Sex and Cardiovascular Function in Relation to Vascular Brain Injury in Patients with Cognitive Complaints

Abstract

Background:

Emerging evidence shows sex differences in manifestations of vascular brain injury in memory clinic patients. We hypothesize that this is explained by sex differences in cardiovascular function.

Objective:

To assess the relation between sex and manifestations of vascular brain injury in patients with cognitive complaints, in interaction with cardiovascular function.

Methods:

160 outpatient clinic patients (68.8±8.5 years, 38% female) with cognitive complaints and vascular brain injury from the Heart-Brain Connection study underwent a standardized work-up, including heart-brain MRI. We calculated sex differences in vascular brain injury (lacunar infarcts, non-lacunar infarcts, white matter hyperintensities [WMHs], and microbleeds) and cardiovascular function (arterial stiffness, cardiac index, left ventricular [LV] mass index, LV mass-to-volume ratio and cerebral blood flow). In separate regression models, we analyzed the interaction effect between sex and cardiovascular function markers on manifestations of vascular brain injury with interaction terms (sex*cardiovascular function marker).

Results:

Males had more infarcts, whereas females tended to have larger WMH-volumes. Males had higher LV mass indexes and LV mass-to-volume ratios and lower CBF values compared to females. Yet, we found no interaction effect between sex and individual cardiovascular function markers in relation to the different manifestations of vascular brain injury (p-values interaction terms > 0.05).

Conclusion:

Manifestations of vascular brain injury in patients with cognitive complaints differed by sex. There was no interaction between sex and cardiovascular function, warranting further studies to explain the observed sex differences in injury patterns.

Keywords

Cardiovascular function cerebrovascular disorders cognitive dysfunction female male sex vascular brain injury

INTRODUCTION

Sex differences in manifestations of cardiovascular disease are well established [1] and there is a growing body of literature that documents sex differences in the etiology and manifestations of dementia [2, 3]. A recent study found that male and female patients with vascular cognitive impairment (VCI) may have different manifestations of vascular brain injury: males had more lacunar and cortical infarcts, whereas females had larger white matter hyperintensity (WMH) volumes [4]. Sex differences in traditional vascular risk factors did not seem to explain these differences. These findings suggest that the underlying etiology of VCI might differ by sex. Currently, vascular brain injury is one of the few preventable factors of dementia [5]. Establishing sex differences in manifestations of vascular brain injury and exploring underlying mechanisms could ultimately direct treatment strategies.

Cardiovascular (dys)function, such as impaired cardiac function, markers of systemic atherosclerotic burden, and large artery stiffness, are associated with both vascular brain injury and cognitive impairment [6 –9]. Moreover, cardiovascular function is known to differ between males and females [10]: e.g., age-related arterial stiffness seems to progress at a faster rate in females than in males [11], and males are known to have lower cerebral blood flow (CBF) values than females at all ages [12, 13]. We hypothesize that sex differences in cardiovascular (dys)function explain the previously observed sex differences in manifestations of vascular brain injury in memory clinic patients.

This study aims to assess the relation between sex and manifestations of vascular brain injury in patients with cognitive complaints and evidence of vascular brain injury on magnetic resonance imaging (MRI), in interaction with cardiovascular function.

MATERIALS AND METHODS

Population

We included patients from the Heart-Brain Connection study (Heart-Brain Connection baseline data version 2, 1-1 2018), an observational multicenter cohort study that focusses on cardiovascular function, cerebral hemodynamics, and cognitive impairment. The rationale and methods of this study were described previously [14].

From September 2014 to September 2017, 160 patients with cognitive complaints and evidence of vascular brain injury on MRI were recruited from outpatient clinics, mostly from memory clinics, in four university medical centers in The Netherlands. Participants were independent in daily living,≥50 years old, and able to undergo cognitive testing. Inclusion criteria were a Clinical Dementia Rating (CDR) score≤1 and a Mini-Mental State Examination (MMSE)≥20 (i.e., we included patients with subjective cognitive decline, mild cognitive impairment, and mild dementia). Evidence of vascular brain injury on MRI had to be present as 1) moderate to severe WMH (Fazekas≥2) and/or (lacunar) infarct(s) or intracerebral (micro)hemorrhage(s), 2) mild WMH (Fazekas = 1) and an increased vascular risk, defined as the presence of at least two vascular risk factors (hypertension, hypercholesterolemia, diabetes mellitus, obesity, smoking. or clinically manifest vascular disease) [14].

The most important exclusion criteria were the inability to undergo an MRI, a life-threatening disease with life expectancy < 3 years (other than VCI, carotid occlusive disease, or heart failure), the presence of a psychiatric or neurologic disease (including neurodegenerative disease other than VCI or Alzheimer’s disease dementia) that affects cognitive functioning. Excluded for technical reasons were patients with atrial fibrillation at the moment of inclusion (of note, previous or paroxysmal atrial fibrillation was not an exclusion criterion) and premature ventricular contraction exceeding 10% of the total number of heartbeats, since these conditions preclude reliable performance of cardiac MRI [14].

All participants provided written informed consent prior to research-related procedures. The Medical Ethics Review Committee of the Leiden University Medical Center performed central approval. Local medical ethical committees of all sites approved the local performance of the study.

Sociodemographic factors, vascular risk factors, and cognitive functioning

Level of education was assessed according to the Verhage criteria, ranging from 1 to 7 (low to high education) [15]. The 15-item Geriatric Depression Scale (GDS-15) was used for screening of depressive symptoms [16].

