Skip to main content
PMC home page
JAMA Network logoLink to JAMA Network
. 2019 May 8;4(6):526–548. doi: 10.1001/jamacardio.2018.4729

Analysis of Outcomes in Ischemic vs Nonischemic Cardiomyopathy in Patients With Atrial Fibrillation

A Report From the GARFIELD-AF Registry

Ramon Corbalan 1,✉, Jean-Pierre Bassand 2,3, Laura Illingworth 3, Giuseppe Ambrosio 4, A John Camm 5, David A Fitzmaurice 6, Keith A A Fox 7, Samuel Z Goldhaber 8, Shinya Goto 9, Sylvia Haas 10, Gloria Kayani 3, Lorenzo G Mantovani 11, Frank Misselwitz 12, Karen S Pieper 3,13, Alexander G G Turpie 14, Freek W A Verheugt 15, Ajay K Kakkar 16, for the GARFIELD-AF Investigators
PMCID: PMC6506904  PMID: 31066873

Key Points

Question

What are the management strategies and outcomes of patients with nonvalvular atrial fibrillation and concomitant congestive heart failure (ischemic or nonischemic cardiomyopathy)?

Findings

In this cohort study of a registry that included 52 014 patients, patients with ischemic cardiomyopathy were less likely to receive oral anticoagulants with or without antiplatelet drugs and more likely to receive antiplatelet drugs alone than patients with nonischemic cardiomyopathy and had a worse outcome in terms of all-cause and cardiovascular death.

Meaning

There is a need for physicians to opt for improved adherence to guidelines-directed treatment of both atrial fibrillation and concomitant congestive heart failure, particularly in patients with ischemic cardiomyopathy.

This cohort study assesses treatment strategies and 1-year clinical outcomes of antithrombotic and CHF therapies for patients with newly diagnosed AF with concomitant congestive heart failure.

Abstract

Importance

Congestive heart failure (CHF) is commonly associated with nonvalvular atrial fibrillation (AF), and their combination may affect treatment strategies and outcomes.

Objective

To assess the treatment strategies and 1-year clinical outcomes of antithrombotic and CHF therapies for patients with newly diagnosed AF with concomitant CHF stratified by etiology (ischemic cardiomyopathy [ICM] vs nonischemic cardiomyopathy [NICM]).

Design, Setting, and Participants

The GARFIELD-AF registry is a prospective, noninterventional registry. A total of 52 014 patients with AF were enrolled between March 2010 and August 2016. A total of 11 738 patients 18 years and older with newly diagnosed AF (≤6 weeks’ duration) and at least 1 investigator-determined stroke risk factor were included. Data were analyzed from December 2017 to September 2018.

Exposures

One-year follow-up rates of death, stroke/systemic embolism, and major bleeding were assessed.

Main Outcomes and Measures

Event rates per 100 person-years were estimated from the Poisson model and Cox hazard ratios (HRs) and 95% confidence intervals.

Results

The median age of the population was 71.0 years, 22 987 of 52 013 were women (44.2%) and 31 958 of 52 014 were white (61.4%). Of 11 738 patients with CHF, 4717 (40.2%) had ICM and 7021 (59.8%) had NICM. Prescription of oral anticoagulant and antiplatelet drugs was not balanced between groups. Oral anticoagulants with or without antiplatelet drugs were used in 2753 patients with ICM (60.1%) and 5082 patients with NICM (73.7%). Antiplatelets were prescribed alone in 1576 patients with ICM (34.4%) and 1071 patients with NICM (15.5%). Compared with patients with NICM, use of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers (72.6% [3439] vs 60.3% [4236]) and of β blockers (63.3% [2988] vs 53.2% [3737]) was higher in patients with ICM. Rates of all-cause and cardiovascular death per 100 patient-years were significantly higher in the ICM group (all-cause death: ICM, 10.2; 95% CI, 9.2-11.1; NICM, 7.0; 95% CI, 6.4-7.6; cardiovascular death: ICM, 5.1; 95% CI, 4.5-5.9; NICM, 2.9; 95% CI, 2.5-3.4). Stroke/systemic embolism rates tended to be higher in ICM groups compared with NICM groups (ICM, 2.0; 95% CI, 1.6-2.5; NICM, 1.5; 95% CI, 1.3-1.9). Major bleeding rates were significantly higher in the ICM group (1.1; 95% CI, 0.8-1.4) compared with the NICM group (0.7; 95% CI, 0.5-0.9).

Conclusions and Relevance

Patients with ICM received oral anticoagulants with or without antiplatelet drugs less frequently and antiplatelets alone more frequently than patients with NICM, but they received angiotensin-converting enzyme inhibitors/angiotensin receptor blockers more often than patients with NICM. All-cause and cardiovascular death rates were higher in patients with ICM than patients with NICM.

Trial Registration

ClinicalTrials.gov Identifier: NCT01090362

Introduction

Congestive heart failure (CHF) is commonly associated with atrial fibrillation (AF), and their combined presentation confers a worse prognosis than either condition alone.1 Treatment of both conditions implies use of specific drugs for CHF plus antithrombotic agents for stroke prevention. In addition, management strategies and outcomes may be affected by the etiology of CHF, namely ischemic cardiomyopathy (ICM) or nonischemic cardiomyopathy (NICM) because the prescription of antithrombotic therapies might be different and could affect prognosis in terms of death, stroke/systemic embolism (SE), and bleeding.2,3 The aim of our study was to assess the treatment strategies in terms of antithrombotic and CHF therapies and 1-year clinical outcomes in patients with newly diagnosed AF and concomitant CHF stratified by etiology (ICM vs NICM) enrolled in the Global Anticoagulant Registry in the Field–Atrial Fibrillation (GARFIELD-AF) registry.

Methods

The design of the GARFIELD-AF registry was reported previously.4,5 Briefly, men and women 18 years and older with AF diagnosed according to standard local procedures within the previous 6 weeks and with at least 1 nonprespecified risk factor for stroke as judged by the local investigator and no valvular disease were eligible for inclusion.5 Patients were enrolled prospectively and consecutively in 35 countries. When random site selection did not generate the required number of sites in a given country, the national lead investigator was asked to recommend sites to make up the numbers (18 of 1317 sites). The sites represent the different care settings in each participating country (office-based practice; hospital departments, including neurology, cardiology, geriatrics, internal medicine, and emergency; anticoagulation clinics; and general or family practice).4,5 Except for Egypt and South Africa, there were no other participant countries from Africa.

