Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

Contents lists available at ScienceDirect

Cerebral Circulation - Cognition and Behavior

journal homepage: [Link]/journal/cerebral-circulation-cognition-and-behavior

Under pressure: A systematic review of the association between blood

pressure variability with depression and anxiety☆
Yuvthi Lutchman a, Rajiv Mahajan b, c, Suzanne M. Cosh a, Katie Harris d, Christophe Tzourio e,
Phillip J. Tully a, f, *
a
School of Psychology, The University of New England, Australia
b
Adelaide Medical School, The University of Adelaide, Australia
c
Department of Cardiology, Lyell McEwin Hospital, Adelaide, Australia
d
The George Institute for Global Health, University of New South Wales, Sydney, Australia
e
Univ. Bordeaux, Inserm, Bordeaux Population Health Research Center, U1219, CHU Bordeaux, F-33000 Bordeaux, France
f
School of Medicine, The University of Adelaide, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Blood pressure variability (BPV) impacts brain health by influencing brain structure and cerebrovascular pa­
Blood pressure variability thologies, though the mechanisms are poorly understood. Changes in the cerebrovasculature may lead to late-
Cognitive health onset depression, cognitive impairment, and dementia, however the relationship between BPV with depres­
Depression
sion and anxiety remains unclear, due to methodological differences and inconsistencies in past research. This
Anxiety
Cerebrovascular disease
review aims to clarify the association between BPV with depression and anxiety in adults to inform un­
Dementia derstandings of the mechanisms implicating BPV in cognitive health. A systematic search from inception through
Hypertension to January 2024 was performed on Embase, PubMed, PsycINFO, and Web of Science. Studies that assessed BPV
quantified by beat-to-beat, 24-hour, or visit-to-visit were eligible if the standardised assessment of depression
and/or anxiety were reported as a linear association, or mean differences across control and affect groups. A total
of 14 articles reporting on 13 samples and N = 5055 persons met the inclusion criteria (median female pro­
portion = 61 %, range 0 % - 76 %). A meta-analysis was not possible due to methodological heterogeneity in BPV
measurements and metrics across studies. Mixed results were observed across depression studies with in­
consistencies and variation in the direction, strength of association, and BPV metric. There was weak evidence
from only three studies to support a linear association between systolic coefficient of variation and anxiety.
Collectively, the findings contribute to understanding the association between BPV and brain health, suggesting
that any relationship between BPV and brain structures critical for cognitive function are independent of
depression and only modestly implicate anxiety.

1. Introduction function beyond average blood pressure (BP) measurements [8].

Emerging research indicates that BPV has broad implications for
The relationship between blood pressure variability (BPV) and neurological health beyond stroke risk, with BPV implicated in brain
adverse health outcomes is a burgeoning area of research. BPV refers to morphology [9,10]. Specifically, BPV, independent of average BP, is
the fluctuations in blood pressure readings quantified across beat-to- associated with arterial stiffness [11], vascular remodelling [12],
beat ([B2B] ultra-short term), ambulatory (short term) or visit-to-visit decreased cerebral perfusion [13,14] and cerebral small vessel disease
([V2V] long term) measurements [1]. BPV has been studied in the [15]. The association between BPV with brain atrophy [16] and the
context of cardiovascular diseases (CVDs) such as hypertension [2,3], progression of white matter lesions [17,18] and cerebral microbleeds
incident stroke [4,5], stroke recurrence [6] and coronary heart disease [18] suggests that BPV is a broad risk factor for cerebrovascular disease.
[7], providing deeper insight into cardiovascular and autonomic However, the association between BPV with dementia [19], Alzheimer’s

☆
PROSPERO: CRD42022312904
* Corresponding author at: School of Psychology, The University of New England, Elm Avenue, Armidale NSW 2351, Australia.
E-mail address: ptully2@[Link] (P.J. Tully).

[Link]
Received 6 March 2024; Received in revised form 5 May 2024; Accepted 31 May 2024
Available online 7 June 2024
2666-2450/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([Link]
nc-nd/4.0/).
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

