Stroke

ORIGINAL CONTRIBUTION

Body Mass Index and Arterial Stiffness Are

Associated With Greater Beat-to-Beat Blood
Pressure Variability After Transient Ischemic
Attack or Minor Stroke
Alastair J.S. Webb , BMBCh, DPhil; Amy Lawson , BSc; Sara Mazzucco , MD, PhD; Linxin Li , DPhil;
Peter M. Rothwell , FMedSci; for the Oxford Vascular Study Phenotyped Cohort

BACKGROUND AND PURPOSE: Blood pressure variability (BPV) from beat to beat is associated with an increased risk of
cardiovascular events and enables rapid assessment of BPV, but the underlying causes of elevated BPV are unclear.

METHODS: In consecutive patients within 4 to 6 weeks of transient ischemic attack or nondisabling stroke (OXVASC [Oxford
Vascular Study]), continuous noninvasive blood pressure was measured beat to beat over 5 minutes (Finometer). Arterial
stiffness was measured by carotid-femoral pulse wave velocity (Sphygmocor). After automated and manual data cleaning,
associations between BPV (residual coefficient of variation), demographic factors, and arterial stiffness were determined for
both systolic and diastolic blood pressure, by ANOVA and linear models. Relationships between demographic factors and
arterial stiffness were determined by interaction terms and mediation.

RESULTS: Among 1013 patients, 54 (5.3%) were in AF, and 51 (5%) had low-quality recordings. In a general linear model
Downloaded from http://ahajournals.org by on March 16, 2021

including the remaining 908 participants, systolic BPV (SBPV) was most strongly associated with age (P=0.00003), body
mass index (BMI; P=0.003), and arterial stiffness (P=0.008), with weaker independent associations with current smoking
(P=0.01) and a low diastolic blood pressure (P=0.046). However, while there was a linear increase in SBPV with BMI in men,
in women, SBPV was lowest for a BMI in the normal range but was greater below 20 or above 30 (ANOVA, P=0.012; BMI-
sex interaction, P=0.03). Although BMI and pulse wave velocity were partially independent, increased pulse wave velocity
mediated ≈32% of the relationship between increased BMI and SBPV (P<0.001).
CONCLUSIONS: Vascular aging, manifest as arterial stiffness, was a strong determinant of increased SBPV and partially
mediated the effect of increased BMI. However, although high BMI was independently associated with SBPV
in both sexes, a low BMI was associated with increased SBPV only in women. SBPV may partially mediate the
relationship between BMI and cardiovascular events, while obesity may provide a modifiable target to reduce SBPV
and cardiovascular events.

GRAPHIC ABSTRACT: An online graphic abstract is available for this article.

Key Words:  blood pressure ◼ body mass index ◼ linear models ◼ systole ◼ vascular stiffness

 
Correspondence to: Alastair J.S. Webb, BMBCh, DPhil, Department of Clinical Neurosciences, Wellcome Trust Clinical Research Career Development Fellowship,
Wolfson Centre for Prevention of Stroke and Dementia, University of Oxford, John Radcliffe Hospital, Headington, Oxford OX3 9DU, United Kingdom. Email alastair.
[email protected]
This manuscript was sent to Emmanuel Touzé, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
The Data Supplement is available with this article at https://www.ahajournals.org/doi/suppl/10.1161/STROKEAHA.120.031179.
For Sources of Funding and Disclosures, see page XXX.
© 2021 The Authors. Stroke is published on behalf of the American Heart Association, Inc., by Wolters Kluwer Health, Inc. This is an open access article under the
terms of the Creative Commons Attribution License, which permits use, distribution, and reproduction in any medium, provided that the original work is properly cited.
Stroke is available at www.ahajournals.org/journal/str

Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179 April 2021   1

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

between beat-to-beat BPV with arterial stiffness and

Original Contribution

Nonstandard Abbreviations and Acronyms major cardiovascular risk factors and which demographic
features determine a likely pathological excess of beat-
BMI body mass index to-beat BPV.
BP blood pressure
BPV blood pressure variability
CV coefficient of variation METHODS
DBP diastolic blood pressure Study Population
DBPV diastolic blood pressure variability Consecutive, consenting patients with transient ischemic
OXVASC Oxford Vascular Study attack or minor stroke (National Institutes of Health Stroke
PWV pulse wave velocity Scale score, <5) were recruited between September 2010 and
SBP systolic blood pressure October 2019, as part of the Phenotyped Cohort of OXVASC
(Oxford Vascular Study).15,16 Participants were recruited at the
SBPV systolic blood pressure variability
OXVASC daily emergency assessment clinic, following a refer-
ral after attendance at the Emergency Department or from pri-

