addition of endovascular thrombectomy | My Assignment Tutor Vol 389 February 11, 2017 641SeminarStrokeGraeme J HankeyIn the past decade, the defi nition of stroke has been revised and major advances have been made for its treatment andprevention. For acute ischaemic stroke, the addition of endovascular thrombectomy of proximal large artery occlusionto intravenous alteplase increases functional independence for a further fi fth of patients. The benefi ts of aspirin inpreventing early recurrent ischaemic stroke are greater than previously recognised. Other strategies to prevent recurrentstroke now include direct oral anticoagulants as an alternative to warfarin for atrial fi brillation, and carotid stenting asan alternative to endarterectomy for symptomatic carotid stenosis. For acute intracerebral haemorrhage, trials areongoing to assess the eff ectiveness of acute blood pressure lowering, haemostatic therapy, minimally invasive surgery,anti-inflammation therapy, and neuroprotection methods. Pharmacological and stem-cell therapies promise to facilitatebrain regeneration, rehabilitation, and functional recovery. Despite declining stroke mortality rates, the global burden ofstroke is increasing. A more comprehensive approach to primary prevention of stroke is required that targets people atall levels of risk and is integrated with prevention strategies for other diseases that share common risk factors.IntroductionThe world is facing an epidemic of stroke. Despite stableincidence rates and declining mortality rates over the pasttwo decades, the number of incident strokes, prevalentstroke survivors, disability-adjusted life-years (DALYs) lostdue to stroke, and stroke-related deaths is increasing.1 ThisSeminar highlights recent developments in the definition,treatment, and prevention of stroke to help clinicians tomanage stroke and reduce its impact on affectedindividuals, their carers, and the population as a whole.EpidemiologyIn 2010, an estimated 16·9 million incident strokesoccurred, which added to a pool of 33 million strokesurvivors worldwide (table 1).1 There were 5·9 milliondeaths and 102 million DALYs lost due to stroke, makingstroke the second leading cause of death after ischaemicheart disease and third leading cause of DALYs lostworldwide. Most of the global burden of stroke, in termsof deaths and DALYs lost, was borne by low-income andmiddle-income countries (LMICs) and caused byhaemorrhagic stroke.2Between 1990 and 2010, the global incidence rate ofstroke remained stable but the number of incident firststrokes increased by 68%. The prevalence of strokeincreased slightly, yet the number of stroke survivorsincreased by 84%. The number of DALYs lost per strokedecreased, but the total number of DALYs lost increasedby 12%. The mortality rate fell, but the number ofstroke-related deaths increased by 26% (table 1).1 Thereduction in rates can probably be attributed to improvedprevention and management of stroke, particularly inhigh-income countries. The increase in numbers, despitereductions in rates, probably refl ects global populationgrowth, increasing life expectancy, and a change in theage structure of most populations.Definition of stroke and transient ischaemic attackThe traditional definition of stroke is clinical and basedon the sudden onset of loss of focal neurological functiondue to infarction or haemorrhage in the relevant part ofthe brain, retina, or spinal cord. Stroke is distinguishedfrom transient ischaemic attack (TIA) if the symptomspersist longer than 24 h (or lead to earlier death). Anupdated defi nition of stroke is an acute episode of focaldysfunction of the brain, retina, or spinal cord lastinglonger than 24 h, or of any duration if imaging (CT orMRI) or autopsy show focal infarction or haemorrhagerelevant to the symptoms.3 The definition includessubarachnoid