multiple acute infarcts in the posterior circulation

Journal of Neurology, Neurosurgery, and Psychiatry 1996;60:289-296

Multiple acute infarcts in the posterior circulation

Andrea Bemasconi, Julien Bogousslavsky, Claudio Bassetti, Franco Regli

AbstractObjective-to evaluate clinical, radiologi-cal, and prognostic features of patientswith multiple acute infarcts in remotearterial territories of the posterior circu-lation.Design-Data analysis from a prospectiveacute stroke registry in a communitybased primary care centre using astandard protocol including MRI andMRA.Results-In three and a half years, 27 ofthe 236 patients (11%) with posterior cir-culation stroke had multiple acuteinfarcts in the posterior circulation asshown by gadolinium enhancement onMRI. Eighteen patients had multipleinfratentorial and supratentorial infarctsincluding the cerebellum and posteriorcerebral artery territory, with coexistingbrainstem involvement in seven patients.Fourteen patients had a rostral basilarartery syndrome and cerebellar signs;four patients had a visual field defect withcerebellar signs. Causes were vertebral(six) or basilar (four) artery atheromato-sis, and cardioembolism (four). Sevenpatients had multiple acute infarcts in theposterior circulation of the cerebellumand lower brainstem. Brainstem andcerebellar signs were found in mostpatients (five); aetiologies were small ves-sel disease (four), cardioembolism (one),and vertebral artery dissection (one). Twopatients with large artery atheromatosishad multiple acute infarcts in the poste-rior circulation in the brainstem and pos-terior cerebral artery territory. Onemonth after stroke more than 25% of thepatients were dependent or had died.There was no difference in the outcomebetween the three groups, and recoverywas linked to the size of infarcts ratherthan to a high number of infarcts.Conclusions-multiple acute infarcts inthe posterior circulation usually involvethe cerebellum. Simultaneous brainstemand posterior cerebral artery territoryinfarcts sparing the cerebellum areuncommon. They can be suspectedclinically before neuroimaging, mainlywhen supratentorial and infratentorialinfarcts coexist. This may be important,because different patterns of infarctionare associated with different causes ofstroke.

(J Neurol Neurosurg Psychiatry 1996;60:289-296)

Keywords: acute multiple infarcts; posterior circula-tion; magnetic resonance imaging

Although most clinical topographic strokestudies have been limited to patients withinvolvement of single vascular territories,acute infarction may involve more than oneterritory. -20

Because of its sensitivity,2' MRI may detectmultiple infarcts that have been overlooked inprevious studies, mainly due to the low yieldof CT in infratentorial infarction.8 22-24We prospectively studied clinical, imaging,

aetiological, and functional characteristics ofall patients with multiple acute infarcts in theposterior circulation who were included in theLausanne Stroke Registry25 over three and ahalf years. We used a standard protocol ofinvestigations including MRI and magneticresonance angiography (MRA).

MethodsWe studied all patients who were admitted toour population based primary care centre over42 months with a first stroke in the posteriorcirculation. These patients represented a sub-sample of 800 patients included consecutivelyin the Lausanne Stroke Registry. The maincriterion for inclusion into the study was thepresence of more than one acute infarctinvolving remote arterial territories in the pos-terior circulation, shown by gadoliniumenhancement on MRI.The infarcts were demonstrated by high

field MRI (Siemens Magnetom SP 63 1-5Tesla) usually conducted within one week andin all patients within 13 days of stroke onset.Studies with MRI included TI, T2, and pro-ton weighted images (5 mm thick slices); con-trast images ( Ti sagittal and transverse 5 mmslices) were obtained after injection of gadolin-ium. Three dimensional MRA of vertebral andbasilar arteries (3D TOF, 64 slices of0 8-1 mm) was also performed in allpatients.20 Basilar artery, intracranial vertebralarteries, the proximal portion of cerebellararteries, and the posterior cerebral arterieswere studied. The extracranial segment of ver-tebral arteries was studied from the changes inthe opacification of the corresponding vesselon MRA with Doppler ultrasound confirma-tion.

Posterior circulation infarcts, otherparenchymal changes, and relevant arteriallesions were recorded and classified into previ-ously defined categories.'720 The nine follow-ing topographic categories were considered:

Department ofNeurology, CentreHospitalierUniversitaire Vaudois,1011 Lausanne,SwitzerlandA BemasconiJ BogousslavskyC BassettiF RegliCorrespondence to:Dr J Bogousslavsky.Received 18 May 1995and in revised form29 August 1995Accepted 19 October 1995