P
mary care, usually within 24 hours. Patients were referred after
atients with episodic hypertension after a cerebro-
transient neurological symptoms or symptoms consistent with
vascular event have a high risk of recurrent stroke,1,2
a minor stroke, not requiring direct admission to hospital. The
residual visit-to-visit variability in blood pressure OXVASC population consists of >92 000 individuals registered
(BPV) on treatment has a poor prognosis despite good with about 100 primary-care physicians in Oxfordshire, United
control of mean blood pressure (BP),3 and benefits of Kingdom.17 All consenting patients had a standardized medical
some antihypertensive drugs in the prevention of stroke history and examination, ECG, blood tests, and a stroke proto-
may partly result from reduced variability in systolic BP col brain magnetic resonance imaging and contrast-enhanced
(SBPV).4–6 Strong associations between visit-to-visit magnetic resonance angiography (or computed tomography of
BPV, cardiovascular events, renal impairment, and cog- the brain and carotid Doppler ultrasound or computed tomogra-
nitive decline have now been demonstrated in popu- phy angiogram), an echocardiogram, and 5-day ambulatory car-
lation-based cohorts,7,8 patients with diabetes,9 renal diac monitor. All patients were assessed by a study physician,
reviewed by the senior study neurologist (P.M.R.), and were
impairment,10 cognitive decline,11 and other groups,12 with
followed up face to face at 1, 3, 6, and 12 months and 2, 5,
similar predictive value of BPV from day to day on home
and 10 years. Medication is standardly prescribed according to
Downloaded from http://ahajournals.org by on March 16, 2021

readings,13–15 with a significant reduction in BPV with guidelines, most commonly with dual antiplatelets (aspirin and
specific antihypertensives.4,6 However, both visit-to-visit clopidogrel), high-dose statins (atorvastatin, 40–80 mg), and a
and home BPV require a prolonged period of assess- combination of perindopril and indapamide, with the possible
ment, good patient compliance, and follow-up visits, addition of amlodipine, to reach a target of <130/80 on home
limiting their use in acute assessment or for assessing telemetric BP monitoring.16 Access to the data will be openly
treatment response in clinical practice and future trials. considered on application to the chief investigator (P.M.R.).
BPV from one beat to the next (beat-to-beat BPV) is As part of the OXVASC Phenotyped Cohort, a routine
also associated with an increased risk of recurrent stroke or prospective cardiovascular physiological assessment is per-
cardiovascular events in patients with a transient ischemic formed at the 1-month follow-up visit between September
2010 and September 2019. Participants were excluded if they
attack or minor stroke, with a similar physiological profile to
were under 18 years of age, had severe cognitive impairment,
home day-to-day BPV,16 and has at least similar predictive
were pregnant, or had autonomic failure, a recent myocar-
value,15 while enabling BPV assessment at a single visit. dial infarction, unstable angina, heart failure (New York Heart
However, the determinants of beat-to-beat BPV in at-risk Association class 3–4 or ejection fraction <40%), or untreated
individuals and its covariance with other major cardiovas- bilateral carotid stenosis (>70%). OXVASC is approved by the
cular risk factors is unclear. Furthermore, beat-to-beat BPV Oxfordshire Research Ethics Committee.
is itself composed of multiple components contributing to After 15 to 20 minutes of supine rest, beat-to-beat BPV
total BPV, from physiological rhythms reflecting breathing was measured over 5 minutes in a quiet, dimly lit, temperature-
and intact autonomic function to increased BPV associ- controlled room (21–23 °C). Continuous ECG and noninvasive
ated with stiff arteries and impaired baroreceptor function BP were acquired at 200 Hz (Finometer; Finapres Medical
in older patients with impaired compensatory mecha- Systems, the Netherlands), via a Powerlab 8/30 with LabChart
Pro software (ADInstruments). Waveforms were preferen-
nisms.16 To assess the potential clinical utility of beat-to-
tially recorded from the middle phalanx of the middle finger.
beat BPV, and to identify populations at an increased risk
Automated calibration (Physiocal) was performed until the
of elevated BPV, it is necessary to understand the determi- recording was stable but turned off during each test. Estimated
nants of beat-to-beat BPV in at-risk populations and how brachial waveforms (Finometer) were calibrated offline by lin-
this varies by demographic characteristics. ear regression to 2 to 3 supine, oscillometric brachial readings,
Therefore, in patients attending a transient ischemic performed immediately before the monitoring period on the
attack and minor stroke clinic, we determined associations contralateral arm, with manual exclusion of artifacts. In patients

2   April 2021 Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

with a significant deterioration in recording quality during the RESULTS

Original Contribution
first 5 minutes, the test was stopped and the calibration pro-
cedure repeated. If necessary, the finger cuff was moved to an One thousand thirty-one assessments were performed in
adjacent finger or the proximal phalanx of the same finger, or 1013 eligible, consecutive, consenting patients between
the hand was warmed with a hand warmer. Before physiological September 2010 and October 2019, with 18 patients
assessment, 2 sitting clinic BPs, 5 minutes apart, were mea- assessed twice after separate clinical presentations. Fifty-
sured at ascertainment and 1 month in the nondominant arm, four of 1013 (5.3%) patients were in atrial fibrillation dur-
by trained personnel. ing the recording, while 51 (5%) patients had inadequate
recordings. Patients with atrial fibrillation or poor record-
Analysis ing quality were older, had higher BP, and were more
BPV on beat-to-beat monitoring was calculated over 5 min- likely to have a history of hypertension (Table 1).
utes. Ectopic beats and artifacts were automatically detected Mean systolic BP (SBP) was strongly correlated with
from the R-R interval of the ECG, visually reviewed and SD of SBP (Figure I in the Data Supplement) but with
removed by linear interpolation of the R-R interval. BP arti- no correlation with CV-SBP or CV-diastolic blood pres-
facts were automatically detected, manually reviewed, and sure (DBP), before or after detrending of the recording
removed by linear interpolation to adjacent normal beats with (residual CV). However, there was an inverse correla-
in-house software. Patients in atrial fibrillation during the
tion between CV-DBP and mean DBP before and after
recording were excluded. SBPV and diastolic BPV (DBPV)
detrending (Figure I in the Data Supplement).
were calculated as the SD and the coefficient of variation (CV;
CV=SD/mean), before and after detrending of the recording Beat-to-beat BPV increased with age (Tables 1 and 2,
about a linear regression across 5 minutes. All recordings and Table III in the Data Supplement) with a slightly
were reviewed blinded to clinical data, after automated and greater SBPV in women, which was not significant
manual data cleaning, to assess for the quality of recording (3, after adjustment for age. The other key univariate and
excellent quality; 2, adequate quality for analysis; 1, unusable, adjusted associations of increased beat-to-beat SBPV
poor quality recording), based upon the presence of artifacts were increased body mass index (BMI) and increased
or drift in the baseline measurement. arterial stiffness (Table 2). After adjustment for age, sex,
Associations with demographic indices were determined and other cardiovascular risk factors, current smoking
by general linear models and by logistic regression for the top was also associated with increased SBPV. A history of
decile of BPV on each index. Models were performed for uni-
hypertension, diabetes, or dyslipidemia was not associ-
variate associations; adjusted for age and sex and for age, sex,
ated. Only age was associated with DBPV in univariate
Downloaded from http://ahajournals.org by on March 16, 2021

