The Exact Science of Stroke Thrombolysis and the Quiet Art of Patient Selection

Joyce S. Balami; Gina Hadley; Brad A. Sutherland; Hasneen Karbalai; Alastair M. Buchan


Brain. 2013;136(12):3528-3553. 

In This Article

Treatment of Acute Ischaemic Stroke

Pharmacological Thrombolysis

Although timely restoration of blood flow is the primary therapeutic goal for patients with acute ischaemic stroke, there is need for careful selection of appropriate patients for thrombolytic therapy. In addition to the various thrombolysis protocols and guidelines, certain key metrics (Box 1) known to be predictive of outcome should help guide clinical judgement, leading to correct diagnosis, appropriate patient selection, and implementation of maximally beneficial therapeutic interventions.

Acute Ischaemic Stroke Thrombolysis Within the Standard Guidelines

Intravenous alteplase (the active ingredient being rt-PA) is the only Food and Drug Administration (FDA)-approved thrombolytic agent for the treatment of acute ischaemic stroke. Rt-PA is a fibrin-selective thrombolytic agent that breaks fibrin down into fibrin degradation products, ultimately dissolving the thrombus and resulting in recanalization of the occluded artery (Balami et al., 2013a). Criteria for using rt-PA for thrombolysis are outlined in Table 2. Evidence for the benefit of rt-PA for acute ischaemic stroke treatment in suitable patients is derived from trials including the NINDS (1995) and European Collaborative Acute Stroke Study (ECASS, Hacke et al., 2008) (Fig. 2).

Figure 2.

Evaluating the importance of time: the effect of rtPA thrombolysis on patient outcome by the end of follow-up. Data were taken from two previously published meta-analyses (ATLANTIS, ECASS and NINDS Investigators, 2004; Wardlaw et al., 2012). Alive and independent, the primary outcome measure, was defined as 0–2 using the modified Rankin Scale or equivalent. Time to treatment was divided between 0–180 min and 181–360 min. A comparison is made between all published trials conducted up to 2004, and all published trials conducted up to 2012. For illustration, IST-3 results were also shown to demonstrate the effects of rtPA in this most recent trial. Unadjusted odds ratios (OR) and their 95% confidence intervals were calculated to compare the effects of rtPA administration with placebo.

Based on results from ECASS III, in 2009 both the AHA/ASA and ESO guidelines extended the time window for treatment of eligible acute ischaemic stroke patients with intravenous rt-PA from 3 h to 4.5 h after the onset of stroke symptoms (ESO, 2008; Del Zoppo et al., 2009; ESO, 2009), reaffirmed in the recent AHA/ASA guidelines (Jauch et al., 2013). Although the FDA has not yet endorsed use of alteplase beyond 3 h, it has been approved for use up to 4.5 h by the European Medicines Agency. The ECASS III trial of patients given intravenous thrombolysis 3–4.5 h after acute ischaemic stroke onset demonstrated a reduced risk of death or dependency at 3 months (number needed to treat = 14, P = 0.04) despite an increase in any intracerebral haemorrhage and symptomatic intracranial haemorrhage (using NINDS definition) (number needed to harm = 10, P = 0.001; number needed to harm = 22, P = 0.006, respectively) (Hacke et al., 2008). The number needed to treat increases from two during the first 90 min (0–1.5 h), to seven during the second 90 min (1.5–3 h), to 14 during the third 90 min (3–4.5 h) after acute ischaemic stroke onset (Hacke et al., 2008). The ECASS III findings are supported by results from the Safe Implementation of Treatments in Stroke-International Stroke Thrombolysis Registry (SITS-ISTR) registry, comparing 664 acute ischaemic stroke patients given intravenous thrombolysis between 3 and 4.5 h with 11 865 patients treated within 3 h (Wahlgren et al., 2008).

A pooled analysis of eight trials showed that intravenous thrombolysis up to 4.5 h from stroke onset can enhance the chance of a favourable outcome. Nonetheless, the analysis showed that the greatest benefit was found in earlier treatment, and that the net benefit diminishes dramatically beyond 4.5 h (Lees et al., 2010). A recently updated systematic review and meta-analysis of 12 trials, including the IST-3, showed that treatment within 6 h of acute ischaemic stroke onset increases patients' chances of being alive and independent by final study follow up [46.3% versus 42.1%, odds ratio (OR) 1.17, 95% confidence interval (CI) 1.06–1.29, P = 0.001], although deaths within 7 days were more likely with thrombolysis (8.9% versus 6.4%, OR 1.44, CI 1.18–1.176, P = 0.0003) (Wardlaw et al., 2012). However, the greatest benefit is still within the 3 h time window (Fig. 2). Further meta-analysis to determine if clinical or imaging scores can serve as independent predictors of outcome is warranted.

Extension of the thrombolysis time window to 4.5 h from acute ischaemic stroke onset creates a new eligible patient population, though it may be marginal in size compared with those presenting within 3 h. A survey of 11 262 patients with acute ischaemic stroke found 14% of <3 h patients, but only 2% of 3–4.5 h patients eligible for thrombolysis (Mihout et al., 2012). There also remain concerns regarding bleeding risk and low recanalization rates in patients treated in the later time window (Rha and Saver, 2007; Hacke et al., 2008; Lees et al., 2010)—only ~50% of patients achieve recanalization (Rha and Saver, 2007), and even if recanalization is achieved, re-occlusion with neurological deterioration has been reported to occur in >30% (Grotta et al., 2001; Alexandrov and Grotta; 2002, Ribo et al., 2006).

Acute Ischaemic Stroke Thrombolysis Outside the Standard Guidelines

Evidence for the thrombolysis guidelines comes primarily from the pivotal clinical trials—NINDS, ECASS and ATLANTIS. These trials had several exclusion criteria designed to limit complications associated with intravenous thrombolysis and increase the chance of a positive outcome. Common exclusions include patients presenting outside the recommended time windows and patients with mild or rapidly improving symptoms (Table 2). Limited alternatives to intravenous thrombolysis have persuaded physicians to sometimes offer treatment to acute ischaemic stroke patients with relative contraindications depending on the risk-to-benefit ratio.

Delayed Thrombolysis up to 6 Hours. Although benefits of thrombolysis gradually diminish with increasing acute ischaemic stroke onset-to-treatment time, useful clinical benefit remains possible. Multimodal imaging techniques (diffusion/perfusion MRI, perfusion CT) may help select patients with salvageable tissue that might benefit from treatment even after 3 or 4.5 h of stroke onset.