289

group.bmj.com on February 7, 2018 - Published by http://jnnp.bmj.com/Downloaded from

http://jnnp.bmj.com/

http://group.bmj.com

Bernascomi, Bogousslavsky, Bassetti, Regli

medulla, pons, cerebellum (posterior inferior/anterior inferior/superior cerebellar artery ter-ritories), midbrain, thalamus (posterior cere-bral artery deep territory), occipital (temporal)region (posterior cerebral artery superficial ter-ritory), and combined involvement. Weassessed the topography of cerebellar infarctson MRI following previously published tem-plates which are currently used in our depart-ment.22 Other parenchymal lesions were alsorecorded, including anterior circulationinfarcts, focal areas of hyperintense signal, anddiffuse leukoaraiosis. Arterial lesions wereclassified into seven categories: dolichoectaticvertebral/basilar arteries, vertebral arterystenosis or occlusion, basilar artery stenosis orocclusion, and occlusion of major branches ofthe basilar artery (including posterior cerebralartery). Acute infarcts were defined asischaemic lesions greater than an enlargedVirchow-Robin perivascular space (whichdemonstrates a low signal intensity on TIweighted images and a high signal on T2weighted images) and showing gadoliniumenhancement. An uninterrupted lesion visibleon two or more neighbouring slices on MRIwas considered to be an infarct in a single arte-rial territory, even if it involved more than oneof our nine topographic categories. Thusinfarcts involving simultaneously two differentcontiguous anatomical structures (for exam-ple, brainstem plus cerebellum or thalamusplus mesencephalon) but corresponding to asingle vascular territory, and infarction over-lapping adjacent territories (for example, pos-terior inferior cerebellar artery and superiorcerebellar artery) were considered a singlelesion. Bilateral infarct in the thalamoperforateterritory (paramedian, thalamic-subthalamicartery) from the P1 segment of the posteriorcerebral artery was also classified as a singlelesion,26 as was infarction in the posterior cere-bral artery territory involving the superficial(occipital-temporal) and deep (mesencephalo-thalamic) region.Thus multiple acute infarctsin the posterior circulation corresponded toinvolvement of two or more of the three mainsequential segments of the posterior circula-tion defined in the New England MedicalCenter classification17 27: proximal (medullaand posterior inferior cerebellar artery cerebel-lum fed by the intracranial vertebral arteriesand their branches); middle (pons, lower mid-brain, and anterior inferior cerebellar arterycerebellum fed by basilar artery and branches);and distal (upper midbrain, thalamus, superiorcerebellar artery cerebellum and temporal andoccipital lobes fed by the distal basilar arteryand superior cerebellar artery, penetrating andposterior cerebral artery branches). Infarctswith a largest diameter < 15 mm were classi-fied as small.

Neurological and neuropsychological fea-tures were assessed by at least two of theauthors including a senior neurologist GB)within five days of stroke. We analysed clinicalfeatures and information about risk factorsdefined following published guidelines25including hypertension (blood pressure higherthan 160/90 mm Hg at least twice before

stroke), diabetes mellitus (fasting blood glu-cose concentrations above 6-0 mmol/l knownto exist before stroke), regular cigarette smok-ing, hypercholesterolaemia (fasting blood cho-lesterol above 6-5 mmol/l), venous packed cellvolume, heart disease (including anginapectoris, old myocardial infarct, chronic non-valvar atrial fibrillation), and previous tran-sient ischaemic attack(s). Apart from MRI andMRA, systematic investigations included brainCT on admission, three lead ECG monitoringfor at least 24 hours after admission, 12 leadECG, standard blood tests, Doppler ultra-sonography with frequency spectral analysis,and B-mode echotomography of the origin ofthe carotid arteries and vertebral arteries.Additional catheter cerebral angiography usingthe Seldinger method was performed in twopatients, and transthoracic/transoesophagealechocardiography was performed in 20selected patients with presumed cardio-embolism.We considered the following potential

causes of stroke: (a) in situ atherosclerosis orartery to artery embolism (large artery disease)in patients with risk factors who had a stenosisof at least 70% of the lumen diameter in theappropriate large artery as shown by threedimensional images on MRA or conventionalangiography using the NASCET method;(b) small artery disease20 was presumed inpatients with longstanding hypertension ordiabetes mellitus (in the absence of potentialarterial or cardiac sources of emboli) and asmall (S 15 mm) infarct limited to the terri-tory of deep perforators; (c) potential cardiacsources of embolism (mainly non-valvar atrialfibrillation, left ventricular akinetic segment,and other less common sources) were consid-ered according to previously defined crite-ria4 25; (d) other and undetermined. Theevolution within the first four weeks afterstroke was recorded in all patients.

ResultsTwenty seven (1 1%) of 236 patients with ver-tebrobasilar ischaemic stroke had multipleacute infarcts in the posterior circulation.There were nine women (33%) and 18 men(67%). The median age was 66 (range 32 to87) years.

GENERAL AND NEUROLOGICAL VASCULARFEATURESNine patients (26%) had a history ofischaemic heart disease and one patient (No26) had non-valvar atrial fibrillation (table 1).Eleven patients (41%) had hypertension, sixpatients (22%) smoked cigarettes regularly,four patients (15%) had hypercholestero-laemia, four patients (15%) had diabetes, 12patients (44%) had multiple risk factors, andsix patients (22%) had none of these risk fac-tors. Two patients (8%) had a patent foramenovale. Overall, age, sex, and risk profiles weresimilar to those of the patients from theLausanne Stroke Registry in general.25 Threepatients (1 1 %) had had transient ischaemicattack(s) (five months, one week, four years

290





before stroke) in the posterior circulation.28Sixteen patients (59%) had a sudden non-progressive onset of symptoms, whereasstroke was progressive in 11 (44%) patients(over 2-24 hours in five patients, over morethan 24 hours to 15 days in six patients).

