Stroke signs and symptoms have traditionally been analyzed

Transcription

1 Supratentorial Age-Related White Matter Changes Predict Outcome in Cerebellar Stroke Eva Grips, MD; Oliver Sedlaczek, MD; Hansjörg Bäzner, MD, PhD; Michael Fritzinger, MD; Michael Daffertshofer, MD, PhD; Michael Hennerici, MD, PhD Background and Purpose Little is known about the relevance of age related white matter lesions (WMLs) concerning outcome after first-ever territorial stroke. Based on an index patient, we hypothesized that age and pre-existent WMLs rather than infarct volume and topography determine outcome. Patients and Methods Thirty-four consecutive patients with magnetic resonance diffusion-weighted imaging proven isolated acute cerebellar infarction were prospectively entered on our stroke data registry. Patients with pre-existent neurological deficits, hemorrhagic, or malignant cerebellar infarction were excluded. Patients were stratified using Rankin and Barthel disability scales into groups: I complete recovery, II moderate, and III significant disability 14 days after stroke onset. Results Initial neurological and functional scores were similar among all the groups with vertigo, nausea, unsteadiness, and limb ataxia being the most common. Infarct volume, vascular territories, and comorbidity did not predict clinical outcome. In contrast, presence and severity of supratentorial WMLs and age significantly determined outcome by functional tests. Conclusions In patients with isolated cerebellar infarction functional outcome correlated with the coexistence of age-related WMLs rather than stroke volume and topography. This reflects the loss of compensatory network integrity as the equivalent of functional incapacity beyond local lesion disturbances. (Stroke. 2005;36: ) Key Words: aging outcome stroke white matter Stroke signs and symptoms have traditionally been analyzed with regard to lesion localization, infarct size, and underlying stroke mechanism. However, more recently, based on the steadily increasing quality of acute stroke imaging, the concept of an immediate relationship of clinical signs and expected anatomical lesion sites is often challenged in patients with multiple lesions. In this context, several situations are principally conceivable: (1) The synchronous occurrence of acute lesions in nonadjacent cerebral territories may cause syndromes such as hemianopia hemiplegia in patients with double infarct in 1 hemisphere. 1 Here, symptoms are consistent with anatomically related function, and unusual findings result from simultaneous infarction in distant vascular territories; (2) the synchronous occurrence of acute lesions in anatomically and functionally related regions either adjacent or remote but symmetrically resulting in more severe deficits than might be expected from the simple addition of 2 infarct syndromes 2 ; (3) unexpected symptoms in the context of an acute infarct, a situation reported previously as synergistic infarct. In a published case, 3 a posterior cortical infarct (aggravated in a patient with a previous frontal lesion) elicited frontal features that had not been expected from a simple sum of effect of these lesions. In a series, 4 a patient exhibited clinical deterioration with a subcortical pattern of deficits fitting with his chronic lacunar lesions rather than with the acute cortical infarct; (4) an acute lesion in exact anatomical symmetry to a previous lesion in the contralateral hemisphere that may induce the reappearance of the previous deficit in addition to the newly acquired symptoms. 5 Thus, concepts of remote network mediated lesions were within reach. In light of these observations, the hypothesis was promoted that new lesions may generate symptoms unrelated to the specific anatomic or functional area affected through the breakdown of widespread neuronal networks. Patients with isolated cerebellar stroke could be considered an ideal model to test whether a disruption of the cerebellar gating function with its influence on supratentorial networks predicts functional outcome. Pre-existing supratentorial white matter lesions (WMLs) could then be considered vice versa to account for a poor prognosis in isolated cerebellar infarction. In consistence with this model, anatomical functional networks have been proposed for a long time and (pre) frontal subcortical circuits have been established. Recently, growing evidence from primate studies confirm the existence of vast anatomical functional circuits through transneuronal trans- Received February 21, 2005; final revision received April 17, 2005; accepted June 13, From the Department of Neurology, Universitätsklinikum Mannheim, University of Heidelberg, Mannheim, Germany. Correspondence to Dr Michael Hennerici, Department of Neurology, Universitätsklinikum Mannheim, University of Heidelberg, Theodor-Kutzer-Ufer 1-3, Mannheim, Germany. hennerici@neuro.ma.uni-heidelberg.de 2005 American Heart Association, Inc. Stroke is available at DOI: /01.STR dc 1988