The presence of vascular risk factors was derived from questionnaires and physical examination. A clinical diagnosis of hypertension was defined as presence in medical history, use of antihypertensive drugs or newly diagnosed hypertension defined as a systolic blood pressure≥140 mm Hg or a diastolic blood pressure≥90 mm Hg. Systolic and diastolic blood pressure were measured on the left and right arm with an automatic blood pressure monitor. The mean of these two readings was used for analyses. Hypercholesterolemia was defined as presence in medical history, use of lipid-lowering drugs or a low-density lipoprotein (LDL) level above 3.5 mmol/L. Diabetes was defined as presence in medical history with or without use of anti-diabetic drugs or a hemoglobin A1c (HbA1c) level above 53 mmol/L. Obesity was defined as a body mass index of≥30. Stroke was defined as presence of an ischemic or hemorrhagic stroke. Ischemic heart disease was defined as a history of myocardial infarction or coronary revascularization (percutaneous coronary intervention or coronary artery bypass graft).

The

MMSE was used as a cognitive screening test [17]. The severity of cognitive symptoms was assessed with the clinical dementia rating (CDR) global score [18]. Cognitive functioning was assessed with a comprehensive test battery (detailed methods are described previously [19] and Supplementary Table 1 shows an overview of the neuropsychological test protocol). All test scores were standardized into z-scores, using a reference group (n = 128, mean age: 65.6±7.4 years, 53% male). We considered a cognitive domain as impaired when the z-score was equal or below –1.5. We classified patients as having no objective cognitive impairment when no cognitive domains were impaired, minor cognitive impairment when one cognitive domain was impaired, and major cognitive impairment when two or more domains were impaired.

Heart-brain MRI protocol

Heart and brain MRIs were acquired on Philips Ingenia, Achieva, and Gemini 3T MRI scanners [14]. The cardiac MRI protocol included short axis multi-slice cine steady-state free precision (SSFP), with the following relevant parameters: resolution = 1.5×1.6×8.0 mm³; TR = 3.1 ms; TE = 1.55; flip angle = 45°; 40 heart phases; 67 phase percentage; breath-hold; number of slices dependent on size of the left ventricle. Additional aorta Q-flow images were performed with a resolution = 2.5×2.5×8 mm³; TR = 4.7 ms; TE = 2.8 ms; flip angle = 10°; velocity encoding = 150 cm/s; number of heart phases was dependent on the heart rate; temporal resolution = 5 ms [14].

For the brain MRI protocol T1-weighted (resolution = 1.0×1.0×1.0 mm³; 3D T1; TR = 8.2 ms; TE = 4.5 ms; shot interval 3.0 ms; flip angle 8°; inversion delay 990 ms), fluid-attenuated inversion recovery (FLAIR) images (resolution = 1.11×1.11×1.11 mm³; TR = 4.8 ms; TE = 313 ms; TI = 1.65 ms; TSE factor = 182) and susceptibility-weighted imaging (SWI) (resolution = 0.8×0.8×0.8 mm³; 3D gradient echo; TR = 45 ms; TE = 31 ms; flip angle 13°; EPI factor = 3) were performed. CBF measures were performed in the same scan session as the structural sequences by means of pseudo-continuous arterial spin labeling (pCASL) with multi-slice 2D echo planar imaging (EPI) readout and background suppression, resolution = 3×3×7 mm³, labeling duration = 1800 ms, post-labeling delay = 1800 ms, single shot EPI readout [20]. Phase contrast flow measures were obtained with a resolution = 1.17×1.17×5 mm³; TR = 12 ms; TE = 8.2 ms; flip angle = 10°; velocity encoding = 200 cm/s; untriggered; 10 averages [14].

Assessment of manifestations of vascular brain injury

Brain volumes, including total brain volume, total intracranial volume (ICV), and WMH-volume, were calculated with an automated pipeline (Quantib Brain, Rotterdam, The Netherlands) taking into account manual segmented infarcts and other pathologies. Results from accuracy and reproducibility tests of automatic brain tissue segmentation are published previously [21, 22]. Repeated infarct segmentations (intra-rater reliability) resulted in similar volumes (paired t-test, gray matter infarct volume: p = 0.92, white matter infarct volume: p = 0.70). Total brain volumes and WMH-volumes were expressed as percentage of the total ICV. Non-lacunar (sub)cortical infarcts, lacunar infarcts, and microbleeds (any, lobar, non-lobar and combination of lobar and non-lobar) were visually rated by an experienced neuroradiologist according to the STRIVE-criteria [23]. Non-lacunar (sub)cortical infarcts are referred to as non-lacunar infarcts.

Assessment of cardiovascular function markers

Both brachial pulse pressure and aortic pulse wave velocity (PWV) were used as markers for arterial stiffness. Aortic PWV was calculated as distance/transit time (in m/s), where distance was defined as length of the aortic arch between imaging planes through the ascending and descending aorta, and transit time was defined as the time between pulse waves at the ascending and descending aorta by half-max method [24].

For cardiac function, LV end-diastolic endocardial and epicardial contours and end-systolic endocardial contours were defined on short-axis cine images semi-automatically with manual correction to calculate left ventricular (LV) end-diastolic and end-systolic volumes, LV ejection fraction, and LV mass. LV stroke volume was calculated according to the formula: LV end-diastolic volume – LV end-systolic volume. LV stroke volume was multiplied by the heart rate to calculate LV cardiac output in L/min. To correct for the effect of body size, LV cardiac output was indexed by body surface area (DuBois & DuBois formula) to calculate cardiac index in L/min/m². LV mass was assessed assuming a density of 1.05 g/cm³ and indexed by body surface to calculate the hypertensive exposure marker LV mass index. The second hypertensive exposure marker, LV mass-to-volume ratio, was calculated by dividing LV mass by LV end-diastolic volume and expressed as percentage [6]. In general, a higher LV mass index or LV mass-to-volume ratio indicates worse cardiovascular health [6]. All analyses were performed using Mass software (version 2017-EXP, Leiden University Medical Center, the Netherlands). LV and PWV analyses were performed separately, both by an experienced reader who was blinded to all other patient data.

