Cardio · arrhythmias
AF rate control: beta-blocker vs digoxin choices
Fellowship-level guide to drug choice for ventricular rate control in atrial fibrillation (AF) under the 2024 ESC AF guideline, the 2023 ACC/AHA/ACCP/HRS AF guideline and the 2026 ESC heart failure guideline: beta-blockers, verapamil and diltiazem, digoxin and digitoxin by left ventricular ejection fraction (LVEF) and comorbidity, acute versus long-term rate control, doses, heart rate targets, atrioventricular node ablation and pacing, pregnancy, and the 2018 NHFA/CSANZ summary.
On this page
Study tools
Your progress
Saved on this device.
Practise this topic
Target exams
- EECC
- ABIM Cardiovascular Disease Certification
Red flags
- AF with acute or worsening haemodynamic instability: ESC 2024 recommends electrical cardioversion to improve immediate patient outcomes (Class I, Level C)
- AF with rapid ventricular response and known moderate or severe LV systolic dysfunction, with or without decompensated HF: ACC/AHA 2023 says intravenous nondihydropyridine calcium channel blockers should not be administered (COR 3: Harm, LOE B-NR)
- Pre-excited AF (anterograde accessory pathway conduction): ACC/AHA 2023 says verapamil, diltiazem, amiodarone, digoxin, adenosine and beta blockers are contraindicated because of the risk of precipitating VF or haemodynamic deterioration (COR 3: Harm, LOE B-NR)
- Permanent AF with risk factors for cardiovascular events: ACC/AHA 2023 says dronedarone should not be used for long-term rate control (COR 3: Harm, LOE B-R)
This topic is about choosing and dosing the rate-control drug, its target and the step to atrioventricular node ablation. Anticoagulation, rhythm control and the rest of the ESC 2024 AF-CARE pathway ([C] comorbidity and risk factor management, [A] avoid stroke and thromboembolism, [R] reduce symptoms by rate and rhythm control, [E] evaluation and dynamic reassessment) are taught in the atrial fibrillation topic.[1]
What rate control is for
ESC 2024 calls limiting tachycardia an integral part of AF management and says it is often sufficient to improve AF-related symptoms.[1] ACC/AHA 2023 states that the overall goals of rate control, in acute and chronic AF with a rapid ventricular response, centre on control of symptoms and the risk of developing LV systolic dysfunction.[3]
For most patients, rate control is not an either-or choice.[1] ESC 2024 observes that most patients need a combination approach of rate and rhythm control, which should be consciously re-evaluated during follow-up.[1] ESC 2024 says its approach to heart rate control (Figure 7 of the guideline) can be used for all types of AF, including paroxysmal, persistent and permanent AF.[1]
ESC 2024 defines permanent AF as a shared decision made between the patient and physician that no further attempts at restoration of sinus rhythm are planned.[1] ESC 2024 text also states that limited evidence exists to inform the best type and intensity of rate control treatment.[1]
Classification: the drug groups and the ejection fraction line
Start with the left ventricle. In the acute setting, ESC 2024 says the choice of drug will depend on the patient's characteristics, presence of heart failure and LVEF, and haemodynamic profile.[1] For long-term control, ESC 2024 says the choice depends on symptoms, comorbidities, and the potential for side effects and interactions.[1] ESC 2024 Table 12 lists beta-blockers, the non-dihydropyridine calcium channel antagonists verapamil and diltiazem, the cardiac glycosides digoxin and digitoxin, and amiodarone.[1]
ESC 2024 AF
Recommendation Table 14
- LVEF above 40%: beta-blockers, diltiazem, verapamil or digoxin as first-choice drugs, to control heart rate and reduce symptoms (Class I, Level B)
- LVEF 40% or less: beta-blockers and/or digoxin, to control heart rate and reduce symptoms (Class I, Level B)
- Table 12: verapamil and diltiazem contraindicated if LVEF is 40% or less
ACC/AHA 2023 AF
Sections 7.2.1, 7.2.2 and 9.2
- Haemodynamically stable rapid ventricular response: beta blockers or nondihydropyridine calcium channel blockers (verapamil, diltiazem; provided that EF is above 40%) for acute rate control (COR 1, LOE B-R)
- LVEF below 40%: nondihydropyridine calcium channel–blocking drugs should not be administered, given their potential to exacerbate HF (COR 3: Harm, LOE C-LD)
- AF and HF: digoxin is reasonable for rate control, in combination with other rate-controlling agents or as monotherapy if other agents are not tolerated (COR 2a, LOE B-R)
ESC 2026 HF
Recommendation Table 15
- HFrEF: LVEF below 50% with symptoms and/or signs of HF
- Stable HFrEF and AF: beta-blockers first-line for short- and long-term rate control (Class I, Level C)
- Stable HFrEF and AF: digoxin should be considered when the rate stays high despite beta-blockers, or beta-blockers are contraindicated or not tolerated, to obtain short- and long-term rate control (Class IIa, Level C)
NHFA/CSANZ 2018
abstract
- β-blockers or non-dihydropyridine calcium channel antagonists remain the first line choice for acute and chronic rate control
The two ESC documents draw their line in different places.[1][4] ESC 2024 AF uses an LVEF cut-off of 40%, while ESC 2026 HF lists HFrEF as expanded to include LVEF up to 50%.[1][4] ESC 2024 AF already advised that patients with HFmrEF (LVEF 41%–49%) and AF should generally be treated according to guidance for HFrEF, albeit with limited evidence to date in AF.[1]
Timing is the other axis. ESC 2024 has separate sections on heart rate control in the acute setting and long-term heart rate control, and ACC/AHA 2023 has separate recommendation tables for acute and long-term rate control.[1][3]
[1] [3] [4]Who needs rate control: epidemiology
- ESC 2026 HF: heart failure and AF frequently coexist, with AF in more than half of patients with HF and HF in more than one-third of patients with AF.[4]
- ESC 2026 HF: the coexistence of HF and AF is associated with a high risk of death, HF hospitalisation and stroke regardless of LVEF.[4]
- ESC 2024 AF: when patients and physicians make a joint decision for rate control, AF is classified as permanent, which the guideline calls the most common ‘type’ of AF in historical registries.[1]
- ACC/AHA 2023: AF is common in patients with COPD, with a prevalence of 11% in a large European registry; the combination of AF and asthma is less common.[3]
- ESC 2024 AF: atrial fibrillation is one of the most common arrhythmias during pregnancy, and AF during pregnancy is associated with an increased risk of death.[1]
- RATE-AF, which compared low-dose digoxin with bisoprolol in 160 patients aged 60 years or older with permanent AF and dyspnoea of NYHA class II or higher, enrolled patients with a mean age of 76 years, 46% women and a mean baseline heart rate of 100/min.[8]
Pathophysiology and pharmacology: how each drug slows the ventricle
ACC/AHA 2023 explains that beta blockers and nondihydropyridine calcium channel blockers both slow conduction through the atrioventricular node.[3] The groups differ in speed of onset and in their effect on contractility, and the two cardiac glycosides differ in how they leave the body.[1][3][4]
| Drug group | Mechanism, as each source states it | Consequence at the bedside |
|---|---|---|
| Beta-blockers | ACC/AHA 2023: slow conduction through the atrioventricular node by blocking beta-1 receptors | ESC 2024: in general, for acute rate control, beta-blockers (for all LVEF) are preferred over digoxin because of their more rapid onset of action and dose-dependent effects |
| Non-dihydropyridine calcium channel blockers (verapamil, diltiazem) | ACC/AHA 2023: slow conduction through the atrioventricular node and have negative inotropic and chronotropic effects | ACC/AHA 2023: diltiazem and verapamil are negative inotropic agents and may not be tolerated in patients with HFrEF |
| Cardiac glycosides (digoxin, digitoxin) | ESC 2024: inhibit the sodium–potassium adenosine triphosphatase and augment parasympathetic tone; ACC/AHA 2023 describes digoxin as having positive inotropic and vagotonic effects | ESC 2024: in general, for acute rate control, beta-blockers (for all LVEF) and diltiazem/verapamil (for LVEF above 40%) are preferred over digoxin because of their more rapid onset of action and dose-dependent effects |
| Intravenous magnesium (adjunct to standard rate-control measures) | ACC/AHA 2023: the mechanism likely stems from blockade of slow inward calcium channels in the sinoatrial and atrioventricular node, slowing the heart rate and prolonging atrioventricular conduction velocity | ACC/AHA 2023: a low adverse effect profile and minimal toxicity make it a favourable option, often as an adjunct to atrioventricular nodal blockers; in AF with rapid ventricular response, adding it to standard rate-control measures is reasonable to achieve and maintain rate control (COR 2a, LOE A) |
Digoxin and digitoxin share a mechanism but not a route out of the body.[1][4] ESC 2026 HF states that both agents share a narrow therapeutic window and require careful monitoring, but digoxin is primarily eliminated through the kidneys whereas digitoxin undergoes hepatic elimination.[4]
[3] [1] [4]Why the ventricle suffers when the rate stays high
ACC/AHA 2023 calls AF the most common cause of arrhythmia-induced cardiomyopathy, in which persistent AF, with or without rapid ventricular rates, can lead to LV dysfunction and HF that can be partially or completely reversed with adequate arrhythmia control.[3] ACC/AHA 2023 separates AF as the only cause of cardiomyopathy (AF-induced) from AF that exacerbates LV dysfunction and HF in concomitant structural heart disease (AF-mediated).[3]
ESC 2026 HF adds that HF induced by AF appears to have a more favourable outcome than HF causing AF.[4]
Why a beta-blocker in AF is not the same drug as in sinus rhythm
Do not assume that the mortality benefit of beta-blockers in heart failure carries over to AF.[21][1] In an individual-patient meta-analysis of ten placebo-controlled heart failure trials (rhythm classified at baseline; mean follow-up 1.5 years), beta-blockers significantly reduced all-cause mortality in sinus rhythm (HR 0.73) but not in AF (HR 0.97), with a significant interaction by baseline rhythm (p=0.002).[21]
