- Email: [email protected]
Ischemic stroke patterns and hemodynamic features in patients with small vertebrobasilar artery
Journal of the Neurological Sciences 287 (2009) 227–235
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Ischemic stroke patterns and hemodynamic features in patients with small vertebrobasilar artery Jong-Ho Park a, Jae-Kyu Roh b,⁎, Hyung-Min Kwon c a b c
Department of Neurology, Myongji Hospital, Kwandong University College of Medicine, Goyang, Republic of Korea Department of Neurology, Seoul National University Hospital, Seoul, Republic of Korea Department of Neurology, Seoul National University Boramae Hospital, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 31 December 2008 Received in revised form 15 June 2009 Accepted 7 July 2009 Available online 13 August 2009 Keywords: Small vertebrobasilar artery Long circumferential artery Ischemic pattern MR imaging Angiography
a b s t r a c t Background: To determine the role of small vertebrobasilar artery (SVBA) in patients with posterior circulation stroke (PCS), we evaluated the ischemic patterns, collateral features, and stroke mechanisms in PCS patients with SVBA. Methods: Ischemic findings on magnetic resonance (MR) imaging were correlated with 3D time-of-flight/ contrast-enhanced MR angiography and/or catheter angiography in 18 patients (mean age, 68.0 ± 11.8 years; 9 males). SVBA (lumen diameter of b 3 mm) was compared with stenotic normal-sized VBA (NVBA) in 14 PCS patients. Results: Ischemic lesions were predominantly observed in the cerebellum and/or medulla (vertebral artery (VA) territory). All subjects had fetal posterior circulation (FPC) from the internal carotid artery to the posterior cerebral artery. Sixteen patients (88.9%) had distal or diffuse VA stenosis/occlusion. In 14 patients (77.8%), long circumferential artery (LCA) was prominently observed. In atherothrombotic patients, infratentorial PCS might occur following artery-to-artery embolism from the low-flowed or stenotic VA to LCA. Ischemic patterns between subjects with and without VA disease were almost similar. As the degree of VA disease increased, the frequency of LCA prominence showed an increased tendency (P = 0.003). Relatively small, scattered infarcts were observed in patients with SVBA than in those with stenotic NVBA. Conclusions: FPC does not protect against infratentorial PCS. Regardless of extensive arterial lesions, relatively small infarcts may be due to previously established collaterals from the LCA, which could compensate for the defects in the infratentorial area. © 2009 Elsevier B.V. All rights reserved.
1. Introduction A small vertebrobasilar artery (SVBA) is infrequently observed in stroke patients, regardless of whether it is directly related to ischemic stroke. It may not be possible to adequately visualize an SVBA on threedimensional (3D) time-of-flight (TOF) magnetic resonance (MR) angiography (MRA); moreover, because its exact nature is unclear, it attracts little attention until posterior circulation stroke (PCS) occurs. Ischemic patterns in patients with SVBA appear to be different from those in subjects with a normal-sized VBA (NVBA). Although it has been reported that SVBA might induce PCS [1], the collaterals that would compensate for the posterior circulation system in the case of ischemic stroke have not been well established. Furthermore, a correlation of neuroimaging findings with vascular pathologies has not been attempted in PCS patients with SVBA. Therefore, we evaluated the ischemic pattern
⁎ Corresponding author. Department of Neurology, Seoul National University College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-744, Republic of Korea. Tel.: +82 2 2072 3265; fax: +82 2 3672 4949. E-mail address: [email protected] (J.-K. Roh). 0022-510X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2009.07.007
of the vascular lesions by magnetic resonance (MR) imaging and the hemodynamic pattern (collateral circulation) by MR angiography (MRA) and/or transfemoral cerebral angiography (TFCA) and assessed the stroke mechanism in PCS patients with SVBA by using the abovementioned techniques together with transcranial Doppler (TCD). 2. Methods 2.1. Patient selection The present study is a case series from 2 academic centers (Seoul National University Hospital (SNUH) and Myongji Hospital (MJH)). We assessed 1959 acute ischemic stroke patients (SNUH: n=1604, between September 2003 and February 2006; MJH: n=355, between March 2006 and September 2007). Within 1 week of symptom onset, patients underwent brain MRI and 3D TOF MRA by a 1.5-Tesla MR system (SNUH: Siemens 1.5 Vision, Erlangen, Germany; MJH: Intera 1.5 T 10.3 version, Eindhoven, Netherlands). The stroke types were categorized as follows: anterior circulation stroke (ACS) =1120 (57.2%), PCS=640 (32.7%), ACS plus PCS=33 (1.7%), transient ischemic attack (TIA)=162 (8.3%), and venous sinus thrombosis=4 (0.2%).
228
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
Table 1 Summary of 18 PCS patients with small vertebrobasilar artery. No
Sex/age
Risk factor(s)
VA lesion
BA lesion
Prominence
FPC
TCD
1 2a 3a 4 5b 6a 7a 8a 9a
F/53 M/69 F/69 F/76 M/87 F/76 M/69 M/73 M/70
HT HT HT, DM, Af HT HT HT Af HT, Af HT, DM, smoking
– – L (diffuse) L (V4) L (V3 and V4) L (V4) R (diffuse) R (V4) R (diffuse)
– – D – – M – – –
PICA (R) – PICA (B), AICA (R) AICA (L) PICA (L) PICA (B) – – –
B B B B B B U B U
10a 11b 12b,c 13a,b,c 14b,c
F/77 F/53 F/58 F/65 M/53
HT, DM HT HT, DM HT HT
R (V4), L (diffuse) B (VBA junction) B (VBA junction) B (V4) B (V4)
– P – – P
PICA PICA PICA PICA PICA
(R) (B) (B) (B) (B)
B B B B B
15a,c
M/87
HT
B (V4)
–
PICA (B)
B
16a,b,c
M/46
–
P
PICA (L), AICA (L)
B
17a,b
F/64
HT
R (diffuse) L (V4) B (V4)
– Decreased MFV and turbulent waves of (B) VBAs (b 30 cm/s) – – Stenosis of BA (N80 cm/s) Stenosis of (B) VBAs (N80 cm/s) Dampened signal of (B) VBAs – Dampened signal of (L) VA and BA Stenosis of (R) VA (N 110 cm/s) Blunted signal of BA – Blunted signal of (B) VBAs Dampened signal of (L) VA and BA bidirectional flow of (R) VA Stenosis of BA (N140 cm/s) Dampened signal of (L) VA Dampened signal of (L) VA Absent signal of (R) VA and BA –
–
AICA (B)
B
M/79
HT
B (diffuse)
P
AICA (B), SCA (B)
B
18
a,b,c
Stenosis of BA (N120 cm/s) Dampened signal of (B) VAs –
PCS, posterior circulation stroke; VA, vertebral artery; BA, basilar artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; FPC, fetal posterior circulation; HT, hypertension; DM, diabetes mellitus; Af, atrial fibrillation; L, left; R, right; U, unilateral; B, bilateral; P, proximal; M, middle; D, distal; TFCA, transfemoral cerebral angiography; MFV, mean flow velocity. a Contrast-enhanced MR angiography performed. b TFCA performed. c Retrograde filling of BA from anterior circulation.
