Risk Factors for and Clinical Relevance of Incident and Progression of Cerebral Small Vessel Disease Markers in an Asian Memory Clinic Population

Abstract

Background:

Cerebral small vessel disease (SVD) is one of the major contributors to cognitive impairment and dementia. However, data on the incidence and progression of SVD in an Asian population are lacking.

Objective:

The present study aims to investigate the incidence, progression, associated risk factors, and clinical relevance of SVD in a memory clinic setting.

Methods:

A prospective case-control study, where 346 patients underwent repeated brain MRI with a mean interval of 24.5 months, accessing white matter hyperintensities (WMH), lacunes and cerebral microbleeds (CMBs). Severity of cognitive impairment was assessed using Clinical Dementia Rating scale and change in clinical diagnosis. Data on demographics, vascular risk factors, and clinical history were collected at baseline.

Results:

The prevalence of significant WMH (Fazekas ≥2) was 56.6% at baseline which progressed to 59.0% at follow-up. Overall prevalence of CMBs increased from 42.2% to 47.4% (9% new cases) and lacunes increased from 31.8% to 33.2% (2.1% new cases). Hypertension was associated with WMH progression (OR: 1.78, 95% CI: 1.01, 2.99) and increasing age was associated with incident CMBs (OR: 1.04, 95% CI: 1.01, 1.08). Moreover, the use of lipid-lowering medications decreased the incidence of lacunes (OR: 0.15, 95% CI: 0.04, 0.61). The major risk factor for incident SVD was baseline SVD lesion load. WMH progression was associated with increased severity of cognitive impairment (OR: 1.95, 95% CI: 1.16, 3.23).

Conclusion:

Vascular risk factors and baseline severity of SVD lesion load were associated with progression of SVD. Furthermore, WMH progression was linked with increased severity of cognitive impairment. Future studies should be aimed to slow cognitive deterioration by preventing SVD related brain damage by targeting vascular risk factors.

Keywords

Cerebral small vessel disease clinical dementia rating cognitive impairment magnetic resonance imaging vascular risk factors

INTRODUCTION

Cerebral small vessel disease (SVD) represents a wide range of pathological processes affecting arteries, arterioles, venules and capillaries, resulting in ischemic, hemorrhagic, and inflammatory damage [1]. The magnetic resonance imaging (MRI) markers of SVD include white matter hyperintensities (WMH), lacunes, and cerebral microbleeds (CMBs). SVD is frequently observed in the general elderly population with a prevalence ranging from 50–98% for WMH, 8–28% for lacunes, and 5–23% for CMBs in persons >60 years [2]. Due to this high prevalence, SVD is reported to be one of the leading causes of cognitive impairment, dementia, and physical disability [1, 3].

Systemic vascular risk factors such as hypertension, hyperlipidemia, and diabetes are highly common among elderly persons and these risk factors have been associated with WMH, lacunes, and CMBs [4 –6]. Previous longitudinal studies have shown that the presence of systemic vascular risk factors and higher burden of SVD at baseline were associated with incident and progression of SVD lesions on follow-up scans [7 –9], with values ranging from 39–74% for WMH, 1.6–19% for lacunes, and 10–18% for CMBs [7 –12]. These differences have been attributed to differences in study population (hospital/memory clinic versus population-based), variations in risk factor profile, and different MRI modalities (1.5 Tesla versus 3 Tesla). Moreover, the prevalence of SVD has been suggested to differ among ethnicities due to differences in vascular risk factors, genetic and environmental susceptibility; for example, it is reported that the burden of SVD is higher in Asians compared to Caucasians due to higher prevalence of cardiovascular risk factors [2]. Hence, we speculate that Asians would have a higher incidence and progression of SVD compared to Caucasians.

Previous studies have shown that WMH, lacunes, and CMBs co-occur and share a common pathophysiological mechanism [2, 13]. However, these studies were mainly focused on WMH progression, with limited data on incident lacunes and CMBs [7 , 12]. It remains unexplored whether baseline presence of one SVD contributes to the incidence and progression of other SVD markers. Furthermore, previous studies have been restricted to Caucasian population with no studies on the progression of SVD in Asia.

In the present study, we report the incidence of SVD as well as the association of baseline vascular risk factors with incidence and progression of SVD in an elderly Asian memory clinic population. Moreover, we investigate the association between progression of SVD and clinical outcomes as assessed by clinical dementia rating (CDR) score and change in clinical diagnosis from baseline to follow-up visit.

MATERIALS AND METHODS

Study population

This study was conducted as part of an ongoing prospective memory clinic study which involves two study sites in Singapore (National University Hospital and Saint Luke’s Hospital). From August 2010 to October 2015, a total of 474 patients were recruited. Following a standard assessment protocol, all patients were classified into the following diagnostic categories: 1) “No cognitive impairment” (NCI): patients with no objective cognitive impairment on formal neuropsychological tests, or functional loss; 2) “Cognitive impairment no dementia” (CIND) was based on clinical judgment and was diagnosed in patients who were impaired in at least one cognitive domain on a formal neuropsychological test battery, but did not meet the Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition (DSM-IV) criteria for dementia; 3) Dementia was diagnosed according to DSM-IV criteria. All participants underwent comprehensive evaluation including physical, medical and neuropsychological assessments along with 3T MRI on the same day. These assessments were performed annually except for MRI which was offered every 2 years.