TOPOGRAPHY OF INFARCTS (TABLES 1, 2, 3,AND 4)Our patients had from two to seven infarcts(median 3 6) including cerebellum (25),brainstem (17), occipital lobe (13), or thala-mus (11). Cerebellar infarcts were bilateral in80% of the patients (20/25), brainsteminfarcts occurred in 24% (4/17) and occipi-tothalamic infarcts were found in 8% ofpatients (2/24). In the 25 patients with cere-bellar infarction the following vascular territo-ries were involved: posterior inferior cerebellarartery in 22 patients (Nos 1-12, 14, 15,18-20, 22-25, 27), bilaterally in 10; superiorcerebellar artery in 15 patients (Nos 1, 3, 5 ,6,10-16, 18, 23, 24, 26), bilaterally in five; ante-rior inferior cerebellar artery in five patients(Nos 4, 5,10, 12, 24). In patients with brain-stem infarction, the territory of the parame-dian and circumferential arteries was involvedin 13 patients (Nos 1, 4, 5, 7, 9, 14, 17-19,21, 22, 23, 25), and the territory of the shortcircumferential arteries from the P2 segmentof the posterior cerebral artery was involved infour patients (Nos 10, 11, 17, 24). The occipi-tal lobe was affected in 13 patients (Nos 1, 2, 3,6, 8, 12-14, 18, 21, 23, 26, 27), bilaterally inonly one (No 6). The thalamus was involvedin 11 patients (Nos 1-3, 10, 11, 12, 15-18,20), bilaterally in two (Nos 16,17) (in sixpatients the infarct was localised to the terri-tory of the thalamogeniculate arteries26 (infero-lateral pedicle29), in five patients to theparamedian territory29 (thalamic-subthalamicarteries26)). No patient had a medullaryinfarct.We identified three clinicotopographic pat-

terns of multiple acute posterior circulationinfarcts (tables 1, 2, and 3 ):

(1) Multiple infratentorial and supratentorialinfarcts of cerebellum and posterior cerebral artery(or proximal andlor middle + distal multipleacute infarcts in the posterior circulation (Nos17,27)): 18 (67%)

Seven patients (39%) also had a brainsteminfarct (table 1). Most of the patients(78% = 14/18) clinically had a rostral basilarartery syndrome730 (including hemianopia inseven patients) and cerebellar signs; four otherpatients had a visual field defect and cerebellarsigns (table 4). A history of ischaemic heartdisease (39%) and hypertension (33%) wascommon. Contrast angiography was normal inone patient (No 15) without risk factors. Thepresumed cause of stroke was large artery dis-ease in nine patients: focal (two) atheromato-sis of the basilar artery; multifocalatheromatosis (one), and focal stenosis (one)or occlusion (two) of one intracranial vertebralartery. One patient (No 8) had a tandemintracranial atheromatous stenosis with diffusenarrowing of the basilar artery associated with astenosis of the posterior cerebral artery. One

patient (No 11) had an occlusion of the proxi-mal right vertebral artery and a distal focalstenosis of the left vertebral artery; one patient(No 23) had coexisting focal stenosis of thebasilar artery and left vertebral artery.Cardioembolism was the likely aetiology infive patients with a history of myocardialinfarct with akinetic left ventricular segment,and non-valvar atrial fibrillation. A 48 year oldman (No 12) died two days after stroke: hehad developed headache, vomiting, left sen-sory-motor hemiparesis, and ataxia, with rapidcoma. Brain MRI showed multiple, large andbilateral infarcts in all three cerebellar andboth posterior cerebral artery territories. BrainMRA suggested an occlusion in the distal por-tion of the basilar artery and left posterior infe-rior cerebellar artery, with otherwise intactarteries. Although transthoracic echocardiog-raphy was normal, a cardioembolic mecha-nism was suspected. Aetiology remainedunknown in three patients, and uncertain intwo patients with only a patent foramen ovale(Nos 24, 27). One month after stroke 13patients (72%) were independent, and four(22%) were still dependent.

(2) Multiple infratentorial infarcts of cerebel-lum and lower brainstem (or proximal + middlemultiple acute infarcts in the posterior circulation(Nos 17,27)):seven (26%)These patients showed brainstem signs

sometimes combined with cerebellar ataxia(table 4). Presumed aetiology was hyperten-sive small artery disease in four patients(57%). A 51 year old patient (No 4) had IgManticardiolipin antibodies without any otherpotential cause of stroke. Another patient (No5) had a proximal occlusion of the right verte-bral artery related to a subintimal dissectiondetected on conventional angiography. Thelikely cause of stroke was cardioembolism in a55 year old woman with coronary heart dis-ease and an akinetic left ventricular segment.All but one patient was independent onemonth after stroke (table 2).

(3) Multiple infarcts of brainstem and posteriorcerebral artery (or proximal + middle multipleacute infarcts in the posterior circulation (Nos17,27)): two (7%).

Only two patients had multiple supratentor-ial and infratentorial infarcts including ponsand both posterior cerebral artery territories,but sparing the cerebellum (table 3). Thesepatients had brainstem signs (table 4); MRAshowed basilar artery or vertebral arteryatheromatosis in both patients. In patient No17 there was an extension of a thrombus fromthe intracranial vertebral artery to the proxi-mal basilar artery.

CONCOMITANT VASCULAR DISEASESNo patient had had a recent ischaemic eventin the anterior circulation, and none had anold infarct in the posterior circulation. Acarotid stenosis was found in one patient, whohad had an episode of amaurosis fugax sixyears before the present stroke (No 10). Anasymptomatic carotid stenosis was detected infour patients, including an intracranial stenosisin one patient (No 8). Nine patients (Nos 3-9,

291




Bernasconi, Bogousslavsky, Bassetti, Regli

Table 1 Infratentorial and supratentorial multiple acute infarcts in the posterior circulation (cerebellum, posterior cerebralartery and brainstem; proximal andlor middle + distal (Nos 17, 27))

Topography Large artery Cardiac Stroke Outcomelesion lesion mechanism

6

8

10

13

14

15

16

18

20

23

24

-ft-1IIr

I.

7,'

7i-

None

None

None

None

None

None Akinesia +

aneurysm

None

LAD

LAD

LAD

LAD

LAD

Independent

Independent

Independent

Independent

Dependent

Cardioembolism Dependent

LAD

None Cardioembolism ?

None None

None

None

None

0.

7l--~ !

None

None

Parietalthrombus

None

None

None

Patent foramenovale

LAD

Independent

Dead

? Independent

Independent

? Independent

Cardioembolism Dependent

? Independent

LAD

LAD

Paradoxicalembolism ?

Independent

Dependent

Independent

None

/s;'7 -

26

27

LAD = large artery disease.

Atrialfibrillation

Patent foramenovale

Cardioembolism Independent

Paradoxicalembolism ?

Independent

292





Table 2 Cerebellum and lower brainstem multiple acute infarcts in the posterior circulation (proximal + middle(Nos 17, 27))


None None Anticardiolipin Independentantibodies

4

0,Q None None LAD Dependent5 ~~~~~~~~~~~~~~~~~~~(dissection)

None None SAD Independent


9


19

f4Q None None SAD Independent

22

None Akinesia Cardioembolism Independent

25

SAD = small artery disease; LAD = large artery disease.

11, 20-22) had leukoaraiosis. Two patients (P < 0-01). Of 20 patients (Nos 1, 2, 4, 5, 6,(Nos 10, 16) with multiple infratentorial and 8, 10-18, 20, 23, 24, 26, 27; 74%) with largesupratentorial infarcts, possibly of cardioem- infarcts (>3 cm, >2 slices on MRI), six (30%)bolic origin, also had a dolichoectatic basilar had a poor functional outcome and one died.artery. We found no patient with arteritis or On the other hand, 12 of 18 patients (67%)another angiopathy. One patient (No 8) with with more than two infarcts were independentperipheral arteriopathy and one patient (No one month after stroke.11) with aortocoronary bypass were underanticoagulant therapy at the onset of stroke.No patient was under antiaggregant therapy at Discussionthe time of the index stroke. Acute multiple infarcts in the posterior circu-

lation were not uncommon, as they repre-EVOLUTION sented over 1 1% of first ever acute ischaemicHaemorrhagic transformation of infarct was stroke in this territory, but we believe that theyonly seen in one non-anticoagulated patient have been commonly overlooked, especiallywho had three infarcts (No 10). One patient without systematic MRI. This is by contrast(No 5) with bilateral large cerebellar infarcts with the absence of a specific study devoted toand a pontine infarct developed a compression that important issue. Barkhof and Valk31 stud-of the fourth ventricle with hydrocephalus. He ied clinicoradiological correlations in fourimproved after external ventricular drainage. patients with multiple lesions in the verte-One month after stroke, 20 patients (74%) brobasilar system detected on MRI and "topwere independent, six were dependent, and of the basilar" syndrome,3 but angiographyone had died. This contrasted with only 19 was not performed and the cause of infarct(9%) of the 209 patients with one single was not considered. Among 14 patients withinfarct being dependent at one month an infarct in the superior cerebellar artery terri-

Table 3 Brainstem and posterior cerebral artery territory multiple acute infarcts in the posterior circulation(proximal + middle (Nos 17, 27))


None LAD Dependent

'<@9@R None LAD Independent

21

LAD =large artery disease.

293





Table 4 Topography, clinicalfeatures, and outcome of 27 patients with multiple acute infarcts in the postenror circulation

Vascular territory No TIA Onset Symptoms

Infratentorial anid supratentorial MAPCI (cerebellumi, PCA and brainsteml)1 PICA, 1 + m; SCA, I + m; p; PCA: Th (il) 7, b s HH, CNP (VII), limb and trunk ataxia, nystagmus2 PICA, 1 + m; SCA, 1; Th (il); PCA 4, b +, s s HH, nystagmus, dysarthria, M + S (f, u, 1), limb and trunk

ataxia, asterixis3 PICA, m; SCA, m; PCA 4, b p HH, vertical gaze palsy, dysarthria, limb ataxia6 PICA, I + m; SCA, + m; PCA 6, b p HH, nystagmus, dysarthria, limb ataxia, NPsy

(asimultagnosia, visual hallucinations, achromatopsia)8 PICA, m; PCA 2, b s HH, vertical gaze palsy, limb and trunk ataxia

10 PICA, 1; AICA; SCA, 1; m (sc); Th (il) 3, u +, o s Somnolence, vertical gaze palsy, CNP (III), dysphagia,dysarthria, M (f, u, 1), limb ataxia