2 Grips et al Network Hypothesis of Stroke 1989 TABLE 1. Patient Demographics Demographics Group I Group II Group III Patients, no Female Age mean (range) (29 67) (23 88) 70 7 (61 77) Risk factors, no. Hypertension Diabetes Hyperlipidemia Smoking Hyperhomocysteinemia Clinical signs, no. Vertigo Nausea Unsteadiness Limb ataxia Dysarthria Gait ataxia Nystagmus Vascular territory, no. PICA SCA AICA Multiple (of the above) Etiology of infarction, no. Cardioembolic Macroangiopathy Dissection Coagulopathy Unknown PICA indicates posterior inferior cerebellar artery; SCA, superior cerebellar artery; AICA, anterior inferior cerebellar artery. port of neurotropic viruses, additionally integrating cerebellar nuclei. 6 Following this model, the persistence of vivid bypasses explains well sudden deterioration in morphologically insignificant network destruction and asymptomatic progress of lesion burden. The detection of subcortical WMLs on MRI in the healthy elderly is common. Its prevalence is estimated from 5% to 20% in several population-based trials. 7 9 Increasing age and vascular risk factors such as hypertension, diabetes, and hyperhomocysteinemia are related to the degree of WMLs. 10,11 The clinical relevance of WMLs is controversial; however, recent data confirm the coexistence of vascular dementia, gait unsteadiness, and urinary incontinence subsumed in the syndrome of subcortical vascular encephalopathy (SVE) with severe WMLs. 12 There is growing evidence that the progression of WMLs is associated with progressive functional deficits such as gait disturbance 13 and upper limb discoordination, 14 and a large ongoing European trial 15 intends to clarify the issue of increasing disability in the elderly on the basis of progressive WMLs. Isolated cerebellar infarcts are well characterized in prospective and retrospective studies with regard to symptoms, territories, and etiology. 16 Outcome has been studied on a long-term basis in patients with stroke of the posterior circulation in general 18,22,23 subgroups of acute malignant or hemorrhagic infarctions of the cerebellum, 24 and has been identified in severely disabled patients. 23 Thus, little is known about acute deficits resulting from loss of cerebellar tissue only. Because acute and chronic lesions in remote infratentorial and supratentorial territories can readily be identified on diffusion-weighted imaging (DWI) and T2- weighted MRI (particularly fluid-attenuated inversion recovery [FLAIR]), prerequisites are available to test our hypothesis in an acute cerebellar stroke for the first time: based on clinical observations, indicating favorable and bad outcome in morphologically similar cerebellar lesions, we hypothesized that age-related WMLs affecting the neuronal network, as represented by supratentorial white matter (WM) structures, might modulate functional outcome and define the individual outcome as well as prognosis rather than the acute infarct for sudden failure of unestablished compensatory synergistic mechanisms. Patients and Methods Thirty-four patients (16 women, ranging from 23 to 88 years of age) with magnetic resonance (MR) DWI proven isolated cerebellar infarction were identified of 275 consecutive patients with posterior circulation stroke diagnosed and documented according to standardized criteria in the Mannheim Stroke Registry within 4 years (1998 to 2002). Exclusion criteria were pre-existing neurological deficits, extracerebellar acute infarctions (particularly brain stem affection), and malignant or hemorrhagic cerebellar lesions. Patients were stratified by Rankin and Barthel disability scales into the following 3 groups: (1) full recovery without persistent deficit 2 days after stroke onset; (2) moderate disability (Rankin 2; Barthel 80) 14 days after stroke onset; and (3) significant disability (Rankin 2; Barthel 80) 14 days after stroke onset. Because outcome might be strongly affected by non-neurological disease, we used the Charlson comorbidity index 25 to rule out significant bias. This validated instrument assigns a weighted score from 0 to 6 to each patient reflecting the number and severity of prespecified comorbid medical conditions (ie, cardiovascular, cerebrovascular, pulmonary, endocrine, renal, and malignant diseases) TABLE 2. Functional Scores Barthel Rankin NIHSS Onset 14 Days Onset 14 Days Onset 14 Days Group I 91 (60 100) (0 3) (0 3) 0 Group II 70 (25 100)* 97 (85 100) 3 (1 5) 1.3 (0 2) 3.2 (1 7) 1.3 (0 3) Group III 40 (10 60) 56 (10 70) 4.7 (4 5) 3.7 (3 5) 5 (2 9) 5.2 (2 15) Mean (range); */ P 0.05/0.01 vs group I (Mann Whitney U test).