The ECASS III trial, which confirmed a clinical benefit within 4.5 h (OR 1.34, CI 1.02–1.76, P = 0.04), showed a wider 95% CI (0.9 to 1.5) at 4.5–6 h in the meta-analysis, suggesting the possibility of benefit from intravenous thrombolysis even beyond 4.5 h (Hacke et al., 2008). A meta-analysis of five trials of patients thrombolysed beyond 3 h showed how delayed thrombolysis in patients selected according to mismatch imaging is associated with increased reperfusion/recanalization (Mishra et al., 2010b), but not to improved clinical outcome, and there was a significant risk of symptomatic intracranial haemorrhage and possibly increased mortality. Recently published IST-3 data show that the primary trial hypothesis, that intravenous thrombolysis given within 6 h of acute ischaemic stroke onset increases the proportion of patients alive and independent at 6 months, is not supported on its own, but integrated with other studies in a meta-analysis, a benefit out to 6 h is suggested. An increase in the percentage of thrombolysed patients alive and independent (Oxford Handicap Score 0–2) at 6 months was found to be non-significant (37% versus 35%, OR 1.13, CI 0.95–1.35, P = 0.181) (Sandercock et al., 2012). Clinical benefit to 4.5 h has been established, but benefits out to 6 h are equivocal, requiring careful patient selection.

Thrombolysis for Mild Acute Ischaemic Stroke/Rapidly Improving Symptoms. Twenty to forty-six per cent of patients with acute ischaemic stroke with mild or rapidly improving stroke symptoms are excluded from intravenous thrombolysis despite presenting within the recommended time window (Barber et al., 2001; Katzan et al., 2004; Kleindorfer et al., 2004; Cocho et al., 2005; Schwamm et al., 2005; Majersik et al., 2007; Schumacher et al., 2007; George et al., 2009; Smith et al., 2011). There are several reasons for this exclusion: first is the presumption that prognosis will be good regardless of whether they are given intravenous thrombolysis, more so as potential intravenous thrombolysis risks may outweigh benefits (Adams et al., 2008; Adams, 2009); second is the fact that the NINDS trials excluded acute ischaemic stroke patients with minor symptoms, no measurable deficit on the NIHSS score, symptoms of isolated motor or sensory stroke, isolated ataxia, isolated dysarthria, isolated facial weakness, or rapidly improving deficit (Magid et al., 2005); third, there was no improvement with intravenous thrombolysis among patients with an NIHSS ≤5 in the NINDS trial (Gladstone et al., 2002); and fourth, many protocols exclude patients who have mild deficits. Consequently, stroke patients with low NIHSS scores are not usually treated in clinical practice, but their outcome is unpredictable, and several may have poor outcomes, as reported in several studies (Nedeltchev et al., 2007; Coutts et al., 2009; Smith et al., 2011). Non-hospital discharge, or disability or death (modified Rankin Score of 2–6), has been reported in 25–64% of acute ischaemic stroke patients with mild stroke or rapidly improving stroke symptoms at hospital discharge (Barber et al., 2001; Smith et al., 2005a; Gonzales et al., 2006; Nedeltchev et al., 2007; Khatri et al., 2008; Coutts et al., 2009), and in 29–58% at 3 months (Khatri et al., 2008; Fischer et al., 2010). In the AHA 'Get With The Guidelines' (GWTG) registry of 93 517 patients who presented within 2 h of symptom onset, 29 200 (31.2%) were not eligible for intravenous thrombolysis solely because of mild or improving symptoms, of which 1% died, 28.3% were not discharged home, and 28.5% were functionally dependent at discharge (Smith et al., 2011). Occasionally, rapid recovery from stroke symptoms is followed by clinical deterioration (Johnston and Easton, 2003; Johnston et al., 2003; Khatri et al., 2012). Persisting large-vessel occlusion, proximal arterial occlusions, and intra- or extracranial carotid stenosis or occlusion substantially increase the risk of early deterioration and are highly predictive of poor functional outcome (Rajajee et al., 2006; Nedeltchev et al., 2007; Coutts et al., 2009).

Post hoc analyis of data from the Canadian Alteplase for Stroke Effectiveness Study (CASES) registry suggest that intravenous thrombolysis in patients with baseline NIHSS ≤5 might be reasonably safe (Steffenhagen et al., 2009). Similarly, the IST-3 showed no significant difference between treated patients with NIHSS ≤5 compared with those >5, although the larger the stroke, the greater the effect of treatment, independent of NIHSS. Although thrombolysis in patients with mild to rapidly improving symptoms remains controversial (Gladstone et al., 2002; NINDS, 2005; Baumann et al., 2006), a subgroup of patients that would benefit from treatment include those with symptoms at presentation perceived to have a disabling deficit despite an NIHSS ≤5, taking into account subjective considerations of the functional impact of the neurological deficit on a specific patient's lifestyle. Examples include isolated aphasia or hemianopia, and established proximal arterial occlusions on intracranial vascular imaging, because of the high chance of poor outcomes (Rajajee et al., 2006; Nedeltchev et al., 2007), in addition to patients with small-vessel events who are at risk of significant disability and are more likely to benefit from intravenous thrombolysis (NINDS, 1995, 2005; Khatri et al., 2010). However, further data on reperfusion effects in this subgroup of patients may be warranted.

Thrombolysis in Patients With a Seizure at Stroke Onset. Seizures usually occur early (within 24 h to 2 weeks) after acute ischaemic stroke onset, but can be delayed (>2 weeks) (Bladin et al., 2000; Lamy et al., 2003; Ryvlin et al., 2006). Occasionally seizures can occur at acute ischaemic stroke onset (Shinton et al., 1988; Selim et al., 2002). Although the reported frequency of early post-acute ischaemic stroke seizures ranges from 2–23%, depending on study design (Burn et al., 1997; Pohlmann-Eden and Bruckmeir, 1997), with late seizure rates of 3–67% (Awada et al., 1999; Camilo and Goldstein, 2004), there are limited data on the frequency of seizures at acute ischaemic stroke onset, although an early study suggested a rate of 5.7% (Shinton et al., 1988).