disease progression [20], and tau accumulation [21] raises the possi­ January 2024, using the following databases from inception: Embase,
bility that BPV is a consequence of cerebrovascular diseases and neu­ PsycINFO, PubMed, and Web of Science. The search strategy comprised
rodegeneration. This highlights the importance of investigating other of a combination of keywords associated with BPV, anxiety and
potential markers of brain health for their potential association with depression (Supplementary Table S1).
BPV.
Among early markers for poorer brain health, late onset depression is 2.2. Inclusion criteria
frequently associated with cerebrovascular changes in white matter
tracts including white matter hyperintensities on magnetic resonance Population and exposure: Studies were required to have a study
imaging (MRI) and cognitive impairment, especially in executive func­ population of human participants aged 18 years or older with clinical
tion [22]. One possibility is that changes to the cerebrovasculature depression and/or anxiety using standardised clinical interviews or
contributes to the development of late onset depression and cognitive validated questionnaires and were derived from samples without sig­
decline. There is strong evidence that elevated BPV is linked to the nificant chronic disease other than cardiovascular disease or
severity and progression of cerebrovascular diseases via pathways hypertension.
including micro-vascular cerebral arterial wall damage and changes to Outcomes: BPV quantified using a standardised and valid device (e.g.
cerebral blood flow [22,23]. Such putative mechanisms may lead to OMRON M4 or Task Force Monitor) from systolic and/or diastolic BP
psychomotor slowing and executive dysfunction in both late onset readings, reported as a known BPV metric (either CoV, or SD, or average
depression and prodromal or early-stage dementia [24]. Additionally, real variability [ARV]) quantified over any time frame, characterised as
fluctuations in BPV might disrupt the central autonomic network, B2B, ambulatory blood pressure measurement (ABPM) over 24 h, or
impacting emotional responses such as anxiety [25] and depression V2V between assessments, including home BP monitoring. Studies were
[26]. required to provide effect sizes from the r or d family, including corre­
While elevated BPV has been implicated in structural and functional lations, standardised regression coefficients or differences in mean
cerebrovascular changes that could potentially influence mood disor­ values, to be included in the review. Effect sizes reported as odds ratios
ders, it remains unclear whether mood and BPV are directly related. or hazard ratios for risk of depression or anxiety attributable to BPV
Confounding variables that influence both mood and BPV include car­ were ineligible.
diovascular health and lifestyle risk factors [27]. There is also selection
biases related to the age groups and comorbidities within prior obser­ 2.3. Exclusion criteria
vational studies making it difficult to infer a direct causal link between
BPV and mood disorders. Current literature is limited by various con­ Ineligible studies were published in a language other than English,
straints, including methodological differences in BPV measurement type the case sample consisted of less than 10 persons, the sample comprised
such as beat-to-beat (B2B), ambulatory blood pressure monitoring of nonhuman sample population, the sampled included children or ad­
(ABPM) and V2V [28], as well as disparities in effect size measures (e.g. olescents <18 years old, experimental designs reporting on response to
correlation, regression and mean difference) and BPV metrics (e.g. co­ postural change or orthostatic challenge or an antihypertensive medi­
efficient of variation [CoV], standard deviation [SD]) [1]. Additionally, cation or other drug trial or CO2 reactivity or laboratory induced stress
B2B depression studies have primarily included younger participants or sadness, were focussed on serious mental illnesses such as bipolar
with lower cerebrovascular disease and dementia risk compared to older disorder or schizophrenia, or were derived solely from populations with
adults [29,30]. As research suggests that depression is linked to an dementia or other neurodegenerative disease, or were a review article,
increased risk of cerebrovascular disease [31] and dementia [24,32,33], letter or editorial or case study.
studies primarily focusing on children and young healthy populations
with low cerebrovascular disease and dementia risk may not fully cap­ 2.4. Study selection
ture the potential longer-term impacts of BPV on affective states. A 2020
systematic review explored the relationship between different affective Abstracts and titles were independently screened for eligibility by a
states with BPV [34]. A limitation of the previous review is that studies single reviewer (Y.L) in Covidence (Covidence, 2021). Following the
relating to nocturnal dipping patterns and orthostatic challenge were initial screening, full texts of relevant articles were assessed against the
included which is discrepant from modern understandings of what inclusion criteria by two reviewers (Y.L and P.J.T). Any discrepancies
constitutes BPV [35]. Additionally, the previous review [34] included were resolved through discussion and consultation with a third author
studies with small sample sizes, adolescents, and a broad inclusion cri­ (S.C). The reference lists of eligible articles were also searched and all
terion encompassing non-disorders states such as hostility as well as papers that met the criteria were within the systematic review. The
bi-polar disorder, resulting in heterogeneity across studies. The lack of study selection process is illustrated in the PRISMA flow diagram
data synthesis for BPV metrics and separate affective disorders further (Fig. 1).
magnified methodological heterogeneity, underscoring the need for
clarity on the putative association between BPV and affect. The present 2.5. Data extraction
systematic review extends beyond the previous review, by synthesizing
BPV studies pertaining to depression and anxiety. By reconciling past All data were extracted by one author (Y.L), which was verified and
limitations, this review aims to clarify whether BPV is associated with cross-checked by a second author (P.J.T). Discrepancies were discussed
depression and anxiety in adults, thereby informing whether BPV’s role until consensus was achieved. The data extracted from the included
in brain health encompasses the common affective states that are studies were study identification details (first author, publication year,
observed in the prodromal stages of dementia [32,36]. country of testing), study design, characteristics of the study sample
(derived from outpatients or community, number of participants, gender
2. Method distribution, mean age), relevant effect sizes, BPV metric (CoV, SD, or
average real variability), and type of depression or anxiety measure.
2.1. Search strategy Data extraction for studies reporting standard mean differences
encompassed both the experimental and comparator groups.
This systematic review adhered to the guidelines outlined in the
Preferred Reporting Items for Systematic reviews and Meta-analysis 2.6. Data synthesis
(PRISMA) [37] and was pre-registered on PROSPERO
(CRD42022312904). A literature search was performed on the 8th Due to heterogeneity in BPV measures (B2B, ABP, V2V), BPV metrics

2
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

Fig. 1. PRISMA flow diagram illustrating the article selection and screening process.

(SD, CoV, ARV), and effect sizes (correlation, regression and mean dif­ Japan, Malaysia, Poland, Taiwan and Turkey, while data from a French
ference), the included studies are synthesized qualitatively and a meta- cohort study was reported in two separate articles. Eleven studies were
analysis was not performed. Data on BPV outcomes were synthesised by cross-sectional in design and two were longitudinal studies. The cumu­
categorising studies by depression and anxiety separately, the reported lative sample was N = 5055 (median sample n = 135), and sample sizes
outcomes sub-divided as linear associations or mean differences, and ranged from 25 to 2297 participants. The proportion of female gender
stratified by B2B, ABPM, and V2V. ranged from 0 % to 76 %. Effect sizes were most commonly regres­
sion/correlation studies (6 studies), five compared groups with stan­
dardized mean differences, and three reported both mean differences
2.7. Quality assessment and correlations. Six studies utilised B2B, while five involved ABPM and
the other three utilized V2V data to quantify BPV. Regarding mental
In evaluating the quality of included studies, each study underwent health, 11 studies investigated depression and seven examined anxiety,
an independent appraisal by authors (Y.L and P.J.T) using the JBI with four studies addressing both depression and anxiety. Seven studies
Critical Appraisal Checklist for Analytical Cross-Sectional Studies. The examining depression reported regression/correlational statistics, while
checklist consists of eight items: identifying inclusion/exclusion criteria, six studies reported differences in mean BPV between a depressed and
specifying study sample, measuring exposure, measuring condition, non-depressed group. For anxiety, five studies reported correlational
identifying, and addressing confounding factors, measuring outcomes, values, while three studies reported mean differences between an anx­
and appropriate statistical analysis. Each criterion was rated using a iety and non-anxiety group (Table 1).
three-point scale (Yes/No/Unclear), which indicated whether the cri­
terion is sufficiently addressed in the study under assessment. All con­
flicts were resolved by discussion. 3.2. Study quality

