aneurysms. J Neurosurg 1982;56(1):53–61. [25] Ohno K, Suzuki R, Masaoka H, Matsushima Y, Monma S, Inaba Y. Unruptured aneurysms in patients with transient ischemic attack or reversible ischemic neurological deficit. Report of eight cases. Clin Neurol Neurosurg 1989;91(3):229–33. [26] Kato M, Kaku Y, Okumura A, Iwama T, Sakai N. Thrombosed unruptured cerebral aneurysm causing brain infarction followed by subarachnoid hemorrhage – case report. Neurol Med Chir (Tokyo) 2005;45(9):472–5. [27] Bruno G, Todor R, Lewis I, Chyatte D. Vascular extracellular matrix remodeling in cerebral aneurysms. J Neurosurg 1998;89(3):431–40. [28] Fontaine V, Jacob MP, Houard X, Rossignol P, Plissonnier D, Angles-Cano E, et al. Involvement of the mural thrombus as a site of protease release and activation in human aortic aneurysms. Am J Pathol 2002;161(5):1701–10. [29] Carrell TW, Burnand KG, Booth NA, Humphries J, Smith A. Intraluminal thrombus enhances proteolysis in abdominal aortic aneurysms. Vascular 2006;14(1):9–16. [30] Jin D, Sheng J, Yang X, Gao B. Matrix metalloproteinases and tissue inhibitors of metalloproteinases expression in human cerebral ruptured and unruptured aneurysm. Surg Neurol 2007;68(Suppl. 2):S11–6 [discussion S16]. [31] Nagashima M, Nemoto M, Hadeishi H, Suzuki A, Yasui N. Unruptured aneurysms associated with ischaemic cerebrovascular diseases. Surgical indication. Acta Neurochir (Wien) 1993;124(2–4):71–8.