Association Between Left Atrial Appendage Morphology Evaluated by Trans-Esophageal Echocardiography and Ischemic Cerebral Stroke in Patients With Atrial Fibrillation Sherif A. Sakr,1 MD, Wagdia A. El-Rasheedy,2 MD, Mahmoud M. Ramadan,1 MD, Ibrahim El-Menshawy,3 MD, Essam Mahfouz,1 MD, and Mohamed Bayoumi,1 MD Summary The left atrial appendage (LAA) represents one of the major sources of cardiac thrombi responsible for embolic stroke in patients with atrial fibrillation (AF). The aim of the present study was to evaluate LAA structure and functions by transesophageal echocardiography (TEE) in patients with AF to investigate the possible association between the different LAA morphologies and the patients’ history of ischemic cerebral stroke. We included 50 patients with non-valvular AF (29 chronic, 21 paroxysmal), 24 patients (13 men) without stroke; and 26 patients (9 men) with a history of ischemic stroke. All patients underwent TEE evaluation of LAA morphology and functions. Compared to patients without stroke, patients with ischemic stroke had significantly higher CHADS2 scores (4.19 ± 0.89 versus 1.67 ± 1.13; P < 0.001) and C-reactive protein levels (8.3 ± 1.6 versus 7.6 ± 0.83 mg/L; P = 0.023), and lower peak filling (21.7 ± 11.3 versus 31.2 ± 9.5 cm/second; P = 0.033) and emptying (22.2 ± 9.7 versus 33.4 ± 13.4 cm/second, P = 0.030) velocities. Triangular LAA morphology had a higher prevalence in patients with stroke (36% versus 12% in non-stroke group); and in half of them an LAA thrombus was present. LAA thrombi were detected in 9 patients (18%) with stroke and in 5 patients (10%) without stroke. On multivariate logistic regression analysis, age (OR = 1.202 [1.042–1.585]; P = 0.041), LAA orifice diameter (OR = 1.275 [1.102–1.748]; P = 0.028), and triangular LAA morphology (OR = 4.53 [1.629– 8.381]; P = 0.011) were significantly and independently associated with ischemic stroke in AF patients. LAA morphology evaluated by TEE may be useful for predicting ischemic cerebral stroke in patients with non-valvular AF, especially in those with a low CHADS2 score. (Int Heart J 2015; 56: 329-334) Key words: Transesophageal, Thrombus, C-reactive protein, CHADS2 score, Brain
A
trial fibrillation (AF) is the most common arrhythmia encountered in clinical practice and its frequency increases with age. Approximately 33% of arrhythmiarelated hospitalizations are for AF, which is also associated with a 5-fold increase in the risk of stroke and 2-fold increase in the risk of all cause mortality.1) AF causes 15-20% of ische mic strokes, and the annual risk of stroke in patients with nonvalvular AF is as high as 5%.2) The left atrial appendage (LAA) has been described as a long, narrow, tubular, wavy, hooked structure with different lobes and a narrow junction and crenellated lumen.3,4) As the LAA represents one of the major sources of cardiac thrombus formation responsible for stroke in patients with AF,5,6) the imaging of the different structures and lobes is of utmost importance to diagnose the presence of thrombus, especially in patients with non-valvular AF.7,8) Multidetector computed tomography 9) and contrast-enhanced magnetic resonance imaging 10) are able to image the LAA; however, it was concluded that magnetic resonance imaging lacks diagnostic accuracy for the detection of LAA
thrombi.10) Transesophageal echocardiography (TEE) is the gold standard to detect LAA thrombus,11) but false positive results still occur.12) In this study, we evaluated the LAA structure and functions by TEE in patients with non-valvular AF (both chronic and paroxysmal types) to study the possible association between the different LAA morphologies and functional changes with the patients’ history of ischemic cerebral stroke.
Methods Study design and participants: This study was conducted on
50 patients with non-valvular AF; including both chronic (29 patients, 58%) and paroxysmal (21 patients, 42%) cases. We included patients with documented paroxysmal AF because large, prospective studies have shown that the risk for thromboembolism does not differ between patients with paroxysmal AF and those with chronic AF.13,14) The patients included in this study were divided into 2 groups: the first group of 26 patients (9 men) were aged 67.5 ±
From the 1 Department of Cardiology, Faculty of Medicine, Mansoura University, 2 Department of Cardiology, Mansoura Students Hospital, and 3 Department of Neurology, Faculty of Medicine, Mansoura University, Mansoura City, Egypt. Address for correspondence: Mahmoud M. Ramadan, MD, Department of Cardiology, Specialized Medicine Hospital, Mansoura University, Gomhoreya Street, Mansoura, Dakahleya 35111, Egypt. E-mail: [email protected] Received for publication November 20, 2014. Revised and accepted December 29, 2014. Released in advance online on J-STAGE April 27, 2015. All rights reserved by the International Heart Journal Association. 329
330
Int Heart J May 2015
SAKR, ET AL
9.1 years with non-valvular AF and a positive documented history of ischemic stroke, and the second group of 24 patients (13 men) were aged 59.8 ± 10.0 years with non-valvular AF and no history of any type of stroke or transient ischemic attacks (TIA). The study was conducted from January 2012 until April 2014 in the Cardiology Department of Mansoura Specialized Medicine Hospital (Mansoura University, Mansoura City, Egypt). Patients with valvular heart disease and dilated cardiomyopathy were excluded. All individuals studied were subjected to thorough history taking and full clinical evaluation. Documented history of stroke was proved by brain computed tomography. A 12-lead ECG was used to document AF. The study was approved by the Institutional Review Board and informed consent was obtained from all patients prior to the start of the study. Transthoracic echocardiography: Standard transthoracic echocardiography (TTE) included two-dimensional, M-mode, and Doppler flow assessments. Left ventricular (LV) internal systolic and diastolic dimensions, posterior wall and interventricular septal thickness, as well as ejection fraction (EF) and fractional shortening were all measured. All Doppler measurements are presented as the average value of 3 consecutive cardiac cycles. Transesophageal echocardiography: All patients underwent TEE (6-MHz multiplane transducer, Sonos 7500) to determine LAA morphology and functions. They were studied in a 4-hour fasting condition, and pharyngeal anesthesia was performed with topical lidocaine 10% spray. Sedation with midazolam was needed in some instances. The LAA was visualized in the mid-esophageal view by rotating the imaging sector from 0° to 180°, and LAA morphology was obtained with the cursor at 120° with counterclockwise rotation at this point, where a transverse section in LAA was obtained to properly assess its morphology. In the longitudinal view (approximately at 90°), maximal LAA area and diameter were measured by the planimetry method. In patients with AF, the maximal and minimal LAA areas were usually not related to the QRS complex. In these patients, 5 to 8 frames were chosen from each cardiac cycle using slow-motion video. The observers performed planimetry of the LAA area for each frame on the monitor screen using the integrated software of the echocardiography machine and printed out the frames with the maximal and minimal areas for further analysis by tracing a line starting from the top of the limbus of the left upper pulmonary vein along the whole LAA endocardial border. The LAA-EF was calculated as the maximal area minus the minimal area divided by the maximal area. LAA blood flow velocities were obtained with pulsed Doppler by positioning the sample volume at the proximal third of the LAA cavity after necessary gain and filter adjustments. The pulsed Doppler sample volume was adjusted into the outlet of the LAA cavity (about 1 cm away from the LA cavity). During the assessment of LAA flow, 2 blood flow velocities were evaluated. Biphasic waves following both the lateral and medial LAA V1 were termed V1 (LAA contraction [emptying] flow velocity) and V2 (filling flow velocity), respectively. The peak flow velocities of V1 and V2 waves were measured. Two independent observers analyzed the peak filling and peak emptying velocities, and values from 5 consecutive cardiac cycles were averaged. Laboratory assessment: Fasting venous blood samples (5 mL
each) were obtained from all patients, and their sera were stored at –70°C until later assessment for high-sensitivity C-reactive protein (hsCRP) levels. Quantitative determination of the serum hsCRP level was performed using an immuno-turbidimetric method (Unimate 3 CP, Roche Milan, Italy). The normal upper reference value for hsCRP with this method is 5 mg/ L. Other routine laboratory measurements (such as blood glucose, renal and liver functions) were performed using an automated auto-analyzer (Cobas Integra, Roche, Germany). Serum lipids were measured by standard enzymatic methods (but lowdensity lipoprotein was calculated by the Friedewald equation). Full blood count was analyzed with an automated cell counter (Sysmex, Roche, Germany). Statistical analysis: All statistical analyses were performed by SPSS for Windows version 22 (SPSS Inc., Chicago, IL). All continuous data are presented as the mean ± SD and were compared using Student’s t-test and one-way analysis of variance (ANOVA) test for two-level and multiple level grouping variables, respectively. Categorical variables are described as count and percent and were compared using Pearson’s chisquare test (or Fisher’s exact test whenever needed). A multivariable logistic regression analysis model was used to identify the significant independent correlates of stroke. All potential confounders were put into the model on the basis of known clinical relevance or statistically significant association observed in univariate analyses. The study population was stratified into 3 subgroups based on the LAA morphology (triangular, rounded, and others), and a sub-analysis was performed to investigate the possible association of LAA morphologic types with ischemic stroke. The odds ratio (OR) and 95% confidence interval (CI) of OR for stroke were computed. All tests were 2-sided, and a P value < 0.05 was considered statistically significant.