2.2. SVBA definition The mean diameter of the normal basilar artery (BA) has been reported to be 3.17 mm [2] and the hypoplastic vertebral artery (VA) was defined to have a lumen diameter of less than 2–3 mm [3,4]. Because there is no consensus on the value of SVBA, we defined SVBA as a diameter of less than 3 mm. For defining the diameter of SVBA, we examined the mid-portion level of the BA and the V2 of the largest VA by using magnified images of MRA on a picture archiving and communication system (PACS). The location of the VA was defined according to standard criteria [5,6] (V1; prevertebral portion, V2; the portion within the vertebral columns, V3; atlantoaxial portion, V4; intracranial portion). 2.3. Study patients We assessed 37 acute stroke patients with SVBA. MRI and MRA were obtained at 2.3 ± 1.1 days after stroke onset. We included patients whose configuration of SVBA was visible on 3D TOF MRA. We also included those who had undergone angiographic work-up (e.g., contrast-enhanced MRA or TFCA) even if their SVBA was not or only faintly visible on routine 3D TOF MRA. Patients were excluded if they showed no visible VBA (n = 7) or only a rudimentary VBA (n = 3) on 3D TOF MRA; moreover, these 10 patients had not undergone detailed angiographic work-up that could enable a diagnosis of SVBA. Among the excluded 10 patients, 9 had been afflicted by PCS and the remaining one by ACS. Those who showed no relevant lesions on diffusion-weighted images (DWI; n = 2) were also excluded. Of the 25 stroke patients with SVBA (7 with ACS and 18 with PCS), 18 PCS patients who underwent MRI and MRA at 2.2 ± 1.2 days after the
onset of symptoms were eventually selected for the study. Acute ischemic lesion was defined by high-signal intensity on DWI with lowsignal intensity on the apparent diffusion coefficient map (only 2 patients [nos. 4 and 18] by T2-weighted image; T2-WI). Of these 18 patients, 12 patients (66.7%; nos. 2, 3, 6–10, 13, and 15–18) underwent contrast-enhanced MRA with the same TOF MR sequence to evaluate the course of the VBA. We assessed 30 acute PCS patients with NVBA between March 2007 and September 2007 at MJH and considered them as the control group. We regarded NVBA as a diameter of more than 3 mm, except in the case of unilateral hypoplastic VA. We obtained MRI and 3D contrast-enhanced MRA scans for the control subjects at 2.2 ± 1.4 days after the onset of symptoms. The topography of the infarcts was determined with reference to the maps, establishing an anatomical correspondence with the dominant arterial territories of the brainstem and cerebellum [7]. TFCA was performed in 8 patients with SVBA (nos. 5, 11–14, and 16–18) and 2 patients with NVBA (nos. 7 and 14) after an informed consent was obtained from each patient. The possibility of VBA dissection could be excluded by the findings of TFCA (string sign) or source view of MRA (double lumen) [8]. The long circumferential artery (LCA) (e.g., posterior inferior [PICA], anterior inferior [AICA], or superior cerebellar artery [SCA]) was regarded to be prominent if it was well depicted on MRA or TFCA compared to its parent SVBA. In patients with NVBA, LCA was regarded to be present if it was adequately visible on MRA or TFCA. The circle of Willis was graded as fetal posterior circulation (FPC) when one of the P2 segments of the posterior cerebral artery was supplied by the internal carotid artery via the posterior communicating artery (absent or hypoplastic P1 segment of the posterior cerebral
Fig. 1. Ischemic stroke patterns according to the various vascular lesions observed in 18 patients with small VBA. Group I: no VA disease, with or without LCA prominence (n = 2), Group II: unilateral VA disease, with (upper) or without (low) LCA prominence (n = 7), and Group III: bilateral VA disease, with LCA prominence VA, vertebral artery; BA, basilar artery; LCA, long circumferential artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; LAA, large artery atherosclerosis; SAO, small artery occlusion; CE, cardiac embolism.
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
229
230
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
Fig. 2. Patient 2. A, Diffusion-weighted imaging shows an infarct in the territory of the right PICA. B-1 and B-2, 3D TOF MRA shows intact VAs caused by possible in situ thrombosis of PICA. Patient 6. A, Diffusion-weighted imaging shows an infarct in the right pons. B, 3D TOF and contrast-enhanced MRA show steno-occlusive lesions of the mid-BA and the left distal VA (V4). Prominence of the right PICA (arrow) is noted. Patient 9. A-1, 2, 3, 4, Diffusion-weighted imaging shows an infarct in the right medial cerebellum, both temporo-occipital lobes, and left medial thalamus (arrows). B, Contrast-enhanced MRA shows diffuse steno-occlusive lesion in the right VA with no prominence of LCA. Patient 12. A-1, 2, Diffusionweighted imaging shows an infarct in the right PICA territory. B, 3D TOF MRA shows no visible vertebrobasilar system. C-1, 2, TFCA depicts prominent PICA ending (arrow) from both distal VAs. Patient 17. A, Diffusion-weighted imaging shows an infarct in the left focal medial cerebellum (PICA territory). B, 3D TOF and contrast-enhanced MRA depict both distal VA stenoses (V4). C, TFCA shows both AICA (white arrow) prominences adjacent to both distal stenotic VAs (black arrow). VA, vertebral artery; BA, basilar artery; LCA, long circumferential artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; TOF, time-of-flight; TFCA, transfemoral cerebral angiography.
Table 2 Associations of VA disease with prominence of LCA in patients with SVBA. Prominence of LCA VA diseaseb
None Unilateral Bilateral
P
None (n = 4)
One (n = 5)
≥ Two (n = 9)a
1 (25.0) 3 (75.0) 0 (0)
1 (20.0) 3 (60.0) 1 (20.0)
0 (0) 1 (11.1) 8 (88.9)
Values are number (percentage) of patients. LCA, long circumferential artery; VA, vertebral artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery. a Including concurrent PICA, AICA, and SCA. b distal or diffuse VA stenosis/occlusion.
0.003
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235 231
Fig. 3. Ischemic stroke patterns in 14 patients with stenotic normal-sized VBA. VA, vertebral artery; BA, basilar artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; LAA, large artery atherosclerosis.
232
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
artery on that side) [9]. MRA or TFCA findings were copied from the original film and analyzed by one of the authors (J.-H.P.) who was blinded to the MRI findings. MRI findings (DWI or T2-WI) were schematically drawn by another author (H.-M.K.) who was blinded to the MRA findings.
respectively. The analyses of the frequencies of LCA prominence according to the number of VA disease were conducted using a linear by linear association test. They were conducted using SPSS 11.5 for Windows (SPSS Inc., Chicago, IL, USA). Two-sided P values of 0.05 or less were considered significant.
2.4. MRI and MRA analyses
3. Results
MRI was performed using two 1.5-Tesla MR systems with T2-WI (repetition time/echo time (TR/TE) of 5000/99 ms in the Siemens system and 4230/100 ms in the Intera system), DWI (TR/TE of 6500/ 110 ms in Siemens and 3737/66 ms in Intera), and a fluid-attenuated inversion recovery (FLAIR) image (TR/TE of 9000/119 ms in Siemens and 11,000/140 ms in Intera). MRA was performed using the 3D TOF technique (older Signa unit, TR/TE of 35/7.2 ms in Siemens and 23/ 6.9 ms in Intera). 3D spoiled gradient-echo acquisition was applied using the following parameters: TR/TE/excitation (37/6.6/1 in Siemens and 21/5/1 in Intera); excitation angle (20° in Siemens and 30° in Intera); field of view (230 × 6/8 in both Siemens and Intera); and acquisition matrix (195 × 512 in Siemens and 368 × 368 in Intera). All MRA procedures included intracranial and neck vessels (except in patient no. 9) and were performed using a 3D TOF technique with or without 3D contrast-enhanced MRA, which was performed after a 10-mL intravenous bolus injection of Magnevist (Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA) with the same TR/TE as each 3D TOF MRA.
