Cognitive Decline After Stroke
Relation to Inflammatory Biomarkers and Hippocampal Volume
Background and Purpose—Inflammation may contribute to cognitive impairment after stroke. Inflammatory markers are associated with hippocampal atrophy. We tested whether markers of inflammation, erythrocyte sedimentation rate (ESR), and serum levels of C-reactive protein are associated with reduced hippocampal volume and poor cognitive performance among stroke survivors.
Methods—We analyzed 368 consecutive cases from our prospective study of first-ever mild–moderate stroke patients. MRI, cognitive tests, and inflammatory markers were determined. Patients were reevaluated 6 and 12 months after the event.
Results—ESR remained unchanged in follow-up examinations, suggesting a chronic inflammation background in some patients. Higher levels of C-reactive protein and ESR were associated with worse performance in cognitive tests, particularly memory scores. This association was maintained for ESR (but not C-reactive protein) after adjustment for confounders (P=0.002). Patients with smaller hippocampi had inferior cognitive results. Moreover, in a multivariate regression model, higher ESR values (but not C-reactive protein) were related to reduced hippocampal volume (P=0.049).
Conclusions—This report shows a strong relationship between ESR and hippocampal volume, as well as with cognitive performance among poststroke patients. This could plausibly relate to incipient cognitive decline via hippocampal pathways.
Low-grade systemic inflammation occurs commonly in aging1 and has been implicated as an important mechanism underlying cognitive impairment in the elderly.2 Inflammatory markers are known to be associated with dementia, particularly Alzheimer’s disease (AD) and stroke,3 but (almost) no data are available in regard to their involvement in vascular cognitive impairment.4 The major risk factors for stroke, such as atherosclerosis, hyperlipidemia, and obesity, are associated with systemic as well as brain inflammation.5 Systemic inflammation may lead to a primed inflammatory environment in the brain prior to stroke occurrence,5 which could aggravate the consequences of ischemia. Additionally, central inflammation may adversely affect learning and memory through hippocampus remodeling.6,7
We hypothesized those subjects who exhibit elevated background inflammation may have smaller hippocampus, or have reduced brain reserve, and may be more prone to develop cognitive impairment after stroke.
Here, we report on the association between inflammatory markers, hippocampal volume, and cognitive functions.
Patients were consecutive eligible participants with mild–moderate stroke or transient ischemic attack in our prospective TABASCO (Tel Aviv Brain Acute Stroke Cohort) study.8 Approval was obtained from the local Ethics Committee, and participants signed informed consent forms.
Between April 2008 and February 2012, a total of 479 consecutive patients were admitted to our center within 72 hours after onset of symptoms with a diagnosis of transient ischemic attack or first-ever mild–moderate stroke and met the study inclusion/exclusion criteria. Neurological assessment includes verification of stroke pathogenesis and the NIH Stroke Scale. Participants who had MRI scans within 7 days from symptoms onset without temporal lobe infarcts and who had no evidence of inflammatory condition within 72 hours were included (n=255; see Figure I in the online-only Data Supplement).
All patients completed a baseline neuropsychological assessment (Montreal Cognitive Assessment (MoCA) and Mindstreams computerized neuropsychological battery).9
Venous blood was drawn for erythrocyte sedimentation rate (ESR), serum C-reactive protein (CRP), plasma fibrinogen, and white blood cell count at baseline and repeated 6 and 12 months later.
To test the hypothesis that elevated ESR and CRP were associated with smaller hippocampal volume and cognitive scores, multiple linear regression models were carried out.
Detailed Methods are provided in the online-only Data Supplement.
Table I in the online-only Data Supplement summarizes study population characteristics.
Associations Between Inflammatory Markers and Hippocampal Volume
The Table shows potential contributors to hippocampal volume at baseline. Among inflammatory markers, only ESR was associated with hippocampal volume. Multiple regression analyses confirmed this relationship (P=0.049, R2=0.543; Table). Patients with smaller hippocampi (lower quartile, 6722.2+479.7 mm3) had significantly higher ESR than those with larger hippocampi (upper quartile, 8866.1+428.9 mm3) at all time points (Figure). Tests with Bonferroni correction analysis of variance revealed that lower and upper quartiles differed significantly across all time points (P=0.002). No association was found between CRP, white blood cell count, or fibrinogen and hippocampal volume.
