Basal Ganglia Perivascular Spaces: A Dual-Risk Biomarker for Ischemic Stroke and Major Bleeding in Antithrombotic Therapy

Highlights

1. High basal ganglia perivascular space (BGPVS) burden (≥21) is associated with a 4-fold increased risk of major bleeding in patients receiving antithrombotic therapy.

**Detailing the Content**

I am now fully immersed in the core content. I’ve finished the “Highlights” section, ensuring the key data points are clear and concise. The data has been rigorously double-checked, confirming the aHR values are correct. I’ve also drafted the introduction, broadened the scope of SVD, and started detailed methodology regarding centralized evaluation. My next focus is the discussion, where I plan to differentiate between arteriolosclerosis and amyloid angiopathy, the two SVD phenotypes.

2. Patients with extensive BGPVS also face a 2.5-fold higher risk of ischemic stroke, independent of other small vessel disease (SVD) markers.

3. BGPVS serves as a potent imaging biomarker for ‘fragile’ cerebral microvasculature, identifying patients at high risk for both thrombotic and hemorrhagic complications.

4. The study highlights the clinical necessity of incorporating BGPVS evaluation into the risk-benefit analysis for long-term oral antithrombotic use.

Introduction: The Clinical Challenge of Antithrombotic Management

In the management of cerebrovascular and cardiovascular diseases, oral antithrombotic therapies—including antiplatelets and anticoagulants—are cornerstones of secondary prevention. However, clinicians often face a therapeutic paradox: the very treatments that prevent ischemic events simultaneously increase the risk of major bleeding, including intracranial hemorrhage (ICH). Identifying reliable biomarkers that can predict this ‘double-edged sword’ risk is a critical priority in personalized medicine.

Cerebral small vessel disease (SVD) has long been recognized as a primary driver of both ischemic and hemorrhagic stroke. While white matter hyperintensities (WMH) and cerebral microbleeds (CMBs) are well-established SVD markers, enlarged perivascular spaces (ePVSs) have historically been overlooked as incidental findings. Recent evidence suggests that ePVSs, particularly in the basal ganglia, may reflect impaired glymphatic clearance and advanced arteriolosclerosis. This study investigates whether the burden of basal ganglia perivascular spaces (BGPVSs) can serve as a predictive tool for patients on antithrombotic therapy.

Study Design and Methodology

The BAT2 study was a prospective, multicenter observational study conducted across 52 hospitals in Japan between 2016 and 2019. The study cohort consisted of 5,065 patients with a median age of 74 years who were either newly starting or continuing oral antithrombotic agents for cardiovascular or cerebrovascular indications.

Baseline multimodal MRI was a prerequisite for enrollment. A centralized committee of neuroradiologists evaluated the images for traditional SVD markers and specifically quantified BGPVSs. The BGPVS burden was categorized into four levels based on the number of visible spaces in the slice of the highest density: 0, 1–10, 11–20, and ≥21. The primary outcomes included major bleeding (defined by International Society on Thrombosis and Haemostasis criteria), intracranial hemorrhage, ischemic stroke, and all-cause mortality. Researchers utilized multivariable Cox proportional hazards models to adjust for age, sex, hypertension, diabetes, previous stroke history, and other SVD markers like CMBs and WMH.

Key Findings: The Predictive Power of BGPVS

Among the 5,065 participants, 75.4% were taking antiplatelet agents and 29.7% were on anticoagulants. The distribution of BGPVS burden showed that more than 14% of the population had a high burden (≥21). Over a median follow-up of 2.0 years, the study documented 266 ischemic events and 92 major bleeding events.

The Risk of Major Bleeding

The most striking finding was the association between BGPVS and hemorrhagic risk. Patients with a BGPVS score of ≥21 exhibited an adjusted hazard ratio (aHR) of 4.04 (95% CI 1.17–13.92) for major bleeding compared to those with no BGPVS. Even when treated as an ordinal variable, each categorical increase in BGPVS burden was associated with a 38% increase in bleeding risk (aHR 1.38, 95% CI 1.07–1.77).

