Glycoprotein IIb-IIIa inhibitors for acute ischaemic stroke

  • Conclusions changed
  • Review
  • Intervention

Authors


Abstract

Background

Glycoprotein (GP) IIb-IIIa inhibitors are antiplatelet agents that act by antagonising GP IIb-IIIa receptors on the platelet surface and block the final common pathway to platelet aggregation by preventing the binding of fibrinogen molecules that form bridges between adjacent platelets. Thus, GP IIb-IIIa inhibitors could favour endogenous thrombolysis by reducing thrombus growth and preventing thrombus re-formation through competitive inhibition with fibrinogen and, due to their mechanism of action, are likely to have a more profound antiplatelet effect with more rapid onset than conventional antiplatelet agents, such as aspirin or clopidogrel. Currently used in clinical practice for the treatment of individuals with acute coronary syndromes and during coronary angioplasty, GP IIb-IIIa inhibitors could also be useful for the treatment of people with acute ischaemic stroke.

Objectives

To assess the use of GP IIb-IIIa inhibitors in people with acute ischaemic stroke to evaluate whether such treatments (1) reduce the proportion of patients who die or remain dependent, and (2) are sufficiently safe for general use. We wished to examine the effects GP IIb-IIIa inhibitors alone or in combination with thrombolytic agents.

Search methods

We searched the Cochrane Stroke Group trials register (last searched 10 June 2013), MEDLINE (1966 to June 2013), EMBASE (1980 to June 2013), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 5, 2013), and major ongoing clinical trials registers (June 2013). We also searched reference lists and contacted trial authors and pharmaceutical companies.

Selection criteria

We aimed to analyse unconfounded randomised controlled trials (RCTs) of GP IIb-IIIa inhibitors in the treatment of people with acute ischaemic stroke. Only individuals who started treatment within six hours of stroke onset were included.

Data collection and analysis

We independently selected trials for inclusion, assessed trial quality and extracted the data.

Main results

We included four trials involving 1365 participants. Three trials compared the intravenous GP IIb-IIIa inhibitor Abciximab with intravenous placebo (1215 participants) and one trial compared the intravenous GP IIb-IIIa inhibitor Tirofiban with intravenous aspirin (150 participants). Treatment with either of these GP IIb-IIIa inhibitors did not significantly reduce long-term death or dependency (odds ratio (OR) 0.97, 95% confidence interval (CI) 0.77 to 1.22, for the comparison between Abciximab and placebo; OR 1.00, 95% CI 0.52 to 1.92, for the comparison between Tirofiban and aspirin) and had no effect on deaths from all causes (OR 1.08, 95% CI 0.77 to 1.53, for the comparison between Abciximab and placebo; OR 1.00, 95% CI 0.35 to 2.82, for the comparison between Tirofiban and aspirin). Abciximab was associated with a significant increase in symptomatic intracranial haemorrhage (OR 4.6, 95% CI 2.01 to 10.54) and with a non-significant increase in major extracranial haemorrhage (OR 1.81, 95% CI 0.96 to 3.41), whereas the only small trial comparing Tirofiban with aspirin showed no increased risk of bleeding complications with Tirofiban (OR 0.32, 95% CI 0.03 to 3.19, for symptomatic intracranial haemorrhage; OR 3.04, 95% CI 0.12 to 75.83, for major extracranial haemorrhages). There was no significant inconsistency across the studies.

Authors' conclusions

The available trial evidence showed that, for individuals with acute ischaemic stroke, GP IIb-IIIa inhibitors are associated with a significant risk of intracranial haemorrhage with no evidence of any reduction in death or disability in survivors. These data do not support their routine use in clinical practice. The conclusion is driven by trials of Abciximab, which contributed 89% of the total number of study participants considered.

Résumé scientifique

Inhibiteurs de la glycoprotéine IIb-IIIa dans l'accident vasculaire cérébral (AVC) ischémique aigu

Contexte

Les inhibiteurs de la glycoprotéine (GP) IIb-IIIa sont des agents antiplaquettaires qui agissent en antagonisant les récepteurs de la GP IIb-IIIa à la surface des plaquettes et bloquent le chemin commun final de l'agrégation plaquettaire en empêchant la fixation des molécules fibrinogènes qui forment des ponts entre les plaquettes adjacentes. Par conséquent, les inhibiteurs de la GP IIb-IIIa pourraient favoriser la thrombolyse endogène en réduisant la croissance des thrombus et en prévenant leur réformation par inhibition compétitive avec le fibrinogène et, en raison de leur mécanisme d'action, sont susceptibles d'avoir un effet antiplaquettaire plus profond avec un début plus rapide que les antiagrégants plaquettaires classiques, tels que l'aspirine ou le clopidogrel. Actuellement utilisés dans la pratique clinique pour le traitement des personnes atteintes de syndromes coronariens aigus et pendant l'angioplastie coronaire, les inhibiteurs de la GP IIb-IIIa pourraient également être utiles pour le traitement des personnes atteintes d'AVC ischémique aigu.

Objectifs

Évaluer l'utilisation des inhibiteurs de la GP IIb-IIIa chez les personnes atteintes d'AVC ischémique aigu afin d'évaluer si ces traitements (1) réduisent la proportion de patients qui décèdent ou restent dépendants, et (2) sont suffisamment sûrs pour une utilisation générale. Nous souhaitions examiner les effets des inhibiteurs de la GP IIb-IIIa seuls ou en association avec les agents thrombolytiques.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre des essais du groupe Cochrane sur les accidents vasculaires cérébraux (dernière recherche le 10 juin 2013), MEDLINE (de 1966 à juin 2013), EMBASE (de 1980 à juin 2013), le registre Cochrane des essais contrôlés (CENTRAL) (Bibliothèque Cochrane, numéro 5, 2013) et les principaux registres d'essais cliniques en cours (juin 2013). Nous avons également consulté les références bibliographiques et contacté les auteurs des essais et des sociétés pharmaceutiques.

Critères de sélection

Notre objectif était d'analyser les essais contrôlés randomisés (ECR) non biaisés sur les inhibiteurs de la GP IIb-IIIa dans le traitement des personnes atteintes d'AVC ischémique aigu. Seuls les patients commençant le traitement dans les six heures suivant le déclenchement de l'AVC étaient inclus.

Recueil et analyse des données

Nous avons indépendamment sélectionné les essais à inclure, évalué leur qualité et extrait les données.

Résultats principaux

Nous avons inclus quatre essais portant sur 1 365 participants. Trois de ces essais comparaient l'administration par voie intraveineuse de l'inhibiteur de la GP IIb-IIIa abciximab à un placebo par voie intraveineuse (1 215 participants) et un essai comparait l'administration par voie intraveineuse de l'inhibiteur de la GP IIb-IIIa tirofiban à l'aspirine par voie intraveineuse (150 participants). Le traitement par l'un de ces inhibiteurs de la GP IIb-IIIa n'a pas permis de réduire significativement les décès ou la dépendance à long terme (rapport des cotes (RC) 0,97, intervalle de confiance (IC) à 95 % de 0,77 à 1,22, pour la comparaison entre abciximab et placebo ; RC 1,00, IC à 95 % 0,52 à 1,92, pour la comparaison entre tirofiban et aspirine) et n'avait aucun effet sur les décès toutes causes confondues (RC 1,08, IC à 95 % 0,77 à 1,53, pour la comparaison entre abciximab et placebo ; RC 1,00, IC à 95 % 0,35 à 2,82, pour la comparaison entre tirofiban et aspirine). L'abciximab était associé à une augmentation significative de l'hémorragie intracrânienne symptomatique (RC 4,6, IC à 95 % 2,01 à 10,54) et à une augmentation non significative de l'hémorragie extracrânienne majeure (RC 1,81, IC à 95 % 0,96 à 3,41), tandis que le seul essai de petite taille comparant le tirofiban à l'aspirine n'a montré aucune augmentation du risque de complications hémorragiques avec le tirofiban (RC 0,32, IC à 95 % 0,03 à 3,19, pour l'hémorragie intracrânienne symptomatique ; RC 3,04, IC à 95 % 0,12 à 75,83, pour l'hémorragie extracrânienne majeure). Il n'y avait aucune incohérence significative entre les études.

