In line with the methodology recommended by the Guidelines Committee of the Spanish Society of Cardiology,1 the present article discusses the innovations and most controversial aspects of the recent guidelines for atrial fibrillation (AF).2 Notably, of 154 recommendations, only 23 (15%) have a level of evidence A and 80 (52%) a level of evidence B, confirming the need for further clinical research into this condition.

Some of the most novel or relevant aspects of the guidelines are summarized in Table, in conjunction with some critical comments.

The guidelines report interesting data on the high incidence and prevalence of AF: 1 in every 4 middle-aged adults in Europe and the United States will develop AF during their lifetime. The estimated prevalence of AF in individuals older than 20 years is 3%, in line with data from the Spanish OFRECE study,3 which showed an AF prevalence of 4.4% in the general population older than 40 years.

The hypothesis that there are different types of AF with distinct pathophysiological bases is somewhat daring. Although the etiological factors can differ, the pathophysiological consequences are probably similar, and there is no evidence for the hypothesized ⿿distinct pathophysiological mechanisms⿿ and ⿿different types of AF⿿, except in very specific disease and clinical situations. The attempts of the guidelines to link distinct etiological factors with specific pathophysiological mechanisms are somewhat speculative but nonetheless interesting because this approach might suggest future lines of research to better understand the pathophysiology of AF.

The use of pulse palpation or electrocardiography for any reason as part of opportunistic screening of at-risk patients can detect about 3% of patients older than 65 years with asymptomatic AF. Asymptomatic AF episodes can also be detected in patients with implanted devices, which can record atrial high rate episodes (AHREs). Such episodes have been associated with a higher risk of embolic events, and it is hypothesized that anticoagulants might effectively prevent stroke. Implantable devices should be interrogated to identify AHREs, and patients with AHRE should undergo AF screening and embolic risk stratification.

A considerable percentage of strokes (about 6.5%) might be due to undetected AF episodes. The prevalence of asymptomatic AF is even higher in selected populations with cryptogenic stroke. There is no clear consensus on the optimal monitoring method or whether it needs to be invasive or not. Continuous monitoring is recommended for 72hours after an ischemic stroke (class of recommendation I, level of evidence B) and, subsequently, ambulatory monitoring with noninvasive systems or implantable loop recorders for all patients who have had an ischemic stroke (IIa B).

The new guidelines classify AF episodes that are cardioverted within 7 days as paroxysmal; in previous guidelines, all episodes requiring cardioversion were considered persistent, even those lasting less than 7 days. This criteria change might be confusing and it remains to be seen whether these definitions will also be adopted by the guidelines of other scientific societies. A new type of AF is introduced⿿long-standing paroxysmal⿿but is insufficiently explained.

The classification of AF based on etiology no longer seems to be relevant from the therapeutic perspective.

The classification is now based on AHRE symptoms, with class 2 subdivided into 2a⿿mild symptoms⿿and 2b⿿severe symptoms. Thus, patients in class 2b could benefit from adequate rhythm control.

The guidelines highlight the risk factors and comorbidities associated with AF and quantify their effect on AF risk. In addition to the usual risk factors, particular emphasis is given to obesity, alcohol consumption, smoking, and regular vigorous exercise.

Some drugs used in the treatment of AF in heart failure patients decrease AF risk (angiotensin-converting enzyme inhibitors, beta-blockers [BBs], and eplerenone), but not others (angiotensin II receptor blockers and neprilysin inhibitors). To reduce recurrence, weight loss is recommended (IIa B) for obese patients, whether via AF ablation or other strategies. For patients with chronic obstructive pulmonary disease, hypoxemia and acidosis correction is recommended in acute episodes (IIa C), as well as obstructive sleep apnea treatment (IIa B). A useful addition would be discussion of the relationship of reduced alcohol consumption and vigorous exercise with AF risk reduction.

A novel recommendation is to abandon the term ⿿nonvalvular AF⿿ and to refer to the specific underlying condition. Although up to 30% of patients with AF have some type of valve involvement, only mechanical heart valve prostheses or rheumatic mitral stenosis impact stroke risk.

The guidelines propose (IIa B) the establishment of specific integrated multidisciplinary management programs for patients with AF to increase adherence to the guidelines⿿ recommendations and improve prognosis. This recommendation is supported by 2 randomized studies that compared this strategy with usual care in tertiary centers, although one of these studies found only a marginal benefit. Various observational studies have also found fewer events with these programs. Although this strategy probably improves the prognosis and treatment of patients with AF, we believe that the current evidence is insufficient.

A controversial aspect is the use of biomarkers (high-sensitivity troponin and N-terminal pro⿿B-type natriuretic peptide) to stratify thromboembolic risk, although the recommendation is IIb B. The diagnostic and prognostic usefulness of these biomarkers has been shown in other cardiovascular diseases but confirmation is required of their additional prognostic value in clinical risk scales. The studies contributing to the inclusion of these biomarkers in the guidelines were performed in patients enrolled in large trials comparing vitamin K antagonists (VKAs) with new oral anticoagulants (NOACs). Accordingly, the role of biomarkers is controversial because all patients were anticoagulated. Their use in low-risk patients has not been validated and their role in patients without anticoagulation remains to be studied.

The guidelines strengthen the usefulness of the CHA2DS2-VASc score for predicting thromboembolic risk. Anticoagulation is not recommended in patients with a CHA2DS2-VASc score of 0, even women (III B). However, anticoagulation is assigned a IIa B recommendation in men and women with a single stroke risk factor and a I A recommendation for men with 2 or more risk factors and women with at least 3. Thus, the female sex is no longer considered an independent risk factor when the anticoagulation indication is being assessed. Men with 1 point and women with 2 points are combined in a special section and the recommendation is for an individual risk evaluation. A recent meta-analysis4 showed that the embolic risk with a CHA2DS2-VASc score of 1 is below the estimated net-benefit threshold for VKAs (1.7% per year), whereas the net-benefit threshold would be lower for NOACs (0.9% per year).

A list of modifiable risk factors for bleeding is proposed but no bleeding risk scale is explicitly recommended. Estimation of bleeding risk was a landmark in the treatment of patients with AF. The HAS-BLED score identifies modifiable risk factors for bleeding, such as the use of antiplatelet agents, uncontrolled hypertension, and poor control of VKA therapy. New bleeding risk scales, such as ATRIA, ORBIT, and ABC,5,6 have replaced the HAS-BLED score with a table of bleeding risk factors grouped into modifiable, potentially modifiable, and nonmodifiable. Another surprising change is the inclusion of biomarkers that are not used in daily clinical practice, such as GDF-15.

VKAs are the sole therapeutic option for patients with rheumatic mitral valve disease or a mechanical heart valve prosthesis. The guidelines note the importance of adequate VKA therapy in anticoagulation control, estimated by the time in therapeutic range (TTR). These antagonists continue to have a I A recommendation when the TTR is adequate. Notably, no cutoff point has been established for the TTR in this section or in the recommendation box. In recent years, a new score, the SAMe-TT2R2, has been shown to be useful in identifying patients failing to achieve optimal anticoagulation control.7 This scale has been validated in a Spanish population.8 Its use would increase the percentage of patients considered suitable for these drugs who are not predicted to achieve stable anticoagulation⿿exceeding 40% of patients in different series⿿and would also avoid the 6-month transition period for VKAs mandated by current Spanish regulations.9

A clear preference is expressed for NOACs over VKAs for nonvalvular AF (I A recommendation). However, it seems illogical to reduce the recommendation for a change to a NOAC when inadequate control of the international normalized ratio (INR) is achieved with VKAs: this recommendation was I B in 2012 and is now IIb A. In addition, the bleeding risk associated with NOACs is similar to that of aspirin10 and they show a greater ability to reduce embolic events.11

A conclusive recommendation is made to avoid the use of aspirin to prevent stroke, independently of patients⿿ embolic risk (III A). The use of antithrombotics is also discouraged if there is no specific indication for antiplatelet therapy.

The guidelines incorporate edoxaban and present a ⿿per-protocol on-treatment analysis⿿, which favors the drug. We believe that it would have been pertinent to discuss the main analysis in more detail. Table 13 shows discrepancies in dosage with the summary of product characteristics of edoxaban. The authors of the guideline have expressly chosen the criteria used in the clinical studies, given the international setting of the guidelines and possible differences in labeling between countries.

Two interesting differences with previous guidelines are the clear declaration of the safety of anticoagulation with NOACs in patients with mild-to-moderate chronic kidney disease and the recommendation to monitor renal function to enable anticoagulant dose modification and risk redefinition.

We must remember that no important clinical trial has validated the utility of acenocoumarol, the most widely used VKA in Spain. Acenocoumarol has a plasma half-life of 8 to 11hours, similar to that of NOACs. Warfarin has a more prolonged half-life, which stabilizes the anticoagulation level.