ESC 2024 AF puts it more cautiously: the prognostic benefit of beta-blockers seen in HFrEF patients with sinus rhythm may not be present in patients with AF.[1] ESC 2024 AF still points out that these drugs have clear proof of safety and there may be other indications for them beyond prognosis, including comorbidity management and symptom improvement.[1]
Clinical presentation: the patient in front of you
Think of three settings: fast AF in the emergency department, permanent AF with persisting symptoms, and AF with a falling ejection fraction. ACC/AHA 2023 notes that for patients needing intravenous rate-control agents, hypotension and/or decompensated HF may limit use of otherwise efficacious agents.[3]
- Unstable patients (ESC 2024) include those with haemodynamic instability caused by the arrhythmia or acute cardiac conditions, and severely ill patients who develop AF (sepsis, trauma, surgery, and particularly cancer-related surgery).[1]
- Arrhythmia-induced cardiomyopathy (ACC/AHA 2023): in an observational study of 24 patients with AF-induced cardiomyopathy and NYHA class III/IV HF, the median time from onset of arrhythmia to cardiomyopathy and HF was 4.2 years; aggressive rate/rhythm control produced significant LVEF recovery in all within 6 months, and the five with recurrent AF had a rapid decline in LVEF within 6 months of recurrence.[3]
- Pregnancy (ESC 2024): rapid atrioventricular conduction of AF may have serious haemodynamic consequences for mother and foetus.[1]
- Pre-excitation (ESC 2024): patients with Wolff–Parkinson–White syndrome and AF are at risk of fast ventricular rates from rapid conduction over the accessory pathway, potentially leading to ventricular fibrillation and sudden death.[1]
Differential diagnosis: what else is driving the rate
Look for the driver before reaching for a drug. ESC 2024 says physicians should always evaluate and manage underlying causes for the initiation of AF before, or in parallel to, instituting acute rate and/or rhythm control, such as treating sepsis, addressing fluid overload or managing cardiogenic shock.[1]
| Situation | What the source says | Why it changes the drug choice |
|---|---|---|
| Pre-excited AF | ACC/AHA 2023: in AF with anterograde accessory pathway conduction, drugs that block atrioventricular nodal conduction (verapamil, diltiazem, amiodarone, digoxin, adenosine or beta blockers) are contraindicated (COR 3: Harm, LOE B-NR) | ACC/AHA 2023 gives the reason as the risk of precipitating VF or haemodynamic deterioration; ESC 2024 Table 12 states that all rate control drugs, and intravenous amiodarone, are contraindicated in Wolff–Parkinson–White syndrome |
| Atrial flutter | ESC 2024: rate control can be difficult to achieve in AFL, despite combination therapy | Combination therapy may still fail to control the rate |
| Sepsis, fluid overload, cardiogenic shock | ESC 2024 names treating sepsis, addressing fluid overload and managing cardiogenic shock among the underlying causes to manage | The cause is treated before, or in parallel to, acute rate control |
| Heart failure with congestion | ESC 2026 HF: diuretic therapy is indicated to relieve congestion and may increase the chances of restoring sinus rhythm by reducing filling pressures and ventricular rate | ESC 2024 AF: achieving euvolaemia with diuretics can also facilitate better control of heart rate in AF |
| Thyroid and other precipitants | ESC 2026 HF says precipitating factors for AF, for example hyperthyroidism, toxic agents including alcohol, valvular heart diseases, infection, diabetes, uncontrolled hypertension, obesity, sleep disorders and volume overload, should be routinely identified and appropriately managed | ESC 2024 AF: patients presenting with new-onset or recurrent AF should be tested for thyroid-stimulating hormone (TSH) levels |
Bedside assessment and investigations
In the acute setting, ESC 2024 says the choice of rate-control drug will depend on the patient's characteristics, presence of heart failure and LVEF, and haemodynamic profile.[1] At the bedside that means asking about the patient, whether there is heart failure, what the LVEF is, and whether the patient is haemodynamically stable.[1]
- Echocardiography: ESC 2024 recommends a transthoracic echocardiogram in patients with an AF diagnosis where this will guide treatment decisions (Class I, Level C).[1]
- ESC 2024 Table 8 (all patients) includes a 12-lead ECG and blood tests (full blood count, kidney function, serum electrolytes, liver function, glucose/HbA1c and thyroid function).[1]
- ESC 2024 Table 8 (selected patients): ambulatory ECG monitoring for assessing AF burden and ventricular rate control, and exercise ECG to evaluate rate control.[1]
- Before digoxin: ESC 2024 Table 12 says to check renal function before starting digoxin and to adapt the dose in CKD patients.[1]
- Combination therapy: ESC 2024 says combining beta-blockers with verapamil or diltiazem should only be performed in secondary care with regular monitoring of heart rate by 24 h ECG to check for bradycardia.[1]
Serum digoxin concentrations
ESC 2026 HF reminds readers that digoxin and digitoxin share a narrow therapeutic window and require careful monitoring.[4] ESC 2024 says serum digoxin concentrations can be monitored to avoid toxicity, especially in patients at higher risk due to older age, renal dysfunction or use of interacting medications.[1]
- ACC/AHA 2023 text: in the small series that defined digoxin toxicity by electrocardiographic abnormalities, toxicity was seen when serum concentrations exceeded 2.0 ng/mL and almost certainly above 3.0 ng/mL.[3]
- ACC/AHA 2023 text: 3 post hoc analyses of the DIG trial suggested that safe use of digoxin is seen at lower serum concentrations.[3]
- The DIG report set its question in chronic heart failure with normal sinus rhythm.[14]
- ESC 2026 HF notes that DIGIT-HF, a trial of low-dose digitoxin in chronic symptomatic HFrEF, also included patients with AF, contrary to the DIG trial.[4]
- ACC/AHA 2023 text, citing a separate reference: serum digoxin concentrations of 0.5 to 0.9 ng/mL were associated with significantly lower all-cause mortality and hospitalisations than concentrations of 1.0 ng/mL or more.[3]
- ACC/AHA 2023 text: another analysis suggested that concentrations above 1.2 ng/mL may be harmful, particularly in women.[3]
- ACC/AHA 2023 text: a post hoc analysis of patients with AF taking digoxin against propensity score-matched controls suggested no increased risk of death below 0.9 ng/mL, a nonsignificant increase at 0.9 to 1.1 ng/mL, and a significant (56%) increased risk of death at 1.2 ng/mL or more.[3]
- ACC/AHA 2023 text: 0.5 to 0.8 ng/mL seems safest in terms of benefit without adverse effects in patients with HFrEF.[3]
- ACC/AHA 2023 Table 21 notes for digoxin: renally eliminated, and increased mortality at plasma concentrations exceeding 1.2 ng/mL.[3]
Management: the unstable patient
Instability changes the question from which drug to whether a drug is the right tool at all. ACC/AHA 2023 stresses that, in the acute setting, recognising patients needing emergency cardioversion is important.[3]
- ESC 2024 recommends electrical cardioversion in AF patients with acute or worsening haemodynamic instability to improve immediate patient outcomes (Class I, Level C).[1]
- ESC 2024 text: emergency electrical cardioversion is still considered the first-choice treatment if sinus rhythm is thought to be beneficial, despite a high rate of immediate relapse; amiodarone is a second-line option because of its delayed activity but may be an appropriate alternative in the acute setting.[1]
- ESC 2026 HF recommends urgent cardioversion in decompensated HF (DHF) with rapid ventricular rates and haemodynamic instability, to restore sinus rhythm (Class I, Level C).[4]
- ESC 2024: intravenous amiodarone, digoxin or landiolol may be considered in patients with AF who have haemodynamic instability or severely depressed LVEF to achieve acute control of heart rate (Class IIb, Level B); the 2025 correction removed esmolol from this row.[1][2]
- ESC 2026 HF: intravenous amiodarone or digoxin may be considered in haemodynamically unstable patients with HFrEF and AF to stabilise the patient and achieve acute control of heart rate (Class IIb, Level C).[4]
- ESC 2026 HF text: in haemodynamically unstable HFrEF, intravenous amiodarone or digoxin may be used to achieve acute rate control and improve haemodynamics, although acute electrical cardioversion to restore sinus rhythm should be first choice.[4]
- ESC 2024 text: the ultra-short acting and highly selective beta-blocker landiolol can safely control rapid AF in patients with low ejection fraction and acutely decompensated heart failure, with limited impact on myocardial contractility or blood pressure.[1]
Management: acute rate control in the stable patient
In the stable patient the choice is about speed and the ventricle. ESC 2024 says that, in general, beta-blockers (for all LVEF) and diltiazem or verapamil (for LVEF above 40%) are preferred over digoxin for acute rate control because of their more rapid onset of action and dose-dependent effects.[1] ESC 2024 adds that more selective beta-1 receptor blockers have a better efficacy and safety profile than unselective beta-blockers.[1]
| ACC/AHA 2023 acute rate control row | COR, LOE |
|---|---|
| In AF with rapid ventricular response who are haemodynamically stable, beta blockers or nondihydropyridine calcium channel blockers (verapamil, diltiazem; provided that EF is above 40%) are recommended for acute rate control | 1, B-R |
| In AF with rapid ventricular response in whom beta blockers and nondihydropyridine calcium channel blockers are ineffective or contraindicated, digoxin can be considered for acute rate control, either alone or in combination with those agents | 2a, B-R |
| In AF with rapid ventricular response, adding intravenous magnesium to standard rate-control measures is reasonable to achieve and maintain rate control | 2a, A |