Conclusions des auteurs

Les preuves disponibles issues d'essais ont montré que, pour les individus atteints d'AVC ischémique aigu, les inhibiteurs de la GP IIb-IIIa sont associés à un risque significatif d'hémorragie intracrânienne, sans preuve aucune d'une réduction des décès ou de l'invalidité chez les survivants. Ces données ne soutiennent pas leur utilisation systématique dans la pratique clinique. Cette conclusion est motivée par les essais sur l'abciximab, qui ont fourni 89 % du nombre total de participants pris en compte ici.

Plain language summary

Glycoprotein IIb-IIIa inhibitors for acute ischaemic stroke

Question: We wanted to evaluate the safety and effectiveness of GP IIb-IIIa inhibitors, alone or in combination with thrombolytic agents, in individuals with acute ischaemic stroke.

Background: Most strokes are due to a sudden blockage of an artery in the brain (this type of stroke is called an ischaemic stroke). In most ischaemic strokes, the blockage is caused by a blood clot. Clots form because platelets circulating in the blood in determined conditions aggregate. Glycoprotein (GP) IIb-IIIa inhibitors are potent, fast and selective blockers of the platelet aggregation and, thus, might help to dissolve blood clots, prevent new clots from forming, and so improve the blood supply to the brain. It is possible, if the drug is given within a few hours of the start of a stroke, that this would reduce brain damage and improve the chances of the individual making a good recovery. However, GP IIb-IIIa inhibitors can also cause bleeding in the brain, which are associated with poorer outcomes. This review aimed to evaluate whether GP IIb-IIIa inhibitors, administered within six hours from stroke onset, reduce the proportion of dead or dependent individuals.

Study characteristics: We identified four trials with a total of 1365 participants in searches conducted up to June 2013: three trials compared the intravenous GP IIb-IIIa inhibitor Abciximab with intravenous placebo and one trial compared the intravenous GP IIb-IIIa inhibitor Tirofiban with intravenous aspirin.

Key results: The results showed that GP IIb-IIIa inhibitors cause bleeding in the brain, and that this complication offset any benefits. Therefore, GP IIb-IIIa inhibitors should be avoided in people with acute ischaemic stroke.

Quality of the evidence: Overall, the studies were considered at low risk of bias.

Résumé simplifié

Les inhibiteurs de la glycoprotéine IIb-IIIa dans l'accident vasculaire cérébral (AVC) ischémique aigu

Question : Nous avons voulu évaluer l'innocuité et l'efficacité des inhibiteurs de la glycoprotéine IIb-IIIa, seuls ou en association avec des agents thrombolytiques, chez les personnes atteintes d'un AVC ischémique aigu.

Contexte : La plupart des AVC sont dus à un blocage soudain d'une artère dans le cerveau (on les appelle alors AVC ischémiques). Dans la majorité des AVC ischémiques, c'est un caillot sanguin qui est à l'origine du blocage. Des caillots se forment parce que les plaquettes circulant dans le sang se rassemblent dans certaines conditions. Les inhibiteurs de la glycoprotéine (GP) IIb-IIIa sont des bloquants puissants, rapides et sélectives de l'agrégation plaquettaire et, par conséquent, ils pourraient aider à dissoudre les caillots de sang, prévenir la formation de nouveaux caillots et donc améliorer l'apport sanguin vers le cerveau. Il est possible, si le médicament est administré dans les heures suivant le déclenchement de l'AVC, que cela réduise les lésions cérébrales et améliore les chances d'une bonne récupération du patient. Cependant, les inhibiteurs de la GP IIb-IIIa peuvent également entraîner des saignements dans le cerveau, qui sont associés à des résultats cliniques défavorables. Cette revue visait à évaluer si les inhibiteurs de la GP IIb-IIIa, administrés dans les six heures suivant le déclenchement de l'AVC, réduisaient la proportion de patients décédés ou dépendants.

Caractéristiques des études : Nous avons identifié quatre essais portant sur un total de 1 365 participants dans les recherches menées jusqu'à juin 2013. Trois de ces essais comparaient l'administration par voie intraveineuse de l'inhibiteur de la GP IIb-IIIa abciximab à un placebo par voie intraveineuse et un essai comparait l'administration par voie intraveineuse de l'inhibiteur de la GP IIb-IIIa tirofiban à l'aspirine par voie intraveineuse.

Résultats principaux : Les résultats ont montré que les inhibiteurs de la GP IIb-IIIa entraînaient des saignements dans le cerveau, et que cette complication annulait tout bénéfice éventuel. Par conséquent, les inhibiteurs de la GP IIb-IIIa devraient être évités chez les personnes atteintes d'AVC ischémique aigu.

Qualité des preuves : Dans l'ensemble, les études étaient considérées à faible risque de biais.

Notes de traduction

Traduit par: French Cochrane Centre 8th July, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

Laienverständliche Zusammenfassung

Glykoprotein-IIb-IIIa-Inhibitoren bei akutem ischämischen Schlaganfall

Frage: Wir wollten die Sicherheit und die Alltagswirksamkeit von Glykprotein(GP)-IIb-IIIa-Inhibitoren allein oder in Kombination mit thrombolytischen Wirkstoffen bei Menschen mit akutem ischämischem Schlaganfall überprüfen.

Hintergrund: Die meisten Schlaganfälle geschehen aufgrund einer plötzlichen Blockierung einer Arterie im Gehirn (dies wird ischämischer Schlaganfall genannt). Bei den meisten ischämischen Schlaganfällen wird diese Blockierung durch ein Blutgerinnsel ausgelöst. Gerinnsel entstehen, weil sich Blutplättchen unter bestimmten Umständen zusammenballen. GP-IIb-IIIa-Inhibitoren sind wirksame, schnelle und selektive Blocker der Zusammenballung von Blutplättchen und könnten daher helfen Blutgerinnsel aufzulösen, der neuen Bildung von Blutgerinnseln vorzubeugen und damit die Blutversorgung des Gehirns zu verbessern. Es ist möglich, dass wenn das Medikament innerhalb von wenigen Stunden nach Beginn eines Schlaganfalls verabreicht wird, Hirnschäden verringert werden und die Wahrscheinlichkeit der Genesung der einzelnen Patienten verbessert wird. Allerdings können GP-IIb-IIIa-Inhibitoren auch zu Blutungen im Gehirn führen, die mit schlechteren Endpunkten assoziiert sind. Dieser Review zielte darauf ab, festzustellen, ob GP-IIb-IIIa-Inhibitoren, die innerhalb von sechs Stunden ab Beginn des Schlaganfalls verabreicht werden, den Anteil der verstorbenen oder von fremder Hilfe abhängigen Personen vermindert.

Studieneigenschaften: Wir identifizierten vier Studien mit insgesamt 1365 Teilnehmern bei der Suche, die bis Juni 2013 durchgeführt wurde. Drei Studien verglichen den intravenösen GP-IIb-IIIa-Inhibitor Abciximab mit intravenösem Placebo und eine Studie verglich den intravenösen GP-IIb-IIIa-Inhibitor Tirofiban mit intravenösem Aspirin.

Die wichtigsten Ergebnisse: Die Ergebnisse zeigten, dass GP-IIb-IIIa-Inhibitoren Blutungen im Gehirn verursachen, und dass diese Komplikation alle Vorteile überwiegt. Daher sollten GP IIb-IIIa-Inhibitoren bei Patienten mit akutem ischämischem Schlaganfall vermieden werden.

Qualität der Evidenz: Insgesamt wurden die Studien mit einem niedrigen Risiko für Bias bewertet.

Anmerkungen zur Übersetzung

I. Töws und C. Bollig, Koordination durch Cochrane Schweiz.

Background

Currently, there is evidence that two pharmacological therapies are effective in treating individuals with acute ischaemic stroke: thrombolysis using recombinant tissue plasminogen activators (rt-PA) and the antiplatelet agent aspirin. The risks and benefits of these two therapies are summarised in two systematic reviews of randomised controlled trials (RCTs) produced by the Cochrane Stroke Group (Sandercock 2008; Wardlaw 2009). Thrombolytic therapy with intravenous rt-PA reduces the risk of death and disability, but not the risk of death alone, is associated with a definite risk of symptomatic cerebral haemorrhage and is deliverable to only a minority of people with stroke (Wardlaw 2009). Aspirin, which is indicated within 48 hours of the onset of a presumed ischaemic stroke, can be administered to the majority of individuals without a major risk of early haemorrhagic complications, but its benefit on an individual level is modest (for every 1000 people treated 13 more were alive and independent at the end of follow up) (Sandercock 2008).

