The target of stroke therapy is the ischaemic penumbra, i.e. the area of brain tissue that after an arterial occlusion has a residual collateral blood flow, insufficient to make neurons function but sufficient to make them survive for some hours after stroke onset. Part of the clinical picture derives from the non-functioning of this area. Two main approaches are currently used to rescue the ischaemic penumbra: arterial reopening with thrombolytic agents, and neuroprotection.
PHASE III TRIALS OF THROMBOLYSIS
A number of clinical trials with i.v. or i.a. thrombolytic agents have been performed during the last decade, and we now have data on more than 5000 patients, allowing some conclusions to be drawn about the advantages and disadvantages of this therapy. The major thrombolytic drugs tested in human stroke today are described below.
The three major trials on intravenous thrombolysis with streptokinase (Australian Stroke Trial, Multicentre Acute Stroke Trial Europe, Multicentre Acute Stroke Trial Italy) were stopped before completion due to an excess of symptomatic haemorrhagic transformation and of both early and late fatality rates (1-3). The rate of death or dependency at the end of follow-up was reduced in treated patients, but not significantly. All three trials excluded patients with minor or rapidly improving symptoms, but included stuporose or comatose patients. No computerized tomography (CT) exclusion criteria were adopted. The causes of this substantial failure of streptokinase have been widely debated in the literature, the treatment of too severely ill patients and the concomitant use of aspirin (1,3) probably being the most likely.
Recombinant tissue plasminogen activator
Large trials with recombinant tissue plasminogen activator (r-TPA) were published between 1995 and 1999 (Table 1). The main differences between study protocols were the 3-h limit for treatment in The National Institute of Neurological Disorders and Stroke (NINDS) trial (7) and the 6-h limit in both the European Cooperative Acute Stroke Studies (ECASS I and II) (8,9), while the Alteplase ThromboLysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) study (10) focused on patients treatable between 3 and 5 h of stroke onset. The drug dose was 0.9 mg/kg, with a maximum of 90 mg, in all but in the ECASS I study, which administered 1.1 mg/kg up to a maximum dose of 100 mg. Patients with slight or rapidly improving symptoms were excluded from all trials, whereas very severe patients were excluded in the ECASS studies but not in the NINDS or ATLANTIS trials. Finally, CT exclusion criteria, i.e. exclusion of patients with early signs of the infarct at CT in more than one-third of middle cerebral artery territory, were adopted in the two ECASS and in the ATLANTIS studies.
The results are definitely more positive than those of the streptokinase trials. In fact, despite an increased incidence of symptomatic haemorrhagic transformation, r-TPA significantly reduced death or dependency at the end of follow-up, not only in patients treated within 3 h but also in the whole population of treated patients. Seen another way, for every 1000 patients treated, r-TPA leads to 20-93 fewer dead or dependent patients when given within 6 h of stroke onset and to 77-203 fewer dead or dependent patients when given within 3 h. Intracranial haemorrhage is determined in 55-99 more patients, indicating the relatively low therapeutic index of r-TPA. But considering that fatal haemorrhages are already included in the death or dependency count, the increased risk of haemorrhagic transformation does not nullify the positive effects of treatment.
A comparison of the absolute effects of thrombolysis in stroke and in myocardial infarction may give a better idea of the potentialities of thrombolysis in stroke (Table 2). For every 1000 patients with acute myocardial infarction given thrombolysis there are 15-40 more patients saved according to the time interval from symptom onset, as opposed to 59 more stroke patients avoiding death or dependence when treated within 6 h and to 180 more when treated within 3 h.
PHASE IV TRIALS OF THROMBOLYSIS
The above-mentioned trials demonstrate that thrombolysis is efficacious, i.e. it provides a net benefit in the ideal conditions of clinical trials, but the question is whether it is effective in the real world. To ascertain this, one parameter of interest is the cost-effectiveness. Based on length of hospital stay and percentage of patients discharged to home in the NINDS trial, a mathematical model estimated that per 1000 treated patients there would be an increase in costs for hospitalization but a decrease in costs for rehabilitation or admission to a nursing home, with a net saving of US$4.5 million (11).
