IMAGING OF ACUTE STROKE AND TRANSIENT ISCHAEMIC ATTACK

Transcription

1 See end of artile for authors affiliations Correspondene to: Dr Keith Muir, Division of Clinial Neurosienes, University of Glasgow, Institute of Neurologial Sienes, Southern General Hospital, Glasgow G51 4TF, UK; IMAGING OF ACUTE STROKE AND TRANSIENT ISCHAEMIC ATTACK ACUTE I KWMuir,CSantosh J Neurol Neurosurg Psyhiatry 2005; 76(Suppl III):iii19 iii28. doi: /jnnp t is inreasingly reognised that both stroke and transient ishaemi attaks (TIA) are medial emergenies and that rapid linial and radiologial evaluation underpin the urgent management of erebrovasular disease. The arbitrary duration based separation of stroke from TIA is felt by many to be redundant in the era of stroke treatment, and the term aute ishaemi erebrovasular syndromes (AICS) is a suggested alternative analogous to ardiologial terminology hanges. However, presentation with an aute stroke syndrome diagnosed linially is not synonymous with an AICS, sine there are a number of pathologies that an produe idential linial pitures, and many stroke mimis. Imaging is an essential omponent of diagnosis. The need for inreasingly early imaging has led to new emphasis on hyperaute hanges on plain omputed tomography (CT) and has also seen the widespread use of more omplex imaging modalities in aute stroke. Immediate brain imaging with CT for all stroke patients on admission is more ost effetive than deferred imaging, even when the possible interventions are limited to aspirin use and stroke unit are. A ost effetiveness analysis has not yet been done to take into aount thrombolyti treatment or modalities other than routine CT. STROKE Around 85% of ases of stroke fulfilling the 1976 World Health Organization definition are ishaemi in origin, with 10% aused by foal haemorrhage and 5% by subarahnoid haemorrhage (SAH). Sine SAH rarely presents with sudden foal symptoms, this review will ignore SAH. Many ishaemi strokes exhibit rapid early improvement, leading liniians to apply the term TIA when stritly speaking this label is attahed only when symptoms resolve entirely within 24 hours. Most true TIAs last minutes, and the longer the symptoms last, the greater the likelihood of a ausative lesion being identified on imaging. Haemorrhage Computed tomography Non-ontrast CT (NCCT) remains the gold standard means of deteting intraranial haemorrhage in aute stroke. Blood is hyperdense beause of its high eletron density (fig 1). As blood is broken down, density on CT delines by approximately 1.5 Hounsfield units (HU) per day. Old haemorrhage appears hypodense on CT within a time sale determined by the volume of the initial haematoma. Small bleeds may be indistinguishable from infarts within days of the event. Anatomial loation is relevant in determining the aetiology of primary intraerebral haemorrhage (PICH) for example, small vessel disease most ommonly auses basal ganglia haemorrhage, while lobar haematoma is most ommonly aused by amyloid angiopathy in the elderly. Lobar haemorrhage in younger patients may be due to underlying pathology for example, bleeds seondary to arteriovenous malformations (AVMs) typially extend from the ortial surfae to the lateral ventriles, superior sagittal sinus thrombosis often gives bilateral parasagittal haemorrhages, and thrombosis of the vein of Labbe auses temporal lobe haemorrhage. Cavernomas may ause pontine or supratentorial lobar bleeds. It is now reognised that a high proportion of haematomas expand within the first hours after onset (fig 1), and that expansion is assoiated with poorer outome. 1 With the preliminary demonstration that reombinant fator VII not only redues haematoma expansion but also improves linial outomes in PICH treated within three hours of onset, 2 early reognition of PICH is likely to beome an important diagnosti goal of aute imaging in its own right, and not simply a neessary step in exlusion before onsidering treatment for an ishaemi event. iii19 J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

