The researchers performed a post hoc analysis of data from a prospective multicenter cohort involving patients treated for minor stroke. Between September 2016 and April 2023, patients receiving intravenous thrombolysis were approached at the First Hospital of Jilin University, and additional recruitment occurred at 15 hospitals in Northeast China from September 2021 to January 2022. A total of 347 patients received intravenous thrombolysis. After 28 patients who did not receive the standard dose of alteplase or bridging therapy were excluded, 319 met the study inclusion criteria. Two patients refused to participate, 39 had incomplete or unqualified blood samples, and six were lost to follow-up at 90 days, leaving 272 patients for analysis.

Eligible participants were older than 18 years, received thrombolysis within 4.5 hours of symptom onset, had an admission National Institutes of Health Stroke Scale (NIHSS) score of no more than 5, and had a pre-stroke Modified Rankin Scale score of 0 or 1. Minor stroke was therefore represented by mild neurological deficits rather than absence of disability. Blood was collected at baseline and 24 hours after thrombolysis. Early neurological deterioration was defined as an increase in the NIHSS score of at least four points within 24 hours after treatment. Twenty of the 272 analyzed patients, or 7.35%, met that definition; 252 did not experience early neurological deterioration.

Baseline GFAP concentrations did not differ significantly between the deterioration and no-deterioration groups. Before thrombolysis, median GFAP was 10.25 pg/mL among patients who later deteriorated, with an interquartile range of 6.81–19.24 pg/mL, compared with 8.54 pg/mL, with an interquartile range of 5.18–14.27 pg/mL, among those who did not. The groups separated more clearly after treatment: median GFAP at 24 hours was 40.05 pg/mL in the deterioration group, with an interquartile range of 22.59–126.24 pg/mL, versus 13.57 pg/mL, with an interquartile range of 8.74–33.80 pg/mL, in the no-deterioration group. The difference was significant at p<0.001. The change from baseline was therefore more closely associated with subsequent deterioration than the single pre-treatment measurement.

The investigators defined the relative GFAP change, ∆GFAP, as the post-thrombolysis concentration divided by the baseline concentration. Median ∆GFAP was 3.57 in the deterioration group, with an interquartile range of 1.40–23.39, and 1.57 in the no-deterioration group, with an interquartile range of 1.00–2.94; the group difference was significant at p=0.006. In univariable logistic regression, each unit increase in the log-transformed change measure was associated with an odds ratio of 3.495 for early neurological deterioration, with a 95% confidence interval of 1.672–7.306. The association remained statistically significant in models adjusted for age and sex, vascular risk factors, and clinical variables. In the most extensively adjusted model, a 2.54-fold increase in ∆GFAP was associated with a 4.64-fold increase in the odds of deterioration, with a 95% confidence interval of 1.86–11.539 and p<0.001.

GFAP is a protein in the cytoskeleton of astrocytes, which contribute to the structural support of neighboring cells and to maintenance of the blood–brain barrier. The source study notes that elevated GFAP during acute ischemic stroke can arise from extensive release associated with astrocyte necrosis in the infarct core and from increased expression associated with reactive gliosis in surrounding tissue. Infarct volume was similar in the two deterioration groups: 1.14 mL, with an interquartile range of 0.70–4.03 mL, among patients who deteriorated and 1.32 mL, with an interquartile range of 0.42–2.80 mL, among those who did not, with p=0.203. On that basis, the authors proposed that the larger GFAP rise may reflect differences in reactive gliosis rather than substantially different amounts of tissue death. This was a biological interpretation proposed by the researchers, not a mechanism directly tested in the cohort.

To assess whether GFAP change added information to conventional clinical variables, the researchers built two prediction models. Model A contained age and admission NIHSS score. Model B added log-transformed ∆GFAP. The clinical-only model had an area under the receiver operating characteristic curve of 0.695, with a 95% confidence interval of 0.584–0.805, while the model including GFAP had an area of 0.770, with a 95% confidence interval of 0.674–0.886. Under 10-fold cross-validation, the corresponding areas were 0.61 and 0.72. A Hosmer–Lemeshow test did not indicate poor model conformity, because its p value was greater than 0.05. The comparison between the two areas under the ROC curves was not statistically significant, but the GFAP-containing model produced significantly higher net reclassification and integrated discrimination values: the net reclassification index was 70.00%, with p=0.003, and the integrated discrimination improvement was 6.62%, with p=0.038. Decision-curve analysis also showed greater net benefit after adding ∆GFAP across a range of threshold probabilities.