Independent ethics committee and hospital-based institutional review board approvals were obtained. The registry is being conducted in accordance with the principles of the Declaration of Helsinki, local regulatory requirements, and the International Conference on Harmonization–Good Pharmacoepidemiological and Clinical Practice guidelines. Written informed consent was obtained from all study participants. Confidentiality and anonymity of all patients recruited into this registry are maintained.

Procedures and Outcome Measures

Baseline characteristics collected at inclusion in the registry included medical history, care setting, type of AF, date and method of diagnosis, symptoms, antithrombotic treatment (ie, vitamin K antagonists, nonvitamin K antagonist oral anticoagulants, and antiplatelets [AP]), and all cardiovascular drugs. Race/ethnicity was classified by the investigator in agreement with the patient.5 Data on components of the CHA2DS2-VASc and HAS-BLED risk stratification schemes were collected. The HAS-BLED scores were calculated excluding fluctuations in international normalized ratio.

Collection of follow-up data was performed every 4 months up to 12 months.4,5 Standardized definitions for clinical events have been reported previously.4,5 In brief, baseline characteristics and treatments and the incidences of death (cardiovascular and noncardiovascular), stroke/SE, and bleeding were recorded. Submitted data were examined for completeness and accuracy by the coordinating center (Thrombosis Research Institute, London, England), and data queries were sent to study sites. The GARFIELD-AF data are captured using an electronic case report form. In accordance with the study protocol, 20% of all electronic case report forms were monitored against source documentation.6 Data for the analysis in this report were extracted from the study database on October 18, 2017.

Definitions

Heart failure was defined as current/prior history of CHF or left ventricular ejection fraction (LVEF) of less than 40%. Congestive heart failure was diagnosed according to clinical criteria at entry and classified according to the New York Heart Association (NYHA) functional class. Information about LVEF assessed by 2-dimensional echocardiography was available in 73% of patients with CHF and was stratified as preserved LVEF when this was 40% or higher and as reduced LVEF when less than 40%. Ischemic cardiomyopathy was defined as patients with a history of coronary artery disease, including known chronic angina pectoris, previous myocardial infarction or unstable angina, coronary artery bypass grafting, or previous percutaneous coronary intervention with or without stenting. Nonischemic cardiomyopathy was the default diagnosis. Worsening HF was defined as progressive or acute decompensation of previously stable CHF. Vascular disease included coronary artery disease with a history of acute coronary syndromes and/or peripheral artery disease. Chronic kidney disease (CKD) was classified according to National Kidney Foundation guidelines into moderate-to-severe (stages 3-5), mild (stages 1 and 2), or none.

Statistical Analysis

Continuous variables were expressed as medians and interquartile ranges (IQR) and categorical variables as frequencies and percentages. Because studies with large sample sizes are prone to producing statistically significant findings in the presence of clinically irrelevant differences, no formal statistical tests are performed for the baseline tables. Instead, clinical interpretations are applied, and statistical tests are reserved for hypotheses of differences in outcomes.

All cohorts of patients had been followed up for a minimum of 1 year. Thus, events and time to events are censored at 365 days. Occurrence of major clinical outcomes was expressed as person-time event rates (per 100 person-years) and 95% confidence intervals. Person-year rates for the first occurrence of the event of interest were estimated using a Poisson model, with the number of events as the dependent variable and the log of time as an offset, ie, a covariate with a known coefficient of 1.

Evaluation of 1-year outcomes by patients with ischemic vs nonischemic HF were performed both as unadjusted and adjusted analyses using Cox proportional hazards models. In addition to the unadjusted person-time event rates, the hazard ratios with 95% confidence intervals are provided for both unadjusted and adjusted results. Adjustment factors, as determined in previous models of the key outcomes, were age with spline knots at 48, 65, 79, and 88 years; sex; race/ethnicity; smoking status; diabetes; hypertension; history of bleeding; moderate-to-severe CKD; use of oral anticoagulation therapy (OAC) at baseline; type of AF; and heavy use of alcohol.

The association of type of HF with outcomes in patients who used AP therapy was also of interest. The inclusion of the main effect of AP alone vs AP therapy plus OAC and the interaction of ischemic vs NICM by AP alone vs AP plus OAC was evaluated in unadjusted and adjusted analyses. The adjustment factors included those listed previously, except OAC use, which appears in the treatment of interest. The test for interaction specifies whether the association between patients with ICM and NICM varies according to treatment received. Hazard ratios with 95% confidence intervals of the difference outcomes for patients with ICM vs NICM within each of the 2 treatment strategies are derived from this model. All statistical analyses were performed using SAS, version 9.4 (SAS Institute), while forest plots were created using SigmaPlot, version 12.5 (Microsoft). The log-rank test was used to compare the survival distribution rates between ICM and NICM groups. A P value of less than .05 was considered statistically significant.

Results

Baseline Characteristics

The population of this study included all 5 cohorts of the GARFIELD-AF registry (eFigure 1 in the Supplement), totaling 52 014 patients recruited between March 2010 and August 2016 and followed up prospectively for a minimum of 1 year. Detailed baseline characteristics of the population are reported in eTable 1 in the Supplement. The median age of the population was 71.0 years, and 22 987 of 52 013 (44.2%) were women. Concomitant CHF was present in 11 738 of 52 007 patients (22.5%), coronary artery disease in 11 232 of 52 007 (21.6%), and history of hypertension in 38 585 of 51 863 (73.4%). Prior stroke/transient ischemic attack was present in 5954 of 52 007 patients (11.4%), history of bleeding in 1317 (2.5%), diabetes in 11 540 (22.2%), and moderate-to-severe CKD in 5373 (11.7%). At enrollment, 4717 patients (9.1%) had ICM and 7021 (13.5%) had NICM (eTable 1 in the Supplement).

Compared with patients with NICM, patients with ICM were more frequently white (74.5% [3465 of 4651] vs 55.8% [3848 of 6897]) and less frequently non-Chinese Asian (12.8% [596] vs 29.5% [2035]), had a higher mean body weight (80.7 kg vs 77.2 kg), had a similar mean LVEF (45.9% vs 46.9%), had a higher rate of depressed LVEF (32.4% [1170 of 3606] vs 37.8% [1892 of 5005]), were less likely to have permanent or persistent AF (30.4% [1434 of 4717] vs 37.3% [2619 of 7021]), and more frequently had NYHA class III/IV CHF (37.0% [1468 of 3964] vs 28.8% [1589 of 5533]). In addition, compared with patients with NICM, patients with ICM were more likely to have carotid occlusive disease (4.3% [199 of 4608] vs 2.4% [169 of 6932]), history of stroke/transient ischemic attack (12.6% [596 of 4717] vs 9.7% [684 of 7021]), hypertension (84.6% [3985 of 4710] vs 73.1% [5121 of 7004]), hypercholesterolemia (55.7% [2530 of 4545] vs 34.0% [2303 of 6782]), type 2 diabetes (28.3% [1335 of 4717] vs 20.2% [1415 of 7021]), and moderate-to-severe CKD (16.2% [764 of 4717] vs 13.7% [960 of 7020]), had a higher mean CHA2DS2-VASc score (4.4 vs 3.8), and were treated more frequently in cardiology (74.5% [3514 of 4717] vs 68.5% [4812 of 7021]) (Table 1).

Table 1. Baseline Characteristics of Patients With Ischemic and Nonischemic Cardiomyopathy.

Characteristic No. (%)
Ischemic Cardiomyopathy (n = 4717) Nonischemic Cardiomyopathy (n = 7021)
Sex
Male 2814 (59.7) 3892 (55.4)
Female 1903 (40.3) 3128 (44.6)
Age at diagnosis, y
Mean (SD) 70.2 (10.6) 70.0 (12.3)
Median (IQR) 71.0 (63.0-78.0) 71.0 (62.0-79.0)
Race/ethnicity
White 3465 (74.5) 3848 (55.8)
Hispanic/Latino 150 (3.2) 557 (8.1)
Afro-Caribbean 11 (0.2) 46 (0.7)
Asian (not Chinese) 596 (12.8) 2035 (29.5)
Chinese 337 (7.2) 267 (3.9)
Mixed/other 92 (2.0) 144 (2.1)
Height, cm
Mean (SD) 167.7 (9.6) 165.5 (11.0)
Median (IQR) 168.0 (160.0-175.0) 165.0 (158.0-173.0)
Weight, kg
Mean (SD) 80.7 (18.7) 77.2 (21.4)
Median (IQR) 80.0 (68.0-92.0) 75.0 (62.0-90.0)
BMI
Mean (SD) 28.6 (5.8) 28.0 (6.4)
Median (IQR) 28.0 (25.0-32.0) 27.0 (24.0-31.0)
Pulse, bpm
Mean (SD) 88.6 (25.1) 95.2 (27.6)
Median (IQR) 82.0 (70.0-100.0) 90.0 (75.0-112.0)
Systolic BP, mm Hg
Mean (SD) 132.0 (19.7) 132.0 (20.9)
Median (IQR) 130.0 (120.0-142.0) 130.0 (120.0-144.0)
Diastolic BP, mm Hg
Mean (SD) 79.8 (12.7) 80.2 (14.1)
Median (IQR) 80.0 (70.0-90.0) 80.0 (70.0-90.0)
LVEF, %
Mean (SD) 46.9 (14.1) 45.9 (15.3)
Median (IQR) 48.0 (35.0-58.0) 45.0 (34.0-59.0)
LVEF category
<40% 1170 (32.4) 1892 (37.8)
≥40% 2436 (67.6) 3113 (62.2)
Care setting specialty at diagnosis
Internal medicine 794 (16.8) 1363(19.4)
Cardiology 3514 (74.5) 4812 (68.5)
Neurology 27 (0.6) 60 (0.9)
Geriatrics 9 (0.2) 43 (0.6)
Primary care/general practice 373 (7.9) 743 (10.6)
Type of AF diagnosed
Permanent 710 (15.1) 1126 (16.0)
Persistent 724 (15.3) 1493 (21.3)
Paroxysmal 1025 (21.7) 1395 (19.9)
New onset (unclassified) 2258 (47.9) 3007 (42.8)
Congestive heart failure NYHA class
I 544 (13.7) 1281 (23.2)
II 1952 (49.2) 2663 (48.1)
III 1285 (32.4) 1348 (24.4)
IV 183 (4.6) 241 (4.4)
Unknown, No. 294 606
Coronary artery disease 4365 (92.5) NA
Acute coronary syndrome 1695 (36.7) NA
Vascular disease 2200 (47.3) 327 (4.7)
Coronary artery bypass graft 490 (10.8) NA
PCI with stenting 1041 (22.1) NA
Carotid occlusive disease 199 (4.2) 169 (2.4)
Stroke/TIA 596 (12.6) 684 (9.7)
Systemic embolization 26 (0.6) 45 (0.6)
History of bleeding 160 (3.4) 193 (2.8)
History of hypertension 3985 (84.6) 5121 (73.1)
Hypercholesterolaemia 2530 (55.7) 2303 (34.0)
Diabetes
Type 1 52 (1.1) 62 (0.9)
Type 2 1335 (28.3) 1415 (20.2)
Alcohol consumption
Abstinent 2284 (55.3) 3411 (55.8)
Light 1456 (35.3) 1822 (29.8)
Moderate 317 (7.7) 667 (10.9)
Heavy 70 (1.7) 216 (3.5)
Unknown, No. 590 905
Smoker
No 2727 (62.0) 4229 (65.4)
Former smoker 1155 (26.3) 1473 (22.8)
Current smoker 518 (11.8) 769 (11.9)
Chronic renal disease
I 816 (20.0) 1545 (24.9)
II 639 (15.6) 882 (14.2)
III 635 (15.5) 793 (12.8)
IV 95 (2.3) 129 (2.1)
V 34 (0.8) 38 (0.6)
None 1868 (45.7) 2821 (45.4)
Moderate-to-severe CKD 764 (16.2) 930 (13.7)
CHA2DS2-VASc score
Mean (SD) 4.4 (1.6) 3.8 (1.5)
Median (IQR) 4.0 (3.0-5.0) 4.0 (3.0-5.0)
HAS-BLED score
Mean (SD) 1.8 (0.9) 1.4 (0.9)
Median (IQR) 2.0 (1.0-2.0) 1.0 (1.0-2.0)
Open in a new tab

Abbreviations: AF, atrial fibrillation; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); BP, blood pressure; CKD, chronic kidney disease; IQR, interquartile range; LVEF, left ventricular ejection fraction; NA, not applicable; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; TIA, transient ischemic attack.

Antithrombotic Therapies

Oral anticoagulants alone or in combination with AP were prescribed in fewer patients with ICM than patients with NICM (alone: ICM, 31.5% [1443 of 4579]; NICM, 61.2% [4225 of 6898]; combined: ICM, 60.1% [2753 of 4579]; NICM, 73.7% [5082 of 6898]). Patients with ICM were more likely to receive AP alone than patients with NICM (34.4% [1576 of 4579] vs 15.5% [1071 of 6898]). Dual AP therapy was given in 580 of 4681 patients with ICM (12.4%) and in 103 of 6961 with NICM (1.5%). No antithrombotic drugs were prescribed in 250 of 4579 patients with ICM (5.5%) and 745 of 6898 with NICM (10.8%) (Table 2). The prescription patterns were remarkably similar across the various categories of risk of stroke as defined by CHA2DS2-VASc score (0-1, 2-4, and >4), with OAC with or without AP rates approximately 60% in patients with ICM vs 70% in patients with NICM and AP only rates approximately 30% in patients with ICM vs 15% in patients with NICM (Figure 1). The prescription patterns were also mostly similar in cardiology, internal medicine, and primary care settings; the rates of AC with or without AP were approximately 60% for patients with ICM and 74% for patients with NICM, and rates of AP were approximately 35% and 15%, respectively. There were too few patients in neurology and geriatrics settings to have a reliable picture of the prescription patterns (eFigure 2 in the Supplement). Lastly, in emergency departments, hospitals, and office settings, rates of AC with or without AP were approximately 60% in patients with ICM and 70% in patients with NICM, and rates of AP were approximately 30% and 15%, respectively. In anticoagulation clinics, most patients received AC with or without AP, primarily in the form of vitamin K antagonists (eFigure 3 in the Supplement).

Table 2. Baseline Treatments (GARFIELD-AF Cohorts 1 to 5) and AF Management at Diagnosis (GARFIELD-AF Cohorts 3 to 5).

Treatment No. (%)
Ischemic Cardiomyopathy Nonischemic Cardiomyopathy
Cohorts 1 to 5
No. 4717 7021
Antithrombotic drugs
AC 1443 (31.5) 4225 (61.2)
AC with or without AP 2753 (60.1) 5082 (73.7)
VKA 880 (19.2) 2440 (35.4)
VKA+AP 887 (19.4) 627 (9.1)
FXA 391 (8.5) 1342 (19.5)
FXA+AP 318 (6.9) 173 (2.5)
DTI 172 (3.8) 443 (6.4)
DTI+AP 105 (2.3) 57 (0.8)
AP only 1576 (34.4) 1071 (15.5)
None 250 (5.5) 745 (10.8)
Unknown, No. 138 123
Heart failure/hypertension drugs
ACEi/ARBs 3439 (72.9) 4236 (60.3)
β-Blockers 2988 (63.3) 3737 (53.2)
Loop and other diuretics 2753 (58.4) 3897 (55.5)
MRA 979 (20.8) 1225 (17.4)
Digoxin and other digitalis 471 (10.0) 662 (9.4)
Other drugs of interest
Oral antidiabetics 461 (13.8) 416 (9.3)
Insulin 241 (7.2) 153 (3.4)
α-Blockers 98 (2.9) 131 (2.9)
Cohorts 3 to 5
No. 3332 4459
AF treatment/strategy initiated
None declared 194 (6.0) 398 (9.2)
Rhythm control 1111 (34.5) 939 (21.7)
Rate control 1262 (39.2) 2197 (51.0)
Both 651 (20.2) 777 (18.0)
Unknown, No. 114 148
Cardioversion
No 2815 (86.5) 3787 (88.4)
Yes 440 (13.5) 498 (11.6)
Currenta 124 (33.8) 234 (62.4)
Pharmacologicala 242 (66.1) 141 (37.6)
Antiarrhythmic drugs
Class Ia 22 (0.7) 22 (0.5)
Class Ic 59 (1.8) 113 (2.5)
Class II (β-blockers) 1492 (44.8) 1842 (41.3)
Class III 568 (17.1) 358 (8.0)
Class IV (calcium-channel blocker) 395 (11.9) 608 (13.6)
Class V (digoxin and other digitalis) 624 (18.7) 939 (21.1)
Open in a new tab

Abbreviations: AC, anticoagulants; ACEi/ARB, angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; AF, atrial fibrillation; AP, antiplatelets; DTI, direct thrombin inhibitor; FXaI, factor Xa inhibitor; MRA, mineralocorticoid receptor antagonist; VKA, vitamin K antagonist.

a

Patients submitted to cardioversion during the first month after inclusion in GARFIELD-AF. Patients who underwent both current and pharmacologic cardioversion are not reported here.

Figure 1. Percentage of Prescription of Anticoagulants and Antiplatelets (AP) and Their Combination in Ischemic Cardiomyopathy (ICM) and Nonischemic Cardiomyopathy (NICM) According to CHA2DS2-VASc Score.

Figure 1.

Open in a new tab

DTI indicates direct thrombin inhibitor; FXA, factor Xa inhibitor; VKA, vitamin K anticoagulation.

Treatment of AF

Complete information about treatment and drug therapies of AF were available for the last 3 cohorts of 34 854 patients (Table 2), of whom 3332 had ICM and 4459 had NICM. Rate control was attempted more frequently than rhythm control in both groups. The uptake of antiarrhythmic drugs was mostly similar in both groups, except that class 3 drugs were more frequently used in patients with ICM. Cardioversion was attempted in a minority of patients, at a similar rate in both groups. Patients with NICM were more likely to receive electric cardioversion than patients with ICM.

HF Drugs

The use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers was higher in patients with ICM vs patients with NICM (72.9% [3439 of 4717] vs 60.3% [4236 of 7021]). The same was observed for β-blocker use (63.3% [2988 of 4717] vs 53.2% [3737 of 7021], respectively). The use of mineralocorticoid receptor antagonists, loop diuretics, and digoxin and other digitalis was similar between the 2 groups (Table 2).

Clinical Outcomes at 1-Year Follow-up

Incidence rates per 100 person-years of all-cause and cardiovascular death were significantly higher in patients with ICM compared with NICM (all-cause death: ICM, 10.2; 95% CI, 9.2-11.1; NICM, 7.0; 95% CI, 6.4-7.6; P <.001; cardiovascular death: ICM, 5.1; 95% CI, 4.5-5.9; NICM, 2.9; 95% CI, 2.5-3.4; P <.001) (Table 3 and Figure 2A). Rates of noncardiovascular death were similar in both cardiomyopathy groups.

Table 3. Incidence Rates per 100 Person-Years of Clinical Outcomes at 1-Year Follow-up.

Variable Ischemic Cardiomyopathy (n = 4717) Nonischemic Cardiomyopathy (n = 7021) P Valuea
All-cause mortality
No. (%) 443 (9.4) 456 (6.5) NA
Incidence rate (95% CI) 10.2 (9.2-11.1) 7.0 (6.4-7.6) <.001
Cardiovascular mortality
No. (%) 224 (4.7) 191 (2.7) NA
Incidence rate (95% CI) 5.1 (4.5-5.9) 2.9 (2.5-3.4) <.001
Noncardiovascular mortality
No. (%) 102 (2.2) 157 (2.2) NA
Incidence rate (95% CI) 2.3 (1.9-2.9) 2.4 (2.1-2.8) .84
Undetermined cause of mortality
No. (%) 117 (2.5) 108 (1.5) NA
Incidence rate (95% CI) 2.7 (2.2-3.2) 1.7 (1.4-2.0) <.001
Stroke/SE
No. (%) 87 (1.8) 99 (1.4) NA
Incidence rate (95% CI) 2.0 (1.6-2.5) 1.5 (1.3-1.9) .06
Major bleeding
No. (%) 46 (1.0) 44 (0.6) NA
Incidence rate (95% CI) 1.1 (0.8-1.4) 0.7 (0.5-0.9) .03
New ACS
No. (%) 81 (1.7) 48 (0.7) NA
Incidence rate (95% CI) 1.9 (1.5-2.3) 0.7 (0.6-1.0) <.001
Worsening CHF
No. (%) 265 (5.6) 444 (6.3) NA
Incidence rate (95% CI) 6.3 (5.6-7.1) 7.1 (6.5-7.8) .79
Open in a new tab

Abbreviations: ACS, acute coronary syndrome; CHF, chronic heart failure; NA, not applicable; SE, systemic embolism.

a

P value is log-rank test.

Figure 2. Unadjusted and Adjusted Hazard Ratios (HRs) for 1-Year Outcomes.

Figure 2.

Open in a new tab

A, Unadjusted and adjusted HRs for clinical outcomes at 1 year. Rates were adjusted for age, sex, race/ethnicity, smoking, diabetes, hypertension, history of bleeding, moderate-to-severe kidney disease, type of medication, type of atrial fibrillation, and heavy alcohol use. B, Adjusted HRs for patients with ischemic cardiomyopathy (ICM) vs patients with nonischemic cardiomyopathy (NICM) according to antiplatelet (AP) treatment. Rates were adjusted for age, sex, race/ethnicity, smoking, diabetes, hypertension, history of bleeding, moderate-to-severe kidney disease, type of atrial fibrillation, and heavy alcohol use. ACS indicates acute coronary syndrome; CHF, congestive heart failure; SE, systemic embolism.

Incidence rates of stroke/SE tended to be higher in the ICM group (2.0; 95% CI, 1.6-2.5) compared with the NICM group (1.5; 95% CI, 1.3-1.9). Incidence rates of major bleeding were significantly higher in the ICM group than in the NICM group. New acute coronary syndromes occurred at a significantly higher rate in the ICM group (1.9; 95% CI, 1.5-2.3) compared with the NICM group (0.7; 95% CI, 0.6-1.0). Worsening CHF occurred at similar rates in both ICM and NICM groups.

The most frequent classified causes of cardiovascular death were CHF, unwitnessed/sudden death, and myocardial infarction, which tended to be balanced in the 2 CHF groups. The causes of deaths classified as noncardiovascular were also balanced between the ICM and NICM groups and were mostly linked to malignancy, respiratory failure, and infection/sepsis. Nevertheless, there was some imbalance in the rates of other and undetermined causes of death between the 2 groups; in particular, the rate of undetermined cause of death was substantially higher in ICM group (eTable 2 in the Supplement).

AP Therapy and 1-Year Clinical Outcomes

The adjusted hazard ratios for ICM vs NICM in patients who received AP therapy and in patients who did not receive AP therapy at enrollment are presented for mortality, stroke/SE, and major bleeding in Figure 2B. Ischemic cardiomyopathy was significantly associated with higher rates of all-cause death, cardiovascular death, and major bleeding in patients who did not receive AP therapy at enrollment. No such excess of risk in ICM vs NICM was observed among patients who received AP therapy.

Discussion

The main findings of this study are that patients with ICM and patients with NICM and concomitant AF substantially differ in terms of baseline characteristics, functional impairment, LV function impairment, and treatments that may be significantly associated with outcomes. They also significantly differ in terms of outcome, with higher rates of death, major bleeding, and new acute coronary syndromes. The known causes of cardiovascular death were proportionally mostly similar in both groups, including death from ischemic stroke (eTable 2 in the Supplement).

In patients without AF, a higher risk of death in patients with ICM compared with patients with NICM was inconsistently reported in observational studies, in retrospective analyses of hospital records, and in reviews of randomized clinical trials.7,8,9,10,11,12,13 The discrepancies between reports were thought to be associated with inconsistency in diagnosing ischemic CHF in the absence of rigorous discriminant definitions of ischemic and nonischemic CHF, with patient selection, or with imbalance in baseline characteristics, with more severe comorbidities in ischemic vs nonischemic CHF.10 The poorer prognosis of ICM was attributed to differences in structural alterations between the 2 entities, with more regional wall motion abnormalities, asynchrony of LV contraction, extensive scar tissue, and poorer LV compliance in addition to a greater extent of coronary artery disease and residual ischemia in ICM vs NICM.14

A limited number of reports addressed the outcome of patients with CHF and concomitant AF. They showed that concomitant AF was associated with a higher risk of death in ICM only.2,3 Atrial fibrillation in patients with ischemic heart disease may contribute to impaired myocardial perfusion, coronary vasoconstriction, and further deterioration of diastolic function. In addition, impaired myocardial perfusion is associated with electrical instability and a higher risk of sudden death.15,16

In this report, patients with ICM had more advanced deterioration of their cardiac condition, as shown by a higher NYHA functional class, a higher rate of depressed LV function (about two-thirds of patients with CHF had preserved LVEF, but patients with ICM had higher rates of reduced LVEF), and higher rates of comorbidities, particularly coronary artery disease and CKD, which might have contributed to a worse prognosis. Treatment strategy of AF and use of antiarrhythmic drugs were mostly similar in both groups, but there was some imbalance between the 2 groups regarding rate or rhythm control and use of class 3 antiarrhythmic drugs. Antithrombotic therapy was suboptimal in the ICM group compared with the NICM group, as shown by a much lower rate of anticoagulation with or without AP therapy and, consequently, a high rate of AP alone prescription even in patients without indication for AP therapy, as already reported17 (Table 2). This may have had a detrimental effect on the risk of death because anticoagulation was shown to be associated with a substantial reduction of the risk of death in previous reports.18,19 In the Reduction of Atherothrombosis for Continued Health Registry (REACH) trial,20 where the incidence of AF in patients with coronary artery disease was 12.5%, it was shown that only 53% of these patients were given anticoagulation therapy whereas 40% were receiving AP agents alone. Additionally, there were higher rates of heart failure and cardiovascular death in the AF population than in the non-AF population.

The reasons for inadequate antithrombotic treatment in the ICM group remain to be determined. These patients, often elderly, need treatments for stroke prevention, heart failure, and coronary disease, resulting in an extensive list of drugs, which may deter the physician from prescribing them all and deter the patient from complying with the prescription.21 Physicians’ adherence to guideline recommendations remains a matter of debate.22,23 In a previous report of the GARFIELD registry,19 physicians’ decision was the most common reason for not prescribing OACs in patients with AF. Lack of adherence may have played a role because the implementation of guidelines was shown to have a favorable association with outcomes in other settings.24,25 In this context, an intriguing fact needs to be mentioned. In patients who did not receive AP therapy, the risks of all-cause and cardiovascular death as well as the risk of major bleeding were significantly higher in those with ICM than in those with NICM; in contrast, in patients who did receive AP therapy, there was no excess of events in those with ICM (Figure 2B). This tends to confirm that AP therapy is associated with a more positive outcome in patients with ischemic heart disease. However, these data were generated from a nonprespecified subgroup analysis. As such, no firm conclusions can be derived, and these data are only hypothesis generating.

There was a substantial but not significant excess of stroke/SE and a significant excess of major bleeding risks in patients with ICM compared with patients with NICM (Table 3 and Figure 2A). Congestive heart failure without AF is associated with a higher risk of stroke, and the combination of CHF and AF further increases this risk irrespective of LV function (preserved or depressed) or of etiology (ischemic or nonischemic).26,27,28 A higher mean CHA2DS2-VASc score and a higher incidence of carotid artery disease in the ICM group may also explain the excess of stroke/SE. Furthermore, as already mentioned, stroke prophylaxis was suboptimal in the ICM group, where twice as many patients received AP therapy alone compared with the NICM group. In addition, prescribing rates of OAC alone or in combination with AP therapy were substantially lower in patients with ICM than in those with NICM (Table 2). These prescription patterns were observed irrespective of CHA2DS2-VASc score, medical specialty, and care setting. Combination therapy is known to increase the risk of bleeding,29,30 and AP therapy alone does not confer any protection against stroke/SE but increases bleeding risk.31,32 This may explain the higher risk of major bleeding in patients with ICM in this study.

Finally, to our knowledge, this study is the first to assess the outcomes of ICM and NICM in patients with AF from an unselected prospective registry population representative of real life. Ischemic cardiomyopathy carried a poorer prognosis than NICM in terms of all-cause and cardiovascular death. It was also associated with a substantial but not significant excess of stroke/SE and significantly higher risk of bleeding. In addition, patients with ICM were suboptimally treated with regards to stroke prophylaxis, which may have affected outcomes. Polypharmacy and/or poor adherence to guideline-recommended therapies may explain the inadequacy of treatment in ischemic CHF.

Limitations

Our study had limitations. Ischemic cardiomyopathy was not defined by the investigator but retrospectively based on the history of coronary artery disease (ie, known chronic angina pectoris, previous myocardial infarction or unstable angina, coronary artery bypass grafting, or previous percutaneous coronary intervention with or without stenting), low LVEF, and signs/symptoms of CHF. Nevertheless, we cannot rule out that some patients in the NICM group had ischemic heart disease. The population included in this analysis is mostly made of white and Asian individuals, so the results may not be relevant for other racial/ethnic groups.

Conclusions

More than 22% of the population of patients with newly diagnosed AF included in the GARFIELD-AF registry had concomitant CHF, of whom around 40% had ICM. They were suboptimally given anticoagulation therapy and more often received AP therapy alone or in combination with anticoagulants. These patients had worse outcomes than patients with NICM in terms of all-cause and cardiovascular death despite the fact that they were more likely to receive angiotensin-converting enzyme inhibitors/angiotensin receptor blockers than patients with NICM. These findings should prompt physicians to opt for active treatment of both AF and CHF, particularly in patients with ICM.

Supplement.

eTable 1. Baseline characteristics of all patients.

eTable 2. Cardiovascular and noncardiovascular causes of death in patients with IHD and non-IHD.

eFigure 1. GARFIELD-AF study population.

eFigure 2. Rates of prescription of anticoagulants and antiplatelet and their combination in ischemic and nonischemic cardiomyopathy according to specialty.

eFigure 3. Rates of prescription of anticoagulants and antiplatelet and their combination in ischemic and nonischemic cardiomyopathy according to care setting.

Click here for additional data file. (151.4KB, pdf)

References

  • 1.Wang TJ, Larson MG, Levy D, et al. . Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart Study. Circulation. 2003;107(23):2920-2925. doi: 10.1161/01.CIR.0000072767.89944.6E [DOI] [PubMed] [Google Scholar]
  • 2.Pedersen OD, Søndergaard P, Nielsen T, et al. ; DIAMOND Study Group Investigators . Atrial fibrillation, ischaemic heart disease, and the risk of death in patients with heart failure. Eur Heart J. 2006;27(23):2866-2870. doi: 10.1093/eurheartj/ehl359 [DOI] [PubMed] [Google Scholar]
  • 3.Raunsø J, Pedersen OD, Dominguez H, et al. ; EchoCardiography and Heart Outcome Study Investigators . Atrial fibrillation in heart failure is associated with an increased risk of death only in patients with ischaemic heart disease. Eur J Heart Fail. 2010;12(7):692-697. doi: 10.1093/eurjhf/hfq052 [DOI] [PubMed] [Google Scholar]
  • 4.Kakkar AK, Mueller I, Bassand JP, et al. ; GARFIELD Registry Investigators . Risk profiles and antithrombotic treatment of patients newly diagnosed with atrial fibrillation at risk of stroke: perspectives from the international, observational, prospective GARFIELD registry. PLoS One. 2013;8(5):e63479. doi: 10.1371/journal.pone.0063479 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kakkar AK, Mueller I, Bassand JP, et al. . International longitudinal registry of patients with atrial fibrillation at risk of stroke: Global Anticoagulant Registry in the FIELD (GARFIELD). Am Heart J. 2012;163(1):13-19.e1. doi: 10.1016/j.ahj.2011.09.011 [DOI] [PubMed] [Google Scholar]
  • 6.Fox KAA, Gersh BJ, Traore S, et al. ; GARFIELD-AF Investigators . Evolving quality standards for large-scale registries: the GARFIELD-AF experience. Eur Heart J Qual Care Clin Outcomes. 2017;3(2):114-122. [DOI] [PubMed] [Google Scholar]
  • 7.Gajanana D, Shah M, Junpapart P, Romero-Corral A, Figueredo VM, Bozorgnia B. Mortality in systolic heart failure revisited: ischemic versus non-ischemic cardiomyopathy. Int J Cardiol. 2016;224:15-17. doi: 10.1016/j.ijcard.2016.08.316 [DOI] [PubMed] [Google Scholar]
  • 8.Bart BA, Shaw LK, McCants CB Jr, et al. . Clinical determinants of mortality in patients with angiographically diagnosed ischemic or nonischemic cardiomyopathy. J Am Coll Cardiol. 1997;30(4):1002-1008. doi: 10.1016/S0735-1097(97)00235-0 [DOI] [PubMed] [Google Scholar]
  • 9.Felker GM, Thompson RE, Hare JM, et al. . Underlying causes and long-term survival in patients with initially unexplained cardiomyopathy. N Engl J Med. 2000;342(15):1077-1084. doi: 10.1056/NEJM200004133421502 [DOI] [PubMed] [Google Scholar]
  • 10.Follath F, Cleland JG, Klein W, Murphy R. Etiology and response to drug treatment in heart failure. J Am Coll Cardiol. 1998;32(5):1167-1172. doi: 10.1016/S0735-1097(98)00400-8 [DOI] [PubMed] [Google Scholar]
  • 11.Lourenço C, Saraiva F, Martins H, et al. . Ischemic versus non-ischemic cardiomyopathy: are there differences in prognosis? experience of an advanced heart failure center. Rev Port Cardiol. 2011;30(2):181-197. [PubMed] [Google Scholar]
  • 12.Mościński M, Szyguła-Jurkiewicz B, Zakliczyński M, et al. . Five-year survival of patients with chronic systolic heart failure of ischemic and non-ischemic etiology: analysis of prognostic factors. Kardiochir Torakochirurgia Pol. 2014;11(1):56-62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ng AC, Sindone AP, Wong HS, Freedman SB. Differences in management and outcome of ischemic and non-ischemic cardiomyopathy. Int J Cardiol. 2008;129(2):198-204. doi: 10.1016/j.ijcard.2007.07.014 [DOI] [PubMed] [Google Scholar]
  • 14.Kwon DH, Hachamovitch R, Popovic ZB, et al. . Survival in patients with severe ischemic cardiomyopathy undergoing revascularization versus medical therapy: association with end-systolic volume and viability. Circulation. 2012;126(11, suppl 1):S3-S8. doi: 10.1161/CIRCULATIONAHA.111.084434 [DOI] [PubMed] [Google Scholar]
  • 15.Hohendanner F, Heinzel FR, Blaschke F, et al. . Pathophysiological and therapeutic implications in patients with atrial fibrillation and heart failure. Heart Fail Rev. 2018;23(1):27-36. doi: 10.1007/s10741-017-9657-9 [DOI] [PubMed] [Google Scholar]
  • 16.Patel RB, Vaduganathan M, Shah SJ, Butler J. Atrial fibrillation in heart failure with preserved ejection fraction: insights into mechanisms and therapeutics. Pharmacol Ther. 2017;176:32-39. doi: 10.1016/j.pharmthera.2016.10.019 [DOI] [PubMed] [Google Scholar]
  • 17.Verheugt FWA, Gao H, Al Mahmeed W, et al. ; GARFIELD-AF Investigators . Characteristics of patients with atrial fibrillation prescribed antiplatelet monotherapy compared with those on anticoagulants: insights from the GARFIELD-AF registry. Eur Heart J. 2018;39(6):464-473. doi: 10.1093/eurheartj/ehx730 [DOI] [PubMed] [Google Scholar]
  • 18.Bassand JP, Accetta G, Al Mahmeed W, et al. ; GARFIELD-AF Investigators . Risk factors for death, stroke, and bleeding in 28,628 patients from the GARFIELD-AF registry: rationale for comprehensive management of atrial fibrillation. PLoS One. 2018;13(1):e0191592. doi: 10.1371/journal.pone.0191592 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bassand JP, Accetta G, Camm AJ, et al. ; GARFIELD-AF Investigators . Two-year outcomes of patients with newly diagnosed atrial fibrillation: results from GARFIELD-AF. Eur Heart J. 2016;37(38):2882-2889. doi: 10.1093/eurheartj/ehw233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Goto S, Bhatt DL, Röther J, et al. ; REACH Registry Investigators . Prevalence, clinical profile, and cardiovascular outcomes of atrial fibrillation patients with atherothrombosis. Am Heart J. 2008;156(5):855-863, 863.e2. doi: 10.1016/j.ahj.2008.06.029 [DOI] [PubMed] [Google Scholar]
  • 21.Cobretti MR, Page RL II, Linnebur SA, et al. . Medication regimen complexity in ambulatory older adults with heart failure. Clin Interv Aging. 2017;12:679-686. doi: 10.2147/CIA.S130832 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Shanbhag D, Graham ID, Harlos K, et al. . Effectiveness of implementation interventions in improving physician adherence to guideline recommendations in heart failure: a systematic review. BMJ Open. 2018;8(3):e017765. doi: 10.1136/bmjopen-2017-017765 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Van Spall HGC, Rahman T, Mytton O, et al. . Comparative effectiveness of transitional care services in patients discharged from the hospital with heart failure: a systematic review and network meta-analysis. Eur J Heart Fail. 2017;19(11):1427-1443. doi: 10.1002/ejhf.765 [DOI] [PubMed] [Google Scholar]
  • 24.Fonarow GC, Albert NM, Curtis AB, et al. . Incremental reduction in risk of death associated with use of guideline-recommended therapies in patients with heart failure: a nested case-control analysis of IMPROVE HF. J Am Heart Assoc. 2012;1(1):16-26. doi: 10.1161/xJAHA.111.000018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Mehta RH, Chen AY, Alexander KP, Ohman EM, Roe MT, Peterson ED. Doing the right things and doing them the right way: association between hospital guideline adherence, dosing safety, and outcomes among patients with acute coronary syndrome. Circulation. 2015;131(11):980-987. doi: 10.1161/CIRCULATIONAHA.114.013451 [DOI] [PubMed] [Google Scholar]
  • 26.Abdul-Rahim AH, Perez AC, Fulton RL, et al. ; Investigators of the Controlled Rosuvastatin Multinational Study in Heart Failure (CORONA); GISSI-Heart Failure (GISSI-HF) Committees and Investigators . Risk of stroke in chronic heart failure patients without atrial fibrillation: analysis of the Controlled Rosuvastatin in Multinational Trial Heart Failure (CORONA) and the Gruppo Italiano per lo Studio della Sopravvivenza nell’Insufficienza Cardiaca-Heart Failure (GISSI-HF) trials. Circulation. 2015;131(17):1486-1494. doi: 10.1161/CIRCULATIONAHA.114.013760 [DOI] [PubMed] [Google Scholar]
  • 27.Abdul-Rahim AH, Perez AC, MacIsaac RL, et al. ; Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity-Preserved (CHARM-Preserved) and the Irbesartan in Heart Failure With Preserved Systolic Function (I-Preserve) Steering Committees . Risk of stroke in chronic heart failure patients with preserved ejection fraction, but without atrial fibrillation: analysis of the CHARM-Preserved and I-Preserve trials. Eur Heart J. 2017;38(10):742-750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kim W, Kim EJ. Heart failure as a risk factor for stroke. J Stroke. 2018;20(1):33-45. doi: 10.5853/jos.2017.02810 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lamberts M, Lip GY, Ruwald MH, et al. . Antithrombotic treatment in patients with heart failure and associated atrial fibrillation and vascular disease: a nationwide cohort study. J Am Coll Cardiol. 2014;63(24):2689-2698. doi: 10.1016/j.jacc.2014.03.039 [DOI] [PubMed] [Google Scholar]
  • 30.Kumar S, Danik SB, Altman RK, et al. . Non-vitamin K antagonist oral anticoagulants and antiplatelet therapy for stroke prevention in patients with atrial fibrillation: a meta-analysis of randomized controlled trials. Cardiol Rev. 2016;24(5):218-223. doi: 10.1097/CRD.0000000000000088 [DOI] [PubMed] [Google Scholar]
  • 31.Golive A, May HT, Bair TL, et al. . The population-based long-term impact of anticoagulant and antiplatelet therapies in low-risk patients with atrial fibrillation. Am J Cardiol. 2017;120(1):75-82. doi: 10.1016/j.amjcard.2017.03.256 [DOI] [PubMed] [Google Scholar]
  • 32.Lip GY, Skjøth F, Rasmussen LH, Larsen TB. Oral anticoagulation, aspirin, or no therapy in patients with nonvalvular AF with 0 or 1 stroke risk factor based on the CHA2DS2-VASc score. J Am Coll Cardiol. 2015;65(14):1385-1394. doi: 10.1016/j.jacc.2015.01.044 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eTable 1. Baseline characteristics of all patients.

eTable 2. Cardiovascular and noncardiovascular causes of death in patients with IHD and non-IHD.

eFigure 1. GARFIELD-AF study population.

eFigure 2. Rates of prescription of anticoagulants and antiplatelet and their combination in ischemic and nonischemic cardiomyopathy according to specialty.

eFigure 3. Rates of prescription of anticoagulants and antiplatelet and their combination in ischemic and nonischemic cardiomyopathy according to care setting.

Click here for additional data file. (151.4KB, pdf)

Articles from JAMA Cardiology are provided here courtesy of American Medical Association

RESOURCES