11 PICA, 1 + m; SCA, I + m; m (sc); Th (il) 6, b p CPN (III), dysarthria, limb and trunk ataxia, NPsy(amnesia, apathy, confusion)

12 PICA, 1; AICA; SCA, I + m; PCA; Th (il) 3, b p Somnolence/coma, nystagmus, M + S (u, 1), limb and trunkataxia

13 SCA, 1 + m; PCA 2, b s HH, limb and trunk ataxia14 PICA, 1; SCA, 1; p; PCA 4, b s HH, dysarthria, limb and trunk ataxia15 PICA, 1 + m; SCA, m; Th (pm) 4, b p Stupor, M (f, u, 1), NPsy (apathy, subcortical aphasia),

asterixis, dysarthria16 SCA, m; Th (pm) 2, b s Stupor, vertical gaze palsy, CNP (III, IV), limb and trunk

ataxia, M (u), NPsy (dementia)18 PICA, 1; SCA, m + 1; p; PCA; Th (pm) 7, b p HH, Homer's syndrome, M (u, 1), dysarthria, NPsy

(confusion, apathy, amnesia)20 PICA, + m; Th (pm) 3, b s Somnolence vertical gaze palsy, M (f, u,l), NPsy

(confusion), dysarthria23 PICA, 1 + m; SCA, 1; p; PCA 4, b s HH, "one and a half' palsy, CNP (V, VII), dysphagia,

dysarthria, M + S (f, u, 1), limb ataxia24 PICA, l + m; SCA; AICA; m (sc) 3, b s Skew deviation, M (u), trunk ataxia, dysarthria, NPsy

(apathy, emotional lability, apraxia)26 SCA, m: PCA 2, b p HH, dyssarthria, limb and trunk ataxia27 PICA, l; PCA; Th (il) 3, u s HH, trunk ataxia, NPsy (alexia, achromatopsia, amnesia), S

(f, u)

4 PICA, 1; AICA; p

5 PICA, l + m; AICA; p

7 PICA, l + m; p

9 PICA, m; p19 PICA, l; p22 PICA, m; p25 PICA, 1; p

17 p; m (sc); Th (tt, pm)

21 p; PCA

Cerebellum and lower brainstem MAAPCI2, u p Saccadic horizontal pursuit, nystagmus, CPN (VIII,

cochlear), M (f, u, 1), limb and trunk ataxia3, b +, s p Somnolence/coma, nystagmus, CNP (V, VI, VII),

dysarthria, limb and trunk ataxia, S (f, u)3, b s Saccadic horizontal pursuit, CNP (VII), limb and trunk

ataxia2, b s Dysarthria, M (ataxic hemiparesis)2, b p Dysarthria, M (f, u, l)2, u s Saccadic horizontal pursuit, dysarthria, M (f, u, 1)2, u s CNP (V, VI), trunk ataxia

Brainstem and PCA ternitory MAPCI5, b +, s p Vertical gaze palsy, Horner's syndrome, pseudo-VI, M

(ataxic hemiparesis), NPsy (visual and auditivehallucinations), nystagmus

3, b s Dysarthria, M (ataxic hemiparesis)

TIA = transient ischaemic attack; No = number of infarcts; u/b = unilateral, bilateral; so = same/other vascular territory;NPsy = neurological disturbance; M/S (f, u, l) = motor weakness (M)/sensory disturbance (face (f), upper (u), lower (1) limb);HH = homonymous hemianopia; CNP = cranial nerve palsy (III-XII).Vascular territories-Cerebellum: PICA, m/l = posterior inferior cerebellar artery, media/lateral branch; AICA = anterior inferiorcerebellar artery; SCA, m/l = superior cerebellar artery, media/lateral branch. Thalamus: th, tt = thalamotuberal artery (polarartery); th, pm = paramedian artery (thalamic-subthalamic artery), th, il = inferolateral artery (thalamogeniculate artery). Occipitallobe: PCA = cortical branches of posterior cerebral artery. Mesencephalon: m, sc = short circumferential arteries from PCA (P2segment). Pons: p = paramedian, short and long circumferential arteries.

tory on MRI or CT, Struck et all'3 reportedthat 11 had associated infarcts involving othercerebellar. brainstem, and supratentorial terri-tories, but the number of patients without an

MRI was not specified. Caplan et al'7 reported10 patients with occlusive disease of the proxi-mal vertebral artery and embolism to the pos-terior circulation: in five subjects MRI showedsupratentorial and infratentorial infarctsinvolving distinct territories. Tettenborn et al12

reported that of 22 patients with postoperativebrainstem and cerebellar infarcts, 10 had mul-tiple infarcts in the posterior circulation.However, these studies did not discuss thetopographic, clinical, and prognostic featuresassociated with multiple infarcts. Morerecently, two reports emphasised that multifo-cal brainstem33 or cerebellar27 infarction hasbeen largely overlooked, but without a system-atic study of acute multiple posterior circula-tion infarcts.Most of our patients (93%) had a cerebellar

component of acute multiple posterior circula-tion infarcts, often bilateral, which involvedthe posterior inferior cerebellar artery territorymore often than in reported series of "pure"

cerebellar infarcts.922 34-36 A supratentorialcomponent of multiple acute posterior circula-tion infarcts was present in 90% of thepatients, and brainstem infarction in twothirds.The main clinical presentation in our 20

patients with acute multiple supratentorial andinfratentorial infarcts (groups 1 and 3) was apartial and reversible rostral basilar artery syn-drome3 7 311 in all but one patient, with cerebellarsigns and a visual field defect in about half ofthe patients. No patient had clinical evidencefor involvement of only one single vascular ter-ritory and none of those with a presumedsmall artery disease (group 2) had a lacunarsyndrome.37

Stroke was seldom preceded by transientischaemic attack compared with other poste-rior circulation infarcts (18%,38 23%,39 57%4').Stroke onset was non-progressive in about twothirds of the patients, without significant dif-ference between the aetiological subgroups(artery to artery embolism, cardioembolism,local thrombosis, small artery disease). Thissuggests that infarcts mostly came on simulta-neously.

