Network Meta-Analysis of Ticagrelor for Stroke Prevention in Patients at High Risk for Cardiovascular or Cerebrovascular Events

Alexandra Bálint, MD; Dániel Tornyos, MD; Oumaima El Alaoui El Abdallaoui, MS; Péter Kupó, MD; András Komócsi, MD, DSc


Stroke. 2021;52(9):2809-2816. 

In This Article


In a multiple-treatment NMA of 26 trials including 124 495 patients, we found further evidence that the choice of antiplatelet strategy influences the risk of stroke in patients with high thrombotic risk. Within this comprehensive analysis of randomized trials testing ticagrelor in a wide range of clinical scenarios, we found that ticagrelor plus aspirin, as compared with aspirin alone, was associated with a significant risk reduction of stroke (20%). Data of this analysis showed an important trade-off between stroke prevention and bleeding risk. Thus, when the risk of major bleeding was taken into consideration, the probability of being the best choice of treatment was the highest for aspirin monotherapy, whereas it was the lowest for A+T. Additionally, this combination significantly increased the risk of intracranial bleeding. We found important clustering of clinical end points among antiplatelet monotherapies and combinations, while in models considering the components of the combinations, the highest stroke prevention potential and the highest bleeding risk were attributable to ticagrelor.

Platelet-driven thrombotic events play a pivotal role in the development of ischemic vascular events. Earlier analyses found favorable results for aspirin as initial therapy in the prevention of ischemic stroke.[43] However, aspirin alone fails to prevent ischemic events in patients at higher risk of recurrences like in cases with recent minor stroke or TIA or in patients with acute coronary syndrome.[44] Later development in antiplatelet therapy aimed at inhibition of alternative pathways including the P2Y12 receptor–mediated activation and in combination with aspirin offering a greater reduction of thromboembolic complications. In TASS (Ticlopidine Aspirin Stroke Study), ticlopidine alone was superior to aspirin with a 21% risk reduction of fatal and nonfatal stroke. However, due to its unfavorable side effects and with the availability of more tolerable ADP inhibitors, ticlopidine is used scarcely in the clinical praxis. Consequently, as no study was performed comparing ticlopidine to ticagrelor, data of ticlopidine studies were not included in our NMA.[45] Moreover, with reassuring data on the reduction in ischemic events seen in acute coronary syndrome, the question was raised whether the intensification of antiplatelet therapy could be similarly beneficial in the prevention of ischemic stroke.

Our findings are partly in line with previous meta-analyses indicating that ticagrelor was more effective in reducing combined ischemic and hemorrhagic stroke compared with other antiplatelet regimens in patients with coronary artery disease, cerebrovascular disease, or peripheral artery disease and extended these with the observation that stroke prevention potential is consistently reflected in trials with ticagrelor treatment regardless of the inclusion condition. Importantly, prevention and bleeding trade-off show clustering at the level of antiplatelet monotherapies and combinations. P2Y12 inhibitor and aspirin combination show more effective stroke prevention, but its use is associated with an increase in the risk of bleeding. This risk includes intracranial bleeding that is significantly higher with ticagrelor and aspirin. The analysis did not show important benefits of ticagrelor-based combination when compared with clopidogrel. Net adverse clinical events showed only benefit among studies with acute coronary syndrome patients.[46]

Our network analysis included some trials that also applied prasugrel—another effective but irreversible P2Y12 blocking agent in combination with aspirin. It is important to note that in the fundamental TRITON-TIMI 38 trial (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel–Thrombolysis in Myocardial Infarction), increasing the risk of bleeding events including fatal bleeding was found in patients with a history of TIA or stroke.[6] Although TRITON-TIMI was not powered for poststroke/TIA events, and only a limited percentage of patients had a history of cerebrovascular disease, prasugrel is contraindicated for them.[6] As all included trials were performed after TRITON-TIMI 38, TIA or stroke was a contraindication for prasugrel treatment while ticagrelor was applied even among the highest risk for intracranial bleeding like those with recent stroke. We believe that the clinical applicability of prasugrel among patients with earlier cerebrovascular events remains to be studied in greater detail. The magnitude of its treatment effect is, however, at the range of the other P2Y12 inhibitors when applied in patients without a cerebrovascular history.

