From Clot to Recovery: New Stroke Research Maps the Whole Patient Journey

Stroke remains among the leading causes of death and acquired disability worldwide, and the clinical problem has quietly changed shape. Reperfusion therapy has been extended to later time windows and to larger ischemic cores, case fatality has fallen across many health systems, and a growing population of survivors now lives for years with cognitive, affective and physical consequences. A new editorial collection, “New horizons in stroke management, volume II,” published in Frontiers in Human Neuroscience by William Huynh of UNSW Sydney and the University of Sydney and Nasser Kashou of Kash Global Tech, gathers 18 contributions spread across eight Frontiers journals spanning neurology, human neuroscience, psychiatry, neuroscience, surgery, pharmacology, endocrinology and nutrition. Its most striking feature, the editors note, is the breadth of disciplines from which the work emerged, with the strongest submissions concerning inflammation, immune competence, metabolism, mood, cognition, surgical judgment and cost, frequently within the same patient.

The first hours and days after onset are represented by four studies, each engaging a decision for which trial evidence remains incompletely defined. Kaur and colleagues implemented an intravenous tirofiban protocol modeled on the RESCUE BT2 trial at a United States comprehensive stroke center, comparing 32 treated patients with a historical standard-care cohort. No symptomatic or asymptomatic intracranial hemorrhage occurred by any of three standard definitions, and no thrombocytopenia was observed. Crucially, and in contrast to the trial on which the protocol was based, more than eighty percent of patients received tirofiban only after early neurological deterioration had already occurred, at a median of 28.8 hours from last known well. The authors therefore describe a rescue strategy for a population in whom current guidelines recommend no targeted therapy, and they appropriately characterize their favorable neurological outcomes as hypothesis-generating.

Registry data supplied a second window on the acute phase. Xie and colleagues analyzed 48,294 thrombolysed patients from the China Stroke Center Alliance, of whom 5,465 had a history of atrial fibrillation, addressing a question on which previous studies have returned conflicting results. After adjustment for age, sex, comorbidity, stroke severity and premorbid function, atrial fibrillation remained associated with in-hospital mortality and, with a larger effect, with cerebral hemorrhage; the same registry showed that pneumonia affected more than a quarter of these thrombolysed patients with atrial fibrillation. Zhang and colleagues, meanwhile, compared stereotactic aspiration of necrotic brain tissue, performed between 24 and 72 hours after onset, with conservative management in 121 patients with malignant middle cerebral artery infarction. Enrollment was restricted to patients aged 61 years and older, a group for whom the landmark decompressive hemicraniectomy trials provide limited guidance. Early cerebral herniation occurred in 8.5 percent of the aspiration group compared with 47.3 percent, and mortality in 19.1 percent compared with 64.9 percent, without significant difference in hemorrhage or intracranial infection.

Cost and accessibility also entered the acute-phase discussion. Xie and colleagues contributed a health-economic dimension infrequently addressed in acute stroke research, modeling edaravone dexborneol delivered as a sublingual tablet against a concentrated solution for injection. Using clinical inputs from the TASTE-SL and TASTE trials, the proportion of patients achieving a modified Rankin Scale score of 1 or less did not differ significantly between formulations, while the oral formulation outperformed on cost-effectiveness. The more consequential implication may be one of access, since an oral route removes the infusion infrastructure otherwise required.

Complications after stroke are not exclusively vascular, and two studies probe the immune and metabolic territory beneath them. Lin and colleagues prospectively sampled peripheral blood on days 1, 3 and 7 following endovascular therapy for acute anterior circulation large vessel occlusion, measuring monocyte human leukocyte antigen-DR expression alongside interleukin-6, C-reactive protein and procalcitonin. After multivariate adjustment, it was reduced mHLA-DR expression at day 1, together with dysphagia, that independently predicted stroke-associated pneumonia. Because mHLA-DR is an established marker of monocyte activity and plays a central role in the immune response, the finding implicates stroke-induced immunodepression rather than excess inflammation, suggesting both a possible biological pathway and a way of identifying susceptible patients early. Zhang and colleagues addressed a largely iatrogenic complication in 305 patients with severe stroke: refeeding syndrome following initiation of enteral nutrition, marked by the characteristic fall in serum phosphorus, was associated with significantly higher rates of new acute liver injury, acute kidney injury, mechanical ventilation and 28-day mortality, together with worse modified Rankin Scale scores at discharge. The implication here is prevention rather than prediction, and nutritional management in severe stroke may deserve more attention than it currently receives.

The largest cluster in the collection, six contributions, concerns depression after stroke, a condition that affects approximately one third of survivors and remains under-recognized. Four studies pursue a common strategy: identifying at-risk patients using laboratory values already obtained at admission. Su and colleagues prospectively followed 318 patients with first-ever acute ischemic stroke and found that those in the highest tertile of the systemic immune-inflammation index were 3.5 times more likely to develop depression at three months. In two complementary analyses of a single cohort at Shanxi Bethune Hospital, Kong and colleagues examined the admission uric acid to high-density lipoprotein cholesterol ratio and the non-high-density lipoprotein to high-density lipoprotein cholesterol ratio, reporting independent associations with 90-day depressive outcomes in both instances; in the latter, each unit increase conferred approximately 35 percent higher odds after adjustment. Luo and colleagues extended the approach to hemorrhagic stroke, showing in 446 patients with spontaneous intracerebral hemorrhage that admission serum pre-albumin was lower among those subsequently diagnosed with depression. Because pre-albumin is a negative acute-phase protein, the direction of this association is concordant with the inflammatory findings. The practical appeal is considerable, since risk stratification may require no new assays but better use of measurements already in the medical record, although the predictive performance of any single index remains modest.