Results TEE was performed for 50 patients with non-valvular AF, aged 63.7 ± 9.6 years, 44% males, mean body mass index (BMI) of 30.6 ± 9.1 kg/m2, 52% with ischemic stroke, LVEF of 53.5 ± 5.0%, 78% with CHADS2 score ≥ 2, 68% hypertensive, 24% diabetic, 26% current/ex-smokers, 50% on oral anticoagulant (warfarin) therapy, 30% on aspirin, 28% on statins, 24% NYHA class 3, 22% NYHA class 2, and 16% NYHA class 1. Table I shows the baseline characteristics of AF patients with and without stroke. As shown in the table, patients with stroke were significantly older but with nearly equal BMI and without significant predilection for either gender or the type of AF (chronic versus paroxysmal). The percentage distribution of diabetes, hypertension, smoking, the use of oral anticoagulants, aspirin and statins did not differ significantly between the 2 patient groups; but the CHADS2 score was, as expected, significantly higher in the stroke group. All laboratory variables (blood count, renal and liver functions, serum lipids, and fasting glucose level) were not statistically different between the 2 groups (not shown in Table I); apart from a significantly higher mean hsCRP level in patients with stroke compared to those without stroke (Table I). Patients with stroke had a wider LAA orifice, longer LAA length, and a larger LAA area compared to patients without stroke; but these differences were statisti-
Vol 56 No 3
331
LAA MORPHOLOGY BY TEE AND STROKE IN AF PATIENTS Table I. Baseline Characteristics Classified According to Event (Stroke) in Patients With Atrial Fibrillation Variables Demographic variables Age (years) Gender Male Female Clinical & laboratory variables AF type Chronic Paroxysmal Diabetes mellitus Hypertension Current or ex-smoking Body mass index (kg/m2) Oral anticoagulant use Aspirin use Statin use CHADS2 score Ejection fraction (%) hsCRP (mg/L) TEE variables LAA orifice diameter (mm) LAA length to apex (mm) Systolic minimum area (cm2) Diastolic maximum area (cm2) Peak filling velocity (cm/second) Peak emptying velocity (cm/second) LAA morphology Triangular Rounded Others Oval Tubular Rectangular Globular Slit-like LAA thrombus In triangular LAA In rounded LAA In other morphologies Oval Tubular Rectangular
Without stroke (n = 24)
With stroke (n = 26)
P
59.8 ± 10.0
67.5 ± 9.0
0.006** 0.077
13 (26%) 11 (22%)
9 (18%) 17 (34%)
11 (22%) 13 (26%) 8 (16%) 15 (30%) 5 (10%) 30.2 ± 9.3 11 (22%) 6 (12%) 8 (16%) 1.67 ± 1.13 55.0 ± 4.9 7.6 ± 0.83
18 (36%) 8 (16%) 4 (8%) 19 (38%) 8 (16%) 31.0 ± 8.8 14 (28%) 9 (18%) 6 (12%) 4.19 ± 0.89 52.0 ± 5.2 8.3 ± 1.6
19.4 ± 5.8 32.1 ± 10.5 3.0 ± 1.2 4.5 ± 1.3 31.2 ± 9.5 33.4 ± 13.4
21.7 ± 3.9 34.8 ± 10.1 3.7 ± 1.4 5.17 ± 1.6 21.7 ± 11.3 22.2 ± 9.7
6 (12%) 11 (22%) 7 (14%) 4 (8%) 0 (0%) 3 (6%) 0 (0%) 0 (0%) 5 (10%) 2 (4%) 1 (2%) 2 (4%) 1 (2%) 1 (2%) 0 (0%)
18 (36%) 1 (2%) 7 (14%) 2 (4%) 3 (6%) 0 (0%) 1 (2%) 1 (2%) 9 (18%) 7 (14%) 0 (0%) 2 (4%) 0 (0%) 1 (2%) 1 (0%)
0.094 0.138 0.423 0.777 0.696 0.843 0.645 0.808 < 0.001** 0.147 0.023* 0.092 0.354 0.280 0.289 0.033* 0.030* 0.001**
0.278
Statistical significance at *P < 0.05 and **P < 0.001. Percentages were computed per the total number of patients (50). AF indicates atrial fibrillation; LAA, left atrial appendage; hsCRP, high-sensitive C-reactive protein; and TEE, transesophageal echocardiography.
cally insignificant. Peak filling and emptying velocities were significantly depressed in patients with stroke compared with the non-stroke group. Among all morphologic varieties, triangular LAA had the highest prevalence in all patients (n = 24, 48%) as well as in those with stroke (36% versus 12% in nonstroke patients); and in half of these triangular forms, thrombi were detected. On the other hand, the most prevalent LAA morphology in patients without stroke was the rounded type (n = 11, 22%; compared to n = 1, 2% in the stroke group). The “other” morphologies included oval (n = 6, 12%), tubular (n = 3, 6%), rectangular (n = 3, 6%), globular (n = 1, 2%), and slitlike (n = 1, 2%) types. LAA thrombi were detected in 9 patients (18%) with stroke and in 5 (10%) non-stroke patients (Table I); and prevailed mainly in the triangular (n = 9, 18%) but rarely (n = 1, 2%) in the rounded LAA morphologic types. Only 4 LAA thr ombi were detected in the remaining LAA morphologic types;
two for tubular, one for oval, and one for rectangular varieties; while globular and slit-like types were free from thrombi. Table II shows the comparative analysis of LAA morphologies in relation to various variables. As shown in the table, the prevalence of ischemic stroke was significantly linked to the triangular LAA type (36%), and was least associated with the rounded type (2%). On variance analysis, there was a statistically significant depression in peak filling and emptying velocities in the triangular LAA morphology compared with the rounded type and other LAA morphologies. After performing univariate logistic regression analysis for each independent variable against the stroke-positive outcome, the following variables were associated with the occurrence of ischemic stroke at P ≤ 0.1 (the entry threshold), and were thus included in the final stepwise logistic regression model: age, gender, AF type (chronic or paroxysmal), LAA orifice diameter, LAA peak filling velocity, LAA peak emptying
332
Int Heart J May 2015
SAKR, ET AL Table II. Comparative Analysis of the Main LAA Morphologies in Relation to Different Variables Variable Demographic variables Age (years) Gender Male Female Clinical and laboratory variables AF type Chronic AF Paroxysmal AF Ischemic stroke Diabetes mellitus Hypertension Smoking (current or ex-smoking) Body mass index (kg/m2) Oral anticoagulant use Aspirin use Statin use CHADS2 score Ejection fraction (%) hsCRP (mg/L) TEE variables LAA orifice diameter (mm) LAA length to apex (mm) Systolic minimum area (cm2) Diastolic maximum area (cm2) Peak filling velocity (cm/second) Peak emptying velocity (cm/second) LAA thrombus
Triangular (n = 24)
Rounded (n = 12)
Others (n = 14)
P
66.3 ± 10.8
56.5 ± 9.5
65.5 ± 8.9
0.021* 0.303
11 (22%) 13 (26%)
7 (14%) 5 (10%)
4 (8%) 10 (20%)
15 (30%) 9 (18%) 18 (36%) 6 (12%) 17 (34%) 4 (8%) 31.1 ± 8.5 12 (24%) 5 (10%) 4 (8%) 3.8 ± 1.47 53.8 ± 4.0 8.3 ± 1.12
4 (8%) 8 (16%) 1 (2%) 2 (4%) 7 (14%) 5 (10%) 32.4 ± 9.3 6 (12%) 7 (14%) 6 (12%) 1.3 ± 0.55 56.3 ± 6.7 6.9 ± 1.06
10 (20%) 4 (8%) 7 (14%) 4 (8%) 10 (20%) 4 (8%) 29.5 ± 9.0 7 (14%) 3 (6%) 5 (10%) 3.0 ± 1.10 52.1 ± 4.9 7.3 ± 0.85
0.002** 0.768 0.712 0.901 0.655 0.454 0.717 0.914 0.678 0.068 0.211
20.7 ± 4.8 33.7 ± 11.3 3.0 ± 1.6 4.6 ± 1.9 22.5 ± 11.6 24.8 ± 11.8 9 (18%)
17.8 ± 6.5 32.6 ± 10.2 3.3 ± 1.6 4.8 ± 1.5 36.5 ± 19.5 41.8 ± 19.5 1 (2%)
21.2 ± 4.2 33.7 ± 9.0 3.9 ± 1.2 5.4 ± 1.4 23.9 ± 16.6 37.0 ± 14.7 4 (8%)
0.087 0.944 0.448 0.561 0.040* 0.006** 0.122*
0.120
Abbreviations as in Table I. The percentages shown were computed per the total number of patients (50). Means ± SD across the 3 groups were compared with one-way analysis of variance (ANOVA) test. Table III. Results of the Final Multivariate Logistic Regression Analysis Model Used to Test the Independent Association of the Input Variables Against the Stroke-Positive Outcome
Age (years) LAA orifice diameter (mm) LAA morphology Other morphologies (used as indicator) Triangular Rounded
Beta-coefficient
Odds ratio
95% confidence interval for odds ratio
P
0.183 0.242
1.202 1.275
1.042 – 1.585 1.102 – 1.748
0.041* 0.028*
1.629 – 8.381 0.470 – 1.225
0.033* 0.011* 0.124
1.504 -1.112
4.530 0.327
Abbreviations as in Table I.
velocity, and LAA morphology (triangular, rounded, and “other” morphologies). It was found that age, LAA orifice diameter, and triangular LAA morphology were the variables significantly and independently associated with ischemic stroke in AF patients (Table III). The final logistic regression model was highly significant (P < 0.001), with good calibration (HosmerLemeshow Goodness-of-fit χ2 = 3.69, P = 0.884) and high discriminative power (C-statistic = 0.86 [0.81–0.93]; P < 0.001, Nagelkerke R2 = 0.713). The model’s correlation matrix was checked to avoid collinearity, which was excluded on the basis of reasonable correlations (generally, r < ±0.5).
Discussion The LAA is an important reservoir for the development of atrial thrombi.7) Recently, it was reported that thromboem-
bolic events could occur even in patients with sinus rhythm as a result of depressed LAA function.15) In this study, we evaluated LAA structure and functions by TEE in patients with nonvalvular paroxysmal and chronic AF, about half of whom had prior ischemic cerebral stroke. Because the presence of thrombus was more common in patients with decreased LAA flow velocities,16) the evaluation of LAA function with TEE in patients with AF or in sinus rhythm may enable the detection of patients prone to thromboembolism. The early detection of this tendency for LAA thrombus formation may be crucial to prevent the embolic consequences.17) Age and LAA morphology were reported as independently related to silent cerebral ischemia burden in AF patients referred for catheter ablation.17) This relationship was first investigated in a multicenter study that reported a protective odds ratio for symptomatic stroke in AF patients with chicken-wing
Vol 56 No 3
LAA MORPHOLOGY BY TEE AND STROKE IN AF PATIENTS
LAA morphology compared with other morphologies.18) This may be explained by the more complex internal anatomical structure (such as that of the windsock and cauliflower LAA morphologies) which more intensively promote local blood stasis and thrombogenesis compared to simpler structures (such as the chicken-wing morphology).17) To date, there are no sufficient data correlating the various LAA morphologies obtained by TEE with the thromboembolic risk of stroke in patients with AF, and our present study is the first to address this topic. We found that the triangular LAA morphology is significantly and independently associated with ischemic stroke when compared with the other LAA morphologies. These findings could be clinically relevant, especially for patients judged to be at low-risk of thromboembolic events, such as those with CHADS2 scores of 0 and 1. In these patients, the presence of a triangular LAA morphology increases the odds for stroke, suggesting the need of a more aggressive antithrombotic therapy. Furthermore, this study may explain the occurrence of stroke in patients with low risk of thromboembolism (CHADS2 score of zero). LAA structure and functions: The LAA is an embryologic remnant that functions during conditions of fluid overload as a blood reservoir.7) The LAA is prone to stasis because of its hooked morphology; for this reason it represents the prevalent site of thrombus formation in patients with AF.7) With the use of TEE, it was reported that up to 98% of atrial thrombi occurring during AF derive from the LAA.5,19) Several studies have showed that among patients with AF, subgroups with low LAA flow velocities have a greater risk of thromboembolic events than those with high flow velocities.20,21) It is now known that AF is coupled with a marked reduction in LAA flow velocity.21,22) Previous studies on patients with AF have shown that subgroups with LAA flow velocity < 20 cm/s,20) < 25 cm/s,21,23) and < 35 cm/s 24) are at greater embolic risk than other patients. Interestingly, patients with LAA flow velocity < 37.0 cm/s and a larger (> 3.5 cm2) LAA orifice had a high incidence of stroke that reached 75%. Therefore, this subpopulation is in need of strict anticoagulation to prevent stroke. These data partially agree with our results which showed that emptying and filling velocities obtained by pulsedwave Doppler were significantly depressed in patients with ischemic stroke compared with non-stroke patients in univariate analyses. However, this association was lost after multivariate adjustment using logistic regression analysis; but the same model confirmed the independent positive association between the width of LAA orifice and the occurrence of stroke. One recent study documented that the LAA length was a determinant of thromboembolism, with a shorter length conferring a greater risk.25) In our study, the increase in LAA length was greater in AF patients with stroke than in the nonstroke group patients; but the difference was statistically insignificant (Table I). Anticoagulation management: The CHADS2 score was introduced into guidelines and implemented into clinical practice to assess individual thromboembolic risk in patients with AF. In patients with a CHADS2 score > 1, the need for oral anticoagulation is not questionable; but in patients with low-intermediate risk for stroke (CHADS2 score = 1), no consensus exists on whether patients should receive oral anticoagulation or antiplatelet therapy.26) Thus, the implementation of LAA morphology may aid the clinical decision towards oral anticoagu-
333
lant or antiplatelet therapy in this subpopulation of patients. In patients with contraindication(s) to oral anticoagulants or because of physician and/or patient preference, it is possible to use antiplatelet therapy (but with conflicting results).27) In this scenario, the identification of LAA morphology associated with a lower risk for stroke may further guide clinicians in the decision process. The findings of the present study suggest that the LAA morphology may be taken into account when planning the anticoagulation management of patients with AF; with the triangular-type LAA mandating an intensified anticoagulation strategy compared to the more benign rounded variety in which antiplatelets only could be sufficient. CRP in the context of AF and stroke: CRP, an acute-phase reactant, is an indicator of underlying systemic inflammation and a novel plasma marker for atherothrombotic disease. The recent use of highly sensitive CRP assays has enhanced the usefulness of CRP as a reliable predictor of cardiovascular events. It has been documented that plasma CRP level was elevated in patients with AF, and plasma CRP was an independent predictor for AF.28) Moreover, elevated plasma CRP levels significantly predicted the risk of future ischemic stroke and TIA in elderly patients independent of other cardiovascular risk factors.29) Therefore, we measured hsCRP to identify any possible association between hsCRP and ischemic stroke in this subset of patients with AF; as any positive independent association could be practically implemented for optimizing anticoagulation therapy. In the current study, the mean plasma hsCRP level was significantly increased in AF patients with ischemic stroke compared to the stroke-negative group (Table I). In this regard, it was shown that the AF patients with ischemic stroke had significantly higher plasma hsCRP levels than the patients with simple AF, and hsCRP was an independent risk factor for ischemic stroke in patients with AF after adjusting for the other cardiovascular risk factors.30) However, this association between hsCRP and ischemic stroke found in univariate analysis in AF patients was lost after multivariate adjustment using the logistic regression model in our study. A similar finding was reported by Potpara, et al 31) where the CHA2DS2-VASc score reliably predicts a short-term, 30-day unfavorable outcome of acute ischemic stroke regardless of the presence or absence of AF, and that adding high-sensitivity TnI or fibrinogen or CRP, or all three biomarkers, to the CHA2DS2-VASc score does not further improve the prediction. Study limitations: First, the small number of patients included in this study is a major limitation. However, our study population was quite homogeneous as only patients with non-valvular AF were included. Secondly, the relatively small number of patients with AF and LAA thrombi may make the statistical comparisons between patients with and without LAA thrombi quite difficult. Thirdly, we were unable to retrieve the anticoagulation status at the time of event especially in the subgroup of patients at high risk of stroke (ie, CHADS2 scores ≥ 2). Finally, it was impossible to determine the duration of AF in the majority of patients; so that this potentially important variable was not included in the analyses. Conclusions: LAA morphology assessed by TEE may be useful for predicting ischemic cerebral stroke in patients with nonvalvular AF, especially in those with a low CHADS2 score.
