- Email: [email protected]
Incidence, Predictive Factors, and Prognostic Impact of Silent Atrial Fibrillation After Transcatheter Aortic Valve Implantation
Incidence, Predictive Factors, and Prognostic Impact of Silent Atrial Fibrillation After Transcatheter Aortic Valve Implantation 1XD XRaphael Robert, D2X XMDa,y, D3X XGuillaume Porot, D4X XMDa,y, D5X XClemence Vernay, D6X XMDa,y, D7X XPhilippe Buffet, D8X XMDa,y, D9X XMarie Fichot, D10X XMDa,y, D1X XCharles Guenancia, D12X XMD, PhDa,b,y, D13X XThibaut Pommier, D14X XMDa,y, D15X XBasile Mouhat, D16X XMDa,y, D17X XYves Cottin, D18X XMD, PhDa,b,y, and D19X XLuc Lorgis, D20X XMD, PhDa,b,y,* New onset atrial fibrillation post-transcatheter aortic valve implantation (TAVI) is common and is associated with adverse outcomes. However, silent atrial fibrillation (AF) is poorly documented in the context. This study sought to evaluate the incidence, predictive factors, and prognostic value of Silent AF post-TAVI. All the consecutive patients with TAVI were prospectively analyzed by continuous electrocardiogram monitoring48 hours after implantation. Silent AF was defined as asymptomatic episodes lasting at least 30 seconds. The population was divided into 3 groups: history of AF, no-AF, and silent AF. Among the 206 patients implanted with TAVI, 19 (16.1%) developed silent AF. Compared with the no-AF group, patients with silent AF shared the same clinical characteristics and cardiovascular risk factors. Procedural success and echography parameters after the device implantation were similar between groups. Left atrial volume was significantly increased (p <0.001) in the silent AF group, together with preimplantation C-reactive protein (CRP) >3 mg/L and glucose (p = 0.048 and p = 0.002). By multivariate analysis, CRP >3 mg/dl and logistic European System for Cardiac Operative Risk Evaluation were identified as independent predictors of silent AF. In-hospital and 1-year mortalities were higher in pre-existing AF patients, whereas no-AF and the silent AF patients share the same prognosis. Our prospective study showed for the first time that silent AF is frequent after TAVI procedures. In conclusion, our work suggests that CRP could help to predict the risk of developing silent AF. However, the onset of silent AF is not associated with worse prognosis in the year following the procedure in our study. © 2018 Elsevier Inc. All rights reserved. (Am J Cardiol 2018;122:446454)
Transcatheter aortic valve implantation (TAVI) has become a therapeutic alternative for patients with severe symptomatic aortic stenosis and with a contraindication to a standard surgery, or high-risk surgical procedure. New onset atrial fibrillation (NOAF) is a well-known complication after this procedure. Its occurrence induces an increase of cardiovascular (CV) death and an increase of arterial embolic events. Ischemic strokes’ prevalence is around 4% after TAVI. About 1/3 of cerebrovascular events occur during the TAVI procedure. However, more of 50% of them occur several days after TAVI,1 suggesting that mechanisms other than those directly related to the procedure may be involved. However, in most cases, atrial fibrillation episodes are asymptomatic and could be unknown by the a Department of Cardiology, University Hospital, Dijon, France; and Laboratory of Cerebro-Vascular Pathophysiology and epidemiology (PEC2) EA 7460, University of Burgundy, Dijon, France. Manuscript received January 24, 2018; revised manuscript received and accepted April 13, 2018. This work was supported by the University Hospital of Dijon, the Faculty of Medicine of Dijon, Dijon, France. See page 453 for disclosure information. *Corresponding author: Tel: (+33) 380293311; fax: (+33) 380293333. E-mail address: [email protected] (L. Lorgis). yThese investigators take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. b
0002-9149/© 2018 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.amjcard.2018.04.029
clinician. Indeed, the large ASSERT prospective study2 found an incidence for silent atrial fibrillation of 10.1% in a cohort of patient without a history of previous atrial fibrillation, 3 months after pacemaker or defibrillators implantation. Few studies have described predictive factors of this arrhythmia. After myocardial infarction (MI), Stamboul et al found that indexed left atrial (LA) diameter and age are independent predictive factors of new onset silent atrial fibrillation.3 In the ASSERT study, investors found that silent atrial fibrillation was associated with a fivefold increased risk of clinical atrial fibrillation. In addition, patients who had silent atrial fibrillation (AF) had twice more risk to have a stroke or an ischemic arterial embolic event during the 2.5 years after. The attributable risk of stroke or systemic embolism associated with silent AF was estimated to 13% in this population. To date, no study has considered the impact and prevalence of silent AF after TAVI procedure. The aim of this prospective study was to assess the incidence, the predictive factors, and the short-term prognosis for the patient who had at least 1 silent atrial fibrillation episode in the 48 hours after TAVI. Methods April 2015 to October 2016, all consecutive patients admitted in intensive care unit after TAVI procedure were www.ajconline.org
Valvular Heart Disease/Silent AF after TAVI
included in this monocentric study conducted in the University Hospital of Dijon. A total of 269 consecutive patients with severe symptomatic aortic stenosis underwent TAVI with a balloon-expandable valve (Edwards SAPIEN, SAPIEN XT, Edwards Lifesciences, Irvine, California) or self-expandable valve Medtronic Corevalve (Medtronic, Inc., Minneapolis, Minnesota) at our institution. Of these, 3 patients with symptomatic AF were excluded, together with 4 who died during the procedure. Details about the TAVI procedure (transfemoral and transapical) have been extensively explained in previous studies.4 Selection of transapical approach was based on the following criteria: small diameter of the iliofemoral arteries, significant peripheral vascular disease, severe calcification or tortuosity of both iliofemoral arteries, severely calcified or porcelain aorta. Patients received full-dose heparin (100 UI/kg, adjusted for an activated clothing time 250 seconds) during the procedure, and aspirin (100 mg/day) + clopidogrel (75 mg/day) were administered following the procedure. The study protocol was approved by the ethics committee of the University Hospital of Dijon, and a written consent form was obtained from the patient or his and/or her legal representative for the France 2 registry inclusion.5 Fifty-six patients with an
447
incomplete data on continuous electrocardiogram monitoring (CEM) 48 hours after TAVI were excluded, leading to a final study population of 206 patients (Figure 1). AF was diagnosed in accordance with the current European Society of Cardiology Guidelines6 as follows: - Absolutely irregular RR intervals; - No distinct P waves; - Atrial cycle length (when visible) 200 ms (atrial rate 300 beats/min). CEM Philips intellivue MP50 was started immediately after admission and recorded continuously for the first 48 hours. The signal was recorded in 2 leads (DII and V1), with a sample frequency of 8 kHz, and a gain resolution of 5 mV/Least Significant Bit. Monitored ECG data were sent to a computer where an unsupervised automated AF episode detection algorithm was applied. Automatic measurements are based on the analysis of RR interval variability, QRS morphology, and P-wave presence screened on 15 beats. Only 15 beats were required to identify AF, and duration of 30 seconds was required to be considered an episode of AF.
Figure 1. Study flow chart.
