Highlight
- A significant spatial and temporal association exists between cortical microinfarcts (CMIs) and cortical superficial siderosis (cSS) in cerebral amyloid angiopathy (CAA).
- Regions of the cortex affected by cSS exhibit markedly higher densities of CMIs, supporting the hypothesis that hemosiderin deposits from cSS may contribute to local ischemic injury.
- Baseline cSS volume independently predicts the development of new cortical microinfarcts over 24 months, whereas baseline CMI burden does not predict cSS progression.
- These findings identify potential mechanistic pathways linking hemorrhagic and ischemic brain injuries in CAA and suggest new targets for preventing cortical injury.
Study Background
Cerebral amyloid angiopathy (CAA) is a common cerebral small vessel disease characterized by deposition of amyloid-β in the walls of cortical and leptomeningeal blood vessels. Clinically, CAA manifests with both hemorrhagic lesions, such as lobar intracerebral hemorrhages and cortical superficial siderosis (cSS), and nonhemorrhagic ischemic lesions, including cortical cerebral microinfarcts (CMIs). Both hemorrhagic and ischemic components contribute to cognitive decline and neurological morbidity. Despite recognition that CMIs and cSS coexist in CAA, the underlying spatial and temporal relationships remain poorly understood. Understanding these associations is crucial for elucidating pathophysiological interactions and identifying interventions to limit cortical injury progression.
Study Design and Methods
This longitudinal observational study prospectively recruited 74 patients with probable CAA as defined by the Boston Criteria version 2.0, who had no prior symptomatic intracerebral hemorrhage or evidence of CAA-related inflammation. Participants were enrolled from a memory clinic-based cohort at Massachusetts General Hospital and underwent high-resolution 3-Tesla MRI scanning at baseline and after 24 months. Imaging assessed the presence and volume of cortical superficial siderosis (cSS), the number and spatial distribution of cortical CMIs, and the presence of lobar lacunes.
Quantitative analyses included multivariable logistic regression and generalized linear models to investigate cross-sectional and longitudinal associations between baseline radiologic markers and radiologic progression. The density of CMIs in cortical regions with versus without overlaid cSS was compared to explore spatial relationships.
Key Findings
Among the 74 patients (median age 73 years; 38% female), 26 (35%) were found to have cortical CMIs (total 135 CMIs), with 39% of these infarcts located beneath areas of cSS. The group with CMIs demonstrated significantly higher prevalence (73% vs 40%; p=0.006) and volume of cSS (median 1.3 mL vs 0; p<0.001), and more lobar lacunes (50% vs 19%; p=0.005) compared to CMI-negative patients.
At baseline, cortical CMI count showed nonlinear positive associations with both cSS volume (exponential coefficient 1.12; 95% CI 1.07-1.18; p<0.001) and lobar lacunes count (exponential coefficient 1.65; 95% CI 1.52-1.80; p<0.001), emphasizing the interrelation between ischemic and hemorrhagic small vessel disease markers.
Spatial analysis revealed that cortical regions covered by cSS had a markedly higher median density of CMIs compared to non-cSS regions (20.84 vs 0.49 CMIs per 1,000 cm2; p=0.010), indicating local vulnerability where hemosiderin deposition occurs.
Follow-up MRI available for 36 patients identified 23 new cortical CMIs over 24 months; notably, 43% of these incident infarcts occurred beneath areas of pre-existing cSS. Importantly, baseline cSS volume was independently associated with incident CMI risk (odds ratio 1.41 per mL increase; 95% CI 1.02-1.95; p=0.036), whereas baseline CMI burden did not predict subsequent cSS progression.
Expert Commentary
This study elegantly elucidates a bidirectional relationship between two hallmark lesion types in CAA—cortical superficial siderosis and cortical microinfarcts—strengthening the concept that hemorrhagic and ischemic injury pathways are intertwined in this disease. The finding that cSS volume predicts new ischemic cortical lesions suggests the pathogenic role of subarachnoid hemorrhage-related iron deposition in promoting focal ischemia, potentially via vessel wall toxicity or inflammation.
However, the longitudinal sample is relatively small, and the observational design limits causal inference. Future work should focus on mechanistic studies to delineate how hemosiderin and related inflammatory processes cause local ischemic damage. Moreover, therapeutic trials targeting cSS-related injury pathways could determine if preventing or reducing cSS mitigates cortical infarction in CAA, thereby preserving cognitive function.
Conclusion
This study identifies a significant spatiotemporal association between cortical superficial siderosis and microinfarcts in cerebral amyloid angiopathy, suggesting that cSS may actively contribute to local ischemic injury. These findings expand our understanding of the complex interplay between hemorrhagic and ischemic brain injuries in CAA, opening avenues for novel biomarker-guided interventions aimed at preventing progressive cortical damage and associated clinical decline. Larger longitudinal studies and mechanistic investigations are warranted to confirm these hypothesis-generating observations and translate them into targeted therapeutics.
Funding and ClinicalTrials.gov
The study was conducted at Massachusetts General Hospital with no direct funding information reported in the original publication. The trial was not registered as this was a prospective observational cohort study.
References
- Losa M, Zanon Zotin MC, van Harten TW, et al. Spatiotemporal Associations Between Cortical Microinfarcts and Cortical Superficial Siderosis in Cerebral Amyloid Angiopathy. Neurology. 2026;107(2):e218-274. doi:10.1212/WNL.0000000000002182
- Greenberg SM, Charidimou A. Diagnosis of cerebral amyloid angiopathy: evolution of the Boston criteria. Stroke. 2018;49(2):491-497.
- van Veluw SJ, Hilal S, Kuijf HJ, et al. Cortical microinfarcts on 3T MRI in cerebral amyloid angiopathy: a cross-validation study. Stroke. 2017;48(3):640-645.
- Charidimou A, Gang Q, Werring DJ. Sporadic cerebral amyloid angiopathy revisited. Brain. 2012;135(Pt 11):3006-3019.