3 1990 Stroke September 2005 TABLE 3. Tentative Outcome Predictors No. Mean Age (range) Cerebellar Territories Affected, no. Infarct Volume, ml Pre-Existing Cortical Lesions Charlson Index Group I (29 67) of8 1 Group II (23 88) of Group III * (61 77) of Mean SD; *P 0.01 vs group I; P 0.05 vs group II; P 0.08 vs group I. and is based on the adjusted mortality risk associated with each comorbid diagnosis; it is a strong predictor of 1-year survival after hospitalization. Stroke imaging in this study was performed on a 1.5-T magnetom vision (Siemens) for 24 to 48 hours after symptom onset and included standard T2, DWI, FLAIR, and MR angiography. MRI was read and evaluated by an independent experienced reader blinded to the clinical outcome. Apart from the identification of pre-existent cerebral lesions, small vessel changes of supratentorial subcortical WM were determined and classified by prominent features such as lesion extent, rims, and frontal caps, from none to severe according to the classification of Fazekas. 26 Cerebellar infarcts were categorized according to the vascular territories as published by Tatu. 20 Infarct volume was measured on DWI after template-based normalization with SPM The image was then read by software (MCID; Imaging Research), and a semiautomatic border based on intensity differences outlined the lesion. This was done for each of the adjacent slices that were calculated, resulting in normalized total cerebellar lesion volume. Statistics Results are expressed as the mean SD. Error probabilities were calculated by comparing 2 patient groups using the Mann Whitney U test and were intended for exploratory data analysis because no Bonferroni correction was performed. The 2 test was used for multiple categorical comparisons. Logistic regression was used to perform a prediction of dichotomous variables from interval or categorical data. Results Of the 34 patients included, 8 patients fulfilled the criteria of group I, 20 of group II, and 6 of group III (Table 1). Clinical symptoms leading to presentation were vertigo (91%), nausea (68%), unsteadiness (68%), and limb ataxia (62%; Table 1). There was no correlation between initial symptoms and outcome and no difference between the frequency of specific symptoms between groups. Symptoms persisting after 2 weeks were mainly vertigo, dysarthria, and unsteadiness. Barthel, Rankin, and National Institutes of Health Stroke Scale (NIHSS) results measured 2 weeks after stroke onset are given in Table 2. Clinical scores at presentation were significantly worse in group III compared with group I, whereas in group II, only the Rankin and NIHSS scores were worse than in group I (P 0.01). Patients in group III were significantly older than the patients in group I; however, there was a large overlap. Comorbidity, as described by the Charlson comorbidity index, was low in all groups and not significantly different (Table 3). The majority of infarctions included the posterior inferior cerebellar artery territory, and 1 territory was affected in 7 patients (Table 1). Distribution of the infarct territories between the groups was similar. This was particularly true for superior cerebellar artery (SCA) infarctions. There was no significant difference in infarct size among the groups; however, there was a trend toward larger infarct volumes in group III (P 0.08; Table 3). Silent territorial infarctions were detected in 6 patients only, again with no intergroup difference. General vascular risk factors and hypertension in particular were much more frequent in group II ( 50%) and group III (100%), with large vessel disease being the suggested etiology of the acute stroke in most of the cases, whereas cardiac embolism (atrial fibrillation, patent foramen ovale) was the leading etiology of stroke in group I (Table 1). However, subcortical WM changes were more frequent in patients with worse outcome: whereas no patient in group I exhibited any WMLs, 7 patients in group II and all of the patients in group III showed subcortical damage (Table 4). Severity of WM damage itself was not identical in groups II and III: mild (grade I) and moderate (grade II) WMLs were each present in 10% of the patients of group II; in group III, half of the patients had moderate (grade II) and the other 50% severe (grade III) WM damage. The overall frequency of WMLs was significantly higher in groups II and III compared with group I ( 2 ; P 0.01 and 0.05, respectively). There was a high correlation between the detection and the degree of WML load and the age of patients (r 0.65; P 0.001). Age also predicted patients group assignment (r 0.41; P 0.05). In a logistic regression combining moderate (group II) and severe persistent disability after 14 days (group III), early recovery (group I) was predicted by the absence of WMLs (Z score 5.591; P 0.018) to a slightly higher degree compared with lower age (Z score 4.514; P 0.034). In a multivariate regressional analysis identifying age and WML-associated initial symptoms, only unsteadiness was predicted by patients older age (t 2.52; P 0.021) and by the detection and degree of WMLs (t 2.90; P 0.007). TABLE 4. Frequency of WMLs WML Grade I II III Total Group I n 8 % Group II n 20 % P 0.01 vs group I Group III n 6 % P 0.05 vs group I 2 P 0.03.