Thrombolysis has now become routine in some centres, which have published encouraging results on sufficiently large samples of patients. In a 17-month period Grond et al., following the NINDS rules, managed to treat 100 patients within 3 h of onset, and obtained results quite comparable to those of the NINDS trial as regards good outcome (53%), fatality rate (12%) and haemorrhagic transformation (11%) (12).
Trouillas et al. also treated 100 patients, but interestingly with a lower dose of r-TPA (0.8 mg/kg) and with a larger therapeutic window of 7 h, and reported data overlapping that of the NINDS trial as regards favourable outcome (45%) and incidence of haemorrhagic transformation (7%), with an additional lower fatality rate (6%) (13).
Finally, the Standard Treatment with Activase to Reverse Stroke (STARS) is an observational study started in 1996 in 18 academic centres and 38 community hospitals in the U.S.A. to follow the use of thrombolysis after FDA approval (14). So far data on 389 patients treated according to the NINDS rules have been communicated, reporting a lower rate of intracranial haemorrhages (11.5%) and of mortality (13%) but also a lower frequency of good outcome (43%) compared with the NINDS results.
However, in this apparently optimistic scenario it is disappointing that, even after FDA approval of r-TPA in the U.S.A., only 1% of patients reaching hospital in time for thrombolysis are currently treated. This substantial scepticism of clinicians may be explained not only by the well-known inertia to transfer a pharmacological success from the experimental setting to clinical practice. In fact, there are still some open questions that influence clinician attitude when weighting the risk/benefit ratio of their therapeutic choices.
The most feared side-effect associated with cerebral thrombolysis is haemorrhagic transformation of the infarct. Extended early CT signs of the index stroke have been repeatedly indicated as reliable predictors of haemorrhagic transformation, in particular in the context of the ECASS I and II studies. Extended early CT signs herald a large infarct, which in turn represents the main risk of secondary bleeding. Early CT signs of the infarct turned out to be predictors of haemorrhagic transformation even in a post hoc analysis of the NINDS trial, but in that study they were retrospectively found in only 5% of patients, compared to 46% of ECASS I and 57% of ECASS II patients. This suggests significant discrepancies in the criteria adopted by different researchers to define the early CT signs, besides stressing significant interobserver variability in early CT reading, a variability that can however be significantly minimized by appropriate training. However, even in the ECASS I study, in which an extremely accurate central CT reading was performed, half of patients who developed symptomatic haemorrhagic transformation did not have early CT signs. Hence, basic and clinical research must continue to investigate mechanisms predisposing to or determining haemorrhagic transformation. It has recently been demonstrated that rat middle cerebral arteries exposed to 2-h ischaemia and then to r-TPA present a markedly reduced reactivity to acetylcholine and serotonin. This suggests an impairment of the cerebral circulation ability to respond to changes in perfusion pressure, which may potentially lead to complications of either brain haemorrhage or oedema. In experimental ischaemia, it has also been demonstrated that a significant loss of basal lamina components of the cerebral microvessels is time dependent and is related to the development of petechial haemorrhages. Many mechanisms may explain this occurrence, such as laminin degradation generated by plasmin (and hence potentially triggered by thrombolysis itself) or transmigration of leukocytes through the vessel wall with release of leukocyte granule enzymes using laminin as substrate. Theoretically, to avoid dangerous intervention this loss of microvascular integrity may be attenuated by a biochemical quantification of the basal lamina damage, for example by quantifying the plasma levels of laminin breakdown products. Moreover pharmacological strategies could be adopted to protect the basal lamina during thrombolysis, such as the use of antibodies with high affinity to basal lamina combined with potent antifibrinolytic agents to protect the exposed subendothelium, or the administration of compounds able to prevent leukocyte adhesion. These latter pertain to the category of neuroprotectant agents and this leads us to consider the second approach to stroke therapy, i.e. neuroprotection.