and cardiovascular risk factors at study entry (current smoking,

history of hypertension, and diabetes). analysis, although BMI was associated with DBPV after
Analyses were performed in R, Matlab r2015, or Windows adjustment for age, sex, and cardiovascular risk factors
Excel. (Table II in the Data Supplement).

Table 1.  Characteristics of Patients Included and Excluded From the Analysis

All Included In AF Poor data P value

1013 908 (90) 54 (5) 51 (5)
Age, y 67 (13.3) 66.1 (13.3) 77.8 (8.9) 72.2 (11.1) <0.001

Sex 427 (42) 387 (43) 18 (33) 22 (44.9) 0.35

Hypertension 453 (44) 390 (43) 36 (67) 27 (55.1) <0.001

Diabetes 116 (11) 104 (11.5) 7 (13) 5 (10.2) 0.88

Smoking
 Current 160 (15) 148 (16) 6 (11) 6 (12.2) 0.39
 Ever 535 (53) 478 (53) 32 (59) 24 (49) 0.62
FHx of stroke 303 (30) 274 (30) 12 (22) 17 (34.7) 0.12
AF
  During test 54 (5.3) … 54 (100) … …
 Ever 74 (7) 37 (4.1) 35 (65) 2 (4.1) <0.001

Dyslipidemia 314 (31) 277 (31) 20 (37) 16 (32.7) 0.67

Heart failure 16 (1.6) 10 (1.1) 6 (11) … <0.001

Systolic BP 133 (19) 132 (18) 137 (24) 145 (27) <0.001

Diastolic BP 78 (12) 78 (11) 83 (15) 77 (8.9) 0.01

Participants were excluded if they were in AF during the assessment or if their recording was of poor quality with a high probability
of artifactual recordings. P values are given for ANOVA for continuous variables and χ2 tests for categorical variables. AF indicates
atrial fibrillation; BP, blood pressure; and FHx, family history.

Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179 April 2021   3

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

Table 2.  Linear Models for Baseline Demographic Measures and SBP Level and Variability
Original Contribution

Mean SBP SBP-rCV

Unadjusted Adj age+sex Adj CV-RFs Unadjusted Adj age+sex Adj CV-RFs

β P value β P value β P value β P value β P value β P value

Age, y 0.34 <0.001* 0.34 <0.001* 0.35 <0.001* 0.02 0.001* 0.02 0.001* 0.02 <0.001*

Women 0.47 0.692 −0.13 0.909 −0.15 0.898 0.27 0.039* 0.24 0.064 0.26 0.057
HTN reported 7.83 <0.001* 5.51 <0.001* 6.33 <0.001* 0.03 0.804 −0.11 0.415 −0.06 0.728
HTN treated 5.04 <0.001* 1.93 0.171 2.26 0.124 −0.07 0.636 −0.26 0.111 −0.25 0.139
Diabetes 1.62 0.377 1.21 0.499 1.09 0.547 0.31 0.123 0.33 0.109 0.39 0.056
Dyslipidemia
 Reported 1.4 0.274 −0.53 0.675 −0.87 0.526 −0.15 0.278 −0.23 0.119 −0.24 0.115
 Treated 1.52 0.313 −0.69 0.645 −1.85 0.239 −0.15 0.383 −0.24 0.162 −0.27 0.123
Smoking
 Ever −0.13 0.91 0.39 0.738 −0.75 0.561 0.04 0.793 0.11 0.418 0.01 0.958
 Current −0.6 0.708 3.09 0.055 3.61 0.041* 0.18 0.311 0.41 0.026* 0.42 0.036*
Height −3.89 0.877 2.83 0.908 3.52 0.885 −1.12 0.689 −0.24 0.932 −0.06 0.982
Weight −0.01 0.732 0.02 0.443 0.01 0.571 0 0.21 0.01 0.024* 0.01 0.032*
BMI −0.1 0.411 0.02 0.842 0.01 0.938 0.03 0.046* 0.03 0.01* 0.04 0.005*
Creatinine 0 0.958 −0.06 0.085 −0.06 0.077 0 0.503 0 0.294 0 0.243
Cholesterol −0.16 0.68 0.1 0.789 0.17 0.654 0.01 0.759 0.01 0.805 0.03 0.506
LDL −0.01 0.555 −0.01 0.564 −0.01 0.593 0 0.282 0 0.229 0 0.285
HDL 1.16 0.323 0.32 0.793 0.71 0.563 −0.05 0.704 −0.2 0.127 −0.2 0.15
TSH 0.23 0.487 0.13 0.684 0.19 0.558 −0.05 0.139 −0.06 0.105 −0.05 0.13
On antihypertensive 4.18 0.002* 1.11 0.42 −5.96 0.168 −0.04 0.805 −0.21 0.175 0.16 0.736
On statin 1.11 0.439 −0.79 0.578 −1.19 0.736 −0.2 0.215 −0.28 0.078 −0.58 0.143
PWV 2.43 <0.001 2.29 <0.001 2.35 <0.001 0.14 <0.001 0.11 0.002 0.11 0.002
Downloaded from http://ahajournals.org by on March 16, 2021