Seizure at acute ischaemic stroke onset is a relative contraindication for thrombolytic therapy (NINDS, 1995), due to the difficulty in differentiating post-ictal Todd's paralysis from ischaemic stroke paralysis, both clinically and radiologically using non-contrast CT scan (Adams et al., 1996b; Selim et al., 2002). However, multimodal imaging using CT angiography, MRI-diffusion-weighted imaging or perfusion weighted imaging/magnetic resonance angiography can be useful in making the distinction, and may guide thrombolyisis decision-making (Schellinger et al., 2000; Sylaja et al., 2006b; Guerrero et al., 2012). The evidence for the safety of thrombolysis in acute ischaemic stroke patients with a seizure at stroke onset has been limited to case series that have suggested that thrombolysis may be used in the presence of confirmed new ischaemic stroke (Selim et al., 2002; Sylaja et al., 2006b). Both the AHA/ASA and ESO guidelines recommend that patients with seizures at stroke onset be treated with intravenous rt-PA if the neurological deficit is related to acute cerebral ischaemia and not a post-ictal event (ESO, 2008; Jauch et al., 2013).

Thrombolysis for Patients With Recent Surgery or Trauma. The current guidelines contraindicate intravenous thrombolysis in patients having undergone surgery or trauma within the past 14 days, but offer various time windows of relative exclusion, ranging from 21 days to 3 months, for different types of surgery or trauma. No randomized controlled trials have been performed, but case studies suggest that although these patients may benefit from intra-arterial interventions, intravenous thrombolysis remains contraindicated (Katzan et al., 1999; Fukuda et al., 2003; Seifert et al., 2011).

Thrombolysis for Patients on Anticoagulation. Controversy surrounds the safety of intravenous thrombolysis in warfarin-treated patients with subtherapeutic international normalized ratio presenting with acute ischaemic stroke (Harrer and Seet, 2012; Miedema et al., 2012). Different centres and regions have variable approaches to the management of such patients. Under AHA/ASA guidelines, patients with acute ischaemic stroke on warfarin are eligible for intravenous thrombolysis if the pretreatment international normalized ratio is ≤1.7, whereas in Europe warfarin-treated stroke patients are not eligible (Harrer and Seet, 2012).

A recent review and meta-analysis of seven intravenous thrombolysis studies, in which 6.6% of patients were on warfarin before acute ischaemic stroke onset, found increased symptomatic intracranial haemorrhage risk in the warfarin group (OR 2.6, CI 1.1–5.9, P = 0.02), but no significant difference in functional outcome (OR 0.9, CI 0.6–1.2) or death from all causes (OR 1.2, CI 0.9–1.8) (Miedema et al., 2012). Similarly, a retrospective observational study using data from the AHA 'Get With The Guidelines' Stroke Registry found intravenous thrombolysis in the 7.7% of acute ischaemic stroke patients on warfarin (international normalized ratio ≤1.7) was not associated with increased symptomatic intracranial haemorrhage risk (adjusted OR 0.78, CI 0.49–1.24) or in-hospital mortality (adjusted OR 0.94, CI 0.79–1.13), (Xian et al., 2012), suggesting the safety of intravenous thrombolysis in these patients. Further, a prospective observational study found no significant increase in symptomatic intracranial haemorrhage or fatal intracerebral bleed rate in thrombolysed acute ischaemic stroke patients on warfarin (international normalized ratio ≤1.7) (Rizos et al., 2012). There is limited information on thrombolysis in patients taking the newer oral anticoagulants such as dabigatran (Connolly et al., 2009), although a drawback of these drugs is possible difficulty in rapidly reversing their effects.

Further research is required on thrombolysis in this subgroup of patients, although the mortality and functional outcomes in existing studies suggest anticoagulation (with international normalized ratio ≤1.7) should not be an absolute contraindication. Even if they are deemed to fall outside the intravenous thrombolysis recommendations, patients with subtherapeutic anticoagulation could potentially benefit from rapid stratification to appropriate intra-arterial interventions.

Thrombolysis in Grey Areas

Certain conditions present challenges in acute ischaemic stroke patients when thrombolysis is a treatment option, because the risks, benefits and efficacy of thrombolysis are not fully established in conditions such as dementia, despite the guidelines not characterizing these conditions as exclusion criteria. Thrombolysis in such situations therefore remains debatable due to this lack of evidence.

Thrombolysis for 'Wake-up Stroke' or Acute Ischaemic Stroke of Indeterminate Onset. Ten to twenty-five per cent of patients wake up with stroke symptoms, or are found unable to communicate the time of stroke onset, the so-called 'wake-up stroke' (Nadeau et al., 2005; Mackey et al., 2011), and are not eligible for thrombolytic therapy. However, numerous imaging and clinical studies have suggested that a considerable fraction of wake-up stroke patients may have had acute ischaemic stroke onset just shortly before or after waking up, and may therefore have potentially salvageable penumbral tissue (Fink et al., 2002; Serena et al., 2003; Todo et al., 2006; Adams et al., 2008; Huisa et al., 2010; Silva et al., 2010). Studies have shown no significant differences in baseline clinical characteristics and stroke severity in wake-up stroke relative to stroke-while-awake (Nadeau et al., 2005; Jimenez-Conde et al., 2007). Some studies have found early ischaemic changes on CT in wake-up stroke patients to be similar to those with known time of symptom onset (Serena et al., 2003; Todo et al., 2006). Similarly, prospective analysis comparing wake-up strokes to control acute ischaemic strokes showed similar initial ASPECTS between patients with wake-up strokes and those presenting within 4 h of acute ischaemic stroke onset (Huisa et al., 2010). However, in a subgroup analysis from the Abciximab in Emergency Stroke Treatment Trial-II (AbESTT-II), more new strokes were detected in head CTs of wake-up stroke patients compared to those with known symptom onset within 6 h, although the difference was not significant (Adams et al., 2008). Early analysis of the IST, including 5152 (29.6%) wake-up strokes, comparing presentations and outcomes of wake-up stroke to stroke-while-awake patients, found that wake-up stroke patients might benefit from acute interventions (Moradiya and Janjua, 2012), and a retrospective review of wake-up stroke patients given intravenous thrombolysis (off-label) demonstrated its safety with no increased risk of intracerebral haemorrhage (Barreto et al., 2009).

