If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding authors at: Department of Medical Imaging, The Second Hospital of Hebei Medical University, 215 West Heping Road, Shijiazhuang 050000, China.
Corresponding authors at: Department of Medical Imaging, The Second Hospital of Hebei Medical University, 215 West Heping Road, Shijiazhuang 050000, China.
Early filling defect in LAA, age and diabetes are independent risk factors for stroke in AF.
•
AF patients with LAA non-thrombotic filling defects on the early-phase of cardiac CT increased ischemic stroke risk.
•
The occurrence of stroke in patients with filling defect may not be related to LAA morphology.
Abstract
Purpose
To investigate the relationship between filling defects in the left atrial appendage restricted to the early phase of cardiac computed tomography (CCT), and ischemic stroke in patients with atrial fibrillation (AF).
Materials and methods
A total of 152 patients with non-valvular AF were retrospectively enrolled and divided into two groups according to the stroke history, as confirmed by brain computed tomography (CT) or magnetic resonance imaging (MRI), as the non-stroke group (n = 89) and stroke group (n = 63), respectively. The numbers of patients with filling defects in the early phase of CCT images without thrombi were recorded. Morphological parameters of the LAA were measured for all participants. All patients with early-phase filling defects (n = 44) were assigned to two groups according to ischemic stroke history: the filling defects with stroke group (n = 28) and the filling defects without stroke group (n = 16). The clinical characteristics and LAA morphological parameters were compared.
Results
Univariate analysis showed that compared with the non-stroke group，LAA volume index and age were higher in the stroke group, and the ratio of early phase filling defect in LAA, hypertension and diabetes were also higher, in the meanwhile the LVEF and BMI were lower (P < 0.05).After adjusting confounding factors by the multivariate logistic regression analysis, filling defect was significantly related with stroke [odds ratio (OR): 4.339, 95% confidence interval (CI): 1.951–9.653, P = 0.000]. LAA morphological parameters were not significantly different between the filling defects with stroke group and the group without stroke.
Conclusion
AF patients with LAA non-thrombotic filling defects in the early-phase of CCT had an increased risk of ischemic stroke compared to those without filling defects. This finding may help to optimize stroke risk stratification in patients with AF.
Atrial fibrillation (AF) is one of the most common arrhythmias occurring worldwide. Stroke is one of the leading causes of functional impairment and increased morbidity and mortality in AF patients.
American Heart Association Statistics CStroke Statistics S Heart disease and stroke statistics-2017 update: a report from the American Heart Association.
The prevention of stroke is the primary clinical task and the focus of ongoing research. Active exploration of potential risk factors for stroke, and effective primary prevention for high-risk groups at an early stage are important measures to reduce the occurrence of stroke events. Thus, assessing and monitoring the risk of thromboembolism in patients is crucial.
Although the CHA2DS2-VASc scoring system is widely used for stroke risk assessment in patients with atrial fibrillation, some patients with a low CHA2DS2-VASc score still suffer from stroke.
Comparative performance of ATRIA, CHADS2, and CHA2DS2-VASc risk scores predicting stroke in patients with atrial fibrillation: results from a National Primary Care Database.
Therefore, it is of great clinical significance to evaluate the impact of other factors beyond the CHA2DS2-VASc scoring system on stroke.
Transient non thrombotic contrast agent filling defects in the LAA during early phase scanning are observed in some patients with AF, and they disappear during delayed phase scanning. The early filling defect represents the state of blood stasis, which may be related to the occurrence of stroke; however, the research on this aspect is still relatively scarce.
At the same time, there is no in-depth study on the relationship between the occurrence of stroke and the morphological changes of the left atrial appendage in patients with early stage filling defects.
This study aimed to investigate the relationship between filling defects in the left atrial appendage restricted to the early phase of cardiac computed tomography (CCT) and ischemic stroke in patients with AF.
2. Materials and methods
2.1 Patients and study design
A total of 152 patients with non-valvular AF between December 2019 and January 2021 were retrospectively enrolled. All patients with AF were clinically diagnosed by physical examination and electrocardiogram (ECG). The patients underwent computed tomography angiography (CTA) examinations before catheter radio-frequency ablation or left atrial appendage occlusion. The baseline demographics of participants were extracted from the electronic medical records. The inclusion criteria were patients without LAA thrombus, and available brain computed tomography (CT) or magnetic resonance imaging (MRI) data. The exclusion criteria were patients with congenital heart disease, rheumatic heart disease, cardiomyopathy, history of cardiac surgery, radio-frequency ablation and LAA closure, pacemaker implantation, moderate to severe mitral stenosis, severe liver and kidney dysfunction, and malignant tumors. The patients were classified into the previous stroke group (n = 63, 35 males, 54 females, age range 50–87 years, mean age 67.25 ± 8.12 years) and non-stroke group (n = 89, 51males, 38 females, age range 32–86 years, mean age 61.64 ± 11.19 years) according to the previous history of ischemic stroke (confirmed by brain CT or MRI).