4 Grips et al Network Hypothesis of Stroke 1991 A through F, Three posterior inferior cerebellar artery infarctions (DWI, left side) and WMLs (FLAIR, right side). A, B, Patient of group I. C, D, Patient of group II. E, F, Patient of group III. Imaging examples characteristic for each group are shown in the Figure (A through F). Discussion This study presents a second level analysis of the prospective Mannheim Stroke Registry, with data on 34 highly selected patients with acute isolated cerebellar infarction. Except for patients age, which cannot be seen as an illness in itself, there is no significant role for comorbidity, the number of cerebellar territories affected, pre-existent silent territorial strokes, or cerebellar infarct size in the prediction of early outcome after first-ever cerebellar stroke. However, the presence and degree of subcortical WM damage was a strong predictor of early recovery. None of the 8 patients experiencing complete clinical restitution within 48 hours after stroke onset exhibited any WMLs. However, 35% of the patients in group II and all of the patients in group III had mild to severe WMLs that were not associated with any clinically apparent

5 1992 Stroke September 2005 neurological deficits before the acute cerebellar stroke. Strongly correlated with the presence of subcortical changes, outcome in these patients was significantly worse, with WMLs being a slightly stronger predictor of nonbeneficial outcome than age. Apart from age, the presence of WMLs was well matched by the identification of vascular risk factors: none of the patients of group I were experiencing hypertension, 50% in group II, and all of the patients in group III. Other vascular risk factors such as diabetes, hyperlipidemia, smoking, and hyperhomocysteinemia were distributed similarly between the 3 groups. Although stroke etiology differs among groups, we did not consider this to be a relevant parameter for outcome in this specific model. Whereas in the study of Kelly et al, 23 the Charlson comorbidity index did significantly predict long-term outcome, no such association was seen in our study population. Therefore, the degree of subcortical damage seems to be a major factor determining recovery from acute cerebellar lesions, suggesting a mutual compensatory function of the cerebellum for progressive vascular damage to functional neuronal networks as hypothesized in SVE. SVE is a clinical entity that, apart from the presence of WMLs, comprises several clinical features such as gait disturbance, dementia, and urinary incontinence. The fact that isolated cerebellar infarctions in general have a good functional outcome compared with infarctions in other vascular territories is also well established. 17,23,24,28,29 To our knowledge, no study has investigated the association of WMLs with acute cerebellar stroke and vice versa with regard to functional outcome. With regard to supratentorial lesions, Boon et al identified the presence of silent subcortical infarcts and its relation to outcome after acute supratentorial stroke. However, they found no influence of subcortical damage to the degree of initial handicap, 30-day fatality and 1-year mortality. 30 To our understanding, the mechanisms underlying the persistence of neurological deficits, namely vertigo, dysarthria, and unsteadiness after acute cerebellar infarcts in patients with WM disease, are related to pre-existing network disruption. Such missing compensation is evident for the motor loops (the cerebellum receiving information from the cerebral cortex via pontine nuclei 31 and cerebellar efferences projecting to the primary motor cortices via the ventrolateral thalamus 31,32 as part of a feedback control 33 ) and their control: in SVE, the disturbance of the circuitry from frontal and parietal cortex to the basal ganglia via WM tracts causes disturbance of gait, 13 disconnecting supplementary motor cortex, and therefore planning and initiating of locomotion from basal ganglia by multilocular diffuse network destruction. Considering the known complexity of motor circuits, the concept of synergistic lesions is very appealing. However, the proof of single lesions and their network disruptions is rarely described, comprising of only a few patients. 