Arterial Stiffness and Progressive Neurological Deficit in Patients With Acute Deep Subcortical Infarction
Background and Purpose—The mechanism of progressive neurological deficit (PND) in patients with ischemic stroke remains unclear. The aim of this study was to clarify whether arterial stiffness, a marker of vascular endothelial impairment and arteriosclerosis, is associated with PND in patients with acute deep subcortical infarction.
Methods—We evaluated 156 consecutive first-ever ischemic stroke patients with acute deep subcortical infarction. PND was defined as an increment of ≥2 points in the National Institute of Health Stroke Scale score or an increase of ≥1 point in the limb weakness score within 7 days of stroke onset. Patients were assessed for risk factors, and infarct size was measured on initial diffusion-weighted magnetic resonance imaging. We measured brachial-ankle pulse wave velocity (baPWV) as a marker of arterial stiffness. We divided patients into 2 groups according to the presence or absence of PND to compare their clinical characteristics.
Results—Fifty-two patients (33%) had PND, and baPWV was significantly higher in patients with than in those without PND. The baPWV cut-off value for PND was 18.24 m/s, with 90% sensitivity and 47% specificity. In multivariable logistic regression analysis, high baPWV (≥18.24 m/s; odds ratio, 8.22; 95% confidence interval, 2.55–31.9), large infarct size (≥15 mm; odds ratio, 2.76; 95% confidence interval, 1.01–7.92), and ≥3 infarct slices on serial axial diffusion-weighted imaging (odds ratio, 3.38; 95% confidence interval, 1.22–10.0) were independently associated with PND.
Conclusions—Arterial stiffness indicated by baPWV is independently associated with PND in patients with acute deep subcortical infarction.
- acute stroke
- arterial stiffness
- blood–brain barrier
- deep subcortical infarct
- progressive neurological deficit
- pulse wave velocity
Deep subcortical infarction sometimes presents with progressive neurological deficit (PND). The etiology of PND is the occlusion of perforating arteries attributable to cerebral small vessel disease.1–8 Small vessel disease is associated with arterial stiffness, a marker of both arteriosclerosis and vascular endothelial impairment, and can cause neurological damage.8–13 A noninvasive measurement of arterial stiffness, brachial-ankle pulse wave velocity (baPWV), is independently associated with cerebral small vessel disease such as silent lacunar infarction12 and white matter disease.13
The mechanism of PND in patients with acute deep subcortical infarcts is still unclear. Arterial stiffness indicates vascular endothelial impairment, which causes blood–brain barrier failure and leads to neurological damage.8 This mechanism could increase the risk of PND. Our hypothesis was that arterial stiffness is independently associated with PND in patients with acute deep subcortical infarcts. To test this hypothesis, we investigated the association between arterial stiffness, conventional risk factors, and PND in patients with a first-ever deep subcortical infarct.
Materials and Methods
This study was approved by the Institutional Review Board at the Hyogo Brain and Heart Center at Himeji. Between October 2003 and March 2010, we enrolled 156 consecutive patients presenting with clinical lacunar stroke syndrome who were admitted to our hospital within 48 hours after stroke onset. All patients had a first-ever ischemic stroke attributable to an acute deep subcortical infarct detected on initial diffusion-weighted imaging.4 Informed consent was obtained from all patients.
Infarcts <15mm in diameter on diffusion-weighted imaging scans were considered small infarcts and those >15mm were considered large infarcts.5 Infarct slice number was first defined as the number of slices accompanied by visible infarcts on serial axial diffusion-weighted imaging scans and then used to evaluate the vertical extension of those infarcts.3 We measured baPWV at 7 days after the onset or stabilization of the neurological deficit using an oscillometric device (Form PWV/ABI; Omron Colin, Tokyo, Japan).10–13 PND was defined as an increase of ≥2 points in the National Institutes of Health Stroke Scale score or increase of ≥1 point in limb weakness in the National Institutes of Health Stroke Scale score during the 7days after stroke onset.3–7 Excellent outcome after stroke onset was defined as a modified Rankin Scale score of 0 to 1 at discharge.9 Univariable and multivariable logistic regression models were used to identify the variables independently associated with PND. Odds ratios are presented along with 95% confidence intervals. All comparisons were 2-tailed and P<0.05 was considered significant. Detailed and Methods are provided in the Online Supplement.
Fifty-two patients (33%) had PND. The average time to detect PND from the onset of ischemic stroke was 24 hours (interquartile range, 17–30 hours). Table 1 summarizes the characteristics of these patients. The baPWV cut-off value to detect PND was 18.24 m/s with 90% sensitivity and 47% specificity. Patients with high baPWV (≥18.24 m/s), those with large infarct (≥15mm), and those with infarct slice number ≥3 had more PND than the rest of the patients (Figure).
In multivariable analysis (Table 2), high baPWV (≥18.24 m/s), infarct size (≥15 mm), and infarct slice number ≥3 were independently associated with PND after adjustment for age, sex, systolic blood pressure on admission, National Institutes of Health Stroke Scale score, estimated glomerular filtration rate, diabetes mellitus, ischemic heart disease, and smoking (model 1). Furthermore, high baPWV (≥18.24 m/s) was associated with PND (odds ratio, 8.47; 95% confidence interval, 2.52–34.5) and inversely associated with excellent stroke outcome (odds ratio, 0.32; 95% confidence interval, 0.10–0.86) after adjustment for both the variables included in model 1 and the treatment factors (the use of cilostazol or clopidogrel, and argatroban hydrate). The details of these results are also provided in the Online Supplement.
The main finding of the present study was that arterial stiffness indicated by baPWV was independently associated with PND in patients with acute deep subcortical infarcts. The association between arterial stiffness and PND had not yet been reported before this study. Several mechanisms might explain this association. Vascular endothelial impairment, which is indicated by arterial stiffness,11–13 causes blood–brain barrier failure and leads to cerebral parenchyma damage,8,14 possibly inducing PND. Decreased production of protective factors such as nitric oxide and endogenous tissue plasminogen activator,7,15 lack of collateral blood flow,1 arteriosclerotic microcirculatory impairment,10 and stepwise occlusion of the cerebral perforating arteries2,7 may accelerate endothelial dysfunction. However, high baPWV could be merely a result of high blood pressure in the acute phase of ischemic stroke.11
The cut-off value of baPWV for PND (18.24 m/s) in the present study was the same as the cut-off of baPWV for cerebral small vessel disease shown in previous studies (cerebral small vessel disease, 17 m/s;11 silent lacunar infarction, 17.24 m/s;12 and white matter disease, 18.29 m/s13). These results support the hypothesis that baPWV is a marker for cerebral microvascular damage.10–13
Arterial stiffness is independently associated with functional outcome after stroke. High carotid-femoral PWV, a conventional measurement of arterial stiffness, is inversely associated with excellent functional outcome in patients with acute ischemic stroke.9 This finding is consistent with our results. The inverse association of baPWV with functional outcome is reasonable, because we also found an association between baPWV and PND.
There were several limitations of this study. A causal relationship between baPWV and PND could not be established, and selection bias was possible because this was an observational study performed in a hospital-based cohort. The low specificity (47%) of elevated baPWV to detect PND suggests that other mechanisms, such as neuroexcitotoxic amino acids, could also play a role in PND. This will need to be addressed in future studies.
Sources of Funding
This study was partially funded by The KAWASAKI Foundation for Medical Science & Medical Welfare.
Dr Saji received a research grant from The KAWASAKI Foundation for Medical Science & Medical Welfare.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.112.670737/-/DC1.
- Received July 20, 2012.
- Accepted August 3, 2012.
- © 2012 American Heart Association, Inc.
- Nakamura K,
- Saku Y,
- Ibayashi S,
- Fujishima M
- Steinke W,
- Ley SC
- Castillo J
- Gasecki D,
- Rojek A,
- Kwarciany M,
- Kubach M,
- Boutouyrie P,
- Nyka W,
- et al