Mechanical thrombectomy, in which physicians thread catheters through blood vessels to physically remove a clot, is the established standard of care for acute ischemic stroke caused by a large vessel occlusion. Intravenous thrombolysis with the drug alteplase is recommended as first-line therapy within four and a half hours of symptom onset, and beyond that window when perfusion imaging shows salvageable brain tissue. When both treatments are used, the combination is known as bridging therapy. Yet the practice remains debated, because intravenous thrombolysis carries a risk of hemorrhage, and well-designed trials comparing direct thrombectomy with bridging therapy have produced mixed results. A recent Cochrane systematic review pooling six randomized controlled trials found no evidence of a difference in functional outcome, mortality, or intracranial hemorrhage between the two strategies, and its authors called for research into time- and person-specific factors that might determine who benefits from the intravenous drug.

Researchers at University Hospital Carl Gustav Carus in Dresden suspected that one such factor might be the biological composition of the clot itself. Clots removed from patients with large vessel occlusion vary considerably in their proportions of fibrin and red blood cells, and these differences fall largely along etiological lines. Clots of cardiac origin, which form when the heart’s rhythm falters and debris travels to the brain, tend to be rich in fibrin and platelets. Clots arising from atherosclerotic disease in the neck or brain arteries tend to be red blood cell rich. Fibrin-rich clots are denser, create more friction against vessel walls, and are known to be less sensitive to thrombolytic drugs, which work by infiltrating the clot and thinning its fibrin fibers. Red blood cell rich clots, with their more loosely packed fibrin networks, allow greater drug penetration and dissolve more readily.

To test whether these biological differences translate into different treatment outcomes, the team conducted a post hoc analysis of their prospective registry of consecutive thrombectomy patients treated between January 2017 and January 2023. The cohort included 798 adults who underwent thrombectomy for anterior circulation large vessel occlusion, either at the Dresden center directly or after transfer from one of thirteen telestroke network spokes or eight partner primary stroke centers in eastern Saxony. Just over half the patients were women, the median age was 77 years, and the median baseline stroke severity on the National Institutes of Health Stroke Scale was 15. Nearly half of the patients, 395 in total, received intravenous alteplase before thrombectomy; no other thrombolytic agents were used during the study period. Stroke etiology was classified according to the TOAST criteria by senior stroke consultants and validated through a secondary central review of all medical records.

The primary outcome was the time from groin puncture to successful angiographic reperfusion, quantified using the modified Thrombolysis in Cerebral Infarction score, with successful reperfusion defined as a score of 2b or higher. The researchers used stratified Cox proportional hazards regression, adjusting for age, baseline stroke severity, tandem occlusions, carotid T occlusions, occlusion site, and distal thrombus migration—a shift of the clot from a proximal to a more distal arterial segment that occurred before any endovascular maneuver. Overall, successful reperfusion was achieved in 85.2 percent of patients, and cardioembolic stroke accounted for 58.9 percent of cases.

The results split cleanly along etiological lines. Among the 330 patients with non-cardioembolic stroke, bridging intravenous thrombolysis was independently associated with a roughly 40 percent higher hazard of achieving successful reperfusion at any point during the procedure, with an adjusted hazard ratio of 1.40. This effect grew stronger in sensitivity analyses: when patients with stroke of undetermined etiology were excluded, the adjusted hazard ratio rose to 2.06, and a sub-analysis restricted to patients with large artery atherosclerotic stroke reproduced the benefit. Consistent with this, patients with atherosclerotic stroke who received the intravenous drug required about 36 percent fewer aspiration attempts during thrombectomy, an effect measured with a negative binomial regression that accounted for age, initial CT severity score, and occlusion site.

Among the 468 patients with cardioembolic stroke, however, the picture was entirely different. Intravenous thrombolysis alone showed no measurable effect on time to reperfusion, with an adjusted hazard ratio of 1.13 that did not reach statistical significance. More troubling, when the clot migrated distally after thrombolysis and before the procedure—a phenomenon observed in 9.7 percent of cardioembolic patients who received the drug versus 0.8 percent of those who did not—the likelihood of successful reperfusion at any given time point dropped by approximately 43 percent, with an adjusted hazard ratio of 0.57. The researchers propose that when thrombolysis fails to soften a fibrin-rich cardiac clot, it may instead promote its displacement into smaller, more distal arterial segments where thrombectomy is technically harder and less effective, a concern underscored by two recent negative randomized trials of thrombectomy for distal vessel occlusions.

The study also quantified the clinical stakes of procedural speed. Each additional 10 minutes of groin-to-recanalization time increased the odds of a worse functional outcome on the modified Rankin scale at 90 days by 8.6 percent, an association that remained robust after adjustment for patient characteristics, stroke features, interventional parameters, and metabolic markers. In other words, a treatment that accelerates reperfusion in one stroke subtype and delays it in another could meaningfully shift outcomes for individual patients, even if population-level averages obscure the difference.

The mechanistic story fits a growing body of evidence. Analyses of the MR CLEAN Registry biobank and other prospective thrombus studies have shown that cardioembolic clots contain a higher proportion of fibrin and platelets, while atherosclerotic clots are predominantly red blood cell rich. Fibrin-rich thrombi demand more recanalization attempts and yield lower reperfusion rates, and laboratory work suggests they are less susceptible to thrombolytic agents because their fibrin is densely packed. The multicenter RESTORE Registry found that clots retrieved in earlier thrombectomy passes had higher red blood cell content, while later passes yielded fibrin-rich material. Beyond simple histology, molecular analyses have identified neutrophil extracellular traps as structural components that stabilize the fibrin network and may contribute to thrombolysis resistance—raising the possibility that partial clot destabilization without effective lysis could promote distal embolization and increase procedural complexity.

The Dresden team is careful to note the limitations of their work. The data come from a regional network registry, histological characterization of retrieved clots was not available, and the subgroup of cardioembolic patients with distal thrombus migration was small, leaving room for residual confounding despite robust statistical adjustment. Cryptogenic stroke, moreover, is a highly heterogeneous category that may mask diverse underlying mechanisms. Still, the consistency of the etiological differences across all multivariable models, and their reproduction in the atherosclerotic sub-analysis, suggest a robust finding. If confirmed in larger studies, the results could help clinicians move beyond a one-size-fits-all approach to bridging therapy, reserving intravenous thrombolysis for the patients whose clots are most likely to yield to it—and sparing others a treatment that may do more harm than good.

Subject of Research: Effect of intravenous thrombolysis before thrombectomy on reperfusion outcomes in cardioembolic versus non-cardioembolic acute ischemic stroke

Article Title: IV Thrombolysis Facilitates Interventional Reperfusion in Non‐Cardioembolic but Not Cardioembolic Stroke

Article References: Sedghi, A., Kaiser, D. P. O., Arndt, M., Diel, N. J., Simon, E., Polanski, W. H., Puetz, V., Huttner, H. B., & Siepmann, T. (2026). IV Thrombolysis Facilitates Interventional Reperfusion in Non‐Cardioembolic but Not Cardioembolic Stroke. Annals of Clinical and Translational Neurology, 13(9), 1893-1902. https://doi.org/10.1002/acn3.70370

Image Credits: AI Generated

DOI: 10.1002/acn3.70370

Keywords: stroke, thrombectomy, intravenous thrombolysis, large vessel occlusion, cardioembolic stroke, clot composition, reperfusion, alteplase, TOAST classification, distal thrombus migration, fibrin-rich thrombi, bridging therapy