Most of the data in the Cochrane review of antiplatelet agents referred to drugs that were given orally (Sandercock 2008); hence, the speed of onset of any antiplatelet effect may have been quite slow. There is therefore a strong case for looking for antithrombotic agents that are less risky than rt-PA but, like aspirin, can be given to the majority of individuals who experience a stroke. Antiplatelet agents that can be given intravenously and have a rapid onset of effect would also be advantageous. Glycoprotein (GP) IIb-IIIa inhibitors, which are already used in clinical practice for the treatment of individuals with coronary disease, potentially fulfil these criteria. These agents were developed on the premise that blockade of the GP IIb-IIIa integrin present in platelets, the final pathway of platelet aggregation, would be more effective than blockade of the activation single pathway of platelet aggregation, the mechanism of action of conventional agents such as aspirin, ticlopidine, or clopidogrel. The GP IIb-IIIa inhibitor class includes the human/murine monoclonal antibody Abciximab, the synthetic peptide Eptifibatide, and the synthetic non-peptides Tirofiban, Lamifiban, Xemilofiban, Orbofiban and Sibrafiban (Bogousslavsky 2001). Currently, GP IIb-IIIa inhibitors are not licensed for any indication in cerebrovascular disease. However, these agents are approved for use, in addition to aspirin and heparin, during percutaneous coronary revascularisation (i.e. coronary angioplasty with or without stent implantation), in individuals with or without an acute coronary syndrome, and in the initial medical management of those with unstable angina or myocardial infarction (MI). Their effects have been analysed in a Cochrane review (Bosch 2010). In individuals undergoing percutaneous coronary revascularisation, the periprocedural parenteral administration of GP IIb-IIIa inhibitors markedly reduced the risk of death or MI, albeit at the price of a moderate increase in the risk of severe bleeding. In people with unstable angina or MI, the risk of death or MI was only slightly reduced, the risk of death alone was not reduced and the risk of severe bleeding was slightly increased with these agents. Because it has been hypothesised that they can work to increase reperfusion of the ischaemic brain, GP IIb-IIIa inhibitors have been investigated for the treatment of individuals with acute ischaemic stroke.

Description of the condition

Stroke is the second most common cause of death and the leading cause of disability in developed countries. Ischaemic stroke (i.e. stroke caused by a sudden blockage of an artery in the brain, usually by a blood clot) is the most common type of stroke, and the quick restoration of blood flow, as soon as the blockage occurs, is crucial for a good long-term outcome.

Description of the intervention

We aimed to evaluate the effect of GP IIb-IIIa inhibitors administered early after an ischaemic stroke, either alone or combined with thrombolytic agents, based on the hypothesis that they may favour clinical improvement, aiding reperfusion of the brain ischaemic area.

How the intervention might work

GP IIb-IIIa inhibitors are potent antiplatelet agents that act quickly by antagonising GP IIb-IIIa receptors on the platelet surface where fibrinogen molecules bind, forming bridges between adjacent platelets. Because GP IIb-IIIa inhibitors inhibit the final pathway of platelet aggregation, they may be more effective than conventional antiplatelet agents, such as aspirin and clopidogrel, which interact with platelet aggregation with different mechanisms involving single pathways of platelet aggregation. This class of drug may favour endogenous thrombolysis by reducing thrombus growth and preventing thrombus re-formation and, therefore, they may aid reperfusion and clinical improvement if administered early after an ischaemic stroke.

Why it is important to do this review

GP IIb-IIIa inhibitors, which are already used in clinical practice for the treatment of individuals with coronary disease, have also been studied for use in the treatment of those with ischaemic stroke. A systematic review of controlled clinical trials appeared necessary in order to assemble all the available data by which to evaluate the potential efficacy and safety of these agents in individuals with acute ischaemic stroke. The previous version of this review (Ciccone 2006) found that there was not enough evidence to draw conclusions regarding the efficacy or safety of GP IIb-IIIa inhibitor therapy in people with acute ischaemic stroke, but results from ongoing trials were awaited.

Objectives

To assess the use of GP IIb-IIIa inhibitors in people with acute ischaemic stroke to evaluate whether such treatments (1) reduce the proportion of patients who die or remain dependent , and (2) are sufficiently safe for general use. We wished to examine the effects GP IIb-IIIa inhibitors alone or in combination with thrombolytic agents

Methods

Criteria for considering studies for this review

Types of studies

All apparently unconfounded RCTs, with or without blinding, of GP IIb-IIIa inhibitor(s) in individuals with acute ischaemic stroke. We included only truly randomised trials, whether or not outcomes were assessed blind to treatment allocation.

Types of participants

People of any age with definite acute ischaemic stroke, for whom treatment was started within six hours of symptom onset.

Types of interventions

We aimed to analyse the following comparisons separately:

  1. GP IIb-IIIa inhibitors versus placebo or open control;

  2. GP IIb-IIIa inhibitors versus rt-PA;

  3. GP IIb-IIIa inhibitors plus rt-PA versus rt-PA.

We considered any GP IIb-IIIa inhibitor, irrespective of the duration of treatment, dosage, or route of administration.

Types of outcome measures

Primary outcomes

Death or dependency: the number of participants who died or remained severely disabled (modified Rankin Scale 3 to 6 or equivalent) at the end of the scheduled follow up, where follow up was performed three months or longer after the stroke.

Secondary outcomes
  1. Death: number of deaths from any cause at the end of follow up.

  2. Symptomatic intracranial haemorrhage (ICH): number of participants with symptomatic fatal or non-fatal ICH (any new ICH or haemorrhagic transformation of a cerebral infarct that developed after randomisation, documented by computed tomography or magnetic resonance scanning, or at autopsy, and that caused clinically manifest deterioration of the neurological condition or resulted in death).

  3. Major extracranial haemorrhage: number of participants with any major extracranial haemorrhage (any bleeding that required transfusion or surgical intervention, or that caused a permanent disabling deficit; e.g. intra-ocular bleeding resulting in blindness).

  4. Any available information about safety in both experimental and control groups.

Search methods for identification of studies

Please refer to the 'Specialized register' section in the Cochrane Stroke Group module. We searched for trials in all languages and we planned to arrange the translation of relevant trial reports published in languages other than English.

Electronic searches

We searched the Cochrane Stroke Group trials register (last searched 10 June 2013), MEDLINE (Ovid) (1966 to June 2013) (Appendix 1), EMBASE (Ovid) (1980 to June 2013) (Appendix 2) and the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 5, 2013) (Appendix 3).

We developed comprehensive search strategies for each database with help of the Cochrane Stroke Group Trials Search Co-ordinator.

We also searched the following international trials registries (June 2013):

Searching other resources

In an effort to identify further published, unpublished and ongoing trials we also:

  1. searched relevant reference lists;

  2. contacted trial authors;

  3. contacted pharmaceutical companies (Centocor Inc., Eli Lilly & Co, Glaxo Wellcome, SmithKline Beecham, Merck Sharp and Dohme, Roche) (June 2005)

  4. used Google (www.google.co.uk/) to search for individual trial names and then searched online resources of the relevant trials (e.g. trial websites).

Data collection and analysis

Selection of studies

We independently read the titles and abstracts of the records identified by the electronic searches and excluded obviously irrelevant studies. We obtained the full text of the remaining papers and the same three authors selected studies for inclusion based on the predetermined review selection criteria. Disagreements were resolved by discussion and consultation with another review author if necessary.

Data extraction and management

Two pairs of review authors (CM and IS; IA and FC) independently extracted information for each included trial about the method of randomisation, blinding of outcome evaluators, balance of baseline prognostic factors (age, stroke severity and time from stroke onset), and whether all the randomised participants were accounted for in the analysis. The fifth author (AC) checked all the data. We contacted the authors of the trials if the above information was not available in the published reports. We used this information to evaluate quality.

In the same manner in which we extracted information for the assessment of study quality, we independently extracted data for outcome and safety assessments, according to measures defined in the Types of outcome measures section. We used intention-to-treat analysis, extracting the number of participants originally allocated to each treatment group irrespective of compliance. We resolved disagreements in the numbers extracted by discussion.

Assessment of risk of bias in included studies

The review authors independently assessed the quality of the trials included, according to Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), by assessing the following items:

  1. allocation concealment (selection bias);

  2. blinding (performance bias and detection bias);

  3. blinding of participants and personnel (performance bias);

  4. blinding of outcome assessment (detection bias);

  5. incomplete outcome data (attrition bias);

  6. selective reporting (reporting bias).