It remains unclear whether the preventive effect of ticagrelor is explainable with its more effective inhibition of P2Y12-dependent platelet activation or with additional effects like increase in adenosine levels due to an additional blockage via ENT-1 (equilibrative nucleoside transporter 1) leading to platelet inhibition, inflammatory milieu modulation, vasodilation, and protection from ischemia and reperfusion injury.[47] Orchestrated by these, ticagrelor may have additional protective effects on cerebral ischemia-reperfusion. Additionally, to the potentially lower bleeding risk due to the reversible P2Y12 inhibition, animal studies indicated neuroprotective effects of ticagrelor through endothelial NO synthase modulation, resulting in increased blood flow and reducing infarct volume.[48]

Both the THALES trial and the subgroup analysis of the SOCRATES trial (Ticagrelor Versus Aspirin in Acute Stroke or Transient Ischemic Attack) support these findings indicating a risk reduction of 32% with ticagrelor and 27% with A+T over aspirin in patients with minor ischemic stroke or high-risk TIA.[11,12] However, it is important to note that these trials also found an important increase of bleeding complications that may reduce or cancel out the ischemic benefit. Functional health status such as disabling stroke outcome (defined as death or modified Rankin Scale score >1) was reported only in the THALES and SOCRATES trials. The analysis of this end point did not explore important differences. The RR was 0.99 (0.93–1.06) in A+T versus aspirin, 0.91 (0.77–1.06) in ticagrelor versus aspirin comparisons, and estimated to be 1.10 (0.92–1.30) in A+T versus ticagrelor (Figure IX in the Data Supplement).


The present analysis should be interpreted considering some limitations. It is important to note that multiple mechanisms may lead to the development of stroke and an important minority of these cases have a cardioembolic origin. Oral anticoagulation remains the treatment of choice of these patients and due to interactions between anticoagulant and antiplatelet therapies cases requiring oral anticoagulation were excluded from these trials.

All analyses were performed by pooling of the active drug arms with various dosages; therefore, it limits our ability to assess how the differential effects of the dosage of these drugs affect the outcomes. Furthermore, the included studies reflected the problems of capturing bleeding events and the lack of one overall accepted bleeding definition system. In 14 trials, the TIMI criteria were used while in 10 other systems like the PLATO (Ticagrelor Compared With Clopidogrel in Patients With Acute Coronary Syndromes trial), BARC (Bleeding Academic Research Consortium), or GUSTO (Global Use of Strategies to Open Occluded Coronary Arteries) were applied often in combinations.

The absence of patient-level data precludes adjustments important confounders that might have remained sources of heterogeneity between trials. This is also true of the aspirin dose since potential ticagrelor-aspirin interaction has been documented. The study-level analysis allows limited insight into the time relations of treatment benefit. Moreover, the optimal time frame may differ among pharmacological treatments and combinations.


Comprehensive analysis of clinical trials supports that the use of ticagrelor as mono- or aspirin combined therapy resulted in more effective stroke prevention in a high-risk patient population. Our analyses also underscore the importance of bleeding associated with intensified antiplatelet therapy. Highlighting the trade-off between bleeding risk and stroke prevention, the data show that besides ischemic risk, bleeding risk should be assessed and considered. This lower risk of ischemic stroke with ticagrelor was counterbalanced with a higher risk of major bleeding including an importantly increased risk of intracranial bleeding. Signal of benefit supporting the use of ticagrelor was neither present in the mortality nor in the combined analyses of ischemic and bleeding events. The decision regarding the choice of antiplatelet agent and its duration should be individualized according to the risks and benefits of the chosen treatment. We believe that the development of scoring tools stratifying patients according to their bleeding risk is essential to improve clinical outcomes in conjunction with intensified antiplatelet stroke prevention.