All patients with early phase filling defects (n = 44) were assigned to two groups according to ischemic stroke history: filling defects with stroke group (n = 28, 15 males, 13 females), mean age: 66.7 ± 8.5, ranged from 50 to 87 years old, and filling defects without stroke group (n = 16, 9 males, 7 females), mean age: 61.0 ± 10.3, ranged from 42 to 77 years old. The clinical characteristics and LAA morphological parameters were compared.
In preparation for radio-frequency ablation or left atrial appendage occlusion, all the participants were treated with an anticoagulant regimen during hospitalization.
The study was approved by the Ethics Committee of the Second Hospital of Hebei Medical University. Written informed consent was obtained from all the patients.
2.2 CCT examination
A Philips 256-slice iCT scanner (Brilliance iCT, Philips Healthcare) was used for CCT examination, with the patient in the supine position. The patients did not receive β-blockers to control the heart rate. The early phase scanning range was from the upper edge of the aortic arch to the diaphragmatic surface of the heart. The non-ionic contrast agent iohexol (350mgI/ml, 0.8 ml/kg) was injected through the antecubital vein at a flow rate of 4-5 ml/s, and then 30 ml of normal saline was injected at the same flow rate. CCT was performed with retrospective ECG gating, and the start of the image acquisition was determined using the bolus-triggering technique. A region of interest (ROI) was placed on the ascending aorta, and the image acquisition was started when the density in the ROI reached 130 Hounsfield units (approximately 10–15 s after contrast injection). The scanning parameters were as follows: tube voltage 120 kV, tube current 280–350 MAS, collimation 128 × 0.625 mm, pitch 0.18, rotation time 330 ms, and matrix 512 × 512.
Delayed scanning covering the LAA area was automatically triggered at 50 s after the injection of the contrast agent. The tube voltage was 100 kV to reduce the scanning dose.
At 45% and 75% of the cardiac phase, the early and delayed scanning images were reconstructed with a layer thickness of 0.9 mm and a section interval of 0.45 mm.
2.3 Measurement of CCT parameters
All images were processed and reviewed using a Philips EBW4.5 workstation. LAA early filling defects were defined as an obviously low-attenuation area, where the LAA cavity showed contrast enhancement less than that of the LA cavity, representing incomplete mixing of contrast medium and blood that appeared only in early phase images (Fig. 1). Multiplanar reconstruction (MPR) was used to obtain the cross-sectional image for measurement of the long diameter, short diameter and area of the LAA orifice, and the distance from the LAA orifice to the first bend of the LAA
(Fig. 2). With the comprehensive cardiac analysis software built into the Philips EBW4.5 workstation, the LAA volume, morphology, depth of the LAA, and the first LAA bend were measured (Fig. 3). LAA morphology was based on the classification by Di Biase et al.
: Cactus, Chicken Wing, Windsock, and Cauliflower (Fig. 4).
Fig. 1Cardiac computed tomography (CCT) of an AF patient showed LAA filling defects on the early phase image (A) and no filling defect in the corresponding position on the delayed phase image(B).
Fig. 2Measurement of the left atrial appendage (LAA) orifice and the distance from LAA orifice to the first bend of the LAA. A-D. The center of the cross positioning line is placed in the transition area between left atrium (LA) and LAA, and the two positioning lines are parallel and vertical to LAA length(A), respectively. On the LAA oblique coronal image, the cross positioning line is adjusted at the atrial wall fold in the oblique coronal image transition area, and the two positioning lines are made parallel and vertical to LAA (B) again to obtain LAA orifice image for measuring the long and short diameter (C) and the orifice area (D), and the distance from LAA orifice to the first bend is measured at the oblique coronal image (B).
Fig. 3The volume, depth, angle of the first bend of LAA are measured by the cardiac function analysis software. A-D. The filling tool is used to fill the LAA on the axial image (A) and coronal image (B) to obtain the 3D image (C) of the LAA. The volume of LAA (C), LAA depth along the LAA direction and the angle of the first bend (D) are measured.
All parameters were analyzed by two experienced radiologists (with 8 and 2 years of experience, respectively) who reached a consensus for each examination.
