6 / 66
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 32, No 3, May/June 2021
116
AFRICA
Cardiovascular Topics
Analysis of risk factors for thrombosis of the left atrium/
left atrial appendage in patients with non-valvular atrial
fibrillation
He Du, Ke Bi, Lisha Xu, Feng Chen, Wenfeng Xiong, Yin Wang
Abstract
Objectives:
Left atrial appendage (LAA) morphology is a
powerful predictor of thrombogenesis of the left atrium
(LA) in patients with non-valvular atrial fibrillation (NVAF).
However, it remains unknown whether LAA morphology is
useful for stroke risk stratification in patients with NVAF.
Methods:
A total of 555 atrial fibrillation patients were
divided into thrombus and non-thrombus groups accord-
ing to transoesophageal echocardiography. We analysed the
correlation between LAA morphology and the CHADS
2
score. We determined the L
2
CHADS
2
score and compared
the ability to predict LA/LAA thrombosis of the CHADS
2
,
L
2
CHADS
2
and CHA
2
DS
2
-VASc scores from the area under
the curve (AUC).
Results:
The odds ratio of non-chicken wing LAA morphol-
ogy was 11.48. Non-chicken wing LAA morphology was
significantly correlated with LA/LAA thrombosis. We incor-
porated LAA morphology into the CHADS
2
score and
named it the L
2
CHADS
2
score. The AUC of the L
2
CHADS
2
score (0.767) in predicting LA /LAA thrombosis was signifi-
cantly higher than that of the CHADS
2
(0.558) or CHA
2
DS
2
-
VASc scores (0.557). The positive and negative predictive
values of the L
2
CHADS
2
score (13.1 and 98.7%) were higher
than those of the CHADS
2
(8.7 and 94.2%) and CHA
2
DS
2
-
VASc scores (6.9 and 6.9%).
Conclusions:
Non-chicken wing LAA morphology was a
powerful predictor of LA/LAA thrombosis in NVAF patients.
The AUC, sensitivity and specificity of the L
2
CHADS
2
score
were higher than those of the CHADS
2
and CHA
2
DS
2
-VASc
scores.
Keywords:
atrial fibrillation, thrombus, stroke, left atrial append-
age morphology
Submitted 30/6/19, accepted 28/12/19
Published online 16/4/21
Cardiovasc J Afr
2021;
32
: 116–122
www.cvja.co.zaDOI: 10.5830/CVJA-2019-071
Atrial fibrillation (AF) is one of the most common arrhythmias,
which is an independent risk factor for stroke. It is estimated
that AF incidence will double by 2035.
1-3
The most valuable
evaluation systems for stroke are the CHADS
2
and CHA
2
DS
2
-
VASc scores, but the risk factors these scoring systems utilise are
relatively limited and less than half of all known possibilities.
This means that some important risk factors have not been
defined, such as low creatinine clearance rate, increased left atrial
(LA) diameter, non-chicken wing left atrial appendages (LAA)
and persistent AF.
1,3,4
As early as 1909, Welch proposed that cardiac stroke
associated with AF was mainly caused by emboli originating
from a LAA thrombus.
4,5
Previous research has shown that LAA
morphology has a close relationship with thrombosis, with the
type with obvious bending in the main lobe less likely to form
thrombus. In 2010, Wang
et al
. first divided LAA morphology
into four types, namely chicken wing, cauliflower, cactus and
windsock, in order to guide LAA closure.
6
Di Biase
et al
. then
reported that these different LAA morphologies are associated
with stroke or transient ischaemic attack (TIA). They also
pointed out that the chicken wing LAA morphology (which has
an obvious bend in the main lobe) is less likely to form thrombus
compared with other LAA morphologies.
7
These discoveries
have attracted extensive attention.
In 2013, Kimura
et al
. found that cauliflower LAA was the
main predictor of stroke/TIA in non-valvular atrial fibrillation
(NVAF) patients with low CHADS
2
scores, and the results
of multivariable logistic regression analysis suggested that
cauliflower LAA was an independent risk factor for stroke or
TIA (OR: 3.4; 95% CI: 1.243–9.055;
p
= 0.017).
8
These studies
Department of Internal Medicine, Shanghai Pulmonary
Hospital, Tongji University, Shanghai, China
He Du, MD
Department of Oncology, Tongji University School of
Medicine, Shanghai, China
Ke Bi
Department of Ultrasound, Shanghai Pulmonary Hospital,
Tongji University, Shanghai, China
Lisha Xu, MD
Yin Wang, PhD,
916276740@qq.comDepartment of Cardiology, Changhai Hospital, Second
Military Medical University, Shanghai, China
Feng Chen, MD
Shanghai Yueyang Integrated Traditional Chinese Medicine
and Western Medicine Hospital, Shanghai, China
Wenfeng Xiong, MD,
xiongwf2010@163.com