Figure 1

Associations between plasma protein levels, acute stroke severity and short- and long-term neurological outcome in 209 ischemic stroke cases, and comparisons of plasma protein levels between these cases and controls at the different time-points. Panels (a–d): Associations between plasma protein levels (Normalized Protein eXpression [NPX] levels) and degree of neurological impairments according to the NIH stroke scale (NIHSS), visualized with forest plots with β values and 95% confidence intervals derived from multivariable linear regression models adjusted for age, sex, and time until the blood draw (model 1). Associations with false discovery rate (FDR) < 0.05 are marked in black. As the NPX value is given on a log2 scale, the yielded β-value corresponds to the predicted change in NIHSS score per each doubling of protein levels. Associations between: (a) plasma protein levels and NIHSS score in the acute stroke phase, (b) plasma protein levels in the acute phase and NIHSS score at 3 months post-stroke, (c) plasma protein levels in the acute phase and NIHSS score at 7 years post-stroke, (d) plasma protein levels at 3 months and NIHSS score at 7 years post-stroke. For BCAN and Nr-CAM (marked in bold), the association between 3-month levels and 7-year outcome withstood additional adjustment for stroke severity. Panels (e–f): Differences in plasma protein levels between cases and controls. The bars correspond to Cohen’s d, i.e. the mean difference in case and control levels divided by the standard deviation for the controls, where a positive value indicates higher levels in cases than controls. Blue bars correspond to comparisons with a statistically significant difference (FDR < 0.05). (e) Comparison between acute phase samples (cases) and controls. (f) Comparison between 3-month samples (cases) and controls.