- Email: [email protected]
Impact of Atrial Fibrillation on Outcomes in Patients Who Underwent Transcatheter Aortic Valve Replacement
Impact of Atrial Fibrillation on Outcomes in Patients Who Underwent Transcatheter Aortic Valve Replacement Abhishek Maan, MD, E. Kevin Heist, MD, PhD, Jonathan Passeri, MD, Ignacio Inglessis, MD, Joshua Baker, MD, Leon Ptaszek, MD, PhD, Gus Vlahakes, MD, Jeremy N. Ruskin, MD, Igor Palacios, MD, Thoralf Sundt, MD, and Moussa Mansour, MD* Transcatheter aortic valve replacement (TAVR) has emerged as an alternative treatment for surgical high-risk patients with severe aortic stenosis. The aim of this study was to determine the impact of atrial fibrillation (AF) on procedural outcomes. Data from 137 patients who underwent TAVR using Edwards SAPIEN valve were reviewed. The predictors of new-onset atrial fibrillation (NOAF) after the procedure were analyzed. In addition, the post-TAVR clinical outcomes and adverse events were compared according to the presence and absence of preprocedural and postprocedural AF. Previous AF was present in 49% of the patients who underwent TAVR. After the procedure, NOAF was detected in 21% of patients, and the cumulative incidence of post-TAVR AF was 60%. After TAVR, 50% of all the episodes of NOAF occurred in the initial 24 hours after the procedure. Transapical approach was observed to an important predictor of NOAF (adjusted odds ratio [OR] 5.05, 95% confidence interval [CI] 1.40 to 18.20, p [ 0.013). The composite outcome of all-cause mortality, stroke, vascular complications, and repeat hospitalization in 1 month after TAVR was significantly higher in patients with previous AF (33 of 67 vs 19 of 70, adjusted OR 2.60, 95% CI 1.22 to 5.54, p [ 0.013) compared with patients who did not have previous AF. The presence of post-TAVR AF led to a prolongation in the duration of intensive care unit stay by an average of 70 hours (95% CI 25 to 114.7 hours, p [ 0.002). Similarly, post-TAVR AF also led to the prolongation in the hospital stay by an average of 6.7 days (95% CI 4.69 to 8.73 days, p <0.0005). In conclusion, our study demonstrates that the presence of AF before TAVR is an important predictor of the composite end point of all-cause mortality, stroke, vascular complications, and repeat hospitalization in 1 month after the procedure. AF after TAVR is more likely to be encountered with the transapical approach and is associated with a prolongation of intensive care unit and hospital stay. Ó 2015 Elsevier Inc. All rights reserved. (Am J Cardiol 2015;115:220e226) Transcatheter aortic valve replacement (TAVR) is a relatively new procedure that has evolved as an alternative treatment option for patients with severe aortic stenosis (AS) who are at a high surgical risk.1e3 Atrial fibrillation (AF) is the most common cardiac arrhythmia with an increased prevalence especially in the elderly population.4,5 Because of the relatively older age of patients with severe AS, the prevalence of AF in this patient population tends to be higher. The presence of severe AS also leads to left ventricular hypertrophy and a state of an increased afterload; both these pathophysiological processes also mediate the development of AF.6e8 Both the presence and development of AF have been observed to be associated with a higher incidence of adverse clinical outcomes and increased mortality in the setting of surgical aortic valve replacement and cardiothoracic Heart Center, Massachusetts General Hospital, Boston, Massachusetts. Manuscript received August 14, 2014; revised manuscript received and accepted October 14, 2014. See page 225 for disclosure information. The work was supported in part by the Deane Institute for Integrative Research in Atrial Fibrillation and Stroke at Massachusetts General Hospital. *Corresponding author: Tel: (617) 726-5557; fax: (617) 724-1241. E-mail address: [email protected] (M. Mansour). 0002-9149/14/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.10.027
surgeries.9e12 AF carries an increased risk of thromboembolic and vascular complications compared with patients in sinus rhythm (SR).13 AF also leads to an atrioventricular dyssynchrony, which further adversely affects cardiac function in AS.14 The aim of this study was to determine the impact of AF on procedural outcomes. Methods The study population consisted of 137 consecutive patients with severe AS who under underwent TAVR at Massachusetts General Hospital between June 2008 and October 2012. The procedure was performed using the balloon-expandable valve (Edwards SAPIEN, SAPIEN XT; Edward Lifesciences, Irvine, California). The preprocedural risk was calculated using the Society of Thoracic Surgeons risk score for Prediction of Mortality15 and the logistic EuroSCORE.16,17 The study design is briefly explained in Figure 1. Data on baseline characteristics on all patients were collected retrospectively. Preprocedural and postprocedural electrocardiograms (ECGs) were analyzed for rhythm, various intervals, and conduction abnormalities. Data on various echocardiographic parameters were obtained from the transthoracic echocardiograms that were obtained before and after TAVR. www.ajconline.org
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
221
Figure 1. Study design and various study subsets.
Patients were evaluated for the presence of AF before the procedure. The presence of AF was confirmed on ECGs and the documentation from inpatient and outpatient medical records. Pre-TAVR rhythm was documented on all patients who underwent the procedure. All the study patients were monitored by continuous telemetry until the day of discharge. New-onset atrial fibrillation (NOAF) was defined as any episode of AF lasting >30 seconds in the patients without any history of this arrhythmia.18,19 In the patients who developed NOAF, collected data included the timing of arrhythmia occurrence, requirement of cardioversion, and the use antithrombotic therapy. The patients were divided into 2 groups: post-TAVR AF (all patients with AF 30 seconds post-TAVR) and post-TAVR SR. Univariate analyses were performed for comparison between various subgroups (previous AF vs no previous AF, post-TAVR AF vs post-TAVR SR). Categorical variables between the 2 subgroups were compared using the Fisher’s exact test, and continuous variables were compared using the 2-tailed unpaired Student’s t test. After the univariate analyses, 4 logistic regression models were prepared for each of the adverse outcomes (NOAF, major vascular complication, composite outcome of all-cause mortality, stroke, vascular complications, and repeat hospitalization in 1 month after TAVR, and all-cause 1-year mortality). We also prepared 2 linear regression models (each for duration of intensive care unit [ICU] stay and duration of hospital stay, respectively) to investigate the impact of post-TAVR AF on these outcomes. The logistic regression models were all tested for the “goodness of fit” of the logistic model with the HosmereLemeshow test. All statistical tests were performed with STATA 9.2 software (Stata Corp LP, College Station, TX). Results The median age was 85 years (range 55 to 96). Previous AF was present in 67 patients (49%). There was no difference
in the baseline clinical variables in the patients with or without previous AF, except that the use of warfarin was significantly more common in the patients with previous AF (59.7% vs 1.4%; p <0.0001). The baseline characteristics of study patients and the comparison between the patients according to the presence of previous AF are listed in Table 1. ECG performed at the time of TAVR revealed the presence of AF in 60% (40 of 67) of the patients with previous AF. The comparison of various echocardiographic parameters between the 2 groups of patients (previous AF vs without previous AF) is listed in Table 1. TAVR was performed using the Edwards SAPIEN (Edwards Lifesciences) balloonexpandable prosthesis with the 2 available sizes (23 and 26 mm). Transapical approach was used in most patients (81 of 137; 59%). To determine the incidence of NOAF, the 67 patients with a history of AF were excluded. In the remaining 70 patients, NOAF occurred in 21 patients (21 of 70; 30%). A total of 50% of all the episodes of NOAF occurred in the initial 24 hours after TAVR (Figure 2). Electrical cardioversion was performed in 3 patients who developed NOAF, and 8 patients converted from AF back to sinus rhythm after receiving amiodarone. The comparison of baseline clinical, echocardiographic, and procedural characteristics of the patients categorized according to the occurrence of NOAF is listed in Table 2. In the multivariate analysis, transapical approach of TAVR was observed to be an important predictor of NOAF after the procedure (adjusted odds ratio [OR] 5.05, 95% confidence interval [CI] 1.40 to 18.20; p ¼ 0.013). The overall 1-year mortality in this study was 12% (16 patients). The standard definition of “improvement in functional status and health-related quality of life after the procedure” was used for determining the success of TAVR.2,20,21 According to this definition, the procedural success of TAVR was 95%. Based on the predischarge echocardiogram, there was a significant decrease in the mean (50.73 16.29 to 11.67 4.52 mm Hg; p <0.001) and peak
222
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of the study population and comparison according to the presence of prior AF Baseline variables Age (years) Male Body mass index (kg/m2) Active smokers STS-PROM score (%) Logistic EuroSCORE (%) Hypertension Hyperlipidemia Diabetes Mellitus Coronary artery disease Congestive heart failure Chronic obstructive pulmonary disease Peripheral vascular disease Pulmonary hypertension Serum creatinine (mg/dl) Cerebrovascular disease Carotid artery disease Previous Procedural Percutaneous coronary intervention Prior pacemaker implantation Balloon valvuloplasty Coronary artery bypass grafting Prior Medication h/o Aspirin Clopidogrel Warfarin Beta-blockers Diuretics ACEi/ARBs Statins Echocardiographic parameters Ejection fraction (%) LVEF < 40% LV EDD (mm) Severe mitral regurgitation Moderate mitral regurgitation Mean aortic gradient (mm Hg) Peak aortic gradient (mm Hg) Left atrial size (mm) Interventricular septum (mm) Aortic valve area (cm2)
Study population (N ¼ 137)
No Prior AF (N ¼ 70)
Prior AF (N ¼ 67)
P value
84.18 6.83 65 (47%) 26.70 5.90 7 (5%) 6.88 3.82 14.33 12.24 109 (80%) 95 (69%) 47 (34%) 99 (72%) 74 (54%) 39 (28%) 42 (31%) 38 (28%) 1.33 0.47 25 (18%) 31 (23%)
83.92 6.96 30 (43%) 26.40 5.87 5 (7%) 6.49 3.18 13.67 12.01 54 (77%) 49 (70%) 27 (39%) 48 (69%) 34 (39%) 18 (26%) 24 (34%) 14 (20%) 1.32 0.49 15 (21%) 20 (29%)
84.46 6.72 35 (52%) 27 5.94 2 (3%) 7.29 4.38 15.02 12.52 55 (82%) 46 (69%) 20 (40%) 51 (76%) 40 (60%) 21 (31%) 18 (27%) 24 (36%) 1.34 0.45 10 (15%) 11 (16%)
0.65 0.31 0.55 0.44 0.23 0.52 0.53 1.00 0.37 0.35 0.23 0.57 0.36 0.06 0.80 0.38 0.10
51 27 24 55
(37%) (20%) (18%) (40%)
23 12 13 24
(33%) (17%) (19%) (34%)
28 15 11 31
(42%) (22%) (16%) (46%)
0.29 0.52 0.83 0.17
113 28 41 90 108 55 106
(82%) (20%) (30%) (66%) (79%) (40%) (79%)
64 16 1 44 56 27 55
(91%) (23%) (1%) (63%) (80%) (39%) (79%)
49 12 40 46 52 28 51
(73%) (18%) (60%) (69%) (78%) (42%) (76%)
0.007 0.53 <0.0001 0.74 0.84 0.73 0.84
55.29 17.10 32 (23%) 44.58 7.14 8 (6%) 60 (44%) 50.73 16.29 89.60 26.53 43.76 6.71 13.38 2.09 0.60 0.13
55.10 17.95 19 (27%) 43.81 7.16 4/70 (6%) 24/66 (36%) 54.36 16.56 92.20 25.51 41.40 6.37 13.40 2.14 0.58 0.13
56.25 13.87 13 (19%) 45.54 7.11 4/67 (6%) 36/63 (57%) 46.94 15.22 80.75 26.49 46.24 6.17 13.36 2.05 0.60 0.13
0.89 0.32 0.16 1.00 0.022 0.0072 0.011 <0.0001 0.91 0.27
ACEi ¼ Angiotensin converting enzyme; ARBs ¼ Angiotensin receptor blockers; LVEDD ¼ Left ventricular end diastolic diameter; PPM ¼ Permanent pacemaker; STS-PROM ¼ Society of Thoracic Surgery-Predictor of Mortality score.
aortic gradients (89.60 26.53 to 23.18 8.57 mm Hg; p <0.001) after the procedure. There was no significant difference in 1-year post-TAVR mortality in the patients according to the presence of previous AF (15% in patients with previous AF vs 9% in patients without previous AF; OR 1.87, 95% CI 0.64 to 5.48, p ¼ 0.29) on univariate analysis. The univariate comparison of clinical outcomes, adverse events, and echocardiographic outcomes in patients grouped according to the presence of previous AF is described in Table 3. The incidence of the combined end point of all-cause 1-year mortality, stroke, vascular complications, and repeat hospitalization in 1-month post-TAVR was also analyzed. On multivariate analysis using logistic regression, previous AF was detected to be an important predictor of this combined end point (adjusted OR 2.60, 95% CI 1.22 to 5.54, p ¼ 0.013).
The cumulative incidence of overall AF after TAVR was also investigated. For this purpose, the patients with a history of AF were also included. The cumulative incidence of overall post-TAVR AF was defined as the combination of patients who had previous AF and persisted in AF throughout their pre- and post-TAVR hospital course and patients who did not have previous AF but developed NOAF after TAVR. According to this definition, the cumulative incidence of post-TAVR AF was found to be 60% (82 of 137 patients). On comparison of baseline clinical and echocardiographic parameters, there were no significant differences in the post-TAVR AF versus post-TAVR SR group. The presence of post-TAVR AF was associated with a significantly longer stay in the hospital compared with the patients in SR after TAVR (16.21 8.06 days vs 9.17 4.29 days; p <0.001).
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
Figure 2. Timing of occurrence of new-onset atrial fibrillation after transcatheter aortic valve replacement.
223
of 6.7 days (95% CI 4.69 to 8.73, p <0.0005). Similarly, postTAVR AF was also an important determinant of the duration of post-TAVR ICU stay. The presence of AF after TAVR led to prolongation of ICU course by an average of 70 hours (95% CI 25.1 to 114.7, p ¼ 0.002). In a logistic regression model, post-TAVR AF was also an important predictor of vascular complications after TAVR (adjusted OR 10.5, 95% CI 1.21 to 91.75; p ¼ 0.033). Patients with post-TAVR AF also showed a statistical trend toward increased all-cause mortality at 1 year (adjusted OR 2.90, 95% CI 0.86 to 9.85; p ¼ 0.087). In addition to preprocedural and postprocedural AF, several other baseline and procedural predictors of adverse outcome after TAVR were also indentified. Patients who had a history of percutaneous coronary intervention were also at increased risk for the composite of clinical outcomes (described previously; adjusted OR 2.10, 95% CI 0.98 to
Table 2 Baseline clinical, echocardiographic and procedural variables in patients according to the development of new-onset AF Variables Age (years) Female Body mass index (Kg/m2) Active smokers Hypertension Hyperlipidemia Diabetes mellitus Coronary artery disease H/o Percutaneous coronary intervention H/o Coronary artery bypass grafting Cerebrovascular disease Congestive heart failure Peripheral Vascular disease H/o Balloon aortic valvuloplasty Chronic obstructive Pulmonary disease Serum creatinine (mg/dl) EGFR (<60 ml/min) Prior Medication h/o Aspirin Clopidogrel Beta-Blockers ACEi/ARBs Statins Echocardiographic parameters Ejection fraction (%) Ejection fraction (<40%) Left ventricular end diastolic diameter (mm) Severe Mitral regurgitation Mean Aortic Gradient (mm Hg) Aortic valve area (cm2) Left atrial size (mm) Right ventricular systolic Pressure (mm Hg) Transapical approach
New onset AF (n ¼ 21)
SR (n ¼ 49)
84.48 6.01 15 (71%) 25.62 5.8 1 (5%) 16 (76%) 13 (62%) 6 (29%) 13 (62%) 6 (29%) 7 (33%) 6 (29%) 10 (48%) 6 (29%) 4 (19%) 5 (24%) 1.34 0.48 15 (71%)
83.69 7.38 27 (55%) 26.73 5.9 4 (8%) 38 (78%) 36 (73%) 21 (43%) 35 (71%) 17 (35%) 17 (35%) 9 (18%) 24 (49%) 18 (37%) 9 (18%) 13 (27%) 1.32 0.49 27 (55%)
19 6 12 7 17
(90%) (29%) (57%) (33%) (81%)
55.95 18.71 6 (29%) 43.43 8.27 3 (14%) 52.28 13.51 0.58 0.14 40.47 7.39 50.57 15.16 18 (86%)
45 10 32 20 38
(92%) (20%) (65%) (41%) (78%)
54.71 17.80 13 (26%) 43.98 6.72 1 (2%) 55.24 17.76 0.57 0.12 41.77 5.94 47.35 11.27 21 (43%)
Odds ratio and 95% CI (Univariate analysis)
P value
2.04; 0.68-6.13
0.67 0.29 0.47 1.00 1.00 0.35 0.29 0.57 0.78 1.00 0.35 1.00 0.59 1.00 1.00 0.86 0.29
0.84; 1.56; 0.71; 0.73; 1.23;
1.00 0.54 0.59 0.60 1.00
2.04; 0.68-6.13 0.56; 0.06-5.36 0.93; 0.28-3.10 0.59; 0.20-1.74 0.53; 0.18-1.61 0.65; 0.22-1.91 0.75; 0.25-2.30 0.94; 0.32-2.78 1.78; 0.54-5.85 0.95; 0.34-2.64 0.69; 0.22-2.10 1.05; 0.28-3.87 0.87; 0.26 -2.84
0.14-5.01 0.48-5.05 0.25-2.02 0.25-2.12 0.34-4.42
1.11; 0.35-3.46 8.0; 0.78-82.03
8.0; 2.1-30.77
0.79 1.00 0.77 0.077 0.49 0.87 0.43 0.33 0.0013
h/o ¼ history of.
The univariate comparison of clinical outcomes and adverse events after TAVR in the patients grouped according to the occurrence of post-TAVR AF is described in Table 4. Using a multivariate analysis and a linear regression model, the presence of post-TAVR AF was observed to be a strong predictor of the prolongation in the hospital stay. The presence of AF after TAVR prolonged the hospital stay by an average
4.50; p ¼ 0.055). Patients who underwent the procedure through a transapical approach were less likely to develop post-TAVR vascular complications (adjusted OR 0.055, 95% CI 0.00064 to 0.47; p ¼ 0.008). In the linear regression model, an elevation in serum creatinine at baseline was also observed to prolong the hospital stay by an average of 2.8 days (95% CI 0.005 to 5.64; p ¼ 0.05). A greater baseline
224
The American Journal of Cardiology (www.ajconline.org)
Table 3 Clinical outcomes in patients undergoing TAVR according to the presence of prior h/o AF Clinical Outcomes AF at the time of procedure Procedural success Valve related complication (embolization, damage to the valve and requirement of 2nd valve) Post procedural AF Myocardial Infarction Development of new conduction block Implantation of new pacemaker Major vascular complication Cerebrovascular events (all) Stroke Repeat hospitalization in 1 month Death within 1 month Death after 1 month Overall Mortality Overall Mortality secondary to cardiac causes Length of hospitalization (days) Composite of clinical outcomes (All-cause mortality, stroke, vascular complications and repeat hospitalization in 1 month) (All-cause mortality, stroke and vascular complications) LVEF prior to discharge from the hospital (%) Mean Ao gradient (mm Hg) Peak Ao gradient (mm Hg) D Mean Aortic gradient (mm Hg) D Peak Aortic gradient (mm Hg)
No h/o AF (n ¼ 70)
Pre-existing AF (n ¼ 67)
OR with 95 % CI (Univariate Analysis)
p value
0/70 67 (96%) 4 (6%)
40 (60%) 63 (94%) 2 (3%)
207.65; 12.33-3498.3 0.71; 0.15-3.28 0.51; 0.089-2.869
<0.0001 0.71 0.68
21 (30%) 1 (1%) 22 (31%) 20 (27%) 2 (3%) 4 (6%) 2 (3%) 9/67 (13%) 3 (4%) 3/67 (5%) 6/70 (9%) 1/70 (1%) 11.12 6.36 19(27%)
61 (91%) 1 (1%) 22 (33%) 11 (16%) 6 (9%) 3 (5%) 3 (5%) 14/61 (23%) 6 (9%) 4/61 (7%) 10/67 (15%) 7/67 (10%) 15.67 8.12 33 (49%)
23.72; 8.88-63.36 1.05; 0.06-17.07 1.07; 0.52-2.19 0.49; 0.21-1.13 3.34; 0.65-17.20 0.77; 0.17- 3.60 1.60; 0.26- 9.90 0.52; 0.21-1.31 2.20; 0.53-9.17 1.50; 0.32-6.98 1.87; 0.64-5.48 8.05; 0.96- 67.35
<0.001 1.00 1.00 0.10 0.16 1.00 0.68 0.18 0.32 0.71 0.29 0.031 0.0005 0.009
11 (16%) 57.51 15.07 12.45 4.56 (n ¼ 67) 24.99 8.83 (n ¼ 67) 41.91 16.32 (n ¼ 67) 66.93 23.89 (n ¼ 67)
20 (30%) 56.25 13.87 10.86 4.41 (n ¼ 63) 21.27 7.90 (n ¼ 63) 36.27 14.19 (n ¼ 63) 59.97 24.92 (n ¼ 63)
2.28; 0.995-5.233
2.61; 1.28-5.31
0.066 0.62 0.0452 0.0126 0.0372 0.1071
Table 4 Comparison of clinical and adverse outcomes in patients with post-procedural AF and SR Clinical Outcome/adverse Events Procedural success Valve related complications (embolization, damage to the valve, requirement of 2nd valve) Myocardial infarction Development of new conduction block Requirement of PPM implantation Major vascular complications Cerebrovascular events (all) Stroke Repeat hospitalization in 1 month Death within 1 month Death after 1 month Overall mortality Overall mortality secondary to cardiac causes Length of hospitalization (days) Composite of clinical outcomes (All-cause mortality, stroke, vascular complications and repeat hospitalization in 1 month)
Post-procedural AF (n¼82)
Post-procedural SR (n¼55)
OR with 95 % CI
P value
76 (93%) 2 (2%)
54 (98%) 4 (7%)
0.23; 0.03-2.00 0.32; 0.06-1.81
0.24 0.22
1 (1%) 25 (30%) 19 (23%) 7 (9%) 4 (5%) 4 (5%) 16/75 (21%) 7 (9%) 5 (6%) 12 (15%) 7 (9%) 16.21 8.06 39 (48%)
1 (2%) 19 (35%) 9 (16%) 1 (2%) 3 (5%) 1 (2%) 7/53 (13%) 1 (2%) 3 (5%) 4 (7%) 1 (2%) 9.17 4.29 13 (24%)
0.67; 0.04-10.90 0.83; 0.40-1.72 1.54; 0.64-3.72 5.04; 0.60-42.19 0.89; 0.19-4.14
1.00 0.709 0.39 0.144 1.00
1.78; 0.68-4.69 5.04; 0.60-42.19 1.12; 0.26-4.92 2.19; 0.67-7.17 5.04; 0.60-42.19
0.350 0.144 1.00 0.278 0.144 <0.0001 0.0068
left ventricular ejection fraction before the procedure was associated with a shortening of hospital stay by an average of 0.10 days (95% CI 0.17 to 0.03; p ¼ 0.006). Discussion This study resulted in 4 important findings. First, the presence of previous AF was an important predictor of the composite of all-cause mortality, stroke, vascular
2.93; 1.37-6.26
complications, and repeat hospitalization in 1 month. Second, NOAF after TAVR was detected in 30% of the patients who did not have a history of AF. Third, the use of transapical approach was an important predictor of development of NOAF after TAVR. Fourth, the presence of post-TAVR AF in patients was a strong predictor of duration of post-TAVR ICU and overall hospital stay and demonstrated a trend toward increased 1-year mortality.
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
In our study, the presence of previous AF was found to be an important predictor of the composite outcomes of allcause mortality, stroke, vascular complications, and repeat hospitalizations within 1 month after TAVR. The findings in this study are consistent with the epidemiologic observations of greater cumulative and cardiovascular mortality in patients with AF.22 From a mechanistic standpoint, the presence of AF may cause a loss of atrioventricular synchrony that can lead to an impairment of cardiac function. The manifestations of such adverse hemodynamic changes could be more profound in the patients with severe AS. This study demonstrated that the incidence of NOAF in patients without a history of this arrhythmia was 30%. The incidence of NOAF within a 30-day follow-up period after TAVR was reported to be 15% in the Placement of Aortic Transcatheter Valve (PARTNER) trial.2 This relatively higher incidence of NOAF in our study could be attributed to a greater number of patients who underwent TAVR through the transapical approach (81 of 37 patients; 59%) compared with 33% of patients in the PARTNER trial who underwent TAVR through this approach. In a single-center based study on 138 patients by Santos et al,23 the incidence of NOAF was reported to be 31%. Another recently published study by Tanawuttiwat et al24 reported a 42% incidence of NOAF in 123 patients. Both these studies identified the transapical approach as an important procedural risk factor for the development of NOAF. Similar to the findings in these studies, the use of transapical approach was a strong predictor for the occurrence of NOAF after TAVR in our study. The use of transapical approach to perform TAVR entails a left mini-thoracotomy and a direct puncture of left ventricular apex. The predisposition to develop NOAF using transapical approach in the setting of TAVR seems similar to the setting of thoracotomy.25 The development of NOAF in patients who underwent thoracotomy seems to be mediated by multiple mechanisms; local trauma during thoracotomy can potentially lead to an injury to the sympathovagal fibers innervating the sinus node that can sensitize the atrial myocardium to catecholamines and, thereby, promote the initiation of AF. Local trauma as a result of direct myocardial puncture of left ventricular apex during the transapical approach can also lead to pain response and ventilator restriction that can further potentiate the hyperadrenergic state and lead to NOAF after TAVR. The occurrence of NOAF relatively early in the postoperative course both in the setting of TAVR and after thoracotomy also may implicate perioperative inflammation as a relevant risk factor.26 In this study, AF after TAVR was identified to be an important predictor of the duration of post-TAVR ICU and hospital stay. Patients with post-TAVR AF were also observed to have a trend toward increased all-cause mortality. We believe that the AF after TAVR could possibly indicate a relatively permanent form of AF that persists throughout the periprocedural and postprocedural course. The findings of an increased duration of ICU stay in these patients might be because of the loss of “atrial kick” that further leads to an impairment of left ventricular systolic function. These adverse hemodynamic changes might lead to a greater requirement for hemodynamic support and, therefore, contribute to a prolonged ICU stay. The presence of post-TAVR AF could also represent the coexistence of left
225
ventricular dysfunction and impaired vascular function in these patients. The pathophysiological changes mediated by AF and the greater incidence of coexisting co-morbid conditions in patients with AF might explain the findings in our study. The findings of increased duration of ICU and overall hospital stay in the patients with AF are also relevant from a health-care economic perspective. To the best of our knowledge, this study is the first to investigate the impact of all 3 aspects of periprocedural AF (previous, new-onset, and post-TAVR AF) on clinical and adverse outcomes after TAVR. Our study is limited by its single-center, retrospective, and nonrandomized design. We believe that the ECGs and telemetry strips might be limited for detection of transient episodes of AF. We also acknowledge that the assessment of echocardiographic parameters is also subjected to interobserver variability. Disclosures Dr. Kevin Heist: (all modest in amount): Biotronik (research grant, honoraria), Boston Scientific (research grant, consultant, honoraria), Medtronic (honoraria), Sanofi (consultant), Sorin (consultant, honoraria), St. Jude Medical (research grant, consultant, honoraria); Dr. Passeri: Edward Lifesciences (proctoring role); Ptaszek: Mast Therapeutics (consultant), St. Jude Medical (consultant), World Care Clinical (consultant); Dr. Ruskin: Advanced Medical Education (consultant), Astellas/Cardiome (consultant, significant), Biosense Webster (consultant, modest) and Fellowship Support (significant), Boston ScientificFellowship Support (significant), Bristol Myers Squibb (consultant), CardioInsight-Scientific Advisory Board (modest), CryoCath-Scientific Steering Committee (no compensation), InfoBionic-Scientific Advisory Board and Equity, Medtronic (consultant, modest) and Fellowship Support (significant), Pfizer-Consultant and Scientific Steering Committee (modest), Portola-Consultant and equity (modest), Sanofi-Consultant (modest), St. Jude MedicalFellowship Support (significant); Dr. Mansour: BiosenseWebster (consultant, research grant), Boston Scientific (research grant), St. Jude (consultant), Cardiofocus (research grant), Endosense (research grant), MC10 (research grant), St. Jude Medical (consultant, research grant), Voyage Medical (research grant); Dr. Sundt: Thrasos (Steering Committee). Dr. Maan reports nothing to disclose. 1. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, Klugmann S, Bedogni F, Maisano F, Marzocchi A, Poli A, Antoniucci D, Napodano M, De Carlo M, Fiorina C, Ussia GP. Incidence and predictors of early and late mortality after transcatheter valve implantation in 663 patients with severe aortic stenosis. Circulation 2011;123:299e308. 2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrman HC, Akin JJ, Anderson WN, Wang D, Pocock SJ; PARTNER Trial Investigators. Transcatheter versus surgical aortic valve-replacement in high-risk patients. N Engl J Med 2011;364: 2187e2198. 3. Thomas M, Schymik G, Walther T, Himbert D, Leferve T, Treede H, Eggebrecht H, Rubino P, Colombo A, Lange R, Schwarz RR, Wendler O. One-year outcomes of cohort 1 in the Edward SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) registry: the European
226
4. 5. 6.
7.
8.
9.
10.
11.
12.
13. 14.
15. 16. 17.
The American Journal of Cardiology (www.ajconline.org) registry of transcatheter aortic valve implantation using the Edwards SAPIEN valve. Circulation 2011;124:425e433. Chugh SS, Blackshear JL, Shen WK, Hammill SC, Gersch BJ. Epidemiology and natural history of atrial fibrillation: clinical implications. J Am Coll Cardiol 2001;37:371e378. Norberg J, Bäckström S, Jansson JH, Johansson L. Estimating the prevalence of atrial fibrillation in a general population using validated electronic health data. Clin Epidemiol 2013;5:475e481. Lee JS, Shim CY, Wi J, Juong B, Ha JW, Lee MH, Pak HN. Left ventricular diastolic dysfunction is closely associated with mechanical function of the left atrium in patients with paroxysmal atrial fibrillation. Circ J 2013;77:697e704. Chrispin J, Jain A, Soliman EZ, Guallar E, Alonso A, Heckbert SR, Bluemke DA, Lima JA, Nazarian S. Association of electrocardiographic and imaging surrogates of left ventricular hypertrophy with incident atrial fibrillation: the Multi-Ethnic Study of Atherosclerosis. J Am Coll Cardiol 2014;63:2007e2013. Bang CN, Dalsgaard M, Greve AM, Kober L, Gohlke-Baerwolf C, Ray S, Rossebo AB, Egstrup K, Wachtell K. Left atrial size and function as predictors of new-onset of atrial fibrillation in patients with asymptomatic aortic stenosis: the Simvastatin and Ezetimibe in Aortic Stenosis study. Int J Cardiol 2013;168:2322e2327. Levy F, Garayalde E, Quere JP, Ianetta-Peltier M, Peltier M, Triboulloy C. Prognostic value of preoperative atrial fibrillation in patients with severe aortic stenosis and low ejection fraction having aortic valve replacement. Am J Cardiol 2006;98:809e811. Saxena A, Dinh DT, Reid CM, Smith JA, Shardey GC, Newcomb AE. Does preoperative atrial fibrillation portend a poorer prognosis in patients undergoing isolated aortic valve replacement? A multicentre Australian study. Can J Cardiol 2013;29:697e703. Ngaage DL, Schaff HV, Barnes SA, Sundt TM III, Mullany CJ, Dearani JA, Daly RC, Orszulak TA. Prognostic implications of preoperative atrial fibrillation in patients undergoing aortic valve replacement: is there an argument for concomitant arrhythmia surgery? Ann Thorac Surg 2006;82:1392e1399. Ngaage DL, Schaff HV, Mullany CJ, Barnes S, Dearani JA, Daly RC, Orszulak TA, Sundt TM III. Does preoperative atria fibrillation influence early and late outcomes of coronary artery bypass grafting? J Thorac Cardiovasc Surg 2007;133:182e189. Dulli DA, Stanko H, Levine RL. Atrial fibrillation is associated with severe acute ischemic stroke. Neuroepidemiology 2003;22:118e123. Levy F, Debry N, Labescat AL, Meimoun P, Malaquin D, Marechaux S, Rusinaru D, Jeu A, Ennezat PV, Castel AL, Tribouilloy C. Echocardiographic prediction of postoperative atrial fibrillation after aortic valve replacement for aortic stenosis: a two-dimensional speckle tracking left ventricular longitudinal strain multicentre pilot study. Arch Cardiovasc Dis 2012;105:499e506. Online STS Risk Calculator- Society of Thoracic Surgeons. Available at: http://riskcalc.sts.org/STSWebRiskCalc273/de.aspx. Accessed on August 08, 2014. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J 2003;24:882e883. Roques F, Nashef SA, Michel P, Gauducheau E, de Vincentiis C, Baudet E, Cortina J, David M, Faichney A, Gabrielle F, Gams E,
18.
19.
20.
21.
22. 23.
24.
25. 26.
Harjula A, Jones MT, Pintor PP, Salamon R, Thulin L. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19,030 patients. Eur J Cardiothorac Surg 1999;15:816e822. Anderson JL, Halperin JL, Albert NM, Bozkurt B, Brinidis RG, Curtis LH, DeMets D, Guyton RA, Hochman JS, Kovacs RJ, Ohman EM, Pressler SJ, Sellke FW, Shen WK, Wann LS, Curtis AB, Ellenbogen KA, Estes NA III, Ezekowitz MD, Jackman WM, January CT, Lowe JE, Page RL, Slotwiner DJ, Stevenson WG, Tracy CM, Fuster V, Ryden LE, Cannom DS, Crinjs HJ, Curtis AB, Ellenbogen KA, Le Heuzey JY, Kay GN, Olsson SB, Prystowsky EN, Tamargo JL, Wann S. Management of patients with atrial fibrillation (compilation of 2006 ACCF/AHA/ESC and 2011 ACCF/AHA/HRS recommendations): a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013;61:1935e1944. El-Chami MF, Kligo PD, Elfstrom KM, Halkos M, Thourani V, Lattouf OM, Delurgio DB, Guyton RA, Leon AR, Puskas JD. Prediction of new-onset atrial fibrillation after cardiac revascularization surgery. Am J Cardiol 2012;110:649e654. Arnold SV, Spertus JA, Lei Y, Green P, Kirtane AJ, Kapadia S, Thourani VH, Herrmann HC, Beohar N, Zajarias A, Mack MJ, Leon MB, Cohen DJ. How to define a poor outcome after transcatheter aortic valve replacement: conceptual framework and empirical observations from the Placement of Aortic Transcatheter Valve (PARTNER) trial. Circ Cardiovasc Qual Outcomes 2013;6:591e597. Reynolds MR, Magnuson EA, Lei Y, Leon MB, Smith CR, Svensson LG, Webb JG, Babaliaros VC, Bowers BS, Fearon WF, Herrmann HC, Kapadia S, Kodali SK, Makkar RR, Pichard AD, Cohen DJ; Placement of Aortic Transcatheter Valves (PARTNER) investigators. Healthrelated quality of life after transcatheter aortic valve replacement in inoperable patients with severe aortic stenosis. Circulation 2011;124: 1964e1972. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham study. N Engl J Med 1982;306:1018e1022. Amat-Santos JJ, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, Villeneuve J, Cote 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 replacement. J Am Coll Cardiol 2012;59: 178e188. Tanawuttiwat T, O’Neill BP, Cohen MG, Chintakanan O, Heldman AW, Martinez CA, Alfonso CE, Mitrani RD, Macon CJ, Carrillo RG, Williams DB, O’Neill WW, Myerburg RJ. New-onset atrial fibrillation after aortic valve replacement: comparison of transfemoral, transapical, transaortic and surgical approaches. J Am Coll Cardiol 2014;63: 1510e1519. Amar D. Prevention and management of perioperative arrhythmias in the thoracic surgical population. Anesthesiol Clin 2008;26: 325e335. Amar D, Goenka A, Zhang H, Park B, Thaler HT. Leuokocytosis and increased risk of atrial fibrillation after general thoracic surgery. Ann Thorac Surg 2006;82:1057e1061.
surgeries.9e12 AF carries an increased risk of thromboembolic and vascular complications compared with patients in sinus rhythm (SR).13 AF also leads to an atrioventricular dyssynchrony, which further adversely affects cardiac function in AS.14 The aim of this study was to determine the impact of AF on procedural outcomes. Methods The study population consisted of 137 consecutive patients with severe AS who under underwent TAVR at Massachusetts General Hospital between June 2008 and October 2012. The procedure was performed using the balloon-expandable valve (Edwards SAPIEN, SAPIEN XT; Edward Lifesciences, Irvine, California). The preprocedural risk was calculated using the Society of Thoracic Surgeons risk score for Prediction of Mortality15 and the logistic EuroSCORE.16,17 The study design is briefly explained in Figure 1. Data on baseline characteristics on all patients were collected retrospectively. Preprocedural and postprocedural electrocardiograms (ECGs) were analyzed for rhythm, various intervals, and conduction abnormalities. Data on various echocardiographic parameters were obtained from the transthoracic echocardiograms that were obtained before and after TAVR. www.ajconline.org
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
221
Figure 1. Study design and various study subsets.
Patients were evaluated for the presence of AF before the procedure. The presence of AF was confirmed on ECGs and the documentation from inpatient and outpatient medical records. Pre-TAVR rhythm was documented on all patients who underwent the procedure. All the study patients were monitored by continuous telemetry until the day of discharge. New-onset atrial fibrillation (NOAF) was defined as any episode of AF lasting >30 seconds in the patients without any history of this arrhythmia.18,19 In the patients who developed NOAF, collected data included the timing of arrhythmia occurrence, requirement of cardioversion, and the use antithrombotic therapy. The patients were divided into 2 groups: post-TAVR AF (all patients with AF 30 seconds post-TAVR) and post-TAVR SR. Univariate analyses were performed for comparison between various subgroups (previous AF vs no previous AF, post-TAVR AF vs post-TAVR SR). Categorical variables between the 2 subgroups were compared using the Fisher’s exact test, and continuous variables were compared using the 2-tailed unpaired Student’s t test. After the univariate analyses, 4 logistic regression models were prepared for each of the adverse outcomes (NOAF, major vascular complication, composite outcome of all-cause mortality, stroke, vascular complications, and repeat hospitalization in 1 month after TAVR, and all-cause 1-year mortality). We also prepared 2 linear regression models (each for duration of intensive care unit [ICU] stay and duration of hospital stay, respectively) to investigate the impact of post-TAVR AF on these outcomes. The logistic regression models were all tested for the “goodness of fit” of the logistic model with the HosmereLemeshow test. All statistical tests were performed with STATA 9.2 software (Stata Corp LP, College Station, TX). Results The median age was 85 years (range 55 to 96). Previous AF was present in 67 patients (49%). There was no difference
in the baseline clinical variables in the patients with or without previous AF, except that the use of warfarin was significantly more common in the patients with previous AF (59.7% vs 1.4%; p <0.0001). The baseline characteristics of study patients and the comparison between the patients according to the presence of previous AF are listed in Table 1. ECG performed at the time of TAVR revealed the presence of AF in 60% (40 of 67) of the patients with previous AF. The comparison of various echocardiographic parameters between the 2 groups of patients (previous AF vs without previous AF) is listed in Table 1. TAVR was performed using the Edwards SAPIEN (Edwards Lifesciences) balloonexpandable prosthesis with the 2 available sizes (23 and 26 mm). Transapical approach was used in most patients (81 of 137; 59%). To determine the incidence of NOAF, the 67 patients with a history of AF were excluded. In the remaining 70 patients, NOAF occurred in 21 patients (21 of 70; 30%). A total of 50% of all the episodes of NOAF occurred in the initial 24 hours after TAVR (Figure 2). Electrical cardioversion was performed in 3 patients who developed NOAF, and 8 patients converted from AF back to sinus rhythm after receiving amiodarone. The comparison of baseline clinical, echocardiographic, and procedural characteristics of the patients categorized according to the occurrence of NOAF is listed in Table 2. In the multivariate analysis, transapical approach of TAVR was observed to be an important predictor of NOAF after the procedure (adjusted odds ratio [OR] 5.05, 95% confidence interval [CI] 1.40 to 18.20; p ¼ 0.013). The overall 1-year mortality in this study was 12% (16 patients). The standard definition of “improvement in functional status and health-related quality of life after the procedure” was used for determining the success of TAVR.2,20,21 According to this definition, the procedural success of TAVR was 95%. Based on the predischarge echocardiogram, there was a significant decrease in the mean (50.73 16.29 to 11.67 4.52 mm Hg; p <0.001) and peak
222
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of the study population and comparison according to the presence of prior AF Baseline variables Age (years) Male Body mass index (kg/m2) Active smokers STS-PROM score (%) Logistic EuroSCORE (%) Hypertension Hyperlipidemia Diabetes Mellitus Coronary artery disease Congestive heart failure Chronic obstructive pulmonary disease Peripheral vascular disease Pulmonary hypertension Serum creatinine (mg/dl) Cerebrovascular disease Carotid artery disease Previous Procedural Percutaneous coronary intervention Prior pacemaker implantation Balloon valvuloplasty Coronary artery bypass grafting Prior Medication h/o Aspirin Clopidogrel Warfarin Beta-blockers Diuretics ACEi/ARBs Statins Echocardiographic parameters Ejection fraction (%) LVEF < 40% LV EDD (mm) Severe mitral regurgitation Moderate mitral regurgitation Mean aortic gradient (mm Hg) Peak aortic gradient (mm Hg) Left atrial size (mm) Interventricular septum (mm) Aortic valve area (cm2)
Study population (N ¼ 137)
No Prior AF (N ¼ 70)
Prior AF (N ¼ 67)
P value
84.18 6.83 65 (47%) 26.70 5.90 7 (5%) 6.88 3.82 14.33 12.24 109 (80%) 95 (69%) 47 (34%) 99 (72%) 74 (54%) 39 (28%) 42 (31%) 38 (28%) 1.33 0.47 25 (18%) 31 (23%)
83.92 6.96 30 (43%) 26.40 5.87 5 (7%) 6.49 3.18 13.67 12.01 54 (77%) 49 (70%) 27 (39%) 48 (69%) 34 (39%) 18 (26%) 24 (34%) 14 (20%) 1.32 0.49 15 (21%) 20 (29%)
84.46 6.72 35 (52%) 27 5.94 2 (3%) 7.29 4.38 15.02 12.52 55 (82%) 46 (69%) 20 (40%) 51 (76%) 40 (60%) 21 (31%) 18 (27%) 24 (36%) 1.34 0.45 10 (15%) 11 (16%)
0.65 0.31 0.55 0.44 0.23 0.52 0.53 1.00 0.37 0.35 0.23 0.57 0.36 0.06 0.80 0.38 0.10
51 27 24 55
(37%) (20%) (18%) (40%)
23 12 13 24
(33%) (17%) (19%) (34%)
28 15 11 31
(42%) (22%) (16%) (46%)
0.29 0.52 0.83 0.17
113 28 41 90 108 55 106
(82%) (20%) (30%) (66%) (79%) (40%) (79%)
64 16 1 44 56 27 55
(91%) (23%) (1%) (63%) (80%) (39%) (79%)
49 12 40 46 52 28 51
(73%) (18%) (60%) (69%) (78%) (42%) (76%)
0.007 0.53 <0.0001 0.74 0.84 0.73 0.84
55.29 17.10 32 (23%) 44.58 7.14 8 (6%) 60 (44%) 50.73 16.29 89.60 26.53 43.76 6.71 13.38 2.09 0.60 0.13
55.10 17.95 19 (27%) 43.81 7.16 4/70 (6%) 24/66 (36%) 54.36 16.56 92.20 25.51 41.40 6.37 13.40 2.14 0.58 0.13
56.25 13.87 13 (19%) 45.54 7.11 4/67 (6%) 36/63 (57%) 46.94 15.22 80.75 26.49 46.24 6.17 13.36 2.05 0.60 0.13
0.89 0.32 0.16 1.00 0.022 0.0072 0.011 <0.0001 0.91 0.27
ACEi ¼ Angiotensin converting enzyme; ARBs ¼ Angiotensin receptor blockers; LVEDD ¼ Left ventricular end diastolic diameter; PPM ¼ Permanent pacemaker; STS-PROM ¼ Society of Thoracic Surgery-Predictor of Mortality score.
aortic gradients (89.60 26.53 to 23.18 8.57 mm Hg; p <0.001) after the procedure. There was no significant difference in 1-year post-TAVR mortality in the patients according to the presence of previous AF (15% in patients with previous AF vs 9% in patients without previous AF; OR 1.87, 95% CI 0.64 to 5.48, p ¼ 0.29) on univariate analysis. The univariate comparison of clinical outcomes, adverse events, and echocardiographic outcomes in patients grouped according to the presence of previous AF is described in Table 3. The incidence of the combined end point of all-cause 1-year mortality, stroke, vascular complications, and repeat hospitalization in 1-month post-TAVR was also analyzed. On multivariate analysis using logistic regression, previous AF was detected to be an important predictor of this combined end point (adjusted OR 2.60, 95% CI 1.22 to 5.54, p ¼ 0.013).
The cumulative incidence of overall AF after TAVR was also investigated. For this purpose, the patients with a history of AF were also included. The cumulative incidence of overall post-TAVR AF was defined as the combination of patients who had previous AF and persisted in AF throughout their pre- and post-TAVR hospital course and patients who did not have previous AF but developed NOAF after TAVR. According to this definition, the cumulative incidence of post-TAVR AF was found to be 60% (82 of 137 patients). On comparison of baseline clinical and echocardiographic parameters, there were no significant differences in the post-TAVR AF versus post-TAVR SR group. The presence of post-TAVR AF was associated with a significantly longer stay in the hospital compared with the patients in SR after TAVR (16.21 8.06 days vs 9.17 4.29 days; p <0.001).
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
Figure 2. Timing of occurrence of new-onset atrial fibrillation after transcatheter aortic valve replacement.
223
of 6.7 days (95% CI 4.69 to 8.73, p <0.0005). Similarly, postTAVR AF was also an important determinant of the duration of post-TAVR ICU stay. The presence of AF after TAVR led to prolongation of ICU course by an average of 70 hours (95% CI 25.1 to 114.7, p ¼ 0.002). In a logistic regression model, post-TAVR AF was also an important predictor of vascular complications after TAVR (adjusted OR 10.5, 95% CI 1.21 to 91.75; p ¼ 0.033). Patients with post-TAVR AF also showed a statistical trend toward increased all-cause mortality at 1 year (adjusted OR 2.90, 95% CI 0.86 to 9.85; p ¼ 0.087). In addition to preprocedural and postprocedural AF, several other baseline and procedural predictors of adverse outcome after TAVR were also indentified. Patients who had a history of percutaneous coronary intervention were also at increased risk for the composite of clinical outcomes (described previously; adjusted OR 2.10, 95% CI 0.98 to
Table 2 Baseline clinical, echocardiographic and procedural variables in patients according to the development of new-onset AF Variables Age (years) Female Body mass index (Kg/m2) Active smokers Hypertension Hyperlipidemia Diabetes mellitus Coronary artery disease H/o Percutaneous coronary intervention H/o Coronary artery bypass grafting Cerebrovascular disease Congestive heart failure Peripheral Vascular disease H/o Balloon aortic valvuloplasty Chronic obstructive Pulmonary disease Serum creatinine (mg/dl) EGFR (<60 ml/min) Prior Medication h/o Aspirin Clopidogrel Beta-Blockers ACEi/ARBs Statins Echocardiographic parameters Ejection fraction (%) Ejection fraction (<40%) Left ventricular end diastolic diameter (mm) Severe Mitral regurgitation Mean Aortic Gradient (mm Hg) Aortic valve area (cm2) Left atrial size (mm) Right ventricular systolic Pressure (mm Hg) Transapical approach
New onset AF (n ¼ 21)
SR (n ¼ 49)
84.48 6.01 15 (71%) 25.62 5.8 1 (5%) 16 (76%) 13 (62%) 6 (29%) 13 (62%) 6 (29%) 7 (33%) 6 (29%) 10 (48%) 6 (29%) 4 (19%) 5 (24%) 1.34 0.48 15 (71%)
83.69 7.38 27 (55%) 26.73 5.9 4 (8%) 38 (78%) 36 (73%) 21 (43%) 35 (71%) 17 (35%) 17 (35%) 9 (18%) 24 (49%) 18 (37%) 9 (18%) 13 (27%) 1.32 0.49 27 (55%)
19 6 12 7 17
(90%) (29%) (57%) (33%) (81%)
55.95 18.71 6 (29%) 43.43 8.27 3 (14%) 52.28 13.51 0.58 0.14 40.47 7.39 50.57 15.16 18 (86%)
45 10 32 20 38
(92%) (20%) (65%) (41%) (78%)
54.71 17.80 13 (26%) 43.98 6.72 1 (2%) 55.24 17.76 0.57 0.12 41.77 5.94 47.35 11.27 21 (43%)
Odds ratio and 95% CI (Univariate analysis)
P value
2.04; 0.68-6.13
0.67 0.29 0.47 1.00 1.00 0.35 0.29 0.57 0.78 1.00 0.35 1.00 0.59 1.00 1.00 0.86 0.29
0.84; 1.56; 0.71; 0.73; 1.23;
1.00 0.54 0.59 0.60 1.00
2.04; 0.68-6.13 0.56; 0.06-5.36 0.93; 0.28-3.10 0.59; 0.20-1.74 0.53; 0.18-1.61 0.65; 0.22-1.91 0.75; 0.25-2.30 0.94; 0.32-2.78 1.78; 0.54-5.85 0.95; 0.34-2.64 0.69; 0.22-2.10 1.05; 0.28-3.87 0.87; 0.26 -2.84
0.14-5.01 0.48-5.05 0.25-2.02 0.25-2.12 0.34-4.42
1.11; 0.35-3.46 8.0; 0.78-82.03
8.0; 2.1-30.77
0.79 1.00 0.77 0.077 0.49 0.87 0.43 0.33 0.0013
h/o ¼ history of.
The univariate comparison of clinical outcomes and adverse events after TAVR in the patients grouped according to the occurrence of post-TAVR AF is described in Table 4. Using a multivariate analysis and a linear regression model, the presence of post-TAVR AF was observed to be a strong predictor of the prolongation in the hospital stay. The presence of AF after TAVR prolonged the hospital stay by an average
4.50; p ¼ 0.055). Patients who underwent the procedure through a transapical approach were less likely to develop post-TAVR vascular complications (adjusted OR 0.055, 95% CI 0.00064 to 0.47; p ¼ 0.008). In the linear regression model, an elevation in serum creatinine at baseline was also observed to prolong the hospital stay by an average of 2.8 days (95% CI 0.005 to 5.64; p ¼ 0.05). A greater baseline
224
The American Journal of Cardiology (www.ajconline.org)
Table 3 Clinical outcomes in patients undergoing TAVR according to the presence of prior h/o AF Clinical Outcomes AF at the time of procedure Procedural success Valve related complication (embolization, damage to the valve and requirement of 2nd valve) Post procedural AF Myocardial Infarction Development of new conduction block Implantation of new pacemaker Major vascular complication Cerebrovascular events (all) Stroke Repeat hospitalization in 1 month Death within 1 month Death after 1 month Overall Mortality Overall Mortality secondary to cardiac causes Length of hospitalization (days) Composite of clinical outcomes (All-cause mortality, stroke, vascular complications and repeat hospitalization in 1 month) (All-cause mortality, stroke and vascular complications) LVEF prior to discharge from the hospital (%) Mean Ao gradient (mm Hg) Peak Ao gradient (mm Hg) D Mean Aortic gradient (mm Hg) D Peak Aortic gradient (mm Hg)
No h/o AF (n ¼ 70)
Pre-existing AF (n ¼ 67)
OR with 95 % CI (Univariate Analysis)
p value
0/70 67 (96%) 4 (6%)
40 (60%) 63 (94%) 2 (3%)
207.65; 12.33-3498.3 0.71; 0.15-3.28 0.51; 0.089-2.869
<0.0001 0.71 0.68
21 (30%) 1 (1%) 22 (31%) 20 (27%) 2 (3%) 4 (6%) 2 (3%) 9/67 (13%) 3 (4%) 3/67 (5%) 6/70 (9%) 1/70 (1%) 11.12 6.36 19(27%)
61 (91%) 1 (1%) 22 (33%) 11 (16%) 6 (9%) 3 (5%) 3 (5%) 14/61 (23%) 6 (9%) 4/61 (7%) 10/67 (15%) 7/67 (10%) 15.67 8.12 33 (49%)
23.72; 8.88-63.36 1.05; 0.06-17.07 1.07; 0.52-2.19 0.49; 0.21-1.13 3.34; 0.65-17.20 0.77; 0.17- 3.60 1.60; 0.26- 9.90 0.52; 0.21-1.31 2.20; 0.53-9.17 1.50; 0.32-6.98 1.87; 0.64-5.48 8.05; 0.96- 67.35
<0.001 1.00 1.00 0.10 0.16 1.00 0.68 0.18 0.32 0.71 0.29 0.031 0.0005 0.009
11 (16%) 57.51 15.07 12.45 4.56 (n ¼ 67) 24.99 8.83 (n ¼ 67) 41.91 16.32 (n ¼ 67) 66.93 23.89 (n ¼ 67)
20 (30%) 56.25 13.87 10.86 4.41 (n ¼ 63) 21.27 7.90 (n ¼ 63) 36.27 14.19 (n ¼ 63) 59.97 24.92 (n ¼ 63)
2.28; 0.995-5.233
2.61; 1.28-5.31
0.066 0.62 0.0452 0.0126 0.0372 0.1071
Table 4 Comparison of clinical and adverse outcomes in patients with post-procedural AF and SR Clinical Outcome/adverse Events Procedural success Valve related complications (embolization, damage to the valve, requirement of 2nd valve) Myocardial infarction Development of new conduction block Requirement of PPM implantation Major vascular complications Cerebrovascular events (all) Stroke Repeat hospitalization in 1 month Death within 1 month Death after 1 month Overall mortality Overall mortality secondary to cardiac causes Length of hospitalization (days) Composite of clinical outcomes (All-cause mortality, stroke, vascular complications and repeat hospitalization in 1 month)
Post-procedural AF (n¼82)
Post-procedural SR (n¼55)
OR with 95 % CI
P value
76 (93%) 2 (2%)
54 (98%) 4 (7%)
0.23; 0.03-2.00 0.32; 0.06-1.81
0.24 0.22
1 (1%) 25 (30%) 19 (23%) 7 (9%) 4 (5%) 4 (5%) 16/75 (21%) 7 (9%) 5 (6%) 12 (15%) 7 (9%) 16.21 8.06 39 (48%)
1 (2%) 19 (35%) 9 (16%) 1 (2%) 3 (5%) 1 (2%) 7/53 (13%) 1 (2%) 3 (5%) 4 (7%) 1 (2%) 9.17 4.29 13 (24%)
0.67; 0.04-10.90 0.83; 0.40-1.72 1.54; 0.64-3.72 5.04; 0.60-42.19 0.89; 0.19-4.14
1.00 0.709 0.39 0.144 1.00
1.78; 0.68-4.69 5.04; 0.60-42.19 1.12; 0.26-4.92 2.19; 0.67-7.17 5.04; 0.60-42.19
0.350 0.144 1.00 0.278 0.144 <0.0001 0.0068
left ventricular ejection fraction before the procedure was associated with a shortening of hospital stay by an average of 0.10 days (95% CI 0.17 to 0.03; p ¼ 0.006). Discussion This study resulted in 4 important findings. First, the presence of previous AF was an important predictor of the composite of all-cause mortality, stroke, vascular
2.93; 1.37-6.26
complications, and repeat hospitalization in 1 month. Second, NOAF after TAVR was detected in 30% of the patients who did not have a history of AF. Third, the use of transapical approach was an important predictor of development of NOAF after TAVR. Fourth, the presence of post-TAVR AF in patients was a strong predictor of duration of post-TAVR ICU and overall hospital stay and demonstrated a trend toward increased 1-year mortality.
Valvular Heart Disease/Impact of AF in Patients Who Underwent TAVR
In our study, the presence of previous AF was found to be an important predictor of the composite outcomes of allcause mortality, stroke, vascular complications, and repeat hospitalizations within 1 month after TAVR. The findings in this study are consistent with the epidemiologic observations of greater cumulative and cardiovascular mortality in patients with AF.22 From a mechanistic standpoint, the presence of AF may cause a loss of atrioventricular synchrony that can lead to an impairment of cardiac function. The manifestations of such adverse hemodynamic changes could be more profound in the patients with severe AS. This study demonstrated that the incidence of NOAF in patients without a history of this arrhythmia was 30%. The incidence of NOAF within a 30-day follow-up period after TAVR was reported to be 15% in the Placement of Aortic Transcatheter Valve (PARTNER) trial.2 This relatively higher incidence of NOAF in our study could be attributed to a greater number of patients who underwent TAVR through the transapical approach (81 of 37 patients; 59%) compared with 33% of patients in the PARTNER trial who underwent TAVR through this approach. In a single-center based study on 138 patients by Santos et al,23 the incidence of NOAF was reported to be 31%. Another recently published study by Tanawuttiwat et al24 reported a 42% incidence of NOAF in 123 patients. Both these studies identified the transapical approach as an important procedural risk factor for the development of NOAF. Similar to the findings in these studies, the use of transapical approach was a strong predictor for the occurrence of NOAF after TAVR in our study. The use of transapical approach to perform TAVR entails a left mini-thoracotomy and a direct puncture of left ventricular apex. The predisposition to develop NOAF using transapical approach in the setting of TAVR seems similar to the setting of thoracotomy.25 The development of NOAF in patients who underwent thoracotomy seems to be mediated by multiple mechanisms; local trauma during thoracotomy can potentially lead to an injury to the sympathovagal fibers innervating the sinus node that can sensitize the atrial myocardium to catecholamines and, thereby, promote the initiation of AF. Local trauma as a result of direct myocardial puncture of left ventricular apex during the transapical approach can also lead to pain response and ventilator restriction that can further potentiate the hyperadrenergic state and lead to NOAF after TAVR. The occurrence of NOAF relatively early in the postoperative course both in the setting of TAVR and after thoracotomy also may implicate perioperative inflammation as a relevant risk factor.26 In this study, AF after TAVR was identified to be an important predictor of the duration of post-TAVR ICU and hospital stay. Patients with post-TAVR AF were also observed to have a trend toward increased all-cause mortality. We believe that the AF after TAVR could possibly indicate a relatively permanent form of AF that persists throughout the periprocedural and postprocedural course. The findings of an increased duration of ICU stay in these patients might be because of the loss of “atrial kick” that further leads to an impairment of left ventricular systolic function. These adverse hemodynamic changes might lead to a greater requirement for hemodynamic support and, therefore, contribute to a prolonged ICU stay. The presence of post-TAVR AF could also represent the coexistence of left
225
ventricular dysfunction and impaired vascular function in these patients. The pathophysiological changes mediated by AF and the greater incidence of coexisting co-morbid conditions in patients with AF might explain the findings in our study. The findings of increased duration of ICU and overall hospital stay in the patients with AF are also relevant from a health-care economic perspective. To the best of our knowledge, this study is the first to investigate the impact of all 3 aspects of periprocedural AF (previous, new-onset, and post-TAVR AF) on clinical and adverse outcomes after TAVR. Our study is limited by its single-center, retrospective, and nonrandomized design. We believe that the ECGs and telemetry strips might be limited for detection of transient episodes of AF. We also acknowledge that the assessment of echocardiographic parameters is also subjected to interobserver variability. Disclosures Dr. Kevin Heist: (all modest in amount): Biotronik (research grant, honoraria), Boston Scientific (research grant, consultant, honoraria), Medtronic (honoraria), Sanofi (consultant), Sorin (consultant, honoraria), St. Jude Medical (research grant, consultant, honoraria); Dr. Passeri: Edward Lifesciences (proctoring role); Ptaszek: Mast Therapeutics (consultant), St. Jude Medical (consultant), World Care Clinical (consultant); Dr. Ruskin: Advanced Medical Education (consultant), Astellas/Cardiome (consultant, significant), Biosense Webster (consultant, modest) and Fellowship Support (significant), Boston ScientificFellowship Support (significant), Bristol Myers Squibb (consultant), CardioInsight-Scientific Advisory Board (modest), CryoCath-Scientific Steering Committee (no compensation), InfoBionic-Scientific Advisory Board and Equity, Medtronic (consultant, modest) and Fellowship Support (significant), Pfizer-Consultant and Scientific Steering Committee (modest), Portola-Consultant and equity (modest), Sanofi-Consultant (modest), St. Jude MedicalFellowship Support (significant); Dr. Mansour: BiosenseWebster (consultant, research grant), Boston Scientific (research grant), St. Jude (consultant), Cardiofocus (research grant), Endosense (research grant), MC10 (research grant), St. Jude Medical (consultant, research grant), Voyage Medical (research grant); Dr. Sundt: Thrasos (Steering Committee). Dr. Maan reports nothing to disclose. 1. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, Klugmann S, Bedogni F, Maisano F, Marzocchi A, Poli A, Antoniucci D, Napodano M, De Carlo M, Fiorina C, Ussia GP. Incidence and predictors of early and late mortality after transcatheter valve implantation in 663 patients with severe aortic stenosis. Circulation 2011;123:299e308. 2. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrman HC, Akin JJ, Anderson WN, Wang D, Pocock SJ; PARTNER Trial Investigators. Transcatheter versus surgical aortic valve-replacement in high-risk patients. N Engl J Med 2011;364: 2187e2198. 3. Thomas M, Schymik G, Walther T, Himbert D, Leferve T, Treede H, Eggebrecht H, Rubino P, Colombo A, Lange R, Schwarz RR, Wendler O. One-year outcomes of cohort 1 in the Edward SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) registry: the European
226
4. 5. 6.
7.
8.
9.
10.
11.
12.
13. 14.
15. 16. 17.
The American Journal of Cardiology (www.ajconline.org) registry of transcatheter aortic valve implantation using the Edwards SAPIEN valve. Circulation 2011;124:425e433. Chugh SS, Blackshear JL, Shen WK, Hammill SC, Gersch BJ. Epidemiology and natural history of atrial fibrillation: clinical implications. J Am Coll Cardiol 2001;37:371e378. Norberg J, Bäckström S, Jansson JH, Johansson L. Estimating the prevalence of atrial fibrillation in a general population using validated electronic health data. Clin Epidemiol 2013;5:475e481. Lee JS, Shim CY, Wi J, Juong B, Ha JW, Lee MH, Pak HN. Left ventricular diastolic dysfunction is closely associated with mechanical function of the left atrium in patients with paroxysmal atrial fibrillation. Circ J 2013;77:697e704. Chrispin J, Jain A, Soliman EZ, Guallar E, Alonso A, Heckbert SR, Bluemke DA, Lima JA, Nazarian S. Association of electrocardiographic and imaging surrogates of left ventricular hypertrophy with incident atrial fibrillation: the Multi-Ethnic Study of Atherosclerosis. J Am Coll Cardiol 2014;63:2007e2013. Bang CN, Dalsgaard M, Greve AM, Kober L, Gohlke-Baerwolf C, Ray S, Rossebo AB, Egstrup K, Wachtell K. Left atrial size and function as predictors of new-onset of atrial fibrillation in patients with asymptomatic aortic stenosis: the Simvastatin and Ezetimibe in Aortic Stenosis study. Int J Cardiol 2013;168:2322e2327. Levy F, Garayalde E, Quere JP, Ianetta-Peltier M, Peltier M, Triboulloy C. Prognostic value of preoperative atrial fibrillation in patients with severe aortic stenosis and low ejection fraction having aortic valve replacement. Am J Cardiol 2006;98:809e811. Saxena A, Dinh DT, Reid CM, Smith JA, Shardey GC, Newcomb AE. Does preoperative atrial fibrillation portend a poorer prognosis in patients undergoing isolated aortic valve replacement? A multicentre Australian study. Can J Cardiol 2013;29:697e703. Ngaage DL, Schaff HV, Barnes SA, Sundt TM III, Mullany CJ, Dearani JA, Daly RC, Orszulak TA. Prognostic implications of preoperative atrial fibrillation in patients undergoing aortic valve replacement: is there an argument for concomitant arrhythmia surgery? Ann Thorac Surg 2006;82:1392e1399. Ngaage DL, Schaff HV, Mullany CJ, Barnes S, Dearani JA, Daly RC, Orszulak TA, Sundt TM III. Does preoperative atria fibrillation influence early and late outcomes of coronary artery bypass grafting? J Thorac Cardiovasc Surg 2007;133:182e189. Dulli DA, Stanko H, Levine RL. Atrial fibrillation is associated with severe acute ischemic stroke. Neuroepidemiology 2003;22:118e123. Levy F, Debry N, Labescat AL, Meimoun P, Malaquin D, Marechaux S, Rusinaru D, Jeu A, Ennezat PV, Castel AL, Tribouilloy C. Echocardiographic prediction of postoperative atrial fibrillation after aortic valve replacement for aortic stenosis: a two-dimensional speckle tracking left ventricular longitudinal strain multicentre pilot study. Arch Cardiovasc Dis 2012;105:499e506. Online STS Risk Calculator- Society of Thoracic Surgeons. Available at: http://riskcalc.sts.org/STSWebRiskCalc273/de.aspx. Accessed on August 08, 2014. Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J 2003;24:882e883. Roques F, Nashef SA, Michel P, Gauducheau E, de Vincentiis C, Baudet E, Cortina J, David M, Faichney A, Gabrielle F, Gams E,
18.
19.
20.
21.
22. 23.
24.
25. 26.
Harjula A, Jones MT, Pintor PP, Salamon R, Thulin L. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19,030 patients. Eur J Cardiothorac Surg 1999;15:816e822. Anderson JL, Halperin JL, Albert NM, Bozkurt B, Brinidis RG, Curtis LH, DeMets D, Guyton RA, Hochman JS, Kovacs RJ, Ohman EM, Pressler SJ, Sellke FW, Shen WK, Wann LS, Curtis AB, Ellenbogen KA, Estes NA III, Ezekowitz MD, Jackman WM, January CT, Lowe JE, Page RL, Slotwiner DJ, Stevenson WG, Tracy CM, Fuster V, Ryden LE, Cannom DS, Crinjs HJ, Curtis AB, Ellenbogen KA, Le Heuzey JY, Kay GN, Olsson SB, Prystowsky EN, Tamargo JL, Wann S. Management of patients with atrial fibrillation (compilation of 2006 ACCF/AHA/ESC and 2011 ACCF/AHA/HRS recommendations): a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2013;61:1935e1944. El-Chami MF, Kligo PD, Elfstrom KM, Halkos M, Thourani V, Lattouf OM, Delurgio DB, Guyton RA, Leon AR, Puskas JD. Prediction of new-onset atrial fibrillation after cardiac revascularization surgery. Am J Cardiol 2012;110:649e654. Arnold SV, Spertus JA, Lei Y, Green P, Kirtane AJ, Kapadia S, Thourani VH, Herrmann HC, Beohar N, Zajarias A, Mack MJ, Leon MB, Cohen DJ. How to define a poor outcome after transcatheter aortic valve replacement: conceptual framework and empirical observations from the Placement of Aortic Transcatheter Valve (PARTNER) trial. Circ Cardiovasc Qual Outcomes 2013;6:591e597. Reynolds MR, Magnuson EA, Lei Y, Leon MB, Smith CR, Svensson LG, Webb JG, Babaliaros VC, Bowers BS, Fearon WF, Herrmann HC, Kapadia S, Kodali SK, Makkar RR, Pichard AD, Cohen DJ; Placement of Aortic Transcatheter Valves (PARTNER) investigators. Healthrelated quality of life after transcatheter aortic valve replacement in inoperable patients with severe aortic stenosis. Circulation 2011;124: 1964e1972. Kannel WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham study. N Engl J Med 1982;306:1018e1022. Amat-Santos JJ, Rodes-Cabau J, Urena M, DeLarochelliere R, Doyle D, Bagur R, Villeneuve J, Cote 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 replacement. J Am Coll Cardiol 2012;59: 178e188. Tanawuttiwat T, O’Neill BP, Cohen MG, Chintakanan O, Heldman AW, Martinez CA, Alfonso CE, Mitrani RD, Macon CJ, Carrillo RG, Williams DB, O’Neill WW, Myerburg RJ. New-onset atrial fibrillation after aortic valve replacement: comparison of transfemoral, transapical, transaortic and surgical approaches. J Am Coll Cardiol 2014;63: 1510e1519. Amar D. Prevention and management of perioperative arrhythmias in the thoracic surgical population. Anesthesiol Clin 2008;26: 325e335. Amar D, Goenka A, Zhang H, Park B, Thaler HT. Leuokocytosis and increased risk of atrial fibrillation after general thoracic surgery. Ann Thorac Surg 2006;82:1057e1061.