Spain is not likely to be able to apply these recommendations on the use of NOACs, largely due to a lack of public funding.

Due to the results of a meta-analysis, surgical closure of the atrial appendage has gone from being supported by expert consensus opinion to having a level of evidence B. Similarly, the guidelines add the possibility of atrial appendage exclusion with thoracoscopy with the same recommendation level. However, despite the reduced embolic risk, the guidelines recommend that oral anticoagulation therapy be maintained after surgical exclusion of the atrial appendage. Percutaneous occlusion of the atrial appendage as an alternative due to contraindication to oral anticoagulation continues to be only a IIb B recommendation, despite recent evidence showing a high success rate and low complication rate (< 5%) with this procedure.

No major changes have been made to these recommendations. For acute rate control, the guidelines continue to preferentially recommend BBs and nondihydropyridine calcium antagonists (CAs) over digitalis. For long-term rate control, the same strength of recommendation is given to these 3 types of drugs (I B); specific drug selection should be based on patient characteristics. The guidelines have removed the recommendation not to use digitalis drugs as the only agents for heart rate control in patients with paroxysmal AF. This advice contrasts with North American guidelines, which only consider BBs and CAs for long-term heart rate control and restrict the use of digitalis to patients with heart failure.13 Digitalis drugs have few negative dromotropic effects in adrenergic states, as usually occur during paroxysmal AF episodes. The data from the FANTASIIA registry14 show that the most widely used drugs for heart rate control in AF patients in Spain are BBs in most patients (60.2%), followed by digoxin (19.5%) and, to a lesser extent, CAs (10.7%). The guidelines recommend the combination of BBs or CAs with digitalis. However, these drug combinations are controversial because the results published on these combinations are discordant.

The guidelines barely consider combined heart rate and rhythm control, despite the possible effects of drug interactions. Combinations of some of these drugs can enhance their negative chronotropic effects. This is the case for the combined use of BBs and the type C antiarrhythmic drugs flecainide and propafenone because both types of drugs profoundly affect sinus node function. They should thus be used with caution in elderly patients or patients with suspected sinus node dysfunction. Recent publications discourage the combination of dronedarone with any agents that depress atrioventricular conduction, such as digoxin, given that the PALLAS trial15 linked their combined use with increased mortality (relative risk up to 7.3 times higher than digoxin or dronedarone in monotherapy). This combination is even discouraged in the European data sheet for dronedarone and it is striking that it has not been reflected in the new guidelines.

Atrioventricular node ablation and pacemaker implantation is one way to achieve heart rate control after drug therapy failure. However, the guidelines have largely ignored the type of cardiac resynchronization device, whether single-chamber, dual-chamber, or triple-chamber.

The first major comment of the guidelines in this area is that the benefits of rhythm control therapy are limited to symptom improvement. Because there are no data on reduced mortality, there are no guidelines for asymptomatic patients. The publication of new studies such as the CABANA (Ablation Versus Anti-arrhythmic Drug Therapy for Atrial Fibrillation Trial) trial (NCT00911508) are eagerly awaited.

There are no changes in the indication or in the usual cardioversion procedure in patients with new-onset AF. Antiarrhythmic drug selection is based on their safety profile in the different heart diseases. The guidelines note that the new drugs, such as vernakalant, do not significantly improve efficacy and recommend which agents are suitable for each disease.

An alternative option is ablation, which is awarded a IIa recommendation. This class of recommendation might be a source of controversy due to the considerable number of randomized studies showing that ablation is superior to antiarrhythmics in patients without previous antiarrhythmic therapy. Because most recurrences are observed in the first months after vascular surgery, the guidelines comment that amiodarone reduces recurrence for 4 weeks, but not for longer times. Intermittent amiodarone use after vascular surgery is associated with greater mortality and more hospitalizations vs continuous use. The effectiveness of new drugs such as ranolazine or combinations of antiarrhythmic drugs such as dronedarone and ranolazine is called into question due to recent publications.

The guidelines affirm that ablation is clearly superior to antiarrhythmic drugs when one of these drugs has previously failed. Although studies comparing ablation and antiarrhythmic drugs in untreated patients are mentioned, the findings are not collected in a specific comment.

The guidelines note that, apart from antrum isolation, no technique has been shown to be superior to catheter ablation for either paroxysmal or persistent FA. The usefulness of certain interesting techniques, such as rotor ablation, remains to be shown in clinical trials. However, cryoablation is not inferior to conventional point-to-point ablation. Although experts indicate that cavotricuspid isthmus ablation should be performed in patients with AF and atrial flutter, there are no conclusive data supporting the effectiveness of this approach.