Regarding total CBF, pCASL images were processed using the automated Iris pipeline for CBF quantification [25]. Quantification of pCASL data into CBF maps was based on a single-compartment model after the subtraction of labeled images from control images [20]. To scale the signal intensities of the subtracted pCASL images to absolute CBF units, a separately acquired proton density weighted image was used. The quantification further included motion-correction of the raw pCASL data [26] and partial volume correction (PVC) [27]. CBF was quantified in normal-appearing gray matter only. To obtain the normal-appearing gray matter mask for each participant, first a binary gray matter segmentation was obtained. Subsequently, PVC-uncorrected pCASL images of all patients were visually inspected [20]. Images with suboptimal quality (i.e., motion artefacts, incomplete ASL-sequence, or labeling errors) and images with dominant vascular artefacts and little tissue perfusion signal were not used for further analyses [28]. As a sensitivity check of the pCASL data we additionally used phase contrast flow measures divided by total brain volume (in ml/100 g/min) for total CBF) (methods are described previously in more detail [29]).

Statistical analysis

Sociodemographic factors, vascular risk factors, manifestations of vascular brain injury and cardiovascular function markers were compared between female and male patients. We used general linear models to obtain age-adjusted female-to-male differences (F-MΔ) with corresponding 95% confidence intervals (CI) and p-values. In these linear regression analyses we used sociodemographic factors, vascular risk factors, manifestations of vascular brain and cardiovascular function markers as dependent variables, and sex and age as independent variables. In a separate model, we adjusted female-to-male differences in manifestations of vascular brain injury additionally for traditional vascular risk factors (diabetes, hypertension, hypercholesterolemia, current smoking, and obesity).

Sensitivity analyses were performed excluding patients without objective cognitive impairment, in order to study sex differences in manifestations of vascular brain injury in patients with minor and major cognitive impairment specifically.

Cardiovascular function markers were evaluated in relation to manifestations of vascular brain injury in males and females separately with linear or logistic regression analyses, depending on the type of the dependent variable. The possible interaction effect between sex and individual cardiovascular function markers in relation to different manifestations of vascular brain jury was investigated in separate linear or logistic regression models for the whole sample with interaction terms (sex*cardiovascular function marker), adjusted for age. Additionally, we applied a Bonferroni adjustment to correct for multiple testing [30].

Because of a non-normal distribution of WMH-volume, LV mass index, and aortic pulse wave velocity, we used the natural logarithm of WMH-volume and LV mass index, and the square root transformation of aortic pulse wave velocity in all linear regression analyses with WMH-volume, LV mass index or aortic pulse wave velocity as dependent variable. Subsequently, to calculate female-to-male differences the output was backtransformed. Associations with a p-value < 0.05 were considered significant. All statistical analyses were performed with R (version 1.3.1093).

RESULTS

Sociodemographic characteristics, vascular risk factors, and cognitive performance Sociodemographic characteristics, vascular risk factors, and cognitive performance of females and males are shown in Table 1. Sixty-one (38%) females and ninety-nine (62%) males with cognitive complaints and evidence of vascular brain injury were included. Mean age was 68.8±8.5 years and was comparable between sexes. Males more often had a clinical diagnosis of hypertension (F-MΔ –13%, 95% CI –26; –1) and tended to have more often a history of stroke or TIA (F-MΔ -14%, 95% CI –30; 2) compared to females. Sociodemographic characteristics, cognitive performance and other vascular risk factors were comparable between sexes.

Table 1

Sex differences in sociodemographic characteristics, vascular risk factors and cognitive performance

	Female (n = 61)	Male (n = 99)	F-M Δ (95% CI)^a	p
Sociodemographics
Age (y), mean (sd)	67.9 (8.9)	69.3 (8.2)	–1.4 (–4.1; 1.3)	0.32
Educational level, mean (sd)^b	5.2 (1.3)	5.3 (1.2)	–0.1 (–0.5; 0.3)	0.67
GDS≥6, n (%)^c	8 (14%)	14 (15%)	–2% (–14; 10)	0.77
Vascular risk factors
Clinical diagnosis of hypertension, n (%)	45 (74%)	86 (87%)	–13% (–26; –1)	0.04
Hypercholesterolemia, n (%)	50 (82%)	82 (83%)	–1% (–13; 11)	0.88
Diabetes mellitus, n (%)	7 (12%)	13 (13%)	–2% (–13; 9)	0.75
Current smoker, n (%)	10 (16%)	19 (19%)	–5% (–17; 8)	0.46
Obesity, n (%)	12 (20%)	16 (16%)	3% (–10; 15)	0.66
History of stroke or TIA, n (%)	28 (46%)	58 (59%)	–14% (–30; 2)	0.09
Ischemic heart disease, n (%)	7 (11%)	20 (20%)	–9% (–21; 3)	0.15
Cognitive performance
MMSE < 24, n (%)	8 (13%)	8 (8%)	6% (–4; 16)	0.22
CDR≥0.5, n (%)	41 (67%)	72 (73%)	–4% (–18; 11)	0.62
Cognitive impairment classification^d
No objective cognitive impairment, n (%)	33 (54%)	54 (55%)	–3% (–18; 12)	0.69
Minor cognitive impairment, n (%)	13 (21%)	23 (23%)	–2% (–15; 12)	0.81
Major cognitive impairment, n (%)	15 (25%)	22 (22%)	5% (–8; 18)	0.47

CDR, clinical dementia rating score; CI, confidence interval; F-M Δ, female to male difference; GDS, geriatric depression scale; MMSE, mini mental state examination; sd, standard deviation. Data are expressed as mean (sd) or number (percentage). Significant (p < 0.05) F-M Δ are printed in bold. ^aSex difference calculated as female – male adjusted for age. ^bEducation was assessed with the system of Verhage, ranging from 1–7 (low to high education). ^c9 missings (females n = 2, males n = 7). ^dPatients were classified as cognitively normal (no impairment), minor cognitive impairment (1 cognitive domain impaired) of major cognitive impairment (>1 domain impaired).