2 iii20 Figure 1 Expansion of aute primary intraerebral haematoma between (A) three hours and (B) five hours after symptom onset, with development of intraventriular haemorrhage and seondary obstrutive hydroephalus. Vasular imaging Surgial evauation may still be onsidered for some haematomas, partiularly superfiial lobar haematomas, and there may be a need to undertake erebral angiography in order to seek an underlying AVM before surgial deompression or evauation an be planned. If surgery is not antiipated, it is usually advisable to defer vasular imaging studies for some months after an aute intraerebral haemorrhage sine mass effet from any residual haematoma may obsure small low pressure AVMs. External arotid studies may need to be inluded in addition to seletive atheterisation of the internal arotid system in order to identify small dural arteriovenous shunts. Magneti resonane imaging Suseptibility weighted MRI sequenes have been ompared to CT in aute stroke and results to date suggest that MRI is a NEUROLOGY IN PRACTICE Figure 2 Mirobleeds (dark dots) on (A) gradient eho magneti resonane imaging (MRI) and (B) fluid attenuated inversion reovery (FLAIR) of the same slie. good alternative for the detetion of haemorrhage. 3 However, further omparative evaluation is needed before MRI an be regarded as a substitute. In investigation of stroke with delayed presentation, gradient eho MRI is the investigation of hoie for exlusion of old haemorrhage. On gradient eho MRI, old bleeds are of low signal. Gradient eho MRI inreasingly identifies mirohaemorrhages in the brain in individuals with no linial history to suggest intraerebral haemorrhage (fig 2). These mirobleeds may be a risk fator for spontaneous bleeds after thrombolyti treatment, and offer an explanation for the ourrene of haematomas that are remote from the site of ishaemia for whih treatment was given. It remains to be established definitively whether the presene of mirobleeds on gradient eho MRI represents a ontraindiation to systemi thrombolysis for ishaemi stroke, although some investigators believe that it does. J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

3 NEUROLOGY IN PRACTICE Figure 3 Evolution of hypodensity on omputed tomography (CT). Right MCA ishaemia at (A) four hours and (B) 24 hours after onset. First CT shows early ishaemi hanges: loss of grey-white matter definition, lentiform nuleus hypodensity, loss of insular ribbon, ompression of lateral ventrile, and hemispheri sulal effaement. Aute ishaemi stroke Computed tomography NCCT remains the mainstay of emergeny imaging of stroke in order to exlude intraranial haemorrhage. NCCT may also identify other intraranial pathologies that mimi stroke suh as tumour or enephalitis. Ishaemi tissue on NCCT appears hypodense beause of a ombination of redued blood volume and ytotoxi oedema. The rate of deline of tissue density is dependent upon severity and duration of ishaemia (fig 3). Within the three hour window for systemi thrombolyti treatment, hypodensity is usually subtle if visible at all. More learly visible hypodensity should always prompt reappraisal of the history around time of onset, sine it suggests a greater duration of ishaemia. Early ishaemi hange (EIC) on NCCT (table 1, fig 4) is a term enompassing hanges that almost ertainly represent a Table 1 Early ishaemi hanges on non-ontrast CT Hyperdense arteries (most ommonly proximal MCA or MCA sylvian dot ) Lentiform nuleus hypodensity Loss of insular ribbon (definition of grey from white matter) Loss of ortial grey-white matter differentiation Hemispheri sulal effaement Loal ompression of lateral ventriles MCA, middle erebral artery. number of different pathologial proesses in aute ishaemia whose signifiane varies. Previous radiologial prejudie that CT within a few hours of stroke onset has low sensitivity is unfounded, at least in middle erebral artery (MCA) olusions, where EICs are present in around 70% of ases within three hours of onset. While the sensitivity of these hanges is ompromised by their subtlety, inter-observer reliability an be improved by systemati CT san evaluation using systems suh as the Alberta stroke programme early CT sore (ASPECTS). 4 Inter-observer agreement is improved signifiantly by linial information being available. A reent large multi-observer omparative study found inter-observer agreement to be greater among neuroradiologists than stroke neurologists or general radiologists. The majority of EICs are features of redued tissue density: it is not defined (and probably indefinable) at what point EIC merges into visible hypodensity. The arbitrary distintion between the two may be of linial signifiane sine visible hypodensity involving a large anatomial volume inreases the risk of poor outome and ompliations of thrombolyti treatment. 