334
SAKR, ET AL
Disclosure Conflicts of interest: No potential conflicts of interest exist.
Acknowledgment The authors thank Dr. Mohammed Osama (consultant of cardiology, National Heart Institute, Imbaba, Egypt) for his sincere help and advice regarding this study.
16. 17. 18.
19.
References
1. Go AS, Mozaffarian D, Roger VL, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Executive summary: heart disease and stroke statistics--2014 update: a report from the American Heart Association. Circulation 2014; 129: 399-410. 2. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation 2006; 114: 119-25. 3. Sharma S, Devine W, Anderson RH, Zuberbuhler JR. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. Br Heart J 1988; 60: 227-31. 4. Tuccillo B, Stümper O, Hess J, et al. Transoesophageal echocardiographic evaluation of atrial morphology in children with congenital heart disease. Eur Heart J 1992; 13: 223-31. 5. Leung DY, Black IW, Cranney GB, Hopkins AP, Walsh WF. Prognostic implications of left atrial spontaneous echo contrast in nonvalvular atrial fibrillation. J Am Coll Cardiol 1994; 24: 755-62. 6. Wang Y, Di Biase L, Horton RP, Nguyen T, Morhanty P, Natale A. Left atrial appendage studied by computed tomography to help planning for appendage closure device placement. J Cardiovasc Electrophysiol 2010; 21: 973-82. 7. Al-Saady NM, Obel OA, Camm AJ. Left atrial appendage: structure, function, and role in thromboembolism. Heart 1999; 82: 54754. (Review) 8. Martinez MW, Kirsch J, Williamson EE, et al. Utility of nongated multidetector computed tomography for detection of left atrial thrombus in patients undergoing catheter ablation of atrial fibrillation. JACC Cardiovasc Imaging 2009; 2: 69-76. 9. Patel A, Au E, Donegan K, et al. Multidetector row computed tomography for identification of left atrial appendage filling defects in patients undergoing pulmonary vein isolation for treatment of atrial fibrillation: comparison with transesophageal echocardiography. Heart Rhythm 2008; 5: 253-60. 10. Mohrs OK, Nowak B, Petersen SE, et al. Thrombus detection in the left atrial appendage using contrast-enhanced MRI: a pilot study. AJR Am J Roentgenol 2006; 186: 198-205. 11. Manning WJ, Weintraub RM, Waksmonski CA, et al. Accuracy of transesophageal echocardiography for identifying left atrial thrombi. A prospective, intraoperative study. Ann Intern Med 1995; 123: 817-22. 12. Stöllberger C, Ernst G, Bonner E, Finsterer J, Slany J. Left atrial appendage morphology: comparison of transesophageal images and postmortem casts. Z Kardiol 2003; 92: 303-8. 13. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994; 154: 1449-57. 14. Predictors of thromboembolism in atrial fibrillation: I. Clinical features of patients at risk. The stroke prevention in atrial fibrillation investigators. Ann Intern Med 1992; 116: 1-5. 15. Manning WJ, Silverman DI, Waksmonski CA, Oettgen P, Douglas PS. Prevalence of residual left atrial thrombi among patients with
20.
21.
22.
23.
24.
25. 26.
27.
28.
29. 30.
31.
Int Heart J May 2015
acute thromboembolism and newly recognized atrial fibrillation. Arch Intern Med 1995; 155: 2193-8. Kamp O, Verhorst PM, Welling RC, Visser CA. Importance of left atrial appendage flow as a predictor of thromboembolic events in patients with atrial fibrillation. Eur Heart J 1999; 20: 979-85. Anselmino M, Scaglione M, Di Biase L, et al. Left atrial appendage morphology and silent cerebral ischemia in patients with atrial fibrillation. Heart Rhythm 2014; 11: 2-7. Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol 2012; 60: 531-8. Uslu N, Nurkalem Z, Orhan AL, et al. Transthoracic echocardiographic predictors of the left atrial appendage contraction velocity in stroke patients with sinus rhythm. Tohoku J Exp Med 2006; 208: 291-8. Zabalgoitia M, Halperin JL, Pearce LA, Blackshear JL, Asinger RW, Hart RG. Transesophageal echocardiographic correlates of clinical risk of thromboembolism in nonvalvular atrial fibrillation. Stroke Prevention in Atrial Fibrillation III Investigators. J Am Coll Cardiol 1998; 31: 1622-6. Mügge A, Kühn H, Nikutta P, Grote J, Lopez JA, Daniel WG. Assessment of left atrial appendage function by biplane transesophageal echocardiography in patients with nonrheumatic atrial fibrillation: identification of a subgroup of patients at increased embolic risk. J Am Coll Cardiol 1994; 23: 599-607. Rubin DN, Katz SE, Riley MF, Douglas PS, Manning WJ. Evaluation of left atrial appendage anatomy and function in recent-onset atrial fibrillation by transesophageal echocardiography. Am J Cardiol 1996; 78: 774-8. Panagiotopoulos K, Toumanidis S, Vemmos K, Saridakis N, Stamatelopoulos S. Secondary prognosis after cardioembolic stroke of atrial origin: the role of left atrial and left atrial appendage dysfunction. Clin Cardiol 2003; 26: 269-74. Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994; 23: 961-9. Khurram IM, Dewire J, Mager M, et al. Relationship between left atrial appendage morphology and stroke in patients with atrial fibrillation. Heart Rhythm 2013; 10: 1843-9. Wann LS, Curtis AB, Ellenbogen KA, et al. 2011 ACCF/AHA/ HRS focused update on the management of patients with atrial fibrillation (update on dabigatran): a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol 2011; 57: 1330-7. European Heart Rhythm Association; European Association for Cardio-Thoracic Surgery, Camm AJ, Kirchhof P, Lip GY, et al; ESC Committee for Practice Guidelines. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010; 12: 1360-420. Chung MK, Martin DO, Sprecher D, et al. C-reactive protein elevation in patients with atrial arrhythmias: inflammatory mechanisms and persistence of atrial fibrillation. Circulation 2001; 104: 2886-91. Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke 2001; 32: 2575-9. You L, Wang P, Lv J, Cianflone K, Wang D, Zhao C. The role of high-sensitivity C-reactive protein, interleukin-6 and cystatin C in ischemic stroke complicating atrial fibrillation. J Huazhong Univ Sci Technolog Med Sci 2010; 30: 648-51. Potpara TS, Polovina MM, Djikic D, Marinkovic JM, Kocev N, Lip GY. The association of CHA2DS2-VASc score and blood biomarkers with ischemic stroke outcomes: the Belgrade stroke study. PLoS One 2014; 9: e106439.
