Higher m-cSVD scores linked to increased stroke risk in hemorrhage-prone patients

This study is a post hoc analysis of a prospective, multicenter, randomized controlled trial. The population consisted of 1,454 hemorrhage-prone patients, defined as those with cerebral microbleeds or prior intracerebral hemorrhage. The setting was a prospective, multicenter, randomized controlled trial, and the mean follow-up was 1.9 years. The intervention was the use of modified cerebral small-vessel disease (m-cSVD) scores, categorized as 1, 2, or 3. The comparator was an m-cSVD score of 1.

The primary outcome was any stroke. Key secondary outcomes included ischemic stroke, hemorrhagic stroke, and major adverse cardiovascular events (MACE). The analysis used multivariable Cox regression to investigate associations between m-cSVD scores and events.

For the primary outcome of any stroke, specific incidence rates were not reported in the provided data. For the key secondary outcome of ischemic stroke, the incidence rate per 100 person-years increased with m-cSVD scores: 1.9 for score 1, 2.8 for score 2, and 5.7 for score 3. The adjusted hazard ratio (HR) for ischemic stroke comparing score 3 to score 1 was 2.72, with a 95% confidence interval (CI) of 1.03 to 7.18.

For the secondary outcome of MACE, score 3 was associated with a higher risk compared to score 1. The adjusted HR was 2.34, with a 95% CI of 1.08 to 5.10. For the secondary outcome of hemorrhagic stroke, the incidence showed a numerical increase: 0.5 for score 1, 0.8 for score 2, and 1.5 for score 3. The unadjusted HR was 3.05, with a 95% CI of 0.66 to 14.14. This association did not reach statistical significance.

Safety and tolerability findings were not reported in the provided data. No adverse events, serious adverse events, discontinuations, or tolerability details were provided.

These results can be compared to prior landmark studies in this therapeutic area. The practice relevance note indicates that high cSVD burden may reflect elevated ischemic risk, warranting careful consideration of ischemic stroke prevention even in patients with hemorrhagic potential. The causality note specifies that the study investigated associations, not causation.

Key methodological limitations include that this is a post hoc analysis, which may introduce bias and limits causal inference. The association between m-cSVD scores and events was investigated via multivariable Cox regression, but the post hoc nature is a significant limitation.

Clinical implications are that these findings suggest a potential role for m-cSVD scores in stratifying ischemic risk among hemorrhage-prone patients. However, the evidence is observational and hypothesis-generating. Practice decisions should not be based solely on these results.

Key questions remain unanswered. The specific incidence rate for the primary outcome of any stroke was not reported. The long-term durability of these associations beyond the 1.9-year mean follow-up is unknown. The mechanisms linking m-cSVD burden to ischemic risk in this population require further investigation. The safety profile of interventions based on these scores was not assessed.