Heart

Atrial Fibrillation Detection During 24-Hour

Ambulatory Blood Pressure Monitoring
Comparison With 24-Hour Electrocardiography
Anastasios Kollias, Antonios Destounis, Petros Kalogeropoulos, Konstantinos G. Kyriakoulis,
Angeliki Ntineri, George S. Stergiou

Abstract—This study assessed the diagnostic accuracy of a novel 24-hour ambulatory blood pressure (ABP) monitor
(Microlife WatchBP O3 Afib) with implemented algorithm for automated atrial fibrillation (AF) detection during each
ABP measurement. One hundred subjects (mean age 70.6±8.2 [SD] years; men 53%; hypertensives 85%; 17 with
permanent AF; 4 paroxysmal AF; and 79 non-AF) had simultaneous 24-hour ABP monitoring and 24-hour Holter
monitoring. Among a total of 6410 valid ABP readings, 1091 (17%) were taken in ECG AF rhythm. In reading-to-
reading ABP analysis, the sensitivity, specificity, and accuracy of ABP monitoring in detecting AF were 93%, 87%,
and 88%, respectively. In non-AF subjects, 12.8% of the 24-hour ABP readings indicated false-positive AF, of whom
27% were taken during supraventricular premature beats. There was a strong association between the proportion
of false-positive AF readings and that of supraventricular premature beats (r=0.67; P<0.001). Receiver operating
characteristic curve revealed that in paroxysmal AF and non-AF subjects, AF-positive readings at 26% during 24-hour
ABP monitoring had 100%/85% sensitivity/specificity (area under the curve 0.91; P<0.01) for detecting paroxysmal
AF. These findings suggest that in elderly hypertensives, a novel 24-hour ABP monitor with AF detector has high
sensitivity and moderate specificity for AF screening during routine ABP monitoring. Thus, in elderly hypertensives, a
24-hour ABP recording with at least 26% of the readings suggesting AF indicates a high probability for AF diagnosis
and should be regarded as an indication for performing 24-hour Holter monitoring.  (Hypertension. 2018;72:110-115.
DOI: 10.1161/HYPERTENSIONAHA.117.10797.) Online Data Supplement •
Key Words: accuracy ◼ arrhythmia ◼ atrial fibrillation ◼ detection ◼ diagnosis ◼ Holter ◼ screening
Downloaded from http://ahajournals.org by on August 30, 2019

A trial fibrillation (AF) is the most common sustained

arrhythmia and a powerful risk factor for stroke.1 AF can
be paroxysmal and may remain undiagnosed for long time
(silent AF) until a stroke event occurs.1 Therefore, screening
for AF is recommended in the elderly using ECG or pulse
palpation, yet the latter has moderate diagnostic accuracy and
both often miss paroxysmal AF.1–4 As screening at a single time
point may not be an effective strategy for preventing AF-related
stroke, recent guidelines support prolonged monitoring for the
detection of paroxysmal AF in high-risk patients such as those
with stroke.1
The vast majority of AF patients are hypertensives, as both
conditions are common in the elderly and hypertension is a
significant risk factor for AF.5 Thus, opportunistic screening
for AF during routine blood pressure (BP) measurement has
considerable potential for early detection of asymptomatic
AF in the elderly. For this purpose, an automated oscillomet-
ric BP monitor equipped with an algorithm specific for AF
detection during automated BP measurement (Microlife Afib;
Microlife AG, Widnau, Switzerland) has been developed.6–8 A
meta-analysis of 6 studies (n=2332) assessing the diagnostic
accuracy of this algorithm during BP measurement showed
pooled sensitivity 0.98 and specificity 0.92.7 In 2013, the UK
National Institute for Health and Care Excellence recom-
mended the Microlife Afib BP monitor for opportunistic AF
screening during routine office BP measurement in primary
care in the elderly (≥65 years).9
Ambulatory BP monitoring (ABPM) is currently regarded
as the most accurate method for hypertension diagnosis.10–13
Several organizations recommend ABPM to be offered to
confirm diagnosis in most or all subjects with suspected
hypertension.10–14 A novel ABPM device (Microlife WatchBP
O3 Afib; Microlife AG, Widnau, Switzerland) with imple-
mented algorithm for automated AF detection during each
BP measurement has been developed. This study assessed the
diagnostic accuracy of this device for AF detection versus ref-
erence 24-hour ECG during routine 24-hour ABPM in elderly
hypertensives.

Received December 29, 2017; first decision January 12, 2018; revision accepted April 7, 2018.
From the Hypertension Center STRIDE-7, National and Kapodistrian University of Athens, School of Medicine, Third Department of Medicine, Sotiria
Hospital, Athens, Greece.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.
117.10797/-/DC1.
Correspondence to George S. Stergiou, Hypertension Center STRIDE-7, National and Kapodistrian University of Athens, School of Medicine, Third
Department of Medicine, Sotiria Hospital, 152 Mesogion Ave, Athens 11527, Greece. E-mail gstergi@med.uoa.gr
© 2018 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.117.10797

110
 Kollias et al   Blood Pressure Monitoring and Atrial Fibrillation   111

Methods
Data, Analytic Methods (Code), and Research
Materials Transparency
The data that support the findings of this study are available from the
corresponding author upon reasonable request.
Participants
This cross-sectional study enrolled subjects referred to a Hypertension
Clinic for the assessment of their BP, treated or untreated for hyper-
tension. Subjects aged ≥65 years, or aged 50 to 64 years with symp-
toms suggesting arrhythmias or with stroke or AF history, were also
included. All subjects had clinical indication for ABPM, either for
confirming the diagnosis of hypertension in untreated subjects or for
the assessment of BP control in treated ones (exclusion of white-coat
or masked uncontrolled hypertension phenomena). Subjects with
pacemaker implantation were excluded. The study protocol was
approved by the hospital scientific committee, and all participants
provided signed informed consent.
Twenty-Four–Hour ABPM and Holter Monitoring
All participants were subjected to simultaneous 24-hour ABPM
and 24-hour Holter monitoring on a routine workday. ABPM was
performed using the validated oscillometric device Microlife
WatchBP O3 Afib with measurements programmed at 20-minute
intervals for 24 hours.15,16 This device has an implemented algo-
rithm for automated AF detection during each BP measurement.
The presence of AF is depicted in the ABPM report with a symbol
D next to BP readings when AF is detected, and the total number
of BP readings with AF detection is reported (Figure S1 in the
online-only Data Supplement). The performance and quality of
the ABPM recording is not affected in any way by the AF detec-
tion function. The AF detector functions as follows: the device
measures the last 10 pulse intervals during cuff deflation and cal-
Downloaded from http://ahajournals.org by on August 30, 2019
confidence intervals [CI] were used) were determined based on true-
and false-positive and true- and false-negative AF ABPM readings
using the MedCalc statistical software (MedCalc Software bvba,
Ostend, Belgium; https://www.medcalc.org; 2018). For an estimated
prevalence of AF readings around 10%, a minimum sample size of
2310 readings (including 231 with AF) would be required to achieve
a minimum power of 80% to detect a change in the percentage value
of sensitivity from 80% to 90%, based on a target significance level of
0.05.21 This minimum sample size would also be sufficient to detect
a change in the value of specificity from 80% to 90%.21 A sequen-
tial analysis by Bayes theorem for estimating post-test likelihood
of AF detection using the pretest likelihood of AF was performed.22
Mann–Whitney or Kruskal–Wallis test was used for comparisons
among 2 or 3 groups, respectively. χ2 test was applied for compar-
ing percentages among groups. Spearman correlation coefficient (r)
was determined for assessing the relationship between the percent-
age of premature ectopic beats and that of false-positive AF read-
ings. Receiver operating characteristic (ROC) curve analyses were
performed for defining the optimal threshold of the percentage of
24-hour AF ABPM readings required to identify accurately: (1) per-
manent or paroxysmal AF, and (2) any ECG abnormality deserv-
ing referral to cardiologist. The percentage threshold of positive AF
ABPM readings was defined as the cutoff point having the highest
Youden index (sensitivity+specificity−1).23 Results are expressed as
mean values with SD. Statistical analysis was performed using the
IBM SPSS Statistics (Version 21.0. Armonk; NY IBM Corp). A P
value less than 0.05 was considered statistically significant.