Most criteria related to study eligibility in the JBI checklist were met
3. Results
by >70 % of the response (Table S4). All included studies used valid and
reliable measures for both exposure (JBI item 3) and outcome variables
3.1. Study selection and characteristics
(JBI item 7), as well as appropriate statistical analysis techniques (JBI
item 8). Some studies displayed moderate methodological quality
A total of 9986 articles were obtained through the database search,
overall, with issues observed in specifying inclusion/exclusion criteria
title and abstracts were screened for 5803 articles, resulting in 53 arti­
(JBI item 1) and addressing confounding variables (JBI Item 5).
cles selected for full-text review. Among these, 39 were excluded
(Table S2), leaving 14 articles included in the systematic review,
reporting on 13 unique study samples [38–51]. The included studies 3.3. Linear association between BPV and depression
were published between 2003 and 2022. Geographically, three studies
were conducted in the USA, two in Germany, one each in Finland, Italy, A total of seven studies reported a linear association between BPV

3
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

Table 1
Comparative overview of studies on BPV, depression and anxiety.
Mood

Author Country Study Design Group N (Women Mean age ± Effect BPV Depression Anxiety Mood
%) SD size Measure instrument

Davydov et al. (2007) USA Cross Clinical, 248 (61) 36.2 ± 10.9 SMD B2B Yes No HAMD
sectional Community
Imaizumi et al. (2016) JPN Cross Outpatient 85 (62) 79.2 ± 5.9 SMD, r ABP Yes No SRQD
sectional
Koklu et al. (2022) TUR Cross Outpatient 88 (43) 45.7 ± 15.1 SMD, r ABP No Yes BAI
sectional
Lin et al. (2020) TWN Longitudinal Outpatient 1112 (0) 32 β V2V Yes Yes BSRS
Sanchez-Gonzalez et al. USA Cross Community 43 (53) 26 ± 1.0 SMD B2B No Yes STAI
(2015) sectional
Schulz et al. (2010) GER Cross Clinical, 114 (68) 30 ± 9.0 SMD B2B Yes No HAMD, BDI
sectional Community
Schumann et al. (2017) GER Cross Inpatient 58 (72) 38.8 ± 12.2 SMD B2B Yes No HAMD, BDI
sectional
Scuteri et al. (2008) ITA Cross Inpatient 135 (76) 78 ± 6.0 SMD, β ABP Yes No GDS
sectional
Shahimi et al. (2022) MYS Cross Community 25 (68) 70.88 ± 7.2 r B2B Yes Yes DASS21
sectional
Sible et al. (2022) USA Cross Clinical 505 (40) 77.7 ± 6.5 β V2V Yes No GDS
sectional
Symonides et al. (2014) POL Cross Outpatient 195 (46) 45.4 ± 15.9 r ABP Yes Yes CECS
sectional
Three-City - (2017) FRA Longitudinal Community 1454 (59) 78.5 ± 3.8 β V2V Yes Yes MINI
Three-City - (2018) FRA Longitudinal Community 2297 (61) 72 SMD V2V Yes Yes MINI + CESD
Virtanen et al. (2003) FIN Cross Community 150 (53) NR β B2B No Yes BSI
sectional

ABP, ambulatory blood pressure; B2B, beat-to-beat; BAI, Beck Anxiety Inventory; BDI, Beck Depression Inventory; β, beta coefficient; BSI, Brief Symptom Inventory;
BSRS, Brief Symptom Rating Scale; CECS, Courtauld Emotional Control Scale; CESD, Centre for Epidemiological Studies – Depression; DASS21, Depression, Anxiety,
and Stress Scale; GDS, Geriatric Depression Scale; HAMD, Hamilton Rating Scale for Depression; MINI, Mini International Neuropsychiatric Interview; NR, not re­
ported; r, correlation coefficient; SMD, standardised mean difference; SRQD, Self-Rating Questionnaire for Depression; STAI, State-Trait Anxiety Inventory; V2V, visit-
to-visit;.

with depression symptom severity (Table 2). Of these, only one study significant associations across B2B [42], ABPM [38,44], and V2V [40].
used B2B measurements to evaluate BPV, while three used ABPM and Significant effect sizes in systolic BPV were found in only two studies,
three conducted V2V assessments. All reported systolic BPV, while five and these were each in different directions. Scuteri et al. [41] found a
also reported diastolic BPV. Three studies reported correlation co­ significant negative association between CoV of systolic BPV and
efficients, while others reported beta-coefficients. Among the seven depression scores (β = − 0.20), indicating higher systolic BPV was
studies, it was generally observed that systolic and diastolic BPV was associated with lower depression severity. In contrast, the Three-City
infrequently associated with depression, with four studies not reporting study [51] found a positive association for systolic BPV and

Table 2
Linear Associations between BPV, depression and anxiety.
Systolic Diastolic

Author (Year) N (Female%) BPV measure SD (CI) CoV (CI) ARV (CI) SD (CI) CoV (CI) ARV (CI)

Depression
Shahimi et al. (2022) 25 (68) B2B r = 0.02 r = 0.04 r = 0.01 r = 0.17 r = 0.15 r = 0.07
(0.01 – 0.03) (0.03 – 0.05) (0.00 – 0.02) (0.16 – 0.18) (0.14 – 0.16) (0.06 – 0.08)
Imaizumi et al. (2016) 85 (62) ABP r = 0.13 r = 0.18 – – – –
(0.12 – 0.14) (0.17 – 0.19)
Scuteri et al. (2008) 135 (64) ABP – β = − 0.20* – – – –
Symonides et al. (2014) 195 (46) ABP r = 0.15 – – r = 0.11 – –
(0.15 – 0.15) (0.11 – 0.11)
Lin et al. (2020) 1112 (0) V2V β = − 0.01 – β = 0.0 β =0.00 – β = 0.00
Sible et al. (2022) 505 (40) V2V – β = 0.11 – – β = 0.16* –
Three-City - (2017) 1454 (59) V2V β = 0.01 β = 0.11 – β = 0.01 β = − 0.01 –
Anxiety
Shahimi et al. (2022) 25 (68) B2B r = − 0.32 r = − 0.39* r = − 0.05 r = − 0.23 r = − 0.40* r = − 0.04
(− 0.33 – − 0.31) (− 0.40 - − 0.38) (0.04 – 0.06) (− 0.24 - − 0.22) (− 0.41 - − 0.39) (− 0.05 - − 0.03)
Virtanen et al. (2003) 150 (53) B2B – β = 0.25* – – – –
Koklu et al. (2022) 88 (43) ABP r = 0.32* r = 0.35* – – – –
(0.31 – 0.33) (0.34 – 0.36)
Symonides et al. (2014) 195 (46) ABP r = 0.15* – – r = 0.11 – –
(0.15 – 0.15) (0.11 – 0.11)
Lin et al. (2020) 1112 (0) V2V β = − 0.01 – β = 0.00 β = 0.00 – β = 0.00
Three-City - (2017) 1454 (59) V2V β = 0.10 β = 0.25* – β = 0.09 β = 0.04 –

ABP, ambulatory blood pressure; ARV, real average variability; β, beta coefficient; B2B, beat-to-beat; BPV, blood pressure variability; CI, 95 % Confidence Intervals;
CoV, coefficient of variation; N, sample size; r, Pearson’s correlation coefficient; SD, standard deviation; V2V, visit-to-visit.
*
Statistical significance <0.05.

