Treatment of Acute Ischaemic Stroke

Summary

Prospective randomised clinical trials have shown how to apply fibrinolytics, heparin and aspirin to the benefit of a patient with acute ischaemic stroke. Fibrinolytics given intravenously are beneficial up to three and in some patients up to six hours after onset of stroke symptoms. The most common and most devastating side effect is intracranial haemorrhage. Therefore, factors unique to each patient which might increase the bleeding tendency have to be considered carefully before fibrinolytics are given. Intraarterial application of fibrinolytics may be safer and more efficacious. However, this treatment modality is restricted to large medical centres and results of randomised trials are not published. If fibrinolytics are contraindicated or if a patient arrives after the time window for fibrinolysis aspirin should be given. Aspirin has a modest beneficial effect and does not increase the risk of an intracranial haemorrhage significantly. Heparin should be avoided in the acute stage of stroke. Its beneficial effect is annihilated by an excess of intra- and extracranial haemorrhages.

Zusammenfassung

Prospektiv randomisierte Studien haben den Nutzen und die Gefahren von Fibrinolytika, Heparin und Aspirin beim akuten Hirnschlag gezeigt. Intravenös verabreichte Fibrinolytika (rt-PA) wirken innerhalb der ersten 3 Stunden und in Einzelfällen auch mehr als 3 Stunden nach Auftreten der klinischen Symptome günstig. Die gefürchtetsten Nebenwirkungen sind intrakranielle Blutungen, und deshalb sollten Faktoren, die auf ein erhöhtes Blutungsrisiko hinweisen, vor der Behandlung sorgfältig gesucht werden. Die intraarterielle Fibrinolyse ist möglicherweise sicherer und wirksamer als die intravenöse. Randomisierte Studien, die dies beweisen, fehlen jedoch, und ausserdem ist die intraarterielle Fibrinolyse nur an grösseren medizinischen Zentren möglich. Bei Kontraindikationen gegen Fibrinolytika oder wenn der Patient erst nach 3 bis 6 Stunden behandelt werden kann, sollte Aspirin gegeben werden. Aspirin hat einen bescheidenen günstigen Effekt und führt im Gegensatz zu Heparin zu keiner signifikanten Erhöhung der Blutungsrate. Eine volle Heparinisierung beim akuten ischämischen Hirninfarkt hat eine erhöhte Blutungsrate zur Folge.

Introduction

In situ thrombosis or embolic occlusion of a cerebral artery is by far the most common cause of ischaemic stroke. Both mechanisms result in focal blockage of the cerebral circulation and initiate an ischaemic cascade that ultimately destroys the neurons. If the blocked vessel is reopened early, oxygen supply and metabolism can be reestablished before the hypoperfused neurons die [1]. Early reperfusion has the potential to salvage much of the hypoperfused tissue. Reperfusion usually saves the border zone of the ischaemic tissue, the socalled penumbra, and in the best scenario even its core [2]. In the acute stage of a stroke it cannot be known in all patients whether thrombus has formed in situ or occlusion has resulted from arterioarterial or cardiogenic embolism. Fortunately this may not play any role for acute stroke treatment, because the goal of treatment is recanalisation both in local atherothrombosis and embolic occlusion of cerebral arteries.

Systemic (Intravenous) Fibrinolysis

Early trials of intravenous fibrinolysis for stroke using streptokinase or urokinase have been performed before the advent of computerised tomography (CT). At that time it was not possible to separate stroke due to ischaemia and stroke due to haemorrhage noninvasively. The results were discouraging [3,4]. Advances in diagnostic techniques and the success of coronary thrombolysis rekindled the interest in thrombolytic treatment of ischaemic stroke. Animal experiments showed that intravenous tissue plasminogen activator (t-PA) and intraarterial urokinase successfully reopened cerebral vessels and reduced neurological damage [5,6]. Soon thereafter the feasibility to recanalise occluded vessels using intraarterial urokinase was demonstrated in humans [7,8,9]. Recanalisation enhanced the chance of clinical improvement. This reawakened the interest of intravenous application of thrombolytics for stroke treatment, because thrombolysis with intravenous agents can begin much sooner after stroke than with intraarterial administration. Intravenous rt-PA given within 6 hours in two series of patients and one small randomised trial opened occluded arteries effectively and without a substantial increase of intracranial haemorrhages [10,11].

With this experience large-scale prospective randomised fibrinolysis trials for stroke treatment were initiated (Table 1) [12,13,14,15,16]. What have we learned from these trials?

First, time is crucial. The NINDS trial demonstrated that rt-PA is efficacious for treatment of stroke. The benefit was a nonsignificant reduction of mortality and a significant decrease of neurological damage. Per 100 patients treated 13 made a full recovery, who would have remained handicapped without fibrinolysis, and there were 4 fewer deaths. Unlike the other trials the time window for patient inclusion in NINDS was 3 hours only. In addition, both in AST and MAST-I patients who were treated within 3 hours had a better outcome than patients enrolled after 3 hours.

Table 1. Intravenous fibrinolysis trials for stroke.