448
The American Journal of Cardiology (www.ajconline.org)
This monitoring system includes the following rhythm alarms: (1)Flat ECG (2)Ventricular fibrillation (3)Couplets or pulsus bigeminus (4)Supra-ventricular extra-systoles
admission, in the left lateral decubitus position, using a commercially available system (Vivid S6, General ElectricMedical Systems, Horton, Norway). Images were obtained by a simultaneous ECG signal, using a 3.5 MHz transducer, in the parasternal and apical views. Standard M-mode and 2-dimensional images were acquired during breath hold and saved in the cine loop format. The systolic function of LV was assessed using the biplane Simpson method in the apical 4-chamber and apical 2-chamber views according to the American Society of Echocardiography.7 In accordance with the guidelines of the American Society of Echocardiography:
(5)Ventricular tachycardia (6)Preset upper and lower heart rate thresholds (100/min and 40/min) In case of AF episodes, ECG source data were reviewed, and patients were systematically interviewed by the investigator of the study for their symptoms, that is, heart failure, palpitations, or dyspnea related to AF. Symptomatic AF group was defined as patients with AF occurring within the 48 hours after the procedure, confirmed on ECG, resulting in clinical symptoms or the need for urgent cardioversion or need to start curative anticoagulation or amiodarone for patient. These patients were excluded from the statistical analysis. Silent AF group was defined as patients with the occurrence of at least 1 episode of AF 30 seconds occurring after the TAVI procedure, and only detected by CEM, for patient without history of AF before TAVI. So, the study population was divided into 3 groups: history of AF, silent AF, and no new AF. Preprocedural and postprocedural were gathered according to the France 2 registry5 model. Data on demographics, CV risk factors (history of hypertension or treated hypertension, diabetes, history of hypercholesterolemia or treated hypercholesterolemia, current smoking, and history of AF), were collected prospectively, along with admission characteristics and hemodynamic parameters. We defined previous MI as at least 1 MI before the admission. Anthropometric parameters (i.e., body mass index [weight (kg)/height (m2)] were measured within 48 hours of admission. Heart failure, defined as a Killip’s class >1, was also gathered. Long-term and acute (first 48 hours) treatments were also collected. We evaluated risk factors for CV surgery using the logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE), which is calculated by a logistic-regression equation (on a scale from 0% to 100%, with higher scores indicating greater risk and a score of more than 20% indicating very high surgical risk). The CHA2DS2-Vasc, as predictive score of stroke in patients with AF, was also calculated. Blood samples were drawn at admission with assessment of levels of C-reactive protein (CRP), glycaemia, thyroidstimulating hormone, hemoglobin, NT-pro-BNP, lipid profile, peak creatinine kinase, troponin I, serum creatinine clearance were estimated by the Cockcroft and Gault formula. CRP level was dichotomized at 3 mg/L for more clinical relevance. All patients included in the study had undergone an echocardiography within the first 48 hours after their
- LA anteriorposterior diameter was calculated in the parasternal long-axis view, using the M-mode technique by measuring from the trailing edge of the posterior wall of the aortic root to the leading edge of the posterior LA wall. - LA area and volume were calculated by the bi-plane Simpson method in the apical 4-chamber and apical 2chamber views and were indexed to the body surface area. The LA appendage, the pulmonary vein origins, and the funnel of the mitral valve were excluded from measurements. LA area and LA volume were calculated at the end of the T wave on the electrocardiogram (ECG), just before the opening of the mitral valve. Left ventricular ejection fraction (LVEF) was dichotomized at 40% for more clinical relevance. Coronary status was based on pre-TAVI coronarography data. A coronary stenosis was considered as significant if it was at least of 50% stenosed. An invasive mean pulmonary arterial pressure was obtained for almost all patients before TAVI. 30-day and 1-year follow-up (median FU was 280 § 153 days) was obtained for the all cohort except 1 patient who was lost to FU, either by telephone interview or mail to the patients, the patient’s relatives or the treating physician and patient’s medical records review. The following adverse outcomes were assessed according to the Valve Academic Research Consortium Classification II8 and adjudicated by a clinical events committee. For the in-hospital stay, we gathered a combined End point that was defined at hospital discharge and included all-cause mortality, TIA or stroke, and MI (within 72 hours after the index procedure, consisting of at least 1 sample after procedure with a peak value exceeding 15£ as the upper). The data were expressed as percentages (n [%]), median (twenty-fifth to -seventy-fifth percentile) or mean § SD as appropriate. The normality of the distribution of continuous data was tested with the KolmogorovSmirnov test. The categorical variables were analyzed using de chi-square or Fisher’s exact test. Continuous variables were analyzed by ANOVA with F test if normally distributed and with KruskalWallis test otherwise. For 2-group comparisons of non-normally distributed variables, values were tested by either the MannWhitney or Wilcoxon rank sum test for unpaired or paired data, respectively. Estimates of silent AF were studied by backward logistic regression analysis using inclusion and exclusion cutoffs at 5% and 1%, respectively. Variables entered into the multivariate model to
Valvular Heart Disease/Silent AF after TAVI
estimate silent AF were chosen according to their relation in univariate analysis (i.e., logistic EuroSCORE, age, female gender, smoking, hypertension, diabetes, previous CAD, or stroke or AF, LVEF <40%, creatinine clearance, CRP >3, indexed LA area, previous PCI, log NT-ProBNP, admission glycaemia, acute and long-term treatments (b blocker, statin, and angiotensin-converting enzyme [ACE] inhibitor or aspirin). Backward multivariate logistic regression analyses were also used to test the association with inhospital and FU-combined end point, including AF in addition to logistic EuroSCORE and LVEF. Statistical analyses were performed using the SPSS 12.0 software (IBM Inc., Armonk, New York). Results April 2015 to October 2016, 206 patients were included in the study, and only 1 was lost to FU. Total 99 (83.9%) patients stayed in sinus along the monitoring, 88 (42.7%) of them had a preexisting AF (paroxysmal AF = 31; permanent AF = 47; persistent AF but sinus rhythm at the time of the procedure = 10). In the population without a history of AF, 19 (16.1%) had at least 1 episode of silent atrial fibrillation. A comparison of the 3 groups is given in Tables 1 and 2. Despite we found no statistical differences regarding demographics and CV risk factors, patients diagnosed with a silent AF after TAVI have a lower logistic EuroSCORE
449
and higher CRP and admission glycaemia levels when compared with others groups. Moreover, clinical presentation, hemodynamic data on admission, and long-term treatments were similar for the 3 groups. Echocardiographic parameters clearly show a higher LA remodeling between AF groups (silent or history of AF) compared with the no AF groups. However, we found no statistical differences when we compared LA parameters between AF groups (silent vs history of AF) (Table 2). Of note, the proportion of patients with a LVEF <40% was similar in the 3 groups. Procedural details and success were strictly identical among groups. The hemodynamic parameters on CEM during the first 48 hours phase were similar in the 3 groups, whatever the FA occurrence. Among the acute therapies, there were no significant differences on b blockers and ACE inhibitors. However, statins were significantly less prescribed in the silent AF group. In the history of AF group, patients were more often on VKA (83%) and amiodarone (25%) (Table 3). In-hospital and FU-combined end point are described in Table 4. Despite we found differences in the frequency of post-TAVI MI and severe arrhythmias, we did not show any variance on the composite in-hospital end point between groups. However, during the FU period, the population with pre-existing AF had a worse short-term prognosis as witnessed by an increased efficacy end point, which is significantly higher than patients unknown for AF before
Table 1 Baseline characteristics of the study population underwent transcatheter aortic valve procedure according to AF occurrence. n(%) or median (twenty-fifth to seventy-fifth) Variable Age (years) Women Hypertension Diabetes mellitus History of CAD Chronic pulmonary disease Prior MI Prior bypass graft BMI (kg/m2) Peripheral vascular disease History of stroke Permanent pacemaker Renal failure (DFG <60) NYHA 3-4 LVEF <40 % Complete revasc Log EuroSCORE (% § SD) CHA2DS2-VASC Score Biological data CRP 3 mg/l Hemoglobin Platelets Glucose, mmol/l Nt-proBNP
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
86 (81-89) 105 (51%) 132 (64%) 56 (27%) 123 (60%) 10 (5%) 7 (3%) 17 (8%) 26.0 (24-29) 33 (16%) 24 (11%) 23 (11%) 103 (51%) 143 (69%) 37 (18%) 38 (18%) 19.8 § 11.6 4.49 § 0.6
86 (82-89) 44 (50%) 58 (66%) 22 (25%) 58 (66%) 5 (6%) 4 (4%) 7 (8%) 26.0 (24.2-29.8) 14 (16%) 10 (11%) 10 (11%) 50 (57%) 65 (73%) 21 (24%) 15 (17%) 22.5 § 12.6*,y 4.52 § 0.11
86 (81-88) 52 (53%) 61 (61%) 28 (28%) 57 (57%) 4 (4%) 3 (3%) 7 (7%) 25.8 (23.4-28.6) 17 (17%) 12 (12%) 12 (12%) 45 (47%) 68 (68%) 13 (13%) 19 (19%) 18.4 § 11.0 4.47 § 0.08
87 (80-88) 9 (48%) 13 (69%) 6 (32%) 8 (42%) 1 (4%) 0 (0%) 3 (16%) 26 (24.3-28.7) 2 (10%) 2 (10%) 1 (5%) 8 (42%) 10 (52%) 3 (16%) 4 (2%1) 14.8 § 6 4.47 § 0.17
0.299 0.892 0.762 0.795 0.133 0.870 0.588 0.445 0.727 0.769 0.974 0.683 0.258 0.186 0.157 0.910 0.010 0.937
71 (34%) 12.1 (10.8-13.2) 206 (168-258) 5.56 (4.81-6.53) 2890 (1211-6287)
34 (39%) 12.1 (10.7-13.2) 201 (154-264) 5.68 (5-6.52)y 3804 (1988-9699)*
29 (29%) 12.2 (10.8-13.2) 207 (168-256) 5.34 (4.65-6.45) 1877 (785-4459)
8 (42%)z 11.9 (11-13.3) 212 (170-240) 6.22 (5.08-11.48)z 2846 (764-4744)
0.065 0.697 0.764 <0.001 0.045
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z : No new AF versus silent AF. *
450
The American Journal of Cardiology (www.ajconline.org)
Table 2 Baseline echocardiographic and procedural data of the study population according to AF occurrence Variable Echographic LVEF (%) Aortic valve area (cm2 § SD) Mean gradient (mm Hg § SD) Indexed LA diam (cm2/m2) Indexed LA area (cm2/m2) Indexed LA Vol (cm3/m2) MR >grade 2 Mean invasive PAP >35 mm Hg Procedure Percutaneous femoral approach Chirurgical approach Conscious sedation Edwards SAPIEN Medtronic CoreValve Contrast volume (ml § SD) Procedural success
Overall (n = 206)
History of AF (n = 88)
No New AF (n = 99)
Silent AF (n = 19)
p
55 (45-60) 0.66 § 0.17 46 § 14 3 (2.5-3.7) 14 (11.6-17) 47 (34-58) 8 (4%) 96 (47%)
55 (45-60) 0.66 § 0.17 43 § 14*,y 3.3 (2.8-3.85) 15 (12.5-19) 51 (39-68)* 5 (6%) 48 (54%)*
55 (50-60) 0.66 § 0.18 47 § 15 2.3 (2.4-3.65) 13 (11-15.7) 40 (30-541) 2 (2%) 40 (40%)
55 (50-60) 0.67 § 0.15 52 § 11 3.05 (2.3-3.7) 14.3 (13-16) 49 (44-57) 1 (5%) 8 (42%)
0.263 0.909 0.019 0.148 0.006 <0.001 0.411 0.141
199 (97%) 7 (3%) 194 (94%) 135 (66%) 71 (34%) 142 § 56 203 (98%)
85 (97%) 3 (3%) 81 (92%) 61 (69%) 27 (30%) 135 § 50y 87 (99%)
96 (97%) 3 (3%) 95 (96%) 63 (64%) 36 (36%) 143 § 61 97 (98%)
18 (95%) 1 (5%) 18 (95%) 11 (58%) 8 (42%) 167 § 47 19 (100%)
0.886 0.886 0.519 0.573 0.547 0.095 0.755
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z: No new AF versus silent AF. *
TAVI: 50% versus 28% versus 16%, respectively (p = 0.001) (Figure 2). Among outcomes gathered, there was a significant higher rate of CV death: 16% versus 5% versus 0%, (p = 0.013), together with a higher rate of rehospitalization, especially for heart failure symptoms: 22% versus 6% versus 0%, (p <0.001). Albeit these results in the history of AF group were expected, the difference of prognosis seen between groups at 30 days FU was reinforce at 1-year FU, without no difference of adverse outcomes between the silent AF groups versus the no new AF group.
By univariate analysis, predictive factors were associated with the occurrence of silent AF were CRP >3 mg/dl and glycaemia on admission (odds ratio [OR] 1.21, 95% confidence intervals [CIs]: 0.99 to 1.07, p = 0.032). There was a trend for the logistic EuroSCORE (OR 0.90, 95% CI: 0.31 to 1.34, p = 0.098) and the mean aortic gradient pre-TAVI (OR 1.03, 95% CI: 0.99 to 1.07, p = 0.057). By multivariate logistic regression analysis, only the CRP >3 mg/dl (OR 10.84 95% CI: 1.37 to 85.9, p = 0.05) and the logistic EuroSCORE (OR 0.94, 95% CI: 0.88 to 1.00, p = 0.05) were
Table 3 Continuous scope data within first 48 hours and discharge therapies of the study population according to AF occurrence Variable Day 1 HR median, beat/min HR max, beat/min HR min, beat/min Number of SVES episodes Day 2 HR median, beat/min HR max, beat/min HR min, beat/min Number of SVES episodes Acute therapies Aspirin Beta blockers ACE inhibitor Statin Vitamin K antagonist Amiodarone p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z : No new AF versus silent AF. *
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
71 (63-83) 92 (80-107) 60 (54-71) 4.5 (0-15)
72 (62-85) 92 (79-108) 63 (56-76)* NA
71 (63-85) 89 (80-105) 58 (54-68) 2 (0-8)
71 (62-80) 105 (83-119)z 61 (55-63) 7 (3-13)
0.612 0.223 0.069 <0.001
71 (66-83) 99 (85-112) 62 (56-70) 5 (1-19)
71 (64-84) 97 (82-114) 62 (53-71) 6 (2-109)*
72 (67-80) 100 (86-112) 62 (57-69) 4 (1-18)
73 (64-89) 101 (87-110) 65 (560-82) 3 (1-6)
0.866 0.999 0.323 0.072
164 (82%) 117 (57%) 140 (68%) 115 (56%) 88 (43%) 26 (13%)
62 (74%)* 48 (55%) 59 (67%) 45 (51%) 73 (83%)*,y 22 (25%)*,y
87 (90%) 57 (57%) 70 (70%) 63 (63%) 14 (14%) 4 (4%)
15 (83%) 12 (63%) 11 (56%) 7 (37%)z 1 (5%) 0 (0%)
0.020 0.771 0.532 0.050 <0.001 <0.001
Valvular Heart Disease/Silent AF after TAVI
451
Table 4 In-hospital and follow-up outcomes to transcatheter aortic valve procedure according to AF occurrence
In hospital Hospital stay (days) Death Composite End point Stroke Myocardial infarction Vascular complication Major bleeding New pacemaker Severe arrhythmia Valvular complication Aortic valve area (cm2 § SD) Mean gradient (mm Hg § SD) 30 days follow-up Efficacy end point Death all cause CV death Stroke or arterial embol Major Bleeding Acute Kidney injury 2 or 3 Rehosp. (heart failure) 1-year follow-up Efficacy End point Death all cause CV death Stroke or arterial embol Major Bleeding Acute Kidney injury 2 or 3 Rehosp. (heart failure)
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
11 § 0.44 6 (2%) 96 (46%) 5 (2%) 41 (20%) 17 (8%) 3 (1.4%) 43 (20%) 8 (4%) 5 (2.4%) 1.72 § 0.08 8.4 § 0.425
12.2 § 0.78* 4 (5%) 41 (47%) 2 (2%) 13 (15%)*,y 8 (9%) 1 (1%) 20 (23%) 2 (2%)* 2 (2%) 1.74 § 0.06 7.5 § 0.42*
9.8 § 0.5 2 (2%) 47 (47%) 2 (2%) 24 (24%) 7 (7%) 2 (2%) 19 (19%) 5 (5%) 3 (3%) 1.72 § 0.08 9.5 § 0.08
10.3 § 1.56 0 (0%) 8 (42%) 1 (5%) 4 (21%) 2 (10%) 0 (0%) 4 (21%) 1 (5%) 0 (0%) 1.70 § 0.30 6.6 § 0.11
0.031 0.432 0.912 0.697 0.267 0.821 0.755 0.838 0.586 0.683 0.977 0.027
51 (24%) 14 (6.7%) 13 (6.2%) 6 (2.9%) 6 (2.9%) 6 (2.9%) 14 (6.7%)
30 (34%)*,y 8 (9%) 10 (11.5%)* 4 (4.7%) 1 (1.2%) 3 (3.5%) 11 (13%)*,y
19 (19%) 5 (5%) 3 (3%) 2 (2%) 5 (5%) 2 (%2) 3 (3%)
2 (10%) 1 (5%) 0 (0%) 0 (0%) 0 (0%) 1 (4%) 0 (0%)
0.020 0.528 0.032 0.432 0.207 0.695 0.017
75 (36%) 40 (19%) 19 (9%) 8 (4%) 9 (4%) 2 (1%) 25 (12%)
44 (50%)*,y 27 (30%) 14 (16%)* 2 (2%) 4 (4%) 1 (1%) 19 (22%)*,y
28 (28%) 12 (12%) 5 (5%) 5 (5%) 5 (5%) 1 (1%) 6 (6%)
3 (16%) 1 (5%) 0 (0%) 1 (5%) 0 (0%) 1 (1%) 0 (0%)
0.001 0.002 0.013 0.586 0.611 0.899 <0.001
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z: No new AF versus silent AF. *
Figure 2. KaplanMeier survival curves for 1-year all-cause and CV mortality, stratified by AF status.
452
The American Journal of Cardiology (www.ajconline.org)
Table 5 Uni and multivariate analysis for predictors of silent AF after TAVI procedure Univariate variable Euroscore logistic Glucose on admission Mean aortic gradient pre TAVI CRP >3 mg/l
Multivariate Odds ratio (95% CI)
p
0.20 (0.31-1.34) 1.21 (1.02-1.44) 1.03 (0.99-1.07) 10.8 (1.38-84)
0.098 0. 032 0.057 0.023
Figure 3. Occurrence of type of AF according to the tertiles of CRP.
independent predictive factors of silent AF after TAVI (Table 5). After dichotomization in tertiles of CRP >3 mg/ dl, we found a linear relation with increasing occurrence of silent AF among the levels of CRP (Figure 3).
Discussion With an incidence of 16.1%, we found that silent AF is not rare in TAVI population. We have identified 2 independent predictive factors for silent AF after TAVI occurrence: a pre-TAVI CRP >3 mg/dl and a trend for the logistic EuroSCORE. Our result is similar with other studies: from France 2 registry, Chopard et al9 who tracked NOAF after TAVI with repeat ECGs found an incidence of 6.0%. Furuta et al10 and Amat-Santos et al,1 who tracked NOAF on CEM until hospital discharge, found an incidence of 7.6% and 31.9%, respectively. We shared the same method as the last 2 studies to detect silent AF, but our time window was shortened (48 hours). Obviously, result of the screening of AF strongly depends of the method used for detection NOAF, and the duration of the monitoring after TAVI procedure.11 Clinical independent predictive factors of AF occurrence after TAVI are very heterogeneous in literature. Despite our logistic regression, to predict silent AF, we have not highlighted a single clinical risk factor. In our study, the silent AF group had a lower logistic EuroSCORE than the permanent sinus rhythm: 14.8 § 6.0 versus 18.4 § 11.0, respectively.
