Supplementary Material for: Perihematomal Diffusion Restriction in Intracerebral Hemorrhage Depends on Hematoma Volume, But Does Not Predict Outcome

<b><i>Background:</i></b> Perihematomal diffusion restriction (PDR) is a frequent finding in primary intracerebral hemorrhage (ICH) on diffusion-weighted MRI. Its frequency, associated clinical and imaging findings and impact on clinical outcome are not well understood. <b><i>Methods:</i></b> This is a retrospective single-center analysis of 172 patients with primary ICH who received MRI within 24 h from symptom onset. PDR was defined as a reduction of apparent diffusion coefficient below 550 × 10<sup>-6</sup> mm<sup>2</sup>/s. Multivariate regression analyses were used to assess independent imaging and clinical predictors of PDR. Clinical outcome was assessed using the modified Rankin scale (mRS) at discharge. <b><i>Results:</i></b> PDR was present in 88 patients (51.2%). Median PDR volume was 1.1 ml (interquartile range 0.2-4.2). Multivariate analyses identified hematoma volume as the key independent predictor of PDR. The volume of perihematomal edema, lobar hematoma location and low diastolic blood pressure at admission were further predictors. Although the occurrence of PDR correlated with in-hospital mortality (75.0 vs. 43.4%, p < 0.001) and moderately severe to severe disability or death at discharge (mRS ≥4; 56.4 vs. 27.8%, p = 0.002), PDR was not an independent predictor of clinical outcome. In contrast, hematoma volume, ventricular extension of hemorrhage and higher age independently predicted an adverse clinical outcome. <b><i>Conclusions:</i></b> PDR is common after primary ICH within 24 h of symptom onset. Hematoma volume was identified as the key predictor of PDR. Although PDR was associated with mortality and severe disability, this effect was confounded by established risk factors. These results do not support a role of early PDR as prognostic factor after ICH independent of hematoma volume.