2.4 Statistical analysis
Statistical analysis was performed using SPSS software (IBM, version 26.0, Chicago, IL, USA). Continuous variables were presented as mean ± standard deviation and tested with an independent t-test if the data followed a normal distribution or as median (interquartile range) and tested with the Mann-Whitney U test if the data had a skewed distribution. Categorical variables are presented as numbers or percentages and compared using the X2-test or Fisher's exact test. Multivariable logistical regression analysis was performed to identify risk factors for stroke. The predictors of stroke identified in the univariate analysis (P < 0.05) were entered as independent variables in multivariate models. P < 0.05 was considered to be statistically significant.
3. Results
3.1 Comparison of Clinical data and CCT variables between the previous stroke group and the non-stroke group in AF patients
Compared with the non-stroke group, patients in the previous stroke group were older, had a lower ejection fraction, and had a lower BMI (P = 0.001, <0.001, and =0.036, respectively). The previous stroke group had a higher proportion of hypertension and diabetes mellitus than the non-stroke group (P = 0.01, =0.017, respectively). There were no significant differences between the two groups in terms of sex, type of AF, time since AF diagnosis, the prevalence of congestive heart failure, dyslipidemia, and coronary artery disease (P > 0.05). Further details concerning the comparison baseline of the characteristics between the groups are shown in Table 1.
Table 1Baseline characteristics of patients in the non-stroke and previous stroke groups.
Variables
Non-stroke (n = 89)
Previous stroke (n = 63)
P value
Age (years)
61.64 ± 11.19
67.25 ± 8.12
0.001
Male
51 (57.30%)
35 (55.56%)
0.830
Time since AF diagnosis (month)
12 (3,60)
24 (3,72)
0.723
Types of atrial fibrillation
0.623
Persistent
36 (40.45%)
28 (44.44%)
Paroxysmal
53 (59.55%)
35 (55.56%)
BMI (kg/cm2)
26.62 ± 3.00
25.49 ± 3.57
0.036
Heart failure
36 (40.45%)
30 (47.62%)
0.380
Hypertension
45 (50.56%)
45 (71.43%)
0.010
Hyperlipidemia
18 (20.22%)
6 (9.52%)
0.075
Coronary heart disease
49 (55.06%)
36 (57.14%)
0.799
Diabetes
16 (18.0%)
22 (34.92%)
0.017
LVEF (%)
64.4 (57.85,68.15)
61.40 (57.30,67.00)
0.000
Note: AF, atrial fibrillation; CCT, cardiac computed tomography; BMI, body mass index; LAA, left atrial appendage. LVEF, left ventricular ejection fraction.
The previous stroke group had a higher proportion of early LAA filling defects and a higher LAA volume index than the non-stroke group (P < 0.001, and P = 0.046, respectively). There were no significant differences between the two groups in the LAA orifice area, LAA orifice long diameter, LAA orifice short diameter, LAA morphology, distance from the LAA ostium to the first LAA bend, angle of the first LAA bend, LAA depth, or LAA volume (Table 2).
Table 2Variables in CCT of AF patients in the non-stroke and previous stroke groups.
Variables
Non stroke (n = 89)
Previous stroke (n = 63)
P value
Early LAA filling defects
16 (17.98%)
28 (44.44%)
0.000
LAA orifice long diameter (mm)
26.08 ± 5.60
25.97 ± 5.66
0.903
LAA orifice short diameter (mm)
17.85 ± 4.51
18.68 ± 4.28
0.259
LAA orifice area (mm2)
347.61 (266.44,445.98)
398.50 (281.24,498.06)
0.131
Distance from LAA orifice to the first bend of the LAA (mm)
5.5 (4.5,6.9)
5.2 (4.2,6.9)
0.400
Angle of the first LAA bend
114 (108,123)
117 (110,127)
0.074
LAA depth (mm)
42.1 (37.15,48.50)
43.50 (38.70,47.90)
0.557
LAA volume (ml)
8.46 ± 3.56
9.44 ± 4.18
0.119
LAA volume index (ml/m2)
4.42 (3.25,5.91)
4.77 (3.71,6.70)
0.046
LAA morphology
0.105
Chicken wing
24 (26.97%)
7 (11.11%)
Cauliflower
36 (40.45%)
34 (53.97%)
Cactus
16 (17.98%)
13 (20.63%)
Windsock
13 (14.61%)
9 (14.29%)
Note: AF, atrial fibrillation; CCT, cardiac computed tomography; BMI, body mass index; LAA, left atrial appendage. LVEF, left ventricular ejection fraction.