Table 1. Intravenous fibrinolysis trials for stroke.

Second, CT findings are markers of the danger of intracranial haemorrhage. In ECASS only patients with hypodensity or sulcal effacement of less than one third of the middle cerebral artery territory showed a better outcome and were not at increased risk of intracranial haemorrhage. Therefore, intravenous fibrinolysis between 4 and 6 hours of onset of symptoms may be hazardous when these CT criteria are not considered. The target population analysis in ECASS suggests that the time window from 4 to 6 hours should be studied further.

Third, there is an interaction between fibrinolytic agents and aspirin or heparin. In AST all patients in the SK and placebo groups received aspirin. In MAST-E most of the patients in both groups received a full dose of intravenous heparin. In MAST-I a 2x2 factor design was used. Patients assigned to streptokinase and aspirin were at higher risk of intracranial haemorrhage and dying than patients treated with streptokinase only. In ECASS low dose subcutaneous heparin was allowed any time and the use of intravenous heparin after 24 hours was at the discretion of the local investigators. Only the protocol of NINDS required that no anticoagulants or antiplatelet agents be given for 24 hours after treatment with rt-PA. I.e. all trials stopped early for safety reasons were using aspirin or heparin in addition to streptokinase in most of the patients, and the only trial with unequivocally positive results was using rt-PA only. These results suggest that aspirin or heparin should be avoided for at least 24 hours after thrombolysis.

Fourth, individual patient data have to be evaluated before the thrombolytic agent is applied. Exclusion criteria for fibrinolysis as used in various trials have to be followed strictly. Patients with a history of hypertension or vascular leukencephalopathy are at increased risk for bleeding, and the risk also increases with age. Data for children are missing.

Fifth, both rt-PA and streptokinase are effective thrombolytic agents for stroke treatment. At present, rt-PA is the preferred agent. Streptokinase as used in MAST-E, MAST-I and AST is not safe. However, the use of streptokinase should and will be pursued further using different doses and trial designs [17]. To summarise, at present there is sufficient evidence to support the use of intravenous rt-PA for acute ischaemic stroke in a patient who fulfils the inclusion and exclusion criteria from the NINDS study [18,19]. The NINDS study has shown efficacy of rt-PA for stroke. If rt-PA given within 3 hours after stroke will be effective under routine conditions in community hospitals is unknown [20]. Therefore it is recommended that the therapy be performed by and limited to neurologists and other specialists with expertise in neurologic emergencies and management of stroke patients and the specific problems associated with stroke [18].

Local Intraarterial Fibrinolysis

Local intraarterial fibrinolysis (LIF) has been shown to recanalise cerebral arteries efficaciously and improve neurological outcome [7,8,9,22,23]. Trials are under way. However, results of completed prospective randomised trials are not available. The PROACT-Trial, a double blind prospective trial, compares recombinant pro-urokinase (r-ProUK) and placebo. First results of 40 patients show, that r-ProUK compared to placebo increased the cure rate by 18 percent [24]. There was no difference of the haemorrhage rate.

The technique of LIF is restricted to large medical centres where an interventional neuroradiologist is part of a stroke team and performs the procedure. A disadvantage of LIF is the loss of time until the intraarterial catheter is placed close to the thrombus or embolus occluding the cerebral vessel. However, there are advantages.

First, prognosis of stroke patients largely depends on the location and extent of vascular occlusion. The diagnostic angiography which has to be performed before LIF allows better classification of patients. It also serves to monitor if recanalisation occurs or not. In carotid artery territory stroke, patients with intracranial carotid bifurcation occlusions (carotid T occlusions) have the worst prognosis. Successful recanalisation is rare. More than half of the patients die in spite of LIF, mostly due to spaceoccupying oedema [25]. Less jeopardised are patients with middle cerebral artery main stem (Ml) occlusions. In Zeumer’s series 3 of 60 patients died and 31 of 60 were cured (Barthel Index > 90). In our own unpublished series 5 of 16 patients, who showed a hyperdense middle cerebral artery on CT as a sign of Ml occlusion and who were treated within 6 hours, had a good outcome (Rankin Scale 0—1). For comparison, intravenous rt-PA given within 90 minutes resulted only in one good outcome of 18 patients with a hyperdense middle cerebral artery on CT [24]. Patients with middle cerebral artery branch occlusions have the best prognosis and a minimal mortality. Three of 4 patients recover completely. In basilar artery occlusion recanalisation increases the chance of survival as well [27]. Young patients with monosegmental embolic occlusion of the basilar artery seem to profit most from LIF.

Second, the haemorrhage rate does not seem to be increased with LIF and less than with systemic thrombolysis. Théron and co-workers noted only one symptomatic haemorrhage among 142 patients treated with LIF. In our own series of 43 patients there were two symptomatic haemorrhages. One of the patients died, the other remained disabled but independent (Rankin Scale 2). In LIF the amount of fibrinolytics used is rarely as high as to produce a systemic lytic effect. This may account for the lower haemorrhage rate than in systemic fibrinolysis.

Third, in acute myocardial infarction percutaneous transluminal angioplasty has been shown to be as efficacious as systemic fibrinolysis. Mechanical recanalisation without or in addition to the administration of fibrinolytics may be advantageous also in the cerebral circulation.

Fourth, when LIF is not successful to reopen a vessel, the performed angiogram helps to decide on the further patient management. When the middle cerebral artery main stem (Ml) before the origin of the lenticulostriate arteries cannot be recanalised patients are at risk of further neurological deterioration and of dying because of space occupying oedema. In these patients hemicraniectomy can be performed early to preserve their lives and to reduce neurological disability [28]. However, the indication of hemicraniectomy for space-occupying oedema is controversial and most of these patients will remain disabled in spite of craniotomy.

LIF has been used successfully for treatment of eye infarction, but prospective randomised trials are missing [24,30]. Five of 17 patients treated within 6 hours after sudden monocular blindness recovered vision to 0.7 or more. The best outcome of vision of 15 controls was 0.2. To summarise, several open and a small randomised study demonstrate that LIF is efficacious for treatment of acute ischaemic stroke. It may be more efficacious and safer than systemic fibrinolysis, but this is not proven. Patients with middle cerebral artery main stem or basilar artery occlusions, who can be recognised on CT because of a dense artery sign, presumably profit most from LIF. However, LIF is less ideal for patients with branch occlusions, who may have as good an outcome as when treated systemically. The best way to deliver fibrinolysis treatment for acute stroke is to organise a stroke team and an algorithm according to the local circumstances to streamline all the processes which are involved into rapid management of stroke victims [31].

Aspirin

When a patient arrives too late at the stroke service to be treated with fibrinolytics, aspirin is the next best alternative. The effects of aspirin on acute stroke has been studied in the International Stroke Trial (IST), the Chinese Acute Stroke Trial (CAST) and MAST-I (MAST-I) [32,33]. Aspirin given within 48 hours reduces mortality and stroke recurrence within the first weeks after stroke (Table 2 and Table 3). The sooner aspirin was given the better was its effect in MAST-I. Aspirin does not increase the risk of cerebral haemorrhage significantly. However, before its administration cerebral haemorrhage should be ruled out by CT. The favorable effect of aspirin seen in the first weeks carries on. It can also be identified six months after the stroke (Table 4): In IST there were 14 fewer deaths or dependent persons per 1000 patients treated. In addition, 10 per 1000 patients reported complete recovery from their strokes who would have remained handicapped without aspirin. At first sight this effect seems to be modest and negligible and it is much lower than the effect of fibrinolysis. However, considering how many strokes occur in industrialised countries each year, even a small effect adds up to many fewer deaths and dependencies.

Table 2. International Stroke Trial (IST) - Main outcome events at 14 days.

Table 2. International Stroke Trial (IST) - Main outcome events at 14 days.

Table 3. Aspirin: short term effects in IST and CAST.

Table 3. Aspirin: short term effects in IST and CAST.

Table 4. International Stroke Trial (1ST)-main outcome events at 6 months.

Table 4. International Stroke Trial (1ST)-main outcome events at 6 months.

Heparin

The effect of heparin (unfractionated standard heparin) was assessed in IST in two different doses: 5000 U or 12500 U subcutaneously twice per day. At two weeks the beneficial effect of heparin was more than offset by an excess of intra- and extracranial haemorrhages (Table 2). There was no subgroup of patients who showed any clear benefit from heparin. Even in patients with atrial fibrillation who had a higher rate of stroke recurrence the excess of haemorrhages annihilated the beneficial effect (Table 5). After IST we use heparin very cautiously. We still use or consider immediate heparinisation in the following situations: cerebral sinuvenous thrombosis, dissection of extracranial arteries, high grade stenosis of cerebral arteries that are not amenable to immediate surgical or endovascular treatment, or intraluminal thrombus that has been visualised. In cardiac embolism it is unclear how long anticoagulation should be delayed. In patients who suffered a stroke from atrial fibrillation we give aspirin and start oral anticoagulation after one week.

Table 5. International Stroke Trial (1ST)-3169 patients with atrial fibrillation. Outcome at I 4 days.

Table 5. International Stroke Trial (1ST)-3169 patients with atrial fibrillation. Outcome at I 4 days.

Low Molecular Weight Heparin

In a prospective randomised trial low molecular weight heparin showed beneficial effects [34]. Mortality was reduced and functional outcome improved. This study was relatively small. Therefore the results need confirmation by a larger study before low molecular weight heparin can generally be recommended for stroke treatment. Orgaran, which acts in a similar way as heparin, has no beneficial effect for stroke [35].

Conclusion

In conclusion, the potential for stroke treatment has been demonstrated. Research to improve efficacy and safety of thrombolysis and to find treatments which can be combined with thrombolysis will be pursued. In addition, to bring thrombolysis to a larger number of stroke victims stroke has to be considered as an emergency with a very high priority. Efforts to make the public aware of stroke symptoms may also have to increase.