Penumbral mismatch might be useful for identifying wake-up stroke patients who might benefit from thrombolysis with an acceptable level of risk. A review of the safety and efficacy of MRI-based thrombolysis in unclear-onset stroke comparing patients with unclear-onset stroke and those with clear-onset stroke suggest that intravenous or intra-arterial thrombolysis based on specific MRI criteria may be safely applied to unclear-onset acute ischaemic stroke patients, as rates of recanalization, early neurological improvement, symptomatic intracranial haemorrhage and 3 month outcome did not differ significantly (Cho et al., 2008). The recently published Tissue Window in Stroke Thrombolysis (TWIST) trial showed how wake-up stroke patients can be safely given intravenous thrombolysis based on a tissue window defined by non-contrast CT and CT angiography/transcranial Doppler, with 34% of wake-up strokes successfully treated, of which 45% had good outcome (modified Rankin Score 0–1). The authors suggested that, given vascular occlusion with a favourable non-contrast CT appearance (minimal early ischaemic changes or high ASPECTS), wake-up stroke patients should be considered for thrombolysis due to the high probability of a salvageable penumbra (Hill et al., 2013). Similarly, a recent observational study in which 56% of wake-up stroke patients were thrombolysed found thrombolysis to be feasible with potential benefit in selected patients based on diffusion weighted imaging–FLAIR mismatch (Manawadu et al., 2013).

Extending the time for Thrombolysis in Emergency Neurological Deficits (EXTEND) is an ongoing trial comparing clinical outcomes of intravenous thrombolysis versus placebo in acute ischaemic stroke patients with significant penumbral mismatch (MRI or CT) from 3–9 h from stroke onset, or wake-up stroke patients within 9 h of midpoint of sleep duration (NCT00887328) ( Efficacy and Safety of MRI-based Thrombolysis in Wake-up Stroke (WAKE-UP) is another ongoing multicentre randomized controlled trial of MRI-based thrombolysis in acute ischaemic stroke, aiming to prove the efficacy and safety of MRI-based intravenous thrombolysis in wake-up stroke patients or those with otherwise unknown symptom onset (NCT01525290) (, hopefully leading to a larger wake-up stroke evidence base. In light of the existing research, wake-up stroke patients should be assessed for eligibility for thrombolysis if they meet clinical and radiological criteria (i.e. minimal early ischaemic changes, high ASPECTS) in the absence of other contraindications.

Thrombolysis for Patients With Dementia. The role of thrombolysis in stroke patients with dementia has remained a grey area, as this condition did not appear in the inclusion or exclusion criteria for major trials. However, most physicians are reluctant to thrombolyse acute ischaemic stroke patients with dementia because of the fear of increased risk of thrombolysis-related intracerebral haemorrhage, due to the higher incidence of silent cerebral microbleeds in the elderly which may be a marker of cerebral small-vessel disease and cerebral amyloid angiopathy, and may therefore increase the risk of intracerebral haemorrhage and poor outcome (Cordonnier et al., 2006; Koennecke, 2006; Henneman et al., 2009). However, a study of patients from CASES found that white matter disease, as diagnosed on CT, although being associated with a higher rate of symptomatic intracranial haemorrhage, did not have an effect on overall clinical outcome at 3 months (Palumbo et al., 2007).

Most physicians are doubtful of the efficacy of rt-PA thrombolysis in older acute ischaemic stroke patients with dementia (Alshekhlee et al., 2011). However, a case-control study of intravenous thrombolysis found no significant difference in intracerebral haemorrhage (5.80%) or hospital mortality (17.4%) in acute ischaemic stroke patients with dementia compared to those without (Alshekhlee et al., 2011). And in a subgroup analysis of data from the Registry of the Canadian Stroke Network (RCSN) and Registered Persons Database (RPDB), dementia was not independently associated with mortality, disability, or institutionalization after acute ischaemic stroke. A benefit was suggested in patients with dementia given intravenous thrombolysis, insofar as there were no differences between those with and without dementia in the risk of intracerebral haemorrhage or mortality or disability at discharge (Saposnik et al., 2012b).

In contrast, an analysis of patients ≥80 years of age who received intravenous thrombolysis or intra-arterial therapy in the 'Get With The Guidelines' database found prestroke dementia to be a powerful independent predictor of in-hospital mortality and poor outcome after acute reperfusion therapy for stroke (OR 3.61, CI 1.39–9.37) (Busl et al., 2011).

With an ageing demographic, an increasing number of elderly patients will present with acute ischaemic stroke and dementia. Further research is needed to fully establish therapeutic protocols, although the studies suggest that at least younger (<80 years old) acute ischaemic stroke patients with dementia may benefit from thrombolysis.

Thrombolysis in Patients With Malignancy. There are limited data on the safety of thrombolysis for acute ischaemic stroke in patients with maliganacy since this condition was an exclusion criterion from clinical trials (NINDS, 1995) and observational studies (Wahlgren et al., 2007) due to the potential risk of symptomatic intracranial haemorrhage (Hsieh and Chen, 2009).

In a retrospective single centre experience of 308 thrombolized acute ischaemic stroke patients, 18 (5.8%) had a concurrent malignancy, and 26 (8.4%) had a remote history of malignancy. Concurrent malignancy was not independently associated with increased in-hospital mortality following thrombolysis, as the mortality was attributable largely to medical comorbidities, not to symptomatic intracranial haemorrhage. However, patients with brain metastases were excluded (Masrur et al., 2011). A case series of thrombolysed acute ischaemic stroke patients with cancer, without cerebral metastasis (Casado-Naranjo et al., 2011) reported on the safety of intravenous rt-PA.

Evidence for thrombolysis in acute ischaemic stroke patients with intracranial neoplasm is limited and based mainly on case reports (Grimm and DeAngelis, 2007; Garcia et al., 2009; Hsieh and Chen, 2009; Casado-Naranjo et al., 2011; Neil and Ovbiagele, 2011), including two cases of safe intravenous thrombolysis in acute ischaemic stroke patients with intracranial neoplasm outside of the brain parenchyma without any associated intratumour haemorrhage (Neil and Ovbiagele, 2011).

Further exploration of the risks and benefits of intravenous thrombolysis for acute ischaemic stroke patients with concurrent malignancy is warranted, although case reports suggest that there may be benefit in the absence of intracranial neoplasm.

