A short course of low-dose intravenous glucocorticoids given around flow-diverter treatment for unruptured brain aneurysms was associated with fewer major cerebrovascular complications and less in-stent narrowing in a prospective multicenter cohort published in Nature Communications. The findings are encouraging, but the study was observational rather than randomized, so they do not yet prove that steroids caused the better outcomes.
Flow diverters treat aneurysms by redirecting blood through the parent artery
An intracranial aneurysm is a weakened, bulging segment of an artery in the brain. Many are discovered before they rupture. When treatment is needed, one option is a flow diverter, a densely braided stent placed across the aneurysm neck.
Instead of filling the aneurysm itself, the device redirects blood along the parent artery. Blood flow into the aneurysm falls, the sac gradually thromboses, and tissue grows across the device.
Flow diversion has become an important treatment for complex unruptured intracranial aneurysms, particularly wide-necked or large lesions. The new paper notes that recent meta-analyses report complete occlusion rates of roughly 75% to 85% at one year.
But the procedure is not risk-free. Ischemic stroke, hemorrhagic stroke, vessel narrowing and other neurological complications can occur during hospitalization or months later. For background on how aneurysms form, rupture risk and warning signs, see our explainer on brain aneurysms and their warning signs.
The study followed patients treated at multiple centers in China
The new analysis was nested within SESIA, the Safety and Efficacy of Stent Deployment for Intracranial Aneurysms prospective multicenter cohort.
The researchers analyzed adults with unruptured intracranial aneurysms who received a flow diverter between January 2018 and August 2024. After applying the study criteria, 2,203 patients were eligible. Of these, 382 had received intravenous glucocorticoids.
Because clinicians decided whether to give glucocorticoids rather than assigning treatment randomly, the researchers used propensity-score matching to make the treated and untreated groups more comparable. Exact matching was also performed for aneurysm location and admission modified Rankin Scale score.
The final matched analysis included 1,327 patients: 360 who received glucocorticoids and 967 who did not.
All patients had been treated with first- or second-generation Pipeline flow diverters.
The steroid regimen began around the time of the procedure
Patients in the glucocorticoid group received either intravenous methylprednisolone or dexamethasone beginning within two hours of the start of flow-diverter implantation.
Methylprednisolone doses ranged from 40 to 160 mg per day, while dexamethasone doses ranged from 10 to 20 mg per day. Treatment continued for at least two days.
The exact dose was chosen by the treating physician, so there was no single standardized steroid protocol across all centers.
Patients also received the antiplatelet therapy required for flow-diverter treatment. In the matched cohort, most received aspirin plus clopidogrel, while a smaller group received aspirin plus ticagrelor.
Major in-hospital cerebrovascular events were substantially less common
The study’s primary outcome combined ischemic stroke, hemorrhagic stroke, transient ischemic attack, silent stroke and neurological death during the postoperative hospital stay.
That composite occurred in 6.9% of glucocorticoid-treated patients versus 13.1% of controls in the propensity-matched analysis. After adjustment, glucocorticoid use was associated with about 47% lower odds of the composite outcome.
The result became more convincing when the researchers focused on major cerebrovascular events, which excluded transient ischemic attack and silent stroke.
Major events occurred in 4.7% of patients given glucocorticoids compared with 12.4% of controls. The adjusted odds ratio was 0.37, corresponding to approximately 63% lower odds in the matched analysis.
The strongest signal was for ischemic stroke
In-hospital ischemic stroke occurred in 2.5% of the glucocorticoid group compared with 10.1% of the non-glucocorticoid group.
The adjusted odds ratio was 0.25, and the association remained significant when the investigators used two additional statistical approaches designed to address confounding.
Hemorrhagic stroke was also numerically less frequent, occurring in 2.5% versus 6.2% of matched patients. However, the hemorrhagic finding was less robust across sensitivity analyses, so it should not be interpreted as established proof that steroids prevent bleeding after flow diversion.
If a stroke does occur, the consequences can extend well beyond the hospital stay. Bites of Bio previously summarized the 2026 stroke rehabilitation guideline and what changed.
The broad primary composite was less robust in sensitivity analyses
This is an important nuance in the paper.
In the propensity-matched analysis, the broad composite of cerebrovascular events favored glucocorticoids. But when the investigators repeated the analysis using overlap weighting and stabilized inverse-probability weighting, that broader composite was no longer statistically significant.
The major-event outcome was more consistent. It remained significantly lower with glucocorticoids across all three analytical approaches.
The authors suggest that part of the difference may come from how minor events were detected. Transient ischemic attacks were numerically more common in the steroid group, and silent strokes were seen only in that group. MRI surveillance was not completely standardized between centers, which could have affected detection of these smaller events.
The association persisted after patients left the hospital
The investigators separately tracked new major cerebrovascular events after discharge so that complications occurring during the initial hospitalization were not simply carried forward into the long-term analysis.
Among matched patients with follow-up data, the median follow-up was about two years.
New major cerebrovascular events occurred in 3.6% of glucocorticoid-treated patients compared with 8.7% of controls.
The adjusted hazard ratio was 0.41. That association also remained significant with overlap weighting and stabilized inverse-probability weighting in the larger unmatched cohort.
