Article Text

Download PDFPDF

Cerebrovascular disease
Endothelial and inflammatory markers in relation to progression of ischaemic cerebral small-vessel disease and cognitive impairment: a 6-year longitudinal study in patients with type 2 diabetes mellitus
  1. T Umemura1,
  2. T Kawamura2,
  3. H Umegaki3,
  4. S Mashita4,
  5. A Kanai2,
  6. T Sakakibara1,
  7. N Hotta2,
  8. G Sobue5
  1. 1Department of Neurology, Chubu Rosai Hospital, Nagoya, Japan
  2. 2Department of Diabetes and Endocrine Internal Medicine, Chubu Rosai Hospital, Nagoya, Japan
  3. 3Department of Geriatrics, Nagoya University Graduate School of Medicine, Nagoya, Japan
  4. 4Department of Radiology, Chubu Rosai Hospital, Nagoya, Japan
  5. 5Department of Neurology, Nagoya University Graduate School of Medicine, Nagoya, Japan
  1. Correspondence to Dr Toshitaka Umemura, Department of Neurology, Chubu Rosai Hospital, 1-10-6 Komei, Minato-ku, Nagoya City, Aichi 455-8530, Japan; t.umemura{at}bg7.so-net.ne.jp

Abstract

Background Progression of silent brain infarctions (SBIs) and white-matter lesions (WMLs) seen on brain MRI is associated with an increased risk of cognitive impairment, but their relation to endothelial and inflammatory markers is unknown in type 2 diabetes mellitus.

Methods In 190 type 2 diabetic outpatients (mean age 62.7 years), the authors related baseline levels of soluble intercellular adhesion molecule-1 (sICAM-1) and high-sensitivity C-reactive protein (hs-CRP) to subsequent brain MRI findings and cognitive function. The authors assessed incident SBIs and changes in periventricular and subcortical WMLs (PVWMLs and SCWMLs) on MRI performed at baseline and 3 and 6 years. Neuropsychological tests were administered to 83 patients older than 65 years at 6 years. This present study represents an extension of the authors' previously published study.

Results SBIs were observed in 46 patients (24.2%), PVWMLs in 93 (48.9%) and SCWMLs in 87 (45.8%) on baseline MRI. After adjustment for age, gender, hypertension, duration of diabetes, baseline MRI findings and medication use, the relative odds associated with a 1SD increase in sICAM-1 levels at baseline were 1.67 (95% CI 1.02 to 3.05) for SBI progression and 2.17 (95% CI 1.29 to 3.62) for PVWML progression at 6 years. In contrast, baseline hs-CRP levels were significantly associated with SBI progression only at 3 years. Significant trends were observed between quartiles of sICAM-1 at baseline and scores in Digit Symbol substitution (p for trend=0.01).

Conclusions The findings suggest that higher sICAM-1 levels are associated with SBI and PVWML progression, and may predict impairment in psychomotor function in type 2 diabetes.

  • Small-vessel disease
  • soluble intercellular adhesion molecule-1
  • high sensitivity C reactive protein
  • diabetes mellitus
  • cognitive impairment
  • cerebrovascular disease
  • MRI
  • vascular dementia

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Silent brain infarctions (SBIs) and cerebral white-matter lesions (WMLs), often found incidentally on brain MRI of older individuals, are thought to be small-vessel disease-related lesions. Several reviews and studies have shown that these lesions are potential predictors of future stroke and cognitive decline.1–4 Ageing and hypertension are widely accepted risk factors for SBIs and WMLs. Although diabetes is an established risk factor for stroke, its relationship to the presence and progression of SBIs and WMLs is still controversial.5 6 In the Rotterdam Scan Study, diabetes was associated with new SBIs in participants without prevalent infarcts.5 Other studies also found that diabetes was a risk factor for the prevalence or progression of WMLs.7 8 However, the mechanism underlying the progression of cerebral small-vessel disease (SVD) in patients with diabetes remains unknown.

Although the pathogenesis of cerebral SVD is incompletely understood, endothelial activation or inflammatory processes may be involved in the pathogenesis of cerebral SVD. In a previous study, we determined the levels of soluble forms of the intercellular adhesion molecule-l (sICAM-l) expressed on vascular endothelial cells during the early phase of atherosclerosis,9 and we reported that an increase in sICAM-1 levels might contribute to the incidence of SBI in older patients with diabetes.10 We also reported that higher levels of sICAM-1 were an independent risk factor for SBI progression in a 3-year longitudinal observational study.11 However, a correlation between endothelial markers and change in WML scale was not assessed in our previous study. In contrast, high-sensitivity C-reactive protein (hs-CRP), which is considered a subclinical inflammatory marker, has attracted great attention in the literature as a predictive factor for future stroke.12–14 Although several studies have indicated an association between hs-CRP levels and the progression of SBIs or WMLs, these results are still controversial.15 16 In addition, other studies have suggested that there could be a relationship between inflammatory markers and cognitive function.17 18

In recent years, the prevalence of diabetes has increased worldwide, and interest has focused on the relationship between diabetes and cognitive decline in the scientific community.19 20 However, an association of endothelial and inflammatory markers with changes in brain MRI findings and cognitive function has not been previously investigated in a long-term study involving patients with diabetes. We therefore sought to conduct the current longitudinal study, which represents an extension of our previously published study. The aim of the present study was to examine whether endothelial and inflammatory markers could serve as predictors for SBI and WML progression and cognitive impairment in patients with type 2 diabetes.

Methods

Study population

For this study, we recruited 219 consecutive patients with type 2 diabetes mellitus who were 45 years of age or older and had no history of prior stroke or transient ischaemic attack from a database at Chubu Rosai Hospital's Diabetic Center between October 1995 and December 2003. After changes in the MRI systems at our hospital between 2000 and 2001, a standard protocol was to change proton images into fluid-attenuated inversion recovery (FLAIR) images. We sought to evaluate changes in WMLs using FLAIR images, and 29 patients who dropped out before 2000 were excluded from this study. Finally, samples were collected from the 190 outpatients with type 2 diabetes (88 males and 102 females; mean age 62.7±8.1; mean duration of diabetes 13.1 years) who were free of dementia at baseline in this longitudinal study. Participants underwent baseline brain MRI scans between 2001 and 2003, and the MRI was repeated after 3 and 6 years to determine the progression of SVD. We simultaneously measured levels of sICAM-1 and hs-CRP. Patients were excluded from the study if they met the following criteria: malignant disease, chronic and/or acute inflammatory disease, cardiovascular disease (including myocardial infarction, unstable angina, and atrial fibrillation), and severe microvascular complications, such as renal failure (serum creatinine >2.0 mg/dl). Individuals with hypertension were defined as those previously diagnosed as having hypertension and currently undergoing treatment with an antihypertensive agent, or as those with a blood pressure greater than 140/90 mm Hg in three consecutive measurements. Informed consent was obtained from each of the participants, and the Ethics Committee of Chubu Rosai Hospital approved the present study. Furthermore, this study was performed in accordance with the principles of the Declaration of Helsinki.

Biochemical characteristics

Various parameters (table 1) were measured, and serum was stored at −70°C until assayed for adhesion molecules. Total cholesterol (T-chol), HDL-cholesterol (HDL-chol), triglycerides and fasting blood glucose were quantified by an autoanalyser using routine enzymatic techniques. HbA1c (normal range 4.3–5.8%) was measured using the latex-enhanced competitive immunoturbidimetric assay. Plasma insulin was assayed by reactive immunoassay.

Table 1

Baseline characteristics of patients with and without small-vessel disease progression at 3 and 6 years

To determine sICAM-l levels, serum samples were analysed in duplicate using the single-step sandwich ELISA method (R&D Systems Europe, Abingdon, UK) with a Model 550 Microplate Reader (Bio-Rad, Hercules, California). Levels of hs-CRP were assayed using a monoclonal antibody coated with polystyrene particles and fixed-timed kinetic nephelometric measurements (BN II, Dade Behring, Marberg, Germany).

Diagnosis of SBIs and WMLs by MRI

An MRI was performed using a 1.5 T Signa Horizon (GE Medical Systems, Milwaukee, Wisconsin). The imaging protocol consisted of T1-weighted images (TR 2380 ms, TE 27.4 ms), T2-weighted images (TR 4017 ms, TE 103 ms) and FLAIR images (TR 8002 ms, TE 146 ms) with a slice thickness of 5 mm and an interslice gap of 2 mm. Baseline and follow-up MRI were conducted using identical protocol and sequences. We defined SBIs as areas of focal hyperintensity larger than 3 mm in diameter detected on T2-weighted images, hypointensity areas on T1-weighted images and areas of hypointensity surrounded by hyperintense rim on FLAIR images. Lesions less than 3 mm in diameter or with a signal intensity similar to that of cerebrospinal fluid on FLAIR images were excluded because of the high possibility of enlarged perivascular spaces, even if hyperintensity on T2-weighted images and hypointensity on T1-weighted images were determined. Progression of SBI was defined as newly diagnosed SBI or an increase in one number or more between baseline and 3- or 6-year follow-up. Baseline WMLs were stratified separately for periventricular and subcortical areas. Periventricular WMLs (PVWMLs) and subcortical WMLs (SCWMLs) were rated according to the Fazekas scale (PVWML: grade 0, no lesion; grade 1, caps or pencil-thin lining; grade 2, smooth halo; grade 3, large confluence extending into deep white matter; and SCWML: grade 0, no lesion; grade 1, punctate; grade 2, early confluent; grade 3, confluent).21 Progression of WMLs was rated on FLAIR images using the modified Rotterdam Progression scale,22 in which the absence or presence of progression (0 or 1, respectively) was rated in three periventricular regions (frontal caps, lateral bands, occipital caps) and four subcortical regions (frontal, parietal, temporal and occipital). Progression of WML change scale was defined as an increase in one point or more on the scale between baseline and 3- or 6-year follow-up. Figure 1 illustrates an example of SVD progression on brain MRI. The neurologist and radiologist, who were blinded to laboratory and clinical data, assessed the existence, location and extension of the PVWML or SCWML on baseline and follow-up MRI. Similar assessments were determined for SBI. Each value of inter-rater reliability for MRI findings, which was expressed as Cohen κ, was within the range of 0.75 to 0.86. A 3-year follow-up MRI was not possible for 18 patients (ischaemic stroke eight, haemorrhagic stroke one, subject refusal one and no response to contact attempts eight) and a 6-year follow-up MRI was not possible for 24 patients (death four, ischaemic stroke four, myocardial infarction four, subject refusal four and no response to contact attempts eight).

Figure 1

Example of small-vessel disease progression on brain MRI. Figures show the baseline and 6-year follow-up MRI (A, D; T2-weighted image, B, C, E, F; fluid-attenuated inversion recovery image). New lacunes in the basal ganglia (arrows) appeared on the follow-up MRI (D). Pencil-thin lining on white-matter lesions at baseline (B). Large confluence extended into deep white-matter lesions at the 6-year follow-up (E). The arrowhead (C) shows a punctate white-matter lesions at baseline, and a confluent white-matter lesions appeared on the same lesions in the follow-up MRI (F).

Assessment of cognitive function

To assess cognitive function, the Mini Mental State Examination (MMSE) was administered to 90 older patients with diabetes who were 65 years of age or older, and informed consent was obtained at the time of a 6-year follow-up. Seven patients with severe cognitive impairment (MMSE score <19 points)23 were not administered the additional detailed cognitive test that we reported in our previous study.24 The Word Recall subset (part of the Alzheimer's Disease Assessment Scale)25 was performed using a verbal memory score range from 0 to 10. The Digit Symbol Substitution Test (DSST), a subtest of the Wechsler Adult Intelligence Scale—Revised, was utilised to evaluate complex psychomotor skills using a score range from 0 to 93.26 The modified Stroop Colour Word Test (mStroop)27 was utilised to assess attention. The seconds required to complete the test were recorded, and the difference between the time required to read the word card and the time required to read the dot card was calculated. The same trained psychological testers administered all neuropsychological tests.

Statistical analysis

Differences between groups were examined by χ2 tests and Fisher exact tests for categorical data and unpaired t tests for continuous data. Because the levels of hs-CRP, immunoreactive insulin and triglycerides were not normally distributed, differences in the levels were tested by non-parametric statistical procedures (Mann–Whitney U test), and the results were shown as the median (IQR 25–75%). We examined the association between markers and lesion progression by multivariate logistic regression analysis, which was adjusted for age, gender, hypertension, duration of diabetes, baseline MRI findings (presence of SBIs and severity of WMLs) and medication use. Hs-CRP levels were analysed as log-transformed values. We also estimated the association between the progression of SVD and cognition scores by analysis of covariance, which was adjusted for age, gender and education level. A two-tailed p value of <0.05 was considered significant. All analyses were performed using SPSS (version 11.0J).

Results

Baseline characteristics of patients with and without cerebral small-vessel disease (SVD) progression at 3 and 6 years

The baseline characteristics of patients with different subtypes of cerebral SVD are shown in table 1. At 6-year follow-up, four patients had died, 17 had suffered stroke or myocardial infarction, and 21 had dropped out of the study. Therefore, follow-up MRI examinations were available for 172 patients at the 3-year and 148 at the 6-year follow-up. Age was significantly higher in patients with SVD progression than in those without at both 3 and 6 years. The frequency of hypertension was significantly higher in patients with SBI progression than in those without at both 3 and 6 years. Furthermore, systolic blood pressure was significantly higher in patients with SBI and SCWML progression than in those without at both 3 and 6 years. As shown in the upper row of figure 2A,B, baseline sICAM-1 levels were significantly higher in patients with SVD than in those without. In contrast, there were no significant relationships between the presence of SVD and baseline hs-CRP levels. As shown in the middle and lower row of figure 2A,B, baseline sICAM-1 levels were significantly higher in patients with progression of SBI and WML at 3 years compared with those without progression of SBI and WML (SBI: 270.7±102.3 vs 219.4±71.6 μg/l, p<0.001; PVWML: 283.8±70.2 vs 224.4±81.8 μg/l, p<0.01; SCWML: 271.9±110.8 vs 223.8±73.9 μg/l, p<0.01). At 6 years, baseline sICAM-1 levels were significantly higher in patients with progression of SBI and PVWML than in those without (SBI: 252.5±66.4 vs 216.8±73.3 μg/l, p<0.05; PVWML: 262.0±78.7 vs 217.4±69.5 μg/l, p<0.01). These associations remained significant after adjustment for age. Between the 3- and 6-year follow-up, sICAM-1 levels at 3 years were only significantly higher in patients with PVWML progression compared with those without PVWML progression (238.0±63.9 vs 208.2±67.8 μg/l, p<0.05; data not shown in figure 2). In contrast, baseline hs-CRP levels were significantly higher in patients with SBI progression at 3 and 6 years compared with those without SBI progression (0.88 (0.47–1.65) vs 0.53 (0.24–0.93) mg/l, p<0.01; 0.74 (0.34–1.31) vs 0.53 (0.24–0.98) mg/l, p<0.05). This association remained significant in patients with SBI progression at 3 years, even after adjustment for age.

Figure 2

Graphic representation of baseline soluble intercellular adhesion molecule-1 levels as dot plots (A) and high-sensitivity C-reactive protein levels as box plots (B) dichotomised by the presence or absence of a small-vessel disease subtype and divided according to whether small-vessel disease progression was present at 3 and 6 years (upper row: baseline; middle row: at 3 years; lower row: at 6 years). (A) Bars show mean (±SD) values. p Values are expressed after adjustment for age. (B) Bars show median values, boxes are the IQRs, and whiskers are ranges excluding statistical outliers. p Values are expressed after adjustment for age. NS, not significant. For the details on small-vessel disease progression, see table 2. PVWML, periventricular white-matter lesion; SBI, silent brain infarction; SCWML, subcortical white-matter lesion.

Changes in MRI findings according to SVD subtype during the 6-year follow-up

SBIs were observed in 46 patients (24.2%), PVWMLs in 93 (48.9%) and SCWMLs in 87 (45.8%) on baseline MRI. Thirty-eight patients (22.1%) had an incident SBI at 6-year follow-up, and approximately two-thirds of these infarcts were observed at the 3-year follow-up. Twenty-one patients (12.2%) had progression of PVWML, and 28 patients (16.3%) had progression of SCWML at the 3-year follow-up. After 6 years, 28 patients (18.9%) and 34 patients (23.0%) demonstrated PVWML and SCWML progression, respectively. The change in the SBI number at both the 3- and 6-year follow-up in patients with no lesions at baseline was less than that in patients with baseline SBI. The proportion of patients with WML progression was higher in patients who already had moderate or severe lesions at baseline than in those without lesions (table 2).

Table 2

Change in silent brain infarction number and white-matter lesion (WML) severity during the 6-year follow-up

RR for the progression of SVD associated with levels of endothelial and inflammatory markers at baseline

We used logistic regression analysis to calculate the ORs per 1SD increase in the level of each marker with the progression of SBI and WML. The ORs associated with a 1SD increase in sICAM-1 levels at baseline were 1.99 (95% CI 1.29 to 2.80) for SBI progression, 2.36 (95% CI 1.30 to 4.30) for PVWML progression and 1.83 (95% CI 1.19 to 2.81) for SCWML progression at the 3-year follow-up. The association remained significant after adjustment for age, gender, hypertension, duration of diabetes, baseline MRI findings and medication use (aspirin, ticlopidine, cilostazol, calcium-channel blockers, angiotensin receptor blockers, angiotensin-converting enzyme-inhibitors and statin). After adjustment for confounding variables, the ORs associated with a 1SD increase in sICAM-1 levels at baseline were 1.67 (95% CI 1.02 to 3.05) for SBI progression and 2.17 (95% CI 1.29 to 3.62) for PVWML progression at the 6-year follow-up. Furthermore, higher levels of sICAM-1 at 3 years were only associated with PVWML progression at the 6-year follow-up after adjustment for age, gender, hypertension, duration of diabetes, baseline MRI findings and medication use (ORs 1.99, 95% CI 1.09 to 3.62). In contrast, only the ORs associated with a 1SD increase in hs-CRP levels at baseline were 1.54 (95% CI 1.01 to 2.36) for SBI progression at the 3-year follow-up after adjustment for confounding variables (table 3).

Table 3

RR for progression of small-vessel disease associated with levels of endothelial and inflammatory markers at baseline

Relationship between SVD progression or endothelial and inflammatory markers and cognitive performance

After adjustment for age, gender and education level, the DSST scores were significantly lower in patients who showed progression of SBI and PVWML, compared with patients without progression of SBI and PVWML (29.1±1.8 vs 35.5±1.4, p<0.01; 29.3±2.1 vs 35.0±1.4, p<0.05, respectively). In addition, the time required to perform the mStroop task was significantly higher in patients who showed progression of SBI and PVWML, compared with patients without progression of SBI and PVWML (24.8±1.4 vs 19.8±1.1, p<0.01; 24.7±1.7 vs 20.2±1.1, p<0.05, respectively). However, there were no significant differences in the MMSE and Word Recall scores between the patients with and without SVD progression. A significant relationship was found between the quartile of sICAM-1 levels at baseline and the DSST score (p for trend=0.01), but there was no significant relationship between the MMSE score and the quartile of sICAM-1 or hs-CRP levels at baseline. The DSST scores for patients in the highest quartile of sICAM-1 levels at baseline were significantly lower than those in the lowest quartile after adjustment for age, gender and education level (figure 3).

Figure 3

Association of baseline soluble intercellular adhesion molecule-1 (sICAM-1) and high-sensitivity C-reactive protein (hs-CRP) quartile levels with cognition scores. Bars represent SEs. Significant differences from the first quartile after adjustment for age, gender and level of education are shown. *p<0.05; **p<0.01. DSST, Digit Symbol Substitution Test; MMSE, Mini-Mental State Examination.

Discussion

We found that baseline sICAM-1 levels were associated with the progression of SBIs and PVWMLs in this 6-year longitudinal study in patients with type 2 diabetes. In contrast, baseline hs-CRP levels were only significantly associated with SBI progression at 3 years. The present study therefore indicates that associations between these two biomarkers with SVD progression differ in clinical implication.

Levels of sICAM-1 attenuated during the follow-up examination because patients developing ischaemic stroke were excluded from 3- to 6-year follow-up analyses. Nevertheless, sICAM-1 levels at baseline remained associated with the progression of SBIs and PVWMLs at both 3 and 6 years, even after adjustment for confounding variables. The strong association between sICAM-1 levels and SBIs in patients with diabetes may result from the following mechanism: ICAM-1 expression is induced in the vascular endothelium by elevated blood glucose levels; ICAM-1 then promotes the adhesion of leucocytes, especially neutrophils, to the vascular endothelium, which causes small vessels in the brain to occlude, and leads to incidence of SBIs. In a previous study, we found that an association between endothelial dysfunction and the presence of SBI may be stronger in patients with diabetes than in subjects without diabetics.28 Several cross-sectional studies29 30 also assessed a correlation between sICAM-1 levels and WMLs. To our knowledge, only one other longitudinal study has explored the relationship between endothelial activation and inflammation with the progression of WMLs. Markus et al31 indicated that sICAM-1 levels were related to progression of white-matter hyperintensities in a 6-year community-based prospective study. In contrast, our study has shown that the relationship between sICAM-1 levels and SCWML progression was only observed at the 3-year follow-up. Wardlaw et al32 hypothesised that activation of the endothelium leads to increased permeability of the blood–brain barrier, which may play a role in the pathogenesis of cerebral SVD. Furthermore, endothelial dysfunction can be exacerbated when oxidative stress enhances blood–brain barrier permeability in patients with diabetes, which may promote development of cerebral SVD. Therefore, our results provide new insights that endothelial dysfunction plays an important role in WML progression in patients with diabetes. Moreover, the present study suggests the possibility of a difference in the degree of endothelial dysfunction between PVWMLs and SCWMLs.

On the other hand, recent studies have shown that hs-CRP levels are elevated in patients with SBIs.13 33 Increased hs-CRP levels are thought to reflect systemic inflammation; its elevation in the microvasculature of the brain may indicate arteriosclerosis progression. However, higher hs-CRP levels were associated with the presence and progression of WMLs, but not with incident lacunar infarcts in the Rotterdam Scan Study.15 These results indicate that elevated CRP levels may be related to WMLs and lacunar infarcts via different mechanisms.15 In our longitudinal study, patients with SBI progression at the 3-year follow-up exhibited higher hs-CRP levels at baseline. However, we failed to demonstrate any significant relationship between hs-CRP levels and WML progression. In this regard, a recent study suggested that a relationship between hs-CRP levels and SVD-related lesions was not apparent in community-based older Japanese subjects.16 Thus, measurement of hs-CRP might have greater value as a predictive marker of vascular events than SVD progression. However, racial differences may potentially have an influence on the different association of hs-CRP levels with the presence and progression of cerebral SVD.

Several studies showed the relationship between changes in MRI findings of SVD and cognitive function.34–38 In the Rotterdam Scan Study, repeat head MRI scans were performed on 629 participants to examine the relationship between MRI changes and decline in cognitive function. SBIs were determined to be associated with memory deterioration and decline in psychomotor function.34 Prins et al35 suggested that cerebral SVD might contribute to cognitive decline in terms of information-processing speed and executive function. Furthermore, de Groot et al36 reported that cognitive decline was associated with periventricular WMLs, but not subcortical WMLs. However, an association of vascular endothelial and inflammatory markers with changes in MRI findings of SVD was not assessed in these studies. Biessels et al39 suggested that microvascular disease was also associated with the pathogenesis of dementia in patients with diabetes. A recent autopsy study also showed that patients with diabetes with dementia had more microvascular infarcts than those without dementia.40 Biomarkers related to SVD progression may therefore play an important role in the accelerated development of cognitive impairment in older people with diabetes. Our results have shown that elevated sICAM-1 levels are associated with lower DSST scores in addition to progression of SBIs and PVWMLs. Therefore, sICAM-1 may be a potentially useful biomarker to predict impairment in psychomotor function in patients with diabetes.

Certain methodological limitations of the present study should be addressed. First, this investigation was limited to patients with diabetes, and therefore, an association between endothelial and inflammatory markers with changes in MRI findings in subjects without diabetes should also be assessed. Second, an association between endothelial and inflammatory markers at baseline with changes in cognitive function could not be assessed because detailed cognitive function was only evaluated at a single time point, a 6-year follow-up examination. Third, the small sample size precluded the identification of a statistically significant association of endothelial and inflammatory markers at baseline with SVD progression according to SBI numbers and WML severity.

In conclusion, our findings suggest that sICAM-1 levels serve as a surrogate marker of the progression of SBIs and PVWMLs, and that this marker may predict impairment in psychomotor function in patients with type 2 diabetes. Therefore, further prospective studies are required to investigate whether reducing plasma sICAM-1 levels might prevent the progression of SVD and cognitive impairment.

References

Footnotes

  • Funding This study was supported by research funds provided to promote the hospital functions of the Japan Labor Health and Welfare Organization.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics approval was provided by the Ethics Committee of Chubu Rosai Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.