| In AF with rapid ventricular response who are critically ill and/or in decompensated HF, in whom beta blockers and nondihydropyridine calcium channel blockers are ineffective or contraindicated, intravenous amiodarone may be considered for acute rate control (footnote: consider the risk of cardioversion and stroke) | 2b, B-NR |
| In AF with rapid ventricular response and known moderate or severe LV systolic dysfunction, with or without decompensated HF, intravenous nondihydropyridine calcium channel blockers should not be administered | 3: Harm, B-NR |
ESC 2024 says combination therapy with digoxin may be required in acute settings, and that combining beta-blockers with diltiazem or verapamil should be avoided except in closely monitored situations.[1]
[3]What the acute trials show
- Siu (Crit Care Med 2009): 150 adults with acute symptomatic AF necessitating hospitalisation and a ventricular rate above 120 bpm were randomised to intravenous diltiazem, digoxin or amiodarone; the trial title calls the population acute uncomplicated AF, and ACC/AHA 2023 describes it as AF without major comorbidities.[18][3]
- In that trial, ventricular rate control at 24 hours was achieved in 90% with diltiazem, 74% with digoxin and 74% with amiodarone.[18]
- In the same trial the median time to rate control was 3 hours with diltiazem, against 6 hours with digoxin and 7 hours with amiodarone; the primary end point was sustained ventricular rate below 90 bpm within 24 hours.[18]
- Perrett meta-analysis (Clin Res Cardiol 2024) of 12 randomised trials in acute AF and atrial flutter (1152 participants, high heterogeneity): across all intravenous beta-blockers there was no difference from comparators in heart rate reduction or in reaching target heart rate.[19]
- In that meta-analysis of acute AF and atrial flutter, conventional selective beta-1 blockers were inferior for target heart rate reduction versus control, super-selective beta-1 blockers were superior, and hypotension and bradycardia were more frequent with non-selective beta-blockers.[19]
- ACC/AHA 2023 text: other agents may be safer and more effective than digoxin for acute rate control, and intravenous diltiazem was more effective than intravenous digoxin in 2 small RCTs.[3]
- ACC/AHA 2023 text: in an RCT of 30 patients with AF and a rapid ventricular response (the cited trial’s title covers atrial fibrillation and flutter), randomised to intravenous diltiazem, digoxin or both, intravenous diltiazem significantly decreased heart rate within 5 minutes against 3 hours with intravenous digoxin.[3]
- ACC/AHA 2023 text: in 52 patients with rapid AF, an intravenous combination of diltiazem and digoxin gave a more rapid and durable response than intravenous diltiazem alone.[3]
- ACC/AHA 2023 text: in the DAAF (Digitalis in Acute Atrial Fibrillation) trial of 239 patients (12% with HF), intravenous digoxin led to a significant decrease in heart rate at 2 hours compared with placebo.[3]
- ACC/AHA 2023 text: a meta-analysis of 6 RCTs (n=745) of intravenous magnesium in rapid AF, given in combination with standard rate-control methods, found it superior to standard rate-control methods for achieving rate control (63% versus 40%; OR 2.49).[3]
- ACC/AHA 2023 text: in one retrospective analysis of patients hospitalised with AF and a rapid ventricular rate, diltiazem was associated with more acute kidney injury within 48 hours in those with LVEF of 50% or less than in those with normal EF (10% versus 3.6%); in another, patients with HFrEF given intravenous diltiazem had more worsening HF symptoms (increased oxygen requirement within 4 hours or initiation of inotropic support within 48 hours) than those given intravenous metoprolol (33% versus 15%); neither analysis noted an increase in in-hospital mortality.[3]
Management: long-term rate control
ESC 2024 says the choice of rate control drugs depends on symptoms, comorbidities, and the potential for side effects and interactions.[1] ESC 2024 lists beta-blockers, diltiazem, verapamil, digoxin or combination therapy as the ways to achieve pharmacological rate control.[1]
| ESC 2024 Recommendation Table 14 row | Class, level |
|---|---|
| Beta-blockers, diltiazem, verapamil or digoxin are recommended as first-choice drugs in patients with AF and LVEF above 40% to control heart rate and reduce symptoms | I, B |
| Beta-blockers and/or digoxin are recommended in patients with AF and LVEF of 40% or less to control heart rate and reduce symptoms | I, B |
| Combination rate control therapy should be considered if a single drug does not control symptoms or heart rate in patients with AF, providing that bradycardia can be avoided, to control heart rate and reduce symptoms | IIa, C |
| ACC/AHA 2023 long-term rate control row | COR, LOE |
|---|---|
| In AF, beta blockers or nondihydropyridine calcium channel blockers (diltiazem, verapamil) are recommended for long-term rate control, with the choice of agent according to underlying substrate and comorbid conditions | 1, B-NR |
| For patients with AF in whom measuring serum digoxin levels is indicated, it is reasonable to target levels below 1.2 ng/mL | 2a, B-NR |
| In AF and HF symptoms, digoxin is reasonable for long-term rate control in combination with other rate-controlling agents, or as monotherapy if other agents are not preferred, not tolerated or contraindicated | 2a, B-R |
| In AF and LVEF below 40%, nondihydropyridine calcium channel–blocking drugs should not be administered given their potential to exacerbate HF | 3: Harm, C-LD |
| In permanent AF with risk factors for cardiovascular events, dronedarone should not be used for long-term rate control | 3: Harm, B-R |
ACC/AHA 2023 summarises that both nondihydropyridine calcium channel blockers and beta blockers are effective for long-term rate control, useful in the absence of preexcitation.[3] ACC/AHA 2023 adds that digoxin may be useful with limited tolerability to other agents, or as adjunct therapy when the ventricular rate is difficult to control.[3]
Beta-blocker or digoxin? The drug-by-drug case
Beta-blockers
- ESC 2024: beta-1 selective adrenoreceptor antagonists are often first-line, largely based on their acute effect on heart rate and benefit in chronic HFrEF
- ESC 2024: the prognostic benefit seen in HFrEF with sinus rhythm may not be present in AF
- ESC 2024 Table 12: in asthma, non-selective beta-blockers should be avoided; contraindicated in acute HF and history of severe bronchospasm
Verapamil and diltiazem
- ESC 2024: a different adverse effect profile makes them useful for those experiencing side effects from beta-blockers
- ESC 2024: in a 60-patient crossover RCT (permanent AF with normal left ventricular function in the cited trial) they did not lead to the same reduction in exercise capacity as beta-blockers, and had a beneficial impact on BNP
- ESC 2024 Table 12: contraindicated if LVEF is 40% or less; adapt doses in hepatic and renal impairment
Digoxin and digitoxin
- ESC 2024: in RCTs there is no association between digoxin and any increase in all-cause mortality; lower doses may be associated with better prognosis
- ESC 2024 Table 12: high plasma levels associated with adverse events; check renal function before starting digoxin and adapt the dose in CKD
- ESC 2026 HF: digoxin is primarily renally eliminated, digitoxin hepatically
Amiodarone
- ESC 2024: reserved as a last option when rate cannot be controlled even with maximal tolerated combination therapy, or for patients who do not qualify for atrioventricular node ablation and pacing
- ESC 2024: many adverse effects relate directly to cumulative dose, restricting its long-term value for rate control
Dronedarone is the drug not to reach for.[1][3] ESC 2024 says dronedarone should not be instituted for rate control because it increases rates of heart failure, stroke and cardiovascular death in permanent AF.[1] ESC 2024 adds that some antiarrhythmic drugs with rate-limiting properties, for example amiodarone and sotalol, should generally be used only for rhythm control.[1]
Verapamil and diltiazem against beta-blockers
Two crossover reports, each in 60 patients with permanent AF, compared four once-daily regimens head to head.[16][17] The drugs were diltiazem 360 mg/day, verapamil 240 mg/day, metoprolol 100 mg/day and carvedilol 25 mg/day, each for 3 weeks in randomised sequence.[16]
- Rate: mean 24-hour heart rate was 96 beats/min untreated, 75 with diltiazem, 81 with verapamil, 82 with metoprolol and 84 with carvedilol; diltiazem gave a lower rate than any other drug.[16]
- Symptoms: arrhythmia-related symptoms were reduced by diltiazem and verapamil but not by the β blockers.[16]
- Exercise (patients with normal left ventricular function): diltiazem or verapamil preserved exercise capacity and reduced NT-proBNP, whereas metoprolol or carvedilol reduced exercise capacity and increased NT-proBNP.[17]
Combinations
- ESC 2024 text: combination therapy of different rate-controlling drugs should be considered only when needed to achieve the target heart rate, with careful follow-up to avoid bradycardia.[1]
- ESC 2024 text: combining beta-blockers with verapamil or diltiazem should only be performed in secondary care with regular monitoring of heart rate by 24 h ECG to check for bradycardia.[1]
- ACC/AHA 2023 text: digoxin may have added efficacy as rate control in conjunction with beta blockers in patients with AF.[3]
- In a randomised, double-blind, placebo-controlled trial of 47 patients with persistent AF and HF (mean LVEF 24%), adding carvedilol to digoxin for four months lowered the 24-hour ventricular rate and improved LVEF and symptom score compared with digoxin alone; in the second phase (six months), there was no significant difference between digoxin alone and carvedilol alone in any variable.[20]
Doses: ESC 2024 Table 12 against ACC/AHA 2023 Table 21
The two tables do not always agree, so quote the body when you quote a dose.[1][3] The rows below are copied from each table; a dash means the drug has no row in that table.
