2485 consecutive patients treated with IVT over a 14 year period at the Helsinki University Central Hospital were evaluated. In addition to the baseline CT completed prior to IVT administration, a 24-hour post-IVT head CT scan was routinely done in this cohort (as well as when ICH was suspected). Symptomatic ICH (sICH) was defined per the ECASS-2 criteria; remote parenchymal hemorrhage was also assessed. WML grading (by four visual rating scales – Gorter scale, van Swieten scale, Blennow rating scale, and Wahlund rating scale) was completed from baseline CT scans by stroke neurologists blinded to patient data and outcomes.
124 patients developed sICH (5%). All of the rating scales (tested as continuous variables and dichotomized with different cut-off points) were associated with an increased risk of sICH (in univariate and multivariate models). A Wahlund scale score >1 (at least moderate focal WML) conferred a 2.7 (95% CI 1.87-3.90) higher odds of developing sICH in univariate analysis. This association remained (OR 2.64, 95% CI 1.71-4.02) even after adjusting for confounding variables. A Blennow score of 5 had the highest association with sICH (OR 3.59, 95% CI 1.72-7.5) in multivariate analysis. A Blennow score >4 (high burden of WML) also was associated with remote parenchymal hemorrhage (multivariable adjusted OR 4.11, 95% CI 2.38-7.1).
This study suggests that IVT-eligible patients with moderate-to-severe WML on baseline CT could have over two-fold higher risk of sICH. It the largest study on this topic; prior reports have evaluated smaller cohorts and demonstrated conflicting results. The notable limitations to this study are that it did not include a non-IVT treated stroke patient cohort and that it was a retrospective analysis. Though multiple factors contribute to sICH risk after IVT, this association should be kept in mind when discussing IVT with patients.
