CLINICAL RESEARCH
Europace (2014) 16, 1284–1290 doi:10.1093/europace/euu077
Atrial fibrillation
Use of antiarrhythmic drug therapy and clinical outcomes in older patients with concomitant atrial fibrillation and coronary artery disease
1 Duke Center for Atrial Fibrillation, Durham, NC, USA; 2Department of Medicine, Duke University Medical Center, Durham, NC, USA; and 3Duke Clinical Research Institute, Duke University Medical Center, PO Box 17969, Durham, NC 27715, USA
Received 12 December 2013; accepted after revision 17 March 2014; online publish-ahead-of-print 21 April 2014
Aims
Atrial fibrillation (AF) and coronary artery disease (CAD) are common in older patients. We aimed to describe the use of antiarrhythmic drug (AAD) therapy and clinical outcomes in these patients. ..................................................................................................................................................................................... Methods We analysed AAD therapy and outcomes in 1738 older patients (age ≥65) with AF and CAD in the Duke Databank for and results cardiovascular disease. The primary outcomes were mortality and rehospitalization at 1 and 5 years. Overall, 35% of patients received an AAD at baseline, 43% were female and 85% were white. Prior myocardial infarction (MI, 31%) and heart failure (41%) were common. Amiodarone was the most common AAD (21%), followed by pure Class III agents (sotalol 6.3%, dofetilide 2.2%). Persistence of AAD was low (35% at 1 year). After adjustment, baseline AAD use was not associated with 1-year mortality [adjusted hazard ratio (HR) 1.23, 95% confidence interval (CI) 0.94–1.60] or cardiovascular mortality (adjusted HR 1.27, 95% CI 0.90–1.80). However, AAD use was associated with increased all-cause rehospitalization (adjusted HR 1.20, 95% CI 1.03–1.39) and cardiovascular rehospitalization (adjusted HR 1.20, 95% CI 1.01–1.43) at 1 year. This association did not persist at 5 years; however, these patients were at very high risk of death (55% for those .75 and on AAD) and all-cause rehospitalization (87% for those .75 and on AAD) at 5 years. ..................................................................................................................................................................................... Conclusions In older patients with AF and CAD, antiarrhythmic therapy was associated with increased rehospitalization at 1 year. Overall, these patients are at high risk of longer-term hospitalization and death. Safer, better-tolerated, and more effective therapies for symptom control in this high-risk population are warranted.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atrial fibrillation † Ischaemic heart disease † Antiarrhythmic drug † Elderly † Outcomes research
Atrial fibrillation (AF) is the most common dysrhythmia in adults, and its incidence increases significantly with age. Ultimately, more than one in four persons over the age of 40 will be diagnosed with AF and 10% of octogenarians carry a diagnosis of AF.1,2 Atrial fibrillation has a negative impact on the quality of life comparable with that observed in patients with ischaemic heart disease, and the effect is largely attributable to significant symptoms including palpitations, fatigue, and exertional limitations.3,4 Randomized data have suggested that maintenance of sinus rhythm (‘rhythm control’) is associated with improved symptoms.5,6
However, antiarrhythmic drug (AAD) therapy in patients with coronary artery disease (CAD) raises several safety concerns, including toxic side effects and the potential for fatal proarrhythmia.7 – 9 Yet most of the evidence in the literature is derived from selected populations and few data are available in older patients. In an effort to assess the effects of AAD on clinical outcomes in older patients with AF and CAD, we performed an analysis of patients in the Duke Databank for Cardiovascular Disease. The objectives of this study were (i) to describe the use of AADs in older patients with AF and CAD, (ii) to assess clinical outcomes in these
* Corresponding author. Tel: +1 919 668 8686; fax: +1 877 991 8498. E-mail address: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2014. For permissions please email: [email protected].
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Benjamin A. Steinberg 1,2,3*, Samuel H. Broderick 3, Renato D. Lopes 2,3, Linda K. Shaw3, Kevin L. Thomas1,2,3, Tracy A. DeWald 2, James P. Daubert 1,2,3, Eric D. Peterson 2,3, Christopher B. Granger 2,3, and Jonathan P. Piccini1,2,3
1285
Antiarrhythmics for AF in older patients with CAD
What’s new?
patients, and (iii) to determine the effect of AADs on mortality and rehospitalization.
Methods Data for the current analysis were obtained from the Duke Databank for Cardiovascular Disease (DDCD), an institution-wide cohort of all patients undergoing catheterization procedures at Duke University Medical Center. All demographics, comorbidities, vital signs, laboratory studies, imaging results, and medications are captured at baseline. Patients with angiographically confirmed CAD are prospectively followed for medication use and outcomes at 6 months, 1 year, and annually thereafter. The design and methods of the DDCD have been described previously.10,11
Outcomes The primary outcomes for this analysis were all-cause mortality and rehospitalization at 1 year. Secondary outcomes were cardiovascular death and cardiovascular rehospitalization from the time of catheterization to 1 year. To further explore associations with outcomes, a landmark analysis of event-free patients was performed at 1 year for the same endpoints at 5 years. Overall, the median follow-up was 3.8 years.
Study population For the present analysis, the overall DDCD cohort was limited to patients undergoing cardiac catheterization with coronary angiography from 2000 to 2010, who were ≥65 years at the time of catheterization, had obstructive (≥50%) or non-obstructive CAD (,50%, angiographicallyconfirmed), and had a diagnosis of AF at baseline, within the prior 12 months. The diagnosis of AF was made by electrocardiogram, DDCD data, or hospital administrative data and patients with incident (new AF) during follow-up were not included. The following patients were excluded: patients who died during the index catheterization hospitalization; those with an index hospitalization of .30 days; patients without medication data available within 30 days after index catheterization; patients receiving quinidine or procainamide at baseline; and patients with a history of ventricular tachycardia or ventricular fibrillation. Patients were stratified by AAD use at baseline. For the purpose of this analysis, AAD therapy was defined as the use of an oral membrane active antiarrhythmic drug (Vaughan– Williams Class I or Class III or a mixed channel blocker) within 30 days of the index catheterization. Only AAD drugs cited in the AHA/ACC/HRS guidelines for rhythm management were considered, including propafenone, flecainide, dofetilide, sotalol, disopyramide, dronedarone, or amiodarone.12 Procainamide
Statistical methods Baseline characteristics are described by absolute rates (percentage) for categorical variables, and medians (interquartile range) for continuous variables. Analytical results were stratified by age group (65 – 75, .75), and then by use of any AAD. Unadjusted 1- and 5-year Kaplan– Meier (KM) event rates were calculated for each sub-group (by age and AAD). Tests for differences across strata were conducted for baseline characteristics. An analysis of variance F-test was used for continuous characteristics. If normality assumptions were violated for any of these tests on continuous variables, a Kruskal – Wallis test was performed for that particular variable where the assumptions were violated. A x2 test for independence was used for categorical baseline characteristics. If the assumptions were violated for a particular x2 test, Fisher’s exact test was performed. Subsequently, Cox proportional hazards regression models for each endpoint were generated for 1-year outcomes. Patients without an outcome event at 1 year and who had .1 year of follow-up were then included in landmark analyses for assessment of 5-year outcomes. Multivariable models observing the relationship between outcomes and the variables of AAD use and age group were constructed. Missing data values were imputed using multiple imputation techniques when ,15% missingness was observed. The result from these techniques is that there are multiple values imputed for each missing data point. The set of multiple values are used in running all of the models and allows the analyses to account for variation due to the fact that some of the data were imputed.13,14 Linearity assumptions were checked by examining the results of cubic polynomial spline plots of the log hazard ratio (HR) of an endpoint against each of the ordinal or continuous adjustment variables. Transformations were performed for variables that had significant non-linear relationships to satisfy this assumption. All statistical analyses of the data were performed by the Duke Clinical Research Institute using SAS software (version 9.2, SAS Institute). This analysis was approved by the Duke Institutional Review Board, and no outside funding was used to perform this analysis. All of the authors had access to the aggregate results and take full responsibility for the results presented herein.
Results From the total DDCD database of 170 629 procedures, 152 395 were excluded for being non-cardiac catheterizations, outside the study age range, in patients with a history of ventricular arrhythmias, or in those without evidence of CAD. Repeat procedures were excluded, as were an additional 10 550 patients without AF (Supplementary material online, Figure S1). The final study cohort included a total of 1738 patients (n ¼ 964 age 65– 75 and n ¼ 774 .75 years old), 609 (35%) of whom were treated with an AAD. Characteristics of the study population, stratified by age and AAD use at baseline, are shown in Table 1. Overall, 43% were female, with women less likely than men to be taking an AAD. Risk factors for atherosclerotic cardiovascular disease, such as hyperlipidaemia and diabetes, were common with lower rates in older patients. Twenty-nine percent (n ¼ 510) had non-obstructive (,50%) CAD, and the remaining 71% (n ¼ 1228) had obstructive (.50%) CAD in 1, 2, or 3 epicardial coronary arteries. Warfarin therapy was prescribed to 34% overall (with 65% having CHADS2 scores ≥2), and further details of
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
† Concomitant atrial fibrillation (AF) and coronary artery disease (CAD) portend a poor prognosis yielding significant morbidity and mortality in older patients. At 5 years, rehospitalization exceeded 75% and mortality exceeded 33% in all groups. † Nearly one-third of patients with AF and CAD are treated with an antiarrhythmic drug (AAD), most commonly amiodarone. † Use of antiarrhythmic therapy was associated with increased risk of rehospitalization at 1 year. † Safer and more effective therapies for symptom control in this population are needed.
and quinidine were excluded. Use of AAD is presented by age group: 65 – 75 and .75 years. All analyses of AAD were conducted according to the treatment at baseline.
1286
B.A. Steinberg et al.
Table 1 Baseline characteristics Overall (n 5 1738) Age 65– 75
Age >75
.............................................. ..............................................
P-value
No AAD (n 5 622) AAD (n 5 342) No AAD (n 5 507) AAD (n 5 267)
............................................................................................................................................................................... Age Female
74 (69, 79)
70 (67, 73)
70 (67, 72)
80 (78, 83)
80 (78, 83)
,0.001 ,0.001
752 (43)
252 (41)
121 (35)
252 (50)
127 (48)
Race White
1453 (85)
497 (81)
302 (88)
421 (84)
233 (88)
Black
206 (12)
86 (14)
30 (9)
63 (13)
27 (10)
42 (2.4)
21 (3.4)
5 (1.5)
13 (2.6)
3 (1.1)
17 (1.0) 1263 (73)
7 (1.1) 442 (71)
5 (1.5) 241 (71)
3 (0.6) 374 (74)
2 (0.8) 206 (77)
Hyperlipidaemia
924 (53)
344 (55)
187 (55)
246 (49)
147 (55)
0.1
Cigarette Smoking Diabetes
654 (38) 458 (26)
266 (43) 198 (32)
140 (41) 94 (28)
172 (34) 118 (23)
76 (29) 48 (18)
,0.001 ,0.001
Body mass index (kg/m2)
27.3 (23.9, 31.3)
28.1 (25.0, 32.2)
28.1 (25.0, 31.6)
25.9 (23.3, 29.4)
Estimated GFR, median (IQR) ≥90
63 (49, 77) 169 (9.7)
Native American
65 (52, 79) 78 (13)
64 (52, 79) 42 (12)
62 (49, 74) 35 (6.9)
26.5 (23.2, 30.1) 57 (46, 69) 14 (5.2)
60 ≤ GFR , 90
759 (44)
276 (44)
140 (41)
238 (47)
105 (39)
30 ≤ GFR , 60 15 ≤ GFR , 30
705 (41) 44 (2.5)
220 (35) 16 (2.6)
135 (40) 6 (1.8)
216 (43) 13 (2.6)
134 (50) 9 (3.4)
60 (3.5)
31 (5.0)
19 (5.6)
5 (1.0)
5 (1.9)
144 (8.5)
92 (15)
52 (15)
0 (0.0)
0 (0.0)
,15 CHADS2 score 0 1 ≥2 Warfarin therapy Heart rate (b.p.m.)
0.2
,0.001 ,0.001
,0.001 454 (27)
226 (37)
120 (36)
76 (15)
32 (12)
1101 (65) 585 (34)
289 (48) 185 (30)
165 (49) 148 (43)
417 (85) 128 (25)
230 (88) 124 (46)
,0.001
73 (62, 86)
73 (63, 85)
72 (61, 90)
73 (63, 85)
74 (61, 83)
0.9
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg)
142 (127, 162) 78 (67, 87)
142 (126, 160) 78 (68, 89)
139 (124, 157) 80 (70, 87)
147 (130, 166) 77 (67, 86)
141 (124, 164) 78 (66, 86)
,0.001 0.09
QTc interval (ms)
446 (421, 476)
442 (418, 470)
450 (421, 477)
446 (422, 479)
455 (430, 484)
,0.001
QRS interval (ms)
97 (86, 122)
95 (86, 114)
99 (87, 116)
98 (86, 128)
101 (85, 133)
0.06
Values are presented as number (%) or median (interquartile range). GFR, glomerular filtration rate.
antithrombotic medications in this population have been described previously.11 Roughly one-third of the cohort had a prior myocardial infarction (MI), and about one-third had prior coronary revascularization (percutaneous or surgical, Table 2). Approximately 40% of patients manifested an acute coronary syndrome on index admission, and AAD use in this subgroup was similar to that of the overall cohort (35%). Forty-one percent had congestive heart failure. The median left-ventricular ejection fraction (EF) was 55% in patients not on AAD vs. 52% for patients on an AAD. The estimated glomerular filtration rate was lower in older patients, and lowest in older patients receiving AAD therapy.
Rate and rhythm therapies Concomitant cardiac medications are shown in Table 3, stratified by age. Eighty-six percent and 84% (ages 65–75 and .75, respectively) of patients receiving AAD were also taking a beta-blocker, compared with 73 and 78% of patients not on an AAD. Calcium-channel blockers (non-dihydropyridines) and digoxin were not commonly used.
The most common AAD in this cohort was amiodarone (60% of those on AAD, 21% overall), followed by sotalol (6.3% overall), and then dofetilide (2.2% overall). More than 75% of the cohort was receiving beta-blocker therapy, including those with mildmoderate and severe heart failure. After 1 year, 35% of patients on AAD at baseline remained on AAD; of patients not on an AAD initially, 16% were on an AAD at 1 year (excluding patients who died).
Unadjusted clinical outcomes After 5 years of follow-up, unadjusted KM rates of all-cause rehospitalization ranged from 79% in non-AAD patients aged 65– 75 to 87% in AAD patients .75 years. Unadjusted KM rates for all-cause mortality ranged from 33% in AAD patients aged 65 –75 to 55% in AAD patients .75 years (Figures 1– 4). Rates of mortality over time were highest in older patients, while rehospitalization rates were highest in patients on AAD of either age group (Supplementary material online, Table S1). A minority of follow-up hospitalization events carried a
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Other Hypertension
0.08
1287
Antiarrhythmics for AF in older patients with CAD
Table 2 Baseline cardiovascular disease Overall (n 5 1738)
Age 65 –75
....................................
Age >75
P-value
....................................
No AAD (n 5 622)
AAD (n 5 342)
No AAD (n 5 507)
174 (28) 213 (34)
104 (30) 99 (29)
173 (34) 150 (30)
AAD (n 5 267)
............................................................................................................................................................................... Prior CAD history History of MI History of revascularization
545 (31) 535 (31)
94 (35) 73 (27)
0.07 0.1
ACS during index hospitalization
0.07 108 (6.2) 253 (14.6)
34 (5.5) 72 (11.6)
24 (7.0) 44 (12.9)
32 (6.3) 90 (17.8) 0
18 (6.7) 47 (17.6)
Unspecified MI
5 (0.3)
2 (0.3)
2 (0.6)
Unstable angina No ACS
378 (21.7) 994 (57.2)
138 (22.2) 376 (60.5)
64 (18.7) 208 (60.8)
118 (23.3) 267 (52.7)
1 (0.4)
Non-obstructive 1-vessel
510 (29) 415 (24)
208 (33) 146 (24)
101 (30) 83 (24)
129 (25) 123 (24)
72 (27) 63 (24)
2-vessel
313 (18)
111 (18)
51 (15)
106 (21)
45 (17)
500 (29) 262 (15)
157 (25) 84 (14)
107 (31) 50 (15)
149 (29) 98 (19)
87 (33) 30 (11)
History of stroke
65 (3.7)
30 (4.8)
13 (3.8)
18 (3.6)
4 (1.5)
EF NYHA heart failure class
55 (40, 62)
55 (41, 63)
52 (39, 60)
55 (43, 63)
58 (21.7) 143 (53.6)
CAD severity
0.07
3-vessel History of cerebrovascular disease
None
0.1
52 (39, 60)
1005 (59)
362 (60)
192 (57)
312 (63)
139 (53)
32 (1.9) 161 (9.5)
11 (1.8) 56 (9.2)
4 (1.2) 41 (12)
11 (2.2) 35 (7.1)
6 (2.3) 29 (11)
III
336 (20)
121 (20)
67 (20)
93 (19)
55 (21)
IV
165 (9.7)
57 (9.4)
33 (9.8)
42 (8.5)
33 (12.6)
I II
0.009 0.004 0.3
Values are presented as number (%) or median (interquartile range). CAD, coronary artery disease; ACS, acute coronary syndrome; MI, myocardial infarction; NYHA, New York Heart Association.
Table 3 Rate and rhythm medications by age and antiarrhythmic drug use Overall
Age 65 –75
.....................................................
No AAD (n 5 622)
AAD (n 5 342)
Age >75
.....................................................
No AAD (n 5 507)
AAD (n 5 267)
............................................................................................................................................................................... ACE inhibitor Beta blocker CCB (verapamil, diltiazem) Digoxin
919 (53)
313 (50)
184 (54)
264 (52)
158 (59)
1369 (79) 330 (19)
455 (73) 107 (17)
295 (86) 68 (20)
394 (78) 93 (18)
225 (84) 62 (23)
282 (16)
109 (18)
48 (14)
88 (17)
Antiarrhythmic Drugs Disopyramide Propafenone Flecainide
342 (100) 22 (1.3) 22 (1.3)
13 (3.8) 14 (4.1)
37 (14) 267 (100) 9 (3.4) 8 (3.0)
20 (1.2)
15 (4.4)
5 (1.9)
Class III Dofetilide
145 (8.3) 38 (2.2)
86 (25) 25 (7.3)
59 (22) 13 (4.9)
Sotalol
109 (6.3)
63 (18)
46 (17)
368 (21) 3 (0.2)
202 (59) 2 (0.6)
166 (62) 1 (0.4)
Amiodarone Dronedarone
AAD, antiarrhythmic drug; ACE, angiotensin-converting enzyme; CCB, calcium-channel blocker.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
STEMI Non-STEMI
1288
B.A. Steinberg et al.
1
1 Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.8
0.8
Probability of death
0.7 0.6 0.5 P-value 75, No AAD Age >75, Taking AAD
0.5
1
1.5
2
2.5 Years
3
3.5
4
4.5
622
361
284
223
195
169
135
107
90
73
5 63
342
174
144
114
97
83
70
55
47
33
31
507
309
237
181
153
129
99
82
64
50
45
267
129
102
79
70
57
44
32
26
14
11
Figure 3 Unadjusted KM event rates for all-cause rehospitalization. AAD, antiarrhythmic drug.
1
1 Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.8
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.9
Probability of cardiovascular hospitalization
0.9
0.7 0.6 0.5 P-value 75, No AAD Age >75, Taking AAD
0.5
1
1.5
2
2.5 Years
3
3.5
622
427
360
310
283
248
213
177
4
342
200
170
147
132
112
103
85
507
344
278
232
207
182
147
127
267
156
133
112
99
88
74
55
48
4.5
5
134
120
77
60
55
105
88
73
36
27
158
Figure 4 Unadjusted KM event rates for cardiovascular rehospitalization. AAD, antiarrhythmic drug.
95% Adjusted hazard ratio Confidence (AAD – No AAD) interval All-cause mortality
1.23
0.94–1.60
Cardiovascular death
1.27
0.90–1.80
Rehospitalization
1.20
1.03–1.39
Cardiovascular Rehospitalization
1.20
1.01–1.43 0 1 2 AAD Better AAD Worse
Figure 5 Forest plot of adjusted KM event rates at 1 year. AAD, antiarrhythmic drug.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
0 0
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
P-value = 0.011
0.3
0.1 0
Probability of cardiovascular death
0.7
0.2
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.9
Probability of hospitalization
0.9
1289
Antiarrhythmics for AF in older patients with CAD
Table S2): all-cause mortality (adjusted HR 0.89, 95% CI 0.72–1.12), cardiovascular death (adjusted HR 0.94, 95% CI 0.69–1.27), any rehospitalization (adjusted HR 1.11, 95% CI 0.89–1.38), and cardiovascular rehospitalization (adjusted HR 1.08, 95% CI 0.84 –1.39).
Discussion
Limitations These data are derived from a single-centre observational cohort, and are thus subject to the limitations of such methods including selection bias. Patients with non-obstructive CAD are queried only for hospitalizations within the Duke Health System, whereas patients with obstructive CAD are queried by mail as to whether any rehospitalization occurred within or outside of Duke. Diagnosis codes for cause of rehospitalization are available only for those events occurring within Duke. Additionally, residual and unmeasured confounding likely persist, and a temporal relationship between AAD and outcome is difficult to establish. Both of these factors limit conclusions of a causal relationship between AAD and outcome. Use of AAD was assessed at a single time point, and thus we cannot discern a potential time-dependent effect of such agents on the outcome. Finally, the breadth of confidence intervals (CIs) around the adjusted HRs for mortality demonstrates that we may lack power to more precisely define the effect of AAD.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
We analysed AAD use and outcomes in over 1700 older patients with AF and concomitant CAD. Nearly one-third of patients were treated with an AAD, most commonly amiodarone. There was a high risk of adverse clinical outcomes in this population, including death and rehospitalization, regardless of AAD use. At 5 years, rehospitalization exceeded 75% and mortality exceeded 33% in all groups. Persistence of AAD therapy at 1 year was low. After multivariable adjustment, mortality was not associated with AAD use; however, patients taking an AAD had a higher risk of rehospitalization and cardiovascular rehospitalization at 1 year when compared with patients not on an AAD. This effect did not appear to persist at 5 years for those patients who were event-free and followed beyond 1 year. Our first observation is the significant mortality and morbidity associated with a concomitant diagnosis of AF and CAD in persons ≥65 years of age. At 5 years, the vast majority had been rehospitalized, and nearly half had died. These findings exaggerate previous descriptions from the REACH registry, where 1-year rates of allcause and cardiovascular mortality in patients with AF and CAD were 4.4 and 3.4%, respectively.15 Our study provides extended, long-term data well beyond 12 months. However, there are several differences between these cohorts; most notably, the REACH study enrolled stable outpatients around the world. In comparison, analyses of randomized trials in ACS have demonstrated the marked poor prognostic effect of AF in acutely-ill patients, and our results are consistent with these rates allowing for age differences.16 This may be attributable to the large proportion with acute coronary syndrome in our cohort (40%), which have been shown to have significantly worse outcomes when complicated by AF.17 There are several important distinctions between our results and those from the AFFIRM trial.18 – 20 The most important are (i) these are contemporary results of AAD treatment, (ii) in a significantly sicker patient cohort (by CAD and heart failure diagnoses), and (iii) they are data from clinical practice—not a well-controlled randomized clinical trial. We demonstrate comparable rates of cardiovascular hospitalization at 5 years—unfortunately, this represents little improvement 10 years later. Similarly, hospitalization risk was increased by AADs in both cohorts, and most common of which remains amiodarone. Amiodarone remains the most frequently used AAD in older patients with AF,21 and there is reliable evidence to support its use in high-risk patients with AF.12,22,23 Yet the significant side effects and toxicities of amiodarone are well-known; these findings underscore not only the urgent need for safer, more effective pharmacotherapies for rhythm control in AF, but the ongoing risk of using suboptimal treatments in high-risk patients. However, the AFFIRM investigators cite a lack of warfarin therapy as a potential contributor to the mortality risk of the rhythm strategy observed in that trial.18 We cannot confirm this and, in fact, our data suggest that it may be more attributable to AAD, as more patients on AADs in our cohort were treated with warfarin. Furthermore, this trend is magnified in our patients .75 years of age (46% on an
AAD received warfarin vs. 25% not on warfarin), a group not well studied in AFFIRM (mean age 70). The isolated early hazard associated with AAD use is likely due to several factors. We cannot exclude adverse patient selection for AAD that is incompletely measured in our multivariable adjustment, leading to higher short-term rehospitalization in this group. Given the age in our population, the competing hazard of unrelated causes of morbidity may overtake any adverse impact associated with AAD at 5 years, by which time all of these patients would be .70 years old. Additionally, there may be an acute hazard associated with AAD use in patients with ACS (roughly 40% of our cohort) that wanes over time.9 However, there was also marked attrition of AAD use in this cohort. Antiarrhythmic agents are often poorly tolerated and, combined with marginal effectiveness, this yields high discontinuation rates. Treatment persistence has been a long-standing problem with AADs,24 including amiodarone.25,26 In one meta-analysis, AADs were discontinued due to side effects in 10.4%, due to treatment failure in 13.5%, and due to non-compliance in 4.2%. Treatment-related death occurred in 0.5%.8 In our data, adverse event rates converged following AAD discontinuation in the majority of patients, suggesting a possible exposure–response relationship. Regardless of the reasons for discontinuation, the need for safe and effective AF treatments remains, particularly for high-risk patients. Dronedarone, which was used infrequently in our cohort, has demonstrated reduced hospitalization in certain high-risk populations;27 however, there have been safety concerns with this drug in other AF groups.28 And while older patients may be candidates for drugs with more long-term toxicities (i.e. amiodarone), use of AADs in his population is fraught with other pitfalls, including dynamic renal function, poly-pharmacy, extensive comorbidity (including CAD), and challenges in surveillance of arrhythmia. As the major objective is frequently symptomatic relief, alternative therapies should be entertained. While novel AADs with improved safety and efficacy profiles may become available, the use of invasive, catheter-based strategies for the management of symptomatic AF is a potential option. Preliminary data support the use of catheter ablation of AF in older, symptomatic patients,29 – 31 and further study is warranted regarding the optimal management of such patients.
1290
Conclusions Older patients with AF and CAD in contemporary practice are at high risk of long-term death and rehospitalization, irrespective of AAD therapy. Treatment with AAD therapy was associated with increased risk of rehospitalization at 1 year. These data highlight the need for improved therapies for symptom control in this population.
Supplementary material Supplementary material is available at Europace online.
Funding This analysis was funded by the Duke Clinical Research Institute; no outside funding was used to perform this analysis.
References 1. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370 –5. 2. Lloyd-Jones DM. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation 2004;110:1042 –6. 3. Dorian P, Jung W, Newman D, Paquette M, Wood K, Ayers GM et al. The impairment of health-related quality of life in patients with intermittent atrial fibrillation: implications for the assessment of investigational therapy. J Am Coll Cardiol 2000;36:1303 –9. 4. Spertus J, Dorian P, Bubien R, Lewis S, Godejohn D, Reynolds MR et al. Development and validation of the Atrial Fibrillation Effect on QualiTy-of-Life (AFEQT) questionnaire in patients with atrial fibrillation. Circ Arrhythm Electrophysiol 2011;4:15–25. 5. Chung MK, Shemanski L, Sherman DG, Greene HL, Hogan DB, Kellen JC et al. Functional status in rate- versus rhythm-control strategies for atrial fibrillation: results of the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) functional status substudy. J Am Coll Cardiol 2005;46:1891 – 9. 6. Guglin M, Chen R, Curtis AB. Sinus rhythm is associated with fewer heart failure symptoms: insights from the AFFIRM trial. Heart Rhythm 2010;7:596 –601. 7. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. The Cardiac Arrhythmia Suppression Trial (CAST) investigators. N Engl J Med 1989;321:406 –12. 8. Calkins H, Reynolds MR, Spector P, Sondhi M, Xu Y, Martin A et al. Treatment of atrial fibrillation with antiarrhythmic drugs or radiofrequency ablation: two systematic literature reviews and meta-analyses. Circ Arrhythm Electrophysiol 2009;2:349 –61. 9. Maisel WH, Kuntz KM, Reimold SC, Lee TH, Antman EM, Friedman PL et al. Risk of initiating antiarrhythmic drug therapy for atrial fibrillation in patients admitted to a university hospital. Ann Intern Med 1997;127:281 –4. 10. Newby LK, LaPointe NM, Chen AY, Kramer JM, Hammill BG, DeLong ER et al. Longterm adherence to evidence-based secondary prevention therapies in coronary artery disease. Circulation 2006;113:203 – 12.
11. Hess CN, Broderick S, Piccini JP, Alexander KP, Newby LK, Shaw LK et al. Antithrombotic therapy for atrial fibrillation and coronary artery disease in older patients. Am Heart J 2012;164:607 –15. 12. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA et al. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation 2011;123:e269 –367. 13. Rubin DB. Inference and missing data. Biometrika 1976;63:581 –92. 14. Little RJA, Rubin DB. Statistical analysis with missing data. New York: John Wiley & Sones; 2002. 15. Goto S, Bhatt DL, Rother J, Alberts M, Hill MD, Ikeda Y et al. Prevalence, clinical profile, and cardiovascular outcomes of atrial fibrillation patients with atherothrombosis. Am Heart J 2008;156:855 –63, 63 e2. 16. Lopes RD, Pieper KS, Horton JR, Al-Khatib SM, Newby LK, Mehta RH et al. Shortand long-term outcomes following atrial fibrillation in patients with acute coronary syndromes with or without ST-segment elevation. Heart 2008;94:867 –73. 17. Crenshaw BS, Ward SR, Granger CB, Stebbins AL, Topol EJ, Califf RM. Atrial fibrillation in the setting of acute myocardial infarction: the GUSTO-I experience. Global utilization of streptokinase and TPA for occluded coronary arteries. J Am Coll Cardiol 1997;30:406 –13. 18. Steinberg JS, Sadaniantz A, Kron J, Krahn A, Denny DM, Daubert J et al. Analysis of cause-specific mortality in the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Circulation 2004;109:1973 – 80. 19. Saksena S, Slee A, Waldo AL, Freemantle N, Reynolds M, Rosenberg Y et al. Cardiovascular outcomes in the AFFIRM trial (Atrial Fibrillation Follow-Up Investigation of Rhythm Management): an assessment of individual antiarrhythmic drug therapies compared with rate control with propensity score-matched analyses. J Am Coll Cardiol 2011;58:1975 –85. 20. Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347:1825 –33. 21. Piccini JP, Mi X, Dewald TA, Go AS, Hernandez AF, Curtis LH. Pharmacotherapy in Medicare beneficiaries with atrial fibrillation. Heart Rhythm 2012;9:1403 –8. 22. Zimetbaum P. Amiodarone for atrial fibrillation. N Engl J Med 2007;356:935 –41. 23. Singh BN, Singh SN, Reda DJ, Tang XC, Lopez B, Harris CL et al. Amiodarone versus sotalol for atrial fibrillation. N Engl J Med 2005;352:1861 –72. 24. Squire A, Goldman ME, Kupersmith J, Stern EH, Fuster V, Schweitzer P. Long-term antiarrhythmic therapy. Problem of low drug levels and patient noncompliance. Am J Med 1984;77:1035 –8. 25. Guerin A, Lin J, Jhaveri M, Wu EQ, Yu AP, Cloutier M et al. Outcomes in atrial fibrillation patients on combined warfarin & antiarrhythmic therapy. Int J Cardiol 2013; 167:564 – 9. 26. Kim MH, Smith PJ, Jhaveri M, Lin J, Klingman D. One-year treatment persistence and potential adverse events among patients with atrial fibrillation treated with amiodarone or sotalol: a retrospective claims database analysis. Clin Ther 2011;33: 1668 –81 e1. 27. Hohnloser SH, Crijns HJ, van Eickels M, Gaudin C, Page RL, Torp-Pedersen C et al. Effect of dronedarone on cardiovascular events in atrial fibrillation. N Engl J Med 2009;360:668 –78. 28. Connolly SJ, Camm AJ, Halperin JL, Joyner C, Alings M, Amerena J et al. Dronedarone in high-risk permanent atrial fibrillation. N Engl J Med 2011;365:2268 –76. 29. Zado E, Callans DJ, Riley M, Hutchinson M, Garcia F, Bala R et al. Long-term clinical efficacy and risk of catheter ablation for atrial fibrillation in the elderly. J Cardiovasc Electrophysiol 2008;19:621 –6. 30. Bunch TJ, Weiss JP, Crandall BG, May HT, Bair TL, Osborn JS et al. Long-term clinical efficacy and risk of catheter ablation for atrial fibrillation in octogenarians. Pacing Clin Electrophysiol 2010;33:146 –52. 31. Piccini JP, Sinner MF, Greiner MA, Hammill BG, Fontes JD, Daubert JP et al. Outcomes of medicare beneficiaries undergoing catheter ablation for atrial fibrillation. Circulation 2012;126:2200 –7.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Conflict of interest: B.A.S. was funded by NIH T-32 training grant #5 T32 HL 7101-37. The following relationships exist related to this presentation: B.A.S., S.H.B., L.K.S., and K.L.T. reported no disclosures. Full disclosure information for J.P.P., R.D.S., E.D.P., and C.G. can be found online at https://www.dcri.org/about-us/conflict-ofinterest/.
B.A. Steinberg et al.
Europace (2014) 16, 1284–1290 doi:10.1093/europace/euu077
Atrial fibrillation
Use of antiarrhythmic drug therapy and clinical outcomes in older patients with concomitant atrial fibrillation and coronary artery disease
1 Duke Center for Atrial Fibrillation, Durham, NC, USA; 2Department of Medicine, Duke University Medical Center, Durham, NC, USA; and 3Duke Clinical Research Institute, Duke University Medical Center, PO Box 17969, Durham, NC 27715, USA
Received 12 December 2013; accepted after revision 17 March 2014; online publish-ahead-of-print 21 April 2014
Aims
Atrial fibrillation (AF) and coronary artery disease (CAD) are common in older patients. We aimed to describe the use of antiarrhythmic drug (AAD) therapy and clinical outcomes in these patients. ..................................................................................................................................................................................... Methods We analysed AAD therapy and outcomes in 1738 older patients (age ≥65) with AF and CAD in the Duke Databank for and results cardiovascular disease. The primary outcomes were mortality and rehospitalization at 1 and 5 years. Overall, 35% of patients received an AAD at baseline, 43% were female and 85% were white. Prior myocardial infarction (MI, 31%) and heart failure (41%) were common. Amiodarone was the most common AAD (21%), followed by pure Class III agents (sotalol 6.3%, dofetilide 2.2%). Persistence of AAD was low (35% at 1 year). After adjustment, baseline AAD use was not associated with 1-year mortality [adjusted hazard ratio (HR) 1.23, 95% confidence interval (CI) 0.94–1.60] or cardiovascular mortality (adjusted HR 1.27, 95% CI 0.90–1.80). However, AAD use was associated with increased all-cause rehospitalization (adjusted HR 1.20, 95% CI 1.03–1.39) and cardiovascular rehospitalization (adjusted HR 1.20, 95% CI 1.01–1.43) at 1 year. This association did not persist at 5 years; however, these patients were at very high risk of death (55% for those .75 and on AAD) and all-cause rehospitalization (87% for those .75 and on AAD) at 5 years. ..................................................................................................................................................................................... Conclusions In older patients with AF and CAD, antiarrhythmic therapy was associated with increased rehospitalization at 1 year. Overall, these patients are at high risk of longer-term hospitalization and death. Safer, better-tolerated, and more effective therapies for symptom control in this high-risk population are warranted.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Atrial fibrillation † Ischaemic heart disease † Antiarrhythmic drug † Elderly † Outcomes research
Atrial fibrillation (AF) is the most common dysrhythmia in adults, and its incidence increases significantly with age. Ultimately, more than one in four persons over the age of 40 will be diagnosed with AF and 10% of octogenarians carry a diagnosis of AF.1,2 Atrial fibrillation has a negative impact on the quality of life comparable with that observed in patients with ischaemic heart disease, and the effect is largely attributable to significant symptoms including palpitations, fatigue, and exertional limitations.3,4 Randomized data have suggested that maintenance of sinus rhythm (‘rhythm control’) is associated with improved symptoms.5,6
However, antiarrhythmic drug (AAD) therapy in patients with coronary artery disease (CAD) raises several safety concerns, including toxic side effects and the potential for fatal proarrhythmia.7 – 9 Yet most of the evidence in the literature is derived from selected populations and few data are available in older patients. In an effort to assess the effects of AAD on clinical outcomes in older patients with AF and CAD, we performed an analysis of patients in the Duke Databank for Cardiovascular Disease. The objectives of this study were (i) to describe the use of AADs in older patients with AF and CAD, (ii) to assess clinical outcomes in these
* Corresponding author. Tel: +1 919 668 8686; fax: +1 877 991 8498. E-mail address: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2014. For permissions please email: [email protected].
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Benjamin A. Steinberg 1,2,3*, Samuel H. Broderick 3, Renato D. Lopes 2,3, Linda K. Shaw3, Kevin L. Thomas1,2,3, Tracy A. DeWald 2, James P. Daubert 1,2,3, Eric D. Peterson 2,3, Christopher B. Granger 2,3, and Jonathan P. Piccini1,2,3
1285
Antiarrhythmics for AF in older patients with CAD
What’s new?
patients, and (iii) to determine the effect of AADs on mortality and rehospitalization.
Methods Data for the current analysis were obtained from the Duke Databank for Cardiovascular Disease (DDCD), an institution-wide cohort of all patients undergoing catheterization procedures at Duke University Medical Center. All demographics, comorbidities, vital signs, laboratory studies, imaging results, and medications are captured at baseline. Patients with angiographically confirmed CAD are prospectively followed for medication use and outcomes at 6 months, 1 year, and annually thereafter. The design and methods of the DDCD have been described previously.10,11
Outcomes The primary outcomes for this analysis were all-cause mortality and rehospitalization at 1 year. Secondary outcomes were cardiovascular death and cardiovascular rehospitalization from the time of catheterization to 1 year. To further explore associations with outcomes, a landmark analysis of event-free patients was performed at 1 year for the same endpoints at 5 years. Overall, the median follow-up was 3.8 years.
Study population For the present analysis, the overall DDCD cohort was limited to patients undergoing cardiac catheterization with coronary angiography from 2000 to 2010, who were ≥65 years at the time of catheterization, had obstructive (≥50%) or non-obstructive CAD (,50%, angiographicallyconfirmed), and had a diagnosis of AF at baseline, within the prior 12 months. The diagnosis of AF was made by electrocardiogram, DDCD data, or hospital administrative data and patients with incident (new AF) during follow-up were not included. The following patients were excluded: patients who died during the index catheterization hospitalization; those with an index hospitalization of .30 days; patients without medication data available within 30 days after index catheterization; patients receiving quinidine or procainamide at baseline; and patients with a history of ventricular tachycardia or ventricular fibrillation. Patients were stratified by AAD use at baseline. For the purpose of this analysis, AAD therapy was defined as the use of an oral membrane active antiarrhythmic drug (Vaughan– Williams Class I or Class III or a mixed channel blocker) within 30 days of the index catheterization. Only AAD drugs cited in the AHA/ACC/HRS guidelines for rhythm management were considered, including propafenone, flecainide, dofetilide, sotalol, disopyramide, dronedarone, or amiodarone.12 Procainamide
Statistical methods Baseline characteristics are described by absolute rates (percentage) for categorical variables, and medians (interquartile range) for continuous variables. Analytical results were stratified by age group (65 – 75, .75), and then by use of any AAD. Unadjusted 1- and 5-year Kaplan– Meier (KM) event rates were calculated for each sub-group (by age and AAD). Tests for differences across strata were conducted for baseline characteristics. An analysis of variance F-test was used for continuous characteristics. If normality assumptions were violated for any of these tests on continuous variables, a Kruskal – Wallis test was performed for that particular variable where the assumptions were violated. A x2 test for independence was used for categorical baseline characteristics. If the assumptions were violated for a particular x2 test, Fisher’s exact test was performed. Subsequently, Cox proportional hazards regression models for each endpoint were generated for 1-year outcomes. Patients without an outcome event at 1 year and who had .1 year of follow-up were then included in landmark analyses for assessment of 5-year outcomes. Multivariable models observing the relationship between outcomes and the variables of AAD use and age group were constructed. Missing data values were imputed using multiple imputation techniques when ,15% missingness was observed. The result from these techniques is that there are multiple values imputed for each missing data point. The set of multiple values are used in running all of the models and allows the analyses to account for variation due to the fact that some of the data were imputed.13,14 Linearity assumptions were checked by examining the results of cubic polynomial spline plots of the log hazard ratio (HR) of an endpoint against each of the ordinal or continuous adjustment variables. Transformations were performed for variables that had significant non-linear relationships to satisfy this assumption. All statistical analyses of the data were performed by the Duke Clinical Research Institute using SAS software (version 9.2, SAS Institute). This analysis was approved by the Duke Institutional Review Board, and no outside funding was used to perform this analysis. All of the authors had access to the aggregate results and take full responsibility for the results presented herein.
Results From the total DDCD database of 170 629 procedures, 152 395 were excluded for being non-cardiac catheterizations, outside the study age range, in patients with a history of ventricular arrhythmias, or in those without evidence of CAD. Repeat procedures were excluded, as were an additional 10 550 patients without AF (Supplementary material online, Figure S1). The final study cohort included a total of 1738 patients (n ¼ 964 age 65– 75 and n ¼ 774 .75 years old), 609 (35%) of whom were treated with an AAD. Characteristics of the study population, stratified by age and AAD use at baseline, are shown in Table 1. Overall, 43% were female, with women less likely than men to be taking an AAD. Risk factors for atherosclerotic cardiovascular disease, such as hyperlipidaemia and diabetes, were common with lower rates in older patients. Twenty-nine percent (n ¼ 510) had non-obstructive (,50%) CAD, and the remaining 71% (n ¼ 1228) had obstructive (.50%) CAD in 1, 2, or 3 epicardial coronary arteries. Warfarin therapy was prescribed to 34% overall (with 65% having CHADS2 scores ≥2), and further details of
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
† Concomitant atrial fibrillation (AF) and coronary artery disease (CAD) portend a poor prognosis yielding significant morbidity and mortality in older patients. At 5 years, rehospitalization exceeded 75% and mortality exceeded 33% in all groups. † Nearly one-third of patients with AF and CAD are treated with an antiarrhythmic drug (AAD), most commonly amiodarone. † Use of antiarrhythmic therapy was associated with increased risk of rehospitalization at 1 year. † Safer and more effective therapies for symptom control in this population are needed.
and quinidine were excluded. Use of AAD is presented by age group: 65 – 75 and .75 years. All analyses of AAD were conducted according to the treatment at baseline.
1286
B.A. Steinberg et al.
Table 1 Baseline characteristics Overall (n 5 1738) Age 65– 75
Age >75
.............................................. ..............................................
P-value
No AAD (n 5 622) AAD (n 5 342) No AAD (n 5 507) AAD (n 5 267)
............................................................................................................................................................................... Age Female
74 (69, 79)
70 (67, 73)
70 (67, 72)
80 (78, 83)
80 (78, 83)
,0.001 ,0.001
752 (43)
252 (41)
121 (35)
252 (50)
127 (48)
Race White
1453 (85)
497 (81)
302 (88)
421 (84)
233 (88)
Black
206 (12)
86 (14)
30 (9)
63 (13)
27 (10)
42 (2.4)
21 (3.4)
5 (1.5)
13 (2.6)
3 (1.1)
17 (1.0) 1263 (73)
7 (1.1) 442 (71)
5 (1.5) 241 (71)
3 (0.6) 374 (74)
2 (0.8) 206 (77)
Hyperlipidaemia
924 (53)
344 (55)
187 (55)
246 (49)
147 (55)
0.1
Cigarette Smoking Diabetes
654 (38) 458 (26)
266 (43) 198 (32)
140 (41) 94 (28)
172 (34) 118 (23)
76 (29) 48 (18)
,0.001 ,0.001
Body mass index (kg/m2)
27.3 (23.9, 31.3)
28.1 (25.0, 32.2)
28.1 (25.0, 31.6)
25.9 (23.3, 29.4)
Estimated GFR, median (IQR) ≥90
63 (49, 77) 169 (9.7)
Native American
65 (52, 79) 78 (13)
64 (52, 79) 42 (12)
62 (49, 74) 35 (6.9)
26.5 (23.2, 30.1) 57 (46, 69) 14 (5.2)
60 ≤ GFR , 90
759 (44)
276 (44)
140 (41)
238 (47)
105 (39)
30 ≤ GFR , 60 15 ≤ GFR , 30
705 (41) 44 (2.5)
220 (35) 16 (2.6)
135 (40) 6 (1.8)
216 (43) 13 (2.6)
134 (50) 9 (3.4)
60 (3.5)
31 (5.0)
19 (5.6)
5 (1.0)
5 (1.9)
144 (8.5)
92 (15)
52 (15)
0 (0.0)
0 (0.0)
,15 CHADS2 score 0 1 ≥2 Warfarin therapy Heart rate (b.p.m.)
0.2
,0.001 ,0.001
,0.001 454 (27)
226 (37)
120 (36)
76 (15)
32 (12)
1101 (65) 585 (34)
289 (48) 185 (30)
165 (49) 148 (43)
417 (85) 128 (25)
230 (88) 124 (46)
,0.001
73 (62, 86)
73 (63, 85)
72 (61, 90)
73 (63, 85)
74 (61, 83)
0.9
Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg)
142 (127, 162) 78 (67, 87)
142 (126, 160) 78 (68, 89)
139 (124, 157) 80 (70, 87)
147 (130, 166) 77 (67, 86)
141 (124, 164) 78 (66, 86)
,0.001 0.09
QTc interval (ms)
446 (421, 476)
442 (418, 470)
450 (421, 477)
446 (422, 479)
455 (430, 484)
,0.001
QRS interval (ms)
97 (86, 122)
95 (86, 114)
99 (87, 116)
98 (86, 128)
101 (85, 133)
0.06
Values are presented as number (%) or median (interquartile range). GFR, glomerular filtration rate.
antithrombotic medications in this population have been described previously.11 Roughly one-third of the cohort had a prior myocardial infarction (MI), and about one-third had prior coronary revascularization (percutaneous or surgical, Table 2). Approximately 40% of patients manifested an acute coronary syndrome on index admission, and AAD use in this subgroup was similar to that of the overall cohort (35%). Forty-one percent had congestive heart failure. The median left-ventricular ejection fraction (EF) was 55% in patients not on AAD vs. 52% for patients on an AAD. The estimated glomerular filtration rate was lower in older patients, and lowest in older patients receiving AAD therapy.
Rate and rhythm therapies Concomitant cardiac medications are shown in Table 3, stratified by age. Eighty-six percent and 84% (ages 65–75 and .75, respectively) of patients receiving AAD were also taking a beta-blocker, compared with 73 and 78% of patients not on an AAD. Calcium-channel blockers (non-dihydropyridines) and digoxin were not commonly used.
The most common AAD in this cohort was amiodarone (60% of those on AAD, 21% overall), followed by sotalol (6.3% overall), and then dofetilide (2.2% overall). More than 75% of the cohort was receiving beta-blocker therapy, including those with mildmoderate and severe heart failure. After 1 year, 35% of patients on AAD at baseline remained on AAD; of patients not on an AAD initially, 16% were on an AAD at 1 year (excluding patients who died).
Unadjusted clinical outcomes After 5 years of follow-up, unadjusted KM rates of all-cause rehospitalization ranged from 79% in non-AAD patients aged 65– 75 to 87% in AAD patients .75 years. Unadjusted KM rates for all-cause mortality ranged from 33% in AAD patients aged 65 –75 to 55% in AAD patients .75 years (Figures 1– 4). Rates of mortality over time were highest in older patients, while rehospitalization rates were highest in patients on AAD of either age group (Supplementary material online, Table S1). A minority of follow-up hospitalization events carried a
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Other Hypertension
0.08
1287
Antiarrhythmics for AF in older patients with CAD
Table 2 Baseline cardiovascular disease Overall (n 5 1738)
Age 65 –75
....................................
Age >75
P-value
....................................
No AAD (n 5 622)
AAD (n 5 342)
No AAD (n 5 507)
174 (28) 213 (34)
104 (30) 99 (29)
173 (34) 150 (30)
AAD (n 5 267)
............................................................................................................................................................................... Prior CAD history History of MI History of revascularization
545 (31) 535 (31)
94 (35) 73 (27)
0.07 0.1
ACS during index hospitalization
0.07 108 (6.2) 253 (14.6)
34 (5.5) 72 (11.6)
24 (7.0) 44 (12.9)
32 (6.3) 90 (17.8) 0
18 (6.7) 47 (17.6)
Unspecified MI
5 (0.3)
2 (0.3)
2 (0.6)
Unstable angina No ACS
378 (21.7) 994 (57.2)
138 (22.2) 376 (60.5)
64 (18.7) 208 (60.8)
118 (23.3) 267 (52.7)
1 (0.4)
Non-obstructive 1-vessel
510 (29) 415 (24)
208 (33) 146 (24)
101 (30) 83 (24)
129 (25) 123 (24)
72 (27) 63 (24)
2-vessel
313 (18)
111 (18)
51 (15)
106 (21)
45 (17)
500 (29) 262 (15)
157 (25) 84 (14)
107 (31) 50 (15)
149 (29) 98 (19)
87 (33) 30 (11)
History of stroke
65 (3.7)
30 (4.8)
13 (3.8)
18 (3.6)
4 (1.5)
EF NYHA heart failure class
55 (40, 62)
55 (41, 63)
52 (39, 60)
55 (43, 63)
58 (21.7) 143 (53.6)
CAD severity
0.07
3-vessel History of cerebrovascular disease
None
0.1
52 (39, 60)
1005 (59)
362 (60)
192 (57)
312 (63)
139 (53)
32 (1.9) 161 (9.5)
11 (1.8) 56 (9.2)
4 (1.2) 41 (12)
11 (2.2) 35 (7.1)
6 (2.3) 29 (11)
III
336 (20)
121 (20)
67 (20)
93 (19)
55 (21)
IV
165 (9.7)
57 (9.4)
33 (9.8)
42 (8.5)
33 (12.6)
I II
0.009 0.004 0.3
Values are presented as number (%) or median (interquartile range). CAD, coronary artery disease; ACS, acute coronary syndrome; MI, myocardial infarction; NYHA, New York Heart Association.
Table 3 Rate and rhythm medications by age and antiarrhythmic drug use Overall
Age 65 –75
.....................................................
No AAD (n 5 622)
AAD (n 5 342)
Age >75
.....................................................
No AAD (n 5 507)
AAD (n 5 267)
............................................................................................................................................................................... ACE inhibitor Beta blocker CCB (verapamil, diltiazem) Digoxin
919 (53)
313 (50)
184 (54)
264 (52)
158 (59)
1369 (79) 330 (19)
455 (73) 107 (17)
295 (86) 68 (20)
394 (78) 93 (18)
225 (84) 62 (23)
282 (16)
109 (18)
48 (14)
88 (17)
Antiarrhythmic Drugs Disopyramide Propafenone Flecainide
342 (100) 22 (1.3) 22 (1.3)
13 (3.8) 14 (4.1)
37 (14) 267 (100) 9 (3.4) 8 (3.0)
20 (1.2)
15 (4.4)
5 (1.9)
Class III Dofetilide
145 (8.3) 38 (2.2)
86 (25) 25 (7.3)
59 (22) 13 (4.9)
Sotalol
109 (6.3)
63 (18)
46 (17)
368 (21) 3 (0.2)
202 (59) 2 (0.6)
166 (62) 1 (0.4)
Amiodarone Dronedarone
AAD, antiarrhythmic drug; ACE, angiotensin-converting enzyme; CCB, calcium-channel blocker.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
STEMI Non-STEMI
1288
B.A. Steinberg et al.
1
1 Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.8
0.8
Probability of death
0.7 0.6 0.5 P-value 75, No AAD Age >75, Taking AAD
0.5
1
1.5
2
2.5 Years
3
3.5
4
4.5
622
361
284
223
195
169
135
107
90
73
5 63
342
174
144
114
97
83
70
55
47
33
31
507
309
237
181
153
129
99
82
64
50
45
267
129
102
79
70
57
44
32
26
14
11
Figure 3 Unadjusted KM event rates for all-cause rehospitalization. AAD, antiarrhythmic drug.
1
1 Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.8
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.9
Probability of cardiovascular hospitalization
0.9
0.7 0.6 0.5 P-value 75, No AAD Age >75, Taking AAD
0.5
1
1.5
2
2.5 Years
3
3.5
622
427
360
310
283
248
213
177
4
342
200
170
147
132
112
103
85
507
344
278
232
207
182
147
127
267
156
133
112
99
88
74
55
48
4.5
5
134
120
77
60
55
105
88
73
36
27
158
Figure 4 Unadjusted KM event rates for cardiovascular rehospitalization. AAD, antiarrhythmic drug.
95% Adjusted hazard ratio Confidence (AAD – No AAD) interval All-cause mortality
1.23
0.94–1.60
Cardiovascular death
1.27
0.90–1.80
Rehospitalization
1.20
1.03–1.39
Cardiovascular Rehospitalization
1.20
1.01–1.43 0 1 2 AAD Better AAD Worse
Figure 5 Forest plot of adjusted KM event rates at 1 year. AAD, antiarrhythmic drug.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
0 0
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
P-value = 0.011
0.3
0.1 0
Probability of cardiovascular death
0.7
0.2
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
Age 65–75, No AAD Age 65–75, Taking AAD Age >75, No AAD Age >75, Taking AAD
0.9
Probability of hospitalization
0.9
1289
Antiarrhythmics for AF in older patients with CAD
Table S2): all-cause mortality (adjusted HR 0.89, 95% CI 0.72–1.12), cardiovascular death (adjusted HR 0.94, 95% CI 0.69–1.27), any rehospitalization (adjusted HR 1.11, 95% CI 0.89–1.38), and cardiovascular rehospitalization (adjusted HR 1.08, 95% CI 0.84 –1.39).
Discussion
Limitations These data are derived from a single-centre observational cohort, and are thus subject to the limitations of such methods including selection bias. Patients with non-obstructive CAD are queried only for hospitalizations within the Duke Health System, whereas patients with obstructive CAD are queried by mail as to whether any rehospitalization occurred within or outside of Duke. Diagnosis codes for cause of rehospitalization are available only for those events occurring within Duke. Additionally, residual and unmeasured confounding likely persist, and a temporal relationship between AAD and outcome is difficult to establish. Both of these factors limit conclusions of a causal relationship between AAD and outcome. Use of AAD was assessed at a single time point, and thus we cannot discern a potential time-dependent effect of such agents on the outcome. Finally, the breadth of confidence intervals (CIs) around the adjusted HRs for mortality demonstrates that we may lack power to more precisely define the effect of AAD.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
We analysed AAD use and outcomes in over 1700 older patients with AF and concomitant CAD. Nearly one-third of patients were treated with an AAD, most commonly amiodarone. There was a high risk of adverse clinical outcomes in this population, including death and rehospitalization, regardless of AAD use. At 5 years, rehospitalization exceeded 75% and mortality exceeded 33% in all groups. Persistence of AAD therapy at 1 year was low. After multivariable adjustment, mortality was not associated with AAD use; however, patients taking an AAD had a higher risk of rehospitalization and cardiovascular rehospitalization at 1 year when compared with patients not on an AAD. This effect did not appear to persist at 5 years for those patients who were event-free and followed beyond 1 year. Our first observation is the significant mortality and morbidity associated with a concomitant diagnosis of AF and CAD in persons ≥65 years of age. At 5 years, the vast majority had been rehospitalized, and nearly half had died. These findings exaggerate previous descriptions from the REACH registry, where 1-year rates of allcause and cardiovascular mortality in patients with AF and CAD were 4.4 and 3.4%, respectively.15 Our study provides extended, long-term data well beyond 12 months. However, there are several differences between these cohorts; most notably, the REACH study enrolled stable outpatients around the world. In comparison, analyses of randomized trials in ACS have demonstrated the marked poor prognostic effect of AF in acutely-ill patients, and our results are consistent with these rates allowing for age differences.16 This may be attributable to the large proportion with acute coronary syndrome in our cohort (40%), which have been shown to have significantly worse outcomes when complicated by AF.17 There are several important distinctions between our results and those from the AFFIRM trial.18 – 20 The most important are (i) these are contemporary results of AAD treatment, (ii) in a significantly sicker patient cohort (by CAD and heart failure diagnoses), and (iii) they are data from clinical practice—not a well-controlled randomized clinical trial. We demonstrate comparable rates of cardiovascular hospitalization at 5 years—unfortunately, this represents little improvement 10 years later. Similarly, hospitalization risk was increased by AADs in both cohorts, and most common of which remains amiodarone. Amiodarone remains the most frequently used AAD in older patients with AF,21 and there is reliable evidence to support its use in high-risk patients with AF.12,22,23 Yet the significant side effects and toxicities of amiodarone are well-known; these findings underscore not only the urgent need for safer, more effective pharmacotherapies for rhythm control in AF, but the ongoing risk of using suboptimal treatments in high-risk patients. However, the AFFIRM investigators cite a lack of warfarin therapy as a potential contributor to the mortality risk of the rhythm strategy observed in that trial.18 We cannot confirm this and, in fact, our data suggest that it may be more attributable to AAD, as more patients on AADs in our cohort were treated with warfarin. Furthermore, this trend is magnified in our patients .75 years of age (46% on an
AAD received warfarin vs. 25% not on warfarin), a group not well studied in AFFIRM (mean age 70). The isolated early hazard associated with AAD use is likely due to several factors. We cannot exclude adverse patient selection for AAD that is incompletely measured in our multivariable adjustment, leading to higher short-term rehospitalization in this group. Given the age in our population, the competing hazard of unrelated causes of morbidity may overtake any adverse impact associated with AAD at 5 years, by which time all of these patients would be .70 years old. Additionally, there may be an acute hazard associated with AAD use in patients with ACS (roughly 40% of our cohort) that wanes over time.9 However, there was also marked attrition of AAD use in this cohort. Antiarrhythmic agents are often poorly tolerated and, combined with marginal effectiveness, this yields high discontinuation rates. Treatment persistence has been a long-standing problem with AADs,24 including amiodarone.25,26 In one meta-analysis, AADs were discontinued due to side effects in 10.4%, due to treatment failure in 13.5%, and due to non-compliance in 4.2%. Treatment-related death occurred in 0.5%.8 In our data, adverse event rates converged following AAD discontinuation in the majority of patients, suggesting a possible exposure–response relationship. Regardless of the reasons for discontinuation, the need for safe and effective AF treatments remains, particularly for high-risk patients. Dronedarone, which was used infrequently in our cohort, has demonstrated reduced hospitalization in certain high-risk populations;27 however, there have been safety concerns with this drug in other AF groups.28 And while older patients may be candidates for drugs with more long-term toxicities (i.e. amiodarone), use of AADs in his population is fraught with other pitfalls, including dynamic renal function, poly-pharmacy, extensive comorbidity (including CAD), and challenges in surveillance of arrhythmia. As the major objective is frequently symptomatic relief, alternative therapies should be entertained. While novel AADs with improved safety and efficacy profiles may become available, the use of invasive, catheter-based strategies for the management of symptomatic AF is a potential option. Preliminary data support the use of catheter ablation of AF in older, symptomatic patients,29 – 31 and further study is warranted regarding the optimal management of such patients.
1290
Conclusions Older patients with AF and CAD in contemporary practice are at high risk of long-term death and rehospitalization, irrespective of AAD therapy. Treatment with AAD therapy was associated with increased risk of rehospitalization at 1 year. These data highlight the need for improved therapies for symptom control in this population.
Supplementary material Supplementary material is available at Europace online.
Funding This analysis was funded by the Duke Clinical Research Institute; no outside funding was used to perform this analysis.
References 1. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV et al. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the AnTicoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA 2001;285:2370 –5. 2. Lloyd-Jones DM. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation 2004;110:1042 –6. 3. Dorian P, Jung W, Newman D, Paquette M, Wood K, Ayers GM et al. The impairment of health-related quality of life in patients with intermittent atrial fibrillation: implications for the assessment of investigational therapy. J Am Coll Cardiol 2000;36:1303 –9. 4. Spertus J, Dorian P, Bubien R, Lewis S, Godejohn D, Reynolds MR et al. Development and validation of the Atrial Fibrillation Effect on QualiTy-of-Life (AFEQT) questionnaire in patients with atrial fibrillation. Circ Arrhythm Electrophysiol 2011;4:15–25. 5. Chung MK, Shemanski L, Sherman DG, Greene HL, Hogan DB, Kellen JC et al. Functional status in rate- versus rhythm-control strategies for atrial fibrillation: results of the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) functional status substudy. J Am Coll Cardiol 2005;46:1891 – 9. 6. Guglin M, Chen R, Curtis AB. Sinus rhythm is associated with fewer heart failure symptoms: insights from the AFFIRM trial. Heart Rhythm 2010;7:596 –601. 7. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. The Cardiac Arrhythmia Suppression Trial (CAST) investigators. N Engl J Med 1989;321:406 –12. 8. Calkins H, Reynolds MR, Spector P, Sondhi M, Xu Y, Martin A et al. Treatment of atrial fibrillation with antiarrhythmic drugs or radiofrequency ablation: two systematic literature reviews and meta-analyses. Circ Arrhythm Electrophysiol 2009;2:349 –61. 9. Maisel WH, Kuntz KM, Reimold SC, Lee TH, Antman EM, Friedman PL et al. Risk of initiating antiarrhythmic drug therapy for atrial fibrillation in patients admitted to a university hospital. Ann Intern Med 1997;127:281 –4. 10. Newby LK, LaPointe NM, Chen AY, Kramer JM, Hammill BG, DeLong ER et al. Longterm adherence to evidence-based secondary prevention therapies in coronary artery disease. Circulation 2006;113:203 – 12.
11. Hess CN, Broderick S, Piccini JP, Alexander KP, Newby LK, Shaw LK et al. Antithrombotic therapy for atrial fibrillation and coronary artery disease in older patients. Am Heart J 2012;164:607 –15. 12. Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA et al. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation 2011;123:e269 –367. 13. Rubin DB. Inference and missing data. Biometrika 1976;63:581 –92. 14. Little RJA, Rubin DB. Statistical analysis with missing data. New York: John Wiley & Sones; 2002. 15. Goto S, Bhatt DL, Rother J, Alberts M, Hill MD, Ikeda Y et al. Prevalence, clinical profile, and cardiovascular outcomes of atrial fibrillation patients with atherothrombosis. Am Heart J 2008;156:855 –63, 63 e2. 16. Lopes RD, Pieper KS, Horton JR, Al-Khatib SM, Newby LK, Mehta RH et al. Shortand long-term outcomes following atrial fibrillation in patients with acute coronary syndromes with or without ST-segment elevation. Heart 2008;94:867 –73. 17. Crenshaw BS, Ward SR, Granger CB, Stebbins AL, Topol EJ, Califf RM. Atrial fibrillation in the setting of acute myocardial infarction: the GUSTO-I experience. Global utilization of streptokinase and TPA for occluded coronary arteries. J Am Coll Cardiol 1997;30:406 –13. 18. Steinberg JS, Sadaniantz A, Kron J, Krahn A, Denny DM, Daubert J et al. Analysis of cause-specific mortality in the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Circulation 2004;109:1973 – 80. 19. Saksena S, Slee A, Waldo AL, Freemantle N, Reynolds M, Rosenberg Y et al. Cardiovascular outcomes in the AFFIRM trial (Atrial Fibrillation Follow-Up Investigation of Rhythm Management): an assessment of individual antiarrhythmic drug therapies compared with rate control with propensity score-matched analyses. J Am Coll Cardiol 2011;58:1975 –85. 20. Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002;347:1825 –33. 21. Piccini JP, Mi X, Dewald TA, Go AS, Hernandez AF, Curtis LH. Pharmacotherapy in Medicare beneficiaries with atrial fibrillation. Heart Rhythm 2012;9:1403 –8. 22. Zimetbaum P. Amiodarone for atrial fibrillation. N Engl J Med 2007;356:935 –41. 23. Singh BN, Singh SN, Reda DJ, Tang XC, Lopez B, Harris CL et al. Amiodarone versus sotalol for atrial fibrillation. N Engl J Med 2005;352:1861 –72. 24. Squire A, Goldman ME, Kupersmith J, Stern EH, Fuster V, Schweitzer P. Long-term antiarrhythmic therapy. Problem of low drug levels and patient noncompliance. Am J Med 1984;77:1035 –8. 25. Guerin A, Lin J, Jhaveri M, Wu EQ, Yu AP, Cloutier M et al. Outcomes in atrial fibrillation patients on combined warfarin & antiarrhythmic therapy. Int J Cardiol 2013; 167:564 – 9. 26. Kim MH, Smith PJ, Jhaveri M, Lin J, Klingman D. One-year treatment persistence and potential adverse events among patients with atrial fibrillation treated with amiodarone or sotalol: a retrospective claims database analysis. Clin Ther 2011;33: 1668 –81 e1. 27. Hohnloser SH, Crijns HJ, van Eickels M, Gaudin C, Page RL, Torp-Pedersen C et al. Effect of dronedarone on cardiovascular events in atrial fibrillation. N Engl J Med 2009;360:668 –78. 28. Connolly SJ, Camm AJ, Halperin JL, Joyner C, Alings M, Amerena J et al. Dronedarone in high-risk permanent atrial fibrillation. N Engl J Med 2011;365:2268 –76. 29. Zado E, Callans DJ, Riley M, Hutchinson M, Garcia F, Bala R et al. Long-term clinical efficacy and risk of catheter ablation for atrial fibrillation in the elderly. J Cardiovasc Electrophysiol 2008;19:621 –6. 30. Bunch TJ, Weiss JP, Crandall BG, May HT, Bair TL, Osborn JS et al. Long-term clinical efficacy and risk of catheter ablation for atrial fibrillation in octogenarians. Pacing Clin Electrophysiol 2010;33:146 –52. 31. Piccini JP, Sinner MF, Greiner MA, Hammill BG, Fontes JD, Daubert JP et al. Outcomes of medicare beneficiaries undergoing catheter ablation for atrial fibrillation. Circulation 2012;126:2200 –7.
Downloaded from http://europace.oxfordjournals.org/ at Virginia Polytechnic Institute and State University on August 30, 2014
Conflict of interest: B.A.S. was funded by NIH T-32 training grant #5 T32 HL 7101-37. The following relationships exist related to this presentation: B.A.S., S.H.B., L.K.S., and K.L.T. reported no disclosures. Full disclosure information for J.P.P., R.D.S., E.D.P., and C.G. can be found online at https://www.dcri.org/about-us/conflict-ofinterest/.
B.A. Steinberg et al.