5 Isodense brain swelling, another EIC, in the aute phase probably represents inreased blood volume, a physiologial vasodilator response to ishaemia indiating metabolially ative tissue. Isodense swollen regions may therefore represent reversible ishaemia. EICs per se were not a predefined exlusion riterion in any thrombolysis trial, and therefore in themselves are not an exlusion from thrombolyti treatment. Extensive visible hypodensity is a risk fator for both poor outome and higher risk of haemorrhage, whih is unsurprising sine the more obvious the hypodensity, the greater the severity (for example, beause of lak of ollateral supply) or the duration of ishaemia. Most defined EIC on CT, and systems suh as ASPECTS are onerned exlusively with stroke aused by olusion of the arotid artery, the main trunk of the MCA, or the major branhes of the MCA. The sensitivity of CT to ishaemia within small penetrating artery territories, the posterior irulation, or sattered multifoal small infarts that are often enountered in emboli stroke, is not established, and tehnial limitations mean that CT sensitivity in these senarios is likely to be poor. Vessel hyperdensity Inreased density of the MCA or other intraranial vessels on NCCT is indiative of thrombus partially or ompletely oluding the vessel. The plane of setion of CT means that main trunk MCA olusions are seen as a linear hyperdensity in the sylvian fissure, while internal arotid artery (ICA) or branh MCA olusions may be seen as hyperdense dots in ross setion. False positive hypderdense MCAs may be seen, partiularly in onditions assoiated with inreased haematorit (for example, polyythaemia) or where hypo- iii21 J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

4 iii22 Figure 4 Two examples of early ishaemi hanges on CT in sub-three hour ishaemi stroke. (A) Aute left middle erebral artery (MCA) ishaemia with loss of insular ribbon, and hypodensity of tail of lentiform nuleus. (B) Similar hanges in a different patient, with hypodensity of entire lentiform nuleus, some hemispheri sulal effaement, and ompression of lateral ventrile aused by hypodensity of audate head. density of brain parenhyma leads to inreased ontrast with normal vessels (for example, herpes enephalitis). Contrast CT Routine use of ontrast enhaned CT is of limited additional diagnosti value in aute stroke and is not reommended, although onerns that blood brain barrier breakdown would lead to ontrast extravasation with risk of stroke worsening are not supported by evidene. Inreased onspiuity of ishaemi lesions within six hours of onset on soure images from CT angiography (CTA) examinations has been reported, but in effet the high dose ontrast administration for CTA yields an image representing erebral blood volume (CBV). Dereased CBV orresponds with infart ore. CT using routine doses of ontrast is not validated in this NEUROLOGY IN PRACTICE respet, and in general the use of ontrast agents should be to aquire additional information from CTA or CT perfusion (CTP), or to address speifi diagnosti onerns about alternative pathologies. CT angiography CTA of intraranial vessels an identify the site of vessel olusion, whih may be of value in linial management deisions. For example, the response to intravenous thrombolyti treatment of tandem olusions of the ipsilateral ICA and MCA, arotid T olusions, or of basilar artery thrombosis, is poor ompared to isolated MCA olusion, and in many entres is onsidered a potential indiation for resue therapy with intra-arterial thrombolytis or mehanial embolus removal. CT perfusion Multidetetor CT sanners allow the aquisition of several slies of brain repeatedly during the intravenous passage of high doses of iodinated ontrast medium. The hanges in the density time urve for eah pixel allow alulation of a number of parameters refleting tissue perfusion by mathematial alulations based around the entral volume priniple. Typial derived parameters inlude mean transit time (MTT), time to bolus peak (TTP), and CBV, from whih erebral blood flow (CBF) an be alulated (as MTT/CBV). TTP and MTT in the first 3 6 hours after stroke onset are preditive of final infart volume in the absene of reperfusion, and represent tissue at risk. Diminished CBV probably represents failure of autoregulatory responses and therefore tissue infartion. The differene between CBV and TTP or MTT lesions an be taken as an estimate of the ishaemi penumbra, the volume of tissue at risk of infartion but still viable (fig 5). 