| Agent | ESC 2024 Table 12: intravenous / usual oral maintenance | ACC/AHA 2023 Table 21: intravenous / oral maintenance |
|---|---|---|
| Metoprolol tartrate | 2.5–5 mg bolus over 2 min, up to 15 mg maximal cumulative dose / 25–100 mg twice daily | 2.5–5 mg bolus over 2 min, up to 3 doses / 25–200 mg twice daily |
| Metoprolol succinate (XL) | N/A / 50–200 mg once daily | N/A / 50–400 mg daily or twice daily in divided doses |
| Bisoprolol | N/A / 1.25–20 mg once daily | N/A / 2.5–10 mg daily |
| Atenolol | N/A / 25–100 mg once daily | N/A / 25–100 mg daily; renally eliminated |
| Carvedilol | N/A / 3.125–50 mg twice daily | N/A / 3.125–25 mg twice daily |
| Nebivolol | N/A / 2.5–10 mg once daily | — |
| Esmolol | 500 µg/kg IV bolus over 1 min, then 50–300 µg/kg/min / N/A | 500 μg/kg bolus over 1 min, then 50–300 μg/kg/min / N/A |
| Landiolol | Optional loading dose of 100 µg/kg IV over 1 min, then 10–40 µg/kg/min; in critically ill patients (cardiac dysfunction, septic shock) start at 1–10 µg/kg/min and titrate according to response / N/A | — |
| Propranolol | — (footnote b: other beta-blockers, e.g. propranolol, are not recommended as specific rate control therapy in AF) | 1 mg over 1 min, repeated as needed every 2 min, up to 3 doses / 10–40 mg 3–4 times daily |
| Verapamil | 2.5–10 mg IV bolus over 5 min / 40 mg twice daily to 480 mg (extended release) once daily | 5–10 mg over at least 2 min (may repeat twice), then 5 mg/h continuous infusion (max 20 mg/h) / 180–480 mg daily (ER); avoid in HFrEF |
| Diltiazem | 0.25 mg/kg IV bolus over 5 min, then 5–15 mg/h / 60 mg three times daily to 360 mg (extended release) once daily | 0.25 mg/kg (actual body weight) IV over 2 min, may repeat 0.35 mg/kg over 2 min, then 5–15 mg/h continuous infusion / 120–360 mg daily (ER); avoid in HFrEF |
| Digoxin | 0.5 mg IV bolus (0.75–1.5 mg over 24 h in divided doses) / 0.0625–0.25 mg once daily | 0.25–0.5 mg over several min, repeat 0.25 mg every 6 h (maximum 1.5 mg/24 h) / 0.0625–0.25 mg daily |
| Digitoxin | 0.4–0.6 mg / 0.05–0.1 mg once daily | — |
| Amiodarone | 300 mg IV in 250 mL 5% dextrose over 30–60 min (preferably via central venous cannula), then 900–1200 mg IV over 24 h in 500–1000 mL via a central venous cannula / 200 mg once daily after loading (loading 200 mg three times daily for 4 weeks, then 200 mg daily or less as appropriate, reducing other rate control drugs according to heart rate) | 150–300 mg IV over 1 h, then 10–50 mg/h over 24 h / 100–200 mg daily (generally IV form used for rate control); loading dose 6–10 g over 2–4 wk |
- ESC 2024 Table 12 uses N/A for not available or not widely available, and its maximum doses are based on the summary of product characteristics of each drug; ACC/AHA 2023 Table 21 uses N/A for not applicable.[1][3]
- ESC 2024 Table 12 footnote: there are no data on atenolol, and it should not be used in heart failure with reduced ejection fraction or in pregnancy.[1]
- ESC 2024 Table 12 footnote: the amiodarone loading regimen may vary, and intravenous dosage should be considered when calculating the total load.[1]
- ESC 2024 Table 12 Contraindicated column, amiodarone: contraindicated in iodine sensitivity; serious potential adverse effects (including pulmonary, ophthalmic, hepatic and thyroid); consider numerous drug interactions.[1]
- RATE-AF doses: digoxin 62.5–250 μg/d (mean 161 μg/d) and bisoprolol 1.25–15 mg/d (mean 3.2 mg/d).[8]
Heart rate targets: lenient first
ESC 2024 says the optimal heart rate target depends on the setting, symptom burden, presence of heart failure, and whether rate control is combined with a rhythm control strategy.[1] ESC 2024 cites RACE II for its lenient target row.[1]
- ESC 2024: lenient rate control with a resting heart rate below 110 b.p.m. should be considered as the initial target for patients with AF, with stricter control reserved for those with continuing AF-related symptoms (Class IIa, Level B).[1]
- ESC 2024 text: lenient rate control is an acceptable initial approach, unless there are ongoing symptoms or suspicion of tachycardia-induced cardiomyopathy, where stricter targets may be indicated.[1]
- ACC/AHA 2023: in AF without HF, in candidates for select rate-control strategies, the heart rate target should be guided by underlying symptoms, in general aiming at a resting heart rate below 100 to 110 bpm (COR 2a, LOE B-R).[3]
- ACC/AHA 2023: in suspected AF-induced cardiomyopathy or refractory HF symptoms on pharmacological rate control, a stricter strategy (below 80 bpm at rest and below 110 bpm during moderate exercise) may be reasonable (COR 2b, LOE B-NR).[3]
- ESC 2026 HF text: lenient rate control, with resting heart rate below 110 b.p.m., should be considered as the initial target in patients with and without HF, with re-evaluation based on symptoms.[4]
- ACC/AHA 2023 text: other populations that may benefit from a low heart rate goal include those with rate-related cardiac dysfunction, ICDs, cardiac resynchronisation therapy and the tachycardia-bradycardia form of sick sinus syndrome.[3]
Randomised: lenient rate control (resting heart rate below 110 beats per minute) vs strict rate control (resting heart rate below 80 beats per minute and below 110 beats per minute during moderate exercise); follow-up at least 2 years, maximum 3 years
Population: 614 patients with permanent atrial fibrillation
Key finding
Primary composite (cardiovascular death, hospitalisation for heart failure, stroke, systemic embolism, bleeding and life-threatening arrhythmic events) at 3 years, estimated cumulative incidence: 12.9% lenient vs 14.9% strict; absolute difference −2.0 percentage points (90% CI −7.6 to 3.5; P below 0.001 for the noninferiority margin). Target met by 97.7% vs 67.0%, with fewer visits in the lenient group
Practice change
Authors: in permanent AF, lenient rate control is as effective as strict rate control and is easier to achieve
The limits of RACE II matter at the viva. ACC/AHA 2023 notes that the groups differed by only 10 bpm, because 32.6% of patients assigned to strict control did not reach target and 78% of lenient-control participants had heart rates below 100 bpm.[3]
ACC/AHA 2023 adds that patients with HF were underrepresented, so whether the results extend to HF is unknown.[3] ESC 2026 HF notes that RACE II and a post-hoc combined AFFIRM and RACE analysis were not focused on HF and only a minority of their patients had HF, and calls the data on strict versus lenient rate control in HF and AF inconclusive.[4]
[9] [1] [3] [4]When drugs fail: atrioventricular node ablation and pacing
ESC 2024 says ablation of the atrioventricular node with pacemaker implantation (‘ablate and pace’) can lower and regularise heart rate in patients with AF, and that the procedure has a low complication rate and a low long-term mortality risk.[1] ACC/AHA 2023 states the trade-off plainly: AVNA provides ventricular rate control effectively and without medications, yet creates dependence on pacing.[3]
| Guideline row | Class or COR, level |
|---|---|
| ESC 2024: atrioventricular node ablation in combination with pacemaker implantation should be considered in patients unresponsive to, or ineligible for, intensive rate and rhythm control therapy, to control heart rate and reduce symptoms | IIa, B |
| ESC 2024: atrioventricular node ablation combined with cardiac resynchronisation therapy should be considered in severely symptomatic patients with permanent AF and at least one hospitalisation for HF, to reduce symptoms, physical limitations, recurrent HF hospitalisation and mortality | IIa, B |
| ESC 2026 HF: atrioventricular node ablation combined with CRT should be considered in patients with severely symptomatic permanent AF and poor rate control despite medical therapy and at least one HF hospitalisation, to reduce symptoms, physical limitations, recurrent HF hospitalisation and death | IIa, B1 |
| ACC/AHA 2023: in AF with uncontrolled rapid ventricular response refractory to rate-control medications (not candidates for, or failed, rhythm control), AVNA can be useful to improve symptoms and QOL | 2a, B-R |
| ACC/AHA 2023: in AF, HFrEF (LVEF below 50%) and refractory rapid ventricular response, in patients who are not candidates for or in whom rhythm control has failed, AVNA and biventricular pacing therapy can be useful to improve symptoms, QOL and EF | 2a, B-R |
| ACC/AHA 2023: in AF, HF and implanted biventricular pacing therapy in whom an effective pacing percentage cannot be achieved with pharmacological therapy, AVNA can be beneficial to improve functional class, reduce the risk of ICD shock and improve survival | 2a, B-NR |
| ACC/AHA 2023: in AF planned for AVNA, implanting a pacemaker before the ablation (before or on the same day) is recommended to ensure adequacy of the pacing leads before ablation | 1, B-NR |
| ACC/AHA 2023: in AF with a persistently rapid ventricular response who undergo AVNA, initial pacemaker lower rate programming should be 80 to 90 bpm to reduce the risk of sudden death | 1, C-LD |
| ACC/AHA 2023: in AF with normal EF undergoing AVNA, conduction system pacing of the His bundle or left bundle area may be reasonable | 2b, C-LD |
| ACC/AHA 2023: in AF and HFrEF undergoing AVNA, conduction system pacing of the His bundle or left bundle branch area may be reasonable as an alternative to biventricular pacing, to improve symptoms, QOL and LV function | 2b, C-LD |
The ESC 2026 HF row differs from the ESC 2024 AF row in two ways.[4][1] The ESC 2026 HF row adds poor rate control despite medical therapy to the entry conditions, and it carries level B1, where the ESC 2024 row carried level B.[4][1] Under the ESC 2024 revision of evidence grading, level B is subdivided into B1 and B2.[7] B1 represents suggestive evidence usually from at least one adequately powered RCT free from major bias, or a meta-analysis of such RCTs, with some evidence against the play of chance (e.g. P below .05 for superiority).[7]
[1] [3] [4]How to do it safely
- Timing, ESC 2024 text: the pacemaker should be implanted a few weeks before the atrioventricular node ablation, with the initial pacing rate after ablation set at 70–90 b.p.m.[1]
- Timing, ACC/AHA 2023: pacemaker before or on the same day as the ablation, to ensure adequacy of the pacing leads before ablation (COR 1, LOE B-NR); with a persistently rapid ventricular response, initial lower rate programming of 80 to 90 bpm, to reduce the risk of sudden death (COR 1, LOE C-LD).[3]
- Why the high early rate, ACC/AHA 2023 text: early observational studies showed sudden death after AVNA in 3% to 7% of patients, from ventricular fibrillation predominantly due to bradycardia, QT prolongation and heterogeneity of repolarisation.[3]
- ACC/AHA 2023 text: current protocols with higher lower-rate pacing in the early postprocedure period have minimised these deaths, and the lower rate is then adjusted over several weeks.[3]
- Leads, ACC/AHA 2023 text: the overall risk of lead dislodgement or failure is approximately 2%, and many operators perform AVNA and pacemaker implant during the same procedure.[3]
- Pacing mode, ESC 2024 text: the choice between right ventricular and biventricular pacing depends on patient characteristics, presence of heart failure and LVEF.[1]
- ACC/AHA 2023 text: HF after AVNA is attributed to the deleterious effects of right ventricular pacing, with risk correlated with baseline cardiac function; right ventricular pacing is advised in patients with preserved EF, because the benefit of biventricular pacing is less and its complication risk higher.[3]
- Conduction system pacing, ESC 2024 text: it may become a potentially useful alternate pacing mode for a pace and ablate strategy once safety and efficacy have been confirmed in larger RCTs.[1]
Who should not be offered it first
- ESC 2024 text: the evidence base has typically included older patients; for younger patients, ablate and pace should only be considered if heart rate remains uncontrolled despite consideration of other pharmacological and non-pharmacological treatment options.[1]
- ACC/AHA 2023 text: long-term data on outcomes after AVNA are limited, and no evidence supports AVNA as first-line therapy.[3]
- ACC/AHA 2023 text: early and late complication rates are not inconsequential, specifically in young patients in view of the risk of pacemaker-mediated cardiomyopathy, and long-term follow-up data are scant.[3]
- ACC/AHA 2023 text: the consequences of lifelong pacemaker implantation, particularly with respect to age and comorbidities, are central to deciding on benefit.[3]
Outcome evidence for ablate and pace
International, open-label, blinded-outcome randomised trial: atrioventricular junction ablation plus biventricular pacing (Ablation + CRT) vs pharmacological rate control; primary endpoint all-cause mortality
Population: 133 patients with severely symptomatic permanent AF for more than 6 months, narrow QRS (110 ms or less) and at least one HF hospitalisation in the previous year; mean age 73 years, 47% female
Key finding
Stopped for efficacy at interim analysis after a median 29 months: all-cause death 11% vs 29% (HR 0.26, 95% CI 0.10–0.65; P = 0.004); estimated death rates 5% vs 21% at 2 years and 14% vs 41% at 4 years; benefit similar with EF 35% or less and above 35%
Practice change
The 2018 report under the same registration (NCT02137187; 102 patients) found a lower rate of the primary composite of HF death, HF hospitalisation or worsening HF with Ablation + CRT at a median 16 months (20% vs 38%; HR 0.38) and a 36% decrease in the specific symptoms and physical limitations of AF at 1 year
Two meta-analyses frame the older evidence.[23][24] A meta-analysis (Circ Arrhythm Electrophysiol 2012) in refractory AF found all-cause mortality similar between AVNA and medical therapy (3.1% versus 3.3%), with low procedure-related mortality (0.27%) and, at a mean follow-up of 26.5 months, a 2.1% incidence of sudden cardiac death after AVNA.[23] A second meta-analysis (Eur J Heart Fail 2012) of four RCTs in patients undergoing AVNA for refractory AF found that biventricular pacing after AVNA did not significantly improve survival compared with right ventricular-only pacing, with a modest but significant structural and functional benefit.[24]
Specific scenarios
Heart failure with reduced ejection fraction
In HFrEF the two ESC documents rank the drugs differently.[4][1] ESC 2024 AF puts beta-blockers and digoxin side by side for LVEF of 40% or less, while ESC 2026 HF makes beta-blockers first-line in stable HFrEF and reserves digoxin for when the rate remains high despite beta-blockers, or beta-blockers are contraindicated or not tolerated.[1][4]
| ESC 2026 HF Recommendation Table 15: rate control rows | Class, level |
|---|---|
| Beta-blockers are recommended in stable patients with HFrEF and AF as first-line therapy for short- and long-term rate control | I, C |
| Digoxin should be considered in stable patients with HFrEF and AF when the ventricular rate remains high despite beta-blockers, or when beta-blockers are contraindicated or not tolerated, to obtain short- and long-term rate control | IIa, C |
| Atrioventricular node ablation combined with CRT should be considered in severely symptomatic permanent AF with poor rate control despite medical therapy and at least one HF hospitalisation, to reduce symptoms, physical limitations, recurrent HF hospitalisation and death | IIa, B1 |
| Intravenous amiodarone or digoxin may be considered in haemodynamically unstable patients with HFrEF and AF to stabilise the patient and achieve acute control of heart rate | IIb, C |
- ESC 2026 HF text: beta-blockers have proven safety and efficacy in AF with HFrEF and are recommended as the first-line approach; if rate control is suboptimal with beta-blockers, or they are not tolerated, digoxin or digitoxin should be considered.[4]
- ESC 2026 HF text: the effects of digoxin in HFrEF with AF have not been studied in RCTs, but digoxin may be useful for AF with rapid ventricular rate when other therapies cannot be used.[4]
- ACC/AHA 2023: in AF and HF, digoxin is reasonable for rate control, in combination with other rate-controlling agents or as monotherapy if other agents are not tolerated (COR 2a, LOE B-R).[3]
- ACC/AHA 2023: in AF and known LVEF below 40%, nondihydropyridine calcium channel-blocking drugs should not be administered because of their potential to exacerbate HF (COR 3: Harm, LOE B-R in the HF table; C-LD in the long-term rate control table).[3]
- ACC/AHA 2023: in AF and HF with rapid ventricular rates in whom beta blockers or calcium channel blockers are contraindicated or ineffective, intravenous amiodarone is reasonable for acute rate control (COR 2a, LOE B-NR), with the same cardioversion and stroke footnote.[3]
- ACC/AHA 2023 text, from MDPIT: among post-infarction patients with a baseline EF below 40%, late HF appeared in more patients on diltiazem than placebo (21% versus 12%), and the excess grew with more severe reductions in baseline EF.[3]
Two glycoside trials sit behind the ESC 2026 HF glycoside row: the newer DIGIT-HF trial of digitoxin (2025) and the older DIG trial of digoxin (1997).[4][13][14] ESC 2026 HF states that a cardiac glycoside (digoxin or digitoxin) should be considered in symptomatic HFrEF with LVEF of 40% or less, despite optimal foundational medical therapy, to reduce the risk of HF hospitalisation (Class IIa, Level B1).[4] The ESC 2026 HF glycoside row aims to reduce the risk of HF hospitalisation; it is not a rate-control row.[4]
International, double-blind, placebo-controlled trial: digitoxin (starting 0.07 mg once daily) vs placebo on top of guideline-directed medical therapy
Population: Chronic HF with LVEF of 40% or less and NYHA class III or IV, or LVEF of 30% or less and NYHA class II; 1212 in the modified intention-to-treat population; ESC 2026 HF notes that, unlike the DIG trial, it included patients with AF
Key finding
Median 36 months: death from any cause or first hospital admission for worsening HF 39.5% vs 44.1% (HR 0.82; 95% CI 0.69–0.98; P = 0.03); all-cause death alone 27.2% vs 29.5% (HR 0.86; 95% CI 0.69–1.07); at least one serious adverse event 4.7% vs 2.8%
Practice change
ESC 2024 AF had listed the DIGIT-HF registration (EudraCT 2013-005326-38) among two ongoing RCTs of digoxin and digitoxin in HFrEF with and without AF
The older DIG trial randomised patients with heart failure and an LVEF of 0.45 or less, in its main trial, to digoxin or placebo on top of diuretics and ACE inhibitors.[14][13] DIG did not reduce overall mortality but reduced hospitalisation both overall and for worsening heart failure; it addressed heart failure with normal sinus rhythm, and ESC 2026 HF notes that DIGIT-HF, unlike DIG, also included patients with AF.[14][4]
Heart failure with preserved ejection fraction
- ESC 2026 HF: there is no evidence on pharmacological strategies for rate control in patients with AF and HFpEF.[4]
- ESC 2024 AF rows are set by ejection fraction: with LVEF above 40%, beta-blockers, diltiazem, verapamil or digoxin are recommended as first-choice drugs to control heart rate and reduce symptoms (Class I, Level B).[1]
- ACC/AHA 2023: in AF with HF symptoms, digoxin is reasonable for long-term rate control, in combination or as monotherapy if other agents are not preferred, not tolerated or contraindicated (COR 2a, LOE B-R).[3]
Asthma and COPD
- ESC 2024 Table 12: in case of asthma, non-selective beta-blockers should be avoided, and beta-blockers are contraindicated with a history of severe bronchospasm.[1]
- ACC/AHA 2023: in AF and COPD, it is reasonable to use cardioselective beta blockers for rate control, especially where other indications exist (eg, MI and HF) (COR 2a, LOE B-R).[3]
- ACC/AHA 2023 text: because beta-2 agonists can be used to treat COPD, clinicians have been reluctant to prescribe beta blockers for fear of antagonising beta-2 agonists or precipitating bronchospasm, but a meta-analysis of RCTs of cardioselective beta blockers found no change in respiratory function, symptoms or response to beta-2 agonists.[3]
- ACC/AHA 2023 text: AF with asthma is less common, and beta blockers are generally avoided to prevent bronchospasm.[3]
After cardiac surgery
- ACC/AHA 2023: in postoperative cardiac surgery patients, beta blockers are recommended to achieve rate control for AF unless contraindicated or ineffective, in which case a nondihydropyridine calcium channel blocker is recommended (COR 1; LOE A for beta blockers, B-R for calcium channel blockers).[3]
- ACC/AHA 2023: in haemodynamically stable cardiac surgery patients with postoperative AF, rate-control (target heart rate below 100 bpm) and/or rhythm-control medications are recommended as initial therapy, chosen by symptoms, haemodynamic consequences and physician preference (COR 1, LOE B-R).[3]
Critical illness and atrial flutter
- ACC/AHA 2023 text: intravenous amiodarone has been shown to be effective in controlling ventricular rates in critically ill patients; in a retrospective study of 38 intensive care patients it was associated with a significant decrease in heart rate without a decrease in BP, compared with intravenous diltiazem and digoxin.[3]
- ESC 2024 Table 12 landiolol row: in critically ill patients (cardiac dysfunction, septic shock), start at 1–10 µg/kg/min and titrate according to response.[1]
- ESC 2024 text: rate control can be difficult to achieve in atrial flutter, despite combination therapy.[1]
Patients with CRT devices
- ESC 2024 text: in CRT recipients, the presence or occurrence of AF is one of the main reasons for suboptimal biventricular pacing.[1]
- ESC 2024 text: improving biventricular pacing is indicated and can be reached by intensifying rate control drugs, atrioventricular node ablation, or rhythm control, depending on patient and AF characteristics.[1]
- ACC/AHA 2023 text: remote monitoring data show a strong association between higher biventricular pacing percentages and reduced mortality, greatest above 98% pacing.[3]
- ACC/AHA 2023 text: strict rate control is also advisable in patients with ICDs and CRT.[3]
Complications and pitfalls
- Giving intravenous verapamil or diltiazem to a patient with moderate or severe LV systolic dysfunction: ACC/AHA 2023 classes this as harm (COR 3: Harm, LOE B-NR), and ESC 2024 Table 12 lists both as contraindicated if LVEF is 40% or less.[3][1]
- Bradycardia from combinations: ESC 2024 says combination therapy should be considered only when needed to achieve the target heart rate, with careful follow-up to avoid bradycardia, and that a beta-blocker with verapamil or diltiazem should only be combined in secondary care with regular monitoring of heart rate by 24 h ECG.[1]
- Digoxin toxicity: ESC 2024 Table 12 links high plasma levels with adverse events; ACC/AHA 2023 Table 21 notes increased mortality at plasma concentrations exceeding 1.2 ng/mL.[1][3]
- Forgetting the kidneys: ESC 2024 says check renal function before starting digoxin and adapt the dose in CKD; ESC 2026 HF notes that digoxin is primarily renally eliminated.[1][4]
- Using dronedarone for rate control: PALLAS (patients at least 65 years old with permanent AF for at least 6 months and risk factors for major vascular events) was stopped for safety after 3236 patients, and dronedarone increased heart failure, stroke and cardiovascular death.[15]
- Amiodarone drift: ESC 2024 reserves it as a last option when heart rate cannot be controlled even with maximal tolerated combination therapy, or for patients who do not qualify for atrioventricular node ablation and pacing, and ACC/AHA 2023 warns about cardioversion and stroke when it is used for rate control.[1][3]
- Atrioventricular nodal blockers in pre-excited AF: contraindicated in ACC/AHA 2023 (COR 3: Harm, LOE B-NR) because of the risk of precipitating VF or haemodynamic deterioration.[3]
- Atenolol in HFrEF or pregnancy: ESC 2024 Table 12 says it should not be used in either.[1]
- Ablate and pace with a slow early pacing rate: ACC/AHA 2023 links early sudden death after AVNA to VF, predominantly due to bradycardia, QT prolongation and heterogeneity of repolarisation.[3]
Prognosis and follow-up
- ESC 2024 AF: prognosis may be affected by LVEF, with the rate of death highest with AF and HFrEF (LVEF of 40% or less), compared with AF and HFpEF (LVEF of 50% or more).[1]
- ESC 2026 HF: in a large observational registry, heart rate in AF was significantly associated with long-term death only in patients with HFrEF and heart rate above 100 b.p.m.[4]
- ESC 2026 HF: a large observational study found lenient and poor rate control associated with a higher risk of death than strict rate control in AF with HF, regardless of LVEF.[4]
- RACE II: the estimated cumulative incidence of the primary composite outcome at 3 years was 12.9% with lenient and 14.9% with strict control.[9]
- Digoxin and death: a systematic review and meta-analysis of observational and controlled trial data, covering studies published from 1960 to July 2014 of digoxin against control (placebo or no treatment), reviewed 52 studies (621,845 patients) and pooled 75 study analyses; the pooled risk ratio for death was 1.76 in unadjusted analyses, 1.61 in adjusted analyses and 1.18 in propensity-matched studies, but 0.99 (0.93 to 1.05) in randomised controlled trials, and ACC/AHA 2023 notes that the 7 RCTs (~8400 patients) in that subanalysis were in patients with concomitant HF.[12][3]
- The same meta-analysis found that on meta-regression, baseline differences between treatment groups had a significant impact on the mortality associated with digoxin, studies with better methods and lower risk of bias were more likely to report a neutral association, and prescription biases limit the value of observational data.[12]
- Ablate and pace: in a Mayo Clinic series of 350 patients, survival after atrioventricular node ablation was similar to that of 229 drug-treated controls with AF, and, in the absence of underlying heart disease, similar to expected survival in the general population.[22]
- Follow-up: ESC 2024 expects combination rate and rhythm approaches to be consciously re-evaluated during follow-up, and ESC 2026 HF advises re-evaluating the lenient target based on symptoms.[1][4]
Special populations
Pregnancy
For pregnancy, the newest ESC rows sit in the 2025 pregnancy guideline, alongside the 2024 AF rows.[5][1] The ESC 2025 pregnancy and ESC 2024 AF recommendation rows agree on the order: a beta-1-selective blocker (not atenolol) first, then digoxin when beta-blockers are ineffective or not tolerated.[5][1] The ESC 2025 row also lists verapamil when beta-blockers fail or are not tolerated, and ESC 2024 text says that if beta-blockers fail, digoxin and verapamil can be considered, with verapamil avoided in the first trimester.[5][1]
| Body and row | Class or COR, level |
|---|---|
| ESC 2024 AF: beta-1 selective blockers are recommended for heart rate control of AF in pregnancy to reduce symptoms and improve maternal and foetal outcomes, excluding atenolol | I, C |
| ESC 2024 AF: digoxin should be considered for heart rate control of AF in pregnancy if beta-blockers are ineffective or not tolerated, to reduce symptoms and improve maternal and foetal outcomes | IIa, C |
| ESC 2025 pregnancy: intravenous beta-blockers (e.g. metoprolol; except atenolol) are recommended as the first-line option for acute rate control in pregnant women with AF or AF with preserved LVEF and rapid ventricular rate | I, C |
| ESC 2025 pregnancy: intravenous digoxin or verapamil (if preserved LVEF) should be considered as a second-line option for initial rate control in pregnant women with AF or AFL and rapid ventricular rate | IIa, C |
| ESC 2025 pregnancy: beta-1-selective blockers (except atenolol) are recommended for rate control in pregnant women with AF, AFL or FAT | I, C |
| ESC 2025 pregnancy: digoxin or verapamil should be considered for rate control in pregnant women with AF, AFL or FAT when beta-blockers fail or are not tolerated | IIa, C |
| ACC/AHA 2023: in pregnant individuals with persistent AF, rate-control agents with a record of safety in pregnancy, such as beta blockers (eg, propranolol or metoprolol) and digoxin, either alone or with beta blockers, are reasonable as first-line agents | 2a, B-NR |
| ESC 2024 AF: immediate electrical cardioversion is recommended in AF during pregnancy with haemodynamic instability or pre-excited AF, to improve maternal and foetal outcomes | I, C |
- ESC 2024 AF text: intravenous selective beta-1 receptor blockers are recommended as first choice for acute heart rate control, not including atenolol, which can lead to intrauterine growth retardation; if beta-blockers fail, digoxin and verapamil can be considered (verapamil should be avoided in the first trimester).[1]
- ESC 2025 pregnancy text: when rate control is needed in (long-standing) persistent or permanent AF, beta-blockers, verapamil or digoxin should be used, also in combination, taking into account the concomitant conditions affecting the mother.[5]
- ESC 2025 pregnancy abbreviations: AFL is atrial flutter and FAT is focal atrial tachycardia.[5]
Older patients, kidney disease and congenital heart disease
- Older patients: ESC 2024 says serum digoxin concentrations can be monitored to avoid toxicity, especially in patients at higher risk due to older age, renal dysfunction or use of interacting medications.[1]
- Older patients: ESC 2024 notes that the evidence base for ablate and pace has typically included older patients.[1]
- Kidney disease: ESC 2024 Table 12 advises adapting verapamil and diltiazem doses in hepatic and renal impairment, and adapting the digoxin dose in CKD.[1]
- Kidney disease: ACC/AHA 2023 Table 21 marks atenolol and digoxin as renally eliminated; ESC 2026 HF notes digitoxin undergoes hepatic elimination.[3][4]
- Congenital heart disease: ESC 2024 says rate control drugs such as selective beta-1 receptor blockers, verapamil, diltiazem and digoxin can be used with caution, with monitoring for bradycardia and hypotension.[1]
Evidence, guidelines and regional differences
Randomised, open-label, blinded end-point trial: low-dose digoxin (62.5–250 μg/d; mean 161 μg/d) vs bisoprolol (1.25–15 mg/d; mean 3.2 mg/d); primary end point SF-36 physical component summary at 6 months
Population: 160 patients aged 60 years or older with permanent AF (no plan to restore sinus rhythm) and NYHA class II or higher dyspnoea; mean age 76 years, 46% women, mean baseline heart rate 100/min
Key finding
No significant difference in SF-36 physical component summary at 6 months (adjusted mean difference 1.4; 95% CI −1.1 to 3.8; P = .28) or in resting heart rate (76.9 vs 74.8/min); 2-class modified EHRA improvement in 53% vs 9%; at 12 months 8 of 20 outcomes were significantly different, all favouring digoxin, including NT-proBNP; at least one adverse event in 25% vs 64%
Practice change
Authors: among patients with permanent AF and symptoms of heart failure, no statistically significant difference in quality of life at 6 months; the findings support potentially basing treatment decisions on other end points
ESC 2024 summarises RATE-AF, a trial in patients with symptomatic permanent AF, as showing no difference between low-dose digoxin and bisoprolol in patient-reported quality of life at 6 months.[1][8] ESC 2024 adds that those randomised to digoxin had fewer adverse effects, a greater improvement in mEHRA and NYHA scores, and a reduction in BNP.[1] ESC 2024 cites RATE-AF in its Class I row for LVEF above 40%.[1]
| Question | ESC 2024 AF | ACC/AHA 2023 AF | ESC 2026 HF |
|---|---|---|---|
| First-line drug, LVEF above 40% | LVEF above 40%: beta-blockers, diltiazem, verapamil or digoxin, to control heart rate and reduce symptoms (I, B); text: HFmrEF (LVEF 41%–49%) with AF should generally be treated according to guidance for HFrEF, albeit with limited evidence to date in AF | Long-term rate control in AF: beta blockers or nondihydropyridine calcium channel blockers (diltiazem, verapamil), chosen by underlying substrate and comorbid conditions (1, B-NR) | HFpEF (LVEF 50% or more): no evidence on pharmacological strategies for rate control (text); LVEF below 50% is HFrEF in this guideline |
| First-line drug, reduced ejection fraction (ESC 2024 AF: LVEF 40% or less; ESC 2026 HF: HFrEF, LVEF below 50%) | LVEF 40% or less: beta-blockers and/or digoxin, to control heart rate and reduce symptoms (I, B) | Long-term row: beta blockers or nondihydropyridine calcium channel blockers by substrate and comorbidity (1, B-NR), but nondihydropyridine calcium channel blockers should not be administered with LVEF below 40% (3: Harm); AF with HF: digoxin is reasonable for rate control, in combination with other rate-controlling agents or as monotherapy if other agents are not tolerated (2a, B-R) | Stable HFrEF: beta-blockers first-line (I, C); digoxin when the rate remains high despite beta-blockers, or beta-blockers are contraindicated or not tolerated (IIa, C) |
| Unstable or severely depressed LVEF | Haemodynamic instability or severely depressed LVEF: intravenous amiodarone, digoxin or landiolol may be considered, to achieve acute control of heart rate (IIb, B; esmolol removed by the 2025 correction) | Critically ill and/or decompensated HF with rapid ventricular response, beta blockers and nondihydropyridine calcium channel blockers ineffective or contraindicated: intravenous amiodarone may be considered for acute rate control (2b, B-NR); AF and HF with rapid ventricular rates, beta blockers or calcium channel blockers contraindicated or ineffective: intravenous amiodarone is reasonable for acute rate control (2a, B-NR); footnote to both rows: consider the risk of cardioversion and stroke when using amiodarone as a rate-control agent | Intravenous amiodarone or digoxin in haemodynamically unstable HFrEF, to stabilise the patient and achieve acute control of heart rate (IIb, C) |
| Resting target | Below 110 b.p.m. initially (IIa, B) | AF without HF, candidates for select rate-control strategies: guided by underlying symptoms, in general a resting rate below 100 to 110 bpm (2a, B-R) | Below 110 b.p.m. initially, with and without HF (text) |
| Ablate and pace | With pacemaker when unresponsive to, or ineligible for, intensive rate and rhythm control, to control heart rate and reduce symptoms (IIa, B); with CRT in severely symptomatic permanent AF after at least one HF hospitalisation, to reduce symptoms, physical limitations, recurrent HF hospitalisation and mortality (IIa, B) | AVNA for refractory rates when rhythm control is not possible or has failed, to improve symptoms and QOL (2a, B-R); pacemaker first, to ensure adequacy of the pacing leads before ablation (1, B-NR); with a persistently rapid ventricular response, initial pacemaker lower rate 80 to 90 bpm after AVNA, to reduce the risk of sudden death (1, C-LD) | With CRT in severely symptomatic permanent AF with poor rate control despite medical therapy and at least one HF hospitalisation, to reduce symptoms, physical limitations, recurrent HF hospitalisation and death (IIa, B1) |
In Australia
The 2018 NHFA/CSANZ guideline was written to assist Australian practitioners in diagnosing and managing adult patients with AF.[6] The abstract of that guideline states that β-blockers or non-dihydropyridine calcium channel antagonists remain the first line choice for acute and chronic rate control.[6] The abstract also highlights deciding between rate and rhythm control at diagnosis and periodically thereafter, and says failure of rate or rhythm control should prompt consideration of percutaneous or surgical ablation.[6]
Only the abstract of that publication is held, so no NHFA/CSANZ recommendation strength is given here.[6] No newer NHFA/CSANZ AF guideline was found when this topic was built, so ESC and ACC/AHA rows carry the detail.
