Thrombolysis in Acute Ischaemic Stroke: An Update

Abstract

Stroke is a major cause of mortality and morbidity, and thrombolysis has served as a catalyst for major changes in the management of acute ischaemic stroke. Intravenous alteplase (recombinant tissue plasminogen activator) is the only approved thrombolytic agent at present indicated for acute ischaemic stoke. While the licensed time window extends to 3h from symptom onset, recent data suggest that the trial window can be extended up to 4.5 h with overall benefit. Nonetheless, 'time is brain' and every effort must be made to reduce the time delay to thrombolysis. Intracranial haemorrhage is the major complication associated with thrombolysis, and key factors increasing risk of haemorrhage include increasing age, high blood pressure, diabetes and stroke severity. Currently, there is no direct evidence to support thrombolysis in patients >80 years of age, with a few case series indicating no overt harm. Identification of viable penumbra based on computed tomography/magnetic resonance imaging may allow future extension of the time window. Adjuvant transcranial Doppler ultrasound has the potential to improve reperfusion rates. While intra-arterial thrombolysis has been in vogue for a few decades, there is no clear advantage over intravenous thrombolysis. The evidence base for thrombolysis in specific situations (e.g. dissection, pregnancy) is inadequate, and individualized decisions are needed, with a clear indication to the patient/carer about the lack of direct evidence, and the risk-benefit balance. Patient-friendly information leaflets may facilitate the process of consent for thrombolysis. This article summarizes the recent advances in thrombolysis for acute ischaemic stroke. Key questions faced by clinicians during the decision-making process are answered based on the evidence available.

Introduction

Stroke is the third largest cause of death responsible for 11% of deaths in England, and the single largest cause of adult disability. About a third of the people who have had a stroke live with long-term disability, with about 300,000 people in England having moderate to severe post-stroke disability. Stroke has a major impact on individual lives and the nation's health and economy. In England it costs nearly £7 billion each year in direct and indirect costs [National Audit Office, 2005]. Acute ischaemia underlies the majority of strokes in the world (72-87%) [Sudlow and Warlow, 1997].

Modern management of stroke includes rapid assessment and admission protocols, thrombolysis for acute ischaemic stroke (AIS), early specialist management in a stroke unit, early use of aspirin in AIS and appropriate physiological monitoring. Thrombolysis for AIS is a key intervention that can reduce disability from stroke. In 1995, the landmark National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial [NINDS Study Group, 1995] showed that patients treated with recombinant tissue plasminogen activator (rt-PA, specifically alteplase) within 3 h of the onset of AIS were at least 30% more likely to have minimal or no disability at 3 months, with no significant change in mortality. Since then, thrombolysis has become well established in most developed countries and many developing countries are in the process of developing organized services to implement thrombolysis. This article aims to provide the reader with an update on the current status of thrombolysis, and answers various key questions faced by today's clinicians when considering thrombolysis in AIS.

Ischaemic penumbra

The concept of ischaemic penumbra was first introduced by Astrup and colleagues in 1981 as ‘hypoperfused brain tissue which has capacity to recover if perfusion is improved’ [Astrup et al. 1981]. Hypoperfusion results in an area of hypoxia and loss of function, but is not enough to cause permanent damage. This area of brain tissue (the so-called ‘ischaemic penumbra’) regains its function if the occluded blood vessel is recanalized and is the key target for thrombolysis. The ischaemic penumbra is a dynamic process, existing even in the centre of the infarct for a short period of time before irreversible necrosis sets in and propagates to the neighbouring tissues over time, and may persist for more than 12 h after the onset of stroke symptoms [Heiss et al. 2001]. The Echoplanar Imaging Thrombolytic Evaluation Trial (EPITHET) showed that the use of alteplase in AIS reduced the relative growth of infarct size and significantly increased reperfusion compared with placebo [Davis et al. 2008].

Discrimination between infarct core and surrounding potentially salvageable penumbra may be useful in identifying patients who will benefit from recanalization of an occluded artery using thrombolysis, and may allow extension of the time window for thrombolysis. Computed tomography (CT) perfusion and magnetic resonance imaging (MRI) diffusion-perfusion mismatch are both attractive modalities for identification of the penumbra due to their availability. Other modalities such as single photon emission computed tomography (SPECT) and positron emission tomography (PET) can also differentiate between infarct core and salvageable brain tissue, but are less accessible.