A
trial fibrillation (AF) is the most common arrhythmia encountered in clinical practice and its frequency increases with age. Approximately 33% of arrhythmiarelated hospitalizations are for AF, which is also associated with a 5-fold increase in the risk of stroke and 2-fold increase in the risk of all cause mortality.1) AF causes 15-20% of ische mic strokes, and the annual risk of stroke in patients with nonvalvular AF is as high as 5%.2) The left atrial appendage (LAA) has been described as a long, narrow, tubular, wavy, hooked structure with different lobes and a narrow junction and crenellated lumen.3,4) As the LAA represents one of the major sources of cardiac thrombus formation responsible for stroke in patients with AF,5,6) the imaging of the different structures and lobes is of utmost importance to diagnose the presence of thrombus, especially in patients with non-valvular AF.7,8) Multidetector computed tomography 9) and contrast-enhanced magnetic resonance imaging 10) are able to image the LAA; however, it was concluded that magnetic resonance imaging lacks diagnostic accuracy for the detection of LAA
thrombi.10) Transesophageal echocardiography (TEE) is the gold standard to detect LAA thrombus,11) but false positive results still occur.12) In this study, we evaluated the LAA structure and functions by TEE in patients with non-valvular AF (both chronic and paroxysmal types) to study the possible association between the different LAA morphologies and functional changes with the patients’ history of ischemic cerebral stroke.
Methods Study design and participants: This study was conducted on
50 patients with non-valvular AF; including both chronic (29 patients, 58%) and paroxysmal (21 patients, 42%) cases. We included patients with documented paroxysmal AF because large, prospective studies have shown that the risk for thromboembolism does not differ between patients with paroxysmal AF and those with chronic AF.13,14) The patients included in this study were divided into 2 groups: the first group of 26 patients (9 men) were aged 67.5 ±
From the 1 Department of Cardiology, Faculty of Medicine, Mansoura University, 2 Department of Cardiology, Mansoura Students Hospital, and 3 Department of Neurology, Faculty of Medicine, Mansoura University, Mansoura City, Egypt. Address for correspondence: Mahmoud M. Ramadan, MD, Department of Cardiology, Specialized Medicine Hospital, Mansoura University, Gomhoreya Street, Mansoura, Dakahleya 35111, Egypt. E-mail: [email protected] Received for publication November 20, 2014. Revised and accepted December 29, 2014. Released in advance online on J-STAGE April 27, 2015. All rights reserved by the International Heart Journal Association. 329
330
Int Heart J May 2015
SAKR, ET AL
9.1 years with non-valvular AF and a positive documented history of ischemic stroke, and the second group of 24 patients (13 men) were aged 59.8 ± 10.0 years with non-valvular AF and no history of any type of stroke or transient ischemic attacks (TIA). The study was conducted from January 2012 until April 2014 in the Cardiology Department of Mansoura Specialized Medicine Hospital (Mansoura University, Mansoura City, Egypt). Patients with valvular heart disease and dilated cardiomyopathy were excluded. All individuals studied were subjected to thorough history taking and full clinical evaluation. Documented history of stroke was proved by brain computed tomography. A 12-lead ECG was used to document AF. The study was approved by the Institutional Review Board and informed consent was obtained from all patients prior to the start of the study. Transthoracic echocardiography: Standard transthoracic echocardiography (TTE) included two-dimensional, M-mode, and Doppler flow assessments. Left ventricular (LV) internal systolic and diastolic dimensions, posterior wall and interventricular septal thickness, as well as ejection fraction (EF) and fractional shortening were all measured. All Doppler measurements are presented as the average value of 3 consecutive cardiac cycles. Transesophageal echocardiography: All patients underwent TEE (6-MHz multiplane transducer, Sonos 7500) to determine LAA morphology and functions. They were studied in a 4-hour fasting condition, and pharyngeal anesthesia was performed with topical lidocaine 10% spray. Sedation with midazolam was needed in some instances. The LAA was visualized in the mid-esophageal view by rotating the imaging sector from 0° to 180°, and LAA morphology was obtained with the cursor at 120° with counterclockwise rotation at this point, where a transverse section in LAA was obtained to properly assess its morphology. In the longitudinal view (approximately at 90°), maximal LAA area and diameter were measured by the planimetry method. In patients with AF, the maximal and minimal LAA areas were usually not related to the QRS complex. In these patients, 5 to 8 frames were chosen from each cardiac cycle using slow-motion video. The observers performed planimetry of the LAA area for each frame on the monitor screen using the integrated software of the echocardiography machine and printed out the frames with the maximal and minimal areas for further analysis by tracing a line starting from the top of the limbus of the left upper pulmonary vein along the whole LAA endocardial border. The LAA-EF was calculated as the maximal area minus the minimal area divided by the maximal area. LAA blood flow velocities were obtained with pulsed Doppler by positioning the sample volume at the proximal third of the LAA cavity after necessary gain and filter adjustments. The pulsed Doppler sample volume was adjusted into the outlet of the LAA cavity (about 1 cm away from the LA cavity). During the assessment of LAA flow, 2 blood flow velocities were evaluated. Biphasic waves following both the lateral and medial LAA V1 were termed V1 (LAA contraction [emptying] flow velocity) and V2 (filling flow velocity), respectively. The peak flow velocities of V1 and V2 waves were measured. Two independent observers analyzed the peak filling and peak emptying velocities, and values from 5 consecutive cardiac cycles were averaged. Laboratory assessment: Fasting venous blood samples (5 mL
each) were obtained from all patients, and their sera were stored at –70°C until later assessment for high-sensitivity C-reactive protein (hsCRP) levels. Quantitative determination of the serum hsCRP level was performed using an immuno-turbidimetric method (Unimate 3 CP, Roche Milan, Italy). The normal upper reference value for hsCRP with this method is 5 mg/ L. Other routine laboratory measurements (such as blood glucose, renal and liver functions) were performed using an automated auto-analyzer (Cobas Integra, Roche, Germany). Serum lipids were measured by standard enzymatic methods (but lowdensity lipoprotein was calculated by the Friedewald equation). Full blood count was analyzed with an automated cell counter (Sysmex, Roche, Germany). Statistical analysis: All statistical analyses were performed by SPSS for Windows version 22 (SPSS Inc., Chicago, IL). All continuous data are presented as the mean ± SD and were compared using Student’s t-test and one-way analysis of variance (ANOVA) test for two-level and multiple level grouping variables, respectively. Categorical variables are described as count and percent and were compared using Pearson’s chisquare test (or Fisher’s exact test whenever needed). A multivariable logistic regression analysis model was used to identify the significant independent correlates of stroke. All potential confounders were put into the model on the basis of known clinical relevance or statistically significant association observed in univariate analyses. The study population was stratified into 3 subgroups based on the LAA morphology (triangular, rounded, and others), and a sub-analysis was performed to investigate the possible association of LAA morphologic types with ischemic stroke. The odds ratio (OR) and 95% confidence interval (CI) of OR for stroke were computed. All tests were 2-sided, and a P value < 0.05 was considered statistically significant.