We resolved discrepancies and disagreements by discussion until consensus was reached.

Measures of treatment effect

The efficacy analyses were based on the results of the individual trials for death or dependency at three months or longer after the stroke. The safety analyses were based on the results for case fatality and the occurrence of symptomatic ICH and major extracranial haemorrhage. The effect measures were based on dichotomous data and were analysed using odds ratios (OR). We used Review Manager 5.2 (RevMan 2012) for all analyses.

Unit of analysis issues

We planned to look for the presence of an interaction between two GP IIb-IIIa inhibitors if data were available from RCTs of a 2 x 2 factorial design (i.e. GP IIb-IIIa inhibitor versus GP IIb-IIIa inhibitor plus thrombolytic agent versus thrombolytic agent versus neither).

Dealing with missing data

If information about excluded participants or participants lost to follow up after randomisation was unavailable from a publication, we decided to seek further information through correspondence with the trialists. Where data about these participants were unavailable, we performed a sensitivity analysis.

Assessment of heterogeneity

We quantified inconsistency across studies using the I2 statistic for heterogeneity (i.e. the variability in the effect estimates that is due to heterogeneity rather than sampling error). We considered a value greater than 50% as indicative of substantial heterogeneity, and calculated the confidence intervals (CI) for the I2 statistic according to Higgins 2002.

Assessment of reporting biases

We used funnel plots to assess reporting bias. We assessed funnel plots qualitatively.

Data synthesis

We calculated a weighted estimate of the treatment effect across trials (Mantel-Haenszel OR) using a fixed-effect model. For interpreting the results we used 95% CI.

Subgroup analysis and investigation of heterogeneity

If we found substantial heterogeneity on the efficacy analysis, our intention was to explore heterogeneity with the following analyses:

  1. comparison between trials with low and high risk of bias;

  2. comparison of the efficacy of GP IIb-IIIa inhibitors in individuals treated within three hours versus those treated between three and six hours of stroke onset;

  3. comparison of the efficacy of different GP IIb-IIIa inhibitors.

To test for differences between subgroups we planned to use the Deeks method, based on the test for heterogeneity (Deeks 2001).

As no substantial heterogeneity on the efficacy analysis was found, we did not carry out these planned analyses.

Sensitivity analysis

If participants were excluded or lost to follow up after randomisation, we sought further information by correspondence with the trialists. If the data about these participants remained unavailable, we provided a worst-case scenario analysis for the composite outcome of 'death or dependency' to ensure the significance of the results. In this analysis, it was assumed that those participants who were lost to follow-up in the treatment group had the worst outcome whereas those lost to follow up in the control group had the best outcome. If the effects of the primary and worst-case meta-analyses were in the same direction and magnitude, a definite conclusion about the treatment effectiveness could be made; otherwise no definite conclusion could be drawn.

Results

Description of studies

Results of the search

Following the previous version of this review (Ciccone 2006), for which the last search was carried out in 2005, a search of the following electronic databases, on June 2013, identified 1323 further references: Cochrane Stroke Group Trials Register (19 references), MEDLINE (181 references), EMBASE (1026 references), and CENTRAL (97 references). After eliminating non-relevant studies, carrying out additional searches on the major ongoing clinical trials registers, using Google to search for individual trial names and then having searched online resources of the relevant trials, we selected seven further possibly relevant studies. A PRISMA flowchart of study selection is shown in Figure 1. We did not find any new ongoing trials from the searches of the trials registries.

Figure 1.

Study flow diagram (2005 to 2013)

Included studies

The previous version of this review (Ciccone 2006) already included two studies (AbESTT 2000; Adams 1998) involving a total of 474 participants. In this review update, we added two new studies (SETIS 2010, and the three cohorts of the AbESTT-II study: AbESTT-II/Companion 5-6 h; AbESTT-II/Primary; AbESTT-II/Wake-Up), bringing the total to four included studies, involving 1365 participants. The AbESTT-II trial contained three study cohorts, which were evaluated independently in our analysis and denoted as primary (AbESTT-II/Primary), companion five to six hours (AbESTT-II/Companion 5-6 h), and wake-up stroke (AbESTT-II/Wake-Up), respectively. See Characteristics of included studies. Three of the four trials (AbESTT 2000; AbESTT-II/Companion 5-6 h; AbESTT-II/Primary; AbESTT-II/Wake-Up; Adams 1998) compared the intravenous GP IIb-IIIa inhibitor Abciximab with intravenous placebo (1215 participants) and one trial (SETIS 2010) compared the intravenous GP IIb-IIIa inhibitor Tirofiban with intravenous aspirin (150 participants). The characteristics of the included trials are summarised in the Characteristics of included studies section.

Excluded studies

We excluded 14 studies (CLEAR 2008; CLEAR-ER 2013; Davalos 2003; FAST-Study; Gahn 2004; Junghans 2002; Mandava 2005; Qureshi 2006; ROSIE-2; SaTIS 2011; Seitz 2003; Siebler 2003; Straub 2004; Velat 2006) for a variety of reasons (see the Characteristics of excluded studies table). In particular, we excluded SaTIS 2011 because participants were included within 22 hours of onset of symptoms and we were unable to obtain any information from the authors on the subgroup of participants treated within six hours. We excluded CLEAR 2008 and CLEAR-ER 2013, which investigated Eptifibatide plus rt-PA versus rt-PA, because they were confounded by two different dosages of rt-PA in the two treatment groups.

Risk of bias in included studies

Figure 2 and Figure 3 show a summary of the risk of bias in all included studies. A 'risk of bias' table for each study is provided in the Characteristics of included studies section.

Figure 2.

'Risk of bias' graph: review authors' judgements about each 'risk of bias' item presented as percentages across all included studies.

Figure 3.

'Risk of bias' summary: review authors' judgements about each 'risk of bias' item for each included study.

Allocation

The method of randomisation provided adequate concealment of allocation in all four trials.

Blinding

All the studies were conducted using a double-blind design.

Incomplete outcome data

No participant was excluded or lost to follow up in the Adams 1998 and SETIS 2010 studies.

In AbESTT 2000, one participant assigned to Abxicimab was withdrawn from the study because he immediately withdrew consent after randomisation; 15 others were lost during follow up. However, the authors stated that "all analyses for efficacy were performed on the intention-to-treat population, regardless of whether the study medication was administered. For patients with missing outcome data, the last observation was carried forward". The participant who withdrew from the study "would not permit use of any data, including those collected at baseline". We considered these explanations to be sufficient and did not contact the authors to seek further information. However, safety analyses were not performed on an intention-to-treat basis but on participants who received at least some study treatment. The same occurred in AbESTT-II/Companion 5-6 h, AbESTT-II/Primary, AbESTT-II/Wake-Up, where all analyses for efficacy were performed on the intention-to-treat population regardless of whether study medication was administered. whereas safety analyses "were performed on all patients who received at least some study treatment according to the actual treatment received" (three participants who were treated but not randomised were included in the safety analyses). We contacted the Executive Committee Members in order to obtain data on the intention-to-treat population, but since this information was unavailable within the time of the review update, we have had to rely on published information only.

Selective reporting

Reports of the studies were free from suggestions concerning selective outcome reporting.

Other potential sources of bias

We included the 43 participants in AbESTT-II/Wake-Up (Characteristics of included studies) even if the study did not comply rigorously with the six-hour time window inclusion criteria (see Types of participants).

Effects of interventions

Primary outcome

Death or dependency at end of follow up

Data for death or dependency (modified Rankin Scale 3 to 6 or equivalent) were available from all the studies for a total of 1365 participants. Treatment with a GP IIb-IIIa inhibitor did not significantly reduce the risk of this outcome, either in the comparison between Abciximab and placebo (AbESTT 2000; AbESTT-II/Companion 5-6 h; AbESTT-II/Primary; AbESTT-II/Wake-Up; Adams 1998) (OR 0.97, 95% CI 0.77 to 1.22) (Analysis 1.1) or in the comparison between Tirofiban and aspirin (SETIS 2010) (OR 1.00, 95% CI 0.52 to 1.92) (Analysis 2.1).

There was no significant inconsistency across the studies with regard to this outcome (I2 = 0%, P value = 0.41).

Assuming that participants who were lost to follow up in the treatment group had the worst outcome, whereas those who were lost to follow up in the placebo group had the best outcome, the worst-case scenario analysis did not significantly change the result (OR 1.05, 95% CI 0.84 to 1.32).