4
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

depression scores (β = 0.11). Significant effect sizes in diastolic BPV 3.6. Between anxiety group differences for BPV
were reported in one study (β = 0.16) [43], and indicated that increased
diastolic BPV was associated with higher depression scores. Among the three studies that explored the mean difference of BPV
between groups with high anxiety (experimental group) versus those
3.4. Linear association between BPV and anxiety with low or no anxiety (control group), only one study reported signif­
icant findings, while the other two did not. Sanchez-Gonzalez et al. [47]
Among the six included studies that reported a linear association found a significant difference in systolic BPV with higher scores among
between BPV and anxiety: two used B2B, two reported ABPM and two individuals with anxiety than the control group. In contrast, the
employed V2V assessments (Table 2). All studies examined the rela­ Three-City study [51] and Koklu et al. [39] did not find significant
tionship between systolic BPV and anxiety, while four studies also re­ differences in either systolic or diastolic BPV between the anxious and
ported diastolic BPV. Significant findings in systolic BPV were reported non-anxious groups (Table 3).
in five studies, three reported correlation coefficients ranging from a
negative association (r = − 0.039) to positive association (r = 0.35) [39, 4. Discussion
42,44] and two reported regression coefficients of β = 0.25 [45,51].
Four of the five studies reported a positive association between systolic This systematic review does not support a consistent relationship
BPV and anxiety scores, Virtanen et al. [45] and Symonides et al. [44] between BPV with depression or anxiety. Unlike a previous review
reported a significant positive association between the CoV of systolic which broadly assessed bipolar disorder, mental illness, hostility, and
BPV and anxiety scores (β = 0.25 and r = 0.15 respectively). Similarly, BPV [34], this review focused specifically on uni-polar depression and
Koklu et al. [39] also reported a statistically significant increase in anxiety in adult populations and contemporary conceptualisations of
systolic BPV-SD (r = 0.23) and systolic BPV-CoV (r = 0.35) in individuals BPV. Our findings suggest that while there are sparse data supporting an
with high anxiety levels measured by the Beck Anxiety Inventory. In association between BPV with depression and anxiety, the findings were
contrast, one study reported a negative correlation between CoV of generally inconsistent, in different directions and strength, and not
systolic BPV and anxiety scores (r = − 0.39) [42], with authors reporting uniform across the BPV metrics. Some evidence implicated systolic BPV
similar findings between CoV of diastolic BPV and anxiety (r = − 0.40). in higher anxiety scores, however this was constrained largely to the
Only one study comprised of 1112 males did not report significant as­ CoV metric, whilst three studies reported a positive linear association,
sociations between systolic or diastolic BPV and anxiety [40]. another reported a negative association. Consequently, the results from
this systematic review provide insights into BPV’s role in cognitive and
3.5. Between depression group differences for bpv cerebrovascular health. As research has previously demonstrated that
BPV is associated with cognitive impairment, the findings of the current
Among the six studies included in this review that explored BPV review tentatively suggest that the relationship between BPV and
between groups with high depression versus those with low or no cognitive function is independent of depression, with only weak residual
depression (experimental versus control group), significant differences variance implicating anxiety.
in systolic and diastolic BPV between experimental and control groups The lack of a consistent association between BPV with common af­
were observed in two studies. Both reported significantly higher systolic fective states prompts a deeper exploration into distinct cognitive
BPV in persons with depression by comparison to persons without mechanism and brain structures beyond those traditionally linked to
depression [48,51], whereas only one study [48] found higher diastolic cognitive function and dementia. BPV has been linked to cerebrovas­
BPV in persons with depression. Otherwise, evidence for elevated BPV in cular changes in brain structures critical for cognitive processing such as
persons with depression was not reported by four other studies [38,41, the prefrontal cortex and subcortical white matter [22,23]. The brain
46,49] (Table 3). areas affected by depression and anxiety that are critical for mood

Table 3
Between group differences for BPV, depression and anxiety.
Systolic Diastolic

Author (Year) Total N Case Group N Control Group N BPV BPV Case Control Case Control
(female%) (female%) Measure Coefficient Group M (SD) Group M (SD)
M (SD) M (SD)

Depression
Davydov et al. (2007) 248 28 (60.7) 220 B2B SD 8.8 (3.1) 8.3 (2.9) – –
Schulz et al. (2010)a 114 57 (68.4) 57 (68.4) B2B SD 6.8 (3.2)* 5.2 (2.9)* 5.6 (2.5)* 4.2 (3.4)*
Schumann et al. (2017) 58 29 (72.4) 29 (72.4) B2B SD 5.51 4.53 (2.12) 3.96 3.22 (1.38)
(2.36) (1.71)
Imaizumi et al. (2016) 85 46 (68) 39 (68) ABP SD 23.3 (4.8) 21.5 (6.5) – –
Scuteri et al. (2008) 135 74 (84) 61 (66) ABP SD 13.3 (3.6) 13.9 (3.3) 9.8 (2.2) 10.0 (1.8)
Three-City - (2018) b 2297 105 (77.1) 2192 (60.6) ABP CoV 11.3 (2.4) 8.9 (2.9)* 10.1 (4.2) 9.1 (3.6)
(61) *
Anxiety
Sanchez-Gonzalez et al. 43 22 (55) 21 (52) B2B CoV 3.5 (0.3)* 1.5 (0.2)* – –
(2015)c
Koklu et al. (2022)d 72 30 42 ABP SD 12.61 13.02 17.02 11.91
(3.24) (3.96) (5.04) (4.21)
Three-City - (2017) 1454 84 (69) 1370 (59.1) ABP CoV 8.3 (2.5) 8.1 (2.9) 8.7 (3.7) 8.3 (3.6)

ABP, ambulatory blood pressure; B2B, beat-to-beat; BPV, blood pressure variability; CoV, coefficient of variation; IQR, interquartile range; N, population; S.E, standard
error coefficient; SD, standard deviation; V2V, visit-to-visit.
*
Statistical significance <0.05.
a
normal-to-normal beat time series.
b
Interquartile range reported.
c
Standard Error reported.
d
Moderate and Severe data combined for BAI case group.