haemorrhage.3 A TIA has been redefinedas focal dysfunction of less than 24 h duration and withno imaging evidence of infarction.3Diagnosis of strokeTypical symptoms of stroke include sudden unilateralweakness, numbness, or visual loss; diplopia; alteredspeech; ataxia; and non-orthostatic vertigo.4 Associatedsymptoms (eg, headache) vary and usually reflect thecause or consequences of the stroke. Atypical symptomsof stroke include isolated vertigo, binocular blindness,amnesia, anosognosia, dysarthria, dysphagia, stridor,foreign accent, or headache; hemiballismus; alien handsyndrome; confusion; and altered consciousness.4Diagnostically, the Face Arm and Speech Test (FAST)aids screening for stroke and is as sensitive and specificas the Recognition of Stroke in the Emergency Room(ROSIER) score.4,5 Non-contrast cranial CT scan hasnear-perfect sensitivity to detect fresh intracranialhaemorrhage, but its sensitivity for diagnosis ofischaemic stroke is poor if ischaemia is recent, small, orin the posterior fossa. Diffusion weighted MRILancet 2017; 389: 641–54Published OnlineSeptember 13, 2016 of Medicine &Pharmacology, The Universityof Western Australia, Perth,WA, Australia(Prof G J Hankey MD);Department of Neurology,Sir Charles Gairdner Hospital,Perth, WA, Australia(Prof G J Hankey); and WesternAustralianNeuroscience ResearchInstitute (WANRI), Perth, WA,Australia (Prof G J Hankey)Correspondence to:Prof Graeme J Hankey, School ofMedicine and Pharmacology,The University of WesternAustralia, Perkins Institute ofMedical Research, QEII MedicalCentre, Perth, WA 6009,[email protected] strategy and selection criteriaI searched the Cochrane Library, PubMed, and MEDLINE usingthe search term “stroke” in combination with the terms“diagnosis”, “risk factors”, “prognosis”, “treatment”, and“prevention” for articles published between Jan 1, 2010, andJune 1, 2016. I also searched the reference lists of articlesidentifi ed by the search. I selected mainly articles published inthe past 5 years, but included older key publications.Seminar642 Vol 389 February 11, 2017(DWI-MRI) detects acute brain ischaemia in about 90%of patients with ischaemic stroke and about a third ofpatients with transient symptoms lasting less than 24 h.6,7DWI-MRI can be suggestive of stroke in patients with astroke mimic (eg, seizures, migraine, hypoglycaemia,tumour, encephalitis, abscess, and multiple sclerosis).Gradient-echo T2-weighted susceptibility MRI is assensitive as CT for acute haemorrhage and more sensitivefor previous haemorrhage. About 20–25% of patientspresenting with a stroke syndrome have a stroke mimic;most commonly seizures, syncope, sepsis, peripheralvestibulopathy, and toxic or metabolic encephalopathy.8The diagnosis of stroke is most diffi cult in the initialhours, particularly if the onset is uncertain, the featuresare atypical or changing, the patient is unwell or agitated,access to imaging is delayed, or brain imaging is normal.Subtypes of strokeClinical ischaemic stroke syndromes include totalanterior circulation syndrome, partial anterior circulationsyndrome, lacunar syndrome, and posterior circulationsyndrome.9 Pathological subtypes comprise ischaemicstroke (cerebral, retinal, and spinal infarction) andhaemorrhagic stroke (intracerebral haemorrhage andsubarachnoid haemorrhage). The proportions ofpathological and aetiological subtypes of stroke varyamong populations of diff erent age, race, ethnic origin,and nationality.Aetiologically, ischaemic stroke is caused by embolismfrom the heart, artery-to-artery embolism, and in-situsmall vessel disease. Aetiological subtypes of ischaemicstroke are classifi ed according to the TOASTclassification,10 the ASCOD phenotyping system(A: atherosclerosis; S: small-vessel disease; C: cardiacpathology; O: other cause; D: dissection),11 and theCausative Classification