No single trial with this category of drugs has so far obtained positive results. This is probably the 'chronicle of an announced death' because the rationale of neuroprotection is to antagonize the noxious mechanisms triggered by ischaemia, waiting for spontaneous reperfusion, or to counteract the potentially harmful effects of reperfusion itself. Hence, in both cases the prerequisite to demonstrating an effect of neuroprotectant agents is spontaneous reperfusion. Actually, this is reported to occur at an increasing rate over the days following stroke onset, and in up to two-thirds of cases within 2 weeks, according to some authors. However, given the dynamic nature of ischaemic penumbra, as time goes by the detrimental effects which at stroke onset are triggered by very low levels of residual blood flow, may subsequently be boosted by flow levels initially compatible with neuronal survival (15). This means that if reperfusion is not rapid and complete, the neuroprotectant agents may lose their efficacy with time.
The logical conclusion of what we have discussed so far is that pharmacological reperfusion and brain protection are probably mutually dependent, because thrombolysis may need the concomitant use of compounds able to protect the vessel wall, and neuroprotectant agents actually need thrombolysis to disclose their effect. The next step forward for clinical research is to test the combination of thrombolysis with one or more neuroprotectant agents, in order to increase the number of potentially treatable patients, to prolong the temporal window of therapeutic opportunity and to increase the effectiveness of treatment, with a possible decrease in side-effects.
1. Donnan GA, Davis SM, Chambers BR, et al. Trials of streptokinase in severe acute ischemic stroke. Lancet
2. The Multicenter Acute Stroke Trial—Europe Study Group. Thrombolytic therapy with streptokinase in acute ischemic stroke. N Engl J Med
3. Multicenter Acute Stroke Trial—Italy Group. Randomized controlled trial of streptokinase, aspirin, and combination of both in treatment of acute ischemic stroke. Lancet
4. Mori E, Yoneda Y, Tabuchi, et al. Intravenous recombinant tissue plasminogen activator in acute carotid artery territory stroke. Neurology
5. Haley EC Jr, Brott TG, Sheppard GL, et al. Pilot randomized trial of tissue plasminogen activator in acute ischaemic stroke. The TPA Bridging Study Group. Stroke
6. Yamaguchi T, Hayakawa T, Kiuchi H for the Japanese Thrombolysis Stduy Group. Intravenous tissue plasminogen activator ameliorates the outcome of hyperacute embolic stroke. Cerebrovasc Dis
7. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med
8. Hacke W, Kaste M, Fieschi C, et al. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. The European Cooperative Acute Stroke Study (ECASS). JAMA
9. Hacke W, Kaste M, Fieschi C, for the second European- Australasian Acute Stroke Study Investigators. Randomized double- blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischemic stroke (ECASS II). Lancet
10. Clark WM, Wissman S, Albers GW, Jhamandas JH, Madden KP, Hamilton S: Recombinant tissue-type plasminogen activator (Alteplase) for ischemic stroke 3 to 5 hours after symptom onset. The ATLANTIS Study: a randomized controlled trial. Alteplase ThromboLysis for Acute Noninterventional Therapy in Ischemic Stroke. JAMA
11. Fagan SC, Morgenstern LB, Petitta A, et al. Cost-effectiveness of tissue plasminogen activator for acute ischemic stroke. NINDS rtPA Stroke Study Group. Neurology
12. Grond M, Stenzel C, Schmulling S, et al. Early intravenous thrombolysis for acute ischemic stroke in a community-based approach. Stroke
13. Trouillas P, Nighoghossian N, Derex L, et al. Thrombolysis with intravenous rtPA in a series of 100 cases of acute carotid territory stroke. Determination of etiological, topographic and radiological outcome factors. Stroke
14. Albers GW, Bates VE, Clark WM, Bell R, Verro P, Hamilton SA. Stanford Stroke Center: Intravenous tissue-type plasminogen activator for treatment of acute stroke: the Standard Treatment with Alteplase to Reverse Stroke (STARS) study. JAMA
15. Hossmann KA. Viability thresholds and the penumbra of focal ischemia. Ann Neurol