Patients who experienced early neurological deterioration had a substantially poorer 90-day outcome profile. At follow-up, 5 of 20 patients in the deterioration group, or 25.0%, had a favorable outcome, defined as a Modified Rankin Scale score of 0 or 1. By comparison, 169 of 252 patients in the no-deterioration group, or 67.1%, had a favorable outcome; the difference was significant at p<0.001. Favorable outcomes also became less common across increasing thirds of ∆GFAP: 76.7% in the lowest third, 63.0% in the middle third and 55.6% in the highest third, with p=0.016. These observations describe an association in this cohort and do not demonstrate that GFAP elevation caused either neurological worsening or poorer functional recovery.

Repeated GFAP assessment was technically feasible with the assay used in the study. Serum GFAP was measured using a magnetic-particle-based automated chemiluminescent enzyme-linked immunosorbent assay system that required 60–80 microliters of serum and had a reported turnaround time of approximately 28 minutes from sample loading to result reporting. The lower limit of quantification was 5 pg/mL, and the assay coefficient of variation was below 8.0%. Laboratory technicians were blinded to patients’ clinical data, and test results were not shared with healthcare providers. The results support further evaluation of a repeated biomarker measurement as part of risk assessment, but the study did not test whether using GFAP to guide surveillance or treatment would prevent deterioration or improve outcomes.

Several limitations constrain the conclusions and the immediate generalizability of the prediction model. This was a post hoc analysis of a prospective multicenter cohort, and the final analysis included 272 of the 347 patients who initially received thrombolysis, after 28 exclusions and further losses from refusal, unusable or incomplete blood samples, and missing 90-day follow-up. The investigators did not distinguish between disabling and non-disabling minor strokes. There was no control group of patients receiving standard medical care without intravenous thrombolysis, so the findings do not isolate the effect of thrombolysis or establish whether the GFAP response would have differed without treatment. The relatively small number of events also prevented more precise prediction of hemorrhagic and ischemic causes of deterioration. External validation in larger, independent prospective cohorts is needed before the model can be considered broadly applicable, and future studies must determine whether biomarker-guided monitoring or intervention can actually reduce early neurological deterioration and improve long-term outcomes.

Subject of Research: Prospective cohort analysis of dynamic serum GFAP for predicting early neurological deterioration after thrombolysis in minor stroke.

Article Title: Dynamic Changes in Serum Glial Fibrillary Acidic Protein Levels Predicts Early Neurological Deterioration After Intravenous Thrombolysis in Minor Stroke

Article References: Zhou, J., Qu, Y., Abuduxukuer, R., Zhang, P., Wang, T., Gao, J.-H., Pan, D.-W., Yuan, Z.-M., Li, C.-Y., Chen, L.-P., Chen, J.-H., Gu, J.-L., Wang, A.-Y., Jiang, L.-G., Liu, L., Jiang, Y.-F., Wang, C.-F., Jia, Y., Xin, H., … the Biomarkers of Brain Injury Investigator Study Group (2026). Dynamic Changes in Serum Glial Fibrillary Acidic Protein Levels Predicts Early Neurological Deterioration After Intravenous Thrombolysis in Minor Stroke. Annals of Clinical and Translational Neurology, Article acn3.70552. https://doi.org/10.1002/acn3.70552

Image Credits: AI Generated

DOI: 10.1002/acn3.70552

Keywords: minor ischemic stroke, intravenous thrombolysis, glial fibrillary acidic protein, GFAP, astrocyte injury, early neurological deterioration, blood biomarker, stroke prognosis, prediction model, Chinese cohort, Dynamic, Changes

The researchers divided admissions from Changsha Central Hospital chronologically rather than randomly. The earliest 70 percent, from January 2020 through November 2024, formed a derivation cohort of 593 patients, including 67 with hemorrhagic transformation. The remaining 254 admissions, recorded from late November 2024 through May 2026, formed the internal temporal-validation cohort and included 27 hemorrhagic transformation events. Overall, 94 of 847 admissions, or 11.1 percent, had imaging-confirmed in-hospital hemorrhagic transformation. The design was intended to test whether laboratory markers added useful information beyond readily available clinical variables under changing real-world conditions.