System.12 A third of ischaemicstrokes remain of undetermined cause (ie, cryptogenic),of which a subgroup is now defi ned as having embolicstrokes of undetermined source.13Haemorrhagic stroke is classifi ed according to itsanatomical site or presumed aetiology. The mostcommon sites of intracerebral haemorrhage aresupratentorial (85–95%), including deep (50–75%) andlobar (25–40%).14 The most common causes arehypertension (30–60%), cerebral amyloid angiopathy(10–30%), anticoagulation (1–20%), and vascularstructural lesions (3–8%); the cause is undetermined inabout 5–20% of cases.14Risk factorsHypertension, hypercholesterolaemia, carotid stenosis,and atrial fi brillation are known to be causal risk factorsfor stroke because clinical trials have shown thattreatment of these conditions reduces the incidence ofstroke.15–18 Cigarette smoking, excessive alcohol use,insulin resistance, and diabetes mellitus are also likelycausal risk factors.19–22 Other risk factors that, if modified,could reduce the incidence of stroke includeenvironmental air pollution, childhood healthcircumstances and fitness, high-risk diet and poornutrition, physical inactivity, obesity, blood pressurevariability, sleep-disordered breathing, chronic inflammation, chronic kidney disease, migraine, hormonalcontraception or hormone replacement therapy,psychosocial stress, depression, job strain, and longworking hours.23,24Besides rare highly penetrant mendelian mutationsthat cause early-onset stroke, several genetic loci havebeen associated with ischaemic stroke (eg, chromosome12q24.12 near ALDH2) and its subtypes—eg, the ZFHX31990 2005 2010 Change from 1990–2010Number ofeventsRate per 100 000person-yearsNumber ofeventsRate per 100 000person-yearsNumber ofeventsRate per 100 000person-yearsChange innumber ofeventsChange in rate in HICs Change in rate in LMICsAll stroke Incidence10 078 935Prevalence17 915 338DALYs lost 86 010 384Deaths4 660 449Ischaemic strokeIncidenceDALYs lost7 238 75832 128 220Deaths2 241 077Haemorrhagic strokeIncidenceDALYs lostDeaths2 840 17753 882 1642 419 372 ····decreaseData in parentheses are 95% CI. HIC=high-income country. LMIC=low-income and middle-income country. DALY=disability-adjusted life-year.Table 1: Age-adjusted annual incidence and mortality rates, prevalence, and DALYs lost for all stroke, ischaemic stroke, and haemorrhagic stroke1, Vol 389 February 11, 2017 643gene on chromosome 16q22 and PITX2 gene onchromosome 4q25 for cardioembolic stroke; the HDAC9gene on chromosome 7p21 and locus on chromosome1p13.2 near the TSPAN2 gene for large-vessel stroke; andchromosome 6p25 near the FOXF2 gene for small-vesseldisease.25,26 Genetic variants ε2 and ε4 within theapolipoprotein E (APOE) gene are risk factors for lobarintracerebral haemorrhage.Ten treatable risk factors account for about 90% of thepopulation-attributable risk of stroke.23,24 Stroke can also betriggered by several activities (eg, neck trauma and coitus)and risk factors (eg, alcohol, amphetamines, infection, andair pollution, and perhaps psychosocial stress).27Prognosis after stroke and TIAThe case fatality rates after all stroke are about 15% at1 month, 25% at 1 year, and 50% at 5 years.28 Afterintracerebral haemorrhage, the case fatality rates areabout 55% at 1 year and 70% at 5 years.29 About 40% ofstroke survivors are disabled (modifi ed Rankin Scale[mRS] score 3–5) between 1 month and 5 years afterstroke; 20% are disabled before the stroke.28Five variables(age, verbal component of the Glasgow Coma Scale, armpower, ability to walk, and pre-stroke dependency) predictindependent survival at 3 months and 12 months afterstroke.30 Other prognostic factors include stroke severity,clinical subtype, employment status, marital status, andrecurrent stroke.28After ischaemic stroke and TIA, the risk of recurrentstroke without treatment is