Results TEE was performed for 50 patients with non-valvular AF, aged 63.7 ± 9.6 years, 44% males, mean body mass index (BMI) of 30.6 ± 9.1 kg/m2, 52% with ischemic stroke, LVEF of 53.5 ± 5.0%, 78% with CHADS2 score ≥ 2, 68% hypertensive, 24% diabetic, 26% current/ex-smokers, 50% on oral anticoagulant (warfarin) therapy, 30% on aspirin, 28% on statins, 24% NYHA class 3, 22% NYHA class 2, and 16% NYHA class 1. Table I shows the baseline characteristics of AF patients with and without stroke. As shown in the table, patients with stroke were significantly older but with nearly equal BMI and without significant predilection for either gender or the type of AF (chronic versus paroxysmal). The percentage distribution of diabetes, hypertension, smoking, the use of oral anticoagulants, aspirin and statins did not differ significantly between the 2 patient groups; but the CHADS2 score was, as expected, significantly higher in the stroke group. All laboratory variables (blood count, renal and liver functions, serum lipids, and fasting glucose level) were not statistically different between the 2 groups (not shown in Table I); apart from a significantly higher mean hsCRP level in patients with stroke compared to those without stroke (Table I). Patients with stroke had a wider LAA orifice, longer LAA length, and a larger LAA area compared to patients without stroke; but these differences were statisti-
Vol 56 No 3
331
LAA MORPHOLOGY BY TEE AND STROKE IN AF PATIENTS Table I. Baseline Characteristics Classified According to Event (Stroke) in Patients With Atrial Fibrillation Variables Demographic variables Age (years) Gender Male Female Clinical & laboratory variables AF type Chronic Paroxysmal Diabetes mellitus Hypertension Current or ex-smoking Body mass index (kg/m2) Oral anticoagulant use Aspirin use Statin use CHADS2 score Ejection fraction (%) hsCRP (mg/L) TEE variables LAA orifice diameter (mm) LAA length to apex (mm) Systolic minimum area (cm2) Diastolic maximum area (cm2) Peak filling velocity (cm/second) Peak emptying velocity (cm/second) LAA morphology Triangular Rounded Others Oval Tubular Rectangular Globular Slit-like LAA thrombus In triangular LAA In rounded LAA In other morphologies Oval Tubular Rectangular
Without stroke (n = 24)
With stroke (n = 26)
P
59.8 ± 10.0
67.5 ± 9.0
0.006** 0.077
13 (26%) 11 (22%)
9 (18%) 17 (34%)
11 (22%) 13 (26%) 8 (16%) 15 (30%) 5 (10%) 30.2 ± 9.3 11 (22%) 6 (12%) 8 (16%) 1.67 ± 1.13 55.0 ± 4.9 7.6 ± 0.83
18 (36%) 8 (16%) 4 (8%) 19 (38%) 8 (16%) 31.0 ± 8.8 14 (28%) 9 (18%) 6 (12%) 4.19 ± 0.89 52.0 ± 5.2 8.3 ± 1.6
19.4 ± 5.8 32.1 ± 10.5 3.0 ± 1.2 4.5 ± 1.3 31.2 ± 9.5 33.4 ± 13.4
21.7 ± 3.9 34.8 ± 10.1 3.7 ± 1.4 5.17 ± 1.6 21.7 ± 11.3 22.2 ± 9.7
6 (12%) 11 (22%) 7 (14%) 4 (8%) 0 (0%) 3 (6%) 0 (0%) 0 (0%) 5 (10%) 2 (4%) 1 (2%) 2 (4%) 1 (2%) 1 (2%) 0 (0%)
18 (36%) 1 (2%) 7 (14%) 2 (4%) 3 (6%) 0 (0%) 1 (2%) 1 (2%) 9 (18%) 7 (14%) 0 (0%) 2 (4%) 0 (0%) 1 (2%) 1 (0%)
0.094 0.138 0.423 0.777 0.696 0.843 0.645 0.808 < 0.001** 0.147 0.023* 0.092 0.354 0.280 0.289 0.033* 0.030* 0.001**
0.278
Statistical significance at *P < 0.05 and **P < 0.001. Percentages were computed per the total number of patients (50). AF indicates atrial fibrillation; LAA, left atrial appendage; hsCRP, high-sensitive C-reactive protein; and TEE, transesophageal echocardiography.
cally insignificant. Peak filling and emptying velocities were significantly depressed in patients with stroke compared with the non-stroke group. Among all morphologic varieties, triangular LAA had the highest prevalence in all patients (n = 24, 48%) as well as in those with stroke (36% versus 12% in nonstroke patients); and in half of these triangular forms, thrombi were detected. On the other hand, the most prevalent LAA morphology in patients without stroke was the rounded type (n = 11, 22%; compared to n = 1, 2% in the stroke group). The “other” morphologies included oval (n = 6, 12%), tubular (n = 3, 6%), rectangular (n = 3, 6%), globular (n = 1, 2%), and slitlike (n = 1, 2%) types. LAA thrombi were detected in 9 patients (18%) with stroke and in 5 (10%) non-stroke patients (Table I); and prevailed mainly in the triangular (n = 9, 18%) but rarely (n = 1, 2%) in the rounded LAA morphologic types. Only 4 LAA thr ombi were detected in the remaining LAA morphologic types;
two for tubular, one for oval, and one for rectangular varieties; while globular and slit-like types were free from thrombi. Table II shows the comparative analysis of LAA morphologies in relation to various variables. As shown in the table, the prevalence of ischemic stroke was significantly linked to the triangular LAA type (36%), and was least associated with the rounded type (2%). On variance analysis, there was a statistically significant depression in peak filling and emptying velocities in the triangular LAA morphology compared with the rounded type and other LAA morphologies. After performing univariate logistic regression analysis for each independent variable against the stroke-positive outcome, the following variables were associated with the occurrence of ischemic stroke at P ≤ 0.1 (the entry threshold), and were thus included in the final stepwise logistic regression model: age, gender, AF type (chronic or paroxysmal), LAA orifice diameter, LAA peak filling velocity, LAA peak emptying
332
Int Heart J May 2015
SAKR, ET AL Table II. Comparative Analysis of the Main LAA Morphologies in Relation to Different Variables Variable Demographic variables Age (years) Gender Male Female Clinical and laboratory variables AF type Chronic AF Paroxysmal AF Ischemic stroke Diabetes mellitus Hypertension Smoking (current or ex-smoking) Body mass index (kg/m2) Oral anticoagulant use Aspirin use Statin use CHADS2 score Ejection fraction (%) hsCRP (mg/L) TEE variables LAA orifice diameter (mm) LAA length to apex (mm) Systolic minimum area (cm2) Diastolic maximum area (cm2) Peak filling velocity (cm/second) Peak emptying velocity (cm/second) LAA thrombus
Triangular (n = 24)
Rounded (n = 12)
Others (n = 14)
P
66.3 ± 10.8
56.5 ± 9.5
65.5 ± 8.9
0.021* 0.303
11 (22%) 13 (26%)
7 (14%) 5 (10%)
4 (8%) 10 (20%)
15 (30%) 9 (18%) 18 (36%) 6 (12%) 17 (34%) 4 (8%) 31.1 ± 8.5 12 (24%) 5 (10%) 4 (8%) 3.8 ± 1.47 53.8 ± 4.0 8.3 ± 1.12
4 (8%) 8 (16%) 1 (2%) 2 (4%) 7 (14%) 5 (10%) 32.4 ± 9.3 6 (12%) 7 (14%) 6 (12%) 1.3 ± 0.55 56.3 ± 6.7 6.9 ± 1.06
10 (20%) 4 (8%) 7 (14%) 4 (8%) 10 (20%) 4 (8%) 29.5 ± 9.0 7 (14%) 3 (6%) 5 (10%) 3.0 ± 1.10 52.1 ± 4.9 7.3 ± 0.85
0.002** 0.768 0.712 0.901 0.655 0.454 0.717 0.914 0.678 0.068 0.211
20.7 ± 4.8 33.7 ± 11.3 3.0 ± 1.6 4.6 ± 1.9 22.5 ± 11.6 24.8 ± 11.8 9 (18%)
17.8 ± 6.5 32.6 ± 10.2 3.3 ± 1.6 4.8 ± 1.5 36.5 ± 19.5 41.8 ± 19.5 1 (2%)
21.2 ± 4.2 33.7 ± 9.0 3.9 ± 1.2 5.4 ± 1.4 23.9 ± 16.6 37.0 ± 14.7 4 (8%)
0.087 0.944 0.448 0.561 0.040* 0.006** 0.122*
0.120
Abbreviations as in Table I. The percentages shown were computed per the total number of patients (50). Means ± SD across the 3 groups were compared with one-way analysis of variance (ANOVA) test. Table III. Results of the Final Multivariate Logistic Regression Analysis Model Used to Test the Independent Association of the Input Variables Against the Stroke-Positive Outcome
Age (years) LAA orifice diameter (mm) LAA morphology Other morphologies (used as indicator) Triangular Rounded
Beta-coefficient
Odds ratio
95% confidence interval for odds ratio
P
0.183 0.242
1.202 1.275
1.042 – 1.585 1.102 – 1.748
0.041* 0.028*
1.629 – 8.381 0.470 – 1.225
0.033* 0.011* 0.124
1.504 -1.112
4.530 0.327
Abbreviations as in Table I.