294

I

I

I

III

I

I

II

II

I

IIII





No particular risk factor correlated with apoor outcome. There was no difference in theoutcome between patients with multipleinfratentorial and supratentorial infarctsincluding the posterior cerebral artery,patients with multiple infratentorial infarcts ofthe cerebellum and brainstem, and the twopatients with multiple infarcts of the brainstem and posterior cerebral artery territory.Nevertheless, multiple acute posterior circula-tion infarcts were clearly associated with aworse prognosis than single infarcts in the pos-terior circulation, as 26% v 9% were dead ordependent one month after stroke. However,regardless of distribution of the infarcts, 75%of our patients with multiple acute infarcts inthe posterior circulation were functionallyindependent one month after stroke.Moreover, there was the same proportion ofpatients with a good outcome in the groupswith either more or less than five infarcts. Onthe other hand, the patients with large infarctshad a poorer evolution, suggesting that sizerather than a large number of infarcts corre-lates with a poor outcome.

Besides atheroma and embolism, multipleinfarction may be related to angiopathies(isolated angiitis, amyloid angiopathy, infec-tious arteritis, polyarteritis nodosa, Wegener'sgranulomatosis, giant cell arteritis, Behcet'sdisease, hypersensitivity vasculitis), coagu-lopathies, and familial conditions (CADASIL,MELAS),4142 but none of our patients hadthese conditions. Atheromatosis of the largearteries was the most frequent presumed causeof infarct in our series (4%, table 1), indepen-dently of the topography of the lesions. This isnot surprising as there was a preponderance ofelderly men. Several studies showed thatocclusion of rostral basilar territory branches(posterior cerebral artery, distal basilar artery,and superior cerebellar artery)43 44 andintracranial vertebral artery/posterior inferiorcerebellar artery is usually embolic, 745 intrin-sic atherosclerotic disease being more uncom-mon.46 As previously underlined in singleinfarction in the posterior circulation,57174447the proximal segment of the vertebral arterywas also suspected to be the embolic source inmost of our patients with supratentorial andinfratentorial infarcts, suggesting thatextracranial vertebral artery disease may playan important part in posterior circulationstroke. On the other hand, no "preferential"localisation40 of atheromatosis along the basilarartery was found in our patients, and none ofthem showed a midbasilar occlusion, whichmay sometimes be associated with a benignclinical course.48 49 Atheromatosis may alsoocclude the origin of vertebral artery and basi-lar artery branches, particularly involving thesmaller penetrating vessels rather than largerbranches (cerebellar or anterior spinal arter-ies).46 Small artery disease was uncommon(15%) in our patients, although it was rela-tively more frequent in the patients with multi-ple, small infarcts (60%).The frequency, location, and severity of

carotid atheroma in patients with verte-brobasilar occlusive disease is variable (40%,'

54%50). We did not find a correlation betweenposterior circulation and carotid arteryatheroma, suggesting that it may develop inde-pendently in each of these locations.

Nearly 20% of our patients had a potentialcardiac source of embolism.'4 This seems sim-ilar to what has been reported for ischaemicstroke in general,'4 5152 including unselectedpatients with posterior circulation infarct.20Indeed cerebellar and posterior cerebral arteryterritory infarcts are commonly due to cardio-genic embolism.222547 In 15% of our patients,the cause of stroke remained unknown.However, transoesophageal echocardiographywas performed in only a few patients and ath-erosclerotic plaques in the aortic arch53 54 werenot ruled out.

Overall, our findings suggest that embolism,from an arterial or a cardiac source, is themain aetiology of multiple acute infarcts in theposterior circulation. Whether embolismcauses multiple territory infarction by proxi-mal occlusion of major vessels with distalhaemodynamic infarction, or it is the result ofmultiple emboli or a single embolus breakingup and dispersing into different sites is notclear from our study.Our findings emphasise the heterogeneity of

clinical, topographic, and aetiological aspectsof multiple acute posterior circulation infarcts.This suggests that early recognition of thesetypes of infarct may have implications foremergency therapy options in the acute phaseof stroke, including acute stroke trials assess-ing fibrinolytic or anti-ischaemic drugs.

1 Castaigne P, Lhermitte F, Gautier J, Escourolle R,Derouesne C, der Agopian P, et al. Arterial occlusion inthe vertebro-basilar system. Brain 1973;96: 133-54.

