Atrial fibrillation (AF) and heart failure (HF) are the two major cardiovascular diseases with a rapidly increasing prevalence worldwide. The prevalence of both diseases increases dramatically with aging.1,2 AF and HF not only frequently coexist but also increase the risk for one another. Both systolic and diastolic dysfunction could result in an increase in atrial pressure and volume overload, cellular hypertrophy, myocardial fibrosis, and conduction disturbances, and they have also been shown to be associated with a higher risk for AF. The prevalence of AF in published large clinical trials varies between 9.7% and 49.8%, depending on the severity of HF.3–7 On the contrary, AF can aggravate HF, leading to hemodynamic compromise via the mechanism of inappropriate acceleration of ventricular rate, loss of atrial contraction, and elevated filling pressures. Previous population-based studies have reported that between 59% and 76% of patients with AF and HF develop AF either before or concurrently with onset of HF.8,9 The recent ORBIT-AF registry showed that incident HF developed in 3.6% of patients with AF during the 2-year follow-up period.10
Although the interrelationships between AF and HF could be associated with a vicious cycle, there is no consensus regarding whether AF is an independent prognostic risk factor for HF or only a marker of merely more advanced disease. In the COMET study, patients with HF and permanent AF developed worse outcomes than patients in sinus rhythm, whereas after multivariate adjustment, AF alone did not predict a higher mortality.11 In contrast, the results of the SOLVD trial, which enrolled 6500 patients with left ventricular ejection fraction (LVEF) <35%, demonstrated that baseline AF was an independent predictor of all-cause mortality, progressive pump failure, and the composite end point of death or hospitalization due to HF.12 Treatment strategies of AF in patients with HF remain controversial. Results of the AF-CHF trial demonstrated that rhythm-control strategy failed to reduce death due to cardiovascular causes compared with rate-control strategy in patients with both AF and HF.13 Potential benefits occurring via the maintenance of sinus rhythm might be neutralized by the adverse effects of antiarrhythmic drugs.
The Taiwan Society of Cardiology-Heart Failure with reduced Ejection Fraction (TSOC-HFrEF) registry specifically enrolled decompensated hospitalized patients with LVEF <40%. This is the largest national database till date involving patients with acute decompensated HFrEF in Taiwan.14 Using the TSOC-HFrEF database, we conducted this study to evaluate the characteristics, treatment, and outcomes of patients with HF with different AF classifications.
2.1. Study design and patient characteristics
The TSOC-HFrEF registry was a prospective, multicenter, observational survey of patients who were hospitalized with either acute new-onset HF or acute decompensation of chronic HFrEF from 21 hospitals between 2013 and 2014 in Taiwan. The institutional review board of each hospital agreed to participate in the registry. The enrollment of patients, the overall characteristics of the patient population, and the management during index hospitalization and follow-up have been described in previous articles.14,15
There were no specific exclusion criteria, except that all patients should be aged >18 years and their LVEF had to be documented at <40% before enrollment. There were no specific protocols or recommendation for evaluation and management of HF during this observational study.
Information about AF was based on medical history. Investigators were asked to state whether there was a history of AF, and if yes, whether the AF was paroxysmal or persistent or permanent. For analysis, patients with a history of AF were divided into the following two groups according to their AF status: patients with paroxysmal AF (PAF group) and patients with nonparoxysmal AF (N-PAF group). Baseline characteristics, comorbidities, in-hospital management, discharge medications, and 1-year outcomes were analyzed in both groups. Specific 1-year outcomes included readmission for worsening HF, HF-related mortality (either due to refractory progressive HF, arrhythmic death, or sudden death), and non-HF-related mortality. Data were collected centrally using an electronic, standardized case report form and sent electronically to the data collection center.
2.2. Electrocardiogram, echocardiography, and laboratory studies
Electrocardiogram (ECG) and echocardiographic examinations were performed for all patients. Complete 12-lead ECG was done by a standard method with a paper speed of 25 mm per second. The first ECGs (either in the emergency room or at admission) of each patient were collected for analysis. Rhythm, heart rate, and QRS duration were recorded, and specific ECG findings such as left bundle branch block, left ventricular hypertrophy, and pathologic Q wave were marked. Echocardiographic images were acquired and measured in each institute. Left ventricular end-diastolic diameter was measured at end-diastole, and left ventricular end-systolic diameter and left atrial anteroposterior dimension were measured at end-systole on parasternal views. LVEF was calculated using the biplane Simpson’s method on apical 4-chamber and 2-chamber views.
Baseline laboratory data (either in the emergency room or at admission) were collected for analysis, except for serum blood urea nitrogen and creatinine levels, which were collected before discharge. Glomerular filtration rate (GFR) was calculated using the modification of diet in renal disease formula.
2.3. Statistical analysis
Quantitative data were expressed as mean ± SD or as median and interquartile range, and categorical variables were presented as percentages. Descriptive summaries were presented for different groups of patients. The Student’s t-test or the Mann–Whitney U test was used for comparisons between continuous data, and a χ2 test was used for comparisons between categorical data. A Kaplan–Meier survival analysis was used to plot the survival curves. Multivariate Cox regression analysis with forward selection was performed to assess the predictability of variables on 1-year outcome presented as hazard ratios (HRs) and 95% CIs using p < 0.1 in univariate analyses for inclusion. p < 0.05 was considered to be statistically significant. All tests were two-sided. All the statistical analyses were performed using the SPSS Statistics 17.0 software (Chicago, IL, USA).
3.1. General information
A total of 1509 hospitalized patients (aged 63.9 ± 16.1 years) were included in the TSOC-HFrEF registry from May 2013 to October 2014. Among these patients, 393 (26%) had AF. These patients were further divided into the following two groups based on their AF status: 117 patients with PAF and 276 patients with non-PAF.
3.2. Differences in baseline characteristics
The differences in baseline characteristics of patients in the TSOC-HFrEF registry based on their AF status are shown in Table 1. In general, age and gender were similar in both groups. Regarding the etiology of HF, patients with PAF more frequently presented with ischemic cardiomyopathy (47.3% vs 29.7%, p = 0.021). Medical history revealed that patients with PAF more frequently presented with a history of myocardial infarction (29.9% vs 20.3%, p = 0.039), peripheral arterial disease (46.2% vs 29.0%, p = 0.001), and chronic kidney disease (46.2% vs 29.0%, p = 0.001). However, the incidence rates of a prior stroke were similar (14.5%) in both groups. Patients in the PAF group showed a slightly increased prevalence of hypertension (35% vs 27.9%) and diabetes mellitus (47.9% vs 38%) compared with those in the N-PAF group, which was not statistically different. Patients with PAF had higher a CHA2DS2-VASc score (4.0 vs 3.6, p = 0.033) and a higher HASBLED score (3.1 vs 2.6, p = 0.015) than patients with N-PAF.
3.3. Differences in cardiovascular examinations and laboratory tests
Differences in cardiovascular examinations of patients in the current registry based on their AF status are presented in Table 2, and differences in their laboratory tests are shown in Table 3. Regarding the ECG parameters, the mean heart rate, QRS duration, the percentages of left bundle branch block, left ventricular hypertrophy, and pathologic Q wave were comparable in both groups. The 24-hour Holter monitor data demonstrated that the PAF group had a significantly higher percentage of baseline sinus rhythm than the N-PAF group (p < 0.001). However, the mean 24-hour heart rate, ventricular premature beats counts, atrial premature beats counts, and the percentage of nonsustained ventricular tachycardia were similar in both groups. Echocardiographic examinations revealed that both groups had similar LVEF and left ventricular diameters, whereas patients with N-PAF had significantly larger left atrial diameter (50.5 vs 47.3 mm, p = 0.004), compared with those of patients with PAF. The percentages of moderate or severe mitral regurgitation, aortic regurgitation, tricuspid regurgitation, and aortic stenosis were comparable in both groups.
Patients with PAF had higher blood urea nitrogen and creatinine levels and lower estimated GFR and hemoglobin levels than patients with N-PAF. The levels of electrolytes, blood glucose, liver function, uric acid, natriuretic peptides, and cardiac troponin-I were similar in both groups.
3.4. Differences in in-hospital managements and guideline-recommended medications at discharge
Results of the comparisons of in-hospital management and discharge medications between both groups are shown in Table 4. Vital signs at admission and discharge were comparable in both groups. Approximately 30% to 40% of patients admitted in the intensive care unit and patients in the PAF group were more likely to receive endotracheal tube intubation and mechanical ventilator support compared with patients with N-PAF (19.7% vs 10.1%, p = 0.031). In-hospital mortality rates were 5.1% in the PAF group and 2.5% in the N-PAF group of patients (p = 0.22).
At discharge, patients with PAF were more likely to receive treatment with amiodarone (31.6% vs 13.8%, p < 0.001) and antiplatelet agents (54.1% vs 42.5%, p = 0.041) but less likely to receive renin–angiotensin system (RAS) blockers (52.3% vs 64.9%, p = 0.021), digoxin (26.1% vs 46.6%, p < 0.001), and anticoagulants (33.3% vs 50%, p = 0.003) compared with patients in the N-PAF group. The prescription rates of beta-blockers, mineralocorticoid receptor antagonists, and diuretics were comparable between patients in both groups.
3.5. One-year outcomes and predictors
At 6 and 12 months after hospital discharge, the all-cause mortality rates were 11.5% and 19.3% and the HF-related mortality rates were 7.8% and 12.0%, respectively. The 1-year all-cause mortality (26.2% vs 16.5%, p = 0.024) and non-HF-related mortality rates (13.1% vs 5.0%, p = 0.005) were significantly higher in patients with PAF, whereas the HF-related mortality rates were similar in both groups (13.1% vs 11.5%, p = 0.701). The Kaplan–Meier survival curves are shown in the Figure.
Among all patients with AF in the TSOC-HFrEF registry, the rates of rehospitalization for worsening HF were 30.7% and 37.1% at 6 and 12 months after index hospitalization, respectively. The 1-year readmission rates for HF in both groups were similar (39.4% in PAF and 36.2% in N-PAF patients, p = 0.55). The numbers of HF rehospitalizations were comparable in both groups (1.8 ± 1.2 times in PAF vs 1.7 ± 0.9 times in N-PAF patients, p = 0.447).
To further clarify the impact of AF type on the 1-year outcome, multivariate Cox regression analysis of factors associated with 1-year mortality was performed, and the results are shown in Table 5. Model 1 included baseline characteristics such as age, AF types, CHA2DS2-VASc score, underlying diseases, and body mass index. AF type (PAF vs N-PAF, HR 1.62, 95% CI 1.02-2.59, p = 0.044) and higher CHA2DS2-VASc score (HR 1.16, 95% CI 1.01-1.33, p = 0.038) could predict the 1-year mortality in this model. In model 2, the same baseline characteristics used in model 1 were included, as well as the medical therapy. Model 2 demonstrated that prescription of less than two types of guideline-recommended medical therapy (including RAS blockers, beta-blockers, and mineralocorticoid receptor antagonists, HR 1.96, 95% CI 1.24-3.11, p = 0.004) and a history of chronic kidney disease (HR 1.77, 95% CI 1.12-2.80, p = 0.014) could independently predict the 1-year outcome, and PAF was no longer associated with 1-year mortality.
4.1. Prevalence of AF in HF studies
The prevalence of AF among patients with HF varies in previously published HF studies, depending on the trials’ inclusion criteria, age, gender, HF severity, and comorbidities. In the CONSENSUS trial, which enrolled patients with the most severe HF (NYHA class IV), AF was detected in 49.8% of patients.7 On the contrary, previous large-scale HF registries, including the ADHERE registry from the United States of America,16 the EHFS II survey from Europe,17 and the ATTEND registry from Japan,18 enrolled not only HFrEF but also 34% to 47% of patients with HF with preserved LV ejection fraction. Consequently, patients included in these three international registries were significantly older and the prevalence rates of female gender and hypertension were significantly higher than those among patients in our TSOC-HFrEF registry.13 Since AF was largely associated with old age, female gender, and diastolic dysfunction, the prevalence of AF from these three registries ranged from 31% to 40% and was therefore higher than the prevalence in our current registry (26%). Our registry enrolled only hospitalized patients with HFrEF.
4.2. Baseline characteristics and potential effects on treatment
Ischemia was universally the most common cause of HF among the registries.14,17,18 In the TSOC-HFrEF registry, we found that patients with PAF had a significantly higher prevalence of atherosclerotic disease, such as ischemic cardiomyopathy, history of myocardial infarction, and peripheral arterial disease than patients with N-PAF. The prevalence of a previous stroke was about 14.5% and was similar in both groups. The CHA2DS2-VASc scores were high in both groups. Although we did not classify stroke as atherosclerotic or embolic, according to the underlying disease and scoring system, the risk of future stroke was very high; hence, aggressive medical treatment to prevent stroke in our patient population was mandatory.
Analysis of discharge medications revealed that patients with PAF were more likely to receive antiplatelet agents (54.1% vs 42.5%, p = 0.041) but less likely to receive anticoagulants (33.3% vs 50%, p = 0.003) compared with patients with N-PAF. Although current guidelines recommend anticoagulants for the prevention of thromboembolism for all patients with AF with a CHA2DS2-VASc score ≥ 2, except in case of contraindications,19,20 physicians in the current registry preferred to prescribe fewer anticoagulants and more antiplatelet agents, especially for patients with PAF who also had a high prevalence of atherosclerotic disease. In parallel to the CHA2DS2-VASc scores, the HASBLED scores were high in the registry population and were even higher in patients with PAF than in patients with N-PAF (3.1 ± 1.6 vs 2.6 ± 1.6, p = 0.015), indicating that patients with PAF had a higher risk of bleeding than those with N-PAF. We were not aware whether the nonprescription of anticoagulants was due to neglect of stroke risk, fear of bleeding, or other adverse drug events because the reasons for nonprescription were not collected in our current registry.
The majority of past history findings and patient characteristics, including smoking, alcohol consumption, histories of previous HF hospitalization, valvular heart disease, hypertension, diabetes, and previous device implantation, were similar in both groups. However, patients with PAF had a significantly higher prevalence of chronic kidney disease than those with N-PAF. Because of this, patients with PAF had higher blood urea nitrogen and creatinine levels and lower estimated GFR than patients with N-PAF. The prevalence of renal impairment and end-stage renal disease in patients in Taiwan is very high. A report from Taiwan Renal Registry Data System demonstrated that the prevalence of renal replacement therapy was 2926 per million of the population in 2012.21 The overall prevalence of chronic renal failure in the TSOC-HFrEF registry was 31.5%,14 which was much higher than the data reported in the previous European and Asian national surveys.17,22–24 Renal dysfunction was found to be associated with underutilization of RAS blockers in the current registry,25 and patients with PAF were less likely to receive RAS blockers (52.3% vs 64.9%, p = 0.021) compared with patients with N-PAF at discharge.
4.3. Rhythm control vs rate control for AF and HF
Although a routine strategy of rhythm control does not reduce the rate of cardiovascular mortality compared with a rate-control strategy,13 the CAFÉ-II study reported that restoration to sinus rhythm was associated with improvement in quality of life and LVEF.26 These conflicting findings suggest that the benefits of rhythm control could be counterbalanced by the adverse effects of antiarrhythmic therapy.27 Since dronedarone and class I antiarrhythmic agents are not recommended for patients with HF,19 and sotalol and dofetilide are not available in most of the hospitals in Taiwan, amiodarone was the only effective drug for rhythm control in our registry.
At discharge, prescription rates of beta-blockers were similar in both groups, but patients with PAF were more likely to receive amiodarone (31.6% vs 13.8%, p < 0.001) and less likely to receive digoxin (26.1% vs 46.6%, p < 0.001) compared with patients in the N-PAF group, indicating that more physicians applied rhythm-control strategy on patients with HF with PAF. Although amiodarone could reduce the incidence of AF, induce pharmacological cardioversion, and more efficiently maintain sinus rhythm in patients with HF and AF,28 the notorious side effects, including thyroid, pulmonary, and corneal complications, limit its usage. At baseline, 4.7% of patients in the TSOC-HFrEF registry had either hyperthyroidism or hypothyroidism, which may have been further deteriorated after treatment with amiodarone.14
4.4. Prognostic significance of AF
The prognostic significance of AF in patients with HF remains controversial. In the V-HeFT study, the presence of AF was not associated with a worse outcome in 1427 patients with mild-to-moderate HF.4 In contrast, the SOLVD trial demonstrated that the odds ratio for total mortality among patients with HF with AF compared with patients in sinus rhythm was 1.81 (p < 0.0001).12 In a retrospective analysis of the COMET study, which included 3029 patients with LVEF < 35%, baseline AF was found to significantly increase the risk for death and HF readmission. However, after adjustment for other predictors, AF was no longer an independent risk factor for mortality.11 In the TRACE study, long-term mortality was increased in all subgroups of patients with AF, except in those with the most advanced disease (LVEF < 25%), suggesting that the independent effect of AF on mortality is inversely related to the severity of HF.29
Few HF studies have compared the outcomes of patients with different AF types. In the current study, we found that patients with PAF had significantly higher 1-year mortality than patients with persistent or permanent AF. This result appeared to be paradoxical from the point of view of disease progression, as patients with PAF have less advanced atrial disease and smaller left atrial size than patients with N-PAF. After multivariate adjustment of baseline characteristics, PAF was found to be still associated with higher 1-year mortality. Similar to our findings, the PARADIGM-HF and ATMOSPHERE trials showed that patients with PAF had higher risks for primary composite endpoint, HF hospitalization, and stroke than patients with N-PAF after multivariate adjustment, although cardiovascular death rates were similar in both groups.30 Regarding the baseline characteristics in these two large-scale randomized trials, patients with PAF presented more frequently with ischemic HF etiology (63.6% vs 50.6%, p < 0.001), history of myocardial infarction (49.2% vs 30.0%, p < 0.001), renal disease (21.8% vs 17.3%, p < 0.001), and higher CHA2DS2-VASc score (4.1 vs 3.9, p = 0.005) compared with patients with N-PAF; these results were similar to the findings in our registry. Even in the randomized controlled trials, medical treatments were not equally prescribed; patients with PAF were more likely to receive amiodarone and antiplatelet agents but less likely to receive digoxin and anticoagulants compared with patients with N-PAF.30 Hence, in the current registry data, if guideline-recommended treatments were taken into consideration in the multivariate analysis, PAF would no longer be associated with 1-year mortality, whereas prescription of less than two types of guideline-recommended medical therapy could predict a worse 1-year outcome. Considering these findings together, we noticed that there is a gap between guideline-recommended therapy and real-world practices, which applies to both the randomized controlled trials and our observational registry. The importance of anticoagulants and other evidential HF treatments in patients with concurrent AF and HF should not be neglected.
4.5. Study limitations
Several limitations were present in this study. First, the design of the current registry was an observational, prospective survey that included only hospitalized patients with reduced ejection fraction. Despite covariate adjustment, other measured or unmeasured factors might also affect outcomes. Second, according to the study design, we collected patients’ death mode as follows: death due to refractory HF, death due to arrhythmia or sudden cardiac death and noncardiac death. Fatal stroke could be categorized into non-HF-related death in this study, but it is an important endpoint for those patients with HF with AF. Hospitalization due to stroke and systemic embolism was not recorded; hence, it was difficult to describe the effect of under-usage of guideline-recommended treatment, especially anticoagulants.
In conclusion, The TSOC-HFrEF registry is the largest national database till date involving patients with acute decompensated HFrEF in Taiwan. In this observational, real-world registry, patients with PAF and HFrEF were less likely to receive anticoagulant therapy and more likely to receive antiarrhythmic agents but they had worse 1-year outcome than their N-PAF counterparts. Our study findings further emphasize the importance of optimal guideline-recommended medical therapy in patients with HFrEF, regardless of the underlying AF types.
1. Friberg J, Buch P, Scharling H, Gadsbphioll N, Jensen GB. Rising rates of hospital admissions for atrial fibrillation.Epidemiology200314666–72
2. Mosterd A, Hoes AW. Clinical epidemiology of heart failure.Heart2007931137–46
3. The SOLVD InvestigatorsEffect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure.N Engl J Med1991325293–302
4. Carson PE, Johnson GR, Dunkman WB, Fletcher RD, Farrell L, Cohn JN. The influence of atrial fibrillation on prognosis in mild to moderate heart failure. The V-HeFT studies. The V-HeFT VA Cooperative Studies Group.Circulation199387VI102–10
5. Pfeffer MA, Swedberg K, Granger CB, Held P, McMurray JJ, Michelson EL, et al. Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programme.Lancet2003362759–66
6. Lechat P, Hulot JS, Escolano S, Mallet A, Leizorovicz A, Werhlen-Grandjean M, et al. Heart rate and cardiac rhythm relationships with bisoprolol benefit in chronic heart failure in CIBIS II Trial.Circulation20011031428–33
7. The CONSENSUS trial study groupEffects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS).N Engl J Med19873161429–35
8. Wang TJ, Larson MG, Levy D, Vasan RS, Leip EP, Wolf PA, et al. Temporal relations of atrial fibrillation congestive heart failure their joint influence on mortality: the Framingham Heart Study.Circulation20031072920–5
9. Smith JG, Newton-Cheh C, Almgren P, Struck J, Morgenthaler NG, Bergmann A, et al. Assessment of conventional cardiovascular risk factors and multiple biomarkers for the prediction of incident heart failure and atrial fibrillation.J Am Coll Cardiol2010561712–9
10. Pandey A, Kim S, Moore C, Thomas L, Gersh B, Allen LA, et al. Predictors and prognostic implications of incident heart failure in patients with prevalent atrial fibrillation.JACC Heart Fail2017544–52
11. Swedberg K, Olsson LG, Charlesworth A, Cleland J, Hanrath P, Komajda M, et al. Prognostic relevance of atrial fibrillation in patients with chronic heart failure on long–term treatment with beta–blockers: results from COMET.Eur Heart J2005261303–8
12. Dries DL, Exner DV, Gersh BJ, Domanski MJ, Waclawiw MA, Stevenson LW. Atrial fibrillation is associated with an increased risk for mortality and heart failure progression in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: a retrospective analysis of the SOLVD trials. Studies of left ventricular dysfunction.J Am Coll Cardiol199832695–703
13. Roy D, Talajic M, Nattel S, Wyse DG, Dorian P, Lee KL, et al. Rhythm control versus rate control for atrial fibrillation and heart failure.N Engl J Med20083582667–77
14. Wang CC, Chang HY, Yin WH, Wu YW, Chu PH, Wu CC, et al. TSOC-HFrEF registry: a registry of hospitalized patients with decompensated systolic heart failure: description of population and management.Acta Cardiol Sin201632400–11
15. Chang HY, Wang CC, We YW, Chu PH, Wu CC, Hsu CH, et al. One-year outcomes of acute decompensated systolic heart failure in Taiwan
: lessons from TSOC-HFrEF registry.Acta Cardiol Sin201733127–38
16. Adams KF Jr, Fonarow GC, Emerman CL, LeJemtel TH, Costanzo MR, Abraham WT, et al. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE).Am Heart J2005149209–16
17. Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Follath F, Harjola VP, et al. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population.Eur Heart J2006272725–36
18. Sato N, Kajimoto K, Keida T, Mizuno M, Minami Y, Yumino D, et al. Clinical features and outcome in hospitalized heart failure in Japan (from the ATTEND Registry).Circ J201377944–51
19. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC.Eur Heart J2016372129–200
20. Wang CC, Chen JH, Yu WC, Cheng JJ, Yin WH, Wu CC, et al. 2012 Guidelines of the Taiwan
society of cardiology (TSOC) for the diagnosis and treatment of heart failure.Acta Cardiol Sin201228161–95
21. Lin YC, Hsu CY, Kao CC, Chen TW, Chen HH, Hsu CC, et al. Incidence and prevalence of ESRD in Taiwan
Renal Registry Data System (TWRDS): 2005–2012.Acta Nephrologica20142865–8
22. Tsuchihashi-Makaya M, Hamaguchi S, Kinugawa S, Yokota T, Goto D, Yokoshiki H, et al. Characteristics and outcomes of hospitalized patients with heart failure and reduced vs preserved ejection fraction: report from the Japanese Cardiac Registry of Heart Failure in Cardiology (JCARE-CARD).Circ J2009731893–900
23. Youn YJ, Yoo BS, Lee JW, Kim JY, Han SW, Jeon ES, et al. Treatment performance measures affect clinical outcomes in patients with acute systolic heart failure: report from the Korean Heart Failure Registry.Circ J2012761151–8
24. Hai JJ, Chan PH, Huang D, Ho MH, Ho CW, Cheung E, et al. Clinical characteristics, management, and outcomes of hospitalized heart failure in a Chinese population-The Hong Kong Heart Failure Registry.J Card Fail201622600–8
25. Chang HY, Wang CC, Wei J, Chang CY, Chuang YC, Huang CL, et al. Gap between guidelines and clinical practice in heart failure with reduced ejection fraction: Results from TSOC-HFrEF registry.J Chin Med Assoc201780750–7
26. Shelton RJ, Clark AL, Goode K, Rigby AS, Houghton T, Kaye GC, et al. A randomized, controlled study of rate versus rhythm control in patients with chronic atrial fibrillation and heart failure: (CAFÉ-II Study).Heart200995924–30
27. Mamas MA, Caldwell JC, Chacko S, Garratt CJ, Fath–Ordoubadi F, Neyses L. A meta-analysis of the prognostic significance of atrial fibrillation in chronic heart failure.Eur J Heart Fail200911676–83
28. Deedwania PC, Singh BN, Ellenbogen K, Fisher S, Fletcher R, Singh SN. Spontaneous conversion and maintenance of sinus rhythm by amiodarone in patients with heart failure and atrial fibrillation: observations from the veterans affairs congestive heart failure survival trial of antiarrhythmic therapy (CHF-STAT). The Department of Veterans Affairs CHF–STAT Investigators.Circulation1998982574–9
29. Pedersen OD, Bagger H, Køber L, Torp-Pedersen C; for the TRACE Study GroupImpact of congestive heart failure and left ventricular systolic function on the prognostic significance of atrial fibrillation and atrial flutter following acute myocardial infarction.Int J Cardiol200510065–71
30. Mogensen UM, Jhund PS, Abraham WT, Desai AS, Dickstein K, Packer M, et al. Type of atrial fibrillation and outcomes in patients with heart failure and reduced ejection fraction.J Am Coll Cardiol2017702490–500