Among 474 patients, 12 did not perform baseline MRI due to claustrophobia and inability to follow instructions while 18 had incomplete or poor quality MRI scans. The remaining 444 patients with baseline MRI were scheduled for follow-up scans between July 2012 and August 2017, however 98 patients did not have follow-up MRI scans, leaving 346 patients (NCI = 86, CIND = 149, Dementia = 111) for final analysis. The reasons for not having follow-up MRI scans were: loss to follow-up (n = 50), death (n = 24), incomplete or poor quality scan (n = 9), refusal (n = 8), MRI contraindications (n = 5), and lack of cooperation (n = 2).

Ethical approval was obtained from the National Healthcare Group Domain-Specific Review Board. This study was conducted in accordance with the Declaration of Helsinki. A written informed consent was obtained from all the participants or their caregivers.

Demographics and vascular risk factors

All patients were administered a standardized demographic questionnaire during the baseline visit and data on age, gender, years of formal education and smoking history were recorded. A vascular risk profile was obtained from the clinical interview, physical examination and review of laboratory and medical records. Systolic and diastolic blood pressures were recorded as the average of 2 readings measured five minutes apart on right arm using a digital automated sphygmomanometer. Hypertension was defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg during the examination or a history of hypertension or the use of antihypertensive medications. Hyperlipidemia was defined as total cholesterol level ≥4.14 mmol/l during examination or a history of hyperlipidemia or the use of lipid-lowering medications. Diabetes mellitus was defined as glycated hemoglobin ≥6.5% during the examination or a history of diabetes mellitus or the use of antidiabetic medication. Heart disease was defined as a previous diagnosis of myocardial infarction, congestive heart failure, atrial fibrillation, or intervention procedures such as angioplasty or stenting. Use of any anti-platelets medicines such as aspirin, clopidogrel, cilostazol, and dipyridamole were categorized as ‘yes’ versus ‘no’. Smoking was categorized as ‘ever’ versus ‘never’.

Neuroimaging

All participants underwent MRI at the Clinical Imaging Research Center of the National University of Singapore, using a 3T Siemens Magnetom Trio Tim Scanner system, with a 32-channel head coil. The standardized neuroimaging protocol includes a three-dimensional T1-weighted, T2-weighted, fluid-attenuated inversion recovery (FLAIR) and susceptibility weighted image (SWI) sequence. Identical MRI protocols with the same 3T scanner were used for both baseline and follow-up scans.

Assessment of MRI markers of SVD

MRI markers were defined based on the Standards for Reporting Vascular Changes on Neuroimaging (STRIVE) criteria [14]. The following MRI markers of SVD were graded:

Lacunes were defined as round or ovoid lesions involving the subcortical regions, 3 to 15 mm in diameter, with a low signal on T1-weighted images and FLAIR, a high signal on T2-weighted images and a hyperintense rim with a center following the cerebrospinal fluid intensity [14].

CMBs were defined as focal, round hypointense lesions with blooming effect on SWI. CMBs were graded using the Microbleed Anatomical Rating Scale (MARS) [15].

WMH were defined as hyperintense on T2 and FLAIR sequences and hypointense on T1 weighted images. WMH were graded using the Modified Fazekas Scale (absent = 0, punctate WMH = 1, early confluent WMH = 2 and confluent WMH = 3) [16]. For further analysis, WMH were divided into two groups; no significant WMH (grade 0 and 1) and significant WMH (grade 2 and 3).

Progression and incidence of SVD

Incident SVD was graded using side by side comparison of baseline and follow-up scans (Fig. 1). WMH progression was assessed using the Modified Rotterdam Progression Scale (presence or absence of WMH progression in 9 brain regions) [9]. Incident CMBs was defined as the presence of a new CMBs on follow-up scan which was absent at baseline scan. Similarly, incident lacunes was defined as the presence of new lacunes on follow-up scan which was absent at baseline scan. For lacunes and CMBs, the final number and location of these lesions were recorded at baseline and follow-up scans. Patients with regression of SVD over time were assigned to no incidence group.

Fig.1

Examples of incident cerebral small vessel disease. Axial T2 FLAIR image at baseline (A) and at follow–up (B) shows white matter hyperintensities progression identified by white arrow. Axial T2 FLAIR image at baseline (C) and at follow–up (D) shows an incident lacune (white arrow) at centrum semiovale. SWI at baseline (E) and at follow–up (F) shows an incident CMB visible in right frontal lobe (white arrow).

All MRI scans (n = 346) were graded by one rater (B.G.) (with two years of experience) blinded to participant’s clinical characteristics. All identified SVD markers were discussed with an experienced rater (S.H.) (with six years of experience) for a final decision. In order to assess intra-rater reliability, a subset of 100 scans were randomly selected and graded. Intra-rater agreement was good to excellent: intra-class correlation coefficient of MRI markers were as follows: 1) baseline scans: WMH = 0.86, lacune = 0.77, CMBs = 0.88; 2) follow-up scans: WMH progression = 0.88, incident lacunes = 0.79, incident CMBs = 0.84.