3.1. Demographic features, risk factors, and stroke subtypes of the study subjects In this study, 9 men and 9 women were included. None of the study patients had experienced a VBA dissection or stenosis (N50%) or occlusion of the internal carotid artery. Table 1 lists the baseline characteristics and angiographic findings of the study patients with SVBA. The mean age of the patients was 68.0 ± 11.8 years (range, 46– 87 years). Most of the patients had vascular risk factors: hypertension in 15 patients (83.3%), diabetes in 4 patients (22.2%), Af in 3 patients (16.7%), and smoking in 1 patient (5.6%). The index stroke subtypes of 18 patients according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification [12] were large artery atherosclerosis (LAA) in 12 patients (66.7%), small artery occlusion (SAO) in 3 patients (16.7%), stroke with undetermined etiology (LAA with cardiogenic embolism [CE]) in 2 patients (11.1%), and CE in 1 patient (5.6%). 3.2. TCD findings in study subjects
2.5. TCD examination TCD was performed in patients with SVBA with a 2-MHz pulsed Doppler instrument (SNUH: PMD 100, Spencer Technologies, Seattle, WA, USA; MJH: Companion III, Viasys Healthcare, Warwick, UK) by an experienced sonographer. The posterior circulation was assessed according to the standard protocol [10]. Stenosis of the VBA was diagnosed if the mean flow velocity was greater than 75 cm/s. The residual flow grades of the SVBA were assessed using the Thrombolysis in Brain Ischemia (TIBI) flow classification [11]. 2.6. Definition of risk factors We assessed the risk factors in each patient for stroke, such as hypertension, diabetes, current smoking, and heart disease. Cardiogenic embolic diseases (e.g., atrial fibrillation (Af), left ventricular akinesia, recent myocardial infarction, or patent foramen ovale with right-to-left shunt) were routinely evaluated in all patients by electrocardiography and transthoracic echocardiography; further studies—such as transechocardiography or 24-h Holter monitoring— were performed if any abnormality was found on initial work-up. 2.7. Statistical analysis The analyses of the general categorical variables and continuous variables were conducted using the χ2 tests and Student's t tests,
Of the 18 patients, 11 (nos. 2, 5–7, 9, 10, 12–15, and 17) underwent TCD examination (Table 1). All patients, except 2 (nos. 6 and 7), had an index stroke subtype of LAA. Those with LAA showed stenosis or poor perfusion state (from blunted to absent signal) of VA and/or BA on TCD. 3.3. Characteristics of ischemic findings Brain MRI findings are presented in Fig. 1. Generally, patients had relatively small-sized infarcts with a scattered pattern (i.e., not distributed throughout the arterial territory). In 13 patients (72.2%), ischemic lesions were predominantly observed in the cerebellum and/ or medulla in the VA territory. In 4 patients (22.2%), ischemic lesions were observed in the territory of the BA or its main branches. In 6 patients (33.3%; nos. 7–9, 11, 14, and 16), ischemic lesions were observed in multiple (i.e., more than 2) arterial territories. Ischemic patterns of the patients with LAA were mostly of the artery-to-artery embolism type. 3.4. Angiographic findings of the study subjects by MRA/TFCA FPC from the internal carotid artery to the posterior cerebral artery was noted in all patients with SVBA (100%). Distal VA disease (stenosis or occlusion) adjacent to the distal portion of the PICA orifice or diffuse VA disease was noted in 16 patients (88.9%; bilateral in 9). BA
Fig. 4. Patient 1. A-1 and A-2, Diffusion-weighted imaging shows an acute infarct in the right PICA territory and left medial occipital lobe. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the right distal VA (V4). Patient 4. A, Diffusion-weighted imaging shows an acute infarct in the left lateral medulla. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the left distal VA (V4 and distal V3). Patient 7. A-1 and A-2, Diffusion-weighted imaging shows an acute infarct in the left lateral medulla (lower portion) and left PICA territory. B-1 and B-2, 3D TOF and contrast-enhanced MRA show diffuse steno-occlusive lesion in the left VA. Patient 8. A, Diffusionweighted imaging shows an acute infarct in both PICA territories. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the right distal VA (V4 and distal V3) and the left focal V4. Right AICA is seen on the contrast-enhanced MRA (arrow). Patient 11. A-1, A-2 and A-3, Diffusion-weighted imaging shows an acute small infarct in the left PICA territory and right pons. C, Contrast-enhanced MRA and TFCA depict focal stenosis (arrow) of the left VA (V2). Patient 12. A, Diffusion-weighted imaging shows an acute small infarct in the right PICA territory. B-2, Contrast-enhanced MRA depicts focal stenosis of the right VA (V3). Patient 13. A, Diffusion-weighted imaging shows an acute small, multiple infarcts in the right PICA and SCA territories. B-1 and B-2, 3D TOF and contrast-enhanced MRA depict focal stenosis in both the VAs (right V2 and left V2 and V4). Patient 14. A-1, A-2, A-3 and A-4, Diffusion-weighted imaging shows multiple acute small infarcts in the right PICA, SCA, and both the occipital lobes. B-1 and B-2, 3D TOF and contrast-enhanced MRA depict focal stenosis of both VAs (right focal V3–4 and left focal V4). TFCA shows right AICA prominence (arrow) and focal stenosis of ipsilateral VA (arrowhead). VA, vertebral artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; TOF, time-of-flight; CE, contrast enhanced; TFCA, transfemoral cerebral angiography.
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
233
234
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
stenosis or occlusion was noted in 6 patients (proximal portion in 4, middle in 1, and distal in 1). Prominence of LCA was noted in 14 patients (77.8%), i.e., PICA was prominent in 11 patients (61.1%); AICA in 5 patients (27.8%); and SCA in 1 patient (5.6%). Of the 8 patients who underwent TFCA, 5 (62.5%; nos. 12–14, 16, and 18) showed retrograde filling of BA from anterior circulation. 3.5. Ischemic stroke patterns according to various vascular lesions Of the patients studied, 16 had VA disease; among these, 12 had distal VA disease, while 6 had diffuse VA disease. Consequently, we classified them as follows (Figs. 1 and 2): Group I: no VA disease, with or without LCA prominence (n = 2, 11.1%); Group II: unilateral VA disease, with or without LCA prominence (n = 7, 38.9%); and Group III: bilateral VA disease, with LCA prominence (n = 9, 50.0%). 3.5.1. No VA disease, with or without LCA prominence (Group I) Only 2 patients did not have VA disease. Of them, 1 showed PICA prominence (no. 1) and 1 did not (no. 2). 3.5.1.1. Presumed etiopathogenesis (subtype). The causes included SAO in 1 patient (no. 1) and LAA in the other (no. 2). 3.5.1.2. MRI findings. The lesions were located in the paramedian pons in 1 patient (no. 1) and in the medial PICA territory in another (no. 2). 3.5.2. Unilateral VA disease, with or without LCA prominence (Group II) Of the 7 patients with unilateral VA disease, 4 showed PICA or concurrent AICA prominence (patient nos. 3–6) and the other 3 showed no prominence (patient nos. 7–9). 3.5.2.1. Presumed etiopathogenesis (subtype). A variety of causes were responsible for the pathogenesis in these patients: LAA was evident in 2 patients (nos. 5 and 9); SAO in 2 patients (nos. 4 and 6); stroke with undetermined etiology (LAA plus CE) in 2 patients (nos. 3 and 8); and CE in 1 patient (no. 7). 3.5.2.2. MRI findings. The lesions were variously located in the center of PICA territory (patient nos. 5, 8, and 9); the medial medulla and AICA territory (patient no. 7); the pons (patient nos. 4 and 6); and the posterior cortical area, with or without deep gray matter involvement (patient nos. 3, 8, and 9). However, the lesions mostly originated in the VA territory (patient nos. 5 and 7–9). Of the 4 patients with PICA or/ concurrent AICA prominence, 2 patients (nos. 3 and 5) had single and relatively small infarcts in the occipital cortex and cerebellum; the other 2 patients had SAO in the pons. Moreover, in the 3 patients lacking PICA or AICA prominence (nos. 7–9), scattered infarcts were observed in multiple territories. 3.5.3. Bilateral VA disease with LCA prominence (Group III) Of the 9 patients with bilateral VA disease, 6 showed PICA prominence (patient nos. 10–15); 1, concurrent PICA and AICA prominence (patient no. 16); 1, AICA prominence (patient no. 17); and 1, concurrent AICA and SCA prominence (patient no. 18). None of the patients were without AICA/SCA/PICA prominence. 3.5.3.1. Presumed etiopathogenesis (subtype). sumed cause for all patients was LAA.
In this group, the pre-
3.5.3.2. MRI findings. Lesions were located predominantly in the cerebellum (PICA territory; patient nos. 10 and 12–18). Almost all subjects with cerebellar lesions had relatively small and scattered patterns (i.e., not all of the territory was involved). Concurrent pons or medial medulla involvement was observed in 2 patients (nos.14 and 16) and bilateral mesencephalo-pontine involvement because of stenoocclusion of the distal VA and proximal BA was observed in 1 (no. 11).