Results
Subjects’ Characteristics
A total of 101 subjects were screened and one was excluded
because of implanted pacemaker. Thus, 100 subjects partici-
pated and were finally included in the analysis, the charac-

culates the mean and SD of the time intervals. Each of the 10 teristics of whom are presented in Table 1. According to the
pulse beat intervals that is 25% longer or 25% shorter than the 24-hour Holter monitoring reports, 44 subjects had sinus
mean time interval is discarded, to reduce the effect of premature
beats. The remaining time intervals are used to calculate the irreg- rhythm (without clinically important arrhythmias), 17 par-
ularity index, defined as the SD divided by the mean of the time ticipants had permanent AF, 4 paroxysmal AF (at least 1 AF
intervals.6 If the irregularity index exceeds a threshold value of episode lasting >30 s), and the rest had other arrhythmias
0.06, an AF symbol is ascribed indicating that the patient has AF.6 (Table S1).
Subjects were instructed to perform their usual daily activities but
remain still with their arm extended and relaxed during each BP
measurement. Day and night periods were defined according to the Table 1.   Characteristics of the Study Participants (n=100)
individual patients’ diaries.
The SpiderView (ELA Medical, Sorin Group) multichannel sys- Non-AF Subjects AF Subjects
tem recorder was used for 24-hour Holter monitoring which was Variable (n=79) (n=21)
performed simultaneously with 24-hour ABPM. Time was synchro-
nized in the 2 devices before each application. A cardiologist (P.K. Age, y 70.7±8.8 70.0±5.5
or A.D.) assessed the recordings using the EasyScope Multiday Body mass index, kg/m 2
29.1±4.2 29.2±6.1
ELA Medical software. Artifacts, falsely interpreted as ectopic
beats, were subtracted from the ECG report when calculating the Males (%) 50.6 61.9
number of ectopic beats. Criteria for abnormal 24-hour ECG re- Stroke (%) 11.4 9.5
cording were the following: flutter or AF episode of any duration;
supraventricular or ventricular ectopic beats >720/24 hours; supra- Coronary heart disease (%) 6.3 9.5
ventricular couplets ≥50/24 hours; supraventricular or ventricular
Diabetes mellitus (%) 21.5 14.3
bigeminy ≥50/24 hours; supraventricular runs ≥20/24 hours or ≥10
beats/run; ventricular tachycardia of any duration; sinus pause >3 Hypertension (%) 86.1 81.0
s; and second- or third-degree atrioventricular block. These criteria
were selected to include all clinically important and potentially haz- Antihypertensive treatment 84.8 71.4
ardous arrhythmias, as well as arrhythmias that increase the risk of (%)
AF and stroke.17–20 Ambulatory BP readings 64.8±6.5 61.5±9.0
Ambulatory systolic/diastolic 124.9±10.7/69.8±7.6 123.0±14.4/73.0±8.3
Statistical Analysis
BP, mm Hg
A reading-to-reading analysis of all 24-hour ABPM values obtained
in each patient was performed versus the corresponding simultane- Ambulatory heart rate, bpm 66.3±7.5 77.3±8.0*
ous 24-hour ECG recording (the latter was assessed 1 minute before,
CHA2DS2-VASc score 3.1±1.2 2.9±1.6
during, and 1 minute after each ABPM reading for the presence of AF,
other arrhythmias, or artifacts). The sensitivity, specificity, and accu- AF indicates atrial fibrillation; and BP, blood pressure.
racy of the ABPM device for AF detection (exact Clopper-Pearson *P<0.05 vs non-AF subgroup.
 112  Hypertension  July 2018
AF Diagnostic Accuracy of Individual ABPM
Readings
The number of valid ABPM readings did not differ between non-
AF, permanent AF, and paroxysmal AF subjects (64.8±6.5 ver-
sus 61.2±9.7 versus 62.5±5.4, respectively, P=NS). Among 6410
valid ABPM readings, 1091 (17%) were obtained during ECG AF.
From the total sample of 6410 ABPM readings, the number
of readings with true-positive AF detection was 1013 (15.8%)
and true-negative AF detection was 4609 (71.9%), implying cor-
rect AF detection in 5622 ABPM readings (87.7%). On the other
hand, the number of ABPM readings with false-positive AF
detection was 710 (11.1%) and with false-negative AF detection
was 78 (1.2%), implying incorrect AF detection in 788 ABPM
readings (12.3%). On the basis of these findings, the sensitivity
of individual ABPM readings in detecting AF during 24-hour
monitoring was 93% (95% CI, 91%–94%), the specificity 87%
(95% CI, 86%–88%), and the accuracy 88% (95% CI, 87%-
89%) (Figure 1). A sequential Bayesian analysis examining
the effect of the prevalence of AF on the post-test probability is
shown in Figure S2. Subanalyses on the diagnostic accuracy of
the test device based on age or BP status are shown in Table S2.