Assessment of clinical outcomes

All participants underwent detailed clinical assessment including administration of the CDR scale and neuropsychological assessments annually. However, in this study only baseline and follow-up data at year 2 were analyzed. These assessments were discussed in weekly consensus meetings attended by neurologists, psychologists, and research staff. Change in CDR scores was defined as one unit increase in sum of boxes (SOB) from baseline to second follow-up. Similarly, change in cognitive status was defined as conversion from NCI to CIND and from CIND to dementia from baseline to second follow-up visits. In this study those patients who changed in clinical diagnosis from NCI to CIND or CIND to dementia were grouped together.

Statistical analysis

In order to examine differences between patients with no follow-up scans compared to those with follow-up scans, Mann-Whitney U Test was performed for skewedly distributed continuous variable (age) and chi-square test for dichotomous variables. Binary logistic regression models were constructed with odds ratios (OR) and 95% confidence intervals (CI) to determine the association between baseline risk factors with incident SVD. All models were first adjusted for age, gender, and follow-up time (model I) and subsequently for hypertension, diabetes mellitus, hyperlipidemia, smoking and, history of heart disease (model II). Additional adjustments were made for use of medications (antihypertensive, antidiabetic, lipid-lowering and anti-platelet medications) added separately into the model consisting of demographic factors, vascular risk factors and baseline MRI markers (presence of WMH, presence of lacunes, and presence of CMBs) (model III). Final models were additionally adjusted for presence of SVD markers (lacunes, CMBs, and WMH) at baseline (model IV). The same regression models were further used to determine the association of incident SVD with change in CDR-SOB scores which was initially adjusted for age, gender and follow-up time (model I) and subsequently adjusted for hypertension, diabetes mellitus, hyperlipidemia, smoking and history of heart disease (model II). Final models were additionally adjusted for history of stroke at first follow-up visit (model III). Furthermore, in order to determine the interaction between baseline vascular risk factors and baseline SVD on progression of SVD, interaction terms [baseline vascular risk factors (diabetes mellitus, hyperlipidemia, smoking and history of heart disease)×baseline SVD markers (WMH, lacunes, CMBs)] were included separately into regression model with additional adjustment for covariates listed above. p < 0.05 was considered statistically significant. All the data were analyzed using SPSS software package (version 25).

RESULTS

Table 1 shows the characteristics of the study population with and without follow-up scan. Patients who had both baseline and follow-up scans were younger compared with those without follow-up scans (p = 0.001). Furthermore, patients without follow-up scans had a higher burden of heart disease (p = 0.004) and WMH (p = 0.014). The mean interval between baseline and follow-up MRI scans was 24.5 months.

Table 1

Baseline characteristics of study population with and without follow-up scan

Baseline Variables	Without follow-up scan n = 98	With follow-up scan n = 346	p
Age, median (IQR)	75 (11)	73 (11)	0.001
Gender (Female), n (%)	49 (50)	193 (55.8)	0.310
Hypertension, n (%)	71 (72.4)	245 (70.8)	0.752
Diabetes mellitus, n (%)	34 (34.7)	122 (35.3)	0.917
Hyperlipidemia, n (%)	69 (70.4)	260 (75.1)	0.345
Smoking (ever), n (%)	32 (32.7)	83 (24.0)	0.084
History of heart disease, n (%)	25 (25.5)	46 (13.3)	0.004
Antihypertensive medications, n (%)	70 (71.4)	205 (59.2)	0.028
Antidiabetic medications, n (%)	30 (30.6)	102 (29.5)	0.829
Lipid-lowering medications, n (%)	63 (64.3)	202 (63.6)	0.898
Antiplatelet medications, n (%)	33 (33.7)	113 (32.7)	0.850
Presence of lacunes, n (%)	34 (34.7)	110 (31.8)	0.588
Presence of cerebral microbleeds, n (%)	47 (48.0)	146 (42.2)	0.310
Presence of WMH, Modified Fazekas≥2, n (%)	69 (70.4)	196 (56.6)	0.014

IQR, interquartile range; WMH, white matter hyperintensities. p-value < 0.05 was considered statistically significant, bold values represents statistically significant association.

At baseline, the prevalence of significant WMH (Modified Fazekas≥2) was 56.6% (n = 196) which increased to 59.0% at follow-up. Overall 59.0% (n = 204) had WMH progression in one or more regions (Modified Rotterdam progression score: 1 = 8.4%, 2 = 29.5%, 3 = 13.6%, 4 = 6.4%, and 5 = 1.2%). Moreover, the overall prevalence of CMBs increased from 42.2% (n = 146) at baseline to 47.4% (n = 164) at follow-up. 80 (23.1%) patients had incident CMBs, of which 47 (58.8%) were multiple (9.5% had 1 incident CMB, 2.9% with 2 incident CMBs, 3.8% with 3 incident CMBs, and 6.9 % with≥4 incident CMBs). Among 80 patients with incident CMBs, 38 (47.5%) had strictly lobar CMBs, 18 (22.5%) had strictly deep CMBs whereas 24 (30.0%) had mixed CMBs. Finally, the overall prevalence of lacunes increased from 31.8% (n = 110) at baseline to 33.2% (n = 115) at follow-up. 15 (4.3%) patients developed incident lacunes, of which 5 (33.3%) were multiple (2.9% had 1 incident lacune, 0.9% with 2 incident lacunes and 0.6% with 3 incident lacunes). Regression of MRI markers was observed in 13 patients where one patient had regression of WMH and 12 had regression of CMBs.