3.6. Associations of VA disease with the LCA prominence in patients with SVBA Table 2 demonstrates the association between VA disease and compensatory collaterals from the LCA (PICA/AICA/SCA). We observed that as the degree of VA disease increased (i.e., from “none” to “unilateral” to “bilateral”), the frequency of LCA prominence (i.e., “none,” “one,” and “two or more”) tended to increase (P = 0.003). 3.7. Comparison of ischemic patterns between patients with stenotic SVBA and those with stenotic NVBA Of the 30 PCS patients with NVBA, 14 had an index stroke subtype of LAA. As most of the study patients had VA disease, we intended to compare the ischemic patterns of the 16 patients with stenotic SVBA (Groups II and III) with those of 14 patients with stenotic NVBA who underwent MR study at 2.1 ± 1.3 days after the onset of symptoms. Supplemental table illustrates that baseline demographics and classic risk factors were similar between the 2 groups, except Af. FPC, bilateral VA disease, and LCA presence were more prevalent in the patients with stenotic SVBA (P b 0.0001, P = 0.052, and P = 0.011, respectively). Brain MRI and MRA findings of the stenotic NVBA group are presented in Figs. 3 and 4. Lesions were variously located in the PICA territory (patient nos. 1–3 and 7–14); the lateral medulla (patient nos. 4–7); SCA territory (patient nos. 13 and 14); the occipital area (patient nos. 1 and 14); and the pons (patient no. 11). The stenotic NVBA group showed relatively large, conglomerate infarct patterns compared with those of stenotic SVBA group. However, the ischemic findings of 4 patients with NVBA (patient nos. 11–14) were similar to those of SVBA group patients. They had common feature that showed extracranial focal VA lesion (below the V3). Three patients (nos. 8, 13, and 14) with bilateral VA diseases showed multiple lesions involving more than 2 territories. 4. Discussion To date, few studies have focused on SVBA, except to assess to what extent SVBA might result in PCS [1]. Our study focused on the MR ischemic patterns, hemodynamic features (collateral circulation), and possible stroke mechanisms in PCS patients who showed small or faintly visible VBA on MRA. FPC is a fetal variant of the posterior cerebral artery (PCA) from the internal carotid artery. The prevalence of FPC is reported to be 32% in the general population [13]. The present study showed the existent varieties of FPC (bilateral occurrence in 88.9% of patients), and suggested that FPC may compensate the posterior circulation zone for the hemodynamic insufficiency caused by SVBA. However, FPC could be protective against occipital lobe infarcts only [14]. Since the cerebellar tentorium impedes the formation of a leptomeningeal connection, FPC does not contribute to the perfusion of the infratentorial area [15]. Consequently, FPC makes the development of leptomeningeal collaterals between the ICA and the vertebrobasilar system impossible [15]. Our study showed that most of the infratentorial lesions originated from the cerebellum and/or medulla (VA territory) or the pons (BA territory). These results are consistent with the observation [15] that FPC would not be able to protect the infratentorial area against PCS. We found that the ischemic patterns in PCS patients were generally characterized by relatively small (i.e., not all of the territory was involved) and/or scattered lesions, while stenotic NVBA group showed relatively large, conglomerate infarct patterns except for extracranial focal VA lesion (below V3). The most common angiographic finding was VA stenosis or occlusion in which distal lesions were predominant. This is why infratentorial lesions mainly originate in the cerebellum and/or medulla. The pattern of ischemic lesions between patients with and without VA disease and between patients with differing grades of VA disease did not differ significantly, as observed
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
by MRI. However, the involvement of multiple arterial territories was noted in 6 patients (nos. 7–9, 11, 14, and 16). All of the 6 patients had VA disease with the etiopathogenesis predominantly being LAA (nos. 9, 11, 14, and 16). Regardless of the presence of extensive arterial lesions, relatively small infarcts can be attributed to the established leptomeningeal collaterals from the LCA (e.g., PICA, AICA, or SCA) that can compensate for the defects in the distal vertebrobasilar arterial territories (i.e., infratentorial area). Thus, the degree of collateral development along with a chronic process of atherothrombosis may determine the pattern (particularly, the size) of an ischemic lesion. However, in patients with unilateral VA disease without LCA prominence (nos. 8 and 9), the PICA/AICA/SCA which could have been well delineated on TFCA was not visible on MRA. Our study may raise the question of whether the SVBA is of congenital origin or a consequence of multiple or longitudinal atherosclerotic narrowing. Embryologically, if the BA does not become the main source of blood supply to the developing posterior cerebral arteries, the FPC might persist and remain large in size [1]. The observations that all the 18 patients had FPC and that the FPC was larger than the VBAs may support the hypothesis that the SVBA is congenitally small rather than acquired. Are the smaller arteries more vulnerable to atherothrombosis? It is known that there is a significant relationship between hypoplastic VA and ipsilateral PCS [16,17]. A small-diameter artery appears to be vulnerable to stenosis or occlusion if vascular risk factors are present [18], as reported in some studies [16,19,20]. A small VA shows lower mean flow volume [17,21,22] and decreased flow velocities [22] in color-coded duplex ultrasonography. It is plausible that SVBA could have low flow and slow velocity and therefore be susceptible to prothrombotic or atherosclerotic processes in the presence of vascular risk factors. In our study, the stroke mechanism was predominantly atherothrombotic embolism (66.7%). Our patients with LAA, for whom we performed TCD, generally showed a low-flow or stenotic state of VBA circulation. Considering the ischemic patterns together with the TCD findings, infratentorial PCS may occur due to an arteryto-artery embolism from the low-flowed or stenotic VA to the LCA. Our study has some limitations. First, our study comprised a small number of patients because of our exclusion criteria; this may have resulted in a selection bias. Second, inconsistent diagnostic imaging for angiography (TOF, contrast-enhanced MRA, or TFCA) might have affected the external validity of the hemodynamic collateral findings. In some patients, advanced arterial narrowing from the VA orifice made it difficult to access the entire VBA by TFCA. Finally, we could not perform detailed comparative analysis between the 2 groups (SVBA vs. NVBA) because there were only 2 patients with NVBA who underwent TFCA. In fact, there is no rationale for performing TFCA in PCS patients with NVBA only for comparing the angiographic findings with those of patients with SVBA. Although this was an observational study, we suggest that relatively small infarcts can be attributed to the establishment of leptomeningeal collaterals from the LCA in PCS patients with extensive arterial lesions.
235
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jns.2009.07.007. References [1] Chaturvedi S, Lukovits TG, Chen W, Gorelick PB. Ischemia in the territory of a hypoplastic vertebrobasilar system. Neurology 1999;52:980–3. [2] Smoker WR, Price MJ, Keyes WD, Corbett JJ, Gentry LR. High-resolution computed tomography of the basilar artery: normal size and position. AJNR Am J Neuroradiol 1986;7:55–60. [3] Fisher CM, Gore I, Okabe N, White PD. Atherosclerosis of the carotid and vertebral arteries—extracranial and intracranial. J Neuropathol Exp Neurol 1965;24:455–76. [4] Touboul PJ, Bousser MG, LaPlane D, Castaigne P. Duplex scanning of normal vertebral arteries. Stroke 1986;17:921–3. [5] Argenson C, Francke JP, Sylla S, Dintimille H, Papasian S, di Marino V. The vertebral arteries (segments V1 and V2). Anat Clin 1980;2:29–41. [6] Francke JP, di Marino V, Pannier M, Argenson C, Libersa C. The vertebral arteries (arteria vertebralis). The V3 atlanto-axial and V4 intracranial segments-collaterals. Anat Clin 1981;2:229–42. [7] Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of human brain; brainstem and cerebellum. Neurology 1996;47:1125–35. [8] Auer A, Felber S, Schmidauer C, Waldenberger P, Aichner F. Magnetic resonance angiographic and clinical features of extracranial vertebral artery dissection. J Neurol Neurosurg Psychiatry 1998;64:474–81. [9] van der Grond J, van Raamt AF, van der Graaf Y, Mali WPTM, Bisschops RHC. A fetal circle of Willis is associated with a decreased deep white matter lesion load. Neurology 2004;63:1452–6. [10] Alexandrov AV. Transcranial Doppler sonography: principles, examination technique, normal values, and waveform patterns. Vasc Ultrasound Today 1998;3:141–60. [11] Demchuk AM, Burgin WS, Christou I, Felberg RA, Barber PA, Hill MD, et al. Thrombolysis in Brain Ischemia (TIBI) transcranial Doppler flow grades predict clinical severity, early recovery, and mortality in patients treated with intravenous tissue plasminogen activator. Stroke 2001;32:89–93. [12] Adams Jr HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al. Classification of subtype of acute ischemic stroke. Definitions for use in multicenter clinical trial. TOAST. Stroke 1993;24:35–41. [13] Krabbe Hartkamp MJ, Van der Grond J, de Leeuw FE, de Groot JC, Algra A, Hillen B, et al. Circle of Willis: morphologic variation on three-dimensional time-of-flight MR angiograms. Radiology 1998;207:103–11. [14] Jongen JC, Franke CL, Ramos LM, Wilmink JT, van Gijn J. Direction of flow in posterior communicating artery on magnetic resonance angiography in patients with occipital lobe infarcts. Stroke 2004;35:104–8. [15] van Raamt AF, Mali WP, van Laar PJ, van der Graaf Y. The fetal variant of the circle of Willis and its influence on the cerebral collateral circulation. Cerebrovasc Dis 2006;22:217–24. [16] Park JH, Kim JM, Roh JK. Hypoplastic vertebral artery: frequency and associations with ischaemic stroke territory. J Neurol Neurosurg Psychiatry 2007;78:954–8. [17] Chuang YM, Huang YC, Hu HH, Yang CY. Toward a further elucidation: role of vertebral artery hypoplasia in acute ischemic stroke. Eur Neurol 2006;55:193–7. [18] Caplan LR. Arterial occlusions: does size matter? J Neurol Neurosurg Psychiatry 2007;78:916. [19] Perren F, Poglia D, Landis T, Sztajzel R. Vertebral artery hypoplasia. A predisposing factor for posterior circulation stroke? Neurology 2007;68:65–7. [20] Giannopoulos S, Markoula S, Kosmidou M, Pelidou HS, Kyritsis A. Lateral medullary ischemic events in young adults with hypoplastic vertebral artery. J Neurol Neurosurg Psychiatry 2007;78:987–9. [21] Schöning M, Hartig B. The development of hemodynamics in the extracranial carotid and vertebral arteries. Ultrasound Med Biol 1998;24:655–62. [22] Bartels E. Vertebral sonography. In: Bartels E, editor. Color-coded duplex ultrasonography of the cerebral vessels: Atlas and manual. New York: Schattauer; 1999. p. 113–55.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Ischemic stroke patterns and hemodynamic features in patients with small vertebrobasilar artery Jong-Ho Park a, Jae-Kyu Roh b,⁎, Hyung-Min Kwon c a b c
Department of Neurology, Myongji Hospital, Kwandong University College of Medicine, Goyang, Republic of Korea Department of Neurology, Seoul National University Hospital, Seoul, Republic of Korea Department of Neurology, Seoul National University Boramae Hospital, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 31 December 2008 Received in revised form 15 June 2009 Accepted 7 July 2009 Available online 13 August 2009 Keywords: Small vertebrobasilar artery Long circumferential artery Ischemic pattern MR imaging Angiography
a b s t r a c t Background: To determine the role of small vertebrobasilar artery (SVBA) in patients with posterior circulation stroke (PCS), we evaluated the ischemic patterns, collateral features, and stroke mechanisms in PCS patients with SVBA. Methods: Ischemic findings on magnetic resonance (MR) imaging were correlated with 3D time-of-flight/ contrast-enhanced MR angiography and/or catheter angiography in 18 patients (mean age, 68.0 ± 11.8 years; 9 males). SVBA (lumen diameter of b 3 mm) was compared with stenotic normal-sized VBA (NVBA) in 14 PCS patients. Results: Ischemic lesions were predominantly observed in the cerebellum and/or medulla (vertebral artery (VA) territory). All subjects had fetal posterior circulation (FPC) from the internal carotid artery to the posterior cerebral artery. Sixteen patients (88.9%) had distal or diffuse VA stenosis/occlusion. In 14 patients (77.8%), long circumferential artery (LCA) was prominently observed. In atherothrombotic patients, infratentorial PCS might occur following artery-to-artery embolism from the low-flowed or stenotic VA to LCA. Ischemic patterns between subjects with and without VA disease were almost similar. As the degree of VA disease increased, the frequency of LCA prominence showed an increased tendency (P = 0.003). Relatively small, scattered infarcts were observed in patients with SVBA than in those with stenotic NVBA. Conclusions: FPC does not protect against infratentorial PCS. Regardless of extensive arterial lesions, relatively small infarcts may be due to previously established collaterals from the LCA, which could compensate for the defects in the infratentorial area. © 2009 Elsevier B.V. All rights reserved.