We did not perform a sensitivity analysis for the comparison between a GP IIb-IIIa inhibitor and aspirin, as there were no participants lost to follow up in the only study considered for this comparison (SETIS 2010).

Secondary outcomes

Death from all causes at end of follow up

All the included studies assessed this outcome for all participants randomised. Treatment with a GP IIb-IIIa inhibitor had no effect on the risk of death from all causes (OR 1.08, 95% CI 0.77 to 1.53, for the comparison between a GP IIb-IIIa inhibitor and placebo (Analysis 1.2); OR 1.00, 95% CI 0.35 to 2.82, for the comparison between a GP IIb-IIIa inhibitor and aspirin (Analysis 2.2)), and there was no significant heterogeneity (I2 = 10%, P value = 0.35).

Symptomatic intracranial haemorrhage

Data were available for 1208 participants who received at least some study treatment.

GP IIb-IIIa inhibitors were associated with a significant increase in the risk of symptomatic ICH. This result was driven by the trials comparing Abciximab with placebo (OR 4.6, 95% CI 2.01 to 10.54; I2 = 24%, 95% CI 0% to 85%) (Analysis 1.3), whereas the small trial comparing Tirofiban with aspirin (SETIS 2010) showed a non-significant reduction in the risk of symptomatic ICH with Tirofiban (OR 0.32, 95% CI 0.03 to 3.19) (Analysis 2.3).

Major extracranial haemorrhage

Data were available for 1201 participants who received at least some study treatment.

GP IIb-IIIa inhibitors were associated with a non-significant increase in the risk of major extracranial haemorrhage (OR 1.81, 95% CI 0.96 to 3.41; I2 = 0%, P value = 0.55, for the group of trials comparing a GP IIb-IIIa inhibitor with placebo (Analysis 1.4); OR 3.04, 95% CI 0.12 to 75.83, for the only trial comparing a GP IIb-IIIa inhibitor with aspirin (Analysis 2.4)).

Other

Thrombocytopenia is a potential adverse effect of Abciximab (Dasgupta 2000).

In Adams 1998, four of the 50 participants treated with Abciximab (7%) had moderate thrombocytopenia (between 50,000 and 100,000/mm3) during the first five days after treatment; and thrombocytopenia was associated with symptomatic intracranial or major extracranial haemorrhage. In AbESTT 2000, three of the 195 participants receiving Abciximab (1.5%) had thrombocytopenia: one had a platelet count of less than 20,000/µL, the others had platelet counts between 50,000 and 100,000/µL; four participants received a platelet transfusion. In the AbESTT-II study (AbESTT-II/Companion 5-6 h; AbESTT-II/Primary; AbESTT-II/Wake-Up), 11 of the 397 participants treated with Abciximab (2.8%) suffered from transitory thrombocytopenia that did not explain the excess risk of ICH. In SETIS 2010 there were no cases of thrombocytopenia.

No case of thrombocytopenia was reported in the control groups of the four studies considered.

Discussion

Summary of main results

We found a three-fold statistically significant excess risk of a symptomatic ICH associated with the use of GP IIb-IIIa inhibitors, which was not offset by any benefit on long-term death or disability. This result is derived mainly from trials of Abciximab, which contributed 89% of the total number of participants considered.

Overall completeness and applicability of evidence

The review indicates high-level evidence for a negative risk to benefit ratio for GP IIb-IIIa inhibitors in the treatment of individuals with acute ischaemic stroke. Such evidence is derived mainly from trials of Abciximab. However, data from the only trial of Tirofiban included in this review did not show any trend for a clinical benefit in 150 participants. If data for participants who started treatment with Tirofiban or placebo within six hours from stroke onset in SaTIS 2011 were available, additional information for this agent would be available. In the meantime, the present evidence does not support the conduct of further RCTs of GP IIb-IIIa inhibitors in individuals with acute ischaemic stroke.

We did not find any unconfounded studies investigating a GP IIb-IIIa inhibitor combined with intravenous rt-PA. We did not include either of the two studies of a GP IIb-IIIa inhibitor combined with intravenous rt-PA compared with intravenous rt-PA alone (CLEAR 2008; CLEAR-ER 2013) because both comparisons were confounded (both trials tested a lower dose of rt-PA in the combined group than in the control group).

Quality of the evidence

The included studies were double-blind RCTs that we judged at low risk of bias.

Potential biases in the review process

We aimed to include all relevant publications in this review by using multiple overlapping searches of a number of databases, and contacting trial authors, colleagues and researchers in the field of study. However, the possibility of missing small trials and trials published in journals with limited distribution cannot be totally excluded. Performing a funnel plot (Figure 4) revealed no suggestion of reporting biases, taking into consideration the small number of studies.

Figure 4.

Funnel plot of comparison for Analysis 1: Glycoprotein IIb-IIIa inhibitor versus control, outcome 1.1: Death or dependency at end of follow up.

We found a letter by Cheung 2000 reporting an ongoing study of Abciximab (see Characteristics of ongoing studies for details). We contacted the author for further information but did not receive a reply.

We included in the review the small AbESTT-II/Wake-Up cohort, which enrolled patients that could be treated within three hours of stroke present on awakening. Because stroke occurred during sleep in this cohort, we do not know how many patients complied with the six-hour time window inclusion criteria set for our review.

Data extracted from AbESTT (AbESTT 2000; AbESTT-II/Companion 5-6 h; AbESTT-II/Primary; AbESTT-II/Wake-Up) for symptomatic ICH and major extracranial haemorrhage were available only for participants who received at least some study treatment and not for the intention-to-treat population.

Agreements and disagreements with other studies or reviews

Since publication of the previous version of this review (Ciccone 2006) we have not identified any other reviews considering GP IIb-IIIa inhibitors in individuals with acute ischaemic stroke.

Authors' conclusions

Implications for practice

The available evidence does not support the use of GP IIb-IIIa inhibitors in people with acute ischaemic stroke.

Implications for research

The question of whether GP IIb-IIIa inhibitors can be combined with intravenous thrombolysis remains open. This question should be answered by trials that use similar doses of rt-PA in the two treatment arms.

Acknowledgements

We would like to thank Hazel Fraser, Managing Editor of the Cochrane Stroke Group, for sending us regular lists of trials identified in the Cochrane Stroke Group's Trials Register; Brenda Thomas, the Cochrane Stroke Group Trials Search Co-ordinator, for help in developing the search strategy; Stefano Ottolini from Centocor, for help in trial identification and in obtaining unpublished data for participants randomised within six hours of stroke onset in the Adams 1998 study; the two members of the Cochrane Consumer Network who commented on the review, for their useful insights; and Peter Sandercock, Lead Editor of this review, for supporting the process of the review update.

If you are aware of any additional trials, please let us know.

Data and analyses

Download statistical data

Comparison 1. Glycoprotein IIb-IIIa inhibitor versus placebo or open control
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Death or dependency at end of follow-up51215Odds Ratio (M-H, Fixed, 95% CI)0.97 [0.77, 1.22]
2 Total deaths from all causes at end of follow-up51215Odds Ratio (M-H, Fixed, 95% CI)1.08 [0.77, 1.53]
3 Symptomatic intracranial haemorrhage during follow-up51208Odds Ratio (M-H, Fixed, 95% CI)4.60 [2.01, 10.54]
4 Total major (including fatal) extracranial haemorrhage during follow-up51201Odds Ratio (M-H, Fixed, 95% CI)1.81 [0.96, 3.41]
Analysis 1.1.

Comparison 1 Glycoprotein IIb-IIIa inhibitor versus placebo or open control, Outcome 1 Death or dependency at end of follow-up.

Analysis 1.2.

Comparison 1 Glycoprotein IIb-IIIa inhibitor versus placebo or open control, Outcome 2 Total deaths from all causes at end of follow-up.

Analysis 1.3.

Comparison 1 Glycoprotein IIb-IIIa inhibitor versus placebo or open control, Outcome 3 Symptomatic intracranial haemorrhage during follow-up.

Analysis 1.4.

Comparison 1 Glycoprotein IIb-IIIa inhibitor versus placebo or open control, Outcome 4 Total major (including fatal) extracranial haemorrhage during follow-up.