Clinical trials so far have failed to show a clear benefit of thrombolysis based on identification of ischaemic penumbra, but further trials are ongoing and may affect a change in clinical practice in the future (as described in the following).

Thrombolytic agent

A recent meta-analysis, which included studies of urokinase, streptokinase and alteplase, confirmed the benefits of thrombolytic therapy for stroke in terms of death or dependency, but the only agent which was associated with clear benefit was alteplase (Figure 1). Thus, intravenous (IV) alteplase is the only licensed thrombolytic agent for AIS, the recommended dose being 0.9 mg alteplase/kg body weight (maximum of 90 mg), 10% of the total dose administered as an initial IV bolus and the rest as an IV infusion over 60 min (based on the NINDS study criteria [NINDS Study Group, 1995]). The results are influenced by differences in trial design, dose of thrombolytic agent used, timing of thrombolysis and definition of haemorrhage.

Figure 1.

Cochrane review of death and dependency in patients randomized within 3 h of the stroke, at the end of follow up. (Reproduced with permission from Wardlaw et al. [2009]).

In the United Kingdom, the licence for alteplase is limited to within 3 h of the onset of stroke in highly selected patients, by a physician trained and experienced in the management of stroke, and in centres with appropriate facilities for aftercare [National Institute of Health and Clinical Excellence, 2007]. The guidelines from the American Stroke Association and European Stroke Organisation (ESO) stress the importance of thrombolysis as soon as possible after symptom onset, and have adopted the extended time window of 4.5 h. The National Stroke Foundation of Australia has very similar recommendations for use of alteplase in Australia and New Zealand, and a recent update in 2010 has increased the time window for thrombolysis up to 4.5 h [National Stroke Foundation, 2010].

Tissue plasminogen activator (t-PA), an endogenous, human serine protease found in the intravascular space, the blood-brain interface and within the brain parenchyma, plays a central role in the homeostasis of the coagulation cascade. It cleaves plasminogen to plasmin which dissolves fibrin-based clots. Unfortunately, it also has a destructive effect on extracellular matrix and endothelial basal lamina leading to compromise of the blood-brain barrier and haemorrhage [Micieli et al. 2009]. If used within the licence, alteplase is associated with a significant net reduction in death and disability after ischaemic stroke despite a small but significant increase in the risk of intracerebral haemorrhage (ICH). The use of alteplase is limited to fewer than 5% of AIS patients [Molina and Saver, 2005] because of the limited therapeutic window, insufficient awareness among public and professionals and undue worries about the risk of ICH [Reeves et al. 2005].

Intracranial haemorrhage postthrombolysis

ICH is the major complication of thrombolysis in AIS. It can be classified using three features: associated symptoms (symptomatic with a ≥4-point deterioration in the NIH Stroke Scale, or asymptomatic); duration from thrombolysis (early, i.e. <24h; or late, i.e. >24h from thrombolysis); and anatomical-radiological (see Table 1).

Table 1.

Categorization of intracranial haemorrhage postthrombolysis [Berger, 2001].

Type	Abbreviation	Definition
Haemorrhagic infarction (HI)	HI1	Small petechiae along the margins of the infarct
	HI2	More confluent petechiae within the infarct, but without space-occupying effect
Parenchymal haematoma (PH)	PH1	Haematoma in ≤30% of the infarcted area with some space-occupying effect
	PH2	Haematoma in >30% of the infarcted area with substantial space-occupying effect OR any haemorrhagic lesion outside the infarcted area

In cases where there is more than one haemorrhagic lesion on the CT scan, the worst possible category to be assumed.

