Cardiovascular Journal of Africa: Vol 32 No 2 (MARCH/APRIL 2021)

50 / 60

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 32, No 2, March/April 2021 104 AFRICA
than five to six minutes were associated with an increased risk
of developing clinical AF.18 Similar results were reported later by
the ASSERT study, where 16% of patients with AHRE longer
than six minutes developed clinical AF. These findings suggest
a pathogenic connection between these two arrhythmic entities.49
In CIED patients, AHRE was associated with a two- to
2.5-fold increase in stroke risk when compared to patients
without AHRE.13,49,50 However, the risk of stroke was smaller
than in patients with clinical AF.14 Kazuo Miyazawa et al.
also showed that patients with AHRE had higher mortality
rates when compared to patients without AHRE.50 However,
the West Birmingham Atrial Fibrillation project showed that
AHRE did not increase the thromboembolic risk, while the
baseline CHA2DS2-VASc score was independently associated
with thromboembolic events.25
Several burden thresholds have been investigated, ranging
from five minutes to 24 hours, to establish a minimum threshold
that increases stroke risk. A minimum burden of five minutes
has been proven to increase the risk of stroke.17,18 Pastori et al.
showed that AHRE ≥ five minutes were associated with a 1.7-fold
increase in major adverse cardiovascular events (MACE) while
episodes of more than 24 hours showed a 2.3-fold increase in
MACE.51 However, a re-analysis of the ASSERT study found
that only episodes longer than 24 hours were associated with an
increased risk of stroke.52
In the SOS AF project, a dichotomised analysis was performed
in order to investigate the risk of stroke when comparing
different cut-off thresholds (five minutes, one, six, 12 and 23
hours). The one-hour threshold was associated with a hazard
ratio of 2.11.53 Data from the RATE registry, a multicentre,
prospective, observational study, which investigated outcomes
of over 5 000 patients with device-detected AF, showed that
short episodes of AF terminating within a single adjudicated
EGM recording (up to 20 seconds) did not increase the risk of a
composite outcome of stroke, TIA, hospitalisation or mortality.
However, approximately 50% of these patients did go on to
develop longer episodes of AF over two years of follow up.22
While the association between AHRE and stroke or systemic
thromboembolism has been proven, the minimum duration
of an AHRE that increases the thromboembolic risk remains
uncertain. Despite the contradictory data, a minimum threshold
of five to six minutes is widely considered to increase the risk of
thromboembolic events (Table 1).
Despite the temporal relationship between AF and stroke,
proven first by the Framingham study and later by the AFFIRM
study, there seems to be a temporal discordance between AHRE
and stroke.15,16,54,55 AHRE was diagnosed in only half of the
patients with stroke or systemic embolism in an analysis of the
TRENDS study. 15 Moreover, 73% of the patients did not present
with an AHRE in the preceding month. A similar analysis
performed on the ASSERT study identified an AHRE in 51% of
the patients with stroke.16 The cause for this temporal discordance
between AHRE and stroke is not completely understood.
AHRE and anticoagulation
Initiation of anticoagulant therapy is difficult in these patients.
The threshold of AHRE duration leading to an elevated
stroke risk is one of the major knowledge gaps in the EHRA
consensus for device-detected subclinical AT.12 Selecting the
most efficient antithrombotic therapy for patients with AHRE
was one of the gaps in evidence identified by the European
Society of Cardiology (ESC) taskforce for the 2016 guidelines
on the management of AF.56 There are however ongoing trials
that compare oral anticoagulation with aspirin in patients with
AHRE9,10 (Table 2).
There are no randomised controlled trials published to date
to guide anticoagulant therapy in patients with AHRE. All
Table 1. Relationship between CIED-detected AHRE and systemic embolism
Trial
Number of
patients Follow-up duration AHRE duration cut-off
Atrial cut-off
rate (bpm)
Hazard ratio for TE
event (p-value)
Ancillary MOST 18 312 27 months (median) > 5 min > 220 6.7 (0.020)
Italian AT500 registry59 725 22 months (median) > 24 h > 174 3.1 (0.044)
Botto et al. 60 568 1 year (mean) CHADS2 and AF burden ≥ 5 min in a day or > 24 h > 174 NA
TRENDS13 2486 1.4 years (mean) ≥ 5.5 h > 175 2.2 (0.060)
Home Monitor CRT61 560 370 days (median) ≥ 3.8 h > 180 9.4 (0.006)
ASSERT49 2580 2.5 years (mean) ≥ 6 min > 190 2.5 (0.007)
SOS AF53 10016 2 years (median) ≥ 1 h > 175 2.11 (0.008)
RATE REGISTRY22 5379 22.9 months (median) NA NA 0.87 (0.51)
Table 2. Comparison of ARTESIA and NOAH trials
Study Identifier Inclusion criteria
Number
of patients Design Endpoint
Current
status
Estimated
completion date
ARTESIA9 Clinicaltrials.gov
NCT01938248
Patients without clinical AF
Pacemaker, ICD or CRT
Age ≥ 65 years
CHA2DS2-VASc score ≥ 4
≥ 1 episode of symptomatic AF ≥ 6 min,
atrial rate > 175 bpm
no single episode > 24 h in duration
4000 Randomised, double-
blind, double-dummy
Randomised to
Apixaban 2 × 5 mg or
2 × 2.5 mg vs aspirin 1
× 81 mg
Composite of stroke
and SSE
Major bleeding
Recruiting 2022
NOAH
AFNET 610
Clinicaltrials.gov
NCT02618577
Only patients without overt AF
Pacemaker or ICD
Age ≥ 65 years
CHA2DS2-VASc ≥ 2
≥ 1 episode of AHRE ≥ 6 min, atrial
rate > 180 bpm, no single episode > 24 h
2686 Randomised, double-
blind double-dummy
Edoxaban 1 × 60 mg
or 1 × 30 mg vs 1 ×
aspirin 100 mg or
placebo
Composite of time to
first stroke, SSE or CV
death
Recruiting 2022
AF, atrial fibrillation; ICD, implantable cardioverter defibrillator; CRT, cardiac resynchronisation therapy; SSE, systemic embolism.