Thrombolysis in Pregnancy. Acute ischaemic stroke during pregnancy and puerperium is a rare but potentially devastating event, accounting for >12% of all maternal deaths (Simolke et al., 1991; Del Zotto et al., 2011). The reported incidence of pregnancy- or puerperium-associated acute ischaemic strokes varies considerably, from 4.3–210/100 000 deliveries (Sharshar et al., 1995). Acute ischaemic stroke treatment in pregnancy can be quite challenging due to concerns about potential harm to the foetus, particularly during the first trimester when there is a potentially higher risk of teratogenicity and adverse outcomes in the mother. There are limited data on the safety of thrombolysis in pregnancy because it was an exclusion criterion in clinical trials due to fear of potential maternal and foetal risks, such as placental abruption, premature labour, abortion, retroplacental haemorrhage, peri-partum and postpartum haemorrhage, or even harm to the foetus (Wiese et al., 2006). Data are limited to case reports and small case series. Of the 11 reported cases of thrombolysis in acute ischaemic stroke patients during pregnancy and puerperium, administered from 1–37 weeks gestation (Dapprich and Boessenecker, 2002; Elford et al., 2002; Johnson et al., 2005; Leonhardt et al., 2006; Murugappan et al., 2006; Wiese et al., 2006) and during the postpartum period (Mendez et al., 2008; Ronning et al., 2010), all were associated with improvement in maternal symptoms and delivery of healthy babies, although two were complicated by intracerebral haemorrhage after thrombolysis (Dapprich and Boessenecker, 2002; Elford et al., 2002), an incidence similar to that in non-pregnant patients.

Intra-arterial therapy may be an effective alternative to intravenous thrombolysis to potentially reduce the risk of abruption. There are reported cases of intra-arterial therapy for acute ischaemic stroke in pregnancy (Elford et al., 2002; Johnson et al., 2005) and in the puerperium (Ronning et al., 2010). Admittedly, the number of reported cases is too small to draw any definitive conclusions, but case reports suggest thrombolysis may be feasible and beneficial in pregnancy. Although alteplase does not cross the placental barrier and there is no evidence of teratogenicity from animal studies, potential adverse foetal effects remain unknown (Kojima et al., 1988; Tanaka et al., 1988; Leonhardt et al., 2006). However, in standard obstetric practice, the health of the mother does take precedence over the health of the foetus. Further study on the safety and efficacy of rt-PA thrombolysis in pregnancy is needed, but, in the absence of definitive evidence, benefits and risks of thrombolysis in pregnant acute ischaemic stroke patients should be carefully balanced on a patient-by-patient basis.

Thrombolysis at Extremes of Age

Thrombolysis in Older People. About 30% of all acute strokes occur in subjects over 80 years of age (Bonita et al., 1994; Di Carlo et al., 2003; Marini et al., 2004; Sylaja et al., 2006a; Saposnik et al., 2009), yet there are limited trial data in the very elderly (>80 years old), as trials demonstrating the benefit of intravenous thrombolysis have either excluded them or randomized very few (NINDS, 1995; Hacke et al., 2008). Nonetheless, thrombolysis has been shown to significantly decrease the risk of mortality following acute ischaemic stroke irrespective of age (Caso et al., 2007; Mateen et al., 2009, 2010). Unfortunately, patients ≥80 years old are often excluded from intravenous thrombolysis in clinical practice (Barber et al., 2001; Hacke et al., 2008; Mishra et al., 2010b), primarily because of fears that older people may be predisposed to a greater risk of intracerebral haemorrhage (Zeevi et al., 2007) due to factors such as impaired rt-PA clearance, increased rates of cardio-embolic stroke, and possible amyloid angiopathy (Simon et al., 2004), in addition to the concern that advancing age is associated with poorer prognosis in terms of increased in-hospital mortality risk (Hacke et al., 2004; Heuschmann et al., 2004; Bateman et al., 2006). However, meta-analyses of thrombolysed patients (Engelter et al., 2006; Pundik et al., 2008; Ford et al., 2010; Mishra et al., 2010b; Costello et al., 2012) did not find increased symptomatic intracranial haemorrhage risk among elderly patients despite their generally less favourable outcomes (Engelter et al., 2006; Ringleb et al., 2007), which were attributable to comorbidities rather than consequences of thrombolysis-related complications (NINDS, 1997b; Derex and Nighoghossian, 2009; Alshekhlee et al., 2010; Fonarow et al., 2010).

A controlled study of the influence of age on thrombolysis outcome in acute ischaemic stroke patients from the Virtual International Stroke Trials Archive (VISTA) study, which analysed patients ≤80 and ≥81 years separately, demonstrated the benefit of thrombolysis in the very elderly (Mishra et al., 2010b). Functional outcome, measured by modified Rankin Score, was better in the thrombolysed group (OR 1.39, CI 1.26–1.54, P < 0.0001), regardless of whether they were young (OR 1.42, CI 1.26–1.59, P < 0.0001) or elderly (OR 1.34, CI 1.05–1.70, P = 0.002). Similarly, comparison data from SITS-ISTR and VISTA showed increasing age is generally associated with poorer outcome, but the significant association between treatment and improved outcome at 3 months is maintained even in the very elderly (Mishra et al., 2010a). Also, in the recently published IST-3, where 53% of 3035 acute ischaemic stroke patients were >80 years old, the adjusted effect of treatment between patients >80 and <80 years demonstrated a significant difference (P = 0.027), indicative of a larger benefit in the older age group, adding substantially to the meta-analysis demonstrating that the benefits of thrombolysis are at least as large in the elderly as in younger patients (Wardlaw et al., 2012).

The available evidence suggests thrombolysis should not be withheld on the basis of advanced age alone. However, recombinant human tissue-type plasminogen activator randomized controlled trials with no upper age limit, which may better determine efficacy in reducing disability and mortality in older stroke patients, are still needed, because the IST-3 subgroup analysis is based on a non-significant treatment effect in their primary outcome: proportion of patients alive and independent at 6 months when treated within a 6 h time window.

Thrombolysis in Children. Thrombolysis in children is not a well-established practice due to a paucity of safety and efficacy data. Risk of bleeding may be higher, especially in neonates in whom plasminogen concentrations are frequently low, in addition to immature haemostatic and fibronolytic mechanisms as well as a changing cerebral vasculature (Adams et al., 1996a). In the only large multicentre observational study, involving 687 children with acute ischaemic stroke, only 15 (2%) were thrombolysed, nine intravenous and six intra-arterial. Four (two intravenous and two intra-arterial) had asymptomatic haemorrhages. Neurological deficit was not defined using any validated scale (Amlie-Lefond et al., 2009). This suggests a need for randomized controlled trials in children, although the infrequency of acute ischaemic stroke at this age will make finding a suitably large patient group difficult.

Pre-thrombolysis Management

Management of physiological variables such as blood pressure, blood glucose and body temperature before thrombolysis is crucial as these variables can affect the clinical outcome of acute ischaemic stroke patients treated with intravenous rt-PA.