In the Cochrane Review of trials using rt-PA, there were 60 [95% confidence interval (CI 50-70)] additional symptomatic intracranial haemorrhages per 1000 patients treated [odds ratio (OR) 3.28, 95% CI 2.48-4.33, p < 0.00001] [Wardlaw et al. 2009]. ICH represents a complex and heterogeneous group of phenomena and involves multiple demographic, clinical, biological and haemodynamic parameters. Table 2 summarizes the risk factors for ICH [Micieli et al. 2009]. The influence of prior use of antiplatelets on the risk of ICH postthrombolysis remains debatable. Bravo and colleagues reported a nonsignificant rise in the incidence of symptomatic ICH [Bravo et al. 2008], and while Dorado and colleagues reported a significant increase in the risk of parenchymal haemorrhage with antiplatelets, there was no significant increase in symptomatic ICH [Dorado et al. 2009]. Analysis of the from first and second European Co-operative Acute Stroke Studies (ECASS 1 and ECASS 2) suggested that only PH-2 types of parenchymal haematomas (homogenous haematoma with mass effect occupying 30% of the ischaemic lesion volume) were associated independently with clinical deterioration and poor prognosis [Berger et al. 2001].

Table 2.

Risk factors for haemorrhagic transformation in acute ischemic stroke following thrombolysis [Micieli et al. 2009].

Clinical factor	Biochemical marker	Radiological factor
Longer door-to-needle time	Elevated blood glucose	Early ischaemic change on CT
Stroke severity	Elevated red cell count	Large volume infarct, oedema, or mass effect on the baseline CT
Advancing age	Elevated matrix metalloproteinase/plasma cellular fibronectin levels [Castellanos et al. 2007]	Presence of lacunae
History of diabetes	Elevated calcium-binding proteins	Reduced blood volume on PWI
Elevated blood pressure	Early fibrinogen degradation coagulopathy	Breakdown of blood-brain barrier
History of cardiac failure	—	Presence of micro bleeds on MRI; leukoariosis on MRI

CT, computed tomography; MRI, magnetic resonance imaging; PWI, perfusion-weighted imaging.

The absence of poor predictive factors implies a higher likelihood of improved outcome. A strong correlation has been reported between recanalization and good outcome [Rha and Saver, 2006], and some studies suggest good outcome in the presence of collateral circulation [Kucinski et al. 2003; DeWitte et al. 1998].

Extending the time window

One of the major limitations to the use of thrombolysis in AIS has been its limited therapeutic time window. Alteplase is licensed for use within 3 h of the onset of stroke symptoms in the majority of countries, based on the criteria used in the NINDS trial in 1995. Since then, two studies published in 2008 have indicated the benefit of alteplase up to 4.5 h after the onset of stroke symptoms.

The Third European Co-operative Acute Stroke Study (ECASS III) was a randomized controlled trial of IV alteplase versus placebo administered 3-4.5 h after the onset of stroke [Hacke et al. 2008]. This trial found that patients treated with alteplase were significantly more likely to have a favourable outcome (mRS score of 0 or 1) than those treated with placebo (52.4% versus 45.2%, respectively; OR, 1.34; 95% CI, 1.02-1.76; p = 0.04). While the rate of symptomatic ICH (sICH) was significantly greater in patients treated with alteplase (2.4% versus 0.2%, p = 0.008), it was less than the rate seen in previous trials of patients treated with alteplase within 3 h of symptom onset, including the NINDS trial (6.4%). The all-cause mortality rate at 90 days was similar between those treated with alteplase and those treated with placebo (7.7% versus 8.4%, p = 0.68). The ECASS III trial excluded patients with very severe stroke as assessed by clinical or imaging criteria, which may explain the lower sICH rate.

The Safe Implementation of Treatments in Stroke-International Stroke Thrombolysis Registry (SITS-ISTR) was a prospective, Internet-based audit of more than 700 clinical centres in Europe that compared patients treated with IV alteplase 0.9 mg/kg 3-4.5 h after symptom onset with patients treated within 3 h of symptom onset [Wahlgren et al. 2008]. This observational study found that after 3 months of treatment, patients receiving alteplase in both groups experienced similar rates of functional independence (modified Rankin score mRS of 0-2) (58% of patients treated 3-4.5 h after symptom onset versus 56.3% of those treated within 3 h, p = 0.42) and excellent recovery (mRS score of 0 or 1) (40.5% versus 39.9%, p = 0.79) and the rates of symptomatic ICH (2.2% versus 1.6%, p = 0.24) and mortality (12.7% versus 12.2%, p = 0.72) were similar.

This trial excluded patients over age 80 years and those with an NIHSS score of > 25.