Manifestations of vascular brain injury

Age-adjusted female-to-male differences in manifestations of vascular brain injury are shown in Table 2. Total brain volumes (as percentage of ICV) did not differ between sexes. Males more often had non-lacunar infarcts (F-MΔ –19%, 95% CI –33; –6) and lacunar infarcts (F-MΔ –19%, 95% CI –35; –2) than females. In contrast, females tended to have larger WMH-volumes (as percentage of ICV) compared to males (median F-MΔ 0.19%, 95% CI –0.01; 0.69). Presence of microbleeds (lobar or non-lobar) did not differ significantly between sexes. When analyzed according to their location, a higher occurrence of strictly non-lobar microbleeds was found in females (F-MΔ 6%, 95% CI 0.3; 12) and a higher occurrence of mixed microbleeds (both lobar and non-lobar) in males (F-MΔ –13%, 95% CI –25; –2). Additional adjustment for vascular risk factors did not markedly change the effect sizes; the female-to-male difference in lacunar infarcts lost significance (F-MΔ –17%, 95% CI –34; 0.1), whereas the female-to-male difference in WMH-volume became significant (F-MΔ as % of ICV 0.29%, 95% CI 0.00; 1.69). Sensitivity analyses excluding patients without objective cognitive impairment showed essentially the same results as the main analyses (Supplementary Table 2).

Table 2

Sex differences in manifestations of vascular brain injury

	Female (n = 57)	Male (n = 91)	F-M Δ (95% CI)^a	p
Total ICV (ml), mean (sd)	1311 (106)	1454 (110)
Total brain volume (% of ICV), mean (sd)	78.2% (3.6)	78.0% (3.8)	0.2% (–1.3; 0.8)	0.70
WMH-volume (% of ICV), median (IQR)	0.58% (0.26;1.28)	0.49% (0.12;1.42)	0.19% (–0.01; 0.69)^b	0.08
Non-lacunar (sub)cortical infarcts, n (%)	6 (11%)	28 (31%)	–19% (–33; –6)	<0.01
Lacunar infarcts, n (%)	22 (39%)	52 (57%)	–19% (–35; –2)	0.03
Microbleeds (lobar or non-lobar), n (%)	16 (28%)	39 (43%)	–13% (–29; 3)	0.10

CI, confidence interval; F-M Δ, female-to-male difference; ICV, intracranial volume; IQR, interquartile range (25th to 75th quartile); sd, standard deviation; WMH, white matter hyperintensity. Data are expressed as mean (sd), median (IQR) or number (percentage). Significant (p < 0.05) F-M Δ are printed in bold. ^aSex difference calculated as female –male adjusted for age. ^bFor calculation of the female-male difference of WMH-volume we used the natural logarithm of WMH-volume (% of ICV), next the output was backtransformed.

Cardiovascular function markers

Age-adjusted female-to-male differences in cardiovascular function markers are presented in Table 3. Indicators of arterial stiffness (aortic PWV, but also brachial pulse pressure) and cardiac function (cardiac index) did not differ between female and male patients. Both LV mass indexes (median F-MΔ –12.8 g/m², 95% CI –14.7; –10.6) and LV mass-to-volume ratios (mean F-MΔ –4%, 95% CI –8; –1) were higher in males compared to females. Total CBF was lower in males than females both when assessed with pCASL (mean F-MΔ 4.4 ml/100 g/min, 95% CI 0.7; 8.2) and with phase contrast flow (mean F-MΔ 5.3 ml/100 g/min, 95% CI 0.2; 10.4).

Table 3

Sex differences in cardiovascular function markers

	Female	Male	F-M Δ (95% CI)^a	p
Arterial stiffness	N = 56	N = 83
Aortic pulse wave velocity (m/s), median (IQR)	9.2 (7.7;11.1)	9.0 (7.3;11.0)	0.5 (–0.5; 1.6)^b	0.37
Cardiac function	N = 55	N = 85
Cardiac index (L/min/m²), mean (sd)	2.8 (0.6)	2.9 (0.5)	–0.1 (–0.3; 0.05)	0.16
LV cardiac output (L/min), mean (sd)	4.8 (1.1)	5.8 (1.1)
Body surface area (m²), mean (sd)	1.7 (0.2)	2.0 (0.2)
Hypertensive exposure markers	N = 55	N = 85
LV mass index (g/m²), median (IQR)	41.4 (36.8;46.2)	51.4 (47.9;59.9)	–12.8 (–14.7; –10.6) ^c	<0.01
LV mass (g), median (IQR)	71.8 (62.9;78.0)	105.1 (94.0;121.9)
Body surface area (m²), mean (sd)	1.7 (0.2)	2.0 (0.2)
	N = 55	N = 86
LV mass-to-volume ratio (as %), mean (sd)	60 (10)	64 (10)	–4 (–8; –1)	0.02
LV mass (g), median (IQR)	71 (63;78)	105 (94;122)
LVEDV (ml), median (IQR)	119 (105;139)	163 (144;198)
Cerebral blood flow	N = 50	N = 77
CBF measured with ASL-MRI (ml/100 g/min), mean (sd)	54.1 (11.5)	49.4 (9.9)	4.4 (0.7; 8.2)	0.02

ASL-MRI, arterial spin labeling MRI; CBF, cerebral blood flow; CI, confidence interval; F-M Δ, female-to-male difference; IQR, interquartile range (25th to 75th quartile); LV, left ventricular; LVEDV, left ventricular end diastolic volume; N, number; sd, standard deviation. Significant (p < 0.05) F-M Δ are printed in bold. Data are expressed as mean (sd) or median (IQR). ^aSex difference calculated as female – male adjusted for age. ^bFor calculation of the female-male difference of aortic pulse wave velocity we used the square root of aortic pulse wave velocity, next the output was backtransformed. ^cFor calculation of the female-male difference of LV mass index we used the natural logarithm of LV mass index, next the output was backtransformed.

Sex differences in cardiovascular function markers in relation to manifestations of vascular brain injury

Figure 1 shows the relation between cardiovascular function markers and manifestations of vascular brain injury in females and males separately. Only for males we found a positive relation between both aortic PWV (OR 0.12, 95% CI 0.03–0.21) and LV mass index (OR 0.04, 95% CI 0.01–0.06) and WMH-volume (Fig. 1A), and a negative relation between cardiac index and presence of lacunar infarcts (OR 0.35, 95% CI 0.12–0.97) (Fig. 1 C). However, after correction for multiple testing these effects did not remain significant. We found no interaction effect between sex and individual cardiovascular function markers in relation to the different manifestations of vascular brain injury (p-value for interaction terms all > 0.05).

Fig. 1

The relation between cardiovascular function markers and manifestations of vascular brain injury in females and males. LV, left ventricular; PWV, pulse wave velocity. *After correction for multiple testing these relations lost significance Plots depict the age-adjusted relation between individual cardiovascular function markers and white matter hyperintensity volume (A), non-lacunar (sub)cortical infarcts (B), lacunar infarcts (C), and microbleeds (D) in females and males separately. Data is presented as beta coefficients with corresponding 95% CI (A) or odds ratios with corresponding 95% CI (B-D). P-interaction indicates the significance of the interaction effects between sex and individual cardiovascular function markers in relation to different manifestations of vascular brain injury in the whole sample.

DISCUSSION

This study shows sex differences in manifestations of vascular brain injury in patients with cognitive complaints and evidence of vascular brain injury: infarcts were more common in males, whereas WMH-volumes were larger in females. Also, cardiovascular function markers differed by sex: males had higher LV mass indexes and LV mass-to-volume ratios, reflecting hypertensive exposure, and lower CBF values compared to females. However, we found no interaction effect between sex and individual cardiovascular function markers in relation to the different manifestations of vascular brain injury.

Our findings on sex differences in manifestations of vascular brain injury are in line with previous findings in a memory clinic cohort [4] and large-scale population-based studies [31 –33]. Those studies found that elderly males more often have non-lacunar and lacunar infarcts while elderly females have more pronounced WMHs. Our study confirms previous findings that male and female patients with cognitive complaints and evidence of vascular brain injury at a memory clinic have different manifestations of vascular brain injury [4]. This implies that sex differences likely affect the underlying etiologies of vascular brain injury in patients with cognitive complaints.

Markers for cardiovascular (dys)function are known to differ between males and females [10]. In line with population-based studies we found higher LV mass-to-volume ratios and LV mass indexes in males compared to females [34], and lower total CBF values in males than in females [35 –37]. Literature on markers for cardiac function and arterial stiffness shows that sex differences are age-dependent. Based on normative data from the Framingham Heart Study offspring cohort, the sex difference in cardiac index seems to become smaller in patients aged 65 years and older [34]. The patients in our study had a mean age of 68.8 years, which might explain why we did not find a significant sex difference in cardiac index. Regarding arterial stiffness, we found no sex differences in two complimentary arterial stiffness markers, i.e., brachial pulse pressure and aortic PWV. Longitudinal studies show that arterial stiffness in older females increases more steeply than in males [11], but cross-sectional studies report inconsistent findings [38]. We probably did not find sex differences in arterial stiffness markers because we could not assess changes of arterial stiffness over time.

While we found sex differences in both manifestations of vascular brain injury and cardiovascular function markers, we found no interaction effect between sex and individual cardiovascular function markers in relation to the different manifestations of vascular brain injury. Only in males we found a relation between a few cardiovascular function markers and vascular brain injury. Aortic pulse wave velocity and LV mass index, two hypertension exposure markers, were related to WMHs in males but not in females. Furthermore, lower cardiac index was related to the presence of lacunar infarcts in males but not in females. These findings became non-significant after correction for multiple testing. Overall, our results provide only little support for a differential relation between cardiovascular function markers and manifestations of vascular brain injury in females and males.

All in all, after careful evaluation of cardiovascular function markers, a major share of the observed sex differences in manifestations of vascular brain injury in patients with cognitive complaints remained unexplained. In addition, previous work [4] and our results, show that also traditional vascular risk factors do not explain the observed sex differences in vascular brain injury. Alternative explanations for the observed differences in manifestations between males and females may include biological aspects or disparities in health care provision. For example, a community-based post mortem study [39], showed that females had more severe arteriosclerosis than males, which might explain the larger burden of WMHs in females. Furthermore, estimated lifetime exposure to estrogen, both endogenous (e.g., years of ovulation) and exogenous (e.g., years and timing of menopausal hormone therapy), might have direct effects on the observed sex differences in manifestations of vascular brain injury. In addition, since sex hormone exposure can influence inflammation [40] and coagulation processes [41], sexual dimorphisms in these biological processes might contribute to sex differences in manifestations of vascular brain injury. Besides biological factors, sex differences in management of vascular risk factors could also partially explain the observed sex differences in manifestations of vascular brain injury. Regarding secondary prevention of coronary heart disease for example, risk factor management is known to be generally worse in females than in males [42].

A strength of this study is the clinical setting in which all patients underwent a heart and brain MRI, providing various markers of vascular brain injury and cardiovascular function. Additional strengths include the standardized and detailed recording of sociodemographic characteristics and vascular risk factors. Also, some limitations have to be taken into account. First, the cross-sectional design of this study precludes us to study temporal associations between cardiovascular dys(function) and vascular brain injury. Second, we did not define a minimal threshold for severity of cognitive dysfunction for inclusion in our cohort. By contrast, most diagnostic criteria for VCI state that this construct only applies to patients with minor cognitive impairment or dementia [43, 44]. The reason for our inclusion criterion is that some patients with cognitive decline as result of vascular brain injury may not present with cognitive deficits that are severe enough to be classified as minor cognitively impaired [45]. Importantly, we performed sensitivity analyses that showed that excluding patients without objective cognitive impairment did not essentially affect the female-to-male differences in manifestations of vascular brain injury. Third, because all patients were selected for presence of vascular brain injury, we were only able to assess sex differences in the nature and burden of vascular brain injury, rather than its presence. Finally, the exclusion of patients with permanent atrial fibrillation might have led to a form of selection bias: since ischemic stroke is known to be higher in females with atrial fibrillation compared to males [46].

In conclusion, we provide further evidence on sex differences in manifestations of vascular brain injury in patients with cognitive complaints. However, these differences were not associated with sex differences in cardiovascular function. Future research is needed to unravel causes of the observed sex differences in patterns of brain injury in patients with VCI. This could eventually lead to improved personalized care and sex-specific recommendations for preventing VCI.

Footnotes

ACKNOWLEDGMENTS

We gratefully acknowledge the contribution of researchers and participants of the HBC (Heart-Brain Connection) Consortium. The HBC (Heart-Brain Connection) Consortium is supported by the Netherlands CardioVascular Research Initiative: the Dutch Heart Foundation (CVON 2018-28 & 2012-06 Heart Brain Connection), Dutch Federation of University Medical Centers, the Netherlands Organization for Health Research and Development, and the Royal Netherlands Academy of Sciences.

L.G. Exalto is supported by Alzheimer Nederland WE.03-2019-15. J. de Bresser is supported by Alzheimer Nederland WE.03-2019-08. E.E. Bron is supported by Dutch Heart Foundation (PPP Allowance, 2018B011)

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Mosca

, Benjamin

, Berra

, Bezanson

, Dolor

, Lloyd-Jones

, Newby

, Piña

, Roger

, Shaw

, Zhao

, Beckie

, Bushnell

, D’Armiento

, Kris-Etherton

, Fang

, Ganiats

, Gomes

, Gracia

, Haan

, Jackson

, Judelson

, Kelepouris

, Lavie

, Moore

, Nussmeier

, Ofili

, Oparil

, Ouyang

, Pinn

, Sherif

, Smith

, Sopko

, Chandra-Strobos

, Urbina

, Vaccarino

, Wenger

(2011) Effectiveness-based guidelines for the prevention of cardiovascular disease in women–2011 update: A guideline from the american heart association, Circulation 123, 1243–1262.

Mazure

, Swendsen

(2016) Sex differences in Alzheimer’s disease and other dementias, Lancet Neurol 15, 451–452.

Nebel

, Aggarwal

, Barnes

, Gallagher

, Goldstein

, Kantarci

, Mallampalli

, Mormino

, Scott

, Yu

, Maki

, Mielke

(2018) Understanding the impact of sex and gender in Alzheimer’s disease: A call to action, Alzheimers Dement 14, 1171–1183.

Exalto

, Boomsma

JMF

, Babapour Mofrad

, Barkhof

, Groeneveld

, Heinen

, Kuijf

, Leeuwis

, Prins

, Biessels

, Vd Flier

, TRACE-VCI study group (2020) Sex differences in memory clinic patients with possible vascular cognitive impairment, Alzheimers Dement (Amst) 12, e12090.

Gorelick

, Counts

, Nyenhuis

(2016) Vascular cognitive impairment and dementia, Biochim Biophys Acta 1862, 860–868.

Amier

, Marcks

, Hooghiemstra

, Nijveldt

, van Buchem

, de Roos

, Biessels

, Kappelle

, van Oostenbrugge

, van der Geest

, Bots

, Greving

, Niessen

, van Osch

MJP

, de Bresser

, van de Ven

, van der Flier

, Brunner-La Rocca

H-P

, van Rossum

, Heart-Brain Connection Consortium (2021) Hypertensive exposure markers by MRI in relation to cerebral small vessel disease and cognitive impairment, JACC Cardiovasc Imaging 14, 176–185.

Frazier

, Seider

, Bettcher

, Mack

, Jastrzab

, Chao

, Weiner

, DeCarli

, Reed

, Mungas

, Chui

, Kramer

(2014) The role of carotid intima-media thickness in predicting longitudinal cognitive function in an older adult cohort, Cerebrovasc Dis 38, 441–447.

Jefferson

, Himali

, Beiser

, Au

, Massaro

, Seshadri

, Gona

, Salton

, DeCarli

, O’Donnell

, Benjamin

, Wolf

, Manning

(2010) Cardiac index is associated with brain aging: The Framingham Heart Study, Circulation 122, 690–697.

Mitchell

, van Buchem

, Sigurdsson

, Gotal

, Jonsdottir

, Kjartansson

, Garcia

, Aspelund

, Harris

, Gudnason

, Launer

(2011) Arterial stiffness, pressure and flow pulsatility and brain structure and function: The Age, Gene/Environment Susceptibility–Reykjavik study, Brain 134, 3398–3407.

10.

Huxley

(2007) Sex and the cardiovascular system: The intriguing tale of how women and men regulate cardiovascular function differently, Adv Physiol Educ 31, 17–22.

11.

DuPont

, Kenney

, Patel

, Jaffe

(2019) Sex differences in mechanisms of arterial stiffness, Br J Pharmacol 176, 4208–4225.

12.

Liu

, Zhu

, Feinberg

, Guenther

, Gregori

, Weiner

, Schuff

(2012) Arterial spin labeling MRI study of age and gender effects on brain perfusion hemodynamics, Magn Reson Med 68, 912–922.

13.

Robison

, Gannon

, Salinero

, Zuloaga

(2019) Contributions of sex to cerebrovascular function and pathology, Brain Res 1710, 43–60.

14.

Hooghiemstra

, Bertens

, Leeuwis

, Bron

, Bots

, Brunner-La Rocca

H-P

, de Craen

AJM

, van der Geest

, Greving

, Kappelle

, Niessen

, van Oostenbrugge

, van Osch

MJP

, de Roos

, van Rossum

, Biessels

, van Buchem

, Daemen

MJAP

, van der Flier

, Heart-Brain Connection Consortium (2017) The missing link in the pathophysiology of vascular cognitive impairment: Design of the Heart-Brain Study, Cerebrovasc Dis Extra 7, 140–152.

15.

Verhage

(1964) Intelligentie En Leeftijd: Onderzoek Bij Nederlanders Van 12-77 Jaar [In Dutch], Van Gorcum, Assen.

16.

Yesavage

, Brink

, Rose

, Lum

, Huang

, Adey

, Leirer

(1982) Development and validation of a geriatric depression screening scale: A preliminary report, J Psychiatr Res 17, 37–49.

17.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician, J Psychiatr Res 12, 189–198.

18.

Hughes

, Berg

, Danziger

, Coben

, Martin

(1982) A new clinical scale for the staging of dementia, Br J Psychiatry 140, 566–572.

19.

Hooghiemstra

, Leeuwis

, Bertens

, Biessels

, Bots

, Brunner-La Rocca

H-P

, Greving

, Kappelle

, van Oostenbrugge

, van Rossum

, van der Flier

(2019) Frequent cognitive impairment in patients with disorders along the heart-brain axis, Stroke 50, 3369–3375.

20.

Alsop

, Detre

, Golay

, Günther

, Hendrikse

, Hernandez-Garcia

, Lu

, MacIntosh

, Parkes

, Smits

, van Osch

MJP

, Wang

DJJ

, Wong

, Zaharchuk

(2015) Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: A consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia, Magn Reson Med 73, 102–116.

21.

de Boer

, Vrooman

, van der Lijn

, Vernooij

, Ikram

, van der Lugt

, Breteler

MMB

, Niessen

(2009) White matter lesion extension to automatic brain tissue segmentation on MRI, Neuroimage 45, 1151–1161.

22.

de Boer

, Vrooman

, Ikram

, Vernooij

, Breteler

MMB

, van der Lugt

, Niessen

(2010) Accuracy and reproducibility study of automatic MRI brain tissue segmentation methods, Neuroimage 51, 1047–1056.

23.

Wardlaw

, Smith

, Biessels

, Cordonnier

, Fazekas

, Frayne

, Lindley

, O’Brien

, Barkhof

, Benavente

, Black

, Brayne

, Breteler

, Chabriat

, Decarli

, de Leeuw

F-E

, Doubal

, Duering

, Fox

, Greenberg

, Hachinski

, Kilimann

, Mok

, Oostenbrugge

R van

, Pantoni

, Speck

, Stephan

BCM

, Teipel

, Viswanathan

, Werring

, Chen

, Smith

, van Buchem

, Norrving

, Gorelick

, Dichgans

, STandards for ReportIng Vascular changes on nEuroimaging (STRIVE v1) (2013) Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration, Lancet Neurol 12, 822–838.

24.

Shahzad

, Shankar

, Amier

, Nijveldt

, Westenberg

JJM

, de Roos

, Lelieveldt

BPF

, van der Geest

, Heart Brain Connection study group (2019) Quantification of aortic pulse wave velocity from a population based cohort: A fully automatic method, J Cardiovasc Magn Reson 21, 27.

25.

Bron

, Steketee

RME

, Houston

, Oliver

, Achterberg

, Loog

, van Swieten

, Hammers

, Niessen

, Smits

, Klein

, Alzheimer’s Disease Neuroimaging Initiative (2014) Diagnostic classification of arterial spin labeling and structural MRI in presenile early stage dementia, Hum Brain Mapp 35, 4916–4931.

26.

Huizinga

, Poot

DHJ

, Guyader

J-M

, Klaassen

, Coolen

, van Kranenburg

, van Geuns

RJM

, Uitterdijk

, Polfliet

, Vandemeulebroucke

, Leemans

, Niessen

, Klein

(2016) PCA-based groupwise image registration for quantitative MRI, Med Image Anal 29, 65–78.

27.

Asllani

, Borogovac

, Brown

(2008) Regression algorithm correcting for partial volume effects in arterial spin labeling MRI, Magn Reson Med 60, 1362–1371.

28.

Leeuwis

, Hooghiemstra

, Bron

, Kuipers

, Oudeman

, Kalay

, Brunner-La Rocca

H-P

, Kappelle

, van Oostenbrugge

, Greving

, Niessen

, van Buchem

, van Osch

MJP

, van Rossum

, Prins

, Biessels

G-J

, Barkhof

, Heart-Brain Connection consortium, van der Flier

(2020) Cerebral blood flow and cognitive functioning in patients with disorders along the heart-brain axis: Cerebral blood flow and the heart-brain axis, Alzheimers Dement (N Y) 6, e12034.

29.

Oudeman

, Bron

, Van den Berg-Vos

, Greving

, Biessels

, Klijn

CJM

, Kappelle

, on behalf of the Heart-Brain Connection Consortium (2019) Cerebral perfusion and the occurrence of nonfocal transient neurological attacks, Cerebrovasc Dis 47, 303–308.

30.

Bonferroni (1936) Teoria statistica delle classi e calcolo delle probabilita, Pubblicazioni del R Istituto Superiore di Scienze Economiche e Commericiali di Firenze 8, 3–62.

31.

Fatemi

, Kantarci

, Graff-Radford

, Preboske

, Weigand

, Przybelski

, Knopman

, Machulda

, Roberts

, Mielke

, Petersen

, Jack

, Vemuri

(2018) Sex differences in cerebrovascular pathologies on FLAIR in cognitively unimpaired elderly, Neurology 90, e466–e473.

32.

Sigurdsson

, Aspelund

, Forsberg

, Fredriksson

, Kjartansson

, Oskarsdottir

, Jonsson

, Eiriksdottir

, Harris

, Zijdenbos

, van Buchem

, Launer

, Gudnason

(2012) Brain tissue volumes in the general population of the elderly: The AGES-Reykjavik study, Neuroimage 59, 3862–3870.

33.

Vinke

, de Groot

, Venkatraghavan

, Klein

, Niessen

, Ikram

, Vernooij

(2018) Trajectories of imaging markers in brain aging: The Rotterdam Study, Neurobiol Aging 71, 32–40.

34.

Yeon

, Salton

, Gona

, Chuang

, Blease

, Han

, Tsao

, Danias

, Levy

, O’Donnell

, Manning

(2015) Impact of age, sex, and indexation method on MR left ventricular reference values in the Framingham Heart Study offspring cohort, J Magn Reson Imaging 41, 1038–1045.

35.

Amen

, Trujillo

, Keator

, Taylor

, Willeumier

, Meysami

, Raji

(2017) Gender-based cerebral perfusion differences in 46,034 functional neuroimaging scans, J Alzheimers Dis 60, 605–614.

36.

Makwana

, Tart-Zelvin

, Xu

, Gunstad

, Cote

, Poppas

, Cohen

, Sweet

(2020) Cerebrovascular perfusion among older adults with and without cardiovascular disease, J Neuroimaging 30, 851–856.

37.

Xing

C-Y

, Tarumi

, Liu

, Zhang

, Turner

, Riley

, Tinajero

, Yuan

L-J

, Zhang

(2017) Distribution of cardiac output to the brain across the adult lifespan, J Cereb Blood Flow Metab 37, 2848–2856.

38.

Avolio

, Kuznetsova

, Heyndrickx

, Kerkhof

PLM

, Li

JK-J

(2018) Arterial flow, pulse pressure and pulse wave velocity in men and women at various ages, Adv Exp Med Biol 1065, 153–168.

39.

Oveisgharan

, Arvanitakis

, Yu

, Farfel

, Schneider

, Bennett

(2018) Sex differences in Alzheimer’s disease and common neuropathologies of aging, Acta Neuropathol 136, 887–900.

40.

Klein

, Flanagan

(2016) Sex differences in immune responses, Nat Rev Immunol 16, 626–638.

41.

Roy-O’Reilly

, McCullough

(2014) Sex differences in stroke: The contribution of coagulation, Exp Neurol 259, 16–27.

42.

Zhao

, Vaartjes

, Graham

, Grobbee

, Spiering

, Klipstein-Grobusch

, Woodward

, Peters

(2017) Sex differences in risk factor management of coronary heart disease across three regions, Heart 103, 1587–1594.

43.

Gorelick

, Scuteri

, Black

, Decarli

, Greenberg

, Iadecola

, Launer

, Laurent

, Lopez

, Nyenhuis

, Petersen

, Schneider

, Tzourio

, Arnett

, Bennett

, Chui

, Higashida

, Lindquist

, Nilsson

, Roman

, Sellke

, Seshadri

, American Heart Association Stroke Council, Council on Epidemiology and Prevention, Council on Cardiovascular Nursing, Council on Cardiovascular Radiology and Intervention, and Council on Cardiovascular Surgery and Anesthesia (2011) Vascular contributions to cognitive impairment and dementia: A statement for healthcare professionals from the American Heart Association/American Stroke Association, Stroke 42, 2672–2713.

44.

Sachdev

, Kalaria

, O’Brien

, Skoog

, Alladi

, Black

, Blacker

, Blazer

, Chen

, Chui

, Ganguli

, Jellinger

, Jeste

, Pasquier

, Paulsen

, Prins

, Rockwood

, Roman

, Scheltens

, Internationlal Society for Vascular Behavioral and Cognitive Disorders (2014) Diagnostic criteria for vascular cognitive disorders: A VASCOG statement, Alzheimer Dis Assoc Disord 28, 206–218.

45.

van der Flier

, Skoog

, Schneider

, Pantoni

, Mok

, Chen

CLH

, Scheltens

(2018) Vascular cognitive impairment, Nat Rev Dis Primers 4, 18003.

46.

Magnussen

, Niiranen

, Ojeda

, Gianfagna

, Blankenberg

, Njølstad

, Vartiainen

, Sans

, Pasterkamp

, Hughes

, Costanzo

, Donati

, Jousilahti

, Linneberg

, Palosaari

, de Gaetano

, Bobak

, den Ruijter

, Mathiesen

, Jørgensen

, Söderberg

, Kuulasmaa

, Zeller

, Iacoviello

, Salomaa

, Schnabel

, BiomarCaRE Consortium (2017) Sex differences and similarities in atrial fibrillation epidemiology, risk factors, and mortality in community cohorts: Results from the BiomarCaRE Consortium (Biomarker for Cardiovascular Risk Assessment in Europe), Circulation 136, 1588–1597.