Associations Between Inflammatory Markers and Cognitive Performance
Elevated ESR was associated with inferior performance in the computerized total cognitive (r=–0.160, P=0.045), memory (r=–0.189, P=0.017), visuospatial (r=–0.164, P=0.042), and the MoCA (r=–163, P=0.023) score.
As hippocampus plays a key role in memory formation, we tested the association of ESR with memory scores after controlling for age, sex, hypertension, diabetes mellitus, dyslipidemia, total intracranial volume, presence of apolipoprotein E ε4 (ApoE ε4) allele, NIH Stroke Scale, years of education, hematocrit, body mass index, and time from symptom to blood sampling, in a multiple regression analysis (β=0.451, SE=6.023, P=0.002; see Table II in the online-only Data Supplement).
Elevated CRP levels were associated with decreased performance in the total cognitive (r=–0.283, P<0.001), memory (r=–0.173, P=0.029), visuospatial (r=–0.259, P=0.001), and verbal function scores (r=–0.236, P=0.003), but the association vanished after adjustment. No association was found between CRP and MoCA scores.
Decreased cognitive performance was associated with elevated white blood cell count in the total cognitive (r=–0.180, P=0.024) and the memory (r=–0.179, P=0.025) scores, but vanished after adjustment. No association was found between white blood cell count and MoCA scores.
Fibrinogen concentrations were related to lower MoCA and visuospatial scores (r=–0.193, P=0.015; r=–0.202, P=0.024, respectively).
Among the inflammatory markers, only ESR remained significant contributor to memory performance after adjustment.
Associations Between Hippocampal Volume and Cognitive Performance
Positive associations was observed between hippocampal volume and performance on total cognitive, memory, executive function, and visuospatial scores (r=0.244, P=0.011; r=0.209, P=0.029; r=0.255, P=0.008; r=0.346, P<0.001, respectively).
We found higher ESR values to be associated with worse performance in cognitive tests and with reduced hippocampal volume in stroke survivors, whereas no association was observed with CRP. As ESR values did not change significantly in 1 year, they likely represent chronic systemic inflammatory processes, rather than a consequence of the acute event, whereas CRP levels were significantly higher on admission compared with 6 and 12 months poststroke. These findings are in line with a previous report that higher ESR levels were associated with decreased cognitive performance in healthy young adults.10
ESR has largely been ignored in the context of inflammation and cognition. It is less subject to rapid changes than other inflammatory markers, such as CRP, and, accordingly, is likely to be a more stable indicator of chronic inflammation.11
These findings raise the possibility that many individuals with cerebrovascular ischemic events harbor a long and persistent proinflammatory background. Elevated ESR levels and reduced hippocampal volume may be related to higher age. Nevertheless, age was entered as a potential covariate, whereas ESR still emerged as a significant contributor to both hippocampal volume and memory scores. The mechanisms by which peripheral inflammatory markers relate to hippocampal volume remain unclear.
Our study only included mild–moderate stroke patients, who are expected to perform cognitive tests and be available for follow-up. We also acknowledge the lack of normal controls.
Future studies should focus on the specific inflammatory patterns associated with aging and brain atrophy to identify modifiable factors that may confer risk for accelerated poststroke cognitive aging.
Whether lowering inflammation and ESR can also prevent poststroke cognitive decline12 needs to be addressed in further clinical trials.
Sources of Funding
This study is supported by grant 3000000-5062 from the Chief Scientist Office, Ministry of Health, Israel and grant RAG11482 from the American Federation for Aging Research.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.111.000536/-/DC1.
- Received December 17, 2012.
- Accepted December 28, 2012.
- © 2013 American Heart Association, Inc.
- Kiecolt-Glaser JK,
- Preacher KJ,
- MacCallum RC,
- Atkinson C,
- Malarkey WB,
- Glaser R
- Rothenburg LS,
- Herrmann N,
- Swardfager W,
- Black SE,
- Tennen G,
- Kiss A,
- et al
- Doniger GM,
- Dwolatzky T,
- Zucker DM,
- Chertkow H,
- Crystal H,
- Schweiger A,
- et al