1. Introduction A small vertebrobasilar artery (SVBA) is infrequently observed in stroke patients, regardless of whether it is directly related to ischemic stroke. It may not be possible to adequately visualize an SVBA on threedimensional (3D) time-of-flight (TOF) magnetic resonance (MR) angiography (MRA); moreover, because its exact nature is unclear, it attracts little attention until posterior circulation stroke (PCS) occurs. Ischemic patterns in patients with SVBA appear to be different from those in subjects with a normal-sized VBA (NVBA). Although it has been reported that SVBA might induce PCS [1], the collaterals that would compensate for the posterior circulation system in the case of ischemic stroke have not been well established. Furthermore, a correlation of neuroimaging findings with vascular pathologies has not been attempted in PCS patients with SVBA. Therefore, we evaluated the ischemic pattern
⁎ Corresponding author. Department of Neurology, Seoul National University College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-744, Republic of Korea. Tel.: +82 2 2072 3265; fax: +82 2 3672 4949. E-mail address: [email protected] (J.-K. Roh). 0022-510X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2009.07.007
of the vascular lesions by magnetic resonance (MR) imaging and the hemodynamic pattern (collateral circulation) by MR angiography (MRA) and/or transfemoral cerebral angiography (TFCA) and assessed the stroke mechanism in PCS patients with SVBA by using the abovementioned techniques together with transcranial Doppler (TCD). 2. Methods 2.1. Patient selection The present study is a case series from 2 academic centers (Seoul National University Hospital (SNUH) and Myongji Hospital (MJH)). We assessed 1959 acute ischemic stroke patients (SNUH: n=1604, between September 2003 and February 2006; MJH: n=355, between March 2006 and September 2007). Within 1 week of symptom onset, patients underwent brain MRI and 3D TOF MRA by a 1.5-Tesla MR system (SNUH: Siemens 1.5 Vision, Erlangen, Germany; MJH: Intera 1.5 T 10.3 version, Eindhoven, Netherlands). The stroke types were categorized as follows: anterior circulation stroke (ACS) =1120 (57.2%), PCS=640 (32.7%), ACS plus PCS=33 (1.7%), transient ischemic attack (TIA)=162 (8.3%), and venous sinus thrombosis=4 (0.2%).
228
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
Table 1 Summary of 18 PCS patients with small vertebrobasilar artery. No
Sex/age
Risk factor(s)
VA lesion
BA lesion
Prominence
FPC
TCD
1 2a 3a 4 5b 6a 7a 8a 9a
F/53 M/69 F/69 F/76 M/87 F/76 M/69 M/73 M/70
HT HT HT, DM, Af HT HT HT Af HT, Af HT, DM, smoking
– – L (diffuse) L (V4) L (V3 and V4) L (V4) R (diffuse) R (V4) R (diffuse)
– – D – – M – – –
PICA (R) – PICA (B), AICA (R) AICA (L) PICA (L) PICA (B) – – –
B B B B B B U B U
10a 11b 12b,c 13a,b,c 14b,c
F/77 F/53 F/58 F/65 M/53
HT, DM HT HT, DM HT HT
R (V4), L (diffuse) B (VBA junction) B (VBA junction) B (V4) B (V4)
– P – – P
PICA PICA PICA PICA PICA
(R) (B) (B) (B) (B)
B B B B B
15a,c
M/87
HT
B (V4)
–
PICA (B)
B
16a,b,c
M/46
–
P
PICA (L), AICA (L)
B
17a,b
F/64
HT
R (diffuse) L (V4) B (V4)
– Decreased MFV and turbulent waves of (B) VBAs (b 30 cm/s) – – Stenosis of BA (N80 cm/s) Stenosis of (B) VBAs (N80 cm/s) Dampened signal of (B) VBAs – Dampened signal of (L) VA and BA Stenosis of (R) VA (N 110 cm/s) Blunted signal of BA – Blunted signal of (B) VBAs Dampened signal of (L) VA and BA bidirectional flow of (R) VA Stenosis of BA (N140 cm/s) Dampened signal of (L) VA Dampened signal of (L) VA Absent signal of (R) VA and BA –
–
AICA (B)
B
M/79
HT
B (diffuse)
P
AICA (B), SCA (B)
B
18
a,b,c
Stenosis of BA (N120 cm/s) Dampened signal of (B) VAs –
PCS, posterior circulation stroke; VA, vertebral artery; BA, basilar artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; FPC, fetal posterior circulation; HT, hypertension; DM, diabetes mellitus; Af, atrial fibrillation; L, left; R, right; U, unilateral; B, bilateral; P, proximal; M, middle; D, distal; TFCA, transfemoral cerebral angiography; MFV, mean flow velocity. a Contrast-enhanced MR angiography performed. b TFCA performed. c Retrograde filling of BA from anterior circulation.
2.2. SVBA definition The mean diameter of the normal basilar artery (BA) has been reported to be 3.17 mm [2] and the hypoplastic vertebral artery (VA) was defined to have a lumen diameter of less than 2–3 mm [3,4]. Because there is no consensus on the value of SVBA, we defined SVBA as a diameter of less than 3 mm. For defining the diameter of SVBA, we examined the mid-portion level of the BA and the V2 of the largest VA by using magnified images of MRA on a picture archiving and communication system (PACS). The location of the VA was defined according to standard criteria [5,6] (V1; prevertebral portion, V2; the portion within the vertebral columns, V3; atlantoaxial portion, V4; intracranial portion). 2.3. Study patients We assessed 37 acute stroke patients with SVBA. MRI and MRA were obtained at 2.3 ± 1.1 days after stroke onset. We included patients whose configuration of SVBA was visible on 3D TOF MRA. We also included those who had undergone angiographic work-up (e.g., contrast-enhanced MRA or TFCA) even if their SVBA was not or only faintly visible on routine 3D TOF MRA. Patients were excluded if they showed no visible VBA (n = 7) or only a rudimentary VBA (n = 3) on 3D TOF MRA; moreover, these 10 patients had not undergone detailed angiographic work-up that could enable a diagnosis of SVBA. Among the excluded 10 patients, 9 had been afflicted by PCS and the remaining one by ACS. Those who showed no relevant lesions on diffusion-weighted images (DWI; n = 2) were also excluded. Of the 25 stroke patients with SVBA (7 with ACS and 18 with PCS), 18 PCS patients who underwent MRI and MRA at 2.2 ± 1.2 days after the
onset of symptoms were eventually selected for the study. Acute ischemic lesion was defined by high-signal intensity on DWI with lowsignal intensity on the apparent diffusion coefficient map (only 2 patients [nos. 4 and 18] by T2-weighted image; T2-WI). Of these 18 patients, 12 patients (66.7%; nos. 2, 3, 6–10, 13, and 15–18) underwent contrast-enhanced MRA with the same TOF MR sequence to evaluate the course of the VBA. We assessed 30 acute PCS patients with NVBA between March 2007 and September 2007 at MJH and considered them as the control group. We regarded NVBA as a diameter of more than 3 mm, except in the case of unilateral hypoplastic VA. We obtained MRI and 3D contrast-enhanced MRA scans for the control subjects at 2.2 ± 1.4 days after the onset of symptoms. The topography of the infarcts was determined with reference to the maps, establishing an anatomical correspondence with the dominant arterial territories of the brainstem and cerebellum [7]. TFCA was performed in 8 patients with SVBA (nos. 5, 11–14, and 16–18) and 2 patients with NVBA (nos. 7 and 14) after an informed consent was obtained from each patient. The possibility of VBA dissection could be excluded by the findings of TFCA (string sign) or source view of MRA (double lumen) [8]. The long circumferential artery (LCA) (e.g., posterior inferior [PICA], anterior inferior [AICA], or superior cerebellar artery [SCA]) was regarded to be prominent if it was well depicted on MRA or TFCA compared to its parent SVBA. In patients with NVBA, LCA was regarded to be present if it was adequately visible on MRA or TFCA. The circle of Willis was graded as fetal posterior circulation (FPC) when one of the P2 segments of the posterior cerebral artery was supplied by the internal carotid artery via the posterior communicating artery (absent or hypoplastic P1 segment of the posterior cerebral
Fig. 1. Ischemic stroke patterns according to the various vascular lesions observed in 18 patients with small VBA. Group I: no VA disease, with or without LCA prominence (n = 2), Group II: unilateral VA disease, with (upper) or without (low) LCA prominence (n = 7), and Group III: bilateral VA disease, with LCA prominence VA, vertebral artery; BA, basilar artery; LCA, long circumferential artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; LAA, large artery atherosclerosis; SAO, small artery occlusion; CE, cardiac embolism.
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
229
230
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
Fig. 2. Patient 2. A, Diffusion-weighted imaging shows an infarct in the territory of the right PICA. B-1 and B-2, 3D TOF MRA shows intact VAs caused by possible in situ thrombosis of PICA. Patient 6. A, Diffusion-weighted imaging shows an infarct in the right pons. B, 3D TOF and contrast-enhanced MRA show steno-occlusive lesions of the mid-BA and the left distal VA (V4). Prominence of the right PICA (arrow) is noted. Patient 9. A-1, 2, 3, 4, Diffusion-weighted imaging shows an infarct in the right medial cerebellum, both temporo-occipital lobes, and left medial thalamus (arrows). B, Contrast-enhanced MRA shows diffuse steno-occlusive lesion in the right VA with no prominence of LCA. Patient 12. A-1, 2, Diffusionweighted imaging shows an infarct in the right PICA territory. B, 3D TOF MRA shows no visible vertebrobasilar system. C-1, 2, TFCA depicts prominent PICA ending (arrow) from both distal VAs. Patient 17. A, Diffusion-weighted imaging shows an infarct in the left focal medial cerebellum (PICA territory). B, 3D TOF and contrast-enhanced MRA depict both distal VA stenoses (V4). C, TFCA shows both AICA (white arrow) prominences adjacent to both distal stenotic VAs (black arrow). VA, vertebral artery; BA, basilar artery; LCA, long circumferential artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; TOF, time-of-flight; TFCA, transfemoral cerebral angiography.
Table 2 Associations of VA disease with prominence of LCA in patients with SVBA. Prominence of LCA VA diseaseb
None Unilateral Bilateral
P
None (n = 4)
One (n = 5)
≥ Two (n = 9)a
1 (25.0) 3 (75.0) 0 (0)
1 (20.0) 3 (60.0) 1 (20.0)
0 (0) 1 (11.1) 8 (88.9)
Values are number (percentage) of patients. LCA, long circumferential artery; VA, vertebral artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery. a Including concurrent PICA, AICA, and SCA. b distal or diffuse VA stenosis/occlusion.
0.003
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235 231
Fig. 3. Ischemic stroke patterns in 14 patients with stenotic normal-sized VBA. VA, vertebral artery; BA, basilar artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; LAA, large artery atherosclerosis.
232
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
artery on that side) [9]. MRA or TFCA findings were copied from the original film and analyzed by one of the authors (J.-H.P.) who was blinded to the MRI findings. MRI findings (DWI or T2-WI) were schematically drawn by another author (H.-M.K.) who was blinded to the MRA findings.
respectively. The analyses of the frequencies of LCA prominence according to the number of VA disease were conducted using a linear by linear association test. They were conducted using SPSS 11.5 for Windows (SPSS Inc., Chicago, IL, USA). Two-sided P values of 0.05 or less were considered significant.
2.4. MRI and MRA analyses
3. Results
MRI was performed using two 1.5-Tesla MR systems with T2-WI (repetition time/echo time (TR/TE) of 5000/99 ms in the Siemens system and 4230/100 ms in the Intera system), DWI (TR/TE of 6500/ 110 ms in Siemens and 3737/66 ms in Intera), and a fluid-attenuated inversion recovery (FLAIR) image (TR/TE of 9000/119 ms in Siemens and 11,000/140 ms in Intera). MRA was performed using the 3D TOF technique (older Signa unit, TR/TE of 35/7.2 ms in Siemens and 23/ 6.9 ms in Intera). 3D spoiled gradient-echo acquisition was applied using the following parameters: TR/TE/excitation (37/6.6/1 in Siemens and 21/5/1 in Intera); excitation angle (20° in Siemens and 30° in Intera); field of view (230 × 6/8 in both Siemens and Intera); and acquisition matrix (195 × 512 in Siemens and 368 × 368 in Intera). All MRA procedures included intracranial and neck vessels (except in patient no. 9) and were performed using a 3D TOF technique with or without 3D contrast-enhanced MRA, which was performed after a 10-mL intravenous bolus injection of Magnevist (Bayer HealthCare Pharmaceuticals, Wayne, NJ, USA) with the same TR/TE as each 3D TOF MRA.
3.1. Demographic features, risk factors, and stroke subtypes of the study subjects In this study, 9 men and 9 women were included. None of the study patients had experienced a VBA dissection or stenosis (N50%) or occlusion of the internal carotid artery. Table 1 lists the baseline characteristics and angiographic findings of the study patients with SVBA. The mean age of the patients was 68.0 ± 11.8 years (range, 46– 87 years). Most of the patients had vascular risk factors: hypertension in 15 patients (83.3%), diabetes in 4 patients (22.2%), Af in 3 patients (16.7%), and smoking in 1 patient (5.6%). The index stroke subtypes of 18 patients according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification [12] were large artery atherosclerosis (LAA) in 12 patients (66.7%), small artery occlusion (SAO) in 3 patients (16.7%), stroke with undetermined etiology (LAA with cardiogenic embolism [CE]) in 2 patients (11.1%), and CE in 1 patient (5.6%). 3.2. TCD findings in study subjects