6 PCT has been validated against other tehniques suh as diffusion and perfusion MRI, and quantitative PET. Claims that CTP is itself apable of quantitative blood flow measurement are not universally aepted. High dose ontrast administration for CTA or CTP arries a risk of renal impairment and also neessitates disontinuation of metformin in diabetis to avoid preipitating lati aidosis, a rare ompliation. There is also a risk of allergi reations. The additional time required for the examination, and the need for a patient to lie still during sanning, may present problems with autely ill patients. Magneti resonane imaging Conventional MRI sequenes suh as T2 weighted images arry little advantage over NCCT in sensitivity to stroke within the first hours. However, newer sequenes, notably diffusion weighted MRI (DWI) and dynami ontrast bolus traking perfusion MRI (generally referred to as perfusion weighted imaging, PWI) offer onsiderable inreases in diagnosti sensitivity and are at present better validated than CT tehniques in defining pathophysiologial parameters suh as tissue viability in aute ishaemi stroke. DWI is based on the detetion of the mobility of water moleules, measured as the apparent diffusion oeffiient (ADC) of water. In ishaemia, energy failure ompromises ellular ion pumps that normally extrude sodium, with resultant entry of sodium and extraellular water into ells (ytotoxi oedema). This is evident as redued ADC signal (intraellular water an diffuse less freely than extraellular J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

5 NEUROLOGY IN PRACTICE spae water) and this is proessed to show as bright on DWI (fig 6). DWI is highly sensitive to ishaemia, perhaps greater than 95% within the first hours, and hanges are doumented within 40 minutes of symptom onset in humans (and within two minutes of onset in animal models). Lesion onspiuity is greatly improved ompared to other sequenes or imaging modalities (fig 7). DWI hanges are not speifi, however, and an be seen in foal seizures, enephalitis, and possibly also migraine. Interpretation should also take into aount the phenomenon of T2 shine through, a term denoting visibility on DWI of non-aute lesions that are bright on T2 weighted sequenes. In order to onfirm whether a DWI lesion represents aute ishaemia, an ADC map should be examined to ensure ADC is redued orrespondingly. The inreased DWI signal gradually fades over around 7 10 days (dependent on severity of ishaemia and on lesion volume) to an isointense bakground, so DWI is most useful in differentiating reent from remote ishaemia. Persistene of DWI lesions in some patients after TIA or minor stroke is reported, extending out to several months after symptoms. The signifiane of protrated DWI lesions is not known. Figure 5 (A) Perfusion CT (time to peak (TTP) map) of aute left MCA ishaemia sub-three hours and (B) equivalent nonontrast CT. Perfusion CT predits final infart volume in the absene of arterial reanalisation: (C) three hour TTP map and (D) 24 hour non-ontrast CT. Perfusion MRI is most ommonly applied as bolus traking during the intravenous administration of gadolinium, with the same priniples as those governing CTP imaging allowing the derivation of TTP, MTT, CBV, and CBF. The signal intensity is redued as gadolinium passes through tissue, rather than the inreased density with iodinated ontrast in CTP. PWI has the advantage of aquisition of perfusion data for the entire brain, whereas the physial size of the CT detetor is limited to smaller volumes (generally 20 mm slie thikness) in most systems in urrent linial use. More widely spaed detetors will enable the entire brain to be imaged simultaneously. MRI PWI is also a better validated tehnique with respet to aute stroke, and is more widely available. Arterial spin labelling, a newer tehnique that measures perfusion without ontrast agents, remains experimental at present. The DWI PWI mismath hypothesis In the first hours after stroke, DWI signal hange is postulated to represent irreversible tissue injury, and therefore indiate the infart ore. By omparison with the perfusion defet on PWI, it is possible to define a diffusion perfusion mismath that is proposed to represent iii23 J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

