Abstract Body: Introduction: Chronic kidney disease (CKD) is a risk factor for intracerebral hemorrhage (ICH) and for worse outcomes following ICH. In this study, we examined the relationship between kidney function, ICH, and brain iron deposition, using quantitative susceptibility mapping (QSM) and cerebral microbleeds (CMB) as intermediate biomarkers. We hypothesized that kidney function is an independent risk factor for QSM/CMB, and that high QSM/CMB burden is an independent risk factor for ICH.

Methods: A retrospective cohort of spontaneous ICH and non-ICH patients with CT/MR imaging was identified. For each patient, eGFR along with age, sex, race, ethnicity, hypertension, diabetes (DM), heart disease, and hyperlipidemia were abstracted. CMB volume as well as intracranial arterial calcification (IAC) volume were extracted using custom convolutional neural networks. QSM was derived using the Graz total generalized variation method. All results were visually verified for accuracy by an expert neuroradiologist. Univariate analysis for QSM/CMB risk factors was assessed using Pearson correlation and t-test, while multivariate analysis was assessed using linear regression. Univariate analysis of ICH risk factors was assessed using t-test and Chi-square, while multivariate analysis was assessed using logistic regression.

Results: Among 155 patients in this study, 39 were ICH patients. In all patients, eGFR was significantly inversely correlated with QSM (r=-0.30, p=0.02) and CMB (r=-0.24, p=0.04); significant differences were also observed across CKD stages for both QSM/CMB (ANOVA; p<0.01). Univariate analysis for QSM risk factors yielded additional statistically significant associations for DM (p=0.01), heart disease (p=0.03), Caucasian race (p=0.01) as well as IAC (p=0.048). In multivariate analysis, eGFR (p=0.04) and Caucasian race (p<0.01) remained significant independent predictors of QSM, while eGFR (p=0.02) remained the only independent predictor of CMB. Furthermore, both QSM (p=0.02) and CMB (p<0.01) were significantly higher in ICH patients. In multivariate analysis, QSM remained a significant independent predictor of ICH (p=0.03) while CMB did not reach significance.

Conclusion: This is the first demonstration that eGFR predicts strictly quantified QSM and CMB, which then predicts ICH. These findings offer novel insights into determinants of brain iron deposition and microbleeds as mechanisms underlying the CKD-ICH relationship.