The baseline model captured age, baseline National Institutes of Health Stroke Scale score, atrial fibrillation, large-vessel occlusion, rapid-glucose measurement and onset-to-treatment time. The candidate additions were the natural logarithm of the neutrophil-to-lymphocyte ratio, or ln(NLR), the atherogenic index of plasma, or AIP, and residual cholesterol. In temporal validation, the clinical model had an AUC of 0.801 with a 95 percent confidence interval of 0.707 to 0.881. Adding ln(NLR) raised the AUC to 0.818, a difference of 0.017 with a DeLong test p value of 0.220. Adding AIP and residual cholesterol produced an AUC of 0.803, while the full nine-variable model reached 0.818, a gain of 0.018 with p = 0.346.

Discrimination was therefore only numerically better after the biomarkers were introduced. Calibration and overall prediction error were virtually unchanged: both the clinical and full models had Brier scores of 0.081 in temporal validation, and their calibration slopes were 1.029 and 1.060, respectively. Decision-curve analysis likewise showed closely overlapping net-benefit trajectories, without a broad or consistently separated advantage for the full model. In other words, the laboratory additions did not demonstrate that they would improve clinical decisions more reliably than the simpler model across the range of decision thresholds examined.

The adjusted analysis identified baseline stroke severity as the clearest independent correlate of the broader hemorrhagic transformation outcome. Each one-point increase in the NIHSS score was associated with 8.4 percent higher odds of hemorrhagic transformation, with an odds ratio of 1.084, a 95 percent confidence interval of 1.044 to 1.126 and p below 0.001. Atrial fibrillation was associated with approximately 2.7-fold higher odds, with an odds ratio of 2.66, a confidence interval of 1.34 to 5.27 and p = 0.005. Large-vessel occlusion had a positive but statistically imprecise association. By contrast, ln(NLR), AIP and residual cholesterol were not independently associated with the primary outcome after the clinical variables were included.

That contrast extended to the unadjusted patient characteristics. Admissions with hemorrhagic transformation had a median baseline NIHSS score of 15, compared with 5 among those without transformation, and large-vessel occlusion was present in 75.9 percent versus 41.4 percent. Atrial fibrillation was also more frequent, at 46.2 percent versus 15.7 percent. Patients with transformation also had higher NLR and other inflammatory indices, lower total and low-density lipoprotein cholesterol, and a lower AIP. These differences did not translate into independent biomarker effects or robust incremental prediction once the clinical model was included, illustrating why an association in a group comparison is not equivalent to useful patient-level discrimination.

A separate exploratory analysis produced a narrower signal for strict symptomatic intracranial hemorrhage, or sICH, defined using the ECASS II framework and restricted to events within seven days. There were 37 sICH events overall, including only 12 in the temporal-validation cohort. In the derivation data, ln(NLR) was associated with sICH after adjustment for NIHSS, atrial fibrillation and large-vessel occlusion, with an odds ratio of 1.89, a 95 percent confidence interval of 1.04 to 3.42 and p = 0.037. Adding NLR to the exploratory clinical model increased its validation AUC from 0.816 to 0.888, but the increment of 0.071 did not reach statistical significance, with p = 0.097. The small number of validation events makes this finding hypothesis-generating rather than evidence of clinical readiness.

The robustness checks did not establish a reproducible advantage from adding biomarkers. Excluding patients who received bridging thrombectomy produced validation AUCs of 0.770 for the clinical model and 0.761 for the full model. Restricting the analysis to patients with known large-vessel occlusion status produced AUCs of 0.800 and 0.815, respectively, while limiting hemorrhagic transformation to events documented within seven days yielded 0.802 and 0.809. Excluding three events whose archived documentation could not be re-verified and adopting a conservative approach to uncertain laboratory timing likewise failed to demonstrate consistent biomarker gains. A separate gradient-boosting analysis within the same cohort also ranked NIHSS, atrial fibrillation and large-vessel occlusion above the three biomarkers, but it was not an independent validation.