6 iii24 Figure 6 Apparent diffusion oeffiient (ADC) map and diffusion weighted MRI (DWI) of subaute stroke (day 3). Restrited diffusion of water leads to dereased signal on ADC, inreased signal on DWI. an MRI signature of the ishaemi penumbra. In perhaps as many as 70% of aute strokes aused by MCA olusion imaged within six hours of onset, a DWI PWI mismath is present, the PWI lesion (hypoperfused) being larger than the DWI ( infart ore ). Over time, the DWI lesion expands to eventually inorporate most of the PWI defet (fig 8). 7 This evolution over time is onsistent anatomially with the progression of penumbra to final infart. The mismath appearane is therefore a potential tool to selet patients in whom there is evidene of potentially salvageable tissue, either for linial trials, or for individualised treatment. In the small DIAS trial, 8 intravenous thrombolyti treatment given to patients seleted on the basis of a mismath improved linial and radiologial outomes even though treatment was between 3 9 hours after stroke onset. Confirmation of the linial utility of the mismath hypothesis will ome from further trials that are planned or ongoing. Reent studies omparing DWI PWI mismath with PET have shown, however, that, while the mismath tissue overlaps onsiderably, it does not orrespond with PET defined metaboli NEUROLOGY IN PRACTICE abnormalities signifying the penumbra, and metaboli patterns within the mismath region are omplex. 9 MR angiography (MRA) in the early stages of stroke an identify the site of arterial olusion in muh the same manner as CTA. Time of flight MRA does not require ontrast, but is longer to perform, and therefore often diffiult in aute stroke patients. Shorter time of flight sequenes are of poorer quality. Contrast enhaned MRA improves quality of imaging and shortens imaging time. Other MR sequenes suh as spetrosopy are of researh value only at the present time. Patient tolerability may be a limiting fator in aute stroke MRI: in addition to the onventional MR ompatibility issues suh as ferromagneti implants, paemakers, and metalli foreign bodies, duration of examination is a onern sine patient monitoring is ompromised by the physial onstraints of the sanner. While vital signs an be monitored with MRI ompatible equipment, it is diffiult to deal with a patient vomiting while undergoing an MRI san. Claustrophobia an be problemati, but more often in onvalesent patients and those with minor strokes. Despite these onerns, areful seletion of sequenes ensures that multiparametri MRI is well tolerated and widely used in aute stroke, and is the investigation of first hoie in many stroke entres worldwide. Single photon emission omputed tomography (SPECT) SPECT blood flow imaging uses traers tagged to moleules that are delivered to tissue and fixed in proportion to blood flow (for example, hexamethylpropylene amine oxime (HMPAO), ethyl ysteinate dimer (ECD)). This produes qualitative CBF data, and has the advantage that uptake reflets blood flow at the time of injetion; the san itself an be deferred for several hours without affeting the ability to image this snapshot of perfusion. The duration of a full SPECT san aquisition (around 40 minutes) is too long for routine linial use, but SPECT has produed valuable researh data. Speifi ligands suh as the neuronal marker iomazenil or the NMDA reeptor traer CNS 1261 are of researh value only at present. Positron emission tomography (PET) Multi-traer PET has been invaluable in defining the pathophysiology of aute stroke, but the tehnique is onfined to researh use beause of several fators, inluding the requirement for a ylotron to produe radiotraers in lose proximity to linial ativity, and need for arterial aess to produe quantitative data inreasingly diffiult in the thrombolyti era. Both PET and SPECT are unsuited to serial imaging in individual patients beause of the radiation dose involved. Xenon inhalation CT (Xe-CT) This is a theoretially attrative tehnique sine it is able to produe quantitative CBF data, based on the inhalation of known onentrations of xenon and hanges in tissue density that are dependent on tissue onentration. While some useful researh data have arued from Xe-CT, diffiulties in administration of xenon (whih has anaestheti properties) in aute patients have limited the use of this modality of investigation. J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