Post-discharge ischemic stroke was also less frequent in the glucocorticoid group. There was no clear difference in all-cause mortality.
In-stent stenosis was also less common
The study examined follow-up angiography in addition to clinical events.
In-stent stenosis, defined as more than 25% narrowing of the flow diverter without significant pre-existing stenosis, occurred in 7.9% of glucocorticoid-treated patients versus 13.2% of controls.
Glucocorticoid use was associated with lower adjusted odds of in-stent stenosis.
Importantly, the steroid group did not have better aneurysm occlusion rates. That suggests the apparent safety benefit was not simply because the aneurysms were more successfully sealed.
The proposed mechanism centers on inflammation around the implanted device
The study was not designed to directly prove why glucocorticoids might help, but the authors propose a thromboinflammatory mechanism.
A flow diverter is a metal implant in direct contact with blood. Its surface can activate complement, coagulation pathways, neutrophils and platelets. Neutrophil extracellular traps can promote platelet adhesion and thrombin formation, while inflammatory cytokines can shift the local vessel environment toward clot formation.
Dual antiplatelet therapy is designed primarily to suppress platelet activation and aggregation. It does not directly block the full inflammatory response around the device.
Glucocorticoids suppress inflammatory signaling, including NF-κB-mediated cytokine transcription, and can reduce neutrophil activation. The authors therefore hypothesize that a short perioperative steroid course may complement antiplatelet therapy by dampening inflammation-linked thrombosis.
They also discuss possible effects on blood-brain barrier stability and aneurysm-wall inflammation, but those mechanisms were not directly demonstrated in this cohort.
The study does not establish that steroids should become routine care
The size of the differences is striking, but the design matters.
This was not a randomized trial. Clinicians decided who received methylprednisolone or dexamethasone, and physician preference may have been influenced by factors that were not fully captured in the dataset.
Propensity matching and multiple sensitivity analyses can reduce measured differences between groups, but they cannot eliminate unknown or unmeasured confounding.
The exact glucocorticoid dose also varied between clinicians. The study therefore cannot define an optimal steroid, dose or treatment duration.
Safety data were not comprehensive enough to rule out steroid harms
The investigators reported that no recorded major or minor adverse events were adjudicated as being related to glucocorticoid use.
However, the registry was not designed to systematically capture the full spectrum of steroid-related complications.
The authors specifically note that hyperglycemia, infection, delirium, neuropsychiatric effects and other metabolic complications may have been underestimated.
That limitation is particularly important when considering preventive treatment for patients with an unruptured aneurysm, many of whom have no symptoms before the procedure.
The results apply to elective unruptured aneurysm treatment
Patients with ruptured aneurysms were excluded.
The findings therefore should not be extrapolated to emergency treatment after aneurysmal subarachnoid hemorrhage.
The study also evaluated first- and second-generation Pipeline devices in a Chinese multicenter cohort. Whether the same association applies to every flow-diverter platform, every health system or every patient subgroup remains uncertain.
Some subgroups, including patients with posterior-circulation aneurysms, a previous stroke, multiple flow diverters or aspirin plus ticagrelor therapy, were relatively small and produced wide confidence intervals.
A randomized trial is the critical next step
The observational results are strong enough to justify a more definitive test.
A multicenter, randomized, double-blind, placebo-controlled study of injectable methylprednisolone after flow-diverter implantation has been registered as NCT07353749. The planned trial is designed to test whether a short steroid course can reduce new strokes and neurological death after treatment of unruptured intracranial aneurysms.
That type of randomized evidence will be needed before perioperative glucocorticoids can be considered an established routine addition to flow-diverter treatment.
The main takeaway is promising, but still observational
In this large prospective cohort, short-term low-dose glucocorticoid use around flow-diverter implantation was associated with fewer major in-hospital cerebrovascular events, fewer ischemic strokes, fewer major events during roughly two years of follow-up and less in-stent stenosis.
The major-event findings were more consistent than the study’s broader composite endpoint, which included transient and silent events.
The next question is no longer whether the signal is interesting. It is whether a randomized trial can reproduce it while also measuring steroid-related adverse effects carefully.
Funding and conflicts
The study was supported by grants from the National Natural Science Foundation of China and the Foundation of the National Health Commission Capacity Building and Continuing Education Center. The authors declared no competing interests.
References
- Wen Z, Chen X, Tong X, et al. Low-Dose Glucocorticoids Adjunctive to Flow Diverter Therapy for Unruptured Intracranial Aneurysms Improve Cerebrovascular Outcomes. Nature Communications. 2026. DOI: 10.1038/s41467-026-78408-w.
- ClinicalTrials.gov. Efficacy and Safety of Methylprednisolone After Flow-Diverter Stent Implantation in Unruptured Intracranial Aneurysms. NCT07353749.
- Dong L, Liu Q, Chen X, et al. Methylprednisolone is related to lower incidence of postoperative bleeding after flow diverter treatment for unruptured intracranial aneurysm. Frontiers in Aging Neuroscience. 2023. DOI: 10.3389/fnagi.2023.1029515.