Surprisingly, despite the recent position document of the EHRA with other societies on periprocedural antithrombotic management,16 this article has not been cited. Due to recent evidence, the guidelines state that patients treated with VKAs should maintain this treatment during the procedure. However, the guidelines fail to note whether the benefit observed in patients with AF is long-lasting. Thus, there are doubts about the general suitability of anticoagulant maintenance, whether VKAs or NOACs, during ablation procedures (IIb).

Because AF and ventricular dysfunction frequently coexist and worsen each other's symptoms, it is important to define the value of ablation in these patients. The guidelines fail to cite any study showing that ablation boosts ventricular function recovery in patients with either adequate or inadequate ventricular heart rate control17 and that the main determinant of ventricular function recovery failure is the presence of myocardial infarction. No convincing evidence is put forward showing that ablation is superior to strict control of ventricular heart rate with node ablation and subsequent resynchronization device implantation. However, the guidelines do state that ablation is superior to amiodarone in patients with ventricular dysfunction and an implantable device.

The most notable contribution in this area concerns ablation that is not combined with another type of surgery. The development of videothoracoscopy and bipolar radiofrequency ablation with cryothermy has simplified the procedure and has reduced complications to such an extent that it can be considered another surgical option for the treatment of AF. Effectiveness rates ranging between 70% and 90% justify its use in patients who have undergone failed percutaneous procedures.

The guidelines discuss 2 aspects of hybrid rhythm control therapy involving antiarrhythmic drugs: their combined use with either catheter ablation or pacemakers. Regarding the first approach, although the guidelines establish that it appears to be common knowledge to use antiarrhythmic drugs in patients with AF recurrence after an ablation procedure, no guidelines or strength of recommendation are established for its use, despite various studies showing a lower incidence of AF and fewer hospitalizations after ablation.18 The guidelines mention the possible use of cavotricuspid isthmus ablation and continued antiarrhythmic drug therapy⿿mainly with type IC antiarrhythmic drugs⿿in patients with AF and organized atrial flutter, but no strength of recommendation is provided. In the European and North American guidelines for supraventricular tachycardia treatment, a strong recommendation (I and IIa) is awarded to this approach.

The guidelines also mention combination hybrid therapy involving a pacemaker and antiarrhythmic drugs, largely due to the negative chronotropic effects of these drugs. Pacemakers permit the use of higher doses of these drugs. However, catheter ablation can sometimes obviate the need for antiarrhythmic drugs and pacemakers. Some of these patients have reverse remodeling of the sinus node, which can return after AF termination with ablation.19 However, early sinus node dysfunction can be unmasked by the antiarrhythmic drugs and would require monitoring.

This section has been reorganized to remove the brief comments of previous guidelines on hyperthyroidism and pulmonary disease.

The guidelines stress the need to use medical records and other tests to evaluate the presence of channelopathies and other cardiomyopathies in young patients without apparent structural heart disease but with AF; systematic genetic studies are not considered necessary. The section on sport and AF is well supported, although it might be imprecise, even confusing, regarding the effects of detraining.

Another development is that the pertinent information on AF (and other atrial arrhythmias) in adult patients with congenital heart disease is organized in a separate section. However, the recommendations are level C only, except for those referring to atrial flutter.

The guidelines highlight 9 specific situations requiring appropriately powered randomized trials to robustly establish an adequate recommendation level. Some of these are especially relevant, such as anticoagulation in patients with AF and severe chronic kidney disease, anticoagulation after a bleeding or stroke event, the optimal timing of cardioversion for new-onset AF without anticoagulation, and atrial appendage occlusion for stroke prevention.

There is a lack of scientific evidence in diverse areas of knowledge supporting some of the recommendations or preventing their implementation. Examples include AHREs, the role of anticoagulation in biological prostheses and rheumatic mitral valve disease, and the role of ablation in persistent AF. Questions remain regarding atrial appendage closure/occlusion, because the indications included in the clinical guidelines are different from those evaluated in the main clinical studies, and the need to maintain anticoagulation after device implantation.

Although the guidelines have a strong clinical focus and have incorporated the relevant available evidence, it is patently clear that many of the situations considered require greater scientific support.