7 NEUROLOGY IN PRACTICE Figure 7 Improved lesion onspiuity of DWI in aute ishaemi stroke. (A) DWI and (B) CT in aute right MCA olusion. CT shows early ishaemi hanges (sulal effaement, loss of greywhite differentiation, ompression of lateral ventrile, loss of posterior lentiform nuleus definition, anterior insular ribbon loss). (C) DWI and (D) onventional T2 weighted MRI in multifoal (post-thrombolyti treatment) right MCA infartion. iii25 J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from Figure 8 Diffusion perfusion mismath in sub-six hour aute stroke. (A) DWI (degraded by movement artefat) shows signal hange onfined to basal ganglia (MCA perforator) territory and a small area of abnormal signal in posterior ortial MCA territory. (B) Mean transit time (MTT) perfusion MRI shows prolonged MTT throughout entire left MCA territory. (C) Day 3 infart on DWI showing expansion of lesion to fill hypoperfused lesion volume. on 22 January 2019 by guest. Proteted by opyright.

8 iii26 Figure 9 CT angiography. (A) Proximal third basilar artery stenosis (4D false olour reonstrution of intraranial CT angiography). (B) Spontaneous dissetion of internal arotid artery leading to tapered olusion. (C, D) 4D and multi-planar reformat of arotid bifuration showing stenosis of the internal arotid artery and assoiated plaque uleration. Transranial Doppler ultrasound (TCD) Pulsed wave 2 MHz ultrasound via the temporal bony window in TCD an provide diagnosis of olusive disease of the major branhes of the irle of Willis, and lends itself well to ontinuous monitoring in the aute phase for example, during thrombolyti treatment to determine whether (and when) reanalisation ours. Reent reports support the proposal that diagnosti TCD may enhane lot lysis by reombinant tissue plasminogen ativator (rt-pa) with higher reanalisation rates in the CLOTBUST linial trial, and are baked by experimental evidene. Higher energy ultrasound systems have led to poorer outome beause of higher rates of intraerebral haemorrhage. However, TCD is very user dependent, and onfident identifiation of the major intraranial vessels may be diffiult, partiularly so when one is oluded. NON-ACUTE STROKE CT shows strutural hanges refleting tissue loss after stroke, but has notable limitations in linial use: these inlude poor lesion visibility in the posterior fossa aused by surrounding bone, poor ability to delineate small ortial infarts adjaent to erebrospinal fluid (CSF) spaes, and inability to distinguish old ishaemi stroke from old haemorrhage. In addition, in the subaute phase (around days after the itus), the phenomenon of fogging may obsure a reent infart; in some instanes the CT may appear normal, even after large hemispheri lesions. The isodense appearane is probably aused by a ombination of petehial bleeding, and tissue infiltration by marophages and other inflammatory ells. MRI avoids many of these problems, provided some thought is given to the sequenes that will provide linially relevant information: routine T2 weighted fast spin eho, T1 weighted and proton density images share many of the limitations of CT. A CSF suppressed sequene suh as fluid attenuated inversion reovery (FLAIR) improves lesion NEUROLOGY IN PRACTICE onspiuity, notably for lesions adjaent to CSF spaes, and may distinguish enlarged Virhow-Robin spaes in the basal ganglia and apsule from launar infarts by demonstrating glioti signal hange. FLAIR on some sanners is less sensitive to posterior fossa lesions, where a routine T2 sequene may be superior. Gradient eho MRI is sensitive to haemoglobin degradation produts and so will reliably detet old areas of haemorrhage. Extraranial vasular imaging Doppler ultrasound of the arotid arteries remains the mainstay of extraranial vasular imaging in the UK sine it is non-invasive and available in most hospitals. However, orret identifiation of the arotids is operator dependent, and grading of stenosis being dependent upon measurement of flow veloity may be impreise. Few departments hek the validity of loal Doppler estimates of stenosis against more objetive tehniques, partiularly as onventional angiography is now seldom performed even preoperatively. The anatomial overage of ultrasound is limited to the region of the arotid bifuration and therefore omits intraranial stenosis and aorti arh disease. Visualisation of the vertebral arteries by ultrasound is poor and limited in range. Ultrasound may be a reasonable sreening test to identify patients with onventionally defined surgial arotid stenosis of. 70%. Surgial pratie in onfirming ultrasound