SBP variability is analyzed as the rCV, unadjusted or adjusted for age and sex or for age, sex, and CV-RFs. Independent variables include HTN reported by patients
at baseline or defined by patient report and current treatment at baseline. Current treatment and PWV measures are given at the time of the physiological assessment.
Adj indicates adjusted; BMI, body mass index; CV-RF, cardiovascular risk factor; DBP, diastolic blood pressure; HDL, high-density lipoprotein; HTN, hypertension; LDL,
low-density lipoprotein; PWV, pulse wave velocity; rCV, residual coefficient of variation; SBP, systolic blood pressure; and TSH, thyroid-stimulating hormone.
*P<0.05.

BMI was the strongest clinical risk factor associ- The difference in the pattern of change with BMI
ated with increased BPV. It was not associated with between sexes was evident in a multivariate model,
mean or maximum SBP but only with SBPV (Figure with a significant association between BMI and SBPV
II in the Data Supplement) and was negatively cor- or DBPV when including all indices significantly associ-
related with age (P<0.0001). There was a significant ated with BPV after adjustment for age, sex, and CV
difference in BPV between standardly defined BMI risk factors (Table 2). Furthermore, there was a signifi-
groups in women (ANOVA P=0.012), with a U-shaped cant interaction between BMI with sex for SBPV, but
relationship (Figure 1), such that SBPV in women with not for DBPV, reflecting the linear increase in SBPV
a BMI between 25 and 30 was significantly lower than in men but not women. This was confirmed on strati-
women with a BMI below 20 (post hoc Tukey honestly fying the population by sex, with a significant associa-
significant difference, P=0.04) or >30 (P=0.0499). tion between BMI and BPV in men, exceeding even the
The elevated SBPV in women with a low BMI did not association between age and BPV (Figure IV in the Data
reflect a specific excess of high BPV in this group, with Supplement), while in women, there was no overall lin-
a similarly shaped distribution of SBPV across differ- ear association with the only strong determinant being
ent levels of BMI. In contrast, there was no difference the relationships between age and low DBP with SBPV
between absolute BMI categories in men (Figure 2), (Table 3), reflecting the nonlinear pattern in women.
with a linear increase in BPV when BMI was defined In addition to BMI, arterial stiffness was the factor
by quintiles (Figure I in the Data Supplement). This most strongly associated with BPV, even compared with
pattern predominantly reflected differences in weight, age. A linear increase in SBPV with arterial stiffness was
with a more linear increase in BPV across quintiles seen in both men and women (Figure 2), with a steeper
of weight in men, with no change across quintiles of relationship in women than in men but with no signifi-
height (Figure III in the Data Supplement). cant interaction with sex (P=0.38). This association was

4   April 2021 Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

Original Contribution
Downloaded from http://ahajournals.org by on March 16, 2021

Figure 1. Relationship between increased carotid-femoral pulse wave velocity beat-to-beat blood pressure variability, stratified
by sex.
Results are presented as the mean and CI for each group. Residual coefficient of variation (rCV) is calculated as the coefficient of variation of
detrended recordings. y axes are set to the 95% range of each measure across the whole population. DBP indicates diastolic blood pressure;
PWV, pulse wave velocity; and SBP, systolic blood pressure.

sufficiently strong that there was no association between sex and smoking (Table 3). There was no interaction
age and SBPV after adjustment for pulse wave veloc- between smoking and BMI for BPV and no association
ity (PWV; adjusted for age and sex: age, P=0.19; sex, between thyroid-stimulating hormone level and either
P=0.007; PWV, P=0.002). BMI or BPV. In a causal mediation analysis allowing for
There was no univariate association between BMI adjustment for age and sex, there was also a significant
and PWV (β=0.03, P=0.17), but after adjustment for indirect mediation effect of the relationship between
age and sex, there was a strong association (β=0.06, BMI and SBPV by PWV (average causal mediation
P<0.001), which persisted after adjustment for other effect: 0.008, P<0.001; average direct effect: 0.016,
cardiovascular risk factors (β=0.05, P=0.008). In a P=0.3; total effect: 0.024), explaining 32% of the rela-
combined model allowing for the sex×BMI interaction, tionship between BMI and SBPV. There was no sig-
BMI and PWV independently predicted SBPV, with no nificant indirect mediation by BMI of the relationship
residual significant association between either age or between PWV and SBPV (average causal mediation
DBP with SBPV, although associations remained for effect, P=0.12; 10% variance explained).

Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179 April 2021   5

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV
Original Contribution

Figure 2. Mean values of residual coefficient of variation (rCV) for systolic blood pressure (SBP) and diastolic blood pressure
(DBP), stratified by sex and standard thresholds for body mass index (BMI).
Results are presented as the mean and CI for each group. rCV is calculated as the coefficient of variation of detrended recordings. There is a
significant difference between groups for SBP-rCV for women only (ANOVA, P=0.014), with significant post hoc tests adjusted for multiple
comparisons shown (Tukey test).

However, there is as yet little evidence identifying

DISCUSSION the determinants and clinical characteristics of patients
In an at-risk population with recent transient ischemic with increased beat-to-beat BPV. We previously demon-
attack or minor stroke, beat-to-beat BPV over 5 minutes strated in a much smaller sample that beat-to-beat BPV
was associated with increased BMI and arterial stiffness. was associated with aortic stiffness and pulsatility, as
There was a linear increase in SBPV with BMI in men, but well as with markers of cardiovascular reactivity,16 and
both low and high BMI were associated with increased that SBPV increases with age with an increased skew of
Downloaded from http://ahajournals.org by on March 16, 2021

SBPV in women. PWV and BMI were independently asso- the distribution of SBPV in a subset of the population. In
ciated with SBPV, but PWV also mediated a proportion of this report, increased BMI was the strongest clinical fac-
the effect of BMI on SBPV. Associations between current tor associated with increased SBPV. However, the rela-
smoking and increased SBPV persisted despite adjust- tionship between BMI and SBPV was complex, with a
ment, but associations between both age and falling DBP linear association in men. However, in women, there was
were not significant after adjustment for PWV, indicating a a U-shaped relationship, with increased SBPV in women
likely primary role for arterial stiffness in mediating the rela- with both a reduced and increased BMI compared with
tionship between age, DBP, or BMI with beat-to-beat BPV. normal BMI, consistent with an increased risk of mortal-
Despite the large number of studies demonstrating ity for patients when both overweight and underweight.
that visit-to-visit and day-to-day BPV are associated with Although this could reflect physiological variability in
an increased risk of cardiovascular events,1–4,7–9,11,12,18 few women with a reduced BMI that could be beneficial,
studies have determined the prognostic significance of the marked positive skew of the distribution of SBPV in
beat-to-beat blood variability,15,19 despite its widespread women with a BMI below 20 suggests that this is not the
use in the assessment of autonomic function in both case, and that the elevation in SBPV is driven by patients
research and clinical practice.20,21 We previously dem- with an excess of pathological SBPV, even in women
onstrated in an earlier report from this population that with a reduced BMI. This may reflect increased auto-
beat-to-beat BPV was associated with a 47% increased nomic instability and sympathetic overactivity in obese
risk of stroke and 37% increased risk of cardiovascu- patients but could reflect reverse causation, with frailer
lar events per SD of beat-to-beat BPV,15 compared with patients being predisposed both to being underweight
24% and 33% for day-to-day BPV. One other study and to having increased BPV. Alternatively, elevated
demonstrated that beat-to-beat BPV was increased SBPV may mediate some of the relationship between
in acute stroke and associated with poor outcome,19,22 obesity or being underweight and cardiovascular risk and
albeit with SD as the principle index of BPV, while beat- represent a new treatment target.
to-beat BPV was associated with markers of end-organ This study confirmed the previously demonstrated
injury and vascular aging, both in this population16 and in association between arterial stiffness and beat-to-beat
limited studies in other populations.23 This was also con- BPV, with no significant association between age and
sistent with limited studies using intra-arterial continuous SBPV after adjustment for PWV. Furthermore, the medi-
BP measurements.24,25 ation analysis suggests that it is partly increased arterial

6   April 2021 Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

Table 3.  Multivariate Associations and Interactions Between BMI and Sex for Variability in SBP and DBP

Original Contribution
Entire population Stratified by sex

SBP DBP SBP DBP

β P value β P value β P value β P value

Model 1 Men
  Age, y 0.02 0.0003† 0.05 0.33 0.02 0.01* 0.05 0.49
 Women 0.22 0.11 0.69 0.65 … … … …
 DBP −0.01 0.046* −0.07 0.21 −0.01 0.29 −0.05 0.48
 Smoking 0.47 0.01* 3.6 0.09 0.39 0.07 3.0 0.23
 BMI 0.04 0.003† 0.35 0.02* 0.07 0.0002† 0.49 0.02*
Model 2 Women
  Age, y 0.02 0.0002† 0.06 0.31 0.03 0.004† 0.07 0.43
 Women 1.78 0.015* 9.1 0.26 … … … …
 DBP −0.01 0.03* −0.07 0.18 −0.02 0.04* −0.12 0.22
 Smoking 0.50 0.01* 3.7 0.08 0.66 0.06 4.9 0.20
 BMI 0.07 0.0003† 0.51 0.02* 0.02 0.45 0.2 0.35
 BMI×women −0.06 0.03* −0.31 0.29 … … … …
Model 3
  Age, y 0.015 0.06 −0.01 0.88
 Women 2.31 0.01* 1.29 0.21
 DBP −0.01 0.07 −0.07 0.29
 Smoking 0.60 0.008* 0.47 0.07
 PWV 0.10 0.008* 0.11 0.01*
 BMI 0.06 0.01* 0.029 0.28
 BMI×women −0.07 0.03* −0.042 0.27
Downloaded from http://ahajournals.org by on March 16, 2021

β-Coefficients and P values are given from general linear models, adjusted for significant covariates and for the interaction between BMI and sex or stratified by sex.
BP variability is measured as the residual coefficient of variation after detrending of data (rCV). BMI indicates body mass index; BP, blood pressure; DBP, diastolic blood
pressure; PWV, pulse wave velocity; rCV, residual coefficient of variation; and SBP, systolic blood pressure.
*P<0.05, †P<0.01.

stiffness in patients with increased BMI that results this risk has been shown to be associated with beat-to-
in increased SBPV, although this does not explain the beat BPV in this group.15 Second, 5% of patients did not
increased SBPV in patients with reduced BMI. One pos- have adequate recordings, despite repeated measures to
sible link between arterial stiffness, increased BMI, and improve quality, particularly in elderly patients who may
SBPV is an enhanced inflammatory cascade in obesity be at a particularly increased risk of stroke. As such, the
leading to endothelial dysfunction, increased arterial prevalence of elevated BPV may be underestimated, with
stiffness and atherosclerosis, and potentially increased consequent underestimation of the risk associated with
SBPV,26,27 either directly or indirectly, with recent tri- elevated BPV. Third, we measured beat-to-beat BPV in
als of anti-inflammatory interventions reducing recur- a highly controlled environment, using expensive equip-
rent cardiovascular events,28 independently of mean BP ment. Development of more cost-effective methods would
effects.29 This also implies that weight loss in the obese be essential to apply beat-to-beat BPV to routine clinical
has the potential to reduce cardiovascular morbidity, and practice. Fourth, we extensively cleaned and detrended
stroke in particular, through reductions in arterial stiff- the data, improving precision of measurement but also
ness and, therefore, reductions in BPV. Furthermore, limiting its direct applicability to clinical practice. As such,
given the negative correlation between age and BMI, further development is required to standardize methods
this supports the hypothesis that the rising incidence of of acquisition, data cleaning, and analysis of beat-to-beat
stroke in younger patients, and women in particular, may BPV in a practical method for use in clinical practice.
partly be driven by increasing weight that may partly be Finally, obesity may cause systematic bias in the assess-
mediated by increased SBPV. ment of PWV through artificially increasing the measured
There are limitations to our study. First, all patients were distance between the carotid and femoral applanation
assessed after a cerebrovascular event, limiting gener- sites, while also affecting accuracy of BP measurement.
alizability to other disease groups. However, this popu- Overall, beat-to-beat SBPV reflected both age-associ-
lation is at an increased risk of recurrent stroke,30 and ated arterial stiffness and changes in BMI. This suggests

Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179 April 2021   7

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

a potential role for weight loss to reduce SBPV in the systolic blood pressure, and episodic hypertension. Lancet. 2010;375:895–
Original Contribution

905. doi: 10.1016/S0140-6736(10)60308-X

obese, with a resulting reduction in cardiovascular risk, 3. Rothwell PM, Howard SC, Dolan E, O’Brien E, Dobson JE, Dahlöf B, Poulter
but further research is required in underweight women NR, Sever PS; ASCOT-BPLA and MRC Trial Investigators. Effects of beta
to determine why SBPV may be increased and whether blockers and calcium-channel blockers on within-individual variability in
blood pressure and risk of stroke. Lancet Neurol. 2010;9:469–480. doi:
this is a marker of frailty or an independent treatable fac- 10.1016/S1474-4422(10)70066-1
tor. Furthermore, it implies that measures that reduce 4. Webb AJ, Fischer U, Mehta Z, Rothwell PM. Effects of antihypertensive-
SBPV, such as amlodipine,4 may be particularly effective drug class on interindividual variation in blood pressure and risk of stroke:
a systematic review and meta-analysis. Lancet. 2010;375:906–915. doi:
at reducing cardiovascular risk in patients who are over- 10.1016/S0140-6736(10)60235-8
weight or women who are underweight. 5. Webb AJ, Fischer U, Rothwell PM. Effects of β-blocker selectivity on
blood pressure variability and stroke: a systematic review. Neurology.
2011;77:731–737. doi: 10.1212/WNL.0b013e31822b007a
6. Webb AJ, Wilson M, Lovett N, Paul N, Fischer U, Rothwell PM. Response
CONCLUSIONS of day-to-day home blood pressure variability by antihypertensive drug
class after transient ischemic attack or nondisabling stroke. Stroke.
Beat-to-beat BPV reflects increased arterial stiffness 2014;45:2967–2973. doi: 10.1161/STROKEAHA.114.005982
and an increased BMI in both sexes and a low BMI 7. Muntner P, Whittle J, Lynch AI, Colantonio LD, Simpson LM, Einhorn PT,
in women alone. This may be a potentially modifiable Levitan EB, Whelton PK, Cushman WC, Louis GT, et al. Visit-to-visit vari-
ability of blood pressure and coronary heart disease, stroke, heart failure,
mechanism resulting in the increased risk of cardio- and mortality: a cohort study. Ann Intern Med. 2015;163:329–338. doi:
vascular events due to increased arterial stiffness and 10.7326/M14-2803
alterations in BMI. 8. Muntner P, Shimbo D, Tonelli M, Reynolds K, Arnett DK, Oparil S. The
relationship between visit-to-visit variability in systolic blood pres-
sure and all-cause mortality in the general population: findings from
NHANES III, 1988 to 1994. Hypertension. 2011;57:160–166. doi:
ARTICLE INFORMATION 10.1161/HYPERTENSIONAHA.110.162255
Received June 3, 2020; final revision received September 28, 2020; accepted 9. Chiriacò M, Pateras K, Virdis A, Charakida M, Kyriakopoulou D, Nannipieri
November 25, 2020. M, Emdin M, Tsioufis K, Taddei S, Masi S, et al. Association between
blood pressure variability, cardiovascular disease and mortality in type 2
Affiliation diabetes: a systematic review and meta-analysis. Diabetes Obes Metab.
Wolfson Centre for Prevention of Stroke and Dementia, Department of Clinical 2019;21:2587–2598. doi: 10.1111/dom.13828
Neurosciences, University of Oxford, United Kingdom. 10. Gosmanova EO, Mikkelsen MK, Molnar MZ, Lu JL, Yessayan LT,
Kalantar-Zadeh K, Kovesdy CP. Association of systolic blood pressure vari-
Acknowledgments ability with mortality, coronary heart disease, stroke, and renal disease. J Am
This work uses data provided by patients and collected by the National Health Coll Cardiol. 2016;68:1375–1386. doi: 10.1016/j.jacc.2016.06.054
Service as part of their care and support and would not have been possible with- 11. Oishi E, Ohara T, Sakata S, Fukuhara M, Hata J, Yoshida D, Shibata M,
Downloaded from http://ahajournals.org by on March 16, 2021

out access to these data. The National Institute for Health Research recognizes Ohtsubo T, Kitazono T, Kiyohara Y, et al. Day-to-day blood pressure vari-
and values the role of patient data, securely accessed and stored, both in under- ability and risk of dementia in a General Japanese Elderly Population:
pinning and leading to improvements in research and care. A.J.S. Webb devised, the Hisayama Study. Circulation. 2017;136:516–525. doi: 10.1161/
acquired, supervised, and analyzed the physiological assessments; devised, ana- CIRCULATIONAHA.116.025667
lyzed, and supervised the statistical analysis; and drafted, analyzed, edited, and 12. Stevens SL, Wood S, Koshiaris C, Law K, Glasziou P, Stevens RJ, McManus
submitted the manuscript. A. Lawson, Dr Mazzucco, and L. Li acquired and ana- RJ. Blood pressure variability and cardiovascular disease: systematic review
lyzed the physiological assessments. P.M. Rothwell established and supervised and meta-analysis. BMJ. 2016;354:i4098. doi: 10.1136/bmj.i4098
OXVASC (Oxford Vascular Study); devised, initiated, and supervised the physio- 13. Johansson JK, Puukka PJ, Virtanen R, Jula AM. Beat-to-beat, ambu-
logical studies and statistical analyses; and edited and supervised the manuscript. latory hour-to-hour, and home day-to-day variabilities in blood pres-
sure, pulse pressure, and heart rate in comparison with each other and
Sources of Funding with target-organ damage. Blood Press Monit. 2015;20:113–120. doi:
10.1097/MBP.0000000000000101
OXVASC (Oxford Vascular Study) is funded by the National Institute for Health
14. Kikuya M, Ohkubo T, Metoki H, Asayama K, Hara A, Obara T, Inoue R,
Research (NIHR) Oxford Biomedical Research Centre, Wellcome Trust, Wolfson
Hoshi H, Hashimoto J, Totsune K, et al. Day-by-day variability of blood
Foundation, British Heart Foundation, and the European Union Horizon 2020
pressure and heart rate at home as a novel predictor of prognosis: the
programme (grant 666881, SVDs@target). P.M. Rothwell is in receipt of an NIHR
Ohasama study. Hypertension. 2008;52:1045–1050. doi: 10.1161/
Senior Investigator award. A.J.S. Webb and this work is funded by a Wellcome
HYPERTENSIONAHA.107.104620
Trust Clinical Research Development Fellowship (206589/Z/17/Z) and British
15. Webb AJS, Mazzucco S, Li L, Rothwell PM. Prognostic significance of
Heart Foundation Project Grant (PG/16/38/32080). The views expressed are
blood pressure variability on beat-to-beat monitoring after transient
those of the authors and not necessarily those of the National Health Service, the
ischemic attack and stroke. Stroke. 2018;49:62–67. doi: 10.1161/
NIHR, or the Department of Health.
STROKEAHA.117.019107
Disclosures 16. Webb AJ, Rothwell PM. Physiological correlates of beat-to-beat, ambula-
tory, and day-to-day home blood pressure variability after transient isch-
None.
emic attack or minor stroke. Stroke. 2014;45:533–538. doi: 10.1161/
Supplemental Materials STROKEAHA.113.003321
17. Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE,
Figures I–IV
Redgrave JN, Bull LM, Welch SJ, Cuthbertson FC, et al; Oxford Vascular Study.
Tables I–III
Population-based study of event-rate, incidence, case fatality, and mortality
for all acute vascular events in all arterial territories (Oxford Vascular Study).
Lancet. 2005;366:1773–1783. doi: 10.1016/S0140-6736(05)67702-1
REFERENCES 18. Shimbo D, Abdalla M, Falzon L, Townsend RR, Muntner P. Studies compar-
1. Rothwell PM. Limitations of the usual blood-pressure hypothesis and ing ambulatory blood pressure and home blood pressure on cardiovascular
importance of variability, instability, and episodic hypertension. Lancet. disease and mortality outcomes: a systematic review. J Am Soc Hypertens.
2010;375:938–948. doi: 10.1016/S0140-6736(10)60309-1 2016;10:224–234.e17. doi: 10.1016/j.jash.2015.12.013
2. Rothwell PM, Howard SC, Dolan E, O’Brien E, Dobson JE, Dahlöf B, Sever 19. Dawson SL, Manktelow BN, Robinson TG, Panerai RB, Potter JF. Which
PS, Poulter NR. Prognostic significance of visit-to-visit variability, maximum parameters of beat-to-beat blood pressure and variability best predict early

