Rhythm Strategy in Patients with Early Atrial Fibrillation: A Geriatric Cardiology Perspective

Geriatric Cardiology Take Home Points:

  1. Age-related structural, functional cardiac remodeling and higher multimorbidity burden increases the prevalence of atrial fibrillation (AF) in older adults. Balancing the benefits of AF treatment with treatment-related harm are substantial challenges in the care of older adults.
  2. The EAST-AFNET41 trial, conducted across 135 sites in 11 European countries between 2011 and 2016, was a well-done trial to confirm the hypothesis that in participants with early AF (AF duration ≤1-year) a rhythm control strategy was superior in reducing the incidence of the primary composite outcome. This was driven by a decrease in cardiovascular (CV) death and stroke without changes in the incidences of heart failure (HF) and acute coronary syndrome (ACS) hospitalizations or total hospital duration. Quality of life questionnaires added further insights for future hypothesis-generating studies from EAST-AFNET4 trial dataset. Of the trial participants, 29.1% were ≥75-years – an acceptable proportion of older adult inclusion.
  3. Pertinent considerations towards the care of older adults revolve around the 5M's (Mind, Mobility, Multimorbidity, Medications [polypharmacy] and Matters Most).

Mind: A diagnosis of mild cognitive impairment was noted in 43.7% - with a wide Montreal Cognitive Assessment (MoCA) range at baseline from 4.0-30.0 (severe cognitive impairment to normal – that should always be clinically correlated). No changes were noted between treatment strategies in the MoCA score but deteriorations were noted in the SF-12 mental score (see below). The possibility of differential change would be expected to develop over larger time periods. Of interest would be the natural history of cognitive impairment in those with atrial fibrillation.

Mobility: No significant differences were noted in the SF-12 physical score (See below re: ADL/IADL's).

Medications: While the baseline proportions of individual medications were shown, further (hypothesis-generating) analysis could be performed to address the association of polypharmacy to outcomes and statistical interactions between treatment strategy and polypharmacy (as a continuous variable).

Multimorbidity: Multimorbidity or multiple chronic conditions is the most common comorbidity. Further knowledge on multimorbidity would be pertinent in the care of older adults. Some scores that have been developed in an attempt to standardize multimorbidity burden include the Elixhauser comorbidity score, Charlson score, and others.

Matters Most: Older adults often need to balance benefit (reduction of major adverse cardiovascular events [MACE]) with harm (increase in treatment-related serious adverse events [SAE]). A discussion of the outcome desired as a function of the treatment burden willing to be endured would also be informative. Some statistical concepts as noted below can be used to further comprehend What Matters Most.

Time to Benefit: The cumulative incidence function curve noted in the original trial manuscript demonstrates a time to benefit of approximately 1 year.
Time to Harm: No data has been shown on time to harm.
Restricted Mean Survival Time (RMST): a translational statistic that provides information on the number of days gained because of reductions in the primary cardiovascular composite outcome and the number of days lost as a result of treatment-related harm could be beneficial for shared decision making treatment considerations.
Competing Risks: It is to be commended that the investigators did not present a Kaplan-Meier curve but rather presented a cumulative incidence function (which takes into account the competing risk of non-cardiovascular death). Addressing the competing risk of mortality in older adults will result in more accurate assessment of the incidence of outcome and usually without a change in the study conclusions.
Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL): An assessment of ADL and IADLs are the "functional" manifestations of the "anatomical" changes in quality-of-life scores and will add further knowledge of patient-centered care.

Commentary based on Kirchhof P, Camm AJ, Goette A, et al. Early rhythm-control therapy in patients with atrial fibrillation. N Engl J Med 2020;383:1305–16.1

Rationale for Study: The aim of the Early Treatment of Atrial Fibrillation for Stroke Prevention (EAST-AFNET4) was to assess the superiority of an early-rhythm control strategy in participants with AF duration <1-year to improve CV and non-CV outcomes.

Funding: Multiple grants from several European federal agencies and from device industry partnerships (St. Jude Medical–Abbott, Sanofi).

Study Methods
Design: International (multi-center), investigator-initiated, parallel-group, randomized, open, blinded outcome with event-driven outcomes. Randomization was 1:1, stratified by site and variable block lengths for assignment concealment.

Inclusion Criteria:  Participants with early AF (diagnosis ≤12 months of enrollment), >75 years old, prior transient ischemic attack (TIA) or stroke. Those who met two of the following criteria: >65 years, female sex, heart failure, hypertension, diabetes mellitus, severe coronary artery disease, chronic kidney disease (CKD) stage 3 or 4, left ventricular (LV) hypertrophy (diastolic septal wall width >15 mm).

Exclusion Criteria: Life expectancy <1 year, pregnancy, or childbearing and not on adequate birth control, breastfeeding women, drug abuse, prior AF percutaneous or surgical AF ablation, amiodarone failure, not suitable for AF rhythm control, severe mitral stenosis, prosthetic mitral valve, uncontrolled hepatic or thyroid dysfunction, CKD stage 5 (requiring hemodialysis [HD] or glomerular filtration rate [GFR] <10 ml/minute).

Exposure: Early rhythm control by antiarrhythmics, cardioversion, AF ablation (type based on local study teams). Participants in early rhythm control – asked to transmit single lead electrocardiogram (ECG) twice weekly and if symptomatic. ECG abnormality triggered algorithm to escalate rhythm-control as indicated. Usual care (initially treated without rhythm control therapy with rhythm control used only for symptoms).

Primary outcome(s): #1: Composite of CV death, stroke (ischemic/hemorrhagic), HF hospitalization or ACS hospitalization; #2: Number of nights spent in the hospital per year.

Secondary outcome(s): Individual components of primary outcomes, rhythm, LV function, QOL (EQ-5D, SF-12, EHRA score MoCA at 2 years)

Statistical Analysis:

  • Event-driven trial
  • Recruitment and follow-up time increased to 65 and 30 months respectively
  • Three unblinded, interim analysis at 25%, 50%, 75% of required events
  • O'Brien-Fleming stopping boundaries used
  • CV deaths were treated as censured
  • Multiple imputation used to replace missing values


2,789 participants underwent randomization (1,395 early rhythm and 1,394 usual care) with an average age of 70.3 ± 8.3 years. Pertinent participant baseline characteristics were as follows (46.4% women, 54.1% were in sinus rhythm, 39.6% had prior cardioversion, approximately 11% had a history of prior TIA or stroke with 43.7% noted to have mild cognitive impairment (MCI) and a mean MoCA score of 25.5 ± 3.8 with a range of 4.0-30.0.

#1 primary composite outcome occurred in 249 (3.9%) patients randomized to early rhythm-control therapy and in 316 (5.0%) patients in the usual care arm (hazard ratio [HR] 0.79; confidence interval [CI] 0.67-0.94; p = 0.005).

Patients randomized to early rhythm-control experienced numerically fewer deaths from CV causes (67 vs. 94; HR 0.72; 95%, CI 0.52-0.98), statistically significant fewer strokes (40 vs. 62; p = 0.03), but not HF or ACS hospitalization and more complications of rhythm-control therapy (68 vs. 19; p <0.0001). The clinical benefit of an early rhythm-control strategy was consistent across subgroups, including asymptomatic patients and those without HF.

#2 primary outcome of hospitalization was not statistically different (HR 1.08, 0.92-1.28).

Secondary outcomes

  • No significant changes in LV ejection fraction, EQ-5D score, SF-12 physical score.
  • Statistically significant decrease in SF-12 mental score (-1.20, -2.04 to -0.37).
  • A higher proportion of participants-maintained sinus rhythm in the early rhythm group with no significant change in the proportion of asymptomatic participants [1.14, 0.93 to 1.40].


  1. Kirchhof P, Camm AJ, Goette A, et al. Early rhythm-control therapy in patients with atrial fibrillation. N Engl J Med 2020;383:1305–16.

Clinical Topics: Acute Coronary Syndromes, Arrhythmias and Clinical EP, Cardiovascular Care Team, Heart Failure and Cardiomyopathies, Prevention, Valvular Heart Disease, Atherosclerotic Disease (CAD/PAD), EP Basic Science, Atrial Fibrillation/Supraventricular Arrhythmias, Acute Heart Failure, Hypertension, Geriatric Cardiology

Keywords: ESC Congress, ESC21, Quality of Life, Activities of Daily Living, Ischemic Attack, Transient, Atrial Fibrillation, Electric Countershock, Amiodarone, Coronary Artery Disease, Prevalence, Multiple Chronic Conditions, Glomerular Filtration Rate, Multimorbidity, Acute Coronary Syndrome, Life Expectancy, Survival Rate, Polypharmacy, Confidence Intervals, Mitral Valve Stenosis, Follow-Up Studies, Decision Making, Shared, Breast Feeding, Mitral Valve, Random Allocation, Stroke Volume, Ventricular Remodeling, Heart Failure, Stroke, Electrocardiography, Mental Status and Dementia Tests, Hospitalization, Renal Insufficiency, Chronic, Renal Dialysis, Diabetes Mellitus, Hospitals, Patient-Centered Care, Algorithms, Contraception, Hypertension, Hypertrophy, Cardiology, Substance-Related Disorders, Cognition

< Back to Listings