measurements with other tehniques is varied, with MRA being the most ommon seond modality. CTA offers greater anatomial overage than ultrasound, being able to inlude the vasular tree from the arh of the aorta to the irle of Willis in a single examination. This offers the potential to sreen for potentially relevant linial fators, suh as intraranial stenosis (fig 9), that may impat upon the assessment of perioperative risk, and also identify vertebrobasilar disease and patterns of ollateral supply that may be linially important. On the negative side, CTA requires large intravenous ontrast doses with attendant risks (see above) and involves proessing by a radiologist. J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

9 NEUROLOGY IN PRACTICE Figure 10 (A) DWI and (B) CT 24 hours after transient ishaemi attak of 30 minutes duration. The presene of an aute DWI lesion may be of prognosti signifiane. MRA offers similar anatomial overage to CTA without the need for ontrast, if time of flight sequenes are used. This may overestimate the degree of stenosis sine turbulene in the region of a severe stenosis leads to loss of signal. Signal dropout adjaent to alifi atheroma may also interfere with assessment. Contrast MRA may avoid some of these problems at the expense of being more invasive. Although intra-arterial angiography is the onventional standard for vasular imaging, onerns over the assoiated morbidity (probably 1 5% in older patients with atherosleroti disease) have led to its replaement by the less invasive tehniques of CTA and MRA. However, the dynami quality of angiography does yield information that may be diffiult to obtain from other tehniques for example, the immediately harateristi appearane of Moya Moya disease, or haraterisation of arteriovenous fistulae. In most ases this tehnique is now reserved for partiular diagnosti Abbreviations ADC: apparent diffusion oeffiient ASPECTS: Alberta stroke programme early CT sore CBF: erebral blood flow CBV: erebral blood volume CSF: erebrospinal fluid CT: omputed tomography CTA: CT angiography CTP: CT perfusion DWI: diffusion weighted MRI ECD: ethyl ysteinate dimer EIC: early ishaemi hange FLAIR: fluid attenuated inversion reovery HMPAO: hexamethylpropylene amine oxime ICA: internal arotid artery MCA: middle erebral artery MRA: magneti resonane angiography MRI: magneti resonane imaging MTT: mean transit time NCCT: non-ontrast CT PET: positron emission tomography PWI: perfusion weighted MRI SPECT: single photon emission omputed tomography rt-pa: reombinant tissue plasminogen ativator TCD: transranial Doppler ultrasound TTP: time to peak Xe-CT: xenon inhalation CT or tehnial diffiulties, most of whih an be resolved by CTA or MRA. TIA (AND MINOR STROKE) The majority of patients with true TIA (that is, omplete resolution of all symptoms within 24 hours) and minor stroke have delayed presentation and attend as outpatients. Reognition of the high early risk in some patients, 10 and the advent of thrombolyti treatment, is hanging the pattern of presentation to a greater proportion being seen early and admitted to hospital, and is also inreasing the pressure to investigate rapidly. NCCT is of limited value, sine haemorrhage is very rarely a ause of TIA, and most ishaemi episodes are aused by very small or sattered lesions to whih CT has poor sensitivity. Immediate MRI, partiularly DWI, performed early is of muh greater potential value sine the yield of ausative lesions is far greater (fig 10). 11 This has both immediate linial value in assisting loalisation often very diffiult based upon history alone, and of onsiderable importane in deisions about surgial endarteretomy and has reently been reognised as having prognosti value. 12 Those patients with DWI lesions (partiularly with evidene of intraranial vessel olusions) had a far higher risk of subsequent stroke than those without in a prospetive series of TIA and minor stroke patients. Extraranial vasular imaging is of partiular importane in these patients, both beause arotid stenosis appears to have a higher early reurrene risk than other stroke aetiologies, and also beause endarteretomy has greatest absolute benefit if performed within two weeks of the event Authors affiliations K W Muir, Division of Clinial Neurosienes, University of Glasgow, Institute of Neurologial Sienes, Southern General Hospital, Glasgow, UK iii27 J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.