Table 2 shows the progression of SVD markers in 10 years age group among persons with presence and absence of SVD at baseline. 52 (34.7%) who had not significant WMH (Modified Fazekas < 2) at baseline progressed to significant WMH (Modified Fazekas≥2), whereas 236 (68.2%) patients had no lacunes at baseline among them 5 (2.1%) patients had incident lacunes at follow-up. Similarly, 200 (57.8%) had no CMBs at baseline among them and 18 (9.0%) had incident CMBs at follow-up. Patients with SVD markers at baseline had new lesions on follow-up compared to those with no SVD markers at baseline (overall progression of SVD markers: 74.2% versus 29%, p < 0.001; WMH progression: 77.6 % versus 34.7%, p < 0.001; incident lacunes: 9.1% versus 2.1%, p = 0.003; incident CMBs: 45.2% versus 9%, p < 0.001).

Table 2

Incidence of SVD in 10-year age group in strata of the presence of SVD on baseline MRI

Age range, years	WMH < 2 on baseline scan		WMH≥2 on baseline scan		No prevalent lacunes on baseline scan		Prevalent lacunes on baseline scan		No prevalent CMBs on baseline scan		Prevalent CMBs on baseline scan
	No.	Incident WMH n (%)	No.	Incident WMH n (%)	No.	Incident lacunes n (%)	No.	Incident lacunes n (%)	No.	Incident CMBs n (%)	No.	Incident CMBs n (%)
50–59	14	7 (50)	9	8 (88.9)	15	0 (0)	8	2 (25)	13	0 (0)	10	5 (50)
60–69	65	23 (35.4)	40	29 (72.5)	72	1 (1.4)	33	1 (3.0)	66	3 (4.5)	39	12 (30.8)
70–79	58	17 (29.3)	104	84 (80.8)	112	4 (3.6)	50	6 (12)	90	10 (11.1)	72	32 (44.4)
80–95	13	5 (38.8)	43	31 (72.1)	37	0 (0)	19	1 (5.3)	31	5 (16.1)	25	13 (52)
Total	150	52 (34.7)	196	152 (77.6)	236	5 (2.1)	110	10 (9.1)	200	18 (9.0)	146	62 (42.5)

No, number of patients; WMH, white matter hyperintensities; CMBs, cerebral microbleeds; SVD, cerebral small vessel disease.

Table 3 presents the association between baseline risk factors and incident SVD lesions. Among demographic and vascular risk factors, increasing age was associated with incident CMBs (OR = 1.04, 95% CI = 1.01–1.08, p = 0.012) and hypertension was associated with WMH progression (OR = 1.87, 95% CI = 1.12–3.13, p = 0.017). Use of lipid lowering agent was associated with lower risk of incident lacune (OR = 0.15, 95% CI = 0.04–0.61, p = 0.008). With regards to baseline MRI markers, significant WMH was found to be associated with WMH progression (OR = 7.68, 95% CI = 4.47–13.18, p < 0.001), incident lacunes (OR = 11.29, 95% CI = 1.36–93.65, p = 0.025) and incident CMBs (OR = 4.67, 95% CI = 2.40–9.09, p < 0.001). The presence of CMBs at baseline was associated with incident CMBs (OR = 8.55, 95% CI = 4.57–15.99, p < 0.001) and incident lacunes (OR = 3.52, 95% CI = 1.04–11.98, p = 0.044) on follow-up. Significantly more participants with lacunes at baseline developed incident lacunes (OR = 3.96, 95% CI = 1.10–14.15, p = 0.035) on follow-up compared to participants without lacunes.