Blood Pressure. Elevated pretreatment blood pressure is independently associated with an increased likelihood of symptomatic intracranial haemorrhage (Tsivgoulis et al., 2010). Retrospective analysis of the SITS-ISTR demonstrated a strong linear association between systolic blood pressure and risk of symptomatic intracranial haemorrhage (Ahmed et al., 2009). Although the management of blood pressure in the setting of acute ischaemic stroke is controversial, it is recommended that thrombolytic agents should only be administered to patients with systolic blood pressure <185 mmHg and diastolic blood pressure <110 mmHg at the time of treatment to avoid haemorrhagic transformation (Adams et al., 2007) or high rates of persisting occlusion and partial recanalization (Tsivgoulis et al., 2007). An observational study suggests that pretreatment of blood pressure in acute ischaemic stroke patients before intravenous rt-PA is not associated with increased rate of haemorrhage or poor functional outcome (Martin-Schild et al., 2008). Elevated blood pressure can be treated with intravenous agents such as labetalol, which is preferred because it is easily titrated and has minimal vasodilatory effects on cerebral blood vessels. Other agents to consider include intravenous nicardipine and transdermal nitroglycerine. If blood pressure does not respond and remains >185/110 mmHg, rt-PA should not be administered. Use of further aggressive measures in patients with blood pressure >185/110 mmHg is not recommended because further appropriate control of blood pressure for 24 h may not be possible (Jauch et al., 2013). However, post-thrombolysis, aggressive measures are appropriate to control blood pressure during and for 24 h following therapy.

Enhanced Control of Hypertension and Thrombolysis Stroke Study (ENCHANTED) is a Phase 3 randomized controlled trial designed to establish the effects of low-dose rt-PA and early intensive blood pressure lowering in acute ischaemic stroke patients, with an estimated completion date of December 2016. ENCHANTED aims to compare the efficacy (clinical outcomes) and safety (symptomatic intracranial haemorrhage rates) of standard dose (0.9 mg/kg) versus lower dose (0.6 mg/kg) intravenous thrombolysis, in addition to comparing the efficacy and safety of current blood pressure control recommendations (<185 mmHg systolic blood pressure) to rapid intensive blood pressure lowering (to below 150 mmHg systolic blood pressure) (NCT01422616) (

Blood Glucose. Hyperglycaemia is one of the most important predictors of poor outcome in acute ischaemic stroke patients, with or without intravenous thrombolysis. Hyperglycaemia >7.7 mmol/l before thrombolysis has been shown to be an independent risk factor for recanalization failure (Ribo et al., 2005), and has been associated with diminished neurological improvement, greater infarct size, and worse clinical outcome at 3 months after treatment (Alvarez-Sabin et al., 2003). In a cohort of thrombolysed acute ischaemic stroke patients from CASES, admission glucose >8.0 mmol/l was independently associated with increased risk of death (adjusted relative risk 1.5, CI 1.2–1.9), symptomatic intracranial haemorrhage (adjusted relative risk 1.69, CI 0.95–3.00), and poor functional outcome at 3 months (adjusted relative risk 0.7, CI 0.5–0.9) (Poppe et al., 2009). Similarly, in the SITS-ISTR registry of intravenous-treated patients, blood glucose >6.6 mmol/l was associated with significantly higher odds for mortality (OR 1.24, CI 1.07–1.44, P = 0.004) and lower odds for independence (OR 0.58, CI 0.48–0.70, P < 0.001), and blood glucose of 10–11 mmol/l was associated with increased risk of symptomatic intracranial haemorrhage (OR 2.86, CI, 1.69–4.83, P < 0.001) (Ahmed et al., 2010). These findings suggest that ultra early glycaemic control pre-intravenous thrombolysis may improve outcomes, and correction with rapidly acting insulin is recommended in most published guidelines (ESO guidelines: to <10 mmol/l; AHA/ASA guidelines: maintain within 7.7–9.9 mmol/l).

The Stroke Hyperglycemia Insulin Network Effort (SHINE) trial is an ongoing multicentre randomized controlled trial of 1400 patients, with results expected in 2018, hypothesizing that treatment of hyperglycaemic acute ischaemic stroke patients to a targeted glucose concentration (4.4–7.15 mmol/l) will be safe and result in improved 3 month outcomes (NCT01369069) (

Post-thrombolysis Management

Following intravenous thrombolysis it is crucial that thrombolized patients, even those with clinical improvement, are closely monitored for possible neurological deterioration. Patients must be admitted to a hyper-acute stroke unit for the first 24 h post-intravenous thrombolysis for strict haemodynamic and neurological monitoring. The concept of specialized stroke units is generally agreed upon as one of the most effective interventions reducing mortality and morbidity after acute ischaemic stroke (Stroke Unit Trialists' Collaboration, 1997, 2007; Saposnik et al., 2008, 2009). One needs close observation and frequent monitoring of patients for early neurological worsening and any signs of symptomatic intracranial haemorrhage or adverse drug reaction. Patients may experience early neurological deterioration, even after initial improvement following intravenous thrombolysis (Saqqur et al., 2007a; Awadh et al., 2010; Delgado et al., 2010), the main causes including haemorrhagic transformation (Berger et al., 2001; Warach and Latour, 2004), cerebral oedema (The Helsinki Stroke Thrombolysis Registry Group, 2012), early recurrent ischaemic stroke (Georgiadis et al., 2006; Awadh et al., 2010), persistent arterial occlusion, partial recanalization, and arterial re-occlusion (Alexandrov et al., 2000; Christou et al., 2000; Grotta et al., 2001; Saqqur et al., 2007a).

Neurological status (using NIHSS) and blood pressure should be monitored every 15 min for the first 2 h, then every 30 min for 6 h, then hourly for 16 h (Adams et al., 2007). Blood pressure monitoring is recommended for early detection of hypotension, most likely due to overtreatment, which can worsen cerebral ischaemia. Patients should also be monitored for hemi-orolingual angioedema, particularly those with pre-existing hypertension who are taking angiotensin-converting enzyme inhibitors (Hill et al., 2003a). This is in addition to monitoring other vital signs such as glucose and oxygen saturation. Data analysis from the Helsinki stroke registry found hyperglycaemia (≥8.0 mmol/l) during the 48 h after thrombolysis independently predicted unfavourable outcome, symptomatic intracranial haemorrhage and death (Putaala et al., 2011), and a prospective study of 80 thrombolized acute ischaemic stroke patients found blood pressure variability was associated with greater diffusion-weighted imaging lesion growth and worse clinical course (Delgado-Mederos et al., 2008).