velocity, and LAA morphology (triangular, rounded, and “other” morphologies). It was found that age, LAA orifice diameter, and triangular LAA morphology were the variables significantly and independently associated with ischemic stroke in AF patients (Table III). The final logistic regression model was highly significant (P < 0.001), with good calibration (HosmerLemeshow Goodness-of-fit χ2 = 3.69, P = 0.884) and high discriminative power (C-statistic = 0.86 [0.81–0.93]; P < 0.001, Nagelkerke R2 = 0.713). The model’s correlation matrix was checked to avoid collinearity, which was excluded on the basis of reasonable correlations (generally, r < ±0.5).
Discussion The LAA is an important reservoir for the development of atrial thrombi.7) Recently, it was reported that thromboem-
bolic events could occur even in patients with sinus rhythm as a result of depressed LAA function.15) In this study, we evaluated LAA structure and functions by TEE in patients with nonvalvular paroxysmal and chronic AF, about half of whom had prior ischemic cerebral stroke. Because the presence of thrombus was more common in patients with decreased LAA flow velocities,16) the evaluation of LAA function with TEE in patients with AF or in sinus rhythm may enable the detection of patients prone to thromboembolism. The early detection of this tendency for LAA thrombus formation may be crucial to prevent the embolic consequences.17) Age and LAA morphology were reported as independently related to silent cerebral ischemia burden in AF patients referred for catheter ablation.17) This relationship was first investigated in a multicenter study that reported a protective odds ratio for symptomatic stroke in AF patients with chicken-wing
Vol 56 No 3
LAA MORPHOLOGY BY TEE AND STROKE IN AF PATIENTS
LAA morphology compared with other morphologies.18) This may be explained by the more complex internal anatomical structure (such as that of the windsock and cauliflower LAA morphologies) which more intensively promote local blood stasis and thrombogenesis compared to simpler structures (such as the chicken-wing morphology).17) To date, there are no sufficient data correlating the various LAA morphologies obtained by TEE with the thromboembolic risk of stroke in patients with AF, and our present study is the first to address this topic. We found that the triangular LAA morphology is significantly and independently associated with ischemic stroke when compared with the other LAA morphologies. These findings could be clinically relevant, especially for patients judged to be at low-risk of thromboembolic events, such as those with CHADS2 scores of 0 and 1. In these patients, the presence of a triangular LAA morphology increases the odds for stroke, suggesting the need of a more aggressive antithrombotic therapy. Furthermore, this study may explain the occurrence of stroke in patients with low risk of thromboembolism (CHADS2 score of zero). LAA structure and functions: The LAA is an embryologic remnant that functions during conditions of fluid overload as a blood reservoir.7) The LAA is prone to stasis because of its hooked morphology; for this reason it represents the prevalent site of thrombus formation in patients with AF.7) With the use of TEE, it was reported that up to 98% of atrial thrombi occurring during AF derive from the LAA.5,19) Several studies have showed that among patients with AF, subgroups with low LAA flow velocities have a greater risk of thromboembolic events than those with high flow velocities.20,21) It is now known that AF is coupled with a marked reduction in LAA flow velocity.21,22) Previous studies on patients with AF have shown that subgroups with LAA flow velocity < 20 cm/s,20) < 25 cm/s,21,23) and < 35 cm/s 24) are at greater embolic risk than other patients. Interestingly, patients with LAA flow velocity < 37.0 cm/s and a larger (> 3.5 cm2) LAA orifice had a high incidence of stroke that reached 75%. Therefore, this subpopulation is in need of strict anticoagulation to prevent stroke. These data partially agree with our results which showed that emptying and filling velocities obtained by pulsedwave Doppler were significantly depressed in patients with ischemic stroke compared with non-stroke patients in univariate analyses. However, this association was lost after multivariate adjustment using logistic regression analysis; but the same model confirmed the independent positive association between the width of LAA orifice and the occurrence of stroke. One recent study documented that the LAA length was a determinant of thromboembolism, with a shorter length conferring a greater risk.25) In our study, the increase in LAA length was greater in AF patients with stroke than in the nonstroke group patients; but the difference was statistically insignificant (Table I). Anticoagulation management: The CHADS2 score was introduced into guidelines and implemented into clinical practice to assess individual thromboembolic risk in patients with AF. In patients with a CHADS2 score > 1, the need for oral anticoagulation is not questionable; but in patients with low-intermediate risk for stroke (CHADS2 score = 1), no consensus exists on whether patients should receive oral anticoagulation or antiplatelet therapy.26) Thus, the implementation of LAA morphology may aid the clinical decision towards oral anticoagu-
333
lant or antiplatelet therapy in this subpopulation of patients. In patients with contraindication(s) to oral anticoagulants or because of physician and/or patient preference, it is possible to use antiplatelet therapy (but with conflicting results).27) In this scenario, the identification of LAA morphology associated with a lower risk for stroke may further guide clinicians in the decision process. The findings of the present study suggest that the LAA morphology may be taken into account when planning the anticoagulation management of patients with AF; with the triangular-type LAA mandating an intensified anticoagulation strategy compared to the more benign rounded variety in which antiplatelets only could be sufficient. CRP in the context of AF and stroke: CRP, an acute-phase reactant, is an indicator of underlying systemic inflammation and a novel plasma marker for atherothrombotic disease. The recent use of highly sensitive CRP assays has enhanced the usefulness of CRP as a reliable predictor of cardiovascular events. It has been documented that plasma CRP level was elevated in patients with AF, and plasma CRP was an independent predictor for AF.28) Moreover, elevated plasma CRP levels significantly predicted the risk of future ischemic stroke and TIA in elderly patients independent of other cardiovascular risk factors.29) Therefore, we measured hsCRP to identify any possible association between hsCRP and ischemic stroke in this subset of patients with AF; as any positive independent association could be practically implemented for optimizing anticoagulation therapy. In the current study, the mean plasma hsCRP level was significantly increased in AF patients with ischemic stroke compared to the stroke-negative group (Table I). In this regard, it was shown that the AF patients with ischemic stroke had significantly higher plasma hsCRP levels than the patients with simple AF, and hsCRP was an independent risk factor for ischemic stroke in patients with AF after adjusting for the other cardiovascular risk factors.30) However, this association between hsCRP and ischemic stroke found in univariate analysis in AF patients was lost after multivariate adjustment using the logistic regression model in our study. A similar finding was reported by Potpara, et al 31) where the CHA2DS2-VASc score reliably predicts a short-term, 30-day unfavorable outcome of acute ischemic stroke regardless of the presence or absence of AF, and that adding high-sensitivity TnI or fibrinogen or CRP, or all three biomarkers, to the CHA2DS2-VASc score does not further improve the prediction. Study limitations: First, the small number of patients included in this study is a major limitation. However, our study population was quite homogeneous as only patients with non-valvular AF were included. Secondly, the relatively small number of patients with AF and LAA thrombi may make the statistical comparisons between patients with and without LAA thrombi quite difficult. Thirdly, we were unable to retrieve the anticoagulation status at the time of event especially in the subgroup of patients at high risk of stroke (ie, CHADS2 scores ≥ 2). Finally, it was impossible to determine the duration of AF in the majority of patients; so that this potentially important variable was not included in the analyses. Conclusions: LAA morphology assessed by TEE may be useful for predicting ischemic cerebral stroke in patients with nonvalvular AF, especially in those with a low CHADS2 score.