Table 3

Association between baseline risk factors with incident and progression of SVD

Baseline Variables	WMH progression^* OR (95% CI)	Incident lacunes^* OR (95% CI)	Incident CMBs^* OR (95% CI)
Model I
Age	1.02 (1.00, 1.05)	0.98 (0.92, 1.05)	1.04 (1.01, 1.08)
Gender	1.22 (0.79, 1.188)	1.10 (0.39, 3.10)	1.19 (0.72, 1.98)
Model II
Hypertension	1.87 (1.12, 3.13)	1.78 (0.44, 7.19)	1.33 (0.71, 2.46)
Diabetes	0.87 (0.54, 1.41)	1.14 (0.37, 3.48)	0.82 (0.47, 1.44)
Hyperlipidemia	1.15 (0.67, 1.97)	0.77 (0.22, 2.75)	0.89 (0.47, 1.67)
Smoking	1.22 (0.67, 2.22)	1.83 (0.47, 7.05)	1.41 (0.72, 2.76)
History of heart disease	1.12 (0.57, 2.20)	0.82 (0.16, 4.06)	0.86 (0.40, 1.86)
Model III
Antihypertensive medications	1.63 (0.80, 3.33)	1.64 (0.33, 8.14)	1.20 (0.53, 2.73)
Antidiabetic medications	1.36 (0.47, 3.93)	0.77 (0.08, 7.83)	0.41 (0.11, 1.49)
Lipid lowering medications	1.09 (0.53, 2.28)	0.15 (0.04, 0.61)	1.81 (0.73, 4.48)
Antiplatelet medications	0.96 (0.54, 1.69)	2.63 (0.76, 9.06)	1.03 (0.53, 1.99)
Model IV
Presence of lacunes	1.14 (0.68, 1.91)	3.96 (1.10, 14.15)	1.62 (0.90, 2.91)
Presence of cerebral microbleeds	1.37 (0.87, 2.17)	3.52 (1.04, 11.98)	8.55 (4.57, 15.99)
Baseline WMH	7.68 (4.47, 13.18)	11.29 (1.36, 93.65)	4.67 (2.40, 9.09)

WMH, white matter hyperintensities; CMBs, cerebral microbleeds; SVD, cerebral small vessel disease. Model I: included age and gender and follow-up time. Model II: included age, gender, follow-up time, hypertension, hyperlipidemia, diabetes mellitus, smoking and history of heart disease. Model III: Model II + baseline MRI markers (presence of lacunes, presence of CMBs and presence of WMH) and antihypertensive medications, antidiabetic medications, lipid lowering medications and antiplatelet medications added separately. Model IV: Model II + antihypertensive medications, antidiabetic medications, lipid lowering medications, antiplatelet medications and each individual MRI markers added separately. Bold values represent statistically significant associations at p < 0.05.

There was no significant interaction between baseline vascular risk factors (hypertension, diabetes mellitus, hyperlipidemia, smoking, and history of heart disease) and baseline SVD markers (WMH, lacunes, CMBs) on progression of SVD (WMH progression, incident lacunes, incident CMBs); p value for interaction >0.05 (data not shown).

Table 4 shows the association between incident SVD and change in CDR scores. Patients with progression of SVD were twice more likely to have a change in CDR-SOB scores (OR = 1.95, 95% CI = 1.16–3.23, p = 0.011). On stratifying by individual SVD markers, only WMH progression was associated with change in CDR-SOB scores (OR = 2.02, 95% CI = 1.23–3.29, p = 0.005). This association was independent of age, gender, and vascular risk factors and history of stroke.

Table 4

Association between incident and progression of cerebral small vessel disease and clinical

Models	Change in CDR-SOB score OR (95% CI)	p
Model I
Progression of SVD	1.95 (1.19, 3.21)	0.008
WMH progression	1.96 (1.22, 3.12)	0.005
Incident lacunes	1.57 (0.54, 4.57)	0.405
Incident CMBs	1.09 (0.64, 1.84)	0.762
Model II
Progression of SVD	1.89 (1.13, 3.14)	0.014
WMH progression	1.93 (1.19, 3.13)	0.008
Incident lacunes	1.34 (0.46, 4.21)	0.553
Incident CMBs	1.07 (0.62, 1.84)	0.809
Model III
Progression of SVD	1.95 (1.16, 3.23)	0.011
WMH progression	2.02 (1.23, 3.29)	0.005
Incident lacunes	1.44 (0.48, 4.37)	0.551
Incident CMBs	1.07 (0.62, 1.86)	0.808

SVD, Cerebral small vessel disease; CDR-SOB, clinical dementia rating -sum-of-boxes, WMH, white matter hyperintensities; CMBs, cerebral microbleeds. Model I: included age, gender and follow-up time. Model II: included age, gender, follow-up time, hypertension, hyperlipidemia, diabetes mellitus, smoking and history of heart disease. Model III: Model II + history of stroke at first follow-up visit. p-value < 0.05 was considered statistically significant, bold values represents statistically significant associations.

During the follow-up period of 24.5 months, 18 NCI (20.9%) patients converted to CIND and 20 CIND (13.4%) converted to dementia. Patients with a change in clinical diagnosis had a higher burden of SVD at baseline and incident SVD on follow-up. However, the change in clinical diagnosis was not statistically significant (p > 0.05).

DISCUSSION

In this study, the overall prevalence of significant WMH was 56.6%, CMBs 42.2%, and lacunes 31.8%. During follow-up, 59% of patients had WMH progression (34.7% of patients with no significant WMH at baseline progressed to significant WMH at follow-up), 23.1% incident CMBs (9% new cases), and 4.3% incident lacunes (2.1% new cases). The most important risk factors for incident SVD lesions were increasing age and hypertension. Patients with SVD markers at baseline were at increased risk of developing new SVD lesions. Furthermore, WMH progression was associated with change in CDR scores over time.