Comparison 2. Glycoprotein IIb-IIIa inhibitor versus aspirin
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Death or dependency at end of follow-up1150Odds Ratio (M-H, Fixed, 95% CI)1.0 [0.52, 1.92]
2 Total deaths from all causes at end of follow-up1150Odds Ratio (M-H, Fixed, 95% CI)1.0 [0.35, 2.82]
3 Symptomatic intracranial haemorrhage during follow-up1150Odds Ratio (M-H, Fixed, 95% CI)0.32 [0.03, 3.19]
4 Total major (including fatal) extracranial haemorrhage during follow-up1150Odds Ratio (M-H, Fixed, 95% CI)3.04 [0.12, 75.83]
Analysis 2.1.

Comparison 2 Glycoprotein IIb-IIIa inhibitor versus aspirin, Outcome 1 Death or dependency at end of follow-up.

Analysis 2.2.

Comparison 2 Glycoprotein IIb-IIIa inhibitor versus aspirin, Outcome 2 Total deaths from all causes at end of follow-up.

Analysis 2.3.

Comparison 2 Glycoprotein IIb-IIIa inhibitor versus aspirin, Outcome 3 Symptomatic intracranial haemorrhage during follow-up.

Analysis 2.4.

Comparison 2 Glycoprotein IIb-IIIa inhibitor versus aspirin, Outcome 4 Total major (including fatal) extracranial haemorrhage during follow-up.

Appendices

Appendix 1. MEDLINE (Ovid) search strategy

Revised June 2013

1. cerebrovascular disorders/ or basal ganglia cerebrovascular disease/ or brain ischemia/ or exp brain infarction/ or hypoxia-ischemia, brain/ or carotid artery diseases/ or carotid artery thrombosis/ or carotid artery, internal, dissection/ or intracranial arterial diseases/ or cerebral arterial diseases/ or infarction, anterior cerebral artery/ or infarction, middle cerebral artery/ or infarction, posterior cerebral artery/ or exp "intracranial embolism and thrombosis"/ or exp stroke/ or vertebral artery dissection/
2. (isch?emi$ adj6 (stroke$ or apoplex$ or cerebral vasc$ or cerebrovasc$ or cva or attack$)).tw.
3. ((brain or cerebr$ or cerebell$ or vertebrobasil$ or hemispher$ or intracran$ or intracerebral or infratentorial or supratentorial or middle cerebr$ or mca$ or anterior circulation) adj5 (isch?emi$ or infarct$ or thrombo$ or emboli$ or occlus$ or hypoxi$)).tw.
4. 1 or 2 or 3
5. Platelet Glycoprotein GPIIb-IIIa Complex/ai, de, tu [Antagonists & Inhibitors, Drug Effects, Therapeutic Use]
6. ((glycoprotein$ iib$ or platelet iib$ or gp iib$ or iib$ receptor or fibrinogen receptor$) adj5 (inhib$ or block$ or antag$)).tw.
7. (abciximab or 7E3 or c7E3 or ReoPro or tirofiban or aggrastat or eptifibatide or integrilin or integrelin or intrifiban or sibrafiban or xubix or lamifiban or xemilofiban or orbofiban or fradafiban or lefradafiban or lotrafiban or roxifiban or ym-337 or ym337).mp.
8. 5 or 6 or 7
9. Randomized Controlled Trials as Topic/
10. random allocation/
11. Controlled Clinical Trials as Topic/
12. control groups/
13. clinical trials as topic/ or clinical trials, phase i as topic/ or clinical trials, phase ii as topic/ or clinical trials, phase iii as topic/ or clinical trials, phase iv as topic/
14. double-blind method/
15. single-blind method/
16. Placebos/
17. placebo effect/
18. Therapies, Investigational/
19. Drug Evaluation/
20. Research Design/
21. randomized controlled trial.pt.
22. controlled clinical trial.pt.
23. (clinical trial or clinical trial phase i or clinical trial phase ii or clinical trial phase iii or clinical trial phase iv).pt.
24. (random$ or RCT or RCTs).tw.
25. (controlled adj5 (trial$ or stud$)).tw.
26. (clinical$ adj5 trial$).tw.
27. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.
28. (quasi-random$ or quasi random$ or pseudo-random$ or pseudo random$).tw.
29. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.
30. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.
31. (placebo$ or sham).tw.
32. trial.ti.
33. (assign$ or allocat$).tw.
34. or/9-33
35. 4 and 8 and 34
36. exp animals/ not humans.sh.
37. 35 not 36

Appendix 2. EMBASE (Ovid) search strategy

Revised June 2013

1. cerebrovascular disease/ or brain infarction/ or brain stem infarction/ or cerebellum infarction/ or exp brain ischemia/ or carotid artery disease/ or exp carotid artery obstruction/ or cerebral artery disease/ or exp cerebrovascular accident/ or exp occlusive cerebrovascular disease/ or stroke patient/
2. (isch?emi$ adj6 (stroke$ or apoplex$ or cerebral vasc$ or cerebrovasc$ or cva or attack$)).tw.
3. ((brain or cerebr$ or cerebell$ or vertebrobasil$ or hemispher$ or intracran$ or intracerebral or infratentorial or supratentorial or middle cerebr$ or mca$ or anterior circulation) adj5 (isch?emi$ or infarct$ or thrombo$ or emboli$ or occlus$ or hypoxi$)).tw.
4. 1 or 2 or 3
5. exp fibrinogen receptor antagonist/ or fibrinogen receptor/ or fibrinogen receptor affecting agent/
6. ((glycoprotein$ iib$ or platelet iib$ or gp iib$ or iib$ receptor or fibrinogen receptor$) adj5 (inhib$ or block$ or antag$)).tw.
7. (abciximab or 7E3 or c7E3 or ReoPro or tirofiban or aggrastat or eptifibatide or integrilin or integrelin or intrifiban or sibrafiban or xubix or lamifiban or xemilofiban or orbofiban or fradafiban or lefradafiban or lotrafiban or roxifiban or ym-337 or ym337).tw.
8. 5 or 6 or 7
9. Randomized Controlled Trial/
10. Randomization/
11. Controlled Study/
12. control group/
13. clinical trial/ or phase 1 clinical trial/ or phase 2 clinical trial/ or phase 3 clinical trial/ or phase 4 clinical trial/ or controlled clinical trial/
14. Double Blind Procedure/
15. Single Blind Procedure/
16. triple blind procedure/
17. placebo/
18. "types of study"/
19. random$.tw.
20. (controlled adj5 (trial$ or stud$)).tw.
21. (clinical$ adj5 trial$).tw.
22. ((control or treatment or experiment$ or intervention) adj5 (group$ or subject$ or patient$)).tw.
23. (quasi-random$ or quasi random$ or pseudo-random$ or pseudo random$).tw.
24. ((control or experiment$ or conservative) adj5 (treatment or therapy or procedure or manage$)).tw.
25. ((singl$ or doubl$ or tripl$ or trebl$) adj5 (blind$ or mask$)).tw.
26. (placebo$ or sham).tw.
27. trial.ti.
28. (assign$ or allocat$).tw.
29. (RCT or RCTs).tw.
30. or/9-29
31. 4 and 8 and 30
32. exp animals/ or exp invertebrate/ or animal experiment/ or animal model/ or animal tissue/ or animal cell/ or nonhuman/
33. human/ or normal human/ or human cell/
34. 32 not 33
35. 31 not 34

Appendix 3. CENTRAL search strategy

Revised June 2013

#1 [mh ^"cerebrovascular disorders"] or [mh ^"basal ganglia cerebrovascular disease"] or [mh ^"brain ischemia"] or [mh "brain infarction"] or [mh ^"hypoxia-ischemia, brain"] or [mh ^"carotid artery diseases"] or [mh ^"carotid artery thrombosis"] or [mh ^"carotid artery, internal, dissection"] or [mh ^"intracranial arterial diseases"] or [mh ^"cerebral arterial diseases"] or [mh ^"infarction, anterior cerebral artery"] or [mh ^"infarction, middle cerebral artery"] or [mh ^"infarction, posterior cerebral artery"] or [mh "intracranial embolism and thrombosis"] or [mh stroke] or [mh ^"vertebral artery dissection"]
#2 isch*mi* near/6 (stroke* or apoplex* or cerebral next vasc* or cerebrovasc* or cva or attack*):ti,ab,kw
#3 (brain or cerebr* or cerebell* or vertebrobasil* or hemispher* or intracran* or intracerebral or infratentorial or supratentorial or middle next cerebr* or mca* or "anterior circulation") near/5 (isch*mi* or infarct* or thrombo* or emboli* or occlus* or hypoxi*):ti,ab,kw
#4 #1 or #2 or #3
#5 [mh ^"Platelet Glycoprotein GPIIb-IIIa Complex"/AI,DE,TU]
#6 (glycoprotein* next iib* or platelet next iib* or gp next iib* or iib* next receptor or fibrinogen next receptor*) near/5 (inhib* or block* or antag*)
#7 abciximab or 7E3 or c7E3 or ReoPro or tirofiban or aggrastat or eptifibatide or integrilin or integrelin or intrifiban or sibrafiban or xubix or lamifiban or xemilofiban or orbofiban or fradafiban or lefradafiban or lotrafiban or roxifiban or "ym-337" or ym337
#8 #5 or #6 or #7
#9 #4 and #8

