Vascular Overload Index Predicts Cerebral Small Vessel Disease in a Neurologically Healthy Population

Article information

J Neurosonol Neuroimag. 2021;13(2):55-63

Publication date (electronic) : 2021 December 29

doi : https://doi.org/10.31728/jnn.2021.00105

Ki-Woong Nam, MD, MSc^*,†, Hyung-Min Kwon, MD, PhD^,*,†, Jin-Ho Park, MD, MPH, PhD^‡,§, Hyuktae Kwon, MD, PhD^‡,§

^*Department of Neurology, Seoul National University College of Medicine, Seoul, Korea

^†Department of Neurology, Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Seoul, Korea

^‡Department of Family Medicine, Seoul National University College of Medicine, Seoul, Korea

^§Department of Family Medicine, Seoul National University Hospital, Seoul, Korea

Correspondence: Hyung-Min Kwon, MD, PhD Department of Neurology, Seoul Metropolitan Government-Seoul National University Boramae Medical Center, Seoul National University College of Medicine, 20 Boramae-ro 5-gil, Dongjak-gu, Seoul 07061, Korea Tel: +82-2-870-2475 Fax: +82-2-831-2826 E-mail: hmkwon@snu.ac.kr

Received 2021 November 5; Revised 2021 December 6; Accepted 2021 December 6.

Abstract

INTRODUCTION

Cerebral small vessel diseases (cSVDs) are subclinical pathologies mainly found in the elderly, and their prevalence is increasing worldwide with the development of brain imaging technology (e.g., magnetic resonance imaging [MRI]).1 Although asymptomatic on their own, cSVDs are clinically important because they increase the risk of stroke and dementia.2,3 cSVD contains several subtypes with different pathologies (e.g., white matter hyperintensity [WMH], lacunes, and cerebral microbleeds [CMBs]), which are usually found coexisting in images.4 Therefore, many studies have been conducted to find a common pathological mechanism to explain the occurrence of different cSVD subtypes. Several possible mechanisms and risk factors have been proposed, one of which is arterial stiffness.5,6

The medial wall of arteries contains rich elastic fibers.7 With aging or prolonged exposure to elevated blood pressure (BP), these elastic fibers are degraded, smooth muscle cells proliferate, and extracellular matrix accumulates, which is called vascular remodeling.6,7 Through this process, the arteries gradually stiff. The index reflecting this, arterial stiffness, has shown a close relationship with stroke or cSVD in previous studies.6,8-10 Vascular overload index (VOI) is an indicator that can be calculated using systolic and diastolic BP, and is a comprehensive parameter that reflects the following three circulatory abnormalities: 1) increased arteriolar resistance, 2) increased large-artery stiffness, and 3) early or premature reflection of arterial pulse wave.11 Therefore, physiologically, this indicator is expected to predict cardiovascular/cerebrovascular risk better than isolated systolic or diastolic BP. However, studies examining the association between VOI and cerebrovascular disease have not yet been properly conducted.12

Strokes, once they occur, often leave a disability because neurological recovery is limited. Therefore, if it is possible to reduce the risk by preventing the progression of cSVD before stroke occurs, it will be helpful in the quality of life of the patient. In this study, we evaluated the association between VOI and cSVD in a neurologically healthy population. In addition, by comparing the association with VOI for each subtype of cSVD, we also evaluated whether arterial stiffness is a common mechanism encompassing all the pathologies of cSVD or has a correlation only with a specific pathology.

SUBJECTS AND METHODS

RESULTS

A total of 3,231 participants were evaluated (mean age 57±9 years, male sex 53.9%). The mean WMH volume were 2.63±6.04 mL. The prevalence of lacunes and CMBs were 242 (7.5%) and 134 (4.1%), respectively. The mean VOI value was 9.27±17.98 mmHg. Other detailed baseline characteristics are presented in Table 1. When looking at the comparison between tertiles, the VOI level showed a correlation with age, male sex, body mass index, hypertension, diabetes, current smoking, systolic and diastolic BP, pulse pressure, hemoglobin A1c, fasting glucose level, total cholesterol level, WBC counts, WMH volume, lacunes, and ICAS (Table 2).

Table 1.

Baseline characteristics of the study population

Table 2.

The differences of baseline characteristics according to the vascular overload index tertiles

In univariate linear regression analysis, WMH volume were related to age, hypertension, diabetes, current smoking, systolic and diastolic BP, VOI, hemoglobin A1c, fasting glucose level, total cholesterol level, WBC counts, ICAS, and ECAS. In multivariable linear regression analysis, VOI was significantly associated with WMH volume after adjusting confounders (β=0.004, 95% confidence interval [CI]=0.002–0.006). Age (β=0.050, 95% CI=0.046–0.054), hypertension (β=0.147, 95% CI=0.062–0.233), diabetes (β=0.174, 95% CI=0.027–0.321), hemoglobin A1c (β=-0.567, 95% CI=-1.067 to -0.067), WBC counts (β=0.049, 95% CI=0.027–0.070), and ECAS (β=0.330, 95% CI=0.011–0.648) were also showed close correlation with WMH volume, being independent from VOI (Table 3).

Table 3.

Simple and multiple linear regression analyses between possible predictors and the square root of white matter hyperintensity volume^*

In multivariable logistic regression analysis, VOI was closely associated with the presence of lacunes (adjusted odds ratio [aOR]=1.01, 95% CI=1.00–1.02) (Table 4). However, VOI did not show any statistical significance with CMBs (aOR=1.00, 95% CI=0.99–1.01) (Table 5). Even when subgroup analyses of CMBs were performed according to the location of the lesions, VOI showed only statistically marginal tendency with deep CMBs, but was not significant. There was no statistical tendency or significance between lobar CMBs and VOI (Table 6).

Table 4.

Univariate and multivariate logistic regression analyses between possible predictors and lacunes

Table 5.

Univariate and multivariate logistic regression analyses between possible predictors and cerebral microbleeds

Table 6.

Multivariable logistic regression analyses between possible predictors and lobar/deep cerebral microbleeds

This close relationships between VOI and WMH volume/lacunes were only found in participants without hypertension. In the presence of hypertension, there was no statistical significance between those factors. CMBs were not associated with VOI with or without hypertension (Table 7).

Table 7.

Multivariable linear and logistic regression analysis to evaluate association between vascular overload index and cerebral small vessel diseases according to the presence of hypertension

DISCUSSION

In this study, we found that high VOI was associated with cSVD in a neurologically healthy population. This close association was more pronounced in participants without hypertension. Therefore, our results can be interpreted as suggesting the possibility that VOI may predict the risk of developing cSVD in people not taking anti-hypertensives.

The exact pathophysiological mechanisms explaining the close association between VOI and cSVD are unclear. However, we could suggest several possible hypotheses. First, VOI is a parameter of arterial stiffness, which could be related to the main mechanisms of WMH and lacunes development. Arteries stiffened by aging and long-standing hypertension do not reduce pulse energy according to the Windkessel effect like normal arteries, but propagate it to the perforating arterioles.6 When cerebral autoregulation is damaged in this way, several pathological brain environments are created as follows: 1) endothelial dysfunction occurs, 2) the brain-blood barrier is disrupted and toxic substances infiltrate into the periventricular space, and 3) solute clearance through the glymphatic pathway is impaired.16-20 If these various mechanisms are affected over a long period of time, WMH or lacunes may develop as a result. Second, it can also be explained by the association between VOI and large-vessel atherosclerosis. As mentioned earlier, VOI is an index that reflects not only arteriolar stiffness or premature reflection of the arterial pulse wave, but also the stiffness of the large artery.11 This is actually evidenced by the close association between VOI and ICAS in our data (Table 2). The chronic diffuse hypoperfusion condition caused by ICAS can sufficiently contribute to the development of WMH and lacunes.21 Last, the association between VOI and cSVD may be a coincidence produced by sharing risk factors. Indeed, VOI showed statistically significant associations with several well-known cerebrovascular risk factors (Table 2).

Interestingly, VOI showed close relationships with WMH or lacunes, but not with CMBs. This is probably due to the heterogeneous nature of CMB pathologies.4 CMB is divided into deep and lobar CMBs, and deep CMB is considered to be a pathology related to hypertensive damage.4 However, lobar CMB is considered to be amyloid angiopathy and therefore has a close association with endothelial dysfunction or glymphatic clearance.22 Therefore, the mere fact that these two pathologies are mixed may explain why CMBs appear unrelated to VOI. Actually, when we separately analyzed deep CMBs and lobar CMBs, it was not statistically significant, but deep CMBs showed statistical tendency at VOI and p=0.090 level (Table 6). In addition, our study population was very young and there were many healthy people with rare cardiovascular risk factors. For this reason, the prevalence of CMB itself is low, which may lose statistical significance. It is necessary to verify this with another study population with a higher prevalence of CMBs.

As mentioned earlier, VOI showed associations with cSVDs only in those without a history of hypertension. This may be because patients with long-standing hypertension already have higher arterial stiffness and advanced cSVDs. In our data, participants with hypertension were older, had a higher VOI value, and a higher burden of cSVD than the rest. If arteriosclerosis and cSVD are already severely advanced due to aging and long-standing high pulse pressure, the influence of current BP may no longer be important. Another possibility is that a single value of VOI cannot serve as a good representative value in hypertensive patients. Patients with hypertension have varying BP depending on the type of anti-hypertensives, the time of taking it, and the dose. Therefore, a single value of VOI calculated using systolic and diastolic BP measured once may not properly represent the patient’s usual state. Nevertheless, VOI is a simple index that can be obtained using only a sphygmomanometer. We can easily measure our BP several times a day, every day. Using this easy accessibility, if the fluctuation- or average-value of VOI over a long period is obtained, it may be possible to more accurately predict the volume change of WMH or the additional occurrence of lacunes.

There are several limitations to interpret our results. First, this study is a retrospective cross-sectional study. Even if VOI is statistically close to WMH volume and lacunes, it only implies an association and does not guarantee a causal relationship. To prove this, further prospective studies are needed. Second, we used a single measure of VOI. BP is a value that changes depending on the measurement time, environment, and physical conditions. Therefore, it may have been possible to interpret the results more interestingly if the average or variation of the VOI values measured several times was used. Third, VOI is certainly a useful indicator, but it is not yet a widely used and familiar indicator. Therefore, it would have been more helpful for readers to understand if it included comparisons with well-known arterial stiffness indexes such as pulsatility index, pulse wave velocity, and augmentation index. Last, our study population is a relatively young and healthy population. Therefore, the influence of other major cardiovascular risk factors may be underestimated and the influence of VOI may be overestimated.

In conclusion, we identified that VOI was closely related to cSVD in a neurologically healthy population. Recently, the American Heart Association/American Stroke Association reclassified patients with BP previously classified as prehypertension as stage 1 hypertension to manage the risk of cerebrovascular and cardiovascular diseases at an early stage.1,23 That is, it is important to manage BP in patients at an early stage before they progress to severe hypertension. In this context, our findings can be interpreted as VOI is valuable as a predictor to evaluate the risk of early cerebrovascular disease in place of solitary systolic or diastolic BP. However, our results should first be validated by further studies.

Article information Continued