about 10% at 1 week, 15%at 1 month, and 18% at 3 months.31 The risk is greateramong individuals with recent symptomatic atherosclerosis and high ABCD³-I and recurrence risk estimatorscores (table 2);32,33 the ABCD² score does not reliablydiscriminate patients at low and high risk.34 With urgentassessment and appropriate treatment, the risk ofrecurrent stroke is 80% lower.35,36The longer-term risk of recurrent stroke is about 10%at 1 year, 25% at 5 years, and 40% at 10 years.37 The riskis higher among individuals with symptomaticatherosclerotic disease, vascular risk factors, or an activesource of thrombosis, or who have discontinuedantiplatelet and antihypertensive drugs. For patientswith atrial fi brillation, the risk of stroke increases withhigher CHADS2, CHA2DS2-VASc, and ABC (age,biomarkers [N-terminal fragment B-type natriureticpeptide (NT-proBNP) and cardiac troponin highsensitivity (cTn-hs)], and clinical history [prior stroke orTIA]) stroke scores.38 After haemorrhagic stroke, theannual risks of recurrent intracerebral haemorrhageand ischaemic stroke are similar, and vary from 1·3% to7·4%.29 The risk of recurrent intracerebral haemorrhageis higher after lobar intracerebral haemorrhage thanafter non-lobar haemorrhage and in patients withinadequate blood pressure control than in those inwhom blood pressure is maintained within prespecifiedlimits.39Specific treatment for acute ischaemic strokeIntravenous alteplase (rtPA), 0·9 mg/kg, administeredwithin 4·5 h of ischaemic stroke, increases the odds ofno signifi cant disability (mRS 0–1) at 3–6 months byabout a third and does not aff ect mortality, despiteincreasing the odds of symptomatic intracerebralhaemorrhage (table 3).40 The proportional benefits ofalteplase are larger with earlier treatment and theproportional risks of symptomatic intracerebralhaemorrhage with alteplase are larger with a highSEDAN score (blood sugar, early infarct signs, [hyper]dense cerebral artery sign, age, and National Institutes ofHealth Stroke Scale [NIHSS] score). 40,49Using a lower dose of alteplase (0·6 mg/kg) reduces theincidence of symptomatic intracerebral haemorrhage butdoes not lead to better functional outcome at 90 dayscompared with standard-dose alteplase.50 Functionaloutcome is also not improved by adjunctive transcranialScore forcharacteristicABCD³-I score32*Age ≥60 years 1Blood pressure ≥140/90 mm Hg 1Clinical featuresSpeech impairment without weakness 1Unilateral weakness 2Duration10–59 min 1≥60 min 2Diabetes mellitus present 1Dual TIA (index TIA plus ≥1 other TIA in preceding 7 days) 2Imaging: ipsilateral ≥50% stenosis of internal carotidartery2Imaging: acute diffusion-weighted imaginghyperintensity2Recurrence risk estimator at 90 days33†‡ClinicalHistory of TIA or stroke within the preceding month ofindex stroke1CCS aetiological stroke subtypeLarge artery atherosclerosis 1 Cardioaortic embolism0Small artery occlusion0Other causes1Undetermined causesBrain MRI within fi rst 72 hMultiple acute infarcts01Simultaneous infarcts in different circulationsMultiple infarcts of different ages11Isolated cortical infarcts1 If the stated criteria are not met, a score of 0 is assigned. NA=not applicable.TIA=transient ischaemic attack. CCS=Causative Classifi cation System for IschemicStroke. *Total range 0–13. †Total range 0–6. ‡Available at: 2: Scores to identify patients at early risk of recurrent stroke32,33Seminar644 Vol 389 February 11, 2017doppler ultrasound,51 hypothermia,52 or desmoteplase.53Tenecteplase is being compared with alteplase in severalphase 3 trials (appendix).54 Concomitant antithromboticdrugs should be avoided for the fi rst 24 h after alteplase tolimit haemorrhagic transformation of any infarcted brain.55The addition of endovascular thrombectomy withsecond-generation devices (eg, stent retrievers) toalteplase within 6 h of