5
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

regulation include the amygdala and hippocampus [25,52]. An associ­ future research and well-designed studies in the field of BPV and mood
ation between longer-term BPV with lower hippocampal volume was are crucial. More focus on longitudinal studies in future research is
recently observed in a narrative synthesis of six studies [9]. However, no needed to establish whether there is a temporal relationship between
consistent association between short-term BPV and hippocampal vol­ BPV and mood, as current evidence heavily relies on cross-sectional and
ume was found [9]. With >70 % of studies in the current review observational data. In future research, ideal studies should comprise of
quantifying short-term BPV such as through B2B or ABPM, it is possible larger sample sizes, with extensive follow-up periods, regular BP mea­
that there is a discrepant association between short- and long-term BPV surements including 24-hour monitoring, and repeated standardised
with structural changes in amygdala and hippocampus, accounting for assessments of depression and anxiety. These studies will significantly
the lack of association between BPV and affective states here. It is also contribute to filling the gap in V2V BPV research and also assist in
plausible that further imaging studies in frontotemporal dementia revealing whether there is a causal association between BPV and mood,
marked by behavioural and personality changes might elucidate asso­ which is particularly crucial given the established link between longer-
ciations between BPV and affect [53]. Further research into the impact term BPV and structural changes in the brain [9]. Researchers should
of longer-term BPV on neurogenerative conditions could provide crucial also attempt to control for confounding factors such as underlying health
insights into preventative and therapeutic interventions [24]. Exploring conditions, medications and demographic variables that may indepen­
the practical applications, in clinical settings BPV monitoring could be dently influence BPV and mood outcomes. Larger and more diverse
integrated into routine assessments for early detection of cognitive cohorts are also needed to improve the generalisability of findings
decline or mood disorders [54] to significantly advance patient care. The alongside comprehensive and validated mood assessments. Future in­
finding that calcium channel blockers, an antihypertensive drug which vestigations could also prioritise reporting more than one BPV metric (e.
reduces BPV more than other drugs, was associated with reduced g. CoV, SD, ARV) at a minimum as well as defining both linear and
depressive symptoms also highlights the potential for clinical relevance between-group comparisons with mood. Such consistency in methods
of BPV interventions to mood [55]. However, without further in­ would help further our understanding of BPV’s association with brain
vestigations on longer-term BPV and affect, in populations with and health and especially mood. Future research could also focus on
without dementia, it remains unclear if V2V measures of BPV over the observing the relationship between affect and BPV through longitudinal
longer term are relevant to depression and anxiety. This highlights the research with growth curve modelling or joint modelling the dynamic
need for targeted research that can validate whether BPV is a potential nature of BP fluctuations over time [59].
biomarker for cognitive and mental health disorders. In conclusion, this review contributes to improving our under­
While no consistent association was evident here between BPV and standing of the relationship between BPV and brain health. By
depression, the broad range of psychometric assessments utilised in the employing a more standardized and rigorous examination approach
included studies may underestimate a potential association between than previous studies, the review provides critical insights to the rela­
BPV with somatic and affective depression subtypes. Differentiating tionship between BPV and brain health. It reveals a lack of consistent
between affective disorders and their subtypes is crucial as recent association between BPV with depression, and only modest evidence for
literature reveals differences in brain activity across major depression a linear association between BPV and anxiety. The results highlight the
subtypes. Specifically, somatic depression involves increased activity in methodological heterogeneity within data, emphasizing the necessity
the brain’s right inferior temporal gyrus, enhancing physical sensation for more consistent research methodologies. Specifically, there is a need
perception, and decreased activity in the left hippocampus, impacting for further longitudinal studies focusing on V2V BPV to thoroughly
emotional regulation and memory, in contrast to non-somatic depres­ investigate the potential temporal relationship with mental health and
sion [56]. fMRI also reveals that somatic symptoms are linked to lower cognitive decline. Such research could significantly enhance our un­
functional connectivity in key areas responsible for emotional process­ derstanding of the mechanisms underlying neurodegeneration and
ing and pain perception [57]. Given the putative link between BPV and inform the development of targeted interventions that could mitigate
brain morphology such as reduced hippocampal volume [9], it remains these effects. This review therefore not only enriches current un­
possible that BPV is associated with only a somatic or affective major derstandings, but also shapes future research that could have profound
depression but not both subtypes. Such a putative association would implications for both clinical practice and public health strategies.
have been masked by the range of psychometric assessments here. This
underscores the requisite need for further research with uniform mea­ Funding
sures of depression and anxiety that facilitate the investigation of sub­
types. By identifying specific BPV patterns associated with subtypes of This research did not receive any specific grant from funding
mood, healthcare providers could tailor treatment more effectively [56, agencies in the public, commercial, or not-for-profit sectors.
57], potentially improving patient outcomes.
The detailed inclusion and exclusion criteria here ensuring only CRediT authorship contribution statement
contemporary measures of BPV in adult populations generally free from
major chronic diseases are key strengths that offer comprehensive in­ Yuvthi Lutchman: Conceptualization, Investigation, Methodology,
sights regarding the current literature into BPV and affect. Dis­ Data curation, Writing – original draft, Writing – review & editing. Rajiv
tinguishing linear relationships from mean differences between groups Mahajan: Writing – review & editing. Suzanne M. Cosh: Supervision,
allow researchers to better understand the heterogeneous nature of the Resources, Writing – review & editing. Katie Harris: Writing – review &
extant research. However, a notable limitation is the heterogeneity of editing. Christophe Tzourio: Writing – review & editing. Phillip J.
BPV measurement instruments used across studies, contributing to the Tully: Conceptualization, Investigation, Methodology, Data curation,
variability in results and complicating the identification of consistent Resources, Supervision, Writing – review & editing.
patterns. This variability, alongside the sparse homogeneous data, pre­
cluded the execution of a meta-analysis, resulting in a narrative review Declaration of competing interest
describing only trends. The paucity of eligible and comparable BPV
studies highlights the need for more research aimed at understanding The authors declare the following financial interests/personal re­
the association between BPV and affect. By comparison, previous re­ lationships which may be considered as potential competing interests:
views on brain health and BPV have included 18 studies pertaining to Dr Mahajan has served on the advisory board of Abbott and Med­
stroke [58], 27 concerning cerebral small vessel disease [10], 20 for tronic. The University of Adelaide reports receiving on behalf of Dr
other brain morphology [9], and 53 for dementia and cognition [54]. Mahajan lecture and/or consulting fees from Abbott, Bayer, Biotronik,
Given the inconsistent methodologies and findings across studies, Medtronic, and Pfizer. The University of Adelaide reports receiving on