2.5. TCD examination TCD was performed in patients with SVBA with a 2-MHz pulsed Doppler instrument (SNUH: PMD 100, Spencer Technologies, Seattle, WA, USA; MJH: Companion III, Viasys Healthcare, Warwick, UK) by an experienced sonographer. The posterior circulation was assessed according to the standard protocol [10]. Stenosis of the VBA was diagnosed if the mean flow velocity was greater than 75 cm/s. The residual flow grades of the SVBA were assessed using the Thrombolysis in Brain Ischemia (TIBI) flow classification [11]. 2.6. Definition of risk factors We assessed the risk factors in each patient for stroke, such as hypertension, diabetes, current smoking, and heart disease. Cardiogenic embolic diseases (e.g., atrial fibrillation (Af), left ventricular akinesia, recent myocardial infarction, or patent foramen ovale with right-to-left shunt) were routinely evaluated in all patients by electrocardiography and transthoracic echocardiography; further studies—such as transechocardiography or 24-h Holter monitoring— were performed if any abnormality was found on initial work-up. 2.7. Statistical analysis The analyses of the general categorical variables and continuous variables were conducted using the χ2 tests and Student's t tests,
Of the 18 patients, 11 (nos. 2, 5–7, 9, 10, 12–15, and 17) underwent TCD examination (Table 1). All patients, except 2 (nos. 6 and 7), had an index stroke subtype of LAA. Those with LAA showed stenosis or poor perfusion state (from blunted to absent signal) of VA and/or BA on TCD. 3.3. Characteristics of ischemic findings Brain MRI findings are presented in Fig. 1. Generally, patients had relatively small-sized infarcts with a scattered pattern (i.e., not distributed throughout the arterial territory). In 13 patients (72.2%), ischemic lesions were predominantly observed in the cerebellum and/ or medulla in the VA territory. In 4 patients (22.2%), ischemic lesions were observed in the territory of the BA or its main branches. In 6 patients (33.3%; nos. 7–9, 11, 14, and 16), ischemic lesions were observed in multiple (i.e., more than 2) arterial territories. Ischemic patterns of the patients with LAA were mostly of the artery-to-artery embolism type. 3.4. Angiographic findings of the study subjects by MRA/TFCA FPC from the internal carotid artery to the posterior cerebral artery was noted in all patients with SVBA (100%). Distal VA disease (stenosis or occlusion) adjacent to the distal portion of the PICA orifice or diffuse VA disease was noted in 16 patients (88.9%; bilateral in 9). BA
Fig. 4. Patient 1. A-1 and A-2, Diffusion-weighted imaging shows an acute infarct in the right PICA territory and left medial occipital lobe. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the right distal VA (V4). Patient 4. A, Diffusion-weighted imaging shows an acute infarct in the left lateral medulla. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the left distal VA (V4 and distal V3). Patient 7. A-1 and A-2, Diffusion-weighted imaging shows an acute infarct in the left lateral medulla (lower portion) and left PICA territory. B-1 and B-2, 3D TOF and contrast-enhanced MRA show diffuse steno-occlusive lesion in the left VA. Patient 8. A, Diffusionweighted imaging shows an acute infarct in both PICA territories. B-1 and B-2, 3D TOF and contrast-enhanced MRA show steno-occlusive lesion in the right distal VA (V4 and distal V3) and the left focal V4. Right AICA is seen on the contrast-enhanced MRA (arrow). Patient 11. A-1, A-2 and A-3, Diffusion-weighted imaging shows an acute small infarct in the left PICA territory and right pons. C, Contrast-enhanced MRA and TFCA depict focal stenosis (arrow) of the left VA (V2). Patient 12. A, Diffusion-weighted imaging shows an acute small infarct in the right PICA territory. B-2, Contrast-enhanced MRA depicts focal stenosis of the right VA (V3). Patient 13. A, Diffusion-weighted imaging shows an acute small, multiple infarcts in the right PICA and SCA territories. B-1 and B-2, 3D TOF and contrast-enhanced MRA depict focal stenosis in both the VAs (right V2 and left V2 and V4). Patient 14. A-1, A-2, A-3 and A-4, Diffusion-weighted imaging shows multiple acute small infarcts in the right PICA, SCA, and both the occipital lobes. B-1 and B-2, 3D TOF and contrast-enhanced MRA depict focal stenosis of both VAs (right focal V3–4 and left focal V4). TFCA shows right AICA prominence (arrow) and focal stenosis of ipsilateral VA (arrowhead). VA, vertebral artery; PICA, posterior inferior cerebellar artery; AICA, anterior inferior cerebellar artery; SCA, superior cerebellar artery; TOF, time-of-flight; CE, contrast enhanced; TFCA, transfemoral cerebral angiography.
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
233
234
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
stenosis or occlusion was noted in 6 patients (proximal portion in 4, middle in 1, and distal in 1). Prominence of LCA was noted in 14 patients (77.8%), i.e., PICA was prominent in 11 patients (61.1%); AICA in 5 patients (27.8%); and SCA in 1 patient (5.6%). Of the 8 patients who underwent TFCA, 5 (62.5%; nos. 12–14, 16, and 18) showed retrograde filling of BA from anterior circulation. 3.5. Ischemic stroke patterns according to various vascular lesions Of the patients studied, 16 had VA disease; among these, 12 had distal VA disease, while 6 had diffuse VA disease. Consequently, we classified them as follows (Figs. 1 and 2): Group I: no VA disease, with or without LCA prominence (n = 2, 11.1%); Group II: unilateral VA disease, with or without LCA prominence (n = 7, 38.9%); and Group III: bilateral VA disease, with LCA prominence (n = 9, 50.0%). 3.5.1. No VA disease, with or without LCA prominence (Group I) Only 2 patients did not have VA disease. Of them, 1 showed PICA prominence (no. 1) and 1 did not (no. 2). 3.5.1.1. Presumed etiopathogenesis (subtype). The causes included SAO in 1 patient (no. 1) and LAA in the other (no. 2). 3.5.1.2. MRI findings. The lesions were located in the paramedian pons in 1 patient (no. 1) and in the medial PICA territory in another (no. 2). 3.5.2. Unilateral VA disease, with or without LCA prominence (Group II) Of the 7 patients with unilateral VA disease, 4 showed PICA or concurrent AICA prominence (patient nos. 3–6) and the other 3 showed no prominence (patient nos. 7–9). 3.5.2.1. Presumed etiopathogenesis (subtype). A variety of causes were responsible for the pathogenesis in these patients: LAA was evident in 2 patients (nos. 5 and 9); SAO in 2 patients (nos. 4 and 6); stroke with undetermined etiology (LAA plus CE) in 2 patients (nos. 3 and 8); and CE in 1 patient (no. 7). 3.5.2.2. MRI findings. The lesions were variously located in the center of PICA territory (patient nos. 5, 8, and 9); the medial medulla and AICA territory (patient no. 7); the pons (patient nos. 4 and 6); and the posterior cortical area, with or without deep gray matter involvement (patient nos. 3, 8, and 9). However, the lesions mostly originated in the VA territory (patient nos. 5 and 7–9). Of the 4 patients with PICA or/ concurrent AICA prominence, 2 patients (nos. 3 and 5) had single and relatively small infarcts in the occipital cortex and cerebellum; the other 2 patients had SAO in the pons. Moreover, in the 3 patients lacking PICA or AICA prominence (nos. 7–9), scattered infarcts were observed in multiple territories. 3.5.3. Bilateral VA disease with LCA prominence (Group III) Of the 9 patients with bilateral VA disease, 6 showed PICA prominence (patient nos. 10–15); 1, concurrent PICA and AICA prominence (patient no. 16); 1, AICA prominence (patient no. 17); and 1, concurrent AICA and SCA prominence (patient no. 18). None of the patients were without AICA/SCA/PICA prominence. 3.5.3.1. Presumed etiopathogenesis (subtype). sumed cause for all patients was LAA.
In this group, the pre-
3.5.3.2. MRI findings. Lesions were located predominantly in the cerebellum (PICA territory; patient nos. 10 and 12–18). Almost all subjects with cerebellar lesions had relatively small and scattered patterns (i.e., not all of the territory was involved). Concurrent pons or medial medulla involvement was observed in 2 patients (nos.14 and 16) and bilateral mesencephalo-pontine involvement because of stenoocclusion of the distal VA and proximal BA was observed in 1 (no. 11).
3.6. Associations of VA disease with the LCA prominence in patients with SVBA Table 2 demonstrates the association between VA disease and compensatory collaterals from the LCA (PICA/AICA/SCA). We observed that as the degree of VA disease increased (i.e., from “none” to “unilateral” to “bilateral”), the frequency of LCA prominence (i.e., “none,” “one,” and “two or more”) tended to increase (P = 0.003). 3.7. Comparison of ischemic patterns between patients with stenotic SVBA and those with stenotic NVBA Of the 30 PCS patients with NVBA, 14 had an index stroke subtype of LAA. As most of the study patients had VA disease, we intended to compare the ischemic patterns of the 16 patients with stenotic SVBA (Groups II and III) with those of 14 patients with stenotic NVBA who underwent MR study at 2.1 ± 1.3 days after the onset of symptoms. Supplemental table illustrates that baseline demographics and classic risk factors were similar between the 2 groups, except Af. FPC, bilateral VA disease, and LCA presence were more prevalent in the patients with stenotic SVBA (P b 0.0001, P = 0.052, and P = 0.011, respectively). Brain MRI and MRA findings of the stenotic NVBA group are presented in Figs. 3 and 4. Lesions were variously located in the PICA territory (patient nos. 1–3 and 7–14); the lateral medulla (patient nos. 4–7); SCA territory (patient nos. 13 and 14); the occipital area (patient nos. 1 and 14); and the pons (patient no. 11). The stenotic NVBA group showed relatively large, conglomerate infarct patterns compared with those of stenotic SVBA group. However, the ischemic findings of 4 patients with NVBA (patient nos. 11–14) were similar to those of SVBA group patients. They had common feature that showed extracranial focal VA lesion (below the V3). Three patients (nos. 8, 13, and 14) with bilateral VA diseases showed multiple lesions involving more than 2 territories. 4. Discussion To date, few studies have focused on SVBA, except to assess to what extent SVBA might result in PCS [1]. Our study focused on the MR ischemic patterns, hemodynamic features (collateral circulation), and possible stroke mechanisms in PCS patients who showed small or faintly visible VBA on MRA. FPC is a fetal variant of the posterior cerebral artery (PCA) from the internal carotid artery. The prevalence of FPC is reported to be 32% in the general population [13]. The present study showed the existent varieties of FPC (bilateral occurrence in 88.9% of patients), and suggested that FPC may compensate the posterior circulation zone for the hemodynamic insufficiency caused by SVBA. However, FPC could be protective against occipital lobe infarcts only [14]. Since the cerebellar tentorium impedes the formation of a leptomeningeal connection, FPC does not contribute to the perfusion of the infratentorial area [15]. Consequently, FPC makes the development of leptomeningeal collaterals between the ICA and the vertebrobasilar system impossible [15]. Our study showed that most of the infratentorial lesions originated from the cerebellum and/or medulla (VA territory) or the pons (BA territory). These results are consistent with the observation [15] that FPC would not be able to protect the infratentorial area against PCS. We found that the ischemic patterns in PCS patients were generally characterized by relatively small (i.e., not all of the territory was involved) and/or scattered lesions, while stenotic NVBA group showed relatively large, conglomerate infarct patterns except for extracranial focal VA lesion (below V3). The most common angiographic finding was VA stenosis or occlusion in which distal lesions were predominant. This is why infratentorial lesions mainly originate in the cerebellum and/or medulla. The pattern of ischemic lesions between patients with and without VA disease and between patients with differing grades of VA disease did not differ significantly, as observed
J.-H. Park et al. / Journal of the Neurological Sciences 287 (2009) 227–235
by MRI. However, the involvement of multiple arterial territories was noted in 6 patients (nos. 7–9, 11, 14, and 16). All of the 6 patients had VA disease with the etiopathogenesis predominantly being LAA (nos. 9, 11, 14, and 16). Regardless of the presence of extensive arterial lesions, relatively small infarcts can be attributed to the established leptomeningeal collaterals from the LCA (e.g., PICA, AICA, or SCA) that can compensate for the defects in the distal vertebrobasilar arterial territories (i.e., infratentorial area). Thus, the degree of collateral development along with a chronic process of atherothrombosis may determine the pattern (particularly, the size) of an ischemic lesion. However, in patients with unilateral VA disease without LCA prominence (nos. 8 and 9), the PICA/AICA/SCA which could have been well delineated on TFCA was not visible on MRA. Our study may raise the question of whether the SVBA is of congenital origin or a consequence of multiple or longitudinal atherosclerotic narrowing. Embryologically, if the BA does not become the main source of blood supply to the developing posterior cerebral arteries, the FPC might persist and remain large in size [1]. The observations that all the 18 patients had FPC and that the FPC was larger than the VBAs may support the hypothesis that the SVBA is congenitally small rather than acquired. Are the smaller arteries more vulnerable to atherothrombosis? It is known that there is a significant relationship between hypoplastic VA and ipsilateral PCS [16,17]. A small-diameter artery appears to be vulnerable to stenosis or occlusion if vascular risk factors are present [18], as reported in some studies [16,19,20]. A small VA shows lower mean flow volume [17,21,22] and decreased flow velocities [22] in color-coded duplex ultrasonography. It is plausible that SVBA could have low flow and slow velocity and therefore be susceptible to prothrombotic or atherosclerotic processes in the presence of vascular risk factors. In our study, the stroke mechanism was predominantly atherothrombotic embolism (66.7%). Our patients with LAA, for whom we performed TCD, generally showed a low-flow or stenotic state of VBA circulation. Considering the ischemic patterns together with the TCD findings, infratentorial PCS may occur due to an arteryto-artery embolism from the low-flowed or stenotic VA to the LCA. Our study has some limitations. First, our study comprised a small number of patients because of our exclusion criteria; this may have resulted in a selection bias. Second, inconsistent diagnostic imaging for angiography (TOF, contrast-enhanced MRA, or TFCA) might have affected the external validity of the hemodynamic collateral findings. In some patients, advanced arterial narrowing from the VA orifice made it difficult to access the entire VBA by TFCA. Finally, we could not perform detailed comparative analysis between the 2 groups (SVBA vs. NVBA) because there were only 2 patients with NVBA who underwent TFCA. In fact, there is no rationale for performing TFCA in PCS patients with NVBA only for comparing the angiographic findings with those of patients with SVBA. Although this was an observational study, we suggest that relatively small infarcts can be attributed to the establishment of leptomeningeal collaterals from the LCA in PCS patients with extensive arterial lesions.
235
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jns.2009.07.007. References [1] Chaturvedi S, Lukovits TG, Chen W, Gorelick PB. Ischemia in the territory of a hypoplastic vertebrobasilar system. Neurology 1999;52:980–3. [2] Smoker WR, Price MJ, Keyes WD, Corbett JJ, Gentry LR. High-resolution computed tomography of the basilar artery: normal size and position. AJNR Am J Neuroradiol 1986;7:55–60. [3] Fisher CM, Gore I, Okabe N, White PD. Atherosclerosis of the carotid and vertebral arteries—extracranial and intracranial. J Neuropathol Exp Neurol 1965;24:455–76. [4] Touboul PJ, Bousser MG, LaPlane D, Castaigne P. Duplex scanning of normal vertebral arteries. Stroke 1986;17:921–3. [5] Argenson C, Francke JP, Sylla S, Dintimille H, Papasian S, di Marino V. The vertebral arteries (segments V1 and V2). Anat Clin 1980;2:29–41. [6] Francke JP, di Marino V, Pannier M, Argenson C, Libersa C. The vertebral arteries (arteria vertebralis). The V3 atlanto-axial and V4 intracranial segments-collaterals. Anat Clin 1981;2:229–42. [7] Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of human brain; brainstem and cerebellum. Neurology 1996;47:1125–35. [8] Auer A, Felber S, Schmidauer C, Waldenberger P, Aichner F. Magnetic resonance angiographic and clinical features of extracranial vertebral artery dissection. J Neurol Neurosurg Psychiatry 1998;64:474–81. [9] van der Grond J, van Raamt AF, van der Graaf Y, Mali WPTM, Bisschops RHC. A fetal circle of Willis is associated with a decreased deep white matter lesion load. Neurology 2004;63:1452–6. [10] Alexandrov AV. Transcranial Doppler sonography: principles, examination technique, normal values, and waveform patterns. Vasc Ultrasound Today 1998;3:141–60. [11] Demchuk AM, Burgin WS, Christou I, Felberg RA, Barber PA, Hill MD, et al. Thrombolysis in Brain Ischemia (TIBI) transcranial Doppler flow grades predict clinical severity, early recovery, and mortality in patients treated with intravenous tissue plasminogen activator. Stroke 2001;32:89–93. [12] Adams Jr HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al. Classification of subtype of acute ischemic stroke. Definitions for use in multicenter clinical trial. TOAST. Stroke 1993;24:35–41. [13] Krabbe Hartkamp MJ, Van der Grond J, de Leeuw FE, de Groot JC, Algra A, Hillen B, et al. Circle of Willis: morphologic variation on three-dimensional time-of-flight MR angiograms. Radiology 1998;207:103–11. [14] Jongen JC, Franke CL, Ramos LM, Wilmink JT, van Gijn J. Direction of flow in posterior communicating artery on magnetic resonance angiography in patients with occipital lobe infarcts. Stroke 2004;35:104–8. [15] van Raamt AF, Mali WP, van Laar PJ, van der Graaf Y. The fetal variant of the circle of Willis and its influence on the cerebral collateral circulation. Cerebrovasc Dis 2006;22:217–24. [16] Park JH, Kim JM, Roh JK. Hypoplastic vertebral artery: frequency and associations with ischaemic stroke territory. J Neurol Neurosurg Psychiatry 2007;78:954–8. [17] Chuang YM, Huang YC, Hu HH, Yang CY. Toward a further elucidation: role of vertebral artery hypoplasia in acute ischemic stroke. Eur Neurol 2006;55:193–7. [18] Caplan LR. Arterial occlusions: does size matter? J Neurol Neurosurg Psychiatry 2007;78:916. [19] Perren F, Poglia D, Landis T, Sztajzel R. Vertebral artery hypoplasia. A predisposing factor for posterior circulation stroke? Neurology 2007;68:65–7. [20] Giannopoulos S, Markoula S, Kosmidou M, Pelidou HS, Kyritsis A. Lateral medullary ischemic events in young adults with hypoplastic vertebral artery. J Neurol Neurosurg Psychiatry 2007;78:987–9. [21] Schöning M, Hartig B. The development of hemodynamics in the extracranial carotid and vertebral arteries. Ultrasound Med Biol 1998;24:655–62. [22] Bartels E. Vertebral sonography. In: Bartels E, editor. Color-coded duplex ultrasonography of the cerebral vessels: Atlas and manual. New York: Schattauer; 1999. p. 113–55.