A follow-up CT or MRI should be obtained at 24 h to exclude haemorrhage before antiplatelets or anticoagulants are commenced. If haemorrhage occurs, particularly symptomatic intracranial haemorrhage, consultation with neurosurgical specialists and administration of blood products, including fresh-frozen plasma and platelets, should be considered. Similarly, persistence of hyperdense middle cerebral artery sign (HMCAS) on the follow-up CT can be a useful early predictor of poor functional outcome (Paliwal et al., 2012) and may suggest the need for aggressive medical or surgical intervention, such as pre-emptive decompressive craniotomy.

It is also crucial to establish the aetiological factors contributing to the acute ischaemic stroke, such as hypertension, carotid artery stenosis, or atrial fibrillation, and to take appropriate long-term preventative measures (e.g. antihypertensives, carotid endarterectomy, anticoagulation, antiplatelet drugs) to avoid stroke recurrence.

Predictors of Clinical Outcome in Intravenous rt-PA-treated Patients

Factors predicting functional outcome after treatment with intravenous rt-PA include baseline NIHSS, age, admission blood glucose, the presence of hyperdense middle cerebral artery sign, and early ischaemic changes on admission head CT or an ASPECTS <7 (Barber et al., 2000; Coutts et al., 2003, 2004b; Kharitonova et al., 2009; Lees et al., 2010).

Clinical Scoring Tools for Predicting Outcome. Several scoring tools using both clinical and imaging parameters available before initiating thrombolysis have been developed to predict clinical response and long-term outcome in acute ischaemic stroke patients. Examples include the DRAGON score, iScore and hemorrhage after thrombolysis (HAT) score (Lou et al., 2008; Strbian et al., 2011; Saposnik et al., 2012a) (Box 2).

Imaging Scoring Tools for Predicting Outcome. A patient with ASPECTS ≤7 has been shown to have a 14-fold increased risk of symptomatic intracranial haemorrhage compared to ASPECTS >7 (Barber et al., 2000). The presence of a hyperdense middle cerebral artery sign and a hyperdense basilar artery sign are associated with less favourable outcomes (Tomsick et al., 1996; Derex et al., 2005; Goldmakher et al., 2009; Nagel et al., 2009), and the hyperdense middle cerebral artery sign on baseline scan may identify patients who require more aggressive therapeutic interventions (Doerfler et al., 1996; Schwab et al., 1998; Manno et al., 2003). Its presence should prompt admission to an ICU or other supervised area. Similarly, hyperdense basilar artery sign can be useful both diagnostically, predicting the presence of basilar artery thrombosis, and prognostically, predicting short- and long-term outcomes in patients with a clinical picture suggestive of posterior circulation infarct (Goldmakher et al., 2009). The presence of hyperdense basilar artery sign will indicate the need for urgent CT angiography and possible transfer to a facility with interventional stroke treatment.

Complications of Intravenous Thrombolytic Therapy

The majority of complications of intravenous thrombolytic therapy, such as bleeding (intracerebral haemorrhage and systemic bleeding), angioedema and reperfusion injury with oedema, are due to the thrombolytic actions of rt-PA. Others, such as reocclusion and secondary embolization, are related to ineffective thrombolysis or redistribution of the lysed clot. Also, rt-PA can act upon the brain parenchyma, leading to neurotoxicity and seizures (Balami et al., 2013b).

Symptomatic intracranial haemorrhage is one of the most unfavourable and feared complications of intravenous thrombolysis, occuring in, depending on the definition used, 1.7–8.0% of treated patients. (NINDS, 1995; Albers et al., 2000; Hacke et al., 2004, 2008; Hill and Buchan, 2005; Wahlgren et al., 2007). The three symptomatic intracranial haemorrhage definitions are the NINDs (1997b), the ECASS III (Hacke et al., 2008), and the SITS-MOST (Wahlgren et al., 2007).

Pooled data of 2775 patients in the large randomized (ATLANTIS, ECASS and NINDS) trials showed a symptomatic intracranial haemorrhage rate of 5.9% for rt-PA treatment versus 1.1% for placebo (P < 0.0001) (Hacke et al., 2004). Similarly, meta-analysis of 2639 rt-PA for acute ischaemic stroke patients in general clinical practice showed a symptomatic intracranial haemorrhage rate of 5.2% (Graham, 2003). SITS-MOST's analysis of 31 627 intravenous thrombolysis patients identified nine independent risk factors for symptomatic intracranial haemorrhage: baseline NIHSS, serum glucose, systolic blood pressure, age, body weight, onset-to-treatment time, aspirin or combined aspirin and clopidogrel, and history of hypertension, with an overall symptomatic intracranial haemorrhage rate of 1.8% (Wahlgren et al., 2007). Other factors include signs of mass effect, brain oedema, hypodensity or extensive early infarct change on pretreatment brain imaging (Larrue et al., 1997; NINDS, 1997a, b; Lansberg et al., 2007; Saver and Yafeh, 2007), permeability changes detected on MRI (Kakuda et al., 2008), pretreatment diffusion weighted imaging lesion volume, and post-thrombolysis blood pressure (Butcher et al., 2010).

Several scoring tools using both clinical and imaging parameters available before initiation of thrombolysis have been developed to predict risk of haemorrhagic transformation (HT) in acute ischaemic stroke patients treated with intravenous rt-PA. SITS symptomatic intracranial haemorrhage risk score, a 12 point score adapted from the SITS-ISTR, can easily be applied to predict symptomatic intracranial haemorrhage risk after intravenous rt-PA—a score ≥10 is associated with a >70-fold increased symptomatic intracranial haemorrhage rate compared to a score of 0 (Mazya et al., 2012). The iScore, HAT and ASPECTS, discussed above, have also been validated in predicting HT risk (Barber et al., 2000; Lou et al., 2008; Saposnik et al., 2012a). Additionally, the SEDAN score, ranging from 0 to 6 points, consists of baseline blood Sugar, Early infarct signs and hyperDense cerebral artery sign, Age and baseline NIHSS. It is used for assessing risk of symptomatic intracranial haemorrhage in intravenous thrombolysis patients with anterior and posterior circulation acute ischaemic stroke (Strbian et al., 2012).