334
SAKR, ET AL
Disclosure Conflicts of interest: No potential conflicts of interest exist.
Acknowledgment The authors thank Dr. Mohammed Osama (consultant of cardiology, National Heart Institute, Imbaba, Egypt) for his sincere help and advice regarding this study.
16. 17. 18.
19.
References
1. Go AS, Mozaffarian D, Roger VL, et al; American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Executive summary: heart disease and stroke statistics--2014 update: a report from the American Heart Association. Circulation 2014; 129: 399-410. 2. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation 2006; 114: 119-25. 3. Sharma S, Devine W, Anderson RH, Zuberbuhler JR. The determination of atrial arrangement by examination of appendage morphology in 1842 heart specimens. Br Heart J 1988; 60: 227-31. 4. Tuccillo B, Stümper O, Hess J, et al. Transoesophageal echocardiographic evaluation of atrial morphology in children with congenital heart disease. Eur Heart J 1992; 13: 223-31. 5. Leung DY, Black IW, Cranney GB, Hopkins AP, Walsh WF. Prognostic implications of left atrial spontaneous echo contrast in nonvalvular atrial fibrillation. J Am Coll Cardiol 1994; 24: 755-62. 6. Wang Y, Di Biase L, Horton RP, Nguyen T, Morhanty P, Natale A. Left atrial appendage studied by computed tomography to help planning for appendage closure device placement. J Cardiovasc Electrophysiol 2010; 21: 973-82. 7. Al-Saady NM, Obel OA, Camm AJ. Left atrial appendage: structure, function, and role in thromboembolism. Heart 1999; 82: 54754. (Review) 8. Martinez MW, Kirsch J, Williamson EE, et al. Utility of nongated multidetector computed tomography for detection of left atrial thrombus in patients undergoing catheter ablation of atrial fibrillation. JACC Cardiovasc Imaging 2009; 2: 69-76. 9. Patel A, Au E, Donegan K, et al. Multidetector row computed tomography for identification of left atrial appendage filling defects in patients undergoing pulmonary vein isolation for treatment of atrial fibrillation: comparison with transesophageal echocardiography. Heart Rhythm 2008; 5: 253-60. 10. Mohrs OK, Nowak B, Petersen SE, et al. Thrombus detection in the left atrial appendage using contrast-enhanced MRI: a pilot study. AJR Am J Roentgenol 2006; 186: 198-205. 11. Manning WJ, Weintraub RM, Waksmonski CA, et al. Accuracy of transesophageal echocardiography for identifying left atrial thrombi. A prospective, intraoperative study. Ann Intern Med 1995; 123: 817-22. 12. Stöllberger C, Ernst G, Bonner E, Finsterer J, Slany J. Left atrial appendage morphology: comparison of transesophageal images and postmortem casts. Z Kardiol 2003; 92: 303-8. 13. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994; 154: 1449-57. 14. Predictors of thromboembolism in atrial fibrillation: I. Clinical features of patients at risk. The stroke prevention in atrial fibrillation investigators. Ann Intern Med 1992; 116: 1-5. 15. Manning WJ, Silverman DI, Waksmonski CA, Oettgen P, Douglas PS. Prevalence of residual left atrial thrombi among patients with
20.
21.
22.
23.
24.
25. 26.
27.
28.
29. 30.
31.
Int Heart J May 2015
acute thromboembolism and newly recognized atrial fibrillation. Arch Intern Med 1995; 155: 2193-8. Kamp O, Verhorst PM, Welling RC, Visser CA. Importance of left atrial appendage flow as a predictor of thromboembolic events in patients with atrial fibrillation. Eur Heart J 1999; 20: 979-85. Anselmino M, Scaglione M, Di Biase L, et al. Left atrial appendage morphology and silent cerebral ischemia in patients with atrial fibrillation. Heart Rhythm 2014; 11: 2-7. Di Biase L, Santangeli P, Anselmino M, et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J Am Coll Cardiol 2012; 60: 531-8. Uslu N, Nurkalem Z, Orhan AL, et al. Transthoracic echocardiographic predictors of the left atrial appendage contraction velocity in stroke patients with sinus rhythm. Tohoku J Exp Med 2006; 208: 291-8. Zabalgoitia M, Halperin JL, Pearce LA, Blackshear JL, Asinger RW, Hart RG. Transesophageal echocardiographic correlates of clinical risk of thromboembolism in nonvalvular atrial fibrillation. Stroke Prevention in Atrial Fibrillation III Investigators. J Am Coll Cardiol 1998; 31: 1622-6. Mügge A, Kühn H, Nikutta P, Grote J, Lopez JA, Daniel WG. Assessment of left atrial appendage function by biplane transesophageal echocardiography in patients with nonrheumatic atrial fibrillation: identification of a subgroup of patients at increased embolic risk. J Am Coll Cardiol 1994; 23: 599-607. Rubin DN, Katz SE, Riley MF, Douglas PS, Manning WJ. Evaluation of left atrial appendage anatomy and function in recent-onset atrial fibrillation by transesophageal echocardiography. Am J Cardiol 1996; 78: 774-8. Panagiotopoulos K, Toumanidis S, Vemmos K, Saridakis N, Stamatelopoulos S. Secondary prognosis after cardioembolic stroke of atrial origin: the role of left atrial and left atrial appendage dysfunction. Clin Cardiol 2003; 26: 269-74. Fatkin D, Kelly RP, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol 1994; 23: 961-9. Khurram IM, Dewire J, Mager M, et al. Relationship between left atrial appendage morphology and stroke in patients with atrial fibrillation. Heart Rhythm 2013; 10: 1843-9. Wann LS, Curtis AB, Ellenbogen KA, et al. 2011 ACCF/AHA/ HRS focused update on the management of patients with atrial fibrillation (update on dabigatran): a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol 2011; 57: 1330-7. European Heart Rhythm Association; European Association for Cardio-Thoracic Surgery, Camm AJ, Kirchhof P, Lip GY, et al; ESC Committee for Practice Guidelines. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010; 12: 1360-420. Chung MK, Martin DO, Sprecher D, et al. C-reactive protein elevation in patients with atrial arrhythmias: inflammatory mechanisms and persistence of atrial fibrillation. Circulation 2001; 104: 2886-91. Rost NS, Wolf PA, Kase CS, et al. Plasma concentration of C-reactive protein and risk of ischemic stroke and transient ischemic attack: the Framingham study. Stroke 2001; 32: 2575-9. You L, Wang P, Lv J, Cianflone K, Wang D, Zhao C. The role of high-sensitivity C-reactive protein, interleukin-6 and cystatin C in ischemic stroke complicating atrial fibrillation. J Huazhong Univ Sci Technolog Med Sci 2010; 30: 648-51. Potpara TS, Polovina MM, Djikic D, Marinkovic JM, Kocev N, Lip GY. The association of CHA2DS2-VASc score and blood biomarkers with ischemic stroke outcomes: the Belgrade stroke study. PLoS One 2014; 9: e106439.