2 Caplan L. Bilateral distal vertebral artery occlusion.Neurology 1983;33:552-8.

3 Caplan L. "Top of the basilar" syndrome. Neurology1980;30:72-9.

4 Graff-Radford N, Damasio H, Yamada T, Elsinger PJ,Damasio AR. Nonhaemorrhagic thalamic infarction:clinical, neuropsychological and electrophysiologicalfindings in four anatomical groups defined by computer-ized tomography. Brain 1985;108:485-516.

5 Koroshetz W, Ropper A. Artery-to-artery embolism caus-ing stroke in the posterior circulation. Neurology 1987;37:292-6.

6 Amarenco P, Hauw J, Henin D, Duyckaerts C, Roulet E,Laplane D, et al. Les infarctus du territoire de l'arterecerebelleuse postero-inferieure. Etude clinico-pathologique de 28 cas. Rev Neurol (Paris) 1989;145:277-86.

7 Mehler M. The rostral basilar artery syndrome. Diagnosis,etiology, prognosis. Neurology 1989;39:9-16.

8 Levine S, Quint D, Pessin M, Boulos RS, Welch KMA.Intraluminal clot in the vertebrobasilar circulation: clinicaland radiologic features. Neurology 1989;39:515-22.

9 Amarenco P, Hauw J. Cerebellar infarction in the territory ofanterior and inferior cerebellar artery. A clinicopathologi-cal study of 20 cases. Brain 1990;113:139-55.

10 Amarenco P, Hauw J. Cerebellar infarction in the territory ofthe superior cerebellar artery: a clinicopathologic study of33 cases. Neurology 1990;40:1383-90.

11 Chambers B, Brooder R, Donnan A. Proximal posteriorcerebral artery occlusion simulating middle cerebralartery occlusion. Neurology 1991;41:385-90.

12 Amarenco P. The spectrum of cerebellar infarctions.Neurology 1991;41:73-979.

13 Struck L, Biller J, Bruno A, Neiman RF, Loftus CM, YuhWTC, et al. Superior cerebellar artery territory infarction.Cerebrovasc Dis 199 1;1:71-5.

14 Bogousslavsky J, Cachin C, Regli F, Despland PA, VanMelle G, Kappenberger L. Cardiac sources of embolismand cerebral infarction: clinical consequences and vascularconcomitants: the Lausanne Stroke Registry. Neurology1991;41:855-9.

15 Caplan L, Tettenborn B. Vertebrobasilar occlusive disease:review of selected aspects. 1. Spontaneous dissection ofextracranial and intracranial posterior circulation arteries.Cerebrovasc Dis 1992;2:256-65.

16 Fisher M, Sotak C, Minematsu K, Li L. New magnetic res-onance techniques for evaluating cerebrovascular disease.Ann Neurol 1992;32:1 15-22.

295





17 Caplan L, Amarenco P, Rosengart A, Lafranchise EF,Teal PA, Belkin M, et al. Embolism from vertebralartery origin occlusive disease. Neurology 1992;42:1505-12.

18 Bogousslavsky J, Fox A, Bamett H, Hachinski VC, VinitskiS, Carey LS. Clinico-topographic correlation of smallvertebrobasilar infarct using magnetic resonance imag-ing. Stroke 1986;17:929-38.

19 Fisher CM. The posterior cerebral artery syndrome. CanNeurol Sci 1986;13:232-9.

20 Bogousslavsky J, Regli F, Maeder P, et al. The etiology ofposterior circulation infarcts: a prospective study usingmagnetic resonance angiography. Neurology 1993;43:1528-33.

21 Daniels DL, Haughton VM, Naidich TP. Cranial and spinalmagnetic resonance imaging. An atlas and guide. New York:Raven Press, 1987.

22 Barth A, Bogousslavsky J, Regli F. The clinical and topo-graphic spectrum of cerebellar infarcts: a clinical-mag-netic resonance imaging correlation study. Ann Neurol1993;33:451-6.

23 Tatemichi TK, Steinke W, Duncan C, Bello JA, Odel JG,Behreus MM, et al. Paramedian thalamopeduncularinfarction: clinical syndromes and magnetic resonanceimaging. Ann Neurol 1992;32:162-71.

24 Amarenco P, Kase CS, Rosengart A, Pessin S, BousserMG, Caplan LR. Very small (border zone) cerebellarinfarcts: distribution, causes, mechanisms and clinicalfeatures. Brain 1993;116:161-86.

25 Bogousslavsky J, Van Melle G, Regli F. The LausanneStroke Registry: analysis of 1,000 consecutive patientswith first stroke. Stroke 1988;19:1083-92.

26 Bogousslavsky J, Caplan LR. Vertebrobasilar occlusive dis-ease: review of selected aspects. 3. Thalamic Infarcts.Cerebrovasc Dis 1993;3:193-205.

27 Chaves CJ, Caplan LR, Chung CS, Tapias J, Amarenco P,Teal P, et al. Cerebellar infarcts in the New EnglandMedical Center Posterior Circulation Registry. Neurology1994;44: 1385-90.

28 Trouillas P, Nighoghossian N, Philippon B. Nuclear hemo-dynamic vertebrobasilar insufficiency. A new approachwith the xenon Xe133 method. Arch Neurol 1991;48:921-9.

29 Bogousslavsky J, Regli F, Uske A. Thalamic infarcts: clinicalsyndromes, etiology, and prognosis. Neurology 1988;38:837-48.