6
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

behalf of Dr Mahajan research funding from Abbott, Bayer, and hyperintensities among older people with hypertension, J. Am. Med. Dir. Assoc. 20
(9) (2019) 1175–1177, [Link] e1.
Medtronic.
[18] W. Liu, R. Liu, W. Sun, Q. Peng, W. Zhang, E. Xu, et al., Different impacts of blood
Dr Tully reports receiving consulting fees from the Asia-Pacific pressure variability on the progression of cerebral microbleeds and white matter
Institute of Psycho-Cardiology Research. lesions, Stroke 43 (11) (2012) 2916–2922, [Link]
strokeaha.112.658369.
[19] A. Alperovitch, M. Blachier, A. Soumare, K. Ritchie, J.F. Dartigues, S. Richard-
Acknowledgements Harston, et al., Blood pressure variability and risk of dementia in an elderly cohort,
the three-city study, Alzheimers. Dement. 10 (5 Suppl) (2014) S330–S337, https://
The authors wish to thank Clara Murray for her administrative [Link]/10.1016/[Link].2013.05.1777.
[20] R.A.A. De Heus, M.G.M. Olde Rikkert, P.J. Tully, B.A. Lawlor, J.A.H.R. Claassen,
support. Blood pressure variability and progression of clinical alzheimer disease,
Hypertension 74 (5) (2019) 1172–1180, [Link]
Supplementary materials HYPERTENSIONAHA.119.13664.
[21] I.J. Sible, D.A. Nation, Visit-to-visit blood pressure variability and longitudinal tau
accumulation in older adults, Hypertension 79 (3) (2022) 629–637, [Link]
Supplementary material associated with this article can be found, in org/10.1161/hypertensionaha.121.18479.
the online version, at doi:10.1016/[Link].2024.100228. [22] S.P. Rensma, T.T. van Sloten, L.J. Launer, C.D.A. Stehouwer, Cerebral small vessel
disease and risk of incident stroke, dementia and depression, and all-cause
mortality: a systematic review and meta-analysis, Neurosci. Biobehav. Rev. 90
References (2018) 164–173, [Link]
[23] L. Wouts, M. Kessel, A.T.F. Beekman, R.M. Marijnissen, R.C. Oude Voshaar,
[1] G. Parati, J.E. Ochoa, C. Lombardi, G. Bilo, Assessment and management of blood- Empirical support for the vascular apathy hypothesis: a structured review, Int. J.
pressure variability, Nature Rev. Cardiol. 10 (3) (2013) 143–155, [Link] Geriatr. Psychiatry 35 (1) (2020) 3–11, [Link]
10.1038/nrcardio.2013.1. [24] Y.J. Jang, C. Kang, W. Myung, S.-W. Lim, Y.K. Moon, H. Kim, et al., Additive
[2] P.M. Rothwell, S.C. Howard, E. Dolan, E. O’Brien, J.E. Dobson, B. Dahlof, et al., interaction of mid- to late-life depression and cerebrovascular disease on the risk of
Prognostic significance of visit-to-visit variability, maximum systolic blood dementia: a nationwide population-based cohort study, Alzheimers. Res. Ther. 13
pressure, and episodic hypertension, Lancet 375 (9718) (2010) 895–905, https:// (1) (2021), [Link] 61-61.
[Link]/10.1016/s0140-6736(10)60308-x. [25] A. Etkin, T.D. Wager, Functional neuroimaging of anxiety: a meta-analysis of
[3] C.E. Hastie, P. Jeemon, H. Coleman, L. McCallum, R. Patel, J. Dawson, et al., Long- emotional processing in ptsd, social anxiety disorder, and specific phobia, Am. J.
term and ultra long-term blood pressure variability during follow-up and mortality Psychiatry 164 (10) (2007) 1476–1488, [Link]
in 14,522 patients with hypertension, Hypertension 62 (4) (2013) 698–705, ajp.2007.07030504.
[Link] [26] A. Jamieson, A.M. Goodwill, M. Termine, S. Campbell, C. Szoeke, Depression
[4] A.J. Webb, U. Fischer, Z. Mehta, P.M. Rothwell, Effects of antihypertensive-drug related cerebral pathology and its relationship with cognitive functioning: a
class on interindividual variation in blood pressure and risk of stroke, Lancet 375 systematic review, J. Affect. Disord. 250 (2019) 410–418, [Link]
(9718) (2010) 906–915, [Link] 10.1016/[Link].2019.03.042.
[5] D. Shimbo, J.D. Newman, A.K. Aragaki, M.J. LaMonte, A.A. Bavry, M. Allison, et [27] C. Nardin, M. Rattazzi, P. Pauletto, Blood pressure variability and therapeutic
al., Association between annual visit-to-visit blood pressure variability and stroke implications in hypertension and cardiovascular diseases, High Blood Pressure
in postmenopausal women: data from the women’s health initiative, Hypertension Cardiovasc. Prevent. 26 (5) (2019) 353–359, [Link]
60 (3) (2012) 625–630, [Link] 019-00339-z.
[6] L. Manning, Y. Hirakawa, H. Arima, X. Wang, J. Chalmers, J. Wang, et al., Blood [28] G. Parati, G. Bilo, A. Kollias, M. Pengo, J.E. Ochoa, P. Castiglioni, et al., Blood
pressure variability and outcome after acute intracerebral haemorrhage: post-hoc pressure variability: methodological aspects, clinical relevance and practical
analysis of interact2, Lancet Neurol. 13 (4) (2014) 364–373, [Link] indications for management - a european society of hypertension position paper,
10.1016/s1474-4422(14)70018-3. J. Hypertens. 41 (4) (2023) 527–544, [Link]
[7] S.L. Stevens, S. Wood, C. Koshiaris, K. Law, P. Glasziou, R.J. Stevens, et al., Blood HJH.0000000000003363.
pressure variability and cardiovascular disease: systematic review and meta- [29] C.S.P.D. Mackenzie, K.B.A. Reynolds, K.-L.P.D. Chou, J.M.A. Pagura, J.M.