Two reviews supply the mechanistic account that renders such associations biologically plausible. Tang and colleagues review impaired hippocampal neurogenesis as a mechanism of post-stroke depression: ischemic injury does increase neural stem cell proliferation acutely, yet the hippocampus derives little benefit, because the proliferation is short-lived, few progenitors survive, cell fate shifts toward astrogenesis, and the neurons produced are often abnormal in shape and location. The review also covers hypothalamic-pituitary-adrenal axis hyperactivity, neuroinflammation, reduced brain-derived neurotrophic factor signaling and monoaminergic disturbance. One pathway stands out: interleukin-6 drives indoleamine 2,3-dioxygenase, which diverts tryptophan away from serotonin synthesis, precisely the biology that would link a raised inflammatory index at admission to depression months later. Li and Gao review depression following intracerebral hemorrhage, reported in between 18 and 60 percent of cases and less well characterized than its ischemic counterpart. Alongside monoamine imbalance and bidirectional neuroinflammation, they emphasize oxidative stress and iron dysregulation arising from the breakdown of extravasated blood, mechanisms specific to hemorrhage that argue against simply applying frameworks developed for ischemic stroke.

Assessment and stratification form a third thread. Tanglay and colleagues compared the Addenbrooke’s Cognitive Examination-III against the stroke-specific Oxford Cognitive Screen in a small cohort assessed within seven days of first-ever stroke. Although the ACE-III showed good discriminatory capacity at the group level, neither the standard nor the stroke-specific cutoff satisfied established minimum accuracy criteria for cognitive screening after stroke: five patients identified by the Oxford Cognitive Screen were missed, predominantly owing to attentional deficits, while two aphasic but cognitively intact patients were misclassified as impaired. Impairments in praxis and calculation were those associated with functional dependence, and both are underrepresented in general cognitive instruments. Lu and colleagues recorded eye movements during free visual exploration of familiar dramatic character faces in 16 patients and 23 controls; prolonged fixation duration, restricted saccadic movement and reduced scanpath length distinguished the stroke group, and a machine learning classifier integrating these features achieved an accuracy of 87.18 percent. The merit of the paradigm lies in its independence from instruction-following, which may make it feasible in patients unable to comply with structured assessment. Hamadi and colleagues applied targeted whole-exome sequencing in a Saudi cohort from Tabuk, identifying eleven novel variants in GSTT4 alongside variants in cholesterol-associated genes and common polymorphisms in MTHFR, KLF14, NOS3 and ACE, a valuable addition given that most stroke genetics derives from populations of European ancestry.

Prognosis and recovery complete the picture. Liu and colleagues analyzed 1,620 patients with acute ischemic stroke and hypertension and found that elevated TyG-BMI was associated with one-year recurrence overall, but the association was confined almost entirely to women, with no evidence of association among men. Ye and colleagues applied latent class mixed modeling to serial modified Rankin Scale measurements in 127 patients with cerebral venous sinus thrombosis, showing that long-term recovery is better described by two distinct trajectories than by a continuum; together with younger age, a higher venous collateral circulation score was strongly protective against belonging to the poor-recovery class, identifying a structural and imageable determinant of outcome. Amirbekova and colleagues reported a randomized controlled trial of the ReHand portable soft exoskeleton in 120 patients during the subacute phase, delivered five times weekly over eight weeks in addition to standard therapy. The minimal clinically important difference of 5.25 points on the Fugl-Meyer Assessment for the Upper Extremity was achieved by 91.7 percent of the intervention group compared with 43.3 percent of controls, with corresponding gains reflected in the Barthel Index and Functional Independence Measure. Because upper limb recovery remains one of the most difficult problems in stroke rehabilitation, and the subacute window represents the period of greatest neuroplastic capacity, trials of this kind address an important gap.

Taken together, the 18 contributions describe a field in which progress is occurring across several distinct areas at once, from rescue pharmacology in deteriorating patients to robotic assistance months later. The most consistent theme across the collection is the effort to extract prognostic information from data that already exist, whether admission blood results, registry records or routine bedside assessment. Such approaches are widely applicable and are of particular relevance to centers without access to advanced imaging or specialized biomarker panels. Low-cost prognostic information is essential if risk stratification is to be feasible at scale, and several studies collected here represent concrete progress toward that objective, while the prognostic markers identified will need to be compared directly and tested prospectively, for which these studies provide a useful foundation.

Subject of Research: A multidisciplinary editorial collection spanning acute stroke treatment, complications, post-stroke depression, cognition, genetics and rehabilitation

Article Title: Editorial: New horizons in stroke management, volume II

Article References: Huynh, W., & Kashou, N. (2026). Editorial: New horizons in stroke management, volume II. Frontiers in Human Neuroscience, 20, Article 2023091. https://doi.org/10.3389/fnhum.2026.2023091

Image Credits: AI Generated

DOI: 10.3389/fnhum.2026.2023091

Keywords: stroke, thrombolysis, tirofiban, post-stroke depression, immunodepression, mHLA-DR, cognitive screening, refeeding syndrome, exoskeleton rehabilitation, stroke-associated pneumonia, malignant MCA infarction, prognosis