A notable inclusion is the integrated management of AF because this approach recognizes the epidemiological importance of the disease, its relationship with other risk factors, the need for early detection, its multidisciplinary treatment, and the role played by the patient.

The largest section deals with anticoagulant therapy and considers various related factors such as the correct indication, bleeding risk, and its combination with antiplatelet agents or atrial appendage exclusion. All of these factors interact and have distinct effects due to differences in clinical situations.

The guidelines consider heart rate and rhythm control therapies, with few changes, and lack definitions in specific situations, such as certain hybrid therapies or the role of ablation. However, the surgical setting is considered.

Special situations are defined, and areas lacking clarity are recognized.

There are some important changes concerning the classifications of paroxysmal and persistent AF, the role of female sex as an embolic risk factor, and the abandonment of the HAS-BLED stratification scale of bleeding risk, among others, which could lead to some confusion upon their incorporation into clinical practice.

These guidelines are welcome for persuasively arguing for a need for systematic care in patients with AF.

However, the following authors declare financial relationships with entities that may have interests in certain aspects of the article: F. Arribas: consulting (Boheringer-Ingelheim, Daiichi Sankyo, Lina Nova), presentations (Boston Scientific, Bayer), educational presentations (Medtronic, Boston Scientific, Biosense), attendance at meetings (Medtronic, Lisa Nova, Boston Scientific). I. Roldán: presentations (AstraZeneca). J. L. Merino: board member (Daichii Sankyo, Biosense, Sanofi), consulting (Biotronik, Cardiome, Daichii Sankyo, Medtronic, Sanofi, Sorin), presentations (Biotronik, Boston Scientific, Medtronic, St. Jude, Sanofi), educational presentations (St. Jude), attendance at meetings (Biotronik, Boston Scientific, Medtronic, St. Jude, Sanofi), symposium sponsorship (Bayer, Boheringer-Ingelheim, Pfizer, Cardiome, Boston, Equipamiento Scientific, Medtronic, St. Jude, Sorin, Microport). R. Ruiz: consulting (Medtronic, Boston Scientific), presentations (Medtronic, Boston Scientific, St. Jude). L. Mont: consulting (St. Jude Medical, Boston Scientific, Johnson & Johnson, Medtronic Boehringer), expert testimony (St. Jude, Boston Scientific, Johnson & Johnson, Medtronic, Boheringer-Ingelheim), presentations (St. Jude, Boston Scientific, Johnson & Johnson, Medtronic, Boehringer), educational presentations (Johnson & Johnson). F. Marín: consulting (Daiichi Sankyo, Bayer), attendance at meetings (Daiichi Sankyo, Bayer), presentations (AstraZeneca), research project (Pfizer/BMS, Boehringer-Ingelheim). V. Barrios: member of the board of directors (Bayer, Boheringer-Ingelheim and Daiichi Sankyo), consulting (Bayer, Boehringer Ingelheim, Daiichi Sankyo), presentations (Bayer, Boheringer-Ingelheim, Daiichi Sankyo, Pfizer). G. Barón: consulting (Bayer, Boheringer-Ingelheim, Daiichi Sankyo, Pfizer), expert testimony (Bayer, Boheringer-Ingelheim, Daiichi Sankyo, Pfizer), presentations, manuscript preparation (Bayer), attendance at meetings (Bayer, Pfizer). E. Díaz: consulting (Medtronic, Boston Scientific), presentations (Medtronic, Boston Scientific, St. Jude Medical, Biotronik), educational presentations (Medtronic, Boston Scientific, St. Jude, Biotronik), attendance at meetings (Medtronic, St. Jude). J. L. Ferreiro: presentations (Daiichi Sankyo, Pfizer). N. Pérez-Castellano: consulting (Medtronic, Boston Scientific, St. Jude Medical, Livanova), presentations (Medtronic, Boston Scientific, St. Jude Medical, Livanova), manuscript preparation (Boston Scientific), attendance at meetings (Boston Scientific). A. Tello: grants (Astra Zeneca), presentations (Amgen, Astra Zeneca, Merk, Biosensors), attendance at meetings (Daichii Sankyo). F. Alfonso: attendance at meetings. I. Ferreira-González: presentations (Bayer, Pfizer, Boheringer-Ingelheim), educational presentations (Bayer, Pfizer, Boehringer, Abbott), consulting (Bayer). L. Rodríguez-Padial: presentations (MSD, Pfizer, Rovi, Daiichi Sankyo, Amgen, Sanofi), attendance at meetings (Menarini, Servier, Sanofi).