In non-AF subjects, 12.8% of the ABPM readings showed
false-positive AF detection. Among these readings, 27% were
performed during supraventricular premature beats, 6% dur-
ing ventricular premature beats, and 14% during physical
activity indicated by artifacts in ECG report. There was strong
association between the frequency (%) of AF false-positive
ABPM readings during 24-hour monitoring and the frequency
of supraventricular premature beats in 24-hour ECG (r=0.67;
Downloaded from http://ahajournals.org by on August 30, 2019
P<0.001; Figure 2). Moreover, there was a borderline asso-
ciation of the CHA2DS2-VASc score (CHADS indicates con-
gestive heart failure, hypertension, age, diabetes, stroke; and
VASc, vascular disease, age, sex) with the frequency of supra-
ventricular premature beats (r=0.22; P=0.05) and that of false-
positive AF readings (r=0.20; P=0.07).
Because a considerable percentage of false-positive
AF ABPM readings was observed during physical activity
Figure 1.  Diagnostic accuracy of individual ambulatory blood
pressure readings in detecting atrial fibrillation.
Figure 2.  Association between the frequency (%) of ambulatory
blood pressure readings with false-positive atrial fibrillation
detection and that of supraventricular ectopic beats in subjects
with paroxysmal atrial fibrillation or sinus rhythm. Small-scale
graph represents scatterplot at low frequency of premature
beats.
(artifacts in monitoring), separate sensitivity analyses were
performed for daytime (awake) and night-time (asleep)
periods. The sensitivity, specificity, and accuracy for day-
time ABPM were 92% (95% CI, 89%–94%), 86% (95% CI,
85%–87%), and 87% (95% CI, 86%–88%), respectively,
and for night-time 95% (95% CI, 92%–97%), 88% (95%
CI, 86%–89%) and 89% (95% CI, 88%–90%; P=NS for
daytime versus night-time comparisons). The percentage
of AF-positive ABPM readings did not differ between day-
time and night-time period (28.3±32.6% versus 27.6±34.9%,
respectively; P=NS).
AF Diagnostic Accuracy of 24-Hour ABPM
Recordings
The average percentage of 24-hour AF-positive ABPM read-
ings per subject in non-AF, paroxysmal AF, and permanent AF
subjects was 13.4%, 41.5%, and 92.8%, respectively (Table 2).
In the total sample, the ROC curve (area under the curve
1.00; 95% CI, 1.00–1.00) revealed a percentage of 79%
AF-positive ABPM readings per subject to have sensitivity
100% and specificity 100% for identifying permanent AF.
In paroxysmal AF and non-AF subjects, the ROC curve
(area under the curve 0.91; 95% CI, 0.83–0.99) revealed a
percentage of 26% AF-positive ABPM readings per subject
to have sensitivity 100% and specificity 85% (highest Youden
index) for identifying paroxysmal AF (AF episodes with dura-
tion >30 s) in 24-hour ECG recording (Figure 3A).
In paroxysmal AF and non-AF subjects, the ROC curve
(area under the curve 0.83; 95% CI, 0.69–0.97) revealed a
percentage of 15% AF-positive ABPM readings per subject to
have sensitivity 90% and specificity 77% for the identification
of AF episodes (of any duration) in 24-hour ECG recording
(Figure 3B).
