Early Rhythm Control and the Risks of Ischemic Stroke, Heart Failure, Mortality, and Adverse Events When Performed Early (<3 Months): A Nationwide Cohort Study of Newly Diagnosed Patients with Atrial Fibrillation

Abstract

Background In the Early Treatment of Atrial Fibrillation for Stroke Prevention Trial (EAST-AFNET 4), early rhythm control was associated with better clinical outcomes for patients with atrial fibrillation (AF). However, the intervention arm had more structured and regular follow-up, and whether the better outcomes of patients assigned to rhythm control were solely due to “early” intervention or because of more regular and structured follow-up was unclear.

Objective In the present study, we aimed to investigate whether the findings of the EAST trial are applicable to the “real-world” clinical setting, where a less structured management protocol is operated.

Methods From 2001 to 2016, 301,064 newly diagnosed AF patients were identified from Taiwan National Health Insurance Research Database. Among these patients, 62,649 AF patients receiving antiarrhythmic drugs or catheter ablation within 1 year after AF being diagnosed (similar to the timing definition of the EAST-AFNET 4 trial) were categorized as the early rhythm control group, and the remaining 238,415 patients were defined as usual care group. Risks of clinical events were compared between the two groups.

Results Compared with usual care, early rhythm control was associated with a lower adjusted risk of ischemic stroke (adjusted hazard ratio [aHR]: 0.771, 95% confidence interval [CI]: 0.751–0.792; p < 0.001), heart failure (aHR: 0.851, 95% CI: 0.806–0.899; p < 0.001), acute myocardial infarction (aHR: 0.915, 95% CI: 0.877–0.955; p < 0.001), mortality (aHR: 0.794, 95% CI: 0.782–0.806; p < 0.001), and composite adverse events (aHR: 0.823, 95% CI: 0.813–0.834; p < 0.001). Compared with usual care, the lower risks of ischemic stroke (aHR: 0.746, 95% CI: 0.717–0.775), heart failure (aHR: 0.819, 95% CI: 0.798–0.841), mortality (aHR: 0.777, 95% CI: 0.759–0.795), and composite adverse events (aHR: 0.802, 95% CI: 0.787–0.818) associated with early rhythm control were even more evident when performed early (<3 months) compared to later periods (3–6 months, 7–9 months, and 10–12 months; p _int < 0.001). Principal results were generally consistent for majority of subgroups studied and among the cohort after propensity matching.

Conclusion In this nationwide cohort study, early rhythm control therapy was associated with a lower risk of adverse events compared with usual care among patients with early AF. Outcomes were even better with earlier (<3 months) intervention.

Keywords

atrial fibrillation - early rhythm control - ischemic stroke - mortality

Introduction

Atrial fibrillation (AF) is the commonest cardiac rhythm disorder, with a high risk of mortality and morbidity from ischemic stroke, heart failure, and myocardial infarction, despite oral anticoagulation.[1] Efforts to improve clinical outcomes in AF, especially the residual risk despite anticoagulation, are needed. Integrated care management is based on appropriate characterization[2] and holistic care according to the ABC (Atrial fibrillation Better Care) pathway,[3] implementing “A” Avoid stroke/Anticoagulation, “B” Better symptom management with patient-centered symptom-directed decisions on rate or rhythm control, and “C” Cardiovascular risk and comorbidity optimization.[4] [5] Older randomized trials comparing rate control or rhythm controls strategies have shown that both are noninferior to each other.[6] [7] However, more recent studies have shown the benefits of rhythm control in selected AF patient subgroups, for example, those with heart failure, as shown in the CASTLE-AF trial.[8]

Early intervention with rhythm control may confer additional benefits by modulating electrical and structural remodeling. In the Early Treatment of Atrial Fibrillation for Stroke Prevention Trial (EAST-AFNET 4), early rhythm control was associated with a significant reduction in the primary composite outcome of death, stroke, or serious adverse events related to rhythm control therapy.[9] However, the intervention arm had more structured follow-up with regular (two times per week) and symptom-driven patient-operated electrocardiography recordings with transmission to the study team triggering an extra clinical visit as needed.

First, whether the findings of the EAST trial are applicable to the “real-world” clinical setting, where a less structured management protocol is operated, requires exploration, for the outcomes of ischemic stroke, heart failure, acute myocardial infarction (AMI), mortality, and composite adverse events. Second, the median time since diagnosis was 36 days in the EAST trial, so we wished to explore whether “early” rhythm control can be considered up to 1 year since AF diagnosis. Third, clinical outcomes of patients managed in different time periods, 2001–2004, 2005–2008, 2009–2012, and 2013–2016, were compared. To address these questions, we performed a nationwide cohort study investigating newly diagnosed patients with AF (i.e., <1 year after incident AF was diagnosed) undergoing early rhythm control compared with usual care.

Methods

Database

This study used the “National Health Insurance Research Database (NHIRD)” provided by the Health and Welfare Data Science Center, Ministry of Health and Welfare, Taiwan. The National Health Insurance (NHI) system is a mandatory universal health insurance program that offers comprehensive medical care coverage to all Taiwanese residents. NHIRD consists of detailed health care data from over 23 million enrollees, representing more than 99% of Taiwan's population. In this cohort dataset, the patients' original identification numbers have been encrypted to protect their privacy, but the encrypting procedure was consistent, so that a linkage of the claims belonging to the same patient was feasible within the NHI database and can be followed continuously. The descriptions about Taiwan NHIRD have been reported in our previous studies.[10] [11] [12] [13] [14] The research reported in this article adhered to the Helsinki Declaration as revised in 2013.

Study Population

From January 1, 2001 to December 31, 2016, a total of 449,015 incident AF patients aged ≥20 years were identified from Taiwan NHIRD as the study population. AF was diagnosed using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (427.31) registered by the physicians responsible for the treatments of patients. The diagnostic accuracy of AF using this definition in NHIRD has been validated previously.[15] Among the study population, patients who experienced mortality (n = 78,601), or ischemic stroke (n = 16,997), or heart failure (n = 55,783), or AMI (n = 4,280) within 1 year after being identified were excluded. Finally, 301,064 patients remained and constituted the study cohort. Among the study cohort, 62,649 of them who received antiarrhythmic drugs (AADs) or catheter ablation within 1 year after AF being diagnosed (similar to the timing definition of the EAST-AFNET 4 trial) were categorized as the early rhythm control group, and the remaining 238,415 patients were defined as the usual care group. The flowchart of study design and patient enrollment is shown in [Fig. 1].

Fig. 1 A flowchart of the enrollment of the study cohort. AF, atrial fibrillation; NHIRD, National Health Insurance Research Database.

Definitions and the Comparisons of Risks of Clinical Endpoints

The clinical endpoints included ischemic stroke, heart failure hospitalizations, AMI, all-cause mortality, and composite risks of ischemic stroke, or heart failure, or AMI, or mortality. Ischemic stroke was diagnosed using ICD-9-CM codes, with concomitant imaging studies, including computed tomography or magnetic resonance imaging. The accuracy of diagnosis of ischemic stroke in Taiwan's NHIRD has been reported to be approximately 94%.[16] Another validation study also demonstrated that the diagnostic accuracy of ischemic stroke in NHIRD was high, with the positive predictive value and sensitivity of 88.4 and 97.3%, respectively.[17] The diagnostic accuracies of heart failure and AMI of Taiwan NHIRD have also been validated previously.[18] [19] Each individual endpoint (ischemic stroke, heart failure, AMI, and mortality) was analyzed independently of the others without being censored.

The risks of clinical events were compared between “early rhythm control” and “usual care” groups among the whole study cohort (years 2001–2016) and those identified in different time intervals (years 2001–2004, 2005–2008, 2009–2012, and 2013–2016). Also, patients in the “early rhythm control” groups were further categorized into four groups based on the timing of initiations of rhythm control treatments with AADs or receiving catheter ablations after AF being diagnosed, that is, <3 months, 3–6 months, 7–9 months, and 10–12 months. The risks of clinical events of patients in these four groups were compared with that of usual care.

Falsification Analysis (Sensitivity Analysis)

To further assess the likelihood of confounding by indication, we analyzed three falsification endpoints which were unlikely to be affected by “early rhythm control” or “usual care”—urinary tract infection, cellulitis, and acute appendicitis. A finding of an association between “early rhythm control” or “usual care” and the falsification endpoints would therefore indicate the presence of unmeasured confounders. On the contrary, if risks of these falsification endpoints of the two groups did not differ significantly, the differences between “early rhythm control” and “usual care” with regard to clinical outcomes in which we were interested may be less likely due to treatment selection bias.

Propensity Match Analysis

We performed propensity score–matched analyses for AF patients receiving “early rhythm control” and “usual care.” We calculated propensity scores for the likelihoods of receiving early rhythm control compared with usual care by multivariate logistic regression analyses, conditional on all baseline covariates except for medications listed in [Table 1]. After that, we matched patients in the early rhythm control group to those in the usual care group with a 1:1 ratio on the basis of the closest propensity score for receiving early control within a threshold of ± 0.01. If more than one patient in the usual care group could be matched to the corresponding subject in the early rhythm control group, one patient from the usual care group was selected randomly without repeat sampling. The matching processes were performed for four times for AF patients diagnosed in different time periods (years 2001–2004, 2005–2008, 2009–2012, and 2013–2016).

Table 1 Baseline characteristics of incident AF patients

Variables	All, n = 301,064	Usual care, n = 238,415	Early rhythm control, n = 62,649	p-Value
Age (y), mean value (SD)	69.44 (13.32)	69.74 (13.58)	68.3 (12.23)	<0.0001
Age > 75 y, n (%)	117,698 (39.09)	97101 (40.73)	20,597 (32.88)	<0.0001
Age 65–74 y, n (%)	74,755 (24.83)	57,372 (24.06)	17,383 (27.75)	<0.0001
Sex (male), n (%)	169,624 (56.34)	134,839 (56.56)	34,785 (55.52)	0.423
CHA2DS2-VASc score, mean (SD)	3.11 (1.96)	3.15 (1.97)	2.99 (1.91)	<0.0001
HAS-BLED score, mean (SD)	2.43 (1.41)	2.42 (1.42)	2.51 (1.39)	<0.0001
Comorbidities, n (%)
Hypertension	194,753 (64.69)	152,603 (64.01)	42,150 (67.28)	<0.0001
Diabetes mellitus	80,032 (26.58)	63,476 (26.62)	16,556 (26.43)	0.4556
Heart failure	73,374 (24.37)	59,099 (24.79)	14,275 (22.79)	<0.0001
Prior stroke/TIA	62,294 (20.69)	50,854 (21.33)	11,440 (18.26)	<0.0001
Vascular diseases	23,298 (7.74)	18,063 (7.58)	5,235 (8.36)	<0.0001
COPD	71,143 (23.63)	57,859 (24.27)	13,284 (21.2)	<0.0001
Hyperlipidemia	95,291 (31.65)	71,903 (30.16)	23,388 (37.33)	<0.0001
Autoimmune diseases	15,259 (5.07)	11,749 (4.93)	3,510 (5.6)	<0.0001
Cancer	27,223 (9.04)	21,461 (9.0)	5,762 (9.2)	0.0022
Abnormal renal function	33,392 (11.09)	26,124 (10.96)	7,268 (11.6)	<0.0001
Abnormal liver function	54,952 (18.25)	42,730 (17.92)	12,222 (19.51)	<0.0001
Anemia	34,471 (11.45)	27,681 (11.61)	6,790 (10.84)	<0.0001
History of bleeding	65,916 (21.89)	52,227 (21.91)	13,689 (21.85)	0.7756
Alcohol excess/abuse	5,128 (1.7)	4285 (1.8)	843 (1.35)	<0.0001
Medications, n (%)
Rate control agents	136,291 (45.27)	107,457 (45.07)	28,834 (46.02)	0.5876
Beta blockers	85,737 (28.48)	65,165 (27.33)	20,572 (32.84)	<0.0001
CCBs	34,050 (11.31)	26,596 (11.16)	7,454 (11.9)	<0.0001
Digoxin	44,046 (14.63)	39,320 (16.49)	4,726 (7.54)	<0.0001
Rhythm control agents	62,413 (20.73)	0 (0)	62,413 (99.62)	<0.0001
Amiodarone	41,857 (13.9)	0 (0)	41,857 (66.81)	<0.0001
Dronedarone	2,126 (0.71)	0 (0)	2,126 (3.39)	<0.0001
Propafenone	18,695 (6.21)	0 (0)	18,695 (29.84)	<0.0001
Flecainide	523 (0.17)	0 (0)	523 (0.83)	<0.0001
Sotalol	866 (0.29)	0 (0)	866 (1.38)	<0.0001
Antiplatelet agents	113,059 (37.55)	83,371 (34.97)	29,688 (47.39)	<0.0001
Oral anticoagulants	47,181 (15.67)	36,017 (15.11)	11,164 (17.82)	<0.0001
Warfarin	36,250 (12.04)	27,885 (11.7)	8,365 (13.35)	<0.0001
NOACs	11,409 (3.79)	8,486 (3.56)	2,923 (4.67)	<0.0001
ACEIs/ARBs	90,345 (30.01)	67,570 (28.34)	22,775 (36.35)	<0.0001
Statins	29,849 (9.91)	21,390 (8.97)	8,459 (13.5)	<0.0001
Electrical cardioversion, n (%)	1,625 (0.54)	0 (0)	1625 (2.59)	<0.0001
Catheter ablation of AF, n (%)	532 (0.18)	0 (0)	532 (0.85)	<0.0001

Abbreviations: ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers; AF, atrial fibrillation; CCBs, calcium channel blockers; COPD, chronic obstructive pulmonary disease; NOACs, non-vitamin K antagonist oral anticoagulants; SD, standard deviation; TIA, transient ischemic attack.

Statistical Analysis

Data were presented as the mean value and standard deviation for continuous variables, and proportions for categorical variables. Differences between continuous values and nominal variables were assessed using the unpaired two-tailed t-test and chi-squared test, respectively. The incidences of clinical events were calculated from dividing the number of event by person-time at risk. The risks of clinical events were assessed using the Cox regression analysis. The cumulative incidence curves of events were plotted via the Kaplan–Meier method with the statistical significance examined by the log-rank test. All statistical significances were set at a p <0.05.

The present study was approved by the Institutional Review Board at Taipei Veterans General Hospital, Taipei, Taiwan.

Results

Demography and clinical characteristics of the whole study population are summarized in [Table 1]. Subjects receiving early rhythm control tended to be younger, with a higher mean CHA₂DS₂-VASc score but marginally lower mean HAS-BLED score. Distribution of comorbidities varied between the two groups, although prior stroke/transient ischemic attack (TIA) was less common in the early rhythm control group. Oral anticoagulant (OAC) use was higher in the early rhythm control arm. The clinical characteristics of the study population in different time periods are shown in [Supplementary Tables S1] to [S4] (available in the online version).

Clinical Outcomes and Time Trends

Kaplan–Meier curves demonstrated better outcomes with early rhythm control compared with usual care, for ischemic stroke, heart failure, mortality, and adverse outcomes (all p < 0.001; [Fig. 2]). [Fig. 3] shows the risks of clinical endpoints in early rhythm control and usual care groups. Risk of ischemic stroke was significantly lower with early rhythm control compared with usual care after adjustments for age, CHA₂DS₂-VASc score, HAS-BLED score, chronic obstructive pulmonary disease, hyperlipidemia, autoimmune diseases, cancer, anemia, use of antiplatelet agents, OACs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers, and statins (adjusted hazard ratio [aHR]: 0.771, 95% confidence interval [CI]: 0.751–0.792; p < 0.001; [Fig. 3]). When outcomes were compared for patients identified in different time periods (2001–2004, 2005–2008, 2009–2012, and 2013–2016), there was a significant statistical interaction, with a significant trend toward an even lower risk of ischemic stroke for patients receiving early rhythm control compared with usual care in more recent periods, e.g., in 2013–2016 (aHR: 0.682, 95% CI: 0.621–0.748).

Fig. 2 Cumulative incidence curves of clinical events of early rhythm control versus usual care. HF, heart failure.

Fig. 3 Early rhythm control and risks of clinical events of incident AF patients. *Adjustment for age, CHA₂DS₂-VASc score, HAS-BLED score, chronic obstructive pulmonary disease, hyperlipidemia, autoimmune diseases, cancer, anemia, use of antiplatelet agents, OACs, ACEIs/ARBs, and statins. ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers; AF, atrial fibrillation; aHR, adjusted hazard ratio; AMI, acute myocardial infarction; CI, confidence interval; OACs, oral anticoagulants.

Compared with usual care, early rhythm control was associated with a lower risk of heart failure (aHR: 0.851, 95% CI: 0.806–0.899; p < 0.001), AMI (aHR: 0.915, 95% CI: 0.877–0.955; p < 0.001), and mortality (aHR: 0.794, 95% CI: 0.782–0.806; p < 0.001) ([Fig. 3]). Risk of composite adverse events was also lower with early rhythm control (aHR: 0.823, 95% CI: 0.813–0.834; p < 0.001) without significant statistical interactions for patients identified from different timing periods ([Fig. 3]).

Clinical Outcomes and Timing of Early Rhythm Control

Compared with usual care, the lower risks of ischemic stroke (aHR: 0.746, 95% CI: 0.717–0.775), heart failure (aHR: 0.819, 95% CI: 0.798–0.841), mortality (aHR: 0.777, 95% CI: 0.759–0.795), and composite adverse events (aHR: 0.802, 95% CI: 0.787–0.818) associated with early rhythm control were even more evident when performed early (<3 months) compared with later periods (3–6 months, 7–9 months, and 10–12 months; p _int < 0.001) ([Fig. 4]).

Fig. 4 Timing of early rhythm control and risks of clinical events of incident AF patients. *Adjustment for age, CHA₂DS₂-VASc score, HAS-BLED score, chronic obstructive pulmonary disease, hyperlipidemia, autoimmune diseases, cancer, anemia, use of antiplatelet agents, OACs, ACEIs/ARBs, and statins. ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers; AF, atrial fibrillation; aHR, adjusted hazard ratio; AMI, acute myocardial infarction; CI, confidence interval; OACs, oral anticoagulants.

Subgroup Analyses

The comparisons between early rhythm control and usual care regarding risks of ischemic stroke, heart failure, mortality, and composite adverse events in different subgroups are shown in [Figs. 5] and [6]. Compared with usual care, the risk of ischemic stroke was consistently lower with early rhythm control when comparing age <75 or ≥75 years, males/females, presence/absence of prior heart failure, abnormal renal function or major bleeding, and use of rate control agents, with no significant statistical interaction (p _int = NS). The risk of ischemic stroke was more evident with early rhythm control for some subgroups, including CHA₂DS₂-VASc score ≥3, HAS-BLED score ≥3, prior stroke/TIA, hypertension, diabetes mellitus, and prior OAC use (p _int < 0.05) ([Fig. 5]). For heart failure, the lower risk with early rhythm control was more evident for age ≥75 years, males, CHA₂DS₂-VASc score ≥3, HAS-BLED score ≥3, no prior stroke/TIA, or heart failure, or abnormal renal function, use of rate control agents, and prior OAC use (p _int < 0.05), while other subgroups showed no statistical interactions ([Fig. 5]).

Fig. 5 Early rhythm control and risks of ischemic stroke and heart failure in different subgroups. *Adjustment for age, CHA₂DS₂-VASc score, HAS-BLED score, chronic obstructive pulmonary disease, hyperlipidemia, autoimmune diseases, cancer, anemia, use of antiplatelet agents, OACs, ACEIs/ARBs, and statins. ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers; AF, atrial fibrillation; aHR, adjusted hazard ratio; AMI, acute myocardial infarction; CI, confidence interval; OACs, oral anticoagulants; TIA, transient ischemic attack.

Fig. 6 Early rhythm control and risks of mortality and composite adverse events in different subgroups. *Adjustment for age, CHA₂DS₂-VASc score, HAS-BLED score, chronic obstructive pulmonary disease, hyperlipidemia, autoimmune diseases, cancer, anemia, use of antiplatelet agents, OACs, ACEIs/ARBs, and statins. ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin II-receptor blockers; AF, atrial fibrillation; aHR, adjusted hazard ratio; AMI, acute myocardial infarction; CI, confidence interval; OACs, oral anticoagulants; TIA, transient ischemic attack.

For mortality, improved outcomes associated with early rhythm control were more evident for age ≥75 years, males, CHA₂DS₂-VASc score ≥3, prior stroke/TIA, hypertension, use of rate control agents, and no diabetes, or abnormal renal function, or prior OAC use (p _int < 0.05) ([Fig. 6]). A lower risk of adverse events associated with early rhythm control was more evident for most of the subgroups (p _int < 0.05), apart from heart failure and diabetes mellitus (p _int = NS).

Falsification Analysis (Sensitivity Analysis)

The risks of three falsification endpoints did not differ significantly between “early rhythm control” or “usual care” groups (aHR: 1.11 [95% CI: 0.93–1.30, p = 0.238] for urinary tract infection, aHR: 1.15 [95% CI: 0.98–1.38, p = 0.278] for cellulitis, and aHR: 0.92 [95% CI: 0.70–1.17, p = 0.298] for acute appendicitis), suggesting the significant differences between “early rhythm control” and “usual care” with regard to clinical outcomes in which we were interested may be less likely due to treatment selection bias.

Propensity Matching

After propensity score matching, clinical characteristics were similar between the early rhythm control and usual care groups ([Supplementary Tables S5]–[S9] (available in the online version). The results of the analyses performed among the cohort after the propensity matching were similar to the principal findings observed among the study population without propensity matching ([Supplementary Figs. S1] and [S2] (available in the online version).

Discussion

In this nationwide cohort study, first, our principal finding was that an early rhythm control was associated with lower risks of ischemic stroke, heart failure, mortality, and composite adverse events compared with usual care. Second, there was a significant trend toward a lower risk of ischemic stroke associated with early rhythm control in more recent periods. Third, early rhythm control was associated with even lower risks of ischemic stroke, heart failure, mortality, and adverse events when performed early (<3 months), compared with later periods. The results were consistent for majority of subgroups studied.

In the EAST-AFNET 4 trial, 2,789 patients with early AF (diagnosed ≤12 months, median time since diagnosis: 36 days) were randomized to early rhythm control or usual care. After a median of 5.1 years of follow-up per patient, there was a 21% reduction in first primary outcome with early rhythm control (3.9 per 100 person-years) compared with usual care (5.0 per 100 person-years) (hazard ratio: 0.79; 96% CI: 0.66–0.94; p = 0.005).[9] In the present study, we found that early rhythm control was associated with lower risks for ischemic stroke, heart failure, AMI, mortality, and adverse events compared with usual care in the daily practice, especially when performed early (<3 months), consistent with that observed in the EAST trial.

Although AF burden may be significantly related to the risk of stroke in AF,[20] previous reports did not show superiority of rhythm control with AADs over rate control.[6] [7] Indeed, even asymptomatic AF patients are associated with a high risk of stroke and mortality.[21] [22] [23] Different from these previous studies, the EAST-AFNET 4 trial included AF ablation as a rhythm control therapy. However, there were only 8% and 19.4% of patients received catheter ablations at baseline and 2 years, respectively, in the early rhythm control arm, and around 7% of patients enrolled in the usual care group also received catheter ablations.[9] In our present study, most of the study subjects in the rhythm control arm received AADs (around 99%) rather than catheter ablations, and a lower risk of clinical events associated with early rhythm control was still observed. Therefore, the adoption of catheter ablation per se may not be able to fully explain the benefits of the early rhythm strategy observed in the EAST-AFNET 4 trial.

In our analysis, we showed that the risk of clinical events was even lower when rhythm control therapy was performed within 3 months after AF was diagnosed. This finding may support the concept of the EAST-AFNET 4 trial, that is, rhythm control therapy may be more beneficial when delivered early. However, it should be noted that the rhythm control arm had more structured follow-up with regular (two times per week) and symptom-driven patient-operated electrocardiogram recordings with transmission to the study team triggering an extra clinical visit as needed in the EAST trial. Such a structured follow-up with holistic integrated care has been associated with improved clinical outcomes in patients with AF compared with usual care.[24] [25] In an analysis of high-risk cohorts (e.g., multimorbidity, polypharmacy), structure care and management based on the ABC pathway was associated with improved outcomes, compared with non-ABC compliant management.[26] [27] Indeed, in an analysis from the ESC EORP-AF program, only one-third of patients were eligible for an early rhythm control strategy, and while the latter was associated with a lower rate of major adverse events, this difference was nonsignificant on multivariate analysis, being mediated by differences in baseline characteristics and clinical risk profile.[28] Also, early rhythm control was associated with a higher use of health-care resources and risk of hospital admission. Importantly, no difference in the primary outcome between early rhythm control and “no rhythm control patients” adherent to the ABC pathway was evident.[28]

Therefore, whether the better outcomes of patients assigned to rhythm control in the EAST-AFNET 4 trial were solely due to “early” intervention or because of more regular and structured follow-up was unclear. Our study was performed based on data acquired from the routine daily care with less structured management protocols, and our findings may provide complimentary “real-world” data to support the generalizability of the results of the EAST-AFNET 4 trial into daily clinical practice.

Limitations

This study is limited by its observational design with several important limitations. First, our population represents an Asian population, and the results may not be generalizable to non-Asians. Second, we did not have imaging data to assess the impacts of early rhythm control on cardiac remodeling. Third, different from the EAST-AFNET 4 trial, amiodarone was the most commonly prescribed AAD and catheter ablation was rarely performed within 1 year after incident AF was diagnosed in the Taiwan AF cohort. However, we reported the picture of clinical practice in Taiwan and the results may further support to extend the findings of the EAST-AFNET 4 trial to a very different AF population regarding the strategy for rhythm control. Fourth, OACs were underused in our study population, which was a common issue in Asian countries. Importantly, the subgroup analyses showed consistent findings favoring “early rhythm control” among patients with or without receiving OACs. Fifth, although we have tried to adjust baselines differences between two groups using multivariable Cox regression and propensity matching analyses, unmeasured confounders may still exist. Besides, some important information, such as income and education levels and types of AF, was lacking in our registry dataset. We were not able to exclude the possibility that patients with paroxysmal AF would be more likely to receive rhythm control therapies than nonparoxysmal AF patients, and therefore, a better clinical outcome was observed in the early rhythm control group. However, we have tried to perform falsification analyses and they did not show significant associations between three falsification endpoints and different treatment groups, which may imply that the results we reported may be less likely to be confounded by other factors whose information was not available in our study. Lastly, although we focused on patients who were diagnosed to have new-onset AF and defined “early rhythm control” as the prescriptions of AADs or performing of catheter ablations within 1 year after AF being diagnosed, the exact durations between the real onset of AF and the establishment of the AF diagnosis were not clear, which were usually difficult to be ascertained in usual clinical practice.

Conclusion

In this nationwide cohort study, early rhythm control therapy was associated with a lower risk of adverse events than usual care among patients with early AF. Outcomes were even better with earlier (<3 months) intervention.

Conflict of Interest

None declared.

^* Drs. Chao and Chan contribute equally to this work and are joint first authors.

^** Joint senior authors.

Note: The review process for this paper was fully handled by Christian Weber, Editor-in-Chief.

Author Contributions

Study concept and design: T.-F.C., G.Y.H.L., S.-A.C. Acquisition of data: T.-J.C., Y.-H.C. Analysis and interpretation of data: T.-F.C., C.-E.C., G.Y.H.L., S.-A.C. Drafting of the manuscript: T.-F.C., G.Y.H.L. Critical revision of the manuscript for important intellectual content: G.Y.H.L., S.-A.C. Statistical analysis: T.-F.C., T.-C.T., J.-N.L. Study supervision: G.Y.H.L., S.-A.C.

Note

This study is based on data from the Health and Welfare Data Science Center (HWDC), Ministry of Health and Welfare (MOHW), Taiwan. The interpretation and conclusions contained herein do not represent those of HWDC, MOHW, Taiwan.

• References

• 1 Chao TF, Liu CJ, Tuan TC. et al. Lifetime risks, projected numbers, and adverse outcomes in asian patients with atrial fibrillation: a report from the taiwan nationwide AF cohort study. Chest 2018; 153 (02) 453-466
• 2 Potpara TS, Lip GYH, Blomstrom-Lundqvist C. et al. The 4S-AF Scheme (Stroke Risk; Symptoms; Severity of Burden; Substrate): a novel approach to in-depth characterization (rather than classification) of atrial fibrillation. Thromb Haemost 2021; 121 (03) 270-278
• 3 Lip GYH. The ABC pathway: an integrated approach to improve AF management. Nat Rev Cardiol 2017; 14 (11) 627-628
• 4 Hindricks G, Potpara T, Dagres N. et al. ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association of Cardio-Thoracic Surgery (EACTS). Eur Heart J 2020
• 5 Chao TF, Joung B, Takahashi Y. et al. 2021 Focused update consensus guidelines of the Asia Pacific Heart Rhythm Society on Stroke Prevention in Atrial Fibrillation: executive summary. Thromb Haemost 2022; 122 (01) 20-47
• 6 Wyse DG, Waldo AL, DiMarco JP. et al; Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med 2002; 347 (23) 1825-1833
• 7 Van Gelder IC, Hagens VE, Bosker HA. et al; Rate Control versus Electrical Cardioversion for Persistent Atrial Fibrillation Study Group. A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation. N Engl J Med 2002; 347 (23) 1834-1840
• 8 Marrouche NF, Brachmann J, Andresen D. et al; CASTLE-AF Investigators. Catheter ablation for atrial fibrillation with heart failure. N Engl J Med 2018; 378 (05) 417-427
• 9 Kirchhof P, Camm AJ, Goette A. et al; EAST-AFNET 4 Trial Investigators. Early rhythm-control therapy in patients with atrial fibrillation. N Engl J Med 2020; 383 (14) 1305-1316
• 10 Chao TF, Chiang CE, Liao JN, Chen TJ, Lip GYH, Chen SA. Comparing the effectiveness and safety of nonvitamin K antagonist oral anticoagulants and warfarin in elderly Asian patients with atrial fibrillation: a nationwide cohort study. Chest 2020; 157 (05) 1266-1277
• 11 Chao TF, Lip GYH, Liu CJ. et al. Relationship of aging and incident comorbidities to stroke risk in patients with atrial fibrillation. J Am Coll Cardiol 2018; 71 (02) 122-132
• 12 Chao TF, Lip GYH, Lin YJ. et al. Age threshold for the use of non-vitamin K antagonist oral anticoagulants for stroke prevention in patients with atrial fibrillation: insights into the optimal assessment of age and incident comorbidities. Eur Heart J 2019; 40 (19) 1504-1514
• 13 Tsai CT, Liao JN, Chiang CE. et al. Association of ischemic stroke, major bleeding, and other adverse events with warfarin use vs non-vitamin K antagonist oral anticoagulant use in patients with atrial fibrillation with a history of intracranial hemorrhage. JAMA Netw Open 2020; 3 (06) e206424
• 14 Chang TY, Chan YH, Chiang CE. et al. Risks and outcomes of gastrointestinal malignancies in anticoagulated atrial fibrillation patients experiencing gastrointestinal bleeding: A nationwide cohort study. Heart Rhythm 2020; 17 (10) 1745-1751
• 15 Chang CH, Lee YC, Tsai CT. et al. Continuation of statin therapy and a decreased risk of atrial fibrillation/flutter in patients with and without chronic kidney disease. Atherosclerosis 2014; 232 (01) 224-230
• 16 Cheng CL, Kao YH, Lin SJ, Lee CH, Lai ML. Validation of the National Health Insurance Research Database with ischemic stroke cases in Taiwan. Pharmacoepidemiol Drug Saf 2011; 20 (03) 236-242
• 17 Hsieh CY, Chen CH, Li CY, Lai ML. Validating the diagnosis of acute ischemic stroke in a National Health Insurance claims database. J Formos Med Assoc 2015; 114 (03) 254-259
• 18 Lin YS, Chen TH, Chi CC. et al. Different implications of heart failure, ischemic stroke, and mortality between nonvalvular atrial fibrillation and atrial flutter-a view from a national cohort study. J Am Heart Assoc 2017; 6 (07) 6
• 19 Cheng CL, Lee CH, Chen PS, Li YH, Lin SJ, Yang YH. Validation of acute myocardial infarction cases in the national health insurance research database in taiwan. J Epidemiol 2014; 24 (06) 500-507
• 20 Kaplan RM, Koehler J, Ziegler PD, Sarkar S, Zweibel S, Passman RS. Stroke risk as a function of atrial fibrillation duration and CHA₂DS₂-VASc score. Circulation 2019; 140 (20) 1639-1646
• 21 Sun W, Freedman B, Martinez C, Wallenhorst C, Yan BP. Atrial fibrillation detected by single time-point handheld electrocardiogram screening and the risk of ischemic stroke. Thromb Haemost 2022; 122 (02) 286-294
• 22 Wallenhorst C, Martinez C, Freedman B. Risk of ischemic stroke in asymptomatic atrial fibrillation incidentally detected in primary care compared with other clinical presentations. Thromb Haemost 2022; 122 (02) 277-285
• 23 Wachter R, Freedman B. Subclinical atrial fibrillation and the risk of recurrent ischemic stroke. Thromb Haemost 2021; 121 (06) 697-699
• 24 Guo Y, Lane DA, Wang L. et al; mAF-App II Trial Investigators. Mobile Health technology to improve care for patients with atrial fibrillation. J Am Coll Cardiol 2020; 75 (13) 1523-1534
• 25 Romiti GF, Pastori D, Rivera-Caravaca JM. et al. Adherence to the ‘Atrial Fibrillation Better Care’ pathway in patients with atrial fibrillation: impact on clinical outcomes-a systematic review and meta-analysis of 285,000 patients. Thromb Haemost 2022; 122 (03) 406-414
• 26 Proietti M, Romiti GF, Olshansky B, Lane DA, Lip GYH. Comprehensive management with the ABC (Atrial Fibrillation Better Care) pathway in clinically complex patients with atrial fibrillation: a post hoc ancillary analysis from the AFFIRM trial. J Am Heart Assoc 2020; 9 (10) e014932
• 27 Yao Y, Guo Y, Lip GYH. mAF-App II Trial investigators. The effects of implementing a mobile health-technology supported pathway on atrial fibrillation-related adverse events among patients with multimorbidity: the mAFA-II randomized clinical trial. JAMA Netw Open 2021; 4 (12) e2140071
• 28 Proietti M, Vitolo M, Harrison SL. et al; ESC-EHRA EORP-AF Long-Term General Registry Investigators. Real-world applicability and impact of early rhythm control for European patients with atrial fibrillation: a report from the ESC-EHRA EORP-AF long-term general registry. Clin Res Cardiol 2022; 111 (01) 70-84

Publication History

Received: 21 March 2022

Accepted: 22 March 2022

Accepted Manuscript online:
23 March 2022

Article published online:
30 July 2022