The Risk of Ischemic Stroke

Parallel to the bleeding risk, BGPVS was also a potent predictor of ischemic stroke. High BGPVS burden (≥21) was associated with a 2.58-fold increased risk of ischemic stroke (aHR 2.58, 95% CI 1.21–5.50). This suggest that BGPVS is not merely a marker of bleeding diathesis but a global indicator of severe vascular frailty and thrombotic potential.

Comparison with Other SVD Markers

Notably, the associations for major bleeding and ischemic stroke remained statistically significant even after adjusting for other imaging markers like cerebral microbleeds and white matter hyperintensities. This indicates that BGPVS provides independent prognostic information that traditional SVD markers might miss. Interestingly, while the risk for major bleeding was high, the study did not find a statistically significant association between BGPVS and intracranial hemorrhage alone or all-cause mortality, likely due to the relatively small number of ICH events (54) during the follow-up period.

Expert Commentary and Mechanistic Insights

The findings from the BAT2 study suggest that BGPVS is a surrogate for advanced arteriolosclerosis and blood-brain barrier (BBB) dysfunction. Perivascular spaces are compartments surrounding the small penetrating arteries and veins in the brain; their enlargement is thought to occur when the drainage of interstitial fluid is hindered by arterial stiffness or protein accumulation. In the basal ganglia, these changes are strongly linked to chronic hypertension.

From a clinical perspective, why does BGPVS predict both bleeding and ischemia? The likely explanation lies in the ‘vessel wall’ hypothesis. Severe arteriolosclerosis leads to luminal narrowing (increasing ischemic risk) and simultaneous weakening of the vessel wall (increasing hemorrhagic risk). In patients on antithrombotics, this fragile microvascular environment is more susceptible to failure.

Clinicians should consider BGPVS as a ‘red flag’ on MRI reports. While current guidelines do not yet mandate BGPVS quantification for drug selection, a high BGPVS burden should prompt a more rigorous control of blood pressure and perhaps a more cautious approach to dual antiplatelet therapy (DAPT) or combinations of anticoagulants and antiplatelets.

Study Limitations

Despite its large scale and prospective design, the study has limitations. The follow-up period of 2.0 years is relatively short for observing low-frequency events like ICH. Additionally, the cohort was entirely Japanese, which may limit generalizability to other ethnic groups where the prevalence and patterns of SVD (such as cerebral amyloid angiopathy vs. hypertensive arteriolosclerosis) may differ. Finally, the study did not differentiate between the types of antithrombotics (e.g., Warfarin vs. DOACs) in its primary BGPVS analysis.

Conclusion

The BAT2 study elevates basal ganglia perivascular spaces from an incidental MRI finding to a clinically relevant biomarker for vascular risk. In patients receiving antithrombotic therapy, a high BGPVS burden identifies a phenotype of severe small vessel disease prone to both ischemic stroke and major systemic bleeding. Incorporating BGPVS assessment into routine clinical practice could refine risk stratification and pave the way for more personalized, safer antithrombotic strategies.

References

Iwamoto S, Miwa K, Koga M, et al. Impact of Basal Ganglia Perivascular Spaces on Ischemic and Hemorrhagic Risks in Patients Taking Antithrombotic Therapies. Neurology. 2026;106(7):e214745. doi:10.1212/WNL.0000000000214745.

Wardlaw JM, Smith C, Dichgans M. Mechanisms of sporadic cerebral small vessel disease: insights from neuroimaging. Lancet Neurol. 2013;12(5):483-497.

Doubal FN, MacLullich AM, Lammie GA, Wardlaw JM. Enlarged perivascular spaces on MRI are a feature of cerebral small vessel disease. Stroke. 2010;41(3):450-454.