analysis, BMJ 354 (2016) i4098, [Link] D. Sareen, Prevalence and correlates of generalized anxiety disorder in a national
[8] G. Parati, C. Torlasco, M. Pengo, G. Bilo, J.E. Ochoa, Blood pressure variability: its sample of older adults, Am. J. Geriatric Psychiatry 19 (4) (2011) 305–315, https://
relevance for cardiovascular homeostasis and cardiovascular diseases, [Link]/10.1097/JGP.0b013e318202bc62.
Hypertension Res. 43 (7) (2020) 609–620, [Link] [30] E. Richard, C. Reitz, L.H. Honig, N. Schupf, M.X. Tang, J.J. Manly, et al., Late-life
0421-5. depression, mild cognitive impairment, and dementia, JAMa Neurol. 70 (3) (2013)
[9] D.S. Gutteridge, P.J. Tully, E.S. Ghezzi, S. Jamadar, A.E. Smith, T. Commerford, et 374–382, [Link]
al., Blood pressure variability and structural brain changes: a systematic review, [31] H.C. Kales, D.F. Maixner, A.M. Mellow, Cerebrovascular disease and late-life
J. Hypertens. 40 (6) (2022) 1060–1070, [Link] depression, Am. J. Geriatr. Psychiatry 13 (2) (2005) 88–98.
HJH.0000000000003133. [32] W.C. Lin, L.Y. Hu, S.J. Tsai, A.C. Yang, C.C. Shen, Depression and the risk of
[10] P.J. Tully, Y. Yano, L.J. Launer, K. Kario, M. Nagai, S.P. Mooijaart, et al., vascular dementia: a population-based retrospective cohort study, Int. J. Geriatr.
Association between blood pressure variability and cerebral small-vessel disease: a Psychiatry 32 (5) (2017) 556–563, [Link]
systematic review and meta-analysis, J. Am. Heart. Assoc. 9 (1) (2020) e013841, [33] M.B. Snowden, D.C. Atkins, L.E. Steinman, J.F. Bell, L.L. Bryant, C. Copeland, et al.,
[Link] Longitudinal association of dementia and depression, Am. J. Geriatr. Psychiatry
[11] D.S. Gutteridge, P.J. Tully, A.E. Smith, T. Loetscher, H.A. Keage, Cross-sectional (2014), [Link]
associations between short and mid-term blood pressure variability, cognition, and [34] N.H. Shahimi, R. Lim, S. Mat, C.-H. Goh, M.P. Tan, E. Lim, Association between
vascular stiffness in older adults, Cereb. Circ. Cogn. Behav. 5 (2023) 100181, mental illness and blood pressure variability: a systematic review, Biomed. Eng.
[Link] Online 21 (1) (2022), [Link] 19-19.
[12] M. Nagai, S. Hoshide, K. Dote, K. Kario, Visit-to-visit blood pressure variability and [35] G.S. Stergiou, G. Parati, C. Vlachopoulos, A. Achimastos, E. Andreadis, R. Asmar, et
dementia, Geriatr. Gerontol. Int. 15 (Suppl 1) (2015) 26–33, [Link] al., Methodology and technology for peripheral and central blood pressure and
10.1111/ggi.12660. blood pressure variability measurement: current status and future directions –
[13] I.J. Sible, B. Yew, S. Dutt, Y. Li, A.E. Blanken, J.Y. Jang, et al., Selective position statement of the european society of hypertension working group on blood
vulnerability of medial temporal regions to short-term blood pressure variability pressure monitoring and cardiovascular variability, J. Hypertens. 34 (9) (2016)
and cerebral hypoperfusion in older adults, Neuroimage Rep. 2 (1) (2022), https:// 1665–1677, [Link]
[Link]/10.1016/[Link].2022.100080. [36] P.J. Tully, S. Debette, B. Mazoyer, C. Tzourio, White matter lesions are associated
[14] I.J. Sible, B. Yew, S. Dutt, K.J. Bangen, Y. Li, D.A. Nation, Visit-to-visit blood with specific depressive symptom trajectories among incident depression and
pressure variability and regional cerebral perfusion decline in older adults, dementia populations: three-city dijon mri study, Am. J. Geriatr. Psychiatry 25
Neurobiol. Aging 105 (2021) 57–63, [Link] (12) (2017) 1311–1321, [Link]
neurobiolaging.2021.04.009. [37] M.J. Page, J.E. McKenzie, P.M. Bossuyt, I. Boutron, T.C. Hoffmann, C.D. Mulrow, et
[15] T. Lohman, I. Sible, A. Kapoor, A.C. Engstrom, J.P. Alitin, A. Gaubert, et al., Blood al., The prisma 2020 statement: an updated guideline for reporting systematic
pressure variability, central autonomic network dysfunction and cerebral small reviews, Revista Española De Cardiología (English ed.) 74 (9) (2021) 790–799,
vessel disease in Apoe4 carriers, medRxiv. (2023), [Link] [Link]
2023.12.13.23299556. [38] Y. Imaizumi, K. Eguchi, T. Murakami, T. Saito, S. Hoshide, K. Kario, Locomotive
[16] Y. Yano, J.P. Reis, D.A. Levine, R.N. Bryan, A.J. Viera, D. Shimbo, et al., Visit-to- syndrome is associated with large blood pressure variability in elderly
visit blood pressure variability in young adulthood and hippocampal volume and hypertensives: the japan ambulatory blood pressure prospective (jamp) substudy,
integrity at middle age: the cardia study (coronary artery risk development in J. Clin. Hypertens. (Greenwich) 19 (4) (2017) 388–394, [Link]
young adults), Hypertension 70 (6) (2017) 1091–1098, [Link] jch.12946.
hypertensionaha.117.10144. [39] M. Koklu, A.F. Yagci, P.M. Coskun, H. Taskan, S. Yasar, S. Gormel, et al., The effect
[17] T. van Middelaar, E. Richard, E.P. Moll van Charante, W.A. van Gool, J.W. van of sleep quality and anxiety level on blood pressure variability, Sleep. Vigil. 7 (1)
Dalen, Visit-to-visit blood pressure variability and progression of white matter (2022) 23–31, [Link]