Previous studies have reported a wide range of prevalence of SVD depending upon the study population. Data from three Asian countries has shown an overall prevalence of WMH to be 36.6%, lacunes 24.6%, and CMBs 26.9% [2] whereas those from Caucasian populations have reported the prevalence of WMH of 50–98%, lacunes 8–28%, and CMBs 5–23% [6 , 17–19]. With respect to the incident SVD, prior data has been mainly restricted to Caucasians with the progression of WMH reported from 39–74%, incident lacunes 12–19%, and incident CMBs 10–18% [7–9 , 20]. To date, this is the first study to examine the progression of SVD lesions in Asia. Our current finding suggests that the WMH progression is within the same range of those reported previously in Caucasian whereas the incidence of lacunes was lower and CMBs higher in our Asian memory clinic population. This dissimilarity may partly be due to differences in MRI modalities, i.e., field strength (1.5T versus 3T) and sequences used as well as the inclusion criteria of the study population.

In the current study, we found that baseline presence of WMH predicted the progression of all three SVD markers (WMH progression, incident lacunes, incident CMBs). Furthermore the presence of lacunes, CMBs, and WMH at baseline was associated with incident lacunes. Previous studies have shown that the WMH, CMBs, and lacunes co-occur and progress together, suggesting a common pathophysiological mechanism underlying these lesions [13]. It is further reported that these SVD lesions develop as result of common vascular risk factors such as hypertension, diabetes mellitus, and hyperlipidemia which may lead to several pathological changes in the small penetrating blood vessels including atherosclerosis, lipohyalinosis, and disruption of the blood-brain barrier [6 , 22]. The brain parenchyma surrounding WMH as well as lacunes is vulnerable to further ischemic damage [23, 24], thus increasing the risk of incident SVD. Moreover, vascular amyloid not only leads to the destruction of vessel wall and development of CMBs, but may also cause occlusion leading to the occurrence of other SVD like lacunes and WMH [13]. Consistent with earlier studies, we found that increasing age, baseline presence of CMBs and WMH severity were associated with incident CMBs [8, 25]. It has been reported that cerebral amyloid angiopathy may cause hypoperfusion leading to WMH through several mechanisms such as vessel stenosis, necrosis of smooth muscles, loss of autoregulation, and vasoactive effect of amyloid-β leading to destruction of vessel wall and incident CMBs as well [8, 26].

In this study there was no significant interaction between baseline presence of SVD and vascular risk factors with incident SVD. Use of lipid-lowering medications was found to have protective effect on incident lacunes. It has been shown that the use of lipid-lowering agents such as statins and fibrates reduce the risk of incident stroke through anti-inflammatory, antioxidant, anti-thrombotic and neuroprotective effect [27, 28]. No effect was observed with other medications suggesting that, a mere longer duration of exposure from vascular risk factors may result in progression of SVD [7]. Second, during the different stages of disease other factors such as inflammation, amyloid, and tau may attenuate the effect those traditional cardiovascular risk factors [29, 30]. With regards to clinical outcome, we showed that the progression of SVD, more specifically WMH progression, was associated with an increase in CDR-SOB. Interestingly, patients with change in clinical diagnosis had incident SVD at follow-up. Hence incident SVD not only affects multiple cognitive domains but also affect function of daily activities. Several studies have previously shown the impact of incident SVD on increased severity of disease, gait disturbances, cognitive decline, and disability in activities of daily living [2 , 31–33]. Previous studies have shown gait disturbances, urinary incontinence, and decline in motor performance are common in elderly and these conditions are found to be associated with SVD [31]. SVD typically affects thalamo-cortical and corticospinal tract disrupting motor and sensory pathways which may leads to increased disability of daily living and functional decline [33].

There are some limitations of this study. First, we were unable to quantify the progression of WMH as we did not use volumetric method. However, it has been shown that the Rotterdam Progression Scale is more reliable and sensitive and correlates better with WMH volume changes [34]. Second, as this is a predominantly memory clinic based study with small sample size; it may limit the generalizability of our findings to the general population. Third, since the patients with follow-up scans were significantly younger compared to those without follow-up scans, this may result in selection bias. Fourth, the effect estimates of age with incident CMBs and hypertension with WMH progression is biased towards null (closer to 1). Hence it is difficult to interpret clinical relevance of these lesions and should be interpreted with caution. Fifth, as MRI was only offered every 2 years, we restricted our analysis for cognitive status and CDR changes to second follow-up which may not capture the cognitive deterioration observed during year 1. Finally, even after adjusting for several risk factors, we cannot exclude the possibility of residual confounding. The strengths of this study include: 1) the first study to be conducted in an Asian memory clinic population; 2) we studied the progression of three common SVD MRI markers; and 3) both baseline and follow-up MRI scans were performed at the same 3T scanner.

Conclusion

The present study showed that the prevalence of significant WMH (Fazekas≥2) was 56.6% at baseline which progressed increased to 59.0% at follow-up. Overall prevalence of CMBs increased from 42.2% to 47.4% (9% new cases) and lacunes increased from 31.8% to 33.2% (2.1% new cases). Baseline vascular risk factors and higher of SVD load was associated with progression of SVD and increased severity of cognitive impairment. Future studies should be aimed to slow cognitive deterioration by preventing SVD related brain damage by targeting vascular risk factors.