Among non-AF subjects (n=79), 35 had an abnormal
24-hour ECG recording (Table S1). In these subjects, the ROC
curve (area under the curve 0.72; 95% CI, 0.61–0.84) revealed
a percentage of 11% false-positive AF readings per subject
 Kollias et al   Blood Pressure Monitoring and Atrial Fibrillation   113
Table 2.  Diagnostic Accuracy of a 24-Hour Ambulatory Blood
Pressure Recording for Atrial Fibrillation Detection in Subjects
With Atrial Fibrillation and Those With Sinus Rhythm
Subjects and Readings Non-AF Paroxysmal AF Permanent AF
Subjects (n=100) n=79 n=4
 Average AF-positive ABP 13.4 41.5
readings per subject (%)
ABP readings (n=6410) n=5119 n=250
 True-positive (%) 0 19.2
 True-negative (%) 87.2 57.6
 False-positive (%) 12.8 21.2
 False-negative (%) 0 2
ABP indicates ambulatory blood pressure; and AF, atrial fibrillation.
to have sensitivity 79% and specificity 71% for identifying
any abnormality in 24-hour ECG recording (mainly increased
supraventricular ectopic activity). When the abovementioned
analysis was restricted in non-AF subjects with CHA2DS2-
VASc score ≥4 (n=26), the ROC curve again revealed a per-
centage of 11% false-positive AF readings per subject for
identifying any abnormality in 24-hour ECG recording, but
with sensitivity 79% and specificity 92% (area under the curve
0.83; 95% CI, 0.66–0.99).
On the basis of the abovementioned ROC curve analyses,
a recommendation could be proposed for a threshold value at
which referral to a cardiologist is justified (Table 3).
Downloaded from http://ahajournals.org by on August 30, 2019
Discussion
This study provides the first evidence on the accuracy of a
novel 24-hour ABPM device in detecting AF during routine
use for the evaluation of hypertension in the elderly. The main
findings are that (1) the device has high sensitivity but moder-
ate specificity in detecting AF, (2) a considerable proportion
of individual BP readings with false-positive AF detection is
because of premature beats, especially supraventricular, and
(3) in subjects without permanent AF, a percentage of AF
readings ≥26% has high probability to indicate the presence
of paroxysmal AF.
Strategies aiming at effective screening for AF are
expected to contribute considerably in early AF detection and
management and thereby efficient stroke prevention. These
data suggest that AF detection using a novel ABPM device
is feasible in elderly hypertensives and has high sensitiv-
ity in detecting AF. Because ABPM is now regarded as the
reference tool for confirming hypertension diagnosis and is
increasingly used in clinical practice,10–14 these findings have
n=17
considerable potential for clinical application. Indeed, ABPM
92.8 offers the opportunity of repeated evaluations for AF during
the entire 24-hour period, which might prove particularly use-
n=1041 ful in detecting paroxysmal AF. However, the specificity of the
92.7 ABPM device in detecting AF was moderate with a consider-
able proportion of false-positive AF readings, mainly attrib-
0.3
uted to the presence of excessive supraventricular ectopic
0 activity. Indeed, a significant association was found between
7 the frequency of supraventricular ectopic beats and that of
false-positive AF detection. It should be noted, however, that
excessive supraventricular ectopic activity has been linked not
only to future risk of AF but also to cardiovascular outcome
and mortality.18–20,24–26
The average percentage of AF-positive ABPM readings
was consistently increased across non-AF, paroxysmal AF,
and permanent AF subjects, respectively. A ROC curve analy-
sis in non-AF and paroxysmal AF subjects revealed a percent-
age of 26% AF-positive ABPM readings as optimal threshold
for identifying paroxysmal AF, and a 15% AF-positive ABPM
readings as optimal threshold for identifying any AF episode
(in non-AF subjects this percentage was mainly because of
supraventricular ectopic activity which often coexists with
AF episodes <30 s). Thus, a threshold at 26% could be set
for identifying paroxysmal AF, whereas a lower threshold
(15%) might be suggested for screening high-risk subjects
(high CHA2DS2-VASc score). Moreover, in non-AF sub-
jects, a percentage of AF readings ≥11% was accompanied
by a high probability of an abnormal ECG report, which was
mostly because of excessive supraventricular ectopic activ-
ity. The specificity of the latter analysis was increased when
applied in subjects with high CHA2DS2-VASc score. A previ-
ous study showed excessive supraventricular ectopic activity
and a high CHADS2 score to independently and synergisti-
cally predict the first appearance of AF in patients in sinus
rhythm, indicating a ≈10-fold higher risk.27 Thus, despite the
moderate specificity of the ABPM device for AF detection,
when considering the collateral detection of other significant
arrhythmias—especially excessive supraventricular ectopic
Figure 3.  Receiver operating
characteristic curve in subjects with
sinus rhythm and paroxysmal atrial
fibrillation for the percentage of atrial
fibrillation-positive readings that predict
the presence of: A, Paroxysmal atrial
fibrillation in electrocardiography
(episodes >30s). B, Any episode of atrial
fibrillation in electrocardiography.