Where the evidence is thin
- ESC 2024: limited evidence exists to inform the best type and intensity of rate control treatment.[1]
- ACC/AHA 2023: limited data exist directly comparing rate-control agents, especially in the setting of long-term rate control.[3]
- ESC 2026 HF: data on strict versus lenient rate control in HF and AF are inconclusive, and there is insufficient evidence to support a rhythm-control strategy over a rate-control strategy in HF.[4]
- Kotecha individual-patient meta-analysis (Lancet 2014): the authors conclude that β blockers should not be used preferentially over other rate-control medications and should not be regarded as standard therapy to improve prognosis in concomitant heart failure and AF.[21]
Exam pearls
- Name the body and the year with every class: ESC 2024 puts beta-blockers and/or digoxin together for LVEF of 40% or less, to control heart rate and reduce symptoms (I, B), and ESC 2026 HF puts beta-blockers first-line (I, C), with digoxin when the rate remains high despite beta-blockers or beta-blockers are contraindicated or not tolerated (IIa, C), for short- and long-term rate control in stable HFrEF.[1][4]
- ESC 2026 HF characterises HFrEF by LVEF below 50%, against the 40% line in ESC 2024 AF.[4][1]
- Esmolol is no longer in the ESC 2024 Class IIb row for haemodynamic instability or severely depressed LVEF: the 2025 correction left intravenous amiodarone, digoxin or landiolol, which may be considered to achieve acute control of heart rate.[1][2]
- Digoxin target below 1.2 ng/mL: in ACC/AHA 2023 where measuring serum digoxin levels is indicated (COR 2a, LOE B-NR), and in ESC 2026 HF text on cardiac glycosides in HFrEF, during ongoing therapy.[3][4]
- Initial post-AVNA pacing rate: 70–90 b.p.m. in the ESC 2024 text; initial lower rate 80 to 90 bpm in the ACC/AHA 2023 Class 1 row for a persistently rapid ventricular response, to reduce the risk of sudden death.[1][3]
- Pacemaker timing: a few weeks before ablation in the ESC 2024 text; before or on the same day in the ACC/AHA 2023 row, to ensure adequacy of the pacing leads before ablation.[1][3]
- Propranolol appears in ACC/AHA 2023 Table 21, but ESC 2024 Table 12 says other beta-blockers such as propranolol and labetalol are not recommended as specific rate control therapy in AF.[3][1]
- In RATE-AF the resting heart rate at 6 months was similar with digoxin and bisoprolol (76.9 vs 74.8/min).[8]
References24ShowHide
- [1]Van Gelder IC, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J, 2024.PMID 39210723
- [2]European Society of Cardiology Correction to: 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): Developed by the task force for the management of atrial fibrillation of the European Society of Cardiology (ESC), with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Endorsed by the European Stroke Organisation (ESO). Eur Heart J, 2025.PMID 40622753
- [3]Joglar JA, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2024.PMID 38033089
- [4]Køber L, et al. 2026 ESC Guidelines for the management of heart failure. Eur Heart J, 2026.PMID 42661420
- [5]De Backer J, et al. 2025 ESC Guidelines for the management of cardiovascular disease and pregnancy. Eur Heart J, 2025.PMID 40878294
- [6]Brieger D, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Australian clinical guidelines for the diagnosis and management of atrial fibrillation 2018. Med J Aust, 2018.PMID 30067936
- [7]Jüni P, et al. 2024 Revision of the level of evidence grading system for ESC clinical practice guideline recommendations I: therapy and prevention. Eur Heart J, 2025.PMID 40116721
- [8]Kotecha D, et al. Effect of Digoxin vs Bisoprolol for Heart Rate Control in Atrial Fibrillation on Patient-Reported Quality of Life: The RATE-AF Randomized Clinical Trial. JAMA, 2020.PMID 33351042
- [9]Van Gelder IC, et al. Lenient versus strict rate control in patients with atrial fibrillation. N Engl J Med, 2010.PMID 20231232
- [10]Brignole M, et al. A randomized controlled trial of atrioventricular junction ablation and cardiac resynchronization therapy in patients with permanent atrial fibrillation and narrow QRS. Eur Heart J, 2018.PMID 30165479
- [11]Brignole M, et al. AV junction ablation and cardiac resynchronization for patients with permanent atrial fibrillation and narrow QRS: the APAF-CRT mortality trial. Eur Heart J, 2021.PMID 34453840
- [12]Ziff OJ, et al. Safety and efficacy of digoxin: systematic review and meta-analysis of observational and controlled trial data. BMJ, 2015.PMID 26321114
- [13]Bavendiek U, et al. Digitoxin in Patients with Heart Failure and Reduced Ejection Fraction. N Engl J Med, 2025.PMID 40879434
- [14]Digitalis Investigation Group The effect of digoxin on mortality and morbidity in patients with heart failure. N Engl J Med, 1997.PMID 9036306
- [15]Connolly SJ, et al. Dronedarone in high-risk permanent atrial fibrillation. N Engl J Med, 2011.PMID 22082198
- [16]Ulimoen SR, et al. Comparison of four single-drug regimens on ventricular rate and arrhythmia-related symptoms in patients with permanent atrial fibrillation. Am J Cardiol, 2013.PMID 23111138
- [17]Ulimoen SR, et al. Calcium channel blockers improve exercise capacity and reduce N-terminal Pro-B-type natriuretic peptide levels compared with beta-blockers in patients with permanent atrial fibrillation. Eur Heart J, 2014.PMID 24135831
- [18]Siu CW, et al. Intravenous diltiazem is superior to intravenous amiodarone or digoxin for achieving ventricular rate control in patients with acute uncomplicated atrial fibrillation. Crit Care Med, 2009.PMID 19487941
- [19]Perrett M, et al. Efficacy and safety of intravenous beta-blockers in acute atrial fibrillation and flutter is dependent on beta-1 selectivity: a systematic review and meta-analysis of randomised trials. Clin Res Cardiol, 2024.PMID 37658166
- [20]Khand AU, et al. Carvedilol alone or in combination with digoxin for the management of atrial fibrillation in patients with heart failure? J Am Coll Cardiol, 2003.PMID 14662257
- [21]Kotecha D, et al. Efficacy of β blockers in patients with heart failure plus atrial fibrillation: an individual-patient data meta-analysis. Lancet, 2014.PMID 25193873
- [22]Ozcan C, et al. Long-term survival after ablation of the atrioventricular node and implantation of a permanent pacemaker in patients with atrial fibrillation. N Engl J Med, 2001.PMID 11287974
- [23]Chatterjee NA, et al. Atrioventricular nodal ablation in atrial fibrillation: a meta-analysis and systematic review. Circ Arrhythm Electrophysiol, 2012.PMID 22187425
- [24]Chatterjee NA, et al. Atrioventricular nodal ablation in atrial fibrillation: a meta-analysis of biventricular vs. right ventricular pacing mode. Eur J Heart Fail, 2012.PMID 22436544