10 iii28 C Santosh, Department of Neuroradiology, Institute of Neurologial Sienes, Southern General Hospital The authors wish to thank Dr Evelyn Teasdale for providing the CTA images in fig 9. REFERENCES 1 Brott T, Broderik J, Kothari R, et al. Early hemorrhage growth in patients with intraerebral hemorrhage. Stroke 1997;28: Mayer SA, Brun NC, Begtrup K, et al. Reombinant ativated fator VII for aute intraerebral hemorrhage. N Engl J Med 2005;352: Kidwell CS, Chalela JA, Saver JL, et al. Comparison of MRI and CT for detetion of aute intraerebral hemorrhage. JAMA 2004;292: Barber PA, Demhuk AM, Zhang J, et al. Validity and reliability of a quantitative omputed tomography sore in prediting outome of hyperaute stroke before thrombolyti therapy. ASPECTS study group. Alberta stroke programme early CT sore. Lanet 2000;355: Larrue V, von Kummer R, del Zoppo G, et al. Hemorrhagi transformation in aute ishemi stroke. Potential ontributing fators in the European ooperative aute stroke study. Stroke 1997;28: Wintermark M, Reihhart M, Thiran JP, et al. Prognosti auray of erebral blood flow measurement by perfusion omputed tomography, at the time of NEUROLOGY IN PRACTICE emergeny room admission, in aute stroke patients. Ann Neurol 2002;51: Beaulieu C, de Crespigny A, Tong DC, et al. Longitudinal magneti resonane imaging study of perfusion and diffusion in stroke: evolution of lesion volume and orrelation with linial outome. Ann Neurol 1999;46: Hake W, Albers G, Al Rawi Y, et al. The desmoteplase in aute ishemi stroke trial (DIAS): a phase II MRI-based 9-hour window aute stroke thrombolysis trial with intravenous desmoteplase. Stroke 2005;36: Guadagno JV, Warburton EA, Aigbirhio FI, et al. Does the aute diffusionweighted imaging lesion represent penumbra as well as ore? A ombined quantitative PET/MRI voxel-based study. J Cereb Blood Flow Metab 2004;24: Johnston SC, Gress DR, Browner WS, et al. Short-term prognosis after emergeny department diagnosis of TIA. JAMA 2000;284: Lee LJ, Kidwell CS, Alger J, et al. Impat on stroke subtype diagnosis of early diffusion-weighted magneti resonane imaging and magneti resonane angiography. Stroke 2000;31: Coutts SB, Simon JE, Eliasziw M, et al. Triaging transient ishemi attak and minor stroke patients using aute magneti resonane imaging. Ann Neurol 2005;57: Rothwell PM, Eliasziw M, Gutnikov SA, et al. Endarteretomy for symptomati arotid stenosis in relation to linial subgroups and timing of surgery. Lanet 2004;363: J Neurol Neurosurg Psyhiatry: first published as /jnnp on 16 August Downloaded from on 22 January 2019 by guest. Proteted by opyright.