 114  Hypertension  July 2018
Table 3.  Interpretation of 24-Hour Ambulatory Blood
Pressure Recordings for Atrial Fibrillation Detection
Percentage of AF-
Positive Readings in
24-Hour ABP Recording Interpretation Recommendation
≥79% Permanent AF very Confirm with ECG
likely
≥26% Paroxysmal AF Confirm with 24-hour
and/or excessive Holter monitoring
supraventricular
ectopic activity very
likely
≥15% Paroxysmal AF Confirm with 24-hour
and/or excessive Holter monitoring in
supraventricular high-risk patients
ectopic activity
possible
ABP indicates ambulatory blood pressure; and AF, atrial fibrillation.
activity—it might prove a useful screening tool during routine
ABPM use for hypertension evaluation in the elderly.
Several factors might affect the diagnostic accuracy of the
test device when applied in clinical practice. The prevalence
of AF in the studied population certainly is an important deter-
mining factor, as suggested by Bayesian analysis for clinical
diagnosis (Figure S2). In this study, subanalyses showed bet-
ter specificity and accuracy in the younger group (aged 50–64
years) compared with those aged ≥65 years, but this might be
related to the lower prevalence of AF and ectopic activity in
Downloaded from http://ahajournals.org by on August 30, 2019
the former group. Another interesting finding in this study is
that routine daily physical activity might, at least partly, affect
the diagnostic accuracy of the ABPM device for AF detec-
tion. Indeed, the values of sensitivity, specificity, and accuracy
tended to be slightly, but nonsignificantly, higher for the night-
time than the daytime period, and a considerable proportion
of false-positive AF readings was observed during activity as
indicated by the presence of artifacts in the ECG recording.
The findings of the present study should be interpreted by
taking into account some limitations. First, this was a cross-
sectional study in a selected sample of elderly subjects, most
of whom had hypertension. However, this is the population
that has indication for both 24-hour ABPM and AF screening.
Second, few cases with paroxysmal AF were included, which
probably is the most attractive target group for applying this
technology. Third, in the absence of guidelines on the definition
of ECG excessive ectopic activity, this was derived from pub-
lished outcome studies.18–20 Fourth, the considerable percent-
age of false-positive results might lead to unjustified referral
for 24-hour Holter monitoring. However, this may not neces-
sarily be a major disadvantage because a large proportion of
false-positive readings are because of excessive supraventricu-
lar ectopic activity, which by itself might be a precursor of AF.
Perspectives
In conclusion, this study provided the first evidence on the
diagnostic accuracy of AF detection during routine 24-hour
ABPM in elderly hypertensives. The diagnostic accuracy
of this novel technology seems to be reasonable with high
sensitivity and moderate specificity. A high frequency of AF
false-positive ABPM readings might be because of exces-
sive supraventricular ectopic activity, which however con-
fers high risk for future AF. Further research is needed to
confirm these findings in other populations with lower AF
prevalence and define the optimal diagnostic approach for
clinical implementation of AF screening during routine
24-hour ABPM in the elderly.
Sources of Funding
Microlife, Widnau, Switzerland provided ambulatory blood pressure
monitors with atrial fibrillation detector for this study, but was not
involved in the study design, analysis, and article preparation.
Disclosures
G.S. Stergiou received consultancy fees and had university research
contracts with Microlife. The other authors report no conflicts.
References
1. Kirchhof P, Benussi S, Kotecha D, et al. 2016 ESC Guidelines for the
management of atrial fibrillation developed in collaboration with EACTS.
Europace. 2016;18:1609–1678. doi: 10.1093/europace/euw295.
2. January CT, Wann LS, Alpert JS, et al; ACC/AHA Task Force Members.
2014 AHA/ACC/HRS guideline for the management of patients with
atrial fibrillation: executive summary: a report of the American College
of Cardiology/American Heart Association Task Force on practice guide-
lines and the Heart Rhythm Society. Circulation. 2014;130:2071–2104.
doi: 10.1161/CIR.0000000000000040.
3. National Institute for Health and Care Excellence 2014. Atrial
Fibrillation: Management. Clinical Guideline [CG180]. www.nice.org.
uk/guidance/cg180. Accessed December 7, 2017.
4. Hobbs FD, Fitzmaurice DA, Mant J, Murray E, Jowett S, Bryan S, Raftery
J, Davies M, Lip G. A randomised controlled trial and cost-effectiveness
study of systematic screening (targeted and total population screening)
versus routine practice for the detection of atrial fibrillation in people aged
65 and over. The SAFE study. Health Technol Assess. 2005;9:iii–iv, ix.
5. Manolis AJ, Rosei EA, Coca A, Cifkova R, Erdine SE, Kjeldsen S, Lip
GY, Narkiewicz K, Parati G, Redon J, Schmieder R, Tsioufis C, Mancia
G. Hypertension and atrial fibrillation: diagnostic approach, preven-
tion and treatment. Position paper of the Working Group ‘Hypertension
Arrhythmias and Thrombosis’ of the European Society of Hypertension. J
Hypertens. 2012;30:239–252. doi: 10.1097/HJH.0b013e32834f03bf.
6. Wiesel J, Fitzig L, Herschman Y, Messineo FC. Detection of atrial fibrilla-
tion using a modified microlife blood pressure monitor. Am J Hypertens.
2009;22:848–852. doi: 10.1038/ajh.2009.98.
7. Verberk WJ, Omboni S, Kollias A, Stergiou GS. Screening for atrial fibril-
lation with automated blood pressure measurement: research evidence
and practice recommendations. Int J Cardiol. 2016;203:465–473. doi:
10.1016/j.ijcard.2015.10.182.
8. Kollias A, Stergiou GS. Automated measurement of office, home and
ambulatory blood pressure in atrial fibrillation. Clin Exp Pharmacol
Physiol. 2014;41:9–15. doi: 10.1111/1440-1681.12103.
9. National Institute for Health and Care Excellence 2013. WatchBP Home
A for Opportunistically Detecting Atrial Fibrillation During Diagnosis and
Monitoring of Hypertension. https://www.nice.org.uk/guidance/MTG13.
Accessed December 7, 2017.
10. Mancia G, Fagard R, Narkiewicz K, et al; Task Force Members.
2013 ESH/ESC Guidelines for the management of arterial hyperten-
sion: the Task Force for the management of arterial hypertension of
the European Society of Hypertension (ESH) and of the European
Society of Cardiology (ESC). J Hypertens. 2013;31:1281–1357. doi:
10.1097/01.hjh.0000431740.32696.cc.
11. National Institute for Health and Care Excellence (NICE). Hypertension
in Adults: Diagnosis and Management. Clinical Guideline [CG127]; 2016.
https://www.nice.org.uk/guidance/CG127. Accessed December 7, 2017.
12. O’Brien E, Parati G, Stergiou G, et al; European Society of Hypertension
Working Group on Blood Pressure Monitoring. European Society of
Hypertension position paper on ambulatory blood pressure monitoring. J
Hypertens. 2013;31:1731–1768. doi: 10.1097/HJH.0b013e328363e964.
13. Piper MA, Evans CV, Burda BU, Margolis KL, O’Connor E, Whitlock EP.
Diagnostic and predictive accuracy of blood pressure screening methods
with consideration of rescreening intervals: a systematic review for the
 Kollias et al   Blood Pressure Monitoring and Atrial Fibrillation   115
U.S. Preventive Services Task Force. Ann Intern Med. 2015;162:192–204.
doi: 10.7326/M14-1539.
14. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/
ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the
prevention, detection, evaluation, and management of high blood pres-
sure in adults: executive summary: a report of the American College
of Cardiology/American Heart Association Task Force on Clinical
Practice Guidelines. Hypertension. 2018;71:1269–1324. doi: 10.1161/
HYP.0000000000000066.
15. Ragazzo F, Saladini F, Palatini P. Validation of the microlife WatchBP
O3 device for clinic, home, and ambulatory blood pressure measurement,
according to the International Protocol. Blood Press Monit. 2010;15:59–
62. doi: 10.1097/MBP.0b013e32833531ca.
16. Stergiou GS, Tzamouranis D, Nasothimiou EG, Karpettas N, Protogerou
A. Are there really differences between home and daytime ambula-
tory blood pressure? Comparison using a novel dual-mode ambula-
tory and home monitor. J Hum Hypertens. 2010;24:207–212. doi:
10.1038/jhh.2009.60.
17. Paudel B, Paudel K. The diagnostic significance of the Holter monitoring
in the evaluation of palpitation. J Clin Diagn Res. 2013;7:480–483. doi:
10.7860/JCDR/2013/4923.2802.
18. Binici Z, Intzilakis T, Nielsen OW, Køber L, Sajadieh A. Excessive
supraventricular ectopic activity and increased risk of atrial
fibrillation and stroke. Circulation. 2010;121:1904–1911. doi:
10.1161/CIRCULATIONAHA.109.874982.
19. Larsen BS, Kumarathurai P, Falkenberg J, Nielsen OW, Sajadieh
A. Excessive atrial ectopy and short atrial runs increase the risk of
stroke beyond incident atrial fibrillation. J Am Coll Cardiol. 2015;66:232–
241. doi: 10.1016/j.jacc.2015.05.018.
Novelty and Significance
What Is New?
Downloaded from http://ahajournals.org by on August 30, 2019
• A novel 24-hour ambulatory blood pressure monitor has been developed
which can detect atrial fibrillation (AF) during each measurement. This
study examined the diagnostic accuracy of this device in detecting AF.
What Is Relevant?
• Ambulatory blood pressure monitoring is currently proposed by several
organizations for confirming the diagnosis of hypertension.
• Effective screening strategies are required in the elderly for early AF de-
tection, aiming to prevent morbidity and mortality.
20. Acharya T, Tringali S, Bhullar M, Nalbandyan M, Ilineni VK, Carbajal E,
Deedwania P. Frequent atrial premature complexes and their association
with risk of atrial fibrillation. Am J Cardiol. 2015;116:1852–1857. doi:
10.1016/j.amjcard.2015.09.025.
21. Bujang MA, Adnan TH. Requirements for minimum sample size for sen-
sitivity and specificity analysis. J Clin Diagn Res. 2016;10:YE01–YE06.
doi: 10.7860/JCDR/2016/18129.8744.
22. van Stralen KJ, Stel VS, Reitsma JB, Dekker FW, Zoccali C, Jager KJ.
Diagnostic methods I: sensitivity, specificity, and other measures of accu-
racy. Kidney Int. 2009;75:1257–1263. doi: 10.1038/ki.2009.92.
23. Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3:32–35.
24. Chong BH, Pong V, Lam KF, Liu S, Zuo ML, Lau YF, Lau CP, Tse HF, Siu
CW. Frequent premature atrial complexes predict new occurrence of atrial
fibrillation and adverse cardiovascular events. Europace. 2012;14:942–
947. doi: 10.1093/europace/eur389.
25. Murakoshi N, Xu D, Sairenchi T, Igarashi M, Irie F, Tomizawa T, Tada
H, Sekiguchi Y, Yamagishi K, Iso H, Yamaguchi I, Ota H, Aonuma K.
Prognostic impact of supraventricular premature complexes in commu-
nity-based health checkups: the Ibaraki Prefectural Health Study. Eur
Heart J. 2015;36:170–178. doi: 10.1093/eurheartj/ehu407.
26. Lin CY, Lin YJ, Chen YY, Chang SL, Lo LW, Chao TF, Chung FP, Hu
YF, Chong E, Cheng HM, Tuan TC, Liao JN, Chiou CW, Huang JL, Chen
SA. Prognostic significance of premature atrial complexes burden in pre-
diction of long-term outcome. J Am Heart Assoc. 2015;4:e002192. doi:
10.1161/JAHA.115.002192.
27. Suzuki S, Sagara K, Otsuka T, et al. Usefulness of frequent supraven-
tricular extrasystoles and a high CHADS2 score to predict first-time
appearance of atrial fibrillation. Am J Cardiol. 2013;111:1602–1607. doi:
10.1016/j.amjcard.2013.01.335.
Summary
The test device seemed to have high sensitivity and moderate
specificity for AF detection during routine 24-hour ambulatory
blood pressure monitoring in the elderly. In elderly hypertensives,
routine 24-hour ambulatory blood pressure monitoring with AF de-
tected in ≥26% of the readings indicates a high probability for AF
diagnosis and should be regarded as an indication for performing
24-hour Holter monitoring.