3.2 Multivariable logistic regression analysis for potential risk factors of stroke in AF patients
The stroke related factors (P < 0.05) determined in the univariate analysis were included in the multivariate logistic regression model as independent variables (age, BMI, hypertension, diabetes, left ventricular ejection fraction, LAA early filling defect, and LAA volume index). After adjusting for these factors, the LAA early filling defect was still an important risk factor for stroke, and the LAA volume index had no significant correlation with ischemic stroke history (Table 3).
Table 3Multivariable logistic regression analysis for potential risk factors of stroke in AF patients.
Variables
OR
95%CI
P value
Age (years)
1.067
1.024–1.111
0.002
BMI (kg/cm2)
0.857
0.756–0.973
0.051
Hypertension
2.762
1.189–6.417
0.053
Diabetes
2.850
1.251–6.494
0.013
LVEF (%)
0.996
0.955–1.038
0.949
LAA volume index (ml/m2)
1.106
0.910–1.345
0.485
Early LAA filling defects
4.339
1.951–9.653
0.000
Note: BMI, body mass index; LAA, left atrial appendage. LVEF, left ventricular ejection fraction; OR, odds ratio; CI, confidence interval.
Comparison of Clinical data and CCT variables between the early-phase filling defects with stroke group and the early-phase filling defects without stroke group.
The clinical characteristics and CCT variables were not significantly different between the two groups (Tables 4 and 5), respectively.
Table 4Baseline characteristics in patients with early-phase filling defects with or without stroke.
Variables
Early-phase filling defects without stroke (n = 16)
Early-phase filling defects with stroke (n = 28)
P value
Age (years)
61.00 ± 10.25
66.71 ± 8.52
0.054
Male
9 (56.2%)
15 (53.6%)
0.864
Time since AF diagnosis (month)
18 (6,75)
18 (4.5,72)
0.971
Types of atrial fibrillation
0.798
Persistent
12 (75.0%)
20 (71.4%)
Paroxysmal
4 (25.0%)
8 (28.6%)
BMI (kg/m2)
26.71 ± 2.26
25.84 ± 4.13
0.368
Heart failure
10 (62.5%)
19 (67.9%)
0.718
Hypertension
6 (37.5%)
19 (67.9%)
0.051
Hyperlipidemia
4 (25.0%)
2 (7.1%)
0.169
Coronary heart disease
8 (50%)
17 (60.7%)
0.490
Diabetes
3 (18.8%)
7 (25%)
0.724
LVEF (%)
n
60.6 (53.35,65.88)
0.311
Note: AF, atrial fibrillation; BMI, body mass index; LAA, left atrial appendage. LVEF, left ventricular ejection fraction.
In this study, after adjusting for confounding factors, the early filling defect of the LAA was still significantly related to the history of ischemic stroke, which is similar to the study by Inoue.
This finding provides a new reference for the risk stratification of stroke in patients with AF.
Early-phase filling defects in the LAA on CCT images show that blood and contrast agents are mixed unevenly. This was a sign of blood stasis. According to Virchow's triad for thrombogenesis,
which consists of endothelial dysfunction, abnormal blood stasis, and a hypercoagulable state, early-phase filling defects in LAA are associated with thrombogenesis, which increases the risk of stroke. Ouchi
Filling defects in the left atrial appendage restricted to the early phase of cardiac computed tomography as a potential risk of left atrial appendage dysfunction.
reported that persistent AF, decreased LAAFV, increased LAA volume index, and decreased LVEF correlated with early phase fill defects in the LAA, indicating that hemodynamics were related to early phase fill defects. Some of these factors, such as decreased LAAFV
were previously shown to be risk factors for stroke. Thus, the early phase filling defect in the LAA is a vital sign for assessing stroke risk in patients with AF. In a study by Inoue et al.,
the early phase LAA enhancement pattern was defined as a poor, intermediate, or good enhancement pattern, based on CCT images of 147 patients with chronic AF. They observed a poor enhancement pattern related to stroke. Our study was based on a group of participants including patients with chronic AF but also paroxysmal AF patients. In addition, we simplified the enhancement pattern to good or bad (filling well or filling defects), which made it easier for physicians to use.
Whether the morphologic parameters of LAA are related to stroke risk has been discussed for a long time, including the morphology of LAA. In our study, the results showed that there was no significant difference in LAA morphology between the previous stroke group and non-stroke groups in AF patients. This may be because the morphology of LAA is extremely complex and heterogeneous. Di Biase et al.
reported four different LAA morphologies (Cactus, Chicken-Wing, Windsock, and Cauliflower) and found that the non-Chicken-Wing morphology is associated with a 3-fold risk of prior thromboembolic events. And many researches also provided similar results.