Several design features strengthen the central conclusion. Preprocessing and multiple-imputation models were fitted in the derivation cohort and then applied to the later cohort, limiting information leakage. The analysis assessed discrimination, calibration, Brier score, decision curves and multiple sensitivity analyses rather than relying on a single performance metric. A 500-resample bootstrap check estimated a mean optimism of 0.031 for the full model’s apparent AUC, producing an optimism-corrected derivation AUC of approximately 0.786. A post-hoc adequacy calculation also indicated that more development observations would have been desirable. The authors therefore characterized the work as an exploratory incremental-value analysis rather than the development of a deployable clinical risk score.

Important limitations temper the conclusions. This was a single-center retrospective study, so institutional practices and patient selection may limit generalizability, and internal temporal validation does not establish transportability to other hospitals. Hemorrhagic transformation was identified from official radiology reports rather than independent, blinded review of archived images, and clinically triggered repeat imaging may have affected ascertainment. Although the analyzed blood and lipid measurements came from initial emergency specimens collected before thrombolysis, exact draw-to-bolus intervals were not retained for every patient, and not all results were necessarily available before treatment. Predictor missingness and biomarker distributions also changed over time. The results therefore support evaluating future markers against strong clinical baselines, but they do not justify withholding intravenous thrombolysis from otherwise eligible patients.

Subject of Research: Incremental prediction of hemorrhagic transformation after intravenous thrombolysis for acute ischemic stroke

Article Title: Clinical severity predominates over routine inflammatory and metabolic biomarkers in predicting hemorrhagic transformation after intravenous thrombolysis: internal temporal validation of incremental biomarker value

Article References: Liang, H., Zeng, B., Wang, J., Yi, L., Huang, L., Lin, Z., Zhou, C., & Yue, Z. (2026). Clinical severity predominates over routine inflammatory and metabolic biomarkers in predicting hemorrhagic transformation after intravenous thrombolysis: internal temporal validation of incremental biomarker value. Frontiers in Neurology, 17, Article 1940110. https://doi.org/10.3389/fneur.2026.1940110

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1940110

Keywords: acute ischemic stroke, intravenous thrombolysis, hemorrhagic transformation, NIHSS, neutrophil-to-lymphocyte ratio, atherogenic index of plasma, residual cholesterol, prediction model, temporal validation, Clinical, severity, predominates

The research, published in Frontiers in Neurology, followed 507 patients with mild acute ischemic stroke who received intravenous thrombolysis between January 2024 and June 2025. Ninety days after treatment, neurologists assessed each patient using the modified Rankin Scale, the standard yardstick of stroke recovery. A score of zero or one, meaning no symptoms or no significant disability, was considered an excellent outcome. Anything from two to six, ranging from mild restriction to death, was classified as non-excellent. The result was sobering: 136 patients, or 26.8 percent, fell into that latter category despite having strokes mild enough that many clinicians would have expected near-complete recovery.

That figure is not an anomaly. Previous studies have estimated that between 20 and 35 percent of patients with minor stroke fail to achieve excellent functional status at three months even after timely thrombolysis. What makes the new work distinctive is its focus. Most existing prediction models lump together patients of all severities, or rely on variables such as in-hospital complications that only become known after treatment has begun. Neither approach helps a physician standing at the bedside deciding how aggressively to monitor a patient whose stroke looks trivial. The Tianjin team deliberately restricted their model to information available at admission, before the first drop of alteplase or tenecteplase entered the vein.

The technical pipeline reflects current best practice in clinical machine learning. The researchers first applied LASSO regression, a statistical technique that shrinks the coefficients of uninformative variables to zero, effectively performing automatic feature selection. From a wide panel of demographic, clinical, laboratory, and neuroimaging variables, LASSO retained just six predictors: age, baseline NIHSS score, platelet count, monocyte count, blood glucose level, and the neutrophil-to-lymphocyte ratio, a widely used marker of systemic inflammation. Six variables, all obtainable from a standard complete blood count and chemistry panel, became the raw material for seven competing algorithms.