variable
Odds ratio (95% CI)
p
Euroscore logistic
0.94 (0.88-1.00)
0.05
CRP >3 mg/l
10.84 (1.37-85.9)
0.024
A smaller study,1 with 138 patients, suggested that atrial size is associated with AF, with a proposed cutoff of indexed atrial diameter at 27 mm/m2 to predict atrial fibrillation occurrence. In our study, despite we found that patients with silent AF had a significant higher LA volume than controls, this was not found as a predictive factor in univariate and multivariate analysis. Finally, our study suggests that CRP is a strong predictive factor of silent atrial fibrillation in TAVI population. Kinoshita et al12, in a cohort of 656 patients who underwent off-pump coronary bypass, found that a high preoperative CRP increased atrial fibrillation risk in the 7 postoperative days. The investigators found a hazard ratio at 3.19 for the association of CRP >3 to 10 mg/L and postoperative atrial fibrillation. If CRP was 1 to 3 mg/L, hazard ratio decreased at 1.59. They proposed a cutoff point of 2.1 mg/L: patients with CRP >2.1 mg/L had a twofold increased risk of developing AF after surgery. In a large Danish prospective cohort of 46,000 patients,13 CRP was found as an independent predictive factor of NOAF. Patients were genotyped for 4 CRP gene polymorphisms associated with up to a 63% increase in plasma CRP levels. When the risk of atrial fibrillation was adjusted to CRP genotype and genotype combinations, CRP was no more associated with an increased risk of atrial fibrillation. That is demonstrating that this is the long-term inflammatory response that causes atrial fibrillation. Finally, TAVI with the TF approach results in a lower activation of inflammatory pathways than SAVR. Despite an attenuate inflammatory pathway, TAVI remains a physiological stress causing a significant inflammatory response that can act as a trigger for an AF occurrence. As expected, in the context of TAVI, a pre-existing AF was associated with increased composite end point during the in-hospital and the follow-up period, and a poor prognosis. Pre-existing AF has been previously linked with adverse ischemic cardiac and cerebrovascular events and mortality after TAVI (9 and 10). We also found a significant higher incidence of CV death (11.5%) and rehospitalization for heart failure (13%). Pre-existing AF is a strong risk marker in this population. Despite we previously showed, on a population of acute MI (AMI), that silent or symptomatic AF, was associated with higher hospital and 1-year mortality3, we did not find any difference in term of outcomes when we compared the silent atrial fibrillation group with the control group in the context of TAVI procedures. NOAF appears to be a risk marker of adverse events, but seems to be less pejorative than pre-existing AF. The different studies who analyzed NOAF after TAVI found a worse 1-month and 1-year prognosis than patients with
Valvular Heart Disease/Silent AF after TAVI
permanent sinus rhythm (9 and 10). However, in these studies, investigators did not make difference between silent NOAF and symptomatic NOAF. So, we have sought to refine the knowledge of the NOAF after TAVI. Our hypothesis that silent AF can help to stratify as an intermediaterisk population is not confirmed by our preliminary data. One can argue that this is not the silent AF occurrence that is associated with adverse outcomes, but the silent atrial fibrillation burden. We can also presume that the risk of death or adverse events and their competing associations are likely to be very different in this population of elderly and frail patients from in the settings of AMI. Thus, the rate of adverse events in the control group of SR patients is 5 to 6 times higher than the same population of AMI patients, but the TAVI group was 24 years older. Of note, our median follow-up is short, but our results are in line with previous published studies that did not find a difference of event at 1-month follow-up1416 after TAVI. Moreover, we clearly have a very low number of adverse events after TAVI in the silent AF group. Indeed, we recently learned from the ASSERT population, that more than the presence or the absence of silent atrial fibrillation, that atrial fibrillation burden greater than 24 hours was a powerful prognostic factor, and associated with a significant increase in the rate of ischemic stroke or systemic embolism.15 Interestingly, such relation was not observed when the duration of silent atrial fibrillation was 6 minutes to 24 hours. The elderly patient population underwent TAVI is predisposed to both thromboembolic and bleeding complication. The question of the prognosis raises the question of the therapeutic attitude to be adopted when such arrhythmia is detected. Although at least 3 ongoing randomized controlled trials (ARTE, ATLANTIS, and GALILEO) and a registry-based prospective study (POPular-TAVI) are evaluating the optimal antithrombotic or antiplatelet strategies after TAVI, none of them have specifically stratified patients based on pre-existing AF or NOAF. Our study had several limits. First, our work is monocentric, and size of our cohort is small. Second, we have collected adverse events during the year after the procedure and results could be very different from a long-term FU. Also, we have excluded the patients with the major procedural complications who require reanimation or and those who died during the procedure. In our study, the distinction between symptomatic and silent AF was defined according to the European Society of Cardiology guidelines,6 and was based on whether AF symptoms were present or not. This definition was therefore not absolutely objective and may over diagnosed asymptomatic arrhythmias. To correct this possible effect, we add to the silent AF definition the careful screening of drugs received during the acute phase of monitoring after TAVI, especially use of anticoagulation and antiarrhythmic. In fact, the incidence of symptomatic AF found in our study (1.4%) is lower than the other studies.3 The possible non-negligible portion of false positive of detection of atrial fibrillation, found in the others studies (5% to 12%), has been corrected by the systematic realization of a 12 leads ECG or a systematic analysis of the continuous monitoring. Silent atrial fibrillation is not
453
rare in patient who underwent TAVI. CRP >3 mg/dl was found as independent predictive factor. In clinical practice, CRP on admission can be determined easily and may be useful for noninvasive risk stratification of developing silent AF after TAVI. Silent atrial fibrillation was not associated with a poor prognosis but we must carry out a larger study because the therapeutic impact may be greater with the necessity or not to start a curative anticoagulation in this frail population. Acknowledgment We wish to thank Anais Hamon, Florence Bichat, and Maud Maza for their research assistance and Philip Bastable for editorial assistance. Disclosures The authors have no conflicts of interest to disclose. 1. Amat-Santos IJ1, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, Villeneuve J, C^ote M, Nombela-Franco L, Philippon F, Pibarot P, Dumont E. Incidence, predictive factors, and prognostic value of new-onset atrial fibrillation following transcatheter aortic valve implantation. J Am Coll Cardiol 2012;59(2):178–188. 2. Healey JS, Connolly SJ, Gold MR, Israel CW, Van Gelder IC, Capucci A, Lau CP, Fain E, Yang S, Bailleul C, Morillo CA, Carlson M, Themeles E, Kaufman ES, Hohnloser SH. ASSERT Investigators. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med 2012;366 (2):120–129. 3. Stamboul K, Zeller M, Fauchier L, Gudjoncik A, Buffet P, Garnier F, Guenancia C, Lorgis L, Beer JC, Touzery C, Cottin Y. Prognosis of silent atrial fibrillation after acute myocardial infarction at 1-year follow-up. Heart 2015;101(11):864–869. 4. Rodes-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, Osten M, Natarajan MK, Velianou JL, Martucci G, DeVarennes B, Chisholm R, Peterson MD, Lichtenstein SV, Nietlispach F, Doyle D, DeLarochelliere R, Teoh K, Chu V, Dancea A, Lachapelle K, Cheema A, Latter D, Horlick E. Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk. Acute and late outcomes of the multicenter Canadian experience. J Am Coll Cardiol 2010;55:1080–1090. 5. Gilard M, Eltchaninoff H, Iung B, Donzeau-Gouge P, Chevreul K, Fajadet J, Leprince P, Leguerrier A, Lievre M, Prat A, Teiger E, Lefevre T, Himbert D, Tchetche D, Carrie D, Albat B, Cribier A, Rioufol G, Sudre A, Blanchard D, Collet F, Dos Santos P, Meneveau N, Tirouvanziam A, Caussin C, Guyon P, Boschat J, Le Breton H, Collart F, Houel R, Delpine S, Souteyrand G, Favereau X, Ohlmann P, Doisy V, Grollier G, Gommeaux A, Claudel JP, Bourlon F, Bertrand B, Van Belle E LaskarM, 2 Investigators FRANCE. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med 2012;366 (18):1705–1715. 6. Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, Castella M, Diener HC, Heidbuchel H, Hendriks J, Hindricks G, Manolis AS, Oldgren J, Popescu BA, Schotten U, Van Putte B, Vardas P, Agewall S, Camm J, Baron Esquivias G, Budts W, Carerj S, Casselman F, Coca A, De Caterina R, Deftereos S, Dobrev D, Ferro JM, Filippatos G, Fitzsimons D, Gorenek B, Guenoun M, Hohnloser SH, Kolh P, Lip GY, Manolis A, McMurray J, Ponikowski P, Rosenhek R, Ruschitzka F, Savelieva I, Sharma S, Suwalski P, Tamargo JL, Taylor CJ, Van Gelder IC, Voors AA, Windecker S, Zamorano JL, Zeppenfeld K. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur Heart J 2016;37:2893–2962. 7. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440–1463.
454
The American Journal of Cardiology (www.ajconline.org)
8. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, Brott TG, Cohen DJ, Cutlip DE, van Es GA, Hahn RT, Kirtane AJ, Krucoff MW, Kodali S, Mack MJ, Mehran R, RodesCabau J, Vranckx P, Webb JG, Windecker S, Serruys PW, Leon MB. Valve Academic Research Consortium-2. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Thorac Cardiovasc Surg 2013;145:6–23. 9. Chopard R, Teiger E, Meneveau N, Chocron S, Gilard M, Laskar M, Eltchaninoff H, Iung B, Leprince P, Chevreul K, Prat A, Lievre M, Leguerrier A, Donzeau-Gouge P, Fajadet J, Mouillet G, Schiele F. FRANCE-2 Investigators. Baseline characteristics and prognostic implications of pre-existing and new-onset atrial fibrillation after transcatheter aortic valve implantation: results from the FRANCE-2 registry. JACC Cardiovasc Interv 2015;8:1346–1355. 10. Furuta A, Lellouche N, Mouillet G, Dhanjal T, Gilard M, Laskar M, Eltchaninoff H, Fajadet J, Iung B, Donzeau-Gouge P, Leprince P, Leuguerrier A, Prat A, Dubois-Rande JL, Teiger E. Prognostic value of new onset atrial fibrillation after transcatheter aortic valve implantation: a FRANCE 2 registry substudy. Int J Cardiol 2016;210:72–79. 11. Seet RC, Friedman PA, Rabinstein AA. Prolonged rhythm monitoring for the detection of occult paroxysmal atrial fibrillation in ischemic stroke of unknown cause. Circulation 2011;124:477–486.
