Alexander E. Merkler, MD

Benedictus MR, Hochart A, Rossi C, Boulouis G, Hénon H, van der Flier WM, and  Cordonnier C. Prognostic Factors for Cognitive Decline After Intracerebral Hemorrhage. Stroke. 2015

The association between stroke and cognitive decline is complex and, as Drs. Benedictus et al point out, pre-existing dementia is often present in patients with stroke, and post-stroke dementia is also highly prevalent. Furthermore, dementia increases the risk of stroke, and as a corollary, stroke also increases the risk of dementia. Disentangling this relationship poses a challenge.

In the present article, Drs. Benedictus and her colleagues evaluate the association and potential risk factors for cognitive decline after intracerebral hemorrhage (ICH). Although ICH accounts for only 10% of all strokes, the mortality rate is over 50% within the first year. Although various scores have been created to help predict which patients will survive and regain functional independence (ICH score and FUNC score), there has been a paucity of data on cognitive outcomes after ICH. 

In order to assess risk factors for cognitive decline after ICH, the authors evaluated a cohort of adult patients with ICH. MRI was used to evaluate for cortical atrophy, white matter hyperintensities, lacunes, and microbleeds. Patients were prospectively followed for cognitive decline using MMSEs at 6 months, 12 months, and then annually after ICH.

Out of a cohort of 560 patients, 167 were included in the final analysis. During a median follow-up of 4 years, 37% of patients developed cognitive decline. Perhaps not surprisingly, pre-existing cognitive impairment was the strongest predictor of cognitive decline after ICH. In addition, both previous stroke/TIA and severity of cortical atrophy (but not presence of white matter hyperintensities, microbleeds, or lacunes) were associated with cognitive decline after ICH. In a post-hoc sensitivity analysis evaluating risk factors of cognitive decline in patients without pre-existing cognitive decline, only severity of cortical atrophy remained associated with development of cognitive decline. Finally, ICH location (lobar or non-lobar) was not associated with development of cognitive decline.

Limitations include: 1) baseline demographics (age) and radiographic risk factors such as cortical atrophy were not equally distributed between patients who had cognitive data available for analysis; 2) cognitive decline was solely based on MMSE which made it impossible to perform in patients with certain neurological sequelae of stroke such as aphasia; and 3) cognitive status was only evaluated in patients who survived for at least 1 year after ICH (since patients needed 2 MMSE performed 6 months apart), which limits generalizability of this study.

Overall, what does this study tell us? First, more than 1/3 of patients with ICH will develop cognitive decline. Second, risk factors for the development of cognitive decline after ICH are already present before ICH occurs – baseline cognitive impairment, prior stroke/TIA, and baseline cortical atrophy. Third, the interaction between neurodegenerative and vascular pathology is complex, but this study emphasizes the need for better stroke prevention, which, in turn, may lead to decreased neurodegeneration and perhaps vice versa.