Early Tirofiban After Stroke Thrombolysis Improves Recovery, Meta-Analysis Finds

Giving the antiplatelet drug tirofiban within 24 hours of intravenous thrombolysis significantly improves the odds of a favorable recovery in patients with acute non-cardiogenic ischemic stroke, according to a systematic review and meta-analysis published in Frontiers in Neurology on 20 October 2026. The analysis, led by Zhiqiang Huang of Dazhou Integrated Traditional Chinese and Western Medicine Hospital with colleagues Miao Chen and Ke Pu, pooled evidence from four studies totaling 1,394 patients. It found that early tirofiban raised the odds of achieving a modified Rankin Scale score of 0 to 1 at 90 days by roughly 78 percent in the randomized trials, without a statistically significant increase in symptomatic intracranial hemorrhage or 90-day mortality.

Current practice, rooted in expert consensus rather than high-quality trials, has been to withhold antiplatelet agents for 24 hours after intravenous thrombolysis with alteplase or tenecteplase, out of fear of triggering symptomatic intracranial hemorrhage. Non-cardiogenic strokes, which arise from large artery atherosclerosis or small vessel occlusion rather than cardiac embolism, produce platelet-rich thrombi that resist fibrinolytic drugs. The review notes that recanalization rates in these thrombi fall below 50 percent, and that early neurological deterioration occurs in 10 to 20 percent of patients after thrombolysis, with arterial reocclusion rates reaching 60 percent, complications tied to endothelial injury and ongoing platelet activation.

To assemble the evidence, the researchers systematically searched PubMed, Embase, and the Cochrane Library for studies published between January 2021 and March 2026, with a supplementary search on May 15, 2026. The review was registered with PROSPERO (CRD42025181234) and conducted under PRISMA guidelines. The database search yielded 1,247 records; after deduplication and screening, 47 articles proceeded to full-text assessment, and the supplementary search added two more records. Six records ultimately met the qualitative inclusion criteria: four studies with extractable outcome data entered the quantitative synthesis, and two ongoing trial protocols, ADVENT and TREND-IVT, were reviewed qualitatively but contributed no outcome data. Eligible studies had to enroll adults with non-cardiogenic ischemic stroke confirmed by TOAST classification or equivalent criteria, treat patients with alteplase or tenecteplase, and compare early single antiplatelet therapy started within 24 hours of thrombolysis against standard delayed therapy, placebo, or no antiplatelet treatment during the first day.

One prominent trial was deliberately excluded. The TAPIS trial, which enrolled 1,382 patients with moderate ischemic stroke, tested early dual antiplatelet therapy with ticagrelor plus aspirin initiated within six hours of stroke onset. Because the review’s prespecified criterion required early single-agent antiplatelet therapy, TAPIS did not qualify, and the authors stress that this exclusion was based on the intervention criterion, not the publication date. TAPIS did report improved functional outcomes with early dual therapy, with a risk ratio of 1.11 (95 percent confidence interval 1.03 to 1.20), but the authors argue that such findings should not be pooled with trials of single-agent therapy because the intervention, population, and safety profile differ, and they call for dedicated meta-analyses of early dual antiplatelet therapy after thrombolysis.

The quantitative core of the review rests on two randomized controlled trials. ASSET-IT, a multicenter double-blind trial of 832 patients, evaluated early tirofiban after alteplase. INSTANT, a randomized trial of 359 patients, assessed tirofiban after tenecteplase. Together they enrolled 1,191 patients. In the RCT-only analysis, which the authors pre-specified as their primary analytic strategy, early tirofiban was associated with significantly higher odds of favorable functional outcome compared with standard antiplatelet therapy, with an odds ratio of 1.78 (95 percent confidence interval 1.38 to 2.30, p < 0.001) and moderate heterogeneity of 42 percent. The same RCT-only analysis found no statistically significant difference in symptomatic intracranial hemorrhage (odds ratio 1.19, 95 percent confidence interval 0.64 to 2.21, p = 0.59, with no heterogeneity) or in 90-day mortality (odds ratio 0.84, 95 percent confidence interval 0.61 to 1.16, p = 0.29, with no heterogeneity).