Those algorithms spanned the spectrum of modern machine learning, from classical logistic regression and decision trees to random forests, support vector machines, artificial neural networks, and two gradient-boosting methods, XGBoost and LightGBM. Each of the six non-linear models underwent hyperparameter optimization through grid search combined with stratified five-fold cross-validation, a procedure that repeatedly partitions the training data to find the settings that generalize best. The dataset was split 70:30 into training and internal validation cohorts using stratified random sampling, with a fixed random seed to guarantee reproducibility. When the dust settled, XGBoost, an algorithm that builds an ensemble of decision trees in which each new tree corrects the errors of its predecessors, emerged as the top performer.

On the validation cohort, XGBoost achieved an area under the receiver operating characteristic curve of 0.753, with a 95 percent confidence interval of 0.660 to 0.840. That number, while not spectacular, is respectable for a prediction made from six routine variables, and it edged out the support vector machine at 0.750 and logistic regression at 0.747. The decision tree fared worst at 0.614. Beyond raw discrimination, the team evaluated calibration, how closely predicted probabilities match observed outcomes, using calibration curves and the Brier score, and clinical usefulness through decision curve analysis, which compares the net benefit of acting on the model’s predictions against treating everyone or no one. XGBoost performed adequately on all three fronts, with a Brier score of 0.165.

What elevates the study above a simple algorithm bake-off is its commitment to interpretability. Black-box predictions are notoriously hard for clinicians to trust, so the researchers turned to SHAP, or Shapley Additive Explanations, a technique borrowed from cooperative game theory that assigns each predictor a contribution value for every individual prediction. The SHAP analysis ranked the baseline NIHSS score as the single most influential predictor, followed by blood glucose, age, platelet count, the neutrophil-to-lymphocyte ratio, and monocyte count. Crucially, the direction of each association was consistent with clinical intuition: higher values of every one of the six predictors pushed the model’s output toward a higher probability of poor recovery.

The biology behind those predictors is plausible and, in several cases, well supported by prior literature. Elevated admission glucose has long been linked to worse outcomes after thrombolysis, partly because stress hyperglycemia amplifies oxidative stress, promotes lactate accumulation, and disrupts the blood-brain barrier, increasing the risk of hemorrhagic transformation. A high neutrophil-to-lymphocyte ratio signals an innate immune system in overdrive: neutrophils infiltrating the ischemic brain release reactive oxygen species, matrix metalloproteinase-9, and neutrophil extracellular traps that damage the blood-brain barrier and clog the microvasculature, while lymphocytes, depleted by stroke-induced immunosuppression, can no longer restrain the damage. Even a successful recanalization of the large occluded artery can be undermined by this no-reflow phenomenon at the capillary level.

Platelet count is the most counterintuitive of the six predictors, since higher platelets might seem protective in a disease defined by clotting. Yet the SHAP dependence plots suggested a threshold-like pattern, with counts above roughly 280 to 300 billion per liter associated with higher predicted risk. The authors point to thromboinflammation: activated platelets adhere to injured endothelium and bind leukocytes through molecules such as P-selectin, forming aggregates that promote microvascular thrombosis and endothelial dysfunction. They are careful to note that this is a model-based predictive association, not proof of causation, and that SHAP attributions can be distorted by correlations among predictors. The same caution applies to the entire model, which the researchers frame as an adjunct to, not a replacement for, clinical judgment.

The study’s limitations are acknowledged candidly. It was retrospective and single-center, with internal validation only, so generalizability across institutions and populations remains unproven. The XGBoost model’s sensitivity of 0.525 means roughly half of the patients who ultimately fared poorly were not flagged, a meaningful gap for any screening tool. The cohort contained only nine patients treated with tenecteplase, precluding any analysis of whether the choice of thrombolytic matters. Still, the authors argue that the model’s real value lies in what it enables at the bedside: patients identified as high risk could receive closer neurological monitoring, earlier rehabilitation assessment, more aggressive management of modifiable factors such as glucose, and tighter follow-up after discharge. For a patient population often dismissed as low-risk, a six-variable algorithm that whispers a warning before the needle goes in could, if validated externally, change how mild strokes are managed.

