Mechanistic Implications of Cortical Superficial Siderosis in Patients With Mixed Location Intracerebral Hemorrhage and Cerebral Microbleeds

Background and Objectives: Hypertensive cerebral small vessel disease (HTN-cSVD) is the predominant microangiopathy in patients with a combination of lobar and deep cerebral microbleeds (CMBs) and intracerebral hemorrhage (mixed ICH). We tested the hypothesis that cerebral amyloid angiopathy (CAA) is also a contributing microangiopathy in mixed ICH patients with cortical superficial siderosis (cSS), a marker strongly associated with CAA.

Methods: Brain MRIs from a prospective database of consecutive non-traumatic ICH patients admitted to a referral center were reviewed for the presence of CMBs, cSS, and non-hemorrhagic CAA markers (lobar lacunes, centrum semiovale enlarged perivascular spaces (CSO-EPVS), and multispot white matter hyperintensity [WMH] pattern). The frequency of CAA markers and left ventricular hypertrophy (LVH), a marker for hypertensive end-organ damage, were compared between mixed ICH patients with cSS (mixed ICH/cSS[+]) and without cSS (mixed ICH/cSS[–]) in univariate and multivariable models.

Results: Of 1791 ICH patients, 40 had mixed ICH/cSS(+) and 256 had mixed ICH/cSS(–). LVH was less common in mixed ICH/cSS(+) patients compared to those with mixed ICH/cSS(–) (34% vs. 59%, p = 0.01). The frequencies of CAA imaging markers, namely multispot pattern (18% vs. 4%, p < 0.01) and severe CSO-EPVS (33% vs. 11%, p < 0.01), were higher in patients with mixed ICH/cSS(+) compared to mixed ICH/cSS(–). In a logistic regression model, older age (aOR 1.04 per year, 95% CI [1.00–1.07], p = 0.04), lack of LVH (aOR 0.41, 95% CI [0.19–0.89], p = 0.02), multispot WMH pattern (aOR 5.25, 95% CI [1.63–16.94], p = 0.01), and severe centrum semiovale EPVS (aOR 4.24, 95% CI [1.78–10.13], p < 0.01) were independently associated with mixed ICH/cSS(+) after further adjustment for hypertension and coronary artery disease. Among ICH survivors, the adjusted hazard ratio of ICH recurrence in mixed ICH/cSS(+) patients was 4.65 (95% CI [1.38–11.38], p < 0.01) compared to mixed ICH/cSS(–) patients.

Discussion: The underlying microangiopathy of mixed ICH/cSS(+) likely includes both HTN-cSVD and CAA, whereas mixed ICH/cSS(–) is likely driven by HTN-cSVD. These imaging-based classifications can be important to stratify ICH risk, but warrant confirmation in studies incorporating advanced imaging/pathology.