30 Mehler MF. A novel basilar artery syndrome with acuteconfusion and benign outcome. Neurology 1984;34:203.

31 Barkhof F, Valk J. "Top of the basilar" syndrome: a com-parison of clinical and MR findings. Neuroradiology,1988;30:293-8.

32 Tettenbom B, Caplan LR, Sloan MA, Estol CJ, PessinMS, De Witt LO, et al. Postoperative brainstem andcerebellar infarcts. Neurology, 1993;43:471-7.

33 Tettenbom B. Multifocal ischemic brains-stem lesions. In:brainstem localization and function. Caplan LR, Hopf HC,eds. Berlin: Springer-Verlag, 1994:23-31.

34 Tohgi H, Takahashi S, Chiba K, Hirata Y for the TohokuCerebellar Infarction Study Group. Cerebellar infarction:clinical and neuroimaging analysis in 239 patients. Stroke1993;24; 1697-701.

35 Milandre L, Brosset C, Gouirand R, Khalil R. Les infarctus

cerebelleux purs: trente observations. Presse Med 1992;21:1562-5.

36 Macdonell RAL, Kalnis RM, Donnan GA. Cerebellarinfarction: natural history, prognosis and pathology.Stroke 1989;18:849-55.

37 Orgogozo JM, Bogousslavsky J. Lacunar syndromes. In:Vinken PJ, Bruyn GW, Klawans HL, eds. Handbook ofclinical neurology. Vol 54. Amsterdam: Elsevier, 1989:235-69.

38 Nadeau S, Jordan J, Mishra S. Clinical presentation as aguide to early prognosis in vertebrobasilar stroke. Stroke1992;23: 165-70.

39 Homig CR, Buettner T, Hoffmann 0, Domdorf W. Short-term prognosis of vertebrobasilar ischemic stroke.Cerebrovasc Dis 1992;2:273-81.

40 Pessin MS, Gorelick PB, Kwan ES, Caplan LR. Basilarartery stenosis: middle and distal segments. Neurology1987;37: 1742-6.

41 Tournier-Lasserve E, Iba-Zizen MT, Romero MT,Bousser MG. Autosomal dominant syndrome withstroke-like episodes and leukoencephalopathy. Stroke199 1;22: 1297-302.

42 Hankey GJ. Isolated angiitis/angiopathy of the central ner-vous system. Cerebrovasc Dis 1991- 1:2-15.

43 Amarenco P, Roullet E, Goujon C, Cheron F, Haw J-J,Bousser MG. Infarction in the anterior rostral cerebellum(territory of the lateral branch of the superior cerebellarartery). Neurology 1991;41:253-8.

44 Caplan L, Tettenborn B. Vertebrobasilar occlusive disease:review of selected aspects. 2. Posterior circulationembolism. Cerebrovasc Dis 1992;2:320-6.

45 Pessin MS, Daneault N, Kwan ES, Eisengart MA, CaplanLR. Local embolism from vertebral artery occlusion.Stroke 1988;19:112-5.

46 Caplan LR Vertebrobasilar occlusive disease. In: BarnettHJ, Mohr JP, Stein B,Yatsu F, eds. Stroke: pathophysiol-ogy, diagnosis, and management. New York: ChurchillLivingstone, 1986:549-619.

47 Amarenco P, Hauw JJ, Gautier JC. Arterial pathology incerebellar infarction. Stroke 1990;21:1299-305.

48 Caplan LR. Occlusion of the vertebral or basilar artery.Follow-up analysis of some patients with benign out-come. Stroke 1979,3:277-82.

49 Labauge R, Pages M, Blard JM. Survie prolongee apresocclusion du tronc basilaire: 4 cas. Rev Neurol 1989;145:789-94.

50 Ueda K, Toole JF, McHenry LC. Carotid and verte-brobasilar transient ischemic attacks: clinical and angio-graphic correlation. Neurology 1979;29: 1094-101.

51 Kittner SJ, Sharkness CM, Price TR, Plotnick GD,Dambrosia JM, Wolf PA, et al. Infarcts with acardiac source of embolism in the NINCIDS StrokeData Bank: historical features. Neurology 1990;40:281-84.

52 Marshall RS, Mohr JP. Current management of ischaemicstroke. Jf Neurol Neurosurg Psychiatry 1993;56:6-16.

53 Amarenco P, Duyckaerts C, Tzourio C, Henin D, BousserMG, Haw J-J. The prevalence of ulcerated plaques in theaortic arch in patients with stroke. N Engl Jf Med1992;326:221-5.

54 Caplan LR, Amarenco P, Rosengart A, Lafranchise EF,Teal PA, Belkin M, et al. Embolism from vertebral arteryorigin occlusive disease. Neurology 1992;42: 1505-12.

296




circulation.Multiple acute infarcts in the posterior

A Bernasconi, J Bogousslavsky, C Bassetti and F Regli

doi: 10.1136/jnnp.60.3.2891996 60: 289-296 J Neurol Neurosurg Psychiatry

http://jnnp.bmj.com/content/60/3/289Updated information and services can be found at:

These include:

serviceEmail alerting

box at the top right corner of the online article. Receive free email alerts when new articles cite this article. Sign up in the

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://jnnp.bmj.com/content/60/3/289

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/



multiple acute infarcts in the posterior circulation

Documents