8   April 2021 Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179

 Webb et al BMI, Arterial Stiffness, and Beat-to-Beat BPV

outcome after acute ischemic stroke? Stroke. 2000;31:463–468. doi: 26. Della Corte V, Tuttolomondo A, Pecoraro R, Di Raimondo D, Vassallo V, Pinto

Original Contribution
10.1161/01.str.31.2.463 A. Inflammation, endothelial dysfunction and arterial stiffness as therapeutic
20. van Wijnen VK, Finucane C, Harms MPM, Nolan H, Freeman RL, Westerhof targets in cardiovascular medicine. Curr Pharm Des. 2016;22:4658–4668.
BE, Kenny RA, Ter Maaten JC, Wieling W. Noninvasive beat-to-beat fin- doi: 10.2174/1381612822666160510124801
ger arterial pressure monitoring during orthostasis: a comprehensive 27. Tuttolomondo A, Casuccio A, Della Corte V, Maida C, Pecoraro R,
review of normal and abnormal responses at different ages. J Intern Med. Di Raimondo D, Vassallo V, Simonetta I, Arnao V, Pinto A. Endothelial func-
2017;282:468–483. doi: 10.1111/joim.12636 tion and arterial stiffness indexes in subjects with acute ischemic stroke:
21. Romano SM, Lazzeri C, Chiostri M, Gensini GF, Franchi F. Beat-to-beat relationship with TOAST subtype. Atherosclerosis. 2017;256:94–99. doi:
analysis of pressure wave morphology for pre-symptomatic detection of 10.1016/j.atherosclerosis.2016.10.044
orthostatic intolerance during head-up tilt. J Am Coll Cardiol. 2004;44:1891– 28. Ridker PM, Libby P, MacFadyen JG, Thuren T, Ballantyne C, Fonseca F,
1897. doi: 10.1016/j.jacc.2004.07.046 Koenig W, Shimokawa H, Everett BM, Glynn RJ. Modulation of the inter-
22. Eames PJ, Blake MJ, Dawson SL, Panerai RB, Potter JF. Dynamic cerebral leukin-6 signalling pathway and incidence rates of atherosclerotic events
autoregulation and beat to beat blood pressure control are impaired in acute and all-cause mortality: analyses from the Canakinumab Anti-Inflammatory
ischaemic stroke. J Neurol Neurosurg Psychiatry. 2002;72:467–472. doi: Thrombosis Outcomes Study (CANTOS). Eur Heart J. 2018;39:3499–3507.
10.1136/jnnp.72.4.467 doi: 10.1093/eurheartj/ehy310
23. Wei FF, Li Y, Zhang L, Xu TY, Ding FH, Wang JG, Staessen JA. Beat-to-beat, 29. Rothman AM, MacFadyen J, Thuren T, Webb A, Harrison DG, Guzik
reading-to-reading, and day-to-day blood pressure variability in relation to TJ, Libby P, Glynn RJ, Ridker PM. Effects of interleukin-1β inhibition on
organ damage in untreated Chinese. Hypertension. 2014;63:790–796. doi: blood pressure, incident hypertension, and residual inflammatory risk: a
10.1161/HYPERTENSIONAHA.113.02681 secondary analysis of CANTOS. Hypertension. 2020;75:477–482. doi:
24. Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relation- 10.1161/HYPERTENSIONAHA.119.13642
ship of 24-hour blood pressure mean and variability to severity of tar- 30. Rothwell PM, Giles MF, Chandratheva A, Marquardt L, Geraghty O,
get-organ damage in hypertension. J Hypertens. 1987;5:93–98. doi: Redgrave JN, Lovelock CE, Binney LE, Bull LM, Cuthbertson FC, et
10.1097/00004872-198702000-00013 al; Early Use of Existing Preventive Strategies for Stroke (EXPRESS)
25. Veerman DP, de Blok K, van Montfrans A. Relationship of steady state and Study. Effect of urgent treatment of transient ischaemic attack and minor
ambulatory blood pressure variability to left ventricular mass and urinary stroke on early recurrent stroke (EXPRESS study): a prospective popu-
albumin excretion in essential hypertension. Am J Hypertens. 1996;9:455– lation-based sequential comparison. Lancet. 2007;370:1432–1442. doi:
460. doi: 10.1016/0895-7061(95)00437-8 10.1016/S0140-6736(07)61448-2
Downloaded from http://ahajournals.org by on March 16, 2021

Stroke. 2021;52:00–00. DOI: 10.1161/STROKEAHA.120.031179 April 2021   9