These two studies reinforce the safety and efficacy of alteplase in AIS patients outside the licensed 3 h window, up to 4.5 h after the onset of symptoms. The National Institute of Health and Clinical Excellence (NICE) in the UK is currently reviewing the evidence and will issue updated guidance in the near future regarding the time window for thrombolysis. The ESO has updated its guidelines in January 2009 to increase the time limit for thrombolysis up to 4.5 h [ESO Guidelines, 2008]. The American Heart Association and American Stroke Association has issued recommendations to extend the time window for thrombolysis in AIS to 4.5 h but with additional exclusion criteria including: age of > 80 years; current use of oral anticoagulants (regardless of International Normalized Ratio [INR]); baseline National Institutes of Health Stroke Scale score of >20; and medical history of both stroke and diabetes.

Attempts have been made to increase the window of eligibility for thrombolysis by using specialized CT or MRI perfusion imaging to identify viable penumbra. While desmoteplase, derived from vampire bat saliva, showed safety and efficacy in pilot studies (DIAS1 [Hacke et al. 2005] and DEDAS [Furlan et al. 2006]), over a longer treatment window (3-9 h) the subsequent Phase III DIAS-2 study did not show any benefit with similar 90-day clinical outcomes in either the high- or low-dose group compared with placebo, and increased mortality in the high-dose group (21% versus 5% low dose; 6% placebo) [Hacke et al. 2009]. However, only 3 of 14 deaths in the high-dose group were felt to be related to trial medication. CT- or MR-based identification of the penumbra to guide thrombolysis with tenecteplase in a small study showed greater reperfusion and recanalization versus alteplase, and more neurological improvement [Parsons et al. 2009]. Currently, there is no evidence to support the routine use of thrombolysis beyond 4.5 h, which may be associated with net harm, as indicated by a pooled analysis of all thrombolysis trials [Lees et al. 2010] (Figure 2).

Figure 2.

Relation of onset to treatment delay with treatment effect. (Reproduced with permission from Lees et al. [2010]).

Age limit for thrombolysis

So far, alteplase is only licensed to be used in patients under the age of 80 years. This is because older people were excluded or only partially represented in clinical trials. A systematic review [Engelter et al. 2006] of all cohort studies where patients 80 years or older underwent thrombolysis showed significant differences in baseline characteristics to the disadvantage of older patients compared with younger patients. Older patients had higher 3-month mortality and were less likely to regain a ‘favourable outcome’. However, this did not appear to be due to increased risk of symptomatic ICH in older patients. While pragmatic use of thrombolysis is routinely considered based on biological age (rather than a strict chronological age cutoff), further randomized evidence is required to confirm the benefit of thrombolysis in AIS in patients over the age of 80 years, which is being explored in the third International Stroke Trial (IST-3).

Imaging before thrombolysis

Brain imaging to exclude ICH is a prerequisite for administration of IV alteplase in AIS. CT is preferred because it is rapid and widely available. In cases of diagnostic doubt, hyperacute changes of infarction may support the clinical suspicion. Another advance is the Alberta Stroke Programme Early CT Scale (ASPECTS), an objective method of quantifying early ischaemic changes on the baseline CT scan, which also has prognostic value (ASPECTS score >7 is associated with reduced mortality and increased benefit to patients who receive thrombolysis [Barber et al. 2000]. One analysis showed that there was no evidence of treatment effect modification by the baseline ASPECTS value in the NINDS rt-PA Stroke Study [Demchuk et al. 2005].

Therefore, exclusion of patients within 3h of stroke symptoms based on early ischaemic change/ASPECTS score is not supported by the evidence.

MRI is potentially better than CT because in addition to excluding ICH (gradient echo echo planar imaging [EPI] sequence), it can also identify acute infarction within 30 min of onset (diffusion weighted imaging), which can be extremely useful where there is clinical uncertainty. It may also identify patients who are at increased risk of ICH by identifying extensive brain infarction. However, MRI is limited by its availability and longer scan time. One centre showed a median delay of 20 min in door-to-needle time compared to CT-based therapy [Kang et al. 2005]. Others have shown no significant difference [Solling et al. 2009]. Another area where MRI can be useful is in patients who wake up with stroke symptoms: 25% of strokes occur during sleep. These patients are usually excluded from thrombolysis because of an uncertain time of onset. MRI can help select patients who will benefit from thrombolysis by identifying salvageable brain tissue. A small study using MRI scans to select patients for thrombolysis in patients who wake up with stroke symptoms showed no significant benefit of thrombolysis and no overt harm [Breuer et al. 2010].

While the concept of penumbral imaging is attractive, there is no clear evidence to support such a strategy to aid decisions about thrombol-ysis at present.

Ultrasound and thrombolysis

Transcranial Doppler (TCD) ultrasound can be used in conjunction with rt-PA to improve reperfusion rates in AIS. Other applications of TCD ultrasound include the detection and localization of arterial occlusion, and monitoring for reperfusion. Ultrasound works by delivering a mechanical force at the stagnant flow and clot interfaces. Mechanical agitation can expose shallow layers of thrombus to circulating rt-PA, thus enhancing fibrinolysis. Ultrasound can be high frequency or low frequency. Low-frequency ultrasound has more mechanical force and is more effective in enhancing the effect of rt-PA but it is associated with significantly higher rates of ICH and cannot be used for imaging the vasculature, which limits its clinical utility. High-frequency ultrasound (1.0-2.2 MHz) includes TCD, transcranial colour Doppler and microspheres. Microspheres are micro bubbles which were initially used as ultrasound contrast but now they can also be used for targeted delivery of drugs and to produce localized ultrasound echoes. High-frequency ultrasound is commonly used for imaging and it can also enhance thrombolysis. A recent meta-analysis showed that use of high-frequency ultrasound was associated with a higher likelihood of complete recanalization (in the short term) when compared with rt-PA alone without any significant increase in the frequency of symptomatic ICH [Tsivgoulis, 2010].

Ultrasound techniques are being refined especially the use of microspheres. At present, the use of TCD ultrasound is limited by the cost of the equipment and the need for training to reduce operator variability.

Intra-arterial thrombolysis

Intra-arterial thrombolysis (IAT) is an unlicensed treatment option for AIS caused by large-vessel occlusion, which has been employed by interventional neuroradiologists for decades. Cerebral angiography is performed to localize the occluded artery, and then a percutaneous catheter is inserted and guided into the thrombus for local delivery of the thrombolytic agent.

Intra-arterial (IA) prourokinase (proUK) for AIS (PROACT II) was the largest randomized trial of IA thrombolysis with 121 patients in the treatment group. Comparison of IAT plus heparin with heparin only showed that treatment with IA recombinant proUK within 6 h of the onset of acute ischemic stroke caused by MCA occlusion significantly improved clinical outcome at 90 days (mRS = 2 in 40% of treatment group versus 25% of control group, p = 0.04) but with an increased risk of ICH (10% versus 2%, p = 0.06) [Furlan et al. 1999]. Questions remain over the safety and efficacy of IAT compared with intravenous rt-PA. Most reports of safety with IAT come from case series [Ciccone et al. 2007].

IAT is attractive, as: (i) it allows for a reduced dose of thrombolytic to be delivered directly to the clot with potentially lower risk of systemic haemorrhagic complications; (ii) it increases the rate of recanalization (however, it must be noted that recanalization does not always equate to improved clinical outcome); and (iii) it can be used with mechanical adjuvant therapy, such as clot retrieval.

The current evidence base is inadequate to support routine clinical application of IAT. In addition, practical issues with regards to time delays and need for specialist expertise limits availability. Ongoing trials (e.g. SYNTHESIS) aim to clarify this issue, but unfortunately the recruitment is very slow. Until there is clearer evidence of benefit, IAT cannot be recommended as a routine treatment and should ideally be administered within the constraints of a structured research trial.

Thrombolysis in practice

Setting up a thrombolysis service

In AIS ‘time is brain’ and for thrombolysis in AIS ‘sooner is better’. There is evidence of benefit for thrombolysis up to 4.5 h after the onset of stroke symptoms, but patients who receive thrombolysis within 90 min receive twice as much benefit as patients who receive thrombolysis between 3 and 4.5 h [Hacke et al. 2004]. Thus, developing a thrombolysis service is based on the key principle of delivering the thrombolysis as soon as possible and certainly within 4.5 h. All professionals involved in the patient pathway for thrombolysis should be involved when setting up a thrombolysis service, including primary care representatives, ambulance dispatchers, paramedics, emergency department staff, and the radiology department.

Current European practice

In Europe, the Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS-MOST) was a mandatory requirement following the licensing of alteplase for ischaemic stroke. This confirmed the safety profile of alteplase in routine clinical practice in 6483 participants, who received thrombolysis within 3 h of stroke onset [Wahlgren et al. 2008]. There was a trend towards reduced symptomatic ICH. Also mortality was significantly less than the active treatment groups in placebo-controlled studies of alteplase (11.3% versus 17.3%).

In the UK, the National Stroke Strategy has resulted in a significant increase in the number of hospitals offering thrombolysis (57%) and also those offering a 24-h thrombolysis service (33%) [Royal College of Physicians of London (Intercollegiate Stroke Working Party), 2010]. In the 2008 clinical audit, about 15% were eligible for thrombolysis, but only 1% received this therapy [Royal College of Physicians of London (Intercollegiate Stroke Working Party), 2009].

Thrombolysis is best delivered in an integrated stroke unit which provides acute stroke care coupled with early rehabilitation. For urban populations, thrombolysis can be centralized at one site. Keeping in mind future developments in the field of thrombolysis, proximity of interventional neuroradiology and advanced vascular imaging should be considered. Stroke physicians might have to reorganize their job plans because of commitments to the thrombolysis service and out of hours work. Where required, stroke physicians should be trained to interpret CT scans in the absence of radiologists. Radiology services might have to be redesigned to accommodate the requirement of rapid brain imaging for potential thrombolysis patients.

The ‘number needed to see’ to thrombolyse one patient is estimated at 3–4 and a significant number of nonstroke diagnoses are encountered on the acute stroke take [Weir and Buchan, 2005]. There should be systems in place to transfer nonstroke patients to general medical wards. Assessment of patients for thrombolysis in the hospital is not a single process but a number of simultaneous parallel processes. Thus, an on-call multidisciplinary team of specialists is vital to the smooth running of a predefined pathway for thrombolysis. Most delays occur before the patient arrives in hospital, and establishing a triage protocol between emergency transfer services and stroke unit expedites the delivery of patients to the hospital and almost doubles the rate of thrombolysis [Gladstone et al. 2009]. Stroke physicians should encourage training of ambulance staff, paramedics, primary care physicians and their staff to rapidly recognize stroke symptoms and act on them. For remote and rural populations telemedicine links to a centre with experienced clinicians can support local services and improve thrombolysis rates [Audebert et al. 2006]. A minimum level of facilities is required even with tele-medicine including staffing, monitoring and imaging. A small percentage of patients might still have to be transferred.

Where thrombolysis services are implemented, it is essential to have robust data collection processes and audit to ensure evidence-based clinical practice and identification of barriers to successful thrombolysis. A recent systematic review concluded that regional collaborations achieve higher rates of thrombolysis than local services working in isolation [Price et al. 2009].

Ethical issues

Ethical issues will be different in different parts of the world depending upon the social and cultural background of that area. Consent is the major ethical issue surrounding thrombolysis in AIS. To give consent, a person should be able to understand information and communicate his or her decision. Stroke can affect a patient's understanding of the information as well as their communication, rendering them incapable of giving informed consent. In such situations, the decision for thrombolysis should be made in the best interests of the patient. The risks and benefits of thrombolysis should be considered in partnership with the next of kin, proxy or carer, in an effort to ascertain the patient's prior wishes.

Another ethical issue is to provide enough information to the patients and relatives in a way that they understand and help them make informed decisions. This has to be done quickly in the setting of an acute stroke, in the emergency department in order to save time before thrombolysis. Various tools have been designed to expedite the process of informed decision making for thrombolysis. These tools offer information to patients and relatives in easily understandable formats. Figure 3 shows one such tool. Another tool can be found at www.sem-bc.com/cvatoolkit.

Figure 3.

Patient information tool. (Acquired from the UCLA Stroke Center).