Subject of Research: Machine learning prediction of 90-day functional outcomes after intravenous thrombolysis in mild acute ischemic stroke

Article Title: Machine learning-based models to predict 90-day non-excellent functional outcomes among patients with mild acute ischemic stroke after intravenous thrombolysis

Article References: Wang, Y., Qiu, X., Wang, Y., & Wang, L. (2026). Machine learning-based models to predict 90-day non-excellent functional outcomes among patients with mild acute ischemic stroke after intravenous thrombolysis. Frontiers in Neurology, 17, Article 1960403. https://doi.org/10.3389/fneur.2026.1960403

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1960403

Keywords: acute ischemic stroke, intravenous thrombolysis, machine learning, XGBoost, SHAP, neutrophil-to-lymphocyte ratio, modified Rankin Scale, blood glucose, platelet count, risk stratification, explainable AI, stroke prognosis

The study, led by Irene Scala of the Cerebrovascular Unit at Fondazione IRCCS Istituto Neurologico Carlo Besta in Milan and conducted on behalf of the Italian Stroke Association (ISA)-Young, set out to answer a deceptively simple question: what actually happens when patients receive repeated intravenous thrombolysis (RIVT) for a recurrent acute ischemic stroke within 90 days of a previous thrombolysis-treated event? The researchers systematically screened the literature and identified 28 reports describing 63 patients who underwent the procedure, with individual patient-level data available for 44 of those cases. That patient-level granularity matters enormously in a field dominated by small case series, because it allowed the team to analyze outcomes at the level of individual strokes rather than pooled study averages, reducing the aggregation bias that has plagued earlier reviews.

The headline finding is striking. Among patients who received a second course of intravenous thrombolysis within the three-month window, 66.7 percent achieved a favorable functional outcome — a proportion that compares respectably with outcomes reported for first-time thrombolysis in routine clinical registries. The median interval between the two treatments was just 7.5 days, meaning most of these patients were re-dosed while their initial infarcts were still biologically fresh, precisely the scenario in which the feared hemorrhagic risk was thought to be highest. Yet across the entire dataset, not a single case of symptomatic intracranial hemorrhage (sICH) — the dreaded complication that the three-month exclusion criterion was designed to prevent — was recorded after repeated thrombolysis.

Formal meta-analysis reinforced the picture. The pooled odds ratio for better functional outcomes after repeated thrombolysis was 1.52, with a 95 percent confidence interval of 0.97 to 2.39 and zero heterogeneity across studies (I² = 0%). The confidence interval crosses one, so the trend toward benefit does not reach statistical significance, but the direction is consistent and the absence of heterogeneity suggests the underlying case reports tell a remarkably uniform story. Equally important, the analysis found no significant increase in functional deterioration between stroke episodes, with a pooled odds ratio of 1.28 (95% CI 0.75–2.19; I² = 0%). In other words, patients did not systematically worsen after their second stroke, and there was no signal that the repeat treatment itself drove neurological decline.

Perhaps the most clinically useful aspect of the analysis is what it ruled out. The researchers tested whether outcomes were influenced by the interval between treatments, patient age, sex, stroke etiology, vascular territory, or the time from symptom onset to treatment — and found that none of these factors significantly shaped results. This null finding directly undermines the logic of a fixed temporal cutoff. If the safety of a second dose does not measurably depend on how many days have elapsed since the first stroke, then a rigid three-month clock may be the wrong instrument for judging risk. Instead, the data suggest clinicians should weigh the individual biology of each recurrence rather than the calendar.

The safety signal was not entirely clean, and the authors are careful to report it. Asymptomatic intracranial or systemic hemorrhages occurred in 15.9 percent of patients — bleeding visible on imaging or detected clinically but not causing neurological worsening. Fatal events occurred in 5 percent of cases, but crucially, these deaths were unrelated to neurological complications. Both hemorrhagic and fatal events clustered predominantly in patients with cardioembolic strokes and more severe initial presentations, hinting that the underlying mechanism of recurrence and stroke severity, rather than the repeat thrombolysis itself, may drive adverse outcomes. This pattern aligns with mechanistic reasoning: cardioembolic strokes, often arising from atrial fibrillation or cardiac thrombi, tend to be larger and more lethal, and patients with severe deficits have more tissue at risk regardless of treatment.