When the two observational studies were added as supportive evidence, the results remained consistent. Murphy and colleagues contributed a retrospective cohort study of 130 patients in which low-dose aspirin, 81 milligrams, was initiated at a mean of 21.1 hours after thrombolysis. Zhang and colleagues ran a prospective pilot study of 73 patients with progressive stroke after thrombolysis, treated with tirofiban. In the combined four-study analysis of 1,394 patients, early antiplatelet therapy was associated with higher odds of favorable functional outcome (odds ratio 1.72, 95 percent confidence interval 1.38 to 2.15, p < 0.001, with 39 percent heterogeneity). Symptomatic intracranial hemorrhage showed no significant difference (odds ratio 1.25, 95 percent confidence interval 0.71 to 2.18, p = 0.44, with no heterogeneity), and mortality likewise did not differ significantly (odds ratio 0.86, 95 percent confidence interval 0.64 to 1.16, p = 0.33). The absolute risk of symptomatic intracranial hemorrhage was 2.2 percent in the early therapy group versus 1.8 percent in the standard group, a risk difference of 0.4 percentage points with a 95 percent confidence interval of –0.3 to 1.1 percent. Leave-one-out sensitivity analyses confirmed robustness, with pooled odds ratios for favorable outcome ranging from 1.67 to 1.79.

Subgroup analyses pointed to a possible agent-specific effect, though the authors urge caution. Tirofiban-based regimens, drawn from three studies and 1,264 patients, showed an odds ratio of 1.81 (95 percent confidence interval 1.41 to 2.33) for favorable outcome, while the single aspirin study of 130 patients yielded an odds ratio of 1.35 (95 percent confidence interval 0.97 to 1.88). The test for interaction between the subgroups did not reach statistical significance (p = 0.11), and the aspirin evidence comes from one small retrospective study. The authors therefore describe the tirofiban signal as exploratory, noting that it does not establish superiority over aspirin and requires confirmation in larger head-to-head trials. They also note that tirofiban’s rapid onset and short half-life of approximately two hours make it particularly suitable for the early post-thrombolysis period, since it targets the final common pathway of platelet aggregation through glycoprotein IIb/IIIa inhibition.

The safety findings demand careful interpretation, and the authors are explicit about this. Although the pooled estimates did not detect a statistically significant increase in symptomatic intracranial hemorrhage, the confidence intervals do not exclude a clinically meaningful increase in bleeding risk, and the small number of hemorrhagic events limits the precision of the safety estimates. Differences in how symptomatic intracranial hemorrhage was defined across studies may also affect the pooled estimate. The authors suggest that the observed absolute risk increase of only 0.4 percentage points indicates the bleeding risk may be lower than historically presumed, but they frame this as a hypothesis requiring confirmation rather than a settled conclusion. All four included studies were rated as having low-to-moderate risk of bias using the Cochrane Risk of Bias tool and the Newcastle-Ottawa Scale.

Several further limitations shape how broadly these results can be applied. Only four studies were available for quantitative synthesis. Variability in the timing of early therapy, ranging from less than 3 to 24 hours, and in the agents used precluded definitive conclusions about the optimal protocol, dosing, or patient selection. Most studies were conducted in Asian populations, which limits generalizability to other regions. The observational component, while used primarily for sensitivity and consistency assessment, introduces potential confounding and selection bias, although the close agreement between the RCT-only and combined estimates suggests confounding is unlikely to have materially altered the overall conclusions. The ongoing ADVENT and TREND-IVT trials have not yet reported outcome data, and their results may refine the evidence base.

The authors argue that their findings, grounded primarily in the two large randomized trials of tirofiban, challenge the traditional practice of withholding antiplatelet therapy for 24 hours after thrombolysis, a practice they trace to expert consensus rather than high-quality evidence. They note that the 2026 AHA/ASA guidelines state that adjunctive antithrombotic therapy has not demonstrated benefit in recent randomized trials, but that those trials included heterogeneous stroke populations, whereas ASSET-IT specifically enrolled non-cardioembolic stroke patients. The review concludes that early tirofiban within 24 hours of intravenous thrombolysis significantly improves functional outcomes in acute non-cardiogenic ischemic stroke, with no statistically significant increase in symptomatic hemorrhage or mortality detected, while emphasizing that the evidence for aspirin and other antiplatelet agents remains very limited. The authors call for large-scale randomized controlled trials with standardized protocols to establish optimal timing, agent selection, dosing, and patient selection criteria before any change to guideline recommendations is made.

Subject of Research: Early antiplatelet therapy after intravenous thrombolysis in acute non-cardiogenic ischemic stroke

Article Title: Early antiplatelet therapy after intravenous thrombolysis in acute non-cardiogenic ischemic stroke: a systematic review and meta-analysis

Article References: Huang, Z., Chen, M., & Pu, K. (2026). Early antiplatelet therapy after intravenous thrombolysis in acute non-cardiogenic ischemic stroke: a systematic review and meta-analysis. Frontiers in Neurology, 17, Article 1862583. https://doi.org/10.3389/fneur.2026.1862583

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1862583

Keywords: ischemic stroke, intravenous thrombolysis, tirofiban, antiplatelet therapy, meta-analysis, symptomatic intracranial hemorrhage, non-cardiogenic stroke, ASSET-IT, INSTANT, functional outcome, alteplase, tenecteplase