Non-responders to rt-PA

Despite the proven benefits of rt-PA in patients with acute ischaemic stroke, only about half of patients respond to intravenous thrombolysis (Rha and Saver, 2007; Hacke et al., 2008; Lees et al., 2010). Clinical response may be affected by a number of factors, including location and extent of arterial occlusion, collateral integrity and characteristics of the clot itself such as burden, age and composition (Tan et al., 2007; Fernandez-Cadenas et al., 2009; von Kummer, 2010). These factors are important in predicting non-response to rt-PA, and may be helpful in choosing other reperfusion techniques.

Site of Arterial Occlusion. Patients with terminal internal carotid artery occlusion, proximal middle cerebral artery occlusion, or tandem lesions might have a poor clinical response to rt-PA (Saqqur et al., 2007b). A study of intravenous thrombolysis for acute ischaemic stroke patients found the rate of complete recanalization to be 44.2% for distal middle cerebral artery (favourable outcome, modified Rankin Score ≤ 1 at 3 months = 52%), 30% for proximal middle cerebral artery (25%), 5.9% for terminal internal carotid artery (18%), 27% for tandem cervical internal carotid artery/middle cerebral artery (21%), and 30% for basilar artery occlusion (25%) (Saqqur et al., 2007b). Although continuous transcranial monitoring was a positive predictor for complete recanalization, pre-rt-PA NIHSS, glucose, systolic blood pressure and thrombolysis in brain ischaemia (TIBI) flow grade at the occlusion site were negative independent predictors. Patients with dampened flow (TIBI 3) at the occlusion site had higher odds of complete recanalization than those with no demonstrable residual flow signals (TIBI 0).

Collateral Integrity. Transcranial Doppler flow findings at the site of intracranial occlusion can predict the clinical response to intravenous thrombolysis—patients with lack of residual flow have a low probability of complete recanalization and recovery after intravenous thrombolysis (Labiche et al., 2003; Saqqur et al., 2009). In an earlier study of pretreatment transcranial Doppler, patients with no detectable residual flow signals had a <20% chance for complete early recanalization with intravenous thrombolysis (Labiche et al., 2003). In another study of transcranial Doppler-detected residual flow with TIBI grading before intravenous rt-PA bolus in patients with proximal arterial occlusion, 17.7% of patients with TIBI 0, 33.1% with TIBI 1, 38.2% with TIBI 2 and 47.7% with TIBI 3 had achieved complete recanalization (P < 0.001). Negative independent predictors of complete recanalization were high NIHSS, glucose, systolic blood pressure and lower TIBI grades. Poor outcome rate at 3 months (modified Rankin Score >2) was 61.3% in patients with TIBI 0, 56.9% for TIBI 1, 51.5% for TIBI 2, and 33.9% for TIBI 3 (P = 0.012) (Saqqur et al., 2009).

Clot Burden. Studies have demonstrated a relationship between the clot burden and recanalization (Tan et al., 2009). More proximal occlusions have been found to carry greater thrombus burden (Saqqur et al., 2007b). Using a CT angiogram grading system, the clot burden score is a 10-point scoring system developed to quantify thrombus burden (i.e. presence of contrast opacification on CT angiography) in a large cohort of patients with anterior circulation acute ischaemic stroke. Two points are subtracted for thrombus found in each of: the proximal M1 segment of the middle cerebral artery trunk, the distal M1 segment, and the supraclinoid internal carotid artery. One point is subtracted for thrombus found in each of the M2 branches, A1, and the infraclinoid internal carotid artery. A score of 10 is normal (clot absent), whereas a score of 0 implies complete multisegment vessel occlusion. Clot burden score <10 was associated with reduced odds of independent functional outcome (OR 0.09 for clot burden score ≤5; OR 0.22 clot burden score 6–7; OR 0.48 clot burden score 8–9; all versus clot burden score 10; P < 0.02 for all). Lower clot burden scores were associated with lower follow-up ASPECTS (P < 0.001) and higher parenchymal haematoma rates (P = 0.008) (Puetz et al., 2008a). Another study found clot burden score and collateral score to be useful additional important markers predicting clinical and radiological outcomes. Higher clot burden score and collateral score demonstrated smaller pretreatment perfusion defects and final infarct volume and better clinical outcome (P < 0.01) (Tan et al., 2009).

Clot Composition and Age. Clot age and composition of thromboembolic material are factors likely to predict response to rt-PA therapy (Fulgham et al., 2004; Kim et al., 2006). Non-contrast CT measurement of thrombus composition based on Hounsfield units may be helpful in predicting response to intravenous thrombolysis: thrombus with lower Hounsfield units count on non-contrast CT (platelet-rich) is more resistant to lysis than thrombus with higher Hounsfield units count (erythrocyte-rich) (Kirchhof et al., 2004; Kim et al., 2006). Emboli composed of cholesterol, calcium, calcific plaque, fat or clots containing other debris may be resistant to enzymatic degradation by rt-PA (Halloran and Bekavac, 2004). Thrombolytics are also less effective in treating both mature embolic clots, due to excessive cross-linking (Fulgham et al., 2004), and hyperacute thromboemboli, which are platelet-rich (Fulgham et al., 2004).

Other Thrombolytics

Although alteplase (rt-PA) is currently the only approved thrombolytic agent for acute ischaemic stroke treatment, its limited fibrin specificity and possible neurotoxicity have fueled the search for other plasminogen activators. Newer thrombolytics with potentially improved half-life, higher target specificity and better safety profile are being evaluated in clinical trials. A phase 2B trial found significantly greater reperfusion (P = 0.004) and better clinical outcomes (P < 0.001) at 24 h in patients given tenecteplase versus alteplase <6 h after acute ischaemic stroke onset (Parsons et al., 2012). Similarly, the phase 2 desmoteplase in acute ischaemic stroke (DIAS) and Dose Escalation Study of Desmoteplase in Acute Ischaemic Stroke (DEDAS) trials demonstrated promising results with desmoteplase given 3–9 h after stroke onset (Hacke et al., 2005; Furlan et al., 2006), but the phase 3 DIAS-2 did not show any clinical benefit at 3 months (Hacke et al., 2009). However, DIAS-3, DIAS-4, and DIAS-J are currently ongoing (NCT00790920, NCT00856661, NCT01104467) ( Other examples with trials in progress include reteplase, plasmin and microplasmin. These have been extensively reviewed elsewhere (Balami et al., 2013a).