Thrombolysis in special circumstances

Carotid artery dissection

Spontaneous internal carotid artery dissection (SICA) is a well-recognized cause of ischaemic stroke in young patients. A major concern about thrombolysis in SICA is an increase in size of the vessel wall haematoma. More than 50 cases have been reported where IV alteplase has been used for SICA. SICA has not been excluded from clinical trials of thrombolysis in AIS. It appears safe and effective to use IV alteplase in SICA. The role of IAT in SICA is not known.

Basilar artery occlusion

Trials of thrombolysis in AIS have focused on anterior circulation strokes, but 20% of ischaemic strokes affect the posterior circulation. Basilar artery occlusion is associated with devastating outcomes. Endovascular thrombolysis is generally preferred in basilar artery occlusions. A meta-analysis comparing IV (76 patients) versus IA (344 patients) thrombolytic treatment found that survival and outcome were roughly equal; a total of 24% of patients treated with IAT and 22% treated with intravenous thrombolysis achieved functional independence [Lindsberg and Mattle, 2006]. The ESO Guidelines recommend IAT for basilar artery occlusion in selected patients, although IV thrombolysis is deemed an acceptable alternative even after 3 h [European Stroke Organisation, 2008].

Seizure

Seizure at the onset of stroke symptoms is usually a contra-indication to thrombolysis because of the difficulty in differentiating between postictal paresis and stroke. Advanced imaging techniques such as MRI, CT perfusion scan, PET and SPECT have made it possible to diagnose ischaemic stroke with confidence in these patients. Patients with seizure at onset may be considered for thrombolysis in select situations, where infarction is clearly identified.

Cerebral venous thrombosis

Heparin is generally considered the mainstay of treatment in cerebral venous thrombosis (CVT). Few case series have demonstrated that endovascular thrombolysis could be an option in severe cases of CVT and in patients who do not respond to conventional medical management [Yue et al. 2010].

Pregnancy

Pregnancy and the puerperium are hypercoagulable states, both increasing the risk of thromboembolic complications including stroke and CVT. Since pre-eclampsia can mimic stroke, diagnosis in the pregnant woman can be difficult. MRI can be used for early identification of infarcts, without radiation exposure. In one case series of stroke in pregnant women, the rates of intra-arterial occlusion (7/20) and venous occlusion (6/20) were similar [DeKoninck et al. 2008].

The evidence base for thrombolysis in pregnancy and the puerperium is very limited and based solely on case reports. The major concern is potential deleterious effects on the placenta, causing premature labour, placental abruption and foetal death. However, transplacental transfer does not occur and no teratogenic effects have been reported in animal studies [Leonhardt et al. 2006]. Anecdotal cases of thrombolysis with good results have been reported [Murugappan et al. 2006]. The intra-arterial route has been the main route of thrombolysis, with very limited experience of IV alteplase in pregnant women. The benefits and risks of thrombolysis in pregnant should be balanced carefully and discussed fully with the patient or proxy.

Menstruation

Active bleeding is a contraindication to thrombolysis in AIS; however, a review of the limited literature has demonstrated that alteplase could be used relatively safely during menstruation [Wein et al. 2002]. It may cause excessive bleeding especially on the first day of menstruation. A blood transfusion may become necessary.

Unanswered questions

Is thrombolysis effective after the first 4.5 h from stroke onset?

Pooled analysis of existing trials suggests that there may be harm from thrombolysis beyond the 4.5 h window; however, this is being explored in IST-3 (http://www.dcn.ed.ac.uk/ist3), which aims to determine whether administration of rt-PA within 6 h of ischaemic stroke increases the proportion of independent survivors at 6 months.

Is thrombolysis effective in patients older than a chronological age of 80 years?

While age has been identified as an independent risk factor for poor outcome and haemorrhagic complications, a few case series have suggested that older patients benefit similar to younger patients from thrombolysis. The IST-3 study also incorporates older patients in a randomized controlled trial to provide robust evidence in this age group. In addition, the Thrombolysis in Elderly Stroke Patients in Italy (TESPI, www.strokecenter.org) study is a randomized open-label study looking at the efficacy and safety of alteplase in patients more than 80 years old.

Is thrombolysis effective beyond the first 3 h, in patients where a viable penumbra is identified using perfusion imaging?