ischaemic stroke doubles the rateof angiographic revascularisation at 24 h and functionalindependence at 90 days (table 3), and increases thelikelihood of improving by 1 point or more on the mRSby 2·5 times, without increasing risk of symptomaticintracerebral haemorrhage or all-cause mortality.41,42 Theeffect is consistent among elderly people (>80 years) andpatients ineligible for intravenous alteplase.Improved outcomes with endovascular thrombectomyin recent trials can be attributed to improved patientselection by CT or MR angiogram to confi rm large arteryocclusion (figure), shorter time to revascularisation, andsecond-generation devices and techniques that enablehigher rates of reperfusion.41,42,56 Ongoing trials areevaluating whether patients with small ischaemic coresand substantial salvageable penumbra, as identified byCT or MR perfusion, could benefit from alteplase andendovascular thrombectomy beyond 6 h (appendix).56Implementation of endovascular therapy in clinicalpractice will require local algorithms to enable emergencymedical services to rapidly and accurately identify, triage,and transport the 10% of stroke patients suitable forendovascular therapy directly to comprehensive strokecentres where resourced, accessible, and specialised stroketeams can restore reperfusion within 90 min of arrival.57Specifi c treatment for acute haemorrhagicstrokeIntensive blood pressure reduction within 3–6 h of onsetof intracerebral haemorrhage to a systolic target of lowerthan 140 mm Hg may not be safe for all patients, nor moreeffective in reducing death and disability, compared to asystolic target of lower than 180 mm Hg (table 3).45,46For intracerebral haemorrhage not associated withantithrombotic therapy, recombinant activated factorVII (rFVIIa) decreases haematoma growth, but increasesthromboembolic events, and does not improve functionaloutcome.58 Platelet transfusion after intracerebralhaemorrhage associated with antiplatelet drugs increasesdeath and dependence at 3 months.59 For spontaneousintracerebral haemorrhage associated with vitamin Kantagonist anticoagulation, reversal of the INR to lowerthan 1·3 and reduction of the systolic blood pressure tolower than 160 mm Hg within 4 h is associated withreduced hematoma enlargement.60 Four-factorprothrombin complex concentrate seems superior to freshfrozen plasma to normalise the INR and reducehaematoma expansion.61 Management of acuteintracerebral haemorrhage associated with directinhibition of thrombin or factor Xa by direct oralanticoagulants requires immediate cessation of the directoral anticoagulants, supportive measures, andconsideration of specifi c reversal agents, such asidarucizumab for dabigatran-associated intracerebralhaemorrhage, or non-specifi c haemostatic agents, such asprothrombin complex concentrate, that show positivelaboratory results.61–65Proportion of patientswith reported functionaloutcomeOdds ratio(95% CI)Absolutedifference(%)Treatmentgroup (%)Controlgroup (%)Ischaemic strokeThrombolysis with alteplase40Good recovery (mRS 0–1) in patientswho received alteplase 0–4·5 h afterstroke34% 28% 1·37 (1·20–1·56) 7%Good recovery (mRS 0–1) in patientswho received alteplase 0–3 h afterstroke33% 23% 1·75 (1·35–2·27) 10%Good recovery (mRS 0–1) in patientswho received alteplase 3–4·5 h afterstroke35% 30% 1·26 (1·05–1·51) 6% Symptomatic ICH within 7 daysFatal ICH within 7 daysEndovascular thrombectomy41,42Second-generation devices42Good recovery (mRS 0–1)Independent (mRS 0–2)All devices41Independent (mRS 0–2)Aspirin43Independent (mRS 0–2)Death or dependence (mRS 3–6)Hemicraniectomy44Death (mRS 6)Severe disability or death (mRS 5–6)Major disability or death (mRS 4–6)Major or severe disability (mRS 4–5)Intracerebral haemorrhageIntensive BP-lowering to target systolicBP


Leave a Reply

Your email address will not be published.