Footnotes

ACKNOWLEDGMENTS

We acknowledge all the Memory Aging & Cognition Centre, National University Hospital coordinators for their contribution to recruitment and data acquisition.

This study was supported by the National Medical Research Council grants; NMRC/CG/NUHS/2010 - R-184-005-184-511, NMRC/CG/013/2013, NMRC/CIRG/1446/2016.

Authors’ disclosures available online ().

References

Pantoni

(2010) Cerebral small vessel disease: From pathogenesis and clinical characteristics to therapeutic challenges. Lancet Neurol 9, 689–701.

Hilal

, Mok

, Youn

, Wong

, Ikram

, Chen

(2017) Prevalence, risk factors and consequences of cerebral small vessel diseases: Data from three Asian countries. J Neurol Neurosurg Psychiatry 88, 669–674.

de Laat

, Tuladhar

, van Norden

, Norris

, Zwiers

, de Leeuw

(2011) Loss of white matter integrity is associated with gait disorders in cerebral small vessel disease. Brain 134, 73–83.

Longstreth

Jr , Manolio

, Arnold

, Burke

, Bryan

, Jungreis

, Enright

, O’Leary

, Fried

(1996) Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people. The Cardiovascular Health Study. Stroke 27, 1274–1282.

Arauz

, Murillo

, Cantú

, Barinagarrementeria

, Higuera

(2003) Prospective study of single and multiple lacunar infarcts using magnetic resonance imaging: Risk factors, recurrence, and outcome in 175 consecutive cases. Stroke 34, 2453–2458.

Poels

, Vernooij

, Ikram

, Hofman

, Krestin

, van der Lugt

, Breteler

(2010) Prevalence and risk factors of cerebral microbleeds: An update of the Rotterdam scan study.S. Stroke 41, 103–106.

van Dijk

, Prins

, Vrooman

, Hofman

, Koudstaal

, Breteler

(2008) Progression of cerebral small vessel disease in relation to risk factors and cognitive consequences: Rotterdam Scan study. Stroke 39, 2712–2719.

Poels

, Ikram

, van der Lugt

, Hofman

, Krestin

, Breteler

, Vernooij

(2011) Incidence of cerebral microbleeds in the general population: The Rotterdam Scan Study. Stroke 42, 656–661.

Gouw

, van der Flier

, Fazekas

, van Straaten

, Pantoni

, Poggesi

, Inzitari

, Erkinjuntti

, Wahlund

, Waldemar

, Schmidt

, Scheltens

, Barkhof

(2008) Progression of white matter hyperintensities and incidence of new lacunes over a 3-year period: The Leukoaraiosis and Disability study. Stroke 39, 1414–1420.

10.

Daugherty

, Raz

(2017) Incident risk and progression of cerebral microbleeds in healthy adults: A multi-occasion longitudinal study. Neurobiol Aging 59, 22–29.

11.

Schmidt

, Seiler

, Loitfelder

(2016) Longitudinal change of small-vessel disease-related brain abnormalities. J Cereb Blood Flow Metab 36, 26–39.

12.

Enzinger

, Fazekas

, Ropele

, Schmidt

(2007) Progression of cerebral white matter lesions—clinical and radiological considerations. J Neurol Sci 257, 5–10.

13.

Akoudad

, Ikram

, Koudstaal

, Hofman

, Niessen

, Greenberg

, van der Lugt

, Vernooij

(2014) Cerebral microbleeds are associated with the progression of ischemic vascular lesions. Cerebrovasc Dis 37, 382–388.

14.

Wardlaw

, Smith

, Biessels

, Cordonnier

, Fazekas

, Frayne

, Lindley

, O’Brien

, Barkhof

, Benavente

, Black

, Brayne

, Breteler

, Chabriat

, Decarli

, de Leeuw

, Doubal

, Duering

, Fox

, Greenberg

, Hachinski

, Kilimann

, Mok

, Oostenbrugge

, Pantoni

, Speck

, Stephan

, Teipel

, Viswanathan

, Werring

, Chen

, Smith

, van Buchem

, Norrving

, Gorelick

, Dichgans

(2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 12, 822–838.

15.

Gregoire

, Chaudhary

, Brown

, Yousry

, Kallis

, Jager

, Werring

(2009) The Microbleed Anatomical Rating Scale (MARS): Reliability of a tool to map brain microbleeds. Neurology 73, 1759–1766.

16.

Poggesi

, Pracucci

, Chabriat

, Erkinjuntti

, Fazekas

, Verdelho

, Hennerici

, Langhorne

, O’Brien

, Scheltens

, Visser

, Crisby

, Waldemar

, Wallin

, Inzitari

, Pantoni

(2008) Urinary complaints in nondisabled elderly people with age-related white matter changes: The Leukoaraiosis And DISability (LADIS) Study. J Am Geriatr Soc 56, 1638–1643.

17.

van Es

, van der Grond

, de Craen

, Admiraal-Behloul

, Blauw

, van Buchem

(2008) Risk factors for cerebral microbleeds in the elderly. Cerebrovasc Dis 26, 397–403.