What's new

DateEventDescription
1 August 2013New citation required and conclusions have changedThe review conclusions have changed: whereas in the previous version there was not enough evidence to draw conclusions regarding the efficacy of glycoprotein IIb-IIIa inhibitor therapy in individuals with acute ischaemic stroke, in the current version we found that these drugs are associated with a significant risk of intracranial haemorrhage, with no evidence of any reduction in death or disability in survivors.
1 August 2013New search has been performedThis review has been updated following a re-run of literature searches in June 2013. Two new studies with 951 participants have been added to the review, bringing the total number of included trials to four, with a total of 1365 participants.

History

Protocol first published: Issue 2, 2005
Review first published: Issue 4, 2006

DateEventDescription
25 August 2008AmendedConverted to new review format.

Contributions of authors

Dr Alfonso Ciccone conceived and wrote most of this review, which was checked and discussed with the other authors who approved the final version. Dr Cristina Motto, Ignazio Santilli, Dr Iosief Abraha, Francesco Cozzolino and Dr Alfonso Ciccone selected the trials, extracted the data and assessed the quality of the trials.

Declarations of interest

Ignazio Santilli and Alfonso Ciccone were involved in the recruitment of participants for the AbESTT-II study.

Sources of support

Internal sources

  • N, Not specified.

External sources

  • Evidence Based Research Fund, Division of Clinical Neurosciences, University of Edinburgh, UK.

    Travel, accomodation expenses and workspace for five working days to produce the final draft of the updated review

Differences between protocol and review

The quality evaluation process has been updated according to Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0.

Calculation of the CI for the I2 statistic was introduced, in line with Higgins 2002, in order to assess heterogeneity between studies.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

AbESTT 2000

MethodsC = randomisation by telephone (stratified)
Double blind
Exclusions during trial: 1 (assigned Abciximab)
Losses to follow up: 15 (5 assigned Abciximab)
Participants

38 sites in Europe and North America
401 participants (201 assigned Abciximab), between May 2000 and February 2002

225 (56%) male
Mean age: 68 years
All participants had a CT scan before randomisation
Adult ischaemic stroke < 6 hours since onset

NIHSS score: > 4 and < 23

InterventionsRx: intravenous Abciximab 0.25 mg/kg bolus followed by a 0.125 mg/kg/min infusion for 12 hours
Control: intravenous placebo
OutcomesDeath
Fatal and non-fatal major ICH
Asymptomatic ICH
Extracranial haemorrhage
Thrombocytopenia
NIHSS at 5 days and 3 months
mRS and BI at 3 months
NotesFunding source: supported by Eli Lilly and Company, and Centocor Research and Development
Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Low riskRandomisation by telephone
Blinding (performance bias and detection bias)
All outcomes
Low riskDouble-blind
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and personnel were given identically appearing 5-mL vials of Abciximab or placebo, the 2 study treatments were administered with the same modality
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low riskAll analyses for efficacy were performed on the intention-to-treat population regardless of whether study medication was administered
Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Unclear riskSafety analyses were not performed on the intention-to-treat population but on participants who received at least some study treatment
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

AbESTT-II/Companion 5-6 h

MethodsC = central randomisation, stratified
Double blind
Exclusions during trial: 2 (Rx)
Losses to follow up: 3 (2 Rx, 1 control)
Participants

112 sites in Australia, Europe, South Africa, South and North America

Companion cohort
Ischaemic stroke within 5 to 6 hours after onset

319 participants (160 assigned Abciximab), between December 2003 and September 2005
173 (54%) male
Mean age: 69 years
100% CT before entry
NIHSS score: 4 to 22

Interventions

Rx: Abciximab 0.25 mg/kg intravenous bolus followed by a 0.125 mg/kg/min infusion for 12 hours
Control: intravenous placebo

Duration: 12 hours

OutcomesDeath at 3 months
ICH (symptomatic and systemic) at 5 days and 3 months
Extracranial haemorrhage at 5 days and 3 months
Thrombocytopenia at 5 days
Functional outcome (mRS and BI) at 3 months
Neurological recovery (NIHSS) at 5 days and 3 months
Notes

Ex: specified by protocol - similar to other acute stroke treatment trials
Participants that might be treated with intravenous rt-PA were excluded
FU: 3 months
The trial was terminated prematurely, after 808 participants of the 1800 planned had been included, due to an unfavourable benefit-risk profile

Funding source: supported by Eli Lilly and Company, and Centocor Research and Development

Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Low riskAssignment of treatments was performed via an interactive voice response system
Blinding (performance bias and detection bias)
All outcomes
Low riskTreatment was administered on a double-blind basis
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and personnel were given identically appearing 5-mL vials of Abciximab or placebo, the 2 study treatments were administered with the same modality
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low riskAll analyses for efficacy were performed on the intention-to-treat population regardless of whether study medication was administered
Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Unclear riskSafety analyses were not performed on the intention-to-treat population but on participants who received at least some study treatment, even if not randomised
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

AbESTT-II/Primary

MethodsC = central randomisation
Double blind
Exclusions during trial: 5 (3 Rx, 2 control)
Losses to follow up: 3 (2 Rx, 1 control)
Participants

112 sites in Australia, Europe, South Africa, South and North America.

Primary cohort

Ischaemic stroke < 5 hours

439 participants (221 assigned Abciximab), between December 2003 and September 2005
239 (54%) male
Mean age: 69 years
100% CT before entry
NIHSS score: 4 to 22

Interventions

Rx: Abciximab 0.25 mg/kg intravenous bolus followed by a 0.125 mg/kg/min infusion for 12 hours
Control: intravenous placebo

Duration: 12 hours

OutcomesDeath at 3 months
ICH (symptomatic and systemic) at 5 days and 3 months
Extracranial haemorrhage at 5 days and 3 months
Thrombocytopenia at 5 days
Functional outcome (mRS and BI) at 3 months
Neurological recovery (NIHSS) at 5 days and 3 months
Notes

Ex: specified by protocol - similar to other acute stroke treatment trials
Individuals who might be treated with intravenous rt-PA were excluded
FU: 3 months
The trial was terminated prematurely, after 808 participants of the 1800 planned had been included, due to an unfavourable benefit-risk profile

Funding source: supported by Eli Lilly and Company, and Centocor Research and Development

Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Low riskAssignment of treatments was performed via an interactive voice response system
Blinding (performance bias and detection bias)
All outcomes
Low riskTreatment was administered on a double-blind basis
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and personnel were given identically appearing 5-mL vials of Abciximab or placebo, the 2 study treatments were administered with the same modality
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low riskAll analyses for efficacy were performed on the intention-to-treat population regardless of whether study medication was administered
Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Unclear riskAll safety analyses were not performed on the intention-to-treat population but on participants who received at least some study treatment, even if not randomised
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

AbESTT-II/Wake-Up

MethodsC = central randomisation, stratified
Double blind
Exclusions during trial: 3 (2 Rx, 1 control)
Losses to follow up: 1 (Rx)
Participants

112 sites in Australia, Europe, South Africa, South and North America

Wake-up stroke cohort
Ischaemic stroke, within 3 hours of awakening with signs of stroke
43 participants (22 assigned Abciximab), between December 2003 and May 2005
29 (67%) male
Mean age: 70 years
100% CT before entry
NIHSS score: 4 to 22

Interventions

Rx: Abciximab 0.25 mg/kg intravenous bolus followed by a 0.125 mg/kg/min infusion for 12 hours
Control: intravenous placebo

Duration: 12 hours

OutcomesDeath at 3 months
ICH (symptomatic and systemic) at 5 days and 3 months
Extracranial haemorrhage at 5 days and 3 months
Thrombocytopenia at 5 days
Functional outcome (mRS and BI) at 3 months
Neurological recovery (NIHSS) at 5 days and 3 months
Notes

Ex: specified by protocol - similar to other acute stroke treatment trials
Individuals that might be treated with intravenous rt-PA were excluded
FU: 3 months
recruitment of
Patients enrolment into this part of the study was halted on May 20, 2005, because the rate of symptomatic intracranial haemorrhages among patients receiving Abciximab exceeded the anticipated safety margins. Subsequently, in September 2005, the Safety and Efficacy Monitoring Committee recommended suspending enrolment in the entire trial.