3 5 One of the few established concepts of remote effects is the crossed cerebellar diaschisis, which is defined as depression of blood flow and metabolism in the cerebellar hemisphere contralateral to the focal supratentorial lesion. 34 First observed in positron emission tomography, the phenomenon has been confirmed by means of functional MRI. 35 Although mutual aspects of crossed cerebellar diaschisis are evident, no study has identified primarily infratentorial lesions. Despite the limited absolute number of patients recruited in a monocenter study from a large stroke population reflecting the high degree of selection and exclusion criteria, these data support our hypothesis of a network concept for stroke prognosis. WMLs were a strong predictor of the persistence of symptoms such as vertigo, dysarthria, and unsteadiness after 14 days; this was independent of infarct size or territory of the acute cerebellar infarct but highly correlated with age. These unspecific clinical signs can be secondary to various pathologies, both non-neurological disorders (acid base disturbance, endocrinological or cardiological problems, etc) and multilocular neurological disease (supratentorial and infratentorial, subcortical and cortical, left and right hemispheric). This nonfocal character suggests an important function of subcortical relay stations and connecting fibers to compensate for cerebellar lesions evidenced by loss of network compensation in age-related WM disease. If the prevention or treatment of WMLs could be shown to be associated with risk factor modulation, this would have an important impact on the burden of stroke-associated disability. References 1. Bogousslavsky J. Double infarction in one cerebral hemisphere. Ann Neurol. 1991;30: Mrabet A, Mrad-Ben Hammouda I, Abroug Z, Smiri W, Haddad A. Bilateral infarction of the caudate nuclei. Rev Neurol (Paris). 1994;150: Wolfe N, Babikian VL, Linn RT, Knoefel JE, D Esposito M, Albert ML. Are multiple cerebral infarcts synergistic? Arch Neurol. 1994;51: Hennerici M. Synergistic and remote stroke syndromes. In: Ginsberg MD, Bogousslavsky J, eds. Cerebrovascular Disease. Malden, Mass: Blackwell Science, Inc.; 1998: Fisher CM. Concerning the mechanism of recovery in stroke hemiplegia. Can J Neurol Sci. 1992;19: Kelly RM, Strick PL. Cerebellar loops with motor cortex and prefrontal cortex of a nonhuman primate. J Neurosci. 2003;23: Fazekas F, Niederkorn K, Schmidt R, Offenbacher H, Horner S, Bertha G, Lechner H. White matter signal abnormalities in normal individuals: correlation with carotid ultrasonography, cerebral blood flow measurements, and cerebrovascular risk factors. Stroke. 1988;19: Longstreth WT Jr, Manolio TA, Arnold A, Burke GL, Bryan N, Jungreis CA, Enright PL, O Leary D, Fried L. Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people. The cardiovascular health study. Stroke. 1996;27: Barkhof F, Scheltens P. Imaging of white matter lesions. Cerebrovasc Dis. 2002;13(suppl 2): Bertsch T, Mielke O, Holy S, Zimmer W, Casarin W, Aufenanger J, Walter S, Muehlhauser F, Kuehl S, Ragoschke A, Fassbender K. Homocysteine in cerebrovascular disease: an independent risk factor for subcortical vascular encephalopathy. Clin Chem Lab Med. 2001;39: Taylor WD, MacFall JR, Provenzale JM, Payne ME, McQuoid DR, Steffens DC, Krishnan KR. Serial MR imaging of volumes of hyperintense white matter lesions in elderly patients: correlation with vascular risk factors. AJR Am J Roentgenol. 2003;181: Schmidt R, Enzinger C, Ropele S, Schmidt H, Fazekas F. Progression of cerebral white matter lesions: 6-year results of the Austrian stroke prevention study. Lancet. 2003;361: Baezner H, Oster M, Daffertshofer M, Hennerici M. Assessment of gait in subcortical vascular encephalopathy by computerized analysis: a crosssectional and longitudinal study. J Neurol. 2000;247: Baezner H, Schanz J, Blahak C, Grips E, Woehrle J, Hennerici M. Differential pattern of hand tapping compromise in vascular vs. idiopathic parkinsonism a study based on computerized movement analysis. Mov Dis. 2005;20:

6 Grips et al Network Hypothesis of Stroke Pantoni L, Basile AM, Pracucci G, Asplund K, Bogousslavsky J, Chabriat H, Erkinjuntti T, Fazekas F, Ferro JM, Hennerici M, O Brien J, Scheltens P, Vissar MC, Wahlund LO, Waldemar G, Wallin A, Inzitari D. Impact of age-related cerebral white matter changes on the transition to disability the LADIS study: rationale, design and methodology. Neuroepidemiology. 2005;24: Barth A, Bogousslavsky J, Regli F. The clinical and topographic spectrum of cerebellar infarcts: a clinical-magnetic resonance imaging correlation study. Ann Neurol. 1993;33: Kase CS, Norrving B, Levine SR, Babikian VL, Chodosh EH, Wolf PA, Welch KM. Cerebellar infarction. Clinical and anatomic observations in 66 cases. Stroke. 1993;24: Malm J, Kristensen B, Carlberg B, Fagerlund M, Olsson T. Clinical features and prognosis in young adults with infratentorial infarcts. Cerebrovasc Dis. 1999;9: Canaple S, Bogousslavsky J. Multiple large and small cerebellar infarcts. J Neurol Neurosurg Psychiatry. 1999;66: Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of human brain: brainstem and cerebellum. Neurology. 1996;47: Lie C, Hirsch JG, Rossmanith C, Hennerici MG, Gass A. Clinicotopographical correlation of corticospinal tract stroke: a color-coded diffusion tensor imaging study. Stroke. 2004;35: Glass TA, Hennessey PM, Pazdera L, Chang HM, Wityk RJ, Dewitt LD, Pessin MS, Caplan LR. Outcome at 30 days in the New England Medical Center Posterior Circulation Registry. Arch Neurol. 2002;59: Kelly PJ, Stein J, Shafqat S, Eskey C, Doherty D, Chang Y, Kurina A, Furie KL. Functional recovery after rehabilitation for cerebellar stroke. Stroke. 2001;32: Jauss M, Krieger D, Hornig C, Schramm J, Busse O. Surgical and medical management of patients with massive cerebellar infarctions: results of the German-Austrian Cerebellar Infarction Study. J Neurol. 1999;246: Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40: Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer s dementia and normal aging. AJR Am J Roentgenol. 1987;149: Friston KJ, Ashburner J, Poline JB, Frith CD, Heather JD. Spatial registration and normalization of images. Hum Brain Mapp. 1995;2: Chen HJ, Lee TC, Wei CP. Treatment of cerebellar infarction by decompressive suboccipital craniectomy. Stroke. 1992;23: Tohgi H, Takahashi S, Chiba K, Hirata Y. Cerebellar infarction. Clinical and neuroimaging analysis in 293 patients. The Tohoku Cerebellar Infarction Study Group. Stroke. 1993;24: Boon A, Lodder J, Heuts-van Raak L, Kessels F. Silent brain infarcts in 755 consecutive patients with a first-ever supratentorial ischemic stroke. Relationship with index-stroke subtype, vascular risk factors, and mortality. Stroke. 1994;25: Brodal P, Bjaalie JG. Salient anatomic features of the cortico-pontocerebellar pathway. Prog Brain Res. 1997;114: Asanuma C, Thach WR, Jones EG. Anatomical evidence for segregated focal groupings of efferent cells and their terminal ramifications in the cerebellothalamic pathway of the monkey. Brain Res. 1983;286: Miall RC, Weir DJ, Wolpert DM, Stein JF. Is the cerebellum a smith predictor? J Mot Behav. 1993;25: Infeld B, Davis SM, Lichtenstein M, Mitchell PJ, Hopper JL. Crossed cerebellar diaschisis and brain recovery after stroke. Stroke. 1995;26: Small SL, Hlustik P, Noll DC, Genovese C, Solodkin A. Cerebellar hemispheric activation ipsilateral to the paretic hand correlates with functional recovery after stroke. Brain. 2002;125: