Circulating Vascular Cell Adhesion Molecule-1 (sVCAM-1) Is Associated with Cerebral Blood Flow Dysregulation, Mobility Impairment, and Falls in Older Adults

Abstract

Soluble Vascular Cell Adhesion Molecule-1 (sVCAM-1) is associated with hypertension, vascular inflammation, and systemic endothelial dysfunction. We evaluated whether elevated plasma sVCAM-1 is associated with impaired cerebrovascular function and mobility impairments in elderly. We studied the cross-sectional relationships between plasma sVCAM-1 level, gait speed, and cerebrovascular hemodynamics, and its longitudinal relationship with falls in 680 community-dwelling participants in the MOBILIZE Boston Study aged 65 and older. Falls were recorded prospectively for one year on daily calendars. sVCAM-1 was measured by ELISA assay and beat-to-beat blood flow velocity (BFV) in the middle cerebral artery during rest and in response to changes in end-tidal CO2 was measured by transcranial Doppler ultrasound.

Soluble VCAM-1 concentration was 1094± 340 ng/mL in normotensives, 1195 ± 438ng/mL in controlled hypertensives and 1250 ± 445ng/mL in uncontrolled hypertensives (p=0.008). The mean resting BFV and cerebral vasomotor range (VMR) were respectively 41.0 ± 10.3 cm/s and 1.3 ± 0.4 cm/sec/mmHg. Elevated sVCAM -1 levels indicative of endothelial dysfunction were associated with reduced resting BFV (p=0.017) and cerebral VMR (p=0.0048). Elevated sVCAM-1 levels were associated with slower gait speed ((< 0.8 m/sec) OR = 3.01, 95 %CI (1.56 – 5.83), p=0.0011) and an increased odds of injurious falls (OR = 2.4, 95 %CI (1.4 – 4.2), p=0.0028).

Elevated sVCAM-1 levels may be a marker of cerebral blood flow dysregulation due to endothelial damage from hypertension. It may also signal the presence of cerebral microvascular disease and its clinical consequences, including slow gait speed and falls.

INTRODUCTION

Cerebrovascular diseases and mobility impairments are increasingly prevalent worldwide in older adults, conferring a major burden on health and health care costs.^{1, 2} Recent studies suggest that these conditions may be linked through the development of cerebral endothelial dysfunction and associated cerebral microvascular disease.^{3, 4} Cerebral endothelial dysfunction may be due to a number of conditions associated with aging, including hypertension and type-2 diabetes.^{5, 6} It may also impair cerebral blood flow regulation and cerebral vasoreactivity, and ultimately lead to mobility impairments, including slowing of gait and falls.^{7, 8}

Soluble Vascular Cell Adhesion Molecule-1 (sVCAM-1) is a well-known biomarker of endothelium dysfunction that is associated with hypertension and atherosclerosis.^{9, 10} It plays an important role in accelerating atherosclerosis by facilitating the attachment of inflammatory cells to the vascular endothelial wall and promoting their subsequent migration through the endothelium.^{11, 12} The resulting inflammatory response, injury, and stiffening of the vascular wall may result in impaired cerebral blood flow regulation, especially endothelium-dependent vasodilatation in response to changes in blood end-tidal pCO2, and ischemic damage to the cerebral microvasculature. This microvascular disease often manifests as white matter hyperintensities in brain imaging studies.

We hypothesized that elevated plasma concentrations of sVCAM-1 associated with hypertension may be a marker of cerebral microvascular disease, characterized by reduced middle cerebral artery blood flow, an impaired vascular responsiveness to CO₂, slow gait speed and increased risk of falls in elderly people. We therefore used transcranial Doppler measures of cerebral blood flow regulation, plasma biomarkers, and mobility data from the MOBILIZE Boston Study (Maintenance Of Balance, Independent Living, Intellect, and Zest in the Elderly) to explore the relationships between plasma levels of VCAM-1, cerebral blood flow dysregulation, and mobility impairment in a community-based population of older adults.

METHODS

Participants

The study sample consisted of 680 individuals from the MOBILIZE Boston Study (MBS), which is a prospective cohort study of a unique set of risk factors for falls in community-dwelling seniors living in the Boston area. The design and methodology for this study have been previously described in detail. ^{13, 14} In brief, 680 persons eligible for this study were enrolled using door-to-door population based recruitment. To be included, individuals had to be > 70 years (or age > 65 if living with a participant), able to understand and communicate in English, able to walk 20 feet without personal assistance (walking aids permitted), expected to live in the area for at least 2 years and provide written informed consent for blood withdrawal . Exclusion criteria included terminal disease, severe vision or hearing deficits, and Mini-Mental State Examination score <18.^{15, 16}

All subjects underwent a complete home and laboratory assessment of demographic characteristics, medical conditions, medications, functional status, gait speed, smoking status, alcohol use, blood pressure, and cerebral hemodynamics, then were followed prospectively for falls using a monthly postcard calendar (see below). Only a subset of 419 Subjects had an adequate temporal acoustic window to obtain reliable Doppler measures of cerebral blood flow velocity. Only those subjects with complete data for the variables of interest were included in the analyses.

Clinical Measures

Hypertension

Three blood pressure (BP) categories were created: Normotension, no history or current evidence of hypertension; Controlled hypertension, A history of hypertension and antihypertensive treatment with normal BP (Systolic BP (SBP) < 140 and Diastolic BP (DBP) < 90 during the baseline clinical assessment); and Uncontrolled hypertension, A history of hypertension with abnormal BP (SBP≥140 or DBP≥90). ^{17, 18}

Gait speed

The walking speed was assessed at the study center on a open part in the lab were a four meter oval course was measured out. Gait speed was based on the fastest time of 2 trials of a usual-paced 4-meter walk. The time to walk 4 meters is a component of the Short Physical Performance Battery (SPPB), which was used to assess gait speed. Gait speed was dichotomized into two groups, slow < 0.80 m/s and fast ≥0.80 m/s based on previously established threshold values for frailty, disability, and falls.^{19, 20}

Falls Detection

During a one-year follow-up period, a fall was defined as unintentionally coming to rest on the ground or other lower level not as a result of a major intrinsic event (e.g., myocardial infarction, stroke, or seizure) or an overwhelming external hazard (e.g., hit by a vehicle).^{13, 21} Participants were instructed to complete and return monthly falls calendar postcards designed to be posted on a refrigerator. On the postcards, participants were to record an “F” for each fall on the day it occurred and an “N” on days when no falls occurred. This approach has been well-validated for use in epidemiological cohort studies and described in previous studies.²² All subjects who reported falls were also called to determine the circumstances of the fall and clinical outcomes, including whether any injuries (e.g. fractures) and hospital visits occurred.

Biomarkers measures

Concentrations of soluble intercellular adhesion molecule-1 (sICAM-1), sVCAM-1, and interleukin-6 (IL-6) were measured by ELISA assay (R&D Systems, Minneapolis, MN). For sICAM-1 this assay has a sensitivity of 0.35 ng/mL, and the day-to-day variability of the assay at concentrations of 64.2, 117, 290 and 453 ng/mL are 10.1, 7.4, 6.0 and 6.1%, respectively. For sVCAM-1 the assay has a sensitivity of 2.0 ng/mL, and the day-to-day variability of the assay at concentrations of 9.8, 24.9 and 49.6 ng/mL are 10.2, 8.5 and 8.9%, respectively. For IL-6, the assay has a sensitivity of 0.094 pg/mL, and the day-to-day variability of the assay at concentrations of 0.49, 2.78 and 5.65 pg/mL are 9.6, 7.2 and 6.5%, respectively. In addition, the concentration of high sensitivity C-reactive protein (hsCRP) was determined using an immunoturbidimetric assay on the Hitachi 917 analyzer (Roche Diagnostics - Indianapolis, IN), using reagents and calibrators from DiaSorin (Stillwater, MN). This high-sensitivity assay has a limit of detection of 0.03 mg/L. The day-to-day variability of the assay at concentrations of 0.91, 3.07 and 13.38 mg/L are 2.81, 1.61 and 1.1%, respectively. Non-fasting venous blood samples were collected from participants during the baseline clinical visit. Blood was collected in a 7.5 ml red-top tube, allowed to coagulate, and then centrifuged and decanted according to standard protocols. Serum samples were aliquoted into “Nunc” (Nalgene NUNC, Thermo Fisher Scientific Inc, Waltham MA USA 02451) cryotubes, labelled, and stored in −80°C freezers. All assays were performed by Dr. Nader Rifai’s group at Boston Children’s Hospital.

Transcranial Doppler Ultrasound Measures

Subjects reported to the Cerebrovascular Laboratory at the Hebrew SeniorLife Institute for Aging Research and were instrumented for heart rate (HR, ECG) and beat-to-beat arterial pressure monitoring (ABP, Finapres, Ohmeda Monitoring Systems, Englewood, CO).²³ End-tidal CO₂ was measured using a Vacumed CO₂ Analyzer (Ventura, CA) attached to a nasal cannula. TCD ultrasonography (MultiDop X4, DWL-Transcranial Doppler Systems Inc., Sterling, VA) was used to measure middle cerebral artery (MCA) mean blood flow velocity (BFV) at rest and in response to changes in end-tidal CO₂, as previously described.²⁴ The MCA signal was identified according to standard criteria²⁵ and recorded at a depth of 50 to 60 mm. The envelope of the velocity waveform, derived from a fast-Fourier analysis of the Doppler frequency signal, was digitized at 500 Hz, displayed simultaneously with the ABP, ECG, and end-tidal CO₂ signals, and stored for later off-line analysis. Previous studies, using a variety of techniques (133Xe, SPECT, MRI) and stimuli, have validated transcranial Doppler measures of relative changes in cerebral flow velocity as representative of changes in cerebral blood flow.^{26, 27}

CO₂ Vasoreactivity Protocol

BFV in the MCA was measured continuously while subjects inspired a gas mixture of 8% CO₂, 21% O₂, and balance nitrogen for 2 minutes and then mildly hyperventilated to an end-tidal CO₂ of approximately 25 mm Hg for 2 minutes. Percent change in MCA BFV from baseline was plotted against end tidal CO₂ while breathing room air, 8% CO₂, or hyperventilating. Cerebral vasoreactivity was measured as the slope of this relationship and expressed as percent change in BFV per mm Hg change in end- tidal CO₂. Cerebral vasomotor range (VMR) was calculated as the difference between the highest BFV during CO₂ inhalation (hypercapnia) and the lowest BFV during hyperventilation (hypocapnia) divided by the change in end-tidal CO₂ in during the 6 min of the vasoreactivity protocol.

Other Covariates

Covariates included sociodemographic characteristics, physiologic risk factors, health status, and amount of physical activity. Sociodemographic characteristics assessed in the home interview included age, sex, race (self-identified), and years of education. We used the validated Physical Activity Scale for the Elderly (PASE) to measure physical activity in the previous week.²⁸ Participants were asked about physician-diagnosed major medical conditions. Details of the study methods were published previously.^{13, 14} Orthostatic hypotension was defined as a drop in SBP of ≥20 mm Hg or a drop in DBP of ≥10 mm Hg within 3 minutes of standing.

Diabetes was defined using an algorithm based on self-reported diabetes, use of antidiabetic medications, and laboratory measures from the baseline clinic visit including random glucose (200 mg/dL) and hemoglobin A1c (7%). Body mass index (calculated as weight in kilograms divided by height in meters squared) was calculated from measured height and weight. Cognitive status was defined as MMSE <24 (Cognitive impairment) vs. MMSE ≥24 (No cognitive impairment). Comorbidity index was the number of comorbidities or medical conditions. Medication use (antihypertensive, antidepressant, and benzodiazepine) were also assessed.

Medication Use

Use of cardiovascular medications was assessed at baseline during an in-person interview by asking the participant to report all medications that were taken for at least a 2-week period. Dose information was recorded from medication bottles. Drug data were coded according to the Iowa Drug Information System (IDIS) and classified into the following 6 categories: alpha blockers, ACEIs, ARBs, beta blockers, diuretics, and calcium channel blockers. In addition, we grouped participants into those taking any cardiovascular medications and those on angiotensin system-blocking medications (ACE or ARBs). Psychotropic medication such as use antidepressant, and benzodiazepine were also assessed.

Ethics Statement

The MOBILIZE Boston Study was reviewed and approved by the Hebrew SeniorLife Institutional Review Board (IRB). Written informed consent was obtained from each participant. The study was conducted according to the principles of the Helsinki Declaration.

Data Analysis

Concentrations of sVCAM-1 and sICAM-1 were log-transformed to approximate a normal distribution prior to modeling as continuous variables. We also divided the distributions of sVCAM-1 into quintiles according to the distribution in the entire study population for categorical analyses. Participants were grouped into those with normotension, controlled hypertension and uncontrolled hypertension; slow (<0.8 m/sec) and fast (> 0.8 m/sec) gait speed; or fall categories (none vs. > 1 fall per year of followup; or no falls, non-injurious, and injurious falls). We compared baseline characteristics of different groups of study participants by using t tests, χ2 tests, or Wilcoxon rank-sum tests. We used multivariate linear regression to examine the cross-sectional relationships between log-transformed sVCAM-1 levels and continuous outcomes (e.g., gait speed and CVMR) and logistic regression to estimate the relative risk and 95% confidence intervals (CIs) for quintiles of sVCAM and binary outcomes (e.g., hypertensive, gait speed, or falls groups). In these analyses, we had at least 85% power to detect a significant linear trend (2-sided P<0.05) across quintiles in which the risk in the highest quintile relative to the lowest was at least 1.5.²⁹

Analyses were adjusted for the following groups of confounders: 1) other Biomarkers (ICAM-1, IL6, C-Reactive Protein); 2) socio-demographic conditions (Age, gender, White race, Education level, Body Mass Index, Current Smoker, Alcohol use); 3) Health conditions (Diabetes, Hypertension, Congestive Heart failure, Hyperlipidemia, Cognitive status, Depression, any cardiovascular medications, Coronary artery disease, Previous stroke) and 4) Physical activity level.

Subjects with missing data for the main outcomes, sVCAM-1, hypertension, gait, falls and BFV were excluded. All analyses were performed using SAS software, version 9.3 (SAS Institute Inc, Cary, North Carolina). A two-sided P value of less than 0.05 was considered indicative of statistical significance.

RESULTS

Participants

Table 1 summarizes the demographic and clinical characteristics of the MBS participants with sVCAM-1 and sICAM-1 measurements. The mean age was 78±5 and 62% were female. 361 (54%) participants had controlled hypertension and 162 (24%) had uncontrolled hypertension. TCD measures with cerebral CO2 vasomotor range data were available in 419 (62%) participants. Data for the 419 subjects seemed representative of the larger population and there was no significant difference in demographics or the number of falls between those with and without a TCD window. sVCAM-1 concentration increased with age (p<.0001) and was higher in men than women (p=0.0007). The geometric mean of IL6 and hs-CRP at baseline were respectively 2.58±0.08 and 1.62±0.07. sVCAM-1 was correlated with IL6, (r=0.21 p<0.0001) but not with hs-CRP (r=0.07 p=0.08).

Table 1.

Characteristics of MOBILIZE BOSTON Study Participants at baseline.

Baseline Characteristics	Total Sample*	TCD data†
Demographics
Age, mean (SD), y	78.1±5.4	77.9±5.3
Women	419 (62.4)	242 (57.8)
White Race	546 (81.3)	367 (87.6)
Educational level, median (IQR), y	15 (12–2)	16 (13–17)
Body mass index, mean (SD), kg/m2,
<25	212 (31.6)	146 (34.8)
25–29.9	287 (42.7)	179 (42.7)
≥30	173 (25.7)	106 (25.3)
Alcohol use (Endorsing ≥2 drinks per week)	174 (25.6)	117 (27.9)
Current smoker	391 (57.6)	236 (56.3)
Physical activity score
0– 66	211 (31.4)	127 (30.3)
66.01–124	226 (33.7)	139 (33.2)
124.01–559	235 (35.0)	156 (37.2)
Medical condition
Comorbidity index, mean (SD)	3.0±1.6	2.9±1.6
Controlled Hypertensive	361 (53.7)	225 (53.7)
Uncontrolled hypertensive	162 (24.1)	88 (21.0)
Systolic Orthostatic Hypotension	54 (8.0)	40 (9.5)
Hyperlipidemia	404 (60.1)	255 (60.9)
Diabetes	126 (18.8)	77 (18.4)
Previous Stroke	66 (9.8)	45 (10.7)
Congestive Heart failure	34 (5.1)	21 (5.0)
Coronary disease	104 (15.5)	63 (15.0)
Gait speed <0.8 m/sec	168 (25.0)	102 (24.3)
Previous falls	260 (38.7)	159 (38.0)
Cognitive impairment (MMSE < 24)	73 (10.9)	46 (11.0)
Depression symptoms (CESD-R Score ≥16)	175 (26.0)	110 (26.3)
Medication
Psychotropic medication	48 (7.1)	29 (6.9)
Cardiovascular medication	461 (68.5)	282 (67.3)
Angiotensin system–blocking medication	298 (44.3)	189 (45.1)
Angiotensin-receptor blocker system	94 (14.0)	62 (14.8)
Angiotensin-converting enzyme inhibitor	210 (31.3)	134 (31.2)
Alpha-blocker	46 (6.8)	30 (7.2)
Beta-blocker	359 (53.4)	230 (54.9)
Any diuretic	338 (50.3)	201 (48.0)
Statin	325 (48.4)	199 (47.5)
Neurophysiologic measures†
Cerebral blood Flow Velocity (CBFV), mean (SD), cm/s		41.0±10.3
CBFV % change over baseline CO2, mean (SD)		3.0± 0.7
CO2 reactivity Vasomotor range, mean (SD),
cm/sec/mmHg		1.3± 0.4
Biomarkers measures
C-Reactive Protein, Geometric mean (SEM)mg/L	1.6± 0.1	1.5±0.1
Interleukin-6, Geometric mean (SEM), pg/mL	2.6± 0.1	2.4±0.1
Soluble ICAM-1, Geometric mean (SEM), ng/mL	245± 3	247±3
Soluble VCAM-1, Geometric mean (SEM), ng/mL	1126± 2	1124±2

^*Total sample (n=672);

^†Transcranial Doppler Data (TCD) (n=419).

Soluble VCAM-1 and Hypertension

The mean concentration of sVCAM-1 was 1049.45 ± 26.03 (SEM) in normotensives, 1129.16 ± 20.21 in controlled hypertensives and 1184.00 ± 32.82 in uncontrolled hypertensives (p=0.008). The final multivariate logistic regression model showed that sVCAM -1 was cross-sectionally associated with hypertension (p=0.014). Compared to the lowest quintile, the fourth and fifth quintiles of sVCAM-1 were also significantly associated with any hypertension (respectively, OR= 2.05 (1.04, 4.05) p=0.03, and OR = 2.61, 95 %CI (1.23 – 5.54) p= 0.0085); they were most strongly associated with uncontrolled hypertension (respectively OR = 2.04 95 %CI (1.23 – 3.38, p= 0.006) and OR = 3.05 95 %CI (1.78 – 5.23), p< 0.0001).

Soluble VCAM-1 and Gait speed

In a cross-sectional linear regression analysis, VCAM-1 plasma concentration was negatively associated with gait speed (r=−0.21, p<0.0001). When gait speed was dichotomized into slow (<0.8 m/sec) and fast (>0.8 m/sec) groups, an adjusted logistic regression analysis also showed a significant relationship with sVCAM-1 concentration (OR 1.16 (1.10–1.21, P=0.0084) (Table 2). However, there was an interaction between sVCAM-1 and hypertension (P=0.0006) such that only those subjects with a history of hypertension (controlled or uncontrolled) and the highest quintile of sVCAM-1 had slow gait speeds (<0.8 m/sec) (OR = 3.01 (1.56–5.83), p=0.001)) (Tables S1 & S2).

Table 2.

Odds ratios (95% confidence intervals) for the cross-sectional association between quintiles of sVCAM-1* and slow gait speed (< 0.8m/sec) stratified by hypertension status, No. = 672^†.

Measure	No.	Odds Ratio	95% Confidence Intervals	P Value
Normotensive
Soluble VCAM-1, ng/mLx100 (Unadjusted analysis)	142	1.06	0.95 – 1.19	0.25
Soluble VCAM-1, ng/mLx100 (Adjusted analysis) ‡	142	0.96	0.83 – 1.12	0.68
Hypertensive (Controlled and Uncontrolled)
Soluble VCAM-1, ng/mLx100 (Unadjusted analysis)	521	1.10	1.05 – 1.14	0.0001
Soluble VCAM-1, ng/mLx100 (Adjusted analysis) ‡	521	1.16	1.10– 1.21	0.0084
Quintiles of sVCAM-1, Median [IQR], ng/mL (Adjusted analysis) ‡
Q1: 736 [660–810]	96	1.00	Reference	Reference
Q2: 937 [899–970]	103	1.06	0.53 –2.10	0.7572
Q3: 1115 [1071–1169]	96	1.17	0.59–2.36	0.4201
Q4: 1332 [1277–1389]	110	1.31	0.66–2.57	0.4686
Q5: 1711 [1567–2011]	116	3.01	1.56–5.83	0.0011
Interaction effect of sVCAM-1 and Hypertension on Gait Group				0.0006

Notes:

^*sVCAM-1: Soluble Vascular Cell Adhesion Molecule; IQR, interquartile range.

^†Missing values=8 from 680;

^‡Adjusted for ICAM-1, Age, gender, White race, Education level, Body Mass Index, Current Smoker, Alcohol use, Diabetes, Hypertension, Stroke, Congestive Heart failure, Hyperlipidemia, Coronary artery disease, Cognitive status, Depression, Cardiovascular medications , Psychotropic medication, Comorbidity index and Physical activities level.

Soluble VCAM-1 and Falls over 1-year

Over 1-year of follow-up, 258 subjects fell with an annual incidence of 38.4%. Subjects who developed falls had significantly elevated levels of sVCAM-1 (OR= 1.04 (1.02, 1.09) p=0.0182). Compared to the lowest quintile, the third, fourth and fifth quintiles of sVCAM-1 were also significantly associated with one or more falls over 1-year (Table 3). We also stratified the subjects into 3 groups of fallers and examined sVCAM-1 levels in each group. Soluble VCAM-1 concentration was 1156.9 ± 421.3 in Non-Fallers, 1215.4 ± 436.6 in Non-Injurious Fallers, and 1277.1 ± 452.9 in Injurious Fallers (p= 0.0019). The third, fourth and fifth quintiles of sVCAM-1 were also significantly associated with injurious falls over 1-year compared to the lowest quintile (Table S3). There was no interaction between sVCAM-1 and hypertension on falls status.

Table 3.

Odds ratios (95% confidence intervals) for the association between quintiles of sVCAM-1* and ≥1 fall in the subsequent year, No. = 672^†

Model	Quintiles of sVCAM-1, Median (IQR), ng/mL					P for Trend
	736 (660–810)	937 (899–970)	1115 (1071–1169)	1332 (1277–1389)	1711 (1567–2011)
No.	132	134	136	134	136
Model 1‡
Odds Ratio (95%CI)	1.00	1.35 (0.80–2.28)	2.02 (1.21–3.37)	2.14 (1.28–3.57)	2.01 (1.20–3.66)
P value	Reference	0.26	0.0071	0.0037	0.0076	0.0133
Model 2||
Odds Ratio (95%CI)	1.00	1.38 (0.81–2.34)	2.16 (1.28–3.62)	2.95 (1.36–3.87)	2.30 (1.37–3.95)
P value	Reference	0.23	0.0038	0.0018	0.0026	0.0054
Model 3¶
Odds Ratio (95%CI)	1.00	1.37 (0.79–2.37)	1.99 (1.15–3.44)	2.38 (1.38–4.10)	2.33 (1.32–4.11)
P value	Reference	0.25	0.0141	0.0017	0.0037	0.0096
Model 4#
Odds Ratio (95%CI)	1.00	1.63 (0.91–2.94)	1.88 (1.05–3.39)	2.61 (1.44–4.72)	2.73 (1.44–5.16)
P value	Reference	0.10	0.0351	0.0015	0.0020	0.0135

Abbreviations:

^*sVCAM-1: Soluble Vascular Cell Adhesion Molecule-1; CI, confidence interval; IQR, interquartile range.

^†Missing values=8 from 680;

^‡Model1 non adjusted;

^||Model2 adjusted for ICAM-1 , IL6, C-Reactive Protein;

^¶Model3 additionally adjusted for Age, gender, White race, Education level, Body Mass Index, Current Smoker, Alcohol use;

^#Model4 additionally adjusted for Diabetes, Hypertension, Orthostatic hypotension, Stroke, Congestive Heart failure, Coronary artery disease, Hyperlipidemia, Cognitive status, Depression, Cardiovascular medication , Psychotropic medication, Comorbidity index and Physical activities level.

Soluble VCAM-1 and Cerebral Blood Flow Regulation (n=419)

To determine whether cerebral endothelial dysfunction might explain the association between sVCAM-1, gait speed, and risk of falls, we examined the relationship between sVCAM-1, BFV, CO₂ cerebral vasomotor range.

Soluble VCAM -1 was negatively correlated with resting cerebral BFV (r=−0.12 p=0.017). Soluble VCAM -1 was also associated with cerebral vasomotor range (VMR) (p= 0.040) (Figure 1). After adjustment for relevant covariates, the multivariate linear regression analysis showed that a 10% increase in sVCAM -1 reduced VMR by 2.40% (p=0.0048) (Table S4).

Figure 1.

Soluble ICAM-1

There were no significant relationship between any of the study outcomes and circulating levels of sICAM-1.

DISCUSSION

Our study results show that elevated plasma levels of sVCAM-1 are associated with 1) slower gait speed in older adults with hypertension, 2) incident falls, and 3) injurious falls. The association of sVCAM-1 with cerebral blood flow dysregulation suggests that sVCAM-1 may be involved in the pathogenesis of cerebral microvascular disease and its clinical consequences, namely slowing of gait and falls, in older people.

The cell adhesion molecule-regulated process of leukocyte recruitment often results in endothelial cell dysfunction, which can be manifested as either impaired endothelium-dependent vasorelaxation in arterioles, excess fluid filtration in capillaries, and/or enhanced protein extravasation in venules. ¹¹ Consequently, sVCAM-1 has been implicated in a variety of vascular disorders (e.g., ischemia/reperfusion, atherosclerosis, allograft dysfunction, and vasculitis) and an enhanced expression of sVCAM-1 has been invoked to explain the exaggerated microvascular dysfunction associated with cardiovascular risk factors (hypertension, hypercholesterolemia, diabetes).^{11, 30}

Cerebral microvascular disease has been reported to be associated with slow gait, executive dysfunction, and depressive symptoms.⁸ This association may be due to ischemic injury to areas of the brain that control motor function and attentional processes. Since sVCAM-1 is known to be associated with ischemic injury and endothelial dysfunction in other areas of the circulation, its relationship to abnormal cerebral vasoreactivity in response to CO₂ is biologically plausible. Our observations are supported by those of Novak, et al., who found an association between sVCAM-1 and CO₂ vasoreactivity, as well as gray matter atrophy, in diabetic subjects and older controls.³¹ Small differences in cerebral vasoreactivity or cerebral blood flow are associated with rather large effects on white matter hyperintensities. ³² Therefore, sVCAM-1 may serve as a clinical biomarker of cerebrovascular disease and its clinical consequences. Furthermore, elevated levels of plasma sVCAM-1 (> 1200 ng/mL) may signal a significant risk of hypertension on the cerebral circulation and its potential adverse effects on mobility. There is no correlation with sICAM-1. This suggests that sVCAM-1 has the predominant effect on the cerebrovascular endothelium.

Our study had some limitations. First, the cross-sectional design precludes investigating the temporal relation between sVCAM-1 elevation and hypertension, cerebral blood flow dysregulation, and slowing of gait speed. However, we were able to use longitudinal falls follow-up data to demonstrate a relationship between sVCAM-1 and falls incidence over 1-year. The second limitation is the smaller number of subjects with adequate TCD data because of an inadequate temporal bone window to obtain reliable Doppler measures of cerebral blood flow velocity. This is a common problem among elderly people, affecting approximately one-third of elderly subjects in previous studies.³³ Therefore, our findings may be not representative of all older adults. The third limitation is that we have examined only an endothelial dysfunction marker and sVCAM-1 may act through other pathways to impair mobility. Inflammatory markers were associated with poor physical performance and muscle strength in older persons. ³⁴ For example by inducing T-cell chemotaxis and inflammatory responses, synovial inflammation in hips or knees may affect gait.³⁵ We tried to address this by controlling for the inflammatory biomarkers CRP and IL6 in the multivariable analysis. Endothelial nitric oxide synthase (eNOS) regulates vascular tone through the production of NO. NO-dependent dilation is also impaired by oxidative stress that is elevated in hypertensive arteries. ³⁶ There may be a host of other factors that could be responsible for the change in VR and impaired gait speed. Finally, direct endothelial function tests may have been more appropriate to prove a connection between endothelial dysfunction, adhesion molecular biomarkers and hypertension or cerebrovascular risk but in this study we only measured serum soluble vascular adhesion molecules as putative biomarkers of endothelial damage and vasoreactivity is indeed a direct measure of endothelial function in the cerebral vessels. ³³

Our study has several strengths: First, we used both a cross-sectional and longitudinal study design in which we assessed incident cases of falls that occurred after the measurement of biomarkers. The adhesion molecule (VCAM and ICAM) concentrations were comparable to those in previous studies.³⁷ We included functional (TCD) measures of the cerebral circulation. We believe our work is novel in identifying a potential biomarker of falls risk, and suggesting a potential vascular mechanism of mobility impairment in older people, associated with ischemic damage to the microvasculature of the brain. Previous studies showed that persistent hypertension was associated with both a slower walking speed and greater decline in walking speed in the elderly. ³⁸ The prevalence of hypertension was nearly 80%, which is similar to that observed in other populations. ^{39, 40} Our study participants were derived from a random population of elderly people over age 70 from a defined geographic area of Massachusetts within a 5-mile radius of the Institute for Aging Research. We believe this is representative of the region.

Finally, our findings may have practical implications for clinicians who in the future may be able to use sVCAM-1 measurements to determine brain endothelial and parenchyma health without the use of expensive imaging studies, and intervene early to prevent the clinical consequences of cerebral microvascular disease In elderly people.

PERSPECTIVES

Elevated plasma levels of sVCAM-1 may be a marker of chronic cerebral blood flow dysregulation due to cerebral endothelial damage from hypertension. It may also signal the clinical consequences of cerebral microvascular disease, including slow gait speed and injurious falls among elderly people. Each of these finding suggests that sVCAM-1 may provide an attractive way to detect cerebral endothelium damage and guide future therapeutic interventions. Additional prospective studies are needed to confirm our findings and determine the sequence of events in the pathogenesis of cerebral microvascular disease and its clinical manifestations.

NOVELTY AND SIGNIFICANCE.

1 What is New?

• Elevated plasma concentrations of sVCAM-1 was associated middle cerebral artery blood flow dysregulation, and impairment of cerebrovascular responsiveness to CO₂,

• Elevated plasma concentrations of sVCAM-1 were associated slow gait speed and increased risk of falls in elderly people.

2) What Is Relevant?

• Endothelial dysfunction due to uncontrolled hypertension impaired cerebrovascular function and mobility in elderly.

3) Summary

Elevated plasma levels of sVCAM-1 may be a marker of chronic cerebral blood flow dysregulation due to cerebral endothelial damage from hypertension. It may also signal the clinical consequences of cerebral microvascular disease, including slow gait speed and injurious falls among elderly people.