Funding source: supported by Eli Lilly and Company, and Centocor Research and Development

Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Low risk

Randomisation 1:1 using minimisation procedure with biased-coin assignment

Assignment of treatments performed via an interactive voice response system

Blinding (performance bias and detection bias)
All outcomes
Low riskTreatment was administered on a double-blind basis
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and personnel were given identically appearing 5-mL vials of Abciximab or placebo, the 2 study treatments were administered with the same modality
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low riskAll analyses for efficacy were performed on the intention-to-treat population regardless of whether study medication was administered
Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Unclear riskAll safety analyses were not performed on the intention-to-treat population but on participants who received at least some study treatment, even if not randomised
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

Adams 1998

MethodsC = computerised randomisation by telephone (stratified)
Double blind
Exclusions during trial: none
Losses to follow up: none
Participants

International
74 participants (54 assigned Abciximab), between February and November 1998

40 (54%) male
Mean age: 66 years
All patients had CT scan before randomisation
Adult ischaemic stroke < 24 hours since onset

NIHSS score: > 4

InterventionsRx: 4 dose tiers of intravenous Abciximab - bolus 0.15 mg/kg, bolus 0.2 mg/kg, bolus 0.2 mg/kg + infusion of 0.125 µg/kg/min for 12 hours, bolus 0.25 mg/kg + infusion of 0.125 µg/kg/min for 12 hours
Control: intravenous placebo
Duration: dependent on dose tier, up to 12 hours
OutcomesDeath
Fatal and non-fatal major ICH
Asymptomatic ICH
Extracranial haemorrhage
Thrombocytopenia
Functional outcome (mRS and BI) at 5 days and at 3 months
Notes

Ex: specified by protocol - includes US participants < 3 hours onset, bleeding risk, stupor or coma, suspected SAH, etc
FU: 3 months
Data for participants treated < 6 hours obtained from authors

Funding source: supported by Centocor Inc and Lilly Research Laboratories

Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Low risk

Computerised randomisation

Treatment allocation by contacting a 24-hour-a-day interactive voice response system

Blinding (performance bias and detection bias)
All outcomes
Low riskTreatment was assigned on a double-blind basis
Blinding of participants and personnel (performance bias)
All outcomes
Low riskParticipants and personnel were given identically appearing vials of Abciximab or placebo, the two study treatments were administered with the same modality
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low riskNo incomplete outcome data
Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Low riskNo incomplete outcome data
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

SETIS 2010

  1. a

    BI: Barthel Index
    C: concealment of allocation
    CT: computed tomography
    Ex: exclusion
    FU: follow up
    ICH: intracranial haemorrhage
    mRS: modified Rankin Scale
    NIHSS: National Institutes of Health Stroke Scale
    rt-PA: recombinant tissue plasminogen activator
    Rx: treatment
    SAH: subarachnoid haemorrhage

MethodsC = 1-way matched-pair randomisation list, matching performed according to gender, age (≤ 70 or > 70 years) and NIHSS score (≤ 14 or > 14)
Double blind
Exclusions during trial: none
Losses to follow up: none
Participants

3 centres in Northern Italy

150 participants (75 assigned Tirofiban), between January 2003 and December 2007

73 (49%) male

Mean age: 73 years

NIHSS score: 5 to 25

Onset of stroke: < 6 hours

100% CT before entry

InterventionsRx: intravenous Tirofiban, 0.6 µg/kg/min for 30 min followed by 0.15 µg/kg/min infusion for 72 hours
Control: intravenous aspirin 300 mg daily bolus for 3 days
Outcomes

Primary: NIHSS score reduction ≥ 4 points at 72 hours after the start of treatment, and mRS score of 0 to 1 at 3 months

Secondary: all-cause mortality (intrahospital and at 3 months), NIHSS 0 to 1 at 3 months, and severity-adjusted mRS score at 3 months (this endpoint was added after a protocol amendment after the study started)

Notes

Stopped prematurely 48 months after the start of the trial, after recruitment of half of the intended participants, because of futility and low recruitment rate

Funding source: spontaneous study

Risk of bias
BiasAuthors' judgementSupport for judgement
Allocation concealment (selection bias)Unclear risk

One-way matched-pair randomisation list

Details about concealment of randomisation

Blinding (performance bias and detection bias)
All outcomes
Low riskTreatment was administered on a double-blind basis
Blinding of participants and personnel (performance bias)
All outcomes
Low riskThe study drugs were prepared in indistinguishable vials
Blinding of outcome assessment (detection bias)
All outcomes
Low riskAssessors did not know the allocated interventions
Incomplete outcome data (attrition bias)
Death or dependency
Low risk

Intention to treat

None lost or excluded

Incomplete outcome data (attrition bias)
Safety (symptomatic intracranial haemorrhage, death, major extracranial haemorrhage)
Low riskIntention-to treat
None lost or excluded
Selective reporting (reporting bias)Low riskNo suggestion of selective reporting

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    Rx: treatment

    rt-PA: recombinant tissue plasminogen activator

CLEAR 2008

Rx: intravenous low-dose rt-PA plus Eptifibatide at different dosages

Control: standard-dose intravenous rt-PA

Randomised trial confounded by the use of a lower dose of rt-PA in the combined group (Eptifibatide plus rt-PA) than in the control group (rt-PA alone)

CLEAR-ER 2013

Rx: intravenous low-dose rt-PA plus Eptifibatide at different dosages

Control: standard-dose intravenous rt-PA

Randomised trial confounded by the use of a lower dose of rt-PA in the combined group (Eptifibatide plus rt-PA) than in the control group (rt-PA alone)

Davalos 2003Not randomised (series of consecutive individuals treated with intravenous rt-PA or Abciximab)
FAST-StudyNot randomised (series of consecutive individuals treated with intravenous Abciximab and low dose intra-arterial rt-PA; historical control group treated with intra-arterial rt-PA)
Gahn 2004Not randomised (series of cases treated with intravenous rt-PA or intravenous t-PA combined with Abciximab)
Junghans 2002Not randomised (series of cases treated with intravenous heparin plus Tirofiban; controls treated with intravenous heparin only, matched with respect to age, sex, arterial territory, side, stroke aetiology, and clinical score)
Mandava 2005No control group (series of cases, ineligible for intravenous rt-PA because anticoagulated, treated with intravenous Abciximab)
Qureshi 2006No control group (dose-ranging, phase I study on intra-arterial reteplase and intravenous Abciximab)
ROSIE-2No control group; randomised ongoing dose-escalation study of intravenous Eptifibatide
SaTIS 2011

Rx: intravenous Tirofiban

Control: intravenous placebo

Randomised trial in which individuals were treated within 3 to 22 hours after symptom onset, no data available for those treated < 6 hours

Seitz 2003No control group (series of individuals who underwent combined intravenous rt-PA and Tirofiban)
Siebler 2003No control group (series of individuals treated with intravenous Tirofiban and studied with transcranial Doppler ultrasound)
Straub 2004No control group (series of individuals who underwent combined intravenous rt-PA and Tirofiban)
Velat 2006No control group (intravenous Abciximab for acute cerebral thromboembolic events during neuroendovascular procedures)

Characteristics of ongoing studies [ordered by study ID]

Cheung 2000

  1. a

    Rx: treatment

Trial name or titleNot known
Methods 
ParticipantsIndividuals with ischaemic stroke < 6 hours since symptom onset in Hong Kong
InterventionsRx: Abciximab 0.22 mg/kg intravenous bolus then intravenous infusion of 9 µg/min for 12 hours
Control: unknown
OutcomesNot known
Starting dateNot known
Contact informationRaymond TF Cheung, Division of Neurology, Department of Medicine, Queen Mary Hospital, Hong Kong
NotesThe author was contacted for further information but did not reply
Planned sample size: not known

Ancillary