To understand why the three-month rule exists at all, one has to look back to the founding trials of stroke thrombolysis. The 1995 NINDS trial of tissue plasminogen activator, which established alteplase as standard care, excluded patients with recent stroke, and subsequent guidelines from the European Stroke Organisation and the American Heart Association/American Stroke Association carried the exclusion forward as a relative contraindication grounded largely in expert consensus rather than dedicated trials of repeat dosing. The biological rationale — that freshly infarcted brain and reperfused vessels are more prone to bleeding — has never been disproven, but it has also never been rigorously quantified in the recurrent-stroke setting. Meanwhile, the clinical dilemma is real and growing: patients with active sources of embolism, unstable atherosclerotic plaques, carotid webs, free-floating thrombi, or basilar artery occlusions can suffer devastating recurrent strokes days or weeks after a first event, and withholding thrombolysis from them may condemn them to disability or death.

The new analysis, registered prospectively in PROSPERO (CRD420251276423) and conducted according to PRISMA 2020 reporting standards, synthesizes a literature that spans two decades of case reports and small series — from early accounts of imaging-guided repeat dosing in 2005 to multicenter case studies of ultra-early re-treatment published in 2023. The authors used the Heidelberg bleeding classification to standardize hemorrhage assessment and performed the meta-analysis with established statistical tooling. By extracting individual patient data wherever possible, they converted a scattered, anecdotal literature into something approaching coherent evidence — a methodological upgrade that previous systematic reviews of the same question, which relied on aggregated study-level data, could not achieve.

The caveats are substantial and the authors acknowledge them plainly. Repeated thrombolysis within 90 days is a rare event, and its efficacy remains understudied; the 63 patients analyzed come from published reports that are inherently subject to publication bias, since dramatic successes and instructive failures are more likely to reach print than unremarkable outcomes. There are no randomized trials, no matched controls who were denied repeat thrombolysis, and no way to fully separate the effect of treatment from the effect of careful patient selection — the very clinicians who chose to re-treat these patients presumably judged them to be favorable candidates. The favorable outcomes observed may partly reflect that selection rather than the safety of the drug itself. Five percent mortality, even if neurologically unrelated, is not negligible in an already fragile population.

Still, the conclusion the authors draw is measured and potentially practice-changing: repeated intravenous thrombolysis within 90 days may be a reasonable reperfusion strategy in carefully selected patients with recurrent acute ischemic stroke, and the interval from the previous stroke should not be treated as an isolated exclusion criterion. In an era when mechanical thrombectomy is increasingly available but not always feasible — particularly for distal vessel occlusions or in centers without endovascular capability — the finding suggests that the pharmacological route should not be automatically closed by the calendar. The evidence is observational and thin, but for a rule that has governed stroke units worldwide on the strength of consensus alone, the burden of proof may now have shifted. Larger, prospective registries of recurrent stroke treatment will be needed to confirm whether the zero-sICH finding holds at scale, but for clinicians facing a deteriorating patient with a fresh recurrence, this analysis offers something they have rarely had: data suggesting that saying yes to a second dose may not be the reckless act the guidelines have long implied.

Subject of Research: Safety and efficacy of repeated intravenous thrombolysis within three months for recurrent acute ischemic stroke

Article Title: Safety and efficacy of repeated intravenous thrombolysis within 3 months for acute ischemic stroke: a systematic review and individual-patient data meta-analysis

Article References: Scala, I., Ciacciarelli, A., Cancelloni, V., Digiovanni, A., Galotto, D., Gardin, A., Giammello, F., & the Italian Stroke Association (ISA)-Young (2026). Safety and efficacy of repeated intravenous thrombolysis within 3 months for acute ischemic stroke: a systematic review and individual-patient data meta-analysis. Journal of Neurology, 273(10), Article 607. https://doi.org/10.1007/s00415-026-14144-x

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14144-x

Keywords: stroke, intravenous thrombolysis, recurrent stroke, alteplase, symptomatic intracranial hemorrhage, meta-analysis, systematic review, cerebrovascular disease, reperfusion therapy, stroke guidelines, neurology, thrombolytic drugs

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