The theoretical prospect of targeted thrombolysis to individuals with a viable penumbra is attractive. The Desmoteplase in Acute Ischaemic Stroke studies (DIAS-3 and DIAS-4) aim to determine whether desmoteplase is effective and safe in the treatment of patients with AIS and identified penumbra (by MRI perfusion imaging), when given within 3–9 h from onset of stroke symptoms.

The Imaging-based Thrombolysis Trial in Acute Ischemic Stroke III (ITAISIII) is recruiting selected acute ischaemic stroke patients with mismatch on CTP/computed tomography angiography-source images (CTA-SI) and without large vessel occlusion or significant stenosis to establish the efficacy (clinical outcomes and cerebral perfusion) and safety of IV thrombolysis in the 3–9 h time window.

Is concomitant administration of antiplatelet therapy safe?

The primary objective of the Antiplatelet Therapy in Combination with rt-PA Thrombolysis in Ischemic Stroke (ARTIS) study is to investigate whether acute antiplatelet therapy in addition to standard rt-PA thrombolysis reduces poor outcome (modified Rankin scale 3-6) 3 months after an AIS.

Is another treatment strategy better than IV thrombolysis alone; for example, combined IAT and/or mechanical devices, combined IV/IA thrombolysis?

The Intra-Arterial Versus Systemic Thrombolysis for Acute Ischemic Stroke (SYNTHESIS EXP) study aims to determine whether locoregional IA t-PA and/or mechanical devices, as compared with systemic IV t-PA within 3 h of ischaemic stroke, increases the proportion of independent survivors at 3 months. The Interventional Management of Stroke (IMS) III Trial aims to determine whether a combined IV/IA approach to recanalization is superior to standard IV rt-PA alone, when initiated within 3 h of stroke onset.

Is an alternative model of thrombolysis delivery possible?

The ‘Mobile Stroke-Unit’ for Reduction of the Response Time in Ischemic Stroke (MSU) study is investigating whether a ‘Mobile Stroke Unit’ (a rescue car with an integrated CT scanner, necessary for essential diagnostics) contributes to improved stroke management by reducing delay to a therapeutic decision.

The future of thrombolysis in acute ischaemic stroke

With the emergence of new evidence there will be expansion of thrombolysis treatment to a wider patient population including those presenting late and those over the age of 80 years. This will also be supported by advancements in imaging techniques. The current thrombolytic agent might be superseded by safer and more effective agents with wider applications. Currently, thrombolysis in AIS is a specialist intervention provided at specialist centres by specialist physicians in contrast to thrombolysis in acute myocardial infarction (MI) which can even be initiated at the patient's own home by paramedics. In acute myocardial infarction, we have seen a change from thrombolysis to percutaneous coronary intervention (PCI) as the first-line treatment. Similarly, IV thrombolysis in AIS may be replaced by more advanced and effective treatment options; for example, IAT and mechanical clot retrieval. Until further evidence is available, intravenous thrombolysis remains the cornerstone of management of AIS.

Conclusion

Thrombolysis is at the forefront of modern management of AIS, with good evidence of its efficacy within 4.5 h of symptom onset. Intravenous alteplase is the only approved thrombolytic agent at present. ICH is the major complication associated with thrombolysis in AIS, and key factors increasing risk of haemorrhage include increasing age, high blood pressure, diabetes and stroke severity. The use of TCD ultrasound to facilitate thrombolysis may improve reperfusion rates. IAT is under investigation but cannot be recommended at this stage for routine use. The role of adjuvant TCD ultrasound and intra-arterial mode of thrombolysis in routine clinical practice is not clear. The evidence base for thrombolysis in specific situations (e.g. dissection, pregnancy) is inadequate to assist clinical decisions, and individualized decisions are needed with open discussion with the patient/carer about the lack of evidence, risks and benefits. Patient-friendly information leaflets may facilitate the process of consent before thrombolysis. However, there remain many unanswered dilemmas for the clinician including: thrombolysis in the older patient; in the patient waking up with stroke symptoms; thrombolysis more than 4.5 h from symptom onset; the best imaging modality prethrombolysis; and the optimal service configuration. These are the subject of ongoing clinical trials, and the ever-increasing body of evidence will help to clarify these uncertainties.