Relation of left atrial appendage morphology determined by computed tomography to prior stroke or to increased risk of stroke in patients with atrial fibrillation.
The different results also lie in the roles of the LAA orifice and volume in stroke. Previous studies have shown that increased LAA orifice area, depth, and volume are associated with an increased risk of stroke in AF.
Additional value of left atrial appendage geometry and hemodynamics when considering anticoagulation strategy in patients with atrial fibrillation with low CHA2DS2-VASc scores.
In our study, there was no significant difference in LAA orifice area, long diameter and angle of the first LAA bend between the previous stroke group and the no stroke group, which was similar with the results of Katarzyna et al.
But our study showed that the distance from LAA ostium to the first bend of the LAA did not associate with stroke risk, which was different with them. Further investigation based on a larger group of participants is required. In this study, the volume index of the LAA in the previous stroke group was significantly higher than that in the non-stroke group on univariate analysis. In further multivariate logistic regression analysis, the volume index of LAA was not an independent risk factor for stroke, which may be caused by the correlation between the early filling defects of LAA and the volume of LAA.
Filling defects in the left atrial appendage restricted to the early phase of cardiac computed tomography as a potential risk of left atrial appendage dysfunction.
In addition, the study further divided the patients with LAA filling defects into two groups according to the history of ischemic stroke. It was found that there were no differences in left atrial appendage morphological indexes between the two groups, which indicated that stroke was not related to LAA morphology in patients with left atrial appendage filling defects.
The CHA2DS2-VASc score is widely used to stratify stroke risk in patients with the purpose to initiate anticoagulation therapy. It focuses mostly on well-known risk factors for ischemic stroke, and none of these factors assess the LAA itself, where most of the thrombi in AF are formed. The 2016 ESC guideline suggests that patients with a CHA2DS2-VASc score ≥ 1 would better choose anticoagulation treatment to reduce stroke risk.
These findings suggest that simple stratification using the CHA2DS2-VASc score is insufficient.
The finding that the early filling defect of the LAA is significantly related to the history of ischemic stroke, is very helpful for patients with atrial fibrillation with a CHA2DS2-VASc score ≤ 1, because these patients are usually considered to be a low-risk group for stroke and neglect prevention, but the presence of LAA early filling defect is associated with a higher risk of thromboembolism. According to our research results, it is also necessary for these patients to use anticoagulant drugs promptly to prevent stroke. However, anticoagulant therapy requires comprehensive clinical evaluation of the patient's condition to prevent the risk of bleeding.
There are some limitations to this study: (1) This study is a retrospective study, which cannot guarantee that all patients in the stroke group suffer from cardiogenic stroke, nor can it be sure that the left atrial appendage hemodynamics of patients at the time of stroke; (2) This study was a single-center, small-sample study, which may have selection bias. The subjects included in the study were non valvular atrial fibrillation patients who planned to undergo radio-frequency ablation or LAA occlusion, and most of them were AF patients with poor drug treatment effects. Therefore, the conclusion of this study may not be applicable to all atrial fibrillation populations; (3) The filling of the LAA contrast agent is related to the iodine concentration in the contrast agent, the injection rate and the amount of contrast agent. Therefore, the repeatability of this study may be reduced owing to the influence of the scanning scheme and the equipment performance. Our research results need to be verified in a large sample of prospective studies.
5. Conclusion
AF patients with LAA non-thrombotic filling defects in the early phase of CCT had an increased risk of ischemic stroke compared to those without filling defects. This finding may help optimize stroke risk stratification in patients with AF.
Comparative performance of ATRIA, CHADS2, and CHA2DS2-VASc risk scores predicting stroke in patients with atrial fibrillation: results from a National Primary Care Database.
Filling defects in the left atrial appendage restricted to the early phase of cardiac computed tomography as a potential risk of left atrial appendage dysfunction.
Relation of left atrial appendage morphology determined by computed tomography to prior stroke or to increased risk of stroke in patients with atrial fibrillation.
Additional value of left atrial appendage geometry and hemodynamics when considering anticoagulation strategy in patients with atrial fibrillation with low CHA2DS2-VASc scores.
00057-8&id=gr1.jpg){kind=link}
00057-8&id=gr2.jpg){kind=link}
00057-8&id=gr3.jpg){kind=link}
00057-8&id=gr4.jpg){kind=link}