12. Kinoshita T, Asai T, Takashima N, Hosoba S, Suzuki T, Kambara A, Matsubayashi K. Preoperative C-reactive protein and atrial fibrillation after off-pump coronary bypass surgery. Eur J Cardiothorac Surg 2011;40:1298–1303. 13. Marott SC, Nordestgaard BG, Zacho J, Friberg J, Jensen GB, Tybjaerg-Hansen A, Benn M. Does elevated C-reactive protein increase atrial fibrillation risk? A Mendelian randomization of 47,000 individuals from the general population. J Am Coll Cardiol 2010;56:789–795. 14. Maan A, Heist EK, Passeri J, Inglessis I, Baker J, Ptaszek L, Vlahakes G, Ruskin JN, Palacios I, Sundt T, Mansour M. Impact of atrial fibrillation on outcomes in patients who underwent transcatheter aortic valve replacement. Am J Cardiol 2015;115:220– 226. 15. Van Gelder IC, Healey JS, Crijns HJGM, Wang J, Hohnloser SH, Gold MR, Capucci A, Lau CP, Morillo CA, Hobbelt AH, Rienstra M, Connolly SJ. Duration of device-detected subclinical atrial fibrillation and occurrence of stroke in ASSERT. Eur Heart J 2017;38(17):1339– 1344. 16. Yankelson L, Steinvil A, Gershovitz L, Leshem-Rubinow E, Furer A, Viskin S, Keren G, Banai S, Finkelstein A. Atrial fibrillation, stroke, and mortality rates after transcatheter aortic valve implantation. Am J Cardiol 2014;114:1861–1866.
Transcatheter aortic valve implantation (TAVI) has become a therapeutic alternative for patients with severe symptomatic aortic stenosis and with a contraindication to a standard surgery, or high-risk surgical procedure. New onset atrial fibrillation (NOAF) is a well-known complication after this procedure. Its occurrence induces an increase of cardiovascular (CV) death and an increase of arterial embolic events. Ischemic strokes’ prevalence is around 4% after TAVI. About 1/3 of cerebrovascular events occur during the TAVI procedure. However, more of 50% of them occur several days after TAVI,1 suggesting that mechanisms other than those directly related to the procedure may be involved. However, in most cases, atrial fibrillation episodes are asymptomatic and could be unknown by the a Department of Cardiology, University Hospital, Dijon, France; and Laboratory of Cerebro-Vascular Pathophysiology and epidemiology (PEC2) EA 7460, University of Burgundy, Dijon, France. Manuscript received January 24, 2018; revised manuscript received and accepted April 13, 2018. This work was supported by the University Hospital of Dijon, the Faculty of Medicine of Dijon, Dijon, France. See page 453 for disclosure information. *Corresponding author: Tel: (+33) 380293311; fax: (+33) 380293333. E-mail address: [email protected] (L. Lorgis). yThese investigators take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. b
0002-9149/© 2018 Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.amjcard.2018.04.029
clinician. Indeed, the large ASSERT prospective study2 found an incidence for silent atrial fibrillation of 10.1% in a cohort of patient without a history of previous atrial fibrillation, 3 months after pacemaker or defibrillators implantation. Few studies have described predictive factors of this arrhythmia. After myocardial infarction (MI), Stamboul et al found that indexed left atrial (LA) diameter and age are independent predictive factors of new onset silent atrial fibrillation.3 In the ASSERT study, investors found that silent atrial fibrillation was associated with a fivefold increased risk of clinical atrial fibrillation. In addition, patients who had silent atrial fibrillation (AF) had twice more risk to have a stroke or an ischemic arterial embolic event during the 2.5 years after. The attributable risk of stroke or systemic embolism associated with silent AF was estimated to 13% in this population. To date, no study has considered the impact and prevalence of silent AF after TAVI procedure. The aim of this prospective study was to assess the incidence, the predictive factors, and the short-term prognosis for the patient who had at least 1 silent atrial fibrillation episode in the 48 hours after TAVI. Methods April 2015 to October 2016, all consecutive patients admitted in intensive care unit after TAVI procedure were www.ajconline.org
Valvular Heart Disease/Silent AF after TAVI
included in this monocentric study conducted in the University Hospital of Dijon. A total of 269 consecutive patients with severe symptomatic aortic stenosis underwent TAVI with a balloon-expandable valve (Edwards SAPIEN, SAPIEN XT, Edwards Lifesciences, Irvine, California) or self-expandable valve Medtronic Corevalve (Medtronic, Inc., Minneapolis, Minnesota) at our institution. Of these, 3 patients with symptomatic AF were excluded, together with 4 who died during the procedure. Details about the TAVI procedure (transfemoral and transapical) have been extensively explained in previous studies.4 Selection of transapical approach was based on the following criteria: small diameter of the iliofemoral arteries, significant peripheral vascular disease, severe calcification or tortuosity of both iliofemoral arteries, severely calcified or porcelain aorta. Patients received full-dose heparin (100 UI/kg, adjusted for an activated clothing time 250 seconds) during the procedure, and aspirin (100 mg/day) + clopidogrel (75 mg/day) were administered following the procedure. The study protocol was approved by the ethics committee of the University Hospital of Dijon, and a written consent form was obtained from the patient or his and/or her legal representative for the France 2 registry inclusion.5 Fifty-six patients with an
447
incomplete data on continuous electrocardiogram monitoring (CEM) 48 hours after TAVI were excluded, leading to a final study population of 206 patients (Figure 1). AF was diagnosed in accordance with the current European Society of Cardiology Guidelines6 as follows: - Absolutely irregular RR intervals; - No distinct P waves; - Atrial cycle length (when visible) 200 ms (atrial rate 300 beats/min). CEM Philips intellivue MP50 was started immediately after admission and recorded continuously for the first 48 hours. The signal was recorded in 2 leads (DII and V1), with a sample frequency of 8 kHz, and a gain resolution of 5 mV/Least Significant Bit. Monitored ECG data were sent to a computer where an unsupervised automated AF episode detection algorithm was applied. Automatic measurements are based on the analysis of RR interval variability, QRS morphology, and P-wave presence screened on 15 beats. Only 15 beats were required to identify AF, and duration of 30 seconds was required to be considered an episode of AF.
Figure 1. Study flow chart.
448
The American Journal of Cardiology (www.ajconline.org)
This monitoring system includes the following rhythm alarms: (1)Flat ECG (2)Ventricular fibrillation (3)Couplets or pulsus bigeminus (4)Supra-ventricular extra-systoles
admission, in the left lateral decubitus position, using a commercially available system (Vivid S6, General ElectricMedical Systems, Horton, Norway). Images were obtained by a simultaneous ECG signal, using a 3.5 MHz transducer, in the parasternal and apical views. Standard M-mode and 2-dimensional images were acquired during breath hold and saved in the cine loop format. The systolic function of LV was assessed using the biplane Simpson method in the apical 4-chamber and apical 2-chamber views according to the American Society of Echocardiography.7 In accordance with the guidelines of the American Society of Echocardiography:
(5)Ventricular tachycardia (6)Preset upper and lower heart rate thresholds (100/min and 40/min) In case of AF episodes, ECG source data were reviewed, and patients were systematically interviewed by the investigator of the study for their symptoms, that is, heart failure, palpitations, or dyspnea related to AF. Symptomatic AF group was defined as patients with AF occurring within the 48 hours after the procedure, confirmed on ECG, resulting in clinical symptoms or the need for urgent cardioversion or need to start curative anticoagulation or amiodarone for patient. These patients were excluded from the statistical analysis. Silent AF group was defined as patients with the occurrence of at least 1 episode of AF 30 seconds occurring after the TAVI procedure, and only detected by CEM, for patient without history of AF before TAVI. So, the study population was divided into 3 groups: history of AF, silent AF, and no new AF. Preprocedural and postprocedural were gathered according to the France 2 registry5 model. Data on demographics, CV risk factors (history of hypertension or treated hypertension, diabetes, history of hypercholesterolemia or treated hypercholesterolemia, current smoking, and history of AF), were collected prospectively, along with admission characteristics and hemodynamic parameters. We defined previous MI as at least 1 MI before the admission. Anthropometric parameters (i.e., body mass index [weight (kg)/height (m2)] were measured within 48 hours of admission. Heart failure, defined as a Killip’s class >1, was also gathered. Long-term and acute (first 48 hours) treatments were also collected. We evaluated risk factors for CV surgery using the logistic European System for Cardiac Operative Risk Evaluation (EuroSCORE), which is calculated by a logistic-regression equation (on a scale from 0% to 100%, with higher scores indicating greater risk and a score of more than 20% indicating very high surgical risk). The CHA2DS2-Vasc, as predictive score of stroke in patients with AF, was also calculated. Blood samples were drawn at admission with assessment of levels of C-reactive protein (CRP), glycaemia, thyroidstimulating hormone, hemoglobin, NT-pro-BNP, lipid profile, peak creatinine kinase, troponin I, serum creatinine clearance were estimated by the Cockcroft and Gault formula. CRP level was dichotomized at 3 mg/L for more clinical relevance. All patients included in the study had undergone an echocardiography within the first 48 hours after their
- LA anteriorposterior diameter was calculated in the parasternal long-axis view, using the M-mode technique by measuring from the trailing edge of the posterior wall of the aortic root to the leading edge of the posterior LA wall. - LA area and volume were calculated by the bi-plane Simpson method in the apical 4-chamber and apical 2chamber views and were indexed to the body surface area. The LA appendage, the pulmonary vein origins, and the funnel of the mitral valve were excluded from measurements. LA area and LA volume were calculated at the end of the T wave on the electrocardiogram (ECG), just before the opening of the mitral valve. Left ventricular ejection fraction (LVEF) was dichotomized at 40% for more clinical relevance. Coronary status was based on pre-TAVI coronarography data. A coronary stenosis was considered as significant if it was at least of 50% stenosed. An invasive mean pulmonary arterial pressure was obtained for almost all patients before TAVI. 30-day and 1-year follow-up (median FU was 280 § 153 days) was obtained for the all cohort except 1 patient who was lost to FU, either by telephone interview or mail to the patients, the patient’s relatives or the treating physician and patient’s medical records review. The following adverse outcomes were assessed according to the Valve Academic Research Consortium Classification II8 and adjudicated by a clinical events committee. For the in-hospital stay, we gathered a combined End point that was defined at hospital discharge and included all-cause mortality, TIA or stroke, and MI (within 72 hours after the index procedure, consisting of at least 1 sample after procedure with a peak value exceeding 15£ as the upper). The data were expressed as percentages (n [%]), median (twenty-fifth to -seventy-fifth percentile) or mean § SD as appropriate. The normality of the distribution of continuous data was tested with the KolmogorovSmirnov test. The categorical variables were analyzed using de chi-square or Fisher’s exact test. Continuous variables were analyzed by ANOVA with F test if normally distributed and with KruskalWallis test otherwise. For 2-group comparisons of non-normally distributed variables, values were tested by either the MannWhitney or Wilcoxon rank sum test for unpaired or paired data, respectively. Estimates of silent AF were studied by backward logistic regression analysis using inclusion and exclusion cutoffs at 5% and 1%, respectively. Variables entered into the multivariate model to
Valvular Heart Disease/Silent AF after TAVI
estimate silent AF were chosen according to their relation in univariate analysis (i.e., logistic EuroSCORE, age, female gender, smoking, hypertension, diabetes, previous CAD, or stroke or AF, LVEF <40%, creatinine clearance, CRP >3, indexed LA area, previous PCI, log NT-ProBNP, admission glycaemia, acute and long-term treatments (b blocker, statin, and angiotensin-converting enzyme [ACE] inhibitor or aspirin). Backward multivariate logistic regression analyses were also used to test the association with inhospital and FU-combined end point, including AF in addition to logistic EuroSCORE and LVEF. Statistical analyses were performed using the SPSS 12.0 software (IBM Inc., Armonk, New York). Results April 2015 to October 2016, 206 patients were included in the study, and only 1 was lost to FU. Total 99 (83.9%) patients stayed in sinus along the monitoring, 88 (42.7%) of them had a preexisting AF (paroxysmal AF = 31; permanent AF = 47; persistent AF but sinus rhythm at the time of the procedure = 10). In the population without a history of AF, 19 (16.1%) had at least 1 episode of silent atrial fibrillation. A comparison of the 3 groups is given in Tables 1 and 2. Despite we found no statistical differences regarding demographics and CV risk factors, patients diagnosed with a silent AF after TAVI have a lower logistic EuroSCORE
449
and higher CRP and admission glycaemia levels when compared with others groups. Moreover, clinical presentation, hemodynamic data on admission, and long-term treatments were similar for the 3 groups. Echocardiographic parameters clearly show a higher LA remodeling between AF groups (silent or history of AF) compared with the no AF groups. However, we found no statistical differences when we compared LA parameters between AF groups (silent vs history of AF) (Table 2). Of note, the proportion of patients with a LVEF <40% was similar in the 3 groups. Procedural details and success were strictly identical among groups. The hemodynamic parameters on CEM during the first 48 hours phase were similar in the 3 groups, whatever the FA occurrence. Among the acute therapies, there were no significant differences on b blockers and ACE inhibitors. However, statins were significantly less prescribed in the silent AF group. In the history of AF group, patients were more often on VKA (83%) and amiodarone (25%) (Table 3). In-hospital and FU-combined end point are described in Table 4. Despite we found differences in the frequency of post-TAVI MI and severe arrhythmias, we did not show any variance on the composite in-hospital end point between groups. However, during the FU period, the population with pre-existing AF had a worse short-term prognosis as witnessed by an increased efficacy end point, which is significantly higher than patients unknown for AF before
Table 1 Baseline characteristics of the study population underwent transcatheter aortic valve procedure according to AF occurrence. n(%) or median (twenty-fifth to seventy-fifth) Variable Age (years) Women Hypertension Diabetes mellitus History of CAD Chronic pulmonary disease Prior MI Prior bypass graft BMI (kg/m2) Peripheral vascular disease History of stroke Permanent pacemaker Renal failure (DFG <60) NYHA 3-4 LVEF <40 % Complete revasc Log EuroSCORE (% § SD) CHA2DS2-VASC Score Biological data CRP 3 mg/l Hemoglobin Platelets Glucose, mmol/l Nt-proBNP
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
86 (81-89) 105 (51%) 132 (64%) 56 (27%) 123 (60%) 10 (5%) 7 (3%) 17 (8%) 26.0 (24-29) 33 (16%) 24 (11%) 23 (11%) 103 (51%) 143 (69%) 37 (18%) 38 (18%) 19.8 § 11.6 4.49 § 0.6
86 (82-89) 44 (50%) 58 (66%) 22 (25%) 58 (66%) 5 (6%) 4 (4%) 7 (8%) 26.0 (24.2-29.8) 14 (16%) 10 (11%) 10 (11%) 50 (57%) 65 (73%) 21 (24%) 15 (17%) 22.5 § 12.6*,y 4.52 § 0.11
86 (81-88) 52 (53%) 61 (61%) 28 (28%) 57 (57%) 4 (4%) 3 (3%) 7 (7%) 25.8 (23.4-28.6) 17 (17%) 12 (12%) 12 (12%) 45 (47%) 68 (68%) 13 (13%) 19 (19%) 18.4 § 11.0 4.47 § 0.08
87 (80-88) 9 (48%) 13 (69%) 6 (32%) 8 (42%) 1 (4%) 0 (0%) 3 (16%) 26 (24.3-28.7) 2 (10%) 2 (10%) 1 (5%) 8 (42%) 10 (52%) 3 (16%) 4 (2%1) 14.8 § 6 4.47 § 0.17
0.299 0.892 0.762 0.795 0.133 0.870 0.588 0.445 0.727 0.769 0.974 0.683 0.258 0.186 0.157 0.910 0.010 0.937
71 (34%) 12.1 (10.8-13.2) 206 (168-258) 5.56 (4.81-6.53) 2890 (1211-6287)
34 (39%) 12.1 (10.7-13.2) 201 (154-264) 5.68 (5-6.52)y 3804 (1988-9699)*
29 (29%) 12.2 (10.8-13.2) 207 (168-256) 5.34 (4.65-6.45) 1877 (785-4459)
8 (42%)z 11.9 (11-13.3) 212 (170-240) 6.22 (5.08-11.48)z 2846 (764-4744)
0.065 0.697 0.764 <0.001 0.045
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z : No new AF versus silent AF. *
450
The American Journal of Cardiology (www.ajconline.org)
Table 2 Baseline echocardiographic and procedural data of the study population according to AF occurrence Variable Echographic LVEF (%) Aortic valve area (cm2 § SD) Mean gradient (mm Hg § SD) Indexed LA diam (cm2/m2) Indexed LA area (cm2/m2) Indexed LA Vol (cm3/m2) MR >grade 2 Mean invasive PAP >35 mm Hg Procedure Percutaneous femoral approach Chirurgical approach Conscious sedation Edwards SAPIEN Medtronic CoreValve Contrast volume (ml § SD) Procedural success
Overall (n = 206)
History of AF (n = 88)
No New AF (n = 99)
Silent AF (n = 19)
p
55 (45-60) 0.66 § 0.17 46 § 14 3 (2.5-3.7) 14 (11.6-17) 47 (34-58) 8 (4%) 96 (47%)
55 (45-60) 0.66 § 0.17 43 § 14*,y 3.3 (2.8-3.85) 15 (12.5-19) 51 (39-68)* 5 (6%) 48 (54%)*
55 (50-60) 0.66 § 0.18 47 § 15 2.3 (2.4-3.65) 13 (11-15.7) 40 (30-541) 2 (2%) 40 (40%)
55 (50-60) 0.67 § 0.15 52 § 11 3.05 (2.3-3.7) 14.3 (13-16) 49 (44-57) 1 (5%) 8 (42%)
0.263 0.909 0.019 0.148 0.006 <0.001 0.411 0.141
199 (97%) 7 (3%) 194 (94%) 135 (66%) 71 (34%) 142 § 56 203 (98%)
85 (97%) 3 (3%) 81 (92%) 61 (69%) 27 (30%) 135 § 50y 87 (99%)
96 (97%) 3 (3%) 95 (96%) 63 (64%) 36 (36%) 143 § 61 97 (98%)
18 (95%) 1 (5%) 18 (95%) 11 (58%) 8 (42%) 167 § 47 19 (100%)
0.886 0.886 0.519 0.573 0.547 0.095 0.755
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z: No new AF versus silent AF. *
TAVI: 50% versus 28% versus 16%, respectively (p = 0.001) (Figure 2). Among outcomes gathered, there was a significant higher rate of CV death: 16% versus 5% versus 0%, (p = 0.013), together with a higher rate of rehospitalization, especially for heart failure symptoms: 22% versus 6% versus 0%, (p <0.001). Albeit these results in the history of AF group were expected, the difference of prognosis seen between groups at 30 days FU was reinforce at 1-year FU, without no difference of adverse outcomes between the silent AF groups versus the no new AF group.
By univariate analysis, predictive factors were associated with the occurrence of silent AF were CRP >3 mg/dl and glycaemia on admission (odds ratio [OR] 1.21, 95% confidence intervals [CIs]: 0.99 to 1.07, p = 0.032). There was a trend for the logistic EuroSCORE (OR 0.90, 95% CI: 0.31 to 1.34, p = 0.098) and the mean aortic gradient pre-TAVI (OR 1.03, 95% CI: 0.99 to 1.07, p = 0.057). By multivariate logistic regression analysis, only the CRP >3 mg/dl (OR 10.84 95% CI: 1.37 to 85.9, p = 0.05) and the logistic EuroSCORE (OR 0.94, 95% CI: 0.88 to 1.00, p = 0.05) were
Table 3 Continuous scope data within first 48 hours and discharge therapies of the study population according to AF occurrence Variable Day 1 HR median, beat/min HR max, beat/min HR min, beat/min Number of SVES episodes Day 2 HR median, beat/min HR max, beat/min HR min, beat/min Number of SVES episodes Acute therapies Aspirin Beta blockers ACE inhibitor Statin Vitamin K antagonist Amiodarone p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z : No new AF versus silent AF. *
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
71 (63-83) 92 (80-107) 60 (54-71) 4.5 (0-15)
72 (62-85) 92 (79-108) 63 (56-76)* NA
71 (63-85) 89 (80-105) 58 (54-68) 2 (0-8)
71 (62-80) 105 (83-119)z 61 (55-63) 7 (3-13)
0.612 0.223 0.069 <0.001
71 (66-83) 99 (85-112) 62 (56-70) 5 (1-19)
71 (64-84) 97 (82-114) 62 (53-71) 6 (2-109)*
72 (67-80) 100 (86-112) 62 (57-69) 4 (1-18)
73 (64-89) 101 (87-110) 65 (560-82) 3 (1-6)
0.866 0.999 0.323 0.072
164 (82%) 117 (57%) 140 (68%) 115 (56%) 88 (43%) 26 (13%)
62 (74%)* 48 (55%) 59 (67%) 45 (51%) 73 (83%)*,y 22 (25%)*,y
87 (90%) 57 (57%) 70 (70%) 63 (63%) 14 (14%) 4 (4%)
15 (83%) 12 (63%) 11 (56%) 7 (37%)z 1 (5%) 0 (0%)
0.020 0.771 0.532 0.050 <0.001 <0.001
Valvular Heart Disease/Silent AF after TAVI
451
Table 4 In-hospital and follow-up outcomes to transcatheter aortic valve procedure according to AF occurrence
In hospital Hospital stay (days) Death Composite End point Stroke Myocardial infarction Vascular complication Major bleeding New pacemaker Severe arrhythmia Valvular complication Aortic valve area (cm2 § SD) Mean gradient (mm Hg § SD) 30 days follow-up Efficacy end point Death all cause CV death Stroke or arterial embol Major Bleeding Acute Kidney injury 2 or 3 Rehosp. (heart failure) 1-year follow-up Efficacy End point Death all cause CV death Stroke or arterial embol Major Bleeding Acute Kidney injury 2 or 3 Rehosp. (heart failure)
Overall (n = 206)
History of AF (n = 88)
No new AF (n = 99)
Silent AF (n = 19)
p
11 § 0.44 6 (2%) 96 (46%) 5 (2%) 41 (20%) 17 (8%) 3 (1.4%) 43 (20%) 8 (4%) 5 (2.4%) 1.72 § 0.08 8.4 § 0.425
12.2 § 0.78* 4 (5%) 41 (47%) 2 (2%) 13 (15%)*,y 8 (9%) 1 (1%) 20 (23%) 2 (2%)* 2 (2%) 1.74 § 0.06 7.5 § 0.42*
9.8 § 0.5 2 (2%) 47 (47%) 2 (2%) 24 (24%) 7 (7%) 2 (2%) 19 (19%) 5 (5%) 3 (3%) 1.72 § 0.08 9.5 § 0.08
10.3 § 1.56 0 (0%) 8 (42%) 1 (5%) 4 (21%) 2 (10%) 0 (0%) 4 (21%) 1 (5%) 0 (0%) 1.70 § 0.30 6.6 § 0.11
0.031 0.432 0.912 0.697 0.267 0.821 0.755 0.838 0.586 0.683 0.977 0.027
51 (24%) 14 (6.7%) 13 (6.2%) 6 (2.9%) 6 (2.9%) 6 (2.9%) 14 (6.7%)
30 (34%)*,y 8 (9%) 10 (11.5%)* 4 (4.7%) 1 (1.2%) 3 (3.5%) 11 (13%)*,y
19 (19%) 5 (5%) 3 (3%) 2 (2%) 5 (5%) 2 (%2) 3 (3%)
2 (10%) 1 (5%) 0 (0%) 0 (0%) 0 (0%) 1 (4%) 0 (0%)
0.020 0.528 0.032 0.432 0.207 0.695 0.017
75 (36%) 40 (19%) 19 (9%) 8 (4%) 9 (4%) 2 (1%) 25 (12%)
44 (50%)*,y 27 (30%) 14 (16%)* 2 (2%) 4 (4%) 1 (1%) 19 (22%)*,y
28 (28%) 12 (12%) 5 (5%) 5 (5%) 5 (5%) 1 (1%) 6 (6%)
3 (16%) 1 (5%) 0 (0%) 1 (5%) 0 (0%) 1 (1%) 0 (0%)
0.001 0.002 0.013 0.586 0.611 0.899 <0.001
p <0.05 for : History of AF versus No new AF; y : History of AF versus Silent AF; z: No new AF versus silent AF. *
Figure 2. KaplanMeier survival curves for 1-year all-cause and CV mortality, stratified by AF status.
452
The American Journal of Cardiology (www.ajconline.org)
Table 5 Uni and multivariate analysis for predictors of silent AF after TAVI procedure Univariate variable Euroscore logistic Glucose on admission Mean aortic gradient pre TAVI CRP >3 mg/l
Multivariate Odds ratio (95% CI)
p
0.20 (0.31-1.34) 1.21 (1.02-1.44) 1.03 (0.99-1.07) 10.8 (1.38-84)
0.098 0. 032 0.057 0.023
Figure 3. Occurrence of type of AF according to the tertiles of CRP.
independent predictive factors of silent AF after TAVI (Table 5). After dichotomization in tertiles of CRP >3 mg/ dl, we found a linear relation with increasing occurrence of silent AF among the levels of CRP (Figure 3).
Discussion With an incidence of 16.1%, we found that silent AF is not rare in TAVI population. We have identified 2 independent predictive factors for silent AF after TAVI occurrence: a pre-TAVI CRP >3 mg/dl and a trend for the logistic EuroSCORE. Our result is similar with other studies: from France 2 registry, Chopard et al9 who tracked NOAF after TAVI with repeat ECGs found an incidence of 6.0%. Furuta et al10 and Amat-Santos et al,1 who tracked NOAF on CEM until hospital discharge, found an incidence of 7.6% and 31.9%, respectively. We shared the same method as the last 2 studies to detect silent AF, but our time window was shortened (48 hours). Obviously, result of the screening of AF strongly depends of the method used for detection NOAF, and the duration of the monitoring after TAVI procedure.11 Clinical independent predictive factors of AF occurrence after TAVI are very heterogeneous in literature. Despite our logistic regression, to predict silent AF, we have not highlighted a single clinical risk factor. In our study, the silent AF group had a lower logistic EuroSCORE than the permanent sinus rhythm: 14.8 § 6.0 versus 18.4 § 11.0, respectively.
variable
Odds ratio (95% CI)
p
Euroscore logistic
0.94 (0.88-1.00)
0.05
CRP >3 mg/l
10.84 (1.37-85.9)
0.024
A smaller study,1 with 138 patients, suggested that atrial size is associated with AF, with a proposed cutoff of indexed atrial diameter at 27 mm/m2 to predict atrial fibrillation occurrence. In our study, despite we found that patients with silent AF had a significant higher LA volume than controls, this was not found as a predictive factor in univariate and multivariate analysis. Finally, our study suggests that CRP is a strong predictive factor of silent atrial fibrillation in TAVI population. Kinoshita et al12, in a cohort of 656 patients who underwent off-pump coronary bypass, found that a high preoperative CRP increased atrial fibrillation risk in the 7 postoperative days. The investigators found a hazard ratio at 3.19 for the association of CRP >3 to 10 mg/L and postoperative atrial fibrillation. If CRP was 1 to 3 mg/L, hazard ratio decreased at 1.59. They proposed a cutoff point of 2.1 mg/L: patients with CRP >2.1 mg/L had a twofold increased risk of developing AF after surgery. In a large Danish prospective cohort of 46,000 patients,13 CRP was found as an independent predictive factor of NOAF. Patients were genotyped for 4 CRP gene polymorphisms associated with up to a 63% increase in plasma CRP levels. When the risk of atrial fibrillation was adjusted to CRP genotype and genotype combinations, CRP was no more associated with an increased risk of atrial fibrillation. That is demonstrating that this is the long-term inflammatory response that causes atrial fibrillation. Finally, TAVI with the TF approach results in a lower activation of inflammatory pathways than SAVR. Despite an attenuate inflammatory pathway, TAVI remains a physiological stress causing a significant inflammatory response that can act as a trigger for an AF occurrence. As expected, in the context of TAVI, a pre-existing AF was associated with increased composite end point during the in-hospital and the follow-up period, and a poor prognosis. Pre-existing AF has been previously linked with adverse ischemic cardiac and cerebrovascular events and mortality after TAVI (9 and 10). We also found a significant higher incidence of CV death (11.5%) and rehospitalization for heart failure (13%). Pre-existing AF is a strong risk marker in this population. Despite we previously showed, on a population of acute MI (AMI), that silent or symptomatic AF, was associated with higher hospital and 1-year mortality3, we did not find any difference in term of outcomes when we compared the silent atrial fibrillation group with the control group in the context of TAVI procedures. NOAF appears to be a risk marker of adverse events, but seems to be less pejorative than pre-existing AF. The different studies who analyzed NOAF after TAVI found a worse 1-month and 1-year prognosis than patients with
Valvular Heart Disease/Silent AF after TAVI
permanent sinus rhythm (9 and 10). However, in these studies, investigators did not make difference between silent NOAF and symptomatic NOAF. So, we have sought to refine the knowledge of the NOAF after TAVI. Our hypothesis that silent AF can help to stratify as an intermediaterisk population is not confirmed by our preliminary data. One can argue that this is not the silent AF occurrence that is associated with adverse outcomes, but the silent atrial fibrillation burden. We can also presume that the risk of death or adverse events and their competing associations are likely to be very different in this population of elderly and frail patients from in the settings of AMI. Thus, the rate of adverse events in the control group of SR patients is 5 to 6 times higher than the same population of AMI patients, but the TAVI group was 24 years older. Of note, our median follow-up is short, but our results are in line with previous published studies that did not find a difference of event at 1-month follow-up1416 after TAVI. Moreover, we clearly have a very low number of adverse events after TAVI in the silent AF group. Indeed, we recently learned from the ASSERT population, that more than the presence or the absence of silent atrial fibrillation, that atrial fibrillation burden greater than 24 hours was a powerful prognostic factor, and associated with a significant increase in the rate of ischemic stroke or systemic embolism.15 Interestingly, such relation was not observed when the duration of silent atrial fibrillation was 6 minutes to 24 hours. The elderly patient population underwent TAVI is predisposed to both thromboembolic and bleeding complication. The question of the prognosis raises the question of the therapeutic attitude to be adopted when such arrhythmia is detected. Although at least 3 ongoing randomized controlled trials (ARTE, ATLANTIS, and GALILEO) and a registry-based prospective study (POPular-TAVI) are evaluating the optimal antithrombotic or antiplatelet strategies after TAVI, none of them have specifically stratified patients based on pre-existing AF or NOAF. Our study had several limits. First, our work is monocentric, and size of our cohort is small. Second, we have collected adverse events during the year after the procedure and results could be very different from a long-term FU. Also, we have excluded the patients with the major procedural complications who require reanimation or and those who died during the procedure. In our study, the distinction between symptomatic and silent AF was defined according to the European Society of Cardiology guidelines,6 and was based on whether AF symptoms were present or not. This definition was therefore not absolutely objective and may over diagnosed asymptomatic arrhythmias. To correct this possible effect, we add to the silent AF definition the careful screening of drugs received during the acute phase of monitoring after TAVI, especially use of anticoagulation and antiarrhythmic. In fact, the incidence of symptomatic AF found in our study (1.4%) is lower than the other studies.3 The possible non-negligible portion of false positive of detection of atrial fibrillation, found in the others studies (5% to 12%), has been corrected by the systematic realization of a 12 leads ECG or a systematic analysis of the continuous monitoring. Silent atrial fibrillation is not
453
rare in patient who underwent TAVI. CRP >3 mg/dl was found as independent predictive factor. In clinical practice, CRP on admission can be determined easily and may be useful for noninvasive risk stratification of developing silent AF after TAVI. Silent atrial fibrillation was not associated with a poor prognosis but we must carry out a larger study because the therapeutic impact may be greater with the necessity or not to start a curative anticoagulation in this frail population. Acknowledgment We wish to thank Anais Hamon, Florence Bichat, and Maud Maza for their research assistance and Philip Bastable for editorial assistance. Disclosures The authors have no conflicts of interest to disclose. 1. Amat-Santos IJ1, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, Villeneuve J, C^ote M, Nombela-Franco L, Philippon F, Pibarot P, Dumont E. Incidence, predictive factors, and prognostic value of new-onset atrial fibrillation following transcatheter aortic valve implantation. J Am Coll Cardiol 2012;59(2):178–188. 2. Healey JS, Connolly SJ, Gold MR, Israel CW, Van Gelder IC, Capucci A, Lau CP, Fain E, Yang S, Bailleul C, Morillo CA, Carlson M, Themeles E, Kaufman ES, Hohnloser SH. ASSERT Investigators. Subclinical atrial fibrillation and the risk of stroke. N Engl J Med 2012;366 (2):120–129. 3. Stamboul K, Zeller M, Fauchier L, Gudjoncik A, Buffet P, Garnier F, Guenancia C, Lorgis L, Beer JC, Touzery C, Cottin Y. Prognosis of silent atrial fibrillation after acute myocardial infarction at 1-year follow-up. Heart 2015;101(11):864–869. 4. Rodes-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, Osten M, Natarajan MK, Velianou JL, Martucci G, DeVarennes B, Chisholm R, Peterson MD, Lichtenstein SV, Nietlispach F, Doyle D, DeLarochelliere R, Teoh K, Chu V, Dancea A, Lachapelle K, Cheema A, Latter D, Horlick E. Transcatheter aortic valve implantation for the treatment of severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk. Acute and late outcomes of the multicenter Canadian experience. J Am Coll Cardiol 2010;55:1080–1090. 5. Gilard M, Eltchaninoff H, Iung B, Donzeau-Gouge P, Chevreul K, Fajadet J, Leprince P, Leguerrier A, Lievre M, Prat A, Teiger E, Lefevre T, Himbert D, Tchetche D, Carrie D, Albat B, Cribier A, Rioufol G, Sudre A, Blanchard D, Collet F, Dos Santos P, Meneveau N, Tirouvanziam A, Caussin C, Guyon P, Boschat J, Le Breton H, Collart F, Houel R, Delpine S, Souteyrand G, Favereau X, Ohlmann P, Doisy V, Grollier G, Gommeaux A, Claudel JP, Bourlon F, Bertrand B, Van Belle E LaskarM, 2 Investigators FRANCE. Registry of transcatheter aortic-valve implantation in high-risk patients. N Engl J Med 2012;366 (18):1705–1715. 6. Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, Castella M, Diener HC, Heidbuchel H, Hendriks J, Hindricks G, Manolis AS, Oldgren J, Popescu BA, Schotten U, Van Putte B, Vardas P, Agewall S, Camm J, Baron Esquivias G, Budts W, Carerj S, Casselman F, Coca A, De Caterina R, Deftereos S, Dobrev D, Ferro JM, Filippatos G, Fitzsimons D, Gorenek B, Guenoun M, Hohnloser SH, Kolh P, Lip GY, Manolis A, McMurray J, Ponikowski P, Rosenhek R, Ruschitzka F, Savelieva I, Sharma S, Suwalski P, Tamargo JL, Taylor CJ, Van Gelder IC, Voors AA, Windecker S, Zamorano JL, Zeppenfeld K. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur Heart J 2016;37:2893–2962. 7. Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440–1463.
454
The American Journal of Cardiology (www.ajconline.org)
8. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, Brott TG, Cohen DJ, Cutlip DE, van Es GA, Hahn RT, Kirtane AJ, Krucoff MW, Kodali S, Mack MJ, Mehran R, RodesCabau J, Vranckx P, Webb JG, Windecker S, Serruys PW, Leon MB. Valve Academic Research Consortium-2. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Thorac Cardiovasc Surg 2013;145:6–23. 9. Chopard R, Teiger E, Meneveau N, Chocron S, Gilard M, Laskar M, Eltchaninoff H, Iung B, Leprince P, Chevreul K, Prat A, Lievre M, Leguerrier A, Donzeau-Gouge P, Fajadet J, Mouillet G, Schiele F. FRANCE-2 Investigators. Baseline characteristics and prognostic implications of pre-existing and new-onset atrial fibrillation after transcatheter aortic valve implantation: results from the FRANCE-2 registry. JACC Cardiovasc Interv 2015;8:1346–1355. 10. Furuta A, Lellouche N, Mouillet G, Dhanjal T, Gilard M, Laskar M, Eltchaninoff H, Fajadet J, Iung B, Donzeau-Gouge P, Leprince P, Leuguerrier A, Prat A, Dubois-Rande JL, Teiger E. Prognostic value of new onset atrial fibrillation after transcatheter aortic valve implantation: a FRANCE 2 registry substudy. Int J Cardiol 2016;210:72–79. 11. Seet RC, Friedman PA, Rabinstein AA. Prolonged rhythm monitoring for the detection of occult paroxysmal atrial fibrillation in ischemic stroke of unknown cause. Circulation 2011;124:477–486.
12. Kinoshita T, Asai T, Takashima N, Hosoba S, Suzuki T, Kambara A, Matsubayashi K. Preoperative C-reactive protein and atrial fibrillation after off-pump coronary bypass surgery. Eur J Cardiothorac Surg 2011;40:1298–1303. 13. Marott SC, Nordestgaard BG, Zacho J, Friberg J, Jensen GB, Tybjaerg-Hansen A, Benn M. Does elevated C-reactive protein increase atrial fibrillation risk? A Mendelian randomization of 47,000 individuals from the general population. J Am Coll Cardiol 2010;56:789–795. 14. Maan A, Heist EK, Passeri J, Inglessis I, Baker J, Ptaszek L, Vlahakes G, Ruskin JN, Palacios I, Sundt T, Mansour M. Impact of atrial fibrillation on outcomes in patients who underwent transcatheter aortic valve replacement. Am J Cardiol 2015;115:220– 226. 15. Van Gelder IC, Healey JS, Crijns HJGM, Wang J, Hohnloser SH, Gold MR, Capucci A, Lau CP, Morillo CA, Hobbelt AH, Rienstra M, Connolly SJ. Duration of device-detected subclinical atrial fibrillation and occurrence of stroke in ASSERT. Eur Heart J 2017;38(17):1339– 1344. 16. Yankelson L, Steinvil A, Gershovitz L, Leshem-Rubinow E, Furer A, Viskin S, Keren G, Banai S, Finkelstein A. Atrial fibrillation, stroke, and mortality rates after transcatheter aortic valve implantation. Am J Cardiol 2014;114:1861–1866.