18.

de Leeuw

, de Groot

, Achten

, Oudkerk

, Ramos

, Heijboer

, Hofman

, Jolles

, van Gijn

, Breteler

(2001) Prevalence of cerebral white matter lesions in elderly people: A population based magnetic resonance imaging study. The Rotterdam Scan Study. J Neurol Neurosurg Psychiatry 70, 9–14.

19.

Vermeer

, Longstreth

Jr , Koudstaal

(2007) Silent brain infarcts: A systematic review. Lancet Neurol 6, 611–619.

20.

Ding

, Sigurdsson

, Garcia

, Phillips

, Eiriksdottir

, Gudnason

, van Buchem

, Launer

(2015) Risk factors associated with incident cerebral microbleeds according to location in older people: The Age, Gene/Environment Susceptibility (AGES)-Reykjavik Study. JAMA Neurol 72, 682–688.

21.

Wardlaw

, Sandercock

, Dennis

, Starr

(2003) Is breakdown of the blood-brain barrier responsible for lacunar stroke, leukoaraiosis, and dementia? Stroke 34, 806–812.

22.

Chui

(2007) Subcortical ischemic vascular dementia (SIVD). Neurol Clin 25, 717vi.

23.

Longstreth

Jr , Dulberg

, Manolio

, Lewis

, Beauchamp

Jr , O’Leary

, Carr

, Furberg

(2002) Incidence, manifestations, and predictors of brain infarcts defined by serial cranial magnetic resonance imaging in the elderly: The Cardiovascular Health Study. Stroke 33, 2376–2382.

24.

Duering

, Csanadi

, Gesierich

, Jouvent

, Hervé

, Seiler

, Belaroussi

, Ropele

, Schmidt

, Chabriat

, Dichgans

(2013) Incident lacunes preferentially localize to the edge of white matter hyperintensities: Insights into the pathophysiology of cerebral small vessel disease. Brain 136, 2717–2726.

25.

Goos

, Henneman

, Sluimer

, Vrenken

, Sluimer

, Barkhof

, Blankenstein

, Scheltens

, van der Flier

(2010) Incidence of cerebral microbleeds: A longitudinal study in a memory clinic population. Neurology 74, 1954–1960.

26.

Roher

, Kuo

Y-M

, Esh

, Knebel

, Weiss

, Kalback

, Luehrs

, Childress

, Beach

, Weller

, Kokjohn

(2003) Cortical and leptomeningeal cerebrovascular amyloid and white matter pathology in Alzheimer’s disease. Mol Med 9, 112–122.

27.

Alpérovitch

, Kurth

, Bertrand

, Ancelin

M-L

, Helmer

, Debette

, Tzourio

(2015) Primary prevention with lipid lowering drugs and long term risk of vascular events in older people: Population based cohort study. BMJ 350, h2335.

28.

Fuentes

, Martinez-Sanchez

, Diez-Tejedor

(2009) Lipid-lowering drugs in ischemic stroke prevention and their influence on acute stroke outcome. Cerebrovasc Dis 27(Suppl 1), 126–133.

29.

van Dijk

, Prins

, Vermeer

, Vrooman

, Hofman

, Koudstaal

, Breteler

(2005) C-reactive protein and cerebral small-vessel disease: The Rotterdam Scan Study. Circulation 112, 900–905.

30.

Shoamanesh

, Preis

, Beiser

, Vasan

, Benjamin

, Kase

, Wolf

, DeCarli

, Romero

, Seshadri

(2015) Inflammatory biomarkers, cerebral microbleeds, and small vessel disease: Framingham Heart Study. Neurology 84, 825–832.

31.

Inzitari

, Simoni

, Pracucci

, Poggesi

, Basile

, Chabriat

, Erkinjuntti

, Fazekas

, Ferro

, Hennerici

, Langhorne

, O’Brien

, Barkhof

, Visser

, Wahlund

, Waldemar

, Wallin

, Pantoni

(2007) Risk of rapid global functional decline in elderly patients with severe cerebral age-related white matter changes: The LADIS study. Arch Intern Med 167, 81–88.

32.

Pinter

, Ritchie

, Doubal

, Gattringer

, Morris

, Bastin

, Del C Valdés Hernández

, Royle

, Corley

, Maniega

(2017) Impact of small vessel disease in the brain on gait and balance. Sci Rep 7, 41637.

33.

Rosano

, Kuller

, Chung

, Arnold

, Longstreth

Jr , Newman

(2005) Subclinical brain magnetic resonance imaging abnormalities predict physical functional decline in high-functioning older adults. J Am Geriatr Soc 53, 649–654.

34.

Gouw

, van der Flier

, van Straaten

, Pantoni

, Bastos-Leite

, Inzitari

, Erkinjuntti

, Wahlund

, Ryberg

, Schmidt

, Fazekas

, Scheltens

, Barkhof

(2008) Reliability and sensitivity of visual scales versus volumetry for evaluating white matter hyperintensity progression. Cerebrovasc Dis 25, 247–253.