7
Y. Lutchman et al. Cerebral Circulation - Cognition and Behavior 6 (2024) 100228

[40] Y.P. Lin, C.H. Fan, K.Z. Tsai, K.H. Lin, C.L. Han, G.M. Lin, Psychological stress and the 3c dijon mri study, Psychol. Med. 48 (9) (2018) 1444–1453, [Link]
long-term blood pressure variability of military young males: the cardiorespiratory 10.1017/S0033291717002756.
fitness and hospitalization events in armed forces study, World J. Cardiol. 12 (12) [51] P.J. Tully, C. Tzourio, Psychiatric correlates of blood pressure variability in the
(2020) 626–633, [Link] elderly: the three city cohort study, Physiol. Behav. 168 (2017) 91–97, [Link]
[41] A. Scuteri, G. Spalletta, M. Cangelosi, W. Gianni, A. Assisi, A.M. Brancati, et al., org/10.1016/[Link].2016.10.024.
Decreased nocturnal systolic blood pressure fall in older subjects with depression, [52] S. Campbell, M. Marriott, C. Nahmias, G.M. MacQueen, Lower hippocampal
Aging Clin. Exp. Res. 21 (4–5) (2009) 292–297, [Link] volume in patients suffering from depression: a meta-analysis, Am. J. Psychiatry
BF03324918. 161 (4) (2004) 598–607, [Link]
[42] N.H. Shahimi, C.H. Goh, S. Mat, R. Lim, V.C.A. Koh, S.R. Nyman, et al., [53] S. Lattanzi, S. Luzzi, L. Provinciali, M. Silvestrini, Blood pressure variability in
Psychological status and physical performance are independently associated with alzheimer’s disease and frontotemporal dementia: the effect on the rate of
autonomic function, Biomed. Eng. Online 21 (1) (2022) 29, [Link] cognitive decline, J. Alzheimers. Dis. 45 (2) (2015) 387–394, [Link]
10.1186/s12938-022-00996-7. 10.3233/jad-142532.
[43] I.J. Sible, J.Y. Jang, D.L. Sultzer, D.A. Nation, Visit-to-visit blood pressure [54] R.A.A. de Heus, C. Tzourio, E.J.L. Lee, M. Opozda, A.D. Vincent, K.J. Anstey, et al.,
variability and subthreshold depressive symptoms in older adults, Am. J. Geriatr. Association between blood pressure variability with dementia and cognitive
Psychiatry 30 (10) (2022) 1110–1119, [Link] impairment: a systematic review and meta-analysis, Hypertension (Dallas, Tex.
jagp.2022.03.006. 1979) 78 (5) (2021) 1478–1489, [Link]
[44] B. Symonides, P. Holas, M.g. Schram, J. leszycka, A. Bogaczewicz, Z. Gaciong, Does HYPERTENSIONAHA.121.17797.
the control of negative emotions influence blood pressure control and its [55] P.J. Tully, R. Peters, K. Pérès, K.J. Anstey, C. Tzourio, Effect of Ssri and calcium
variability? Blood Press. 23 (6) (2014) 323–329, [Link] channel blockers on depression symptoms and cognitive function in elderly persons
08037051.2014.901006. treated for hypertension: three city cohort study, Int. Psychogeriatr. 30 (9) (2018)
[45] R. Virtanen, A. Jula, J.K. Salminen, L.M. Voipio-Pulkki, H. Helenius, T. Kuusela, et 1345–1354, [Link]
al., Anxiety and hostility are associated with reduced baroreflex sensitivity and [56] R. Yan, S. Tao, H. Liu, Y. Chen, J. Shi, Y. Yang, et al., Abnormal alterations of
increased beat-to-beat blood pressure variability, Psychosom. Med. 65 (5) (2003) regional spontaneous neuronal activity in inferior frontal orbital gyrus and
751–756. corresponding brain circuit alterations: a resting-state fmri study in somatic
[46] D.M. Davydov, D. Shapiro, I.A. Cook, I. Goldstein, Baroreflex mechanisms in major depression, Front. Psychiatry 10 (2019), [Link]
depression, Prog. Neuropsychopharmacol. Biol. Psychiatry 31 (1) (2007) 164–177, fpsyt.2019.00267, 267-267.
[Link] [57] R. Yan, J.T. Geng, Y.H. Huang, H.W. Zou, X.M. Wang, Y. Xia, et al., Aberrant
[47] M.A. Sanchez-Gonzalez, P. Guzik, R.W. May, A.P. Koutnik, R. Hughes, S. Muniz, et functional connectivity in insular subregions in somatic depression: a resting-state
al., Trait anxiety mimics age-related cardiovascular autonomic modulation in fmri study, BMC. Psychiatry 22 (1) (2022), [Link]
young adults, J. Hum. Hypertens. 29 (4) (2015) 274–280, [Link] 03795-5, 146-146.
10.1038/jhh.2014.72. [58] L.S. Manning, P.M. Rothwell, J.F. Potter, T.G. Robinson, Prognostic significance of
[48] S. Schulz, M. Koschke, K.J. Bar, A. Voss, The altered complexity of cardiovascular short-term blood pressure variability in acute stroke: systematic review, Stroke 46
regulation in depressed patients, Physiol. Meas. 31 (3) (2010) 303–321, https:// (9) (2015) 2482–2490, [Link]
[Link]/10.1088/0967-3334/31/3/003. (1970).
[49] A. Schumann, C. Andrack, K.J. Bar, Differences of sympathetic and [59] H. De Courson, L. Ferrer, A. Barbieri, P.J. Tully, M. Woodward, J. Chalmers, et al.,
parasympathetic modulation in major depression, Prog. Neuropsychopharmacol. Impact of model choice when studying the relationship between blood pressure
Biol. Psychiatry 79 (Pt B) (2017) 324–331, [Link] variability and risk of stroke recurrence, Hypertension 78 (5) (2021) 1520–1526,
pnpbp.2017.07.009. [Link]
[50] P.J. Tully, S. Debette, C. Tzourio, The association between systolic blood pressure
variability with depression, cognitive decline and white matter hyperintensities: