Challenges in atrial fibrillation: detection, alert systems, fibrosis, and infection

For the podcast associated with this article, please visit https://academic.oup.com/eurheartj/pages/Podcasts. Atrial fibrillation is the most common cardiac arrhythmia and its prevalence is predicted to double over the next decades due to an ageing population and the rise in prevalence of risk factors such as hypertension,1 and diabetes and heart failure.2 Aside from an impairment of quality of life, atrial fibrillation is associated with stroke3 and dementia,4 which makes its detection an important medical need, as Barbara Casadei et al. from the John Radcliffe Hospital in Oxford, UK outline in their review ‘Screening for atrial fibrillation: a call for evidence’.5 The authors remind us that atrial fibrillation carries a five-fold increased stroke risk, while anticoagulation in eligible patients can reduce this risk by ∼65%.6,7 Unfortunately, in many, atrial fibrillation remains undetected and untreated, either because the patients are asymptomatic or due to the paroxysmal nature of the arrythmia. Screening might be one approach to increase detection rates and reduce ischaemic strokes through timely anticoagulation. However, international task forces currently recommend against screening due to costs and uncertain benefits.8 A number of large randomized controlled trials have commenced to determine the cost-effectiveness and benefits of screening. The recent AppleWatch study demonstrated how advances in technology are providing the public with self-screening devices that are increasingly affordable and accessible.9 Healthcare professionals should be aware of the implications of these diagnostic pathways and treatments. Once detected, anticoagulation is underprescribed in atrial fibrillation.10 In their article ‘Alert-based computerized decision support for high-risk hospitalized patients with atrial fibrillation not prescribed anticoagulation: a randomized, controlled trial (AF-ALERT)’ Gregory Piazza and colleagues from the Brigham and Women’s Hospital in Boston, Massachusetts, USA tested an alert-based computerized decision-support strategy to increase anticoagulation prescription in 458 hospitalized atrial fibrillation patients with a CHA2DS2-VASc score ≥1 who were not yet prescribed anticoagulant therapy. Patients were randomized to alert-based computerized decision support or no notification. In the alert group, 25.8% were prescribed anticoagulation during the hospitalization, but only 9.5% in the control group; at discharge, these percentages were 23.8% vs. 12.9% and at 90 days 27.7% vs. 17.1%. The alert intervention reduced death, myocardial infarction, cerebrovascular events, and systemic embolism at 90 days to 11.3% compared with 21.9% in controls (Figure 1). The alert group reduced myocardial infarction at 90 days by 87% and cerebrovascular events or systemic embolism at 90 days by 88% (Figure 1). Thus, an alert-based computerized decision-support strategy increased anticoagulation in atrial fibrillation and reduced major adverse cardiovascular events (MACE), including myocardial infarction and stroke. These impressive findings are further discussed in an Editorial by Jonathan P. Piccini from the Duke University Medical Center in Durham, North Carolina, USA.11 Kaplan–Meier curves for freedom from death, myocardial infarction, cerebrovascular accident (stroke or transient ischaemic attack), and systemic embolic event at 90 days (A), Kaplan–Meier curves for freedom from major adverse events at 90 days (death, myocardial infarction, cerebrovascular accident, systemic embolic event, and major or clinically relevant non-major bleeding) (B), cumulative incidence of myocardial infarction at 90 days (C), and cumulative incidence of cerebrovascular accident or systemic embolic event at 90 days (D) (from Piazza G, Hurwitz S, Galvin CE, Harrigan L, Baklla S, Hohlfelder B, Carroll B, Landman AB, Emani S, Goldhaber SZ. Alert-based computerized decision support for high-risk hospitalized patients with atrial fibrillation not prescribed anticoagulation: a randomized, controlled trial (AF-ALERT). See pages 1086–1096). Kaplan–Meier curves for freedom from death, myocardial infarction, cerebrovascular accident (stroke or transient ischaemic attack), and systemic embolic event at 90 days (A), Kaplan–Meier curves for freedom from major adverse events at 90 days (death, myocardial infarction, cerebrovascular accident, systemic embolic event, and major or clinically relevant non-major bleeding) (B), cumulative incidence of myocardial infarction at 90 days (C), and cumulative incidence of cerebrovascular accident or systemic embolic event at 90 days (D) (from Piazza G, Hurwitz S, Galvin CE, Harrigan L, Baklla S, Hohlfelder B, Carroll B, Landman AB, Emani S, Goldhaber SZ. Alert-based computerized decision support for high-risk hospitalized patients with atrial fibrillation not prescribed anticoagulation: a randomized, controlled trial (AF-ALERT). See pages 1086–1096). Unfortunately, anticoagulation with either non-vitamin K antagonist oral anticoagulants (NOACs) or vitamin K antagonists is not without side effects. Besides the well-known bleeding risks,12 other unwanted drug effects have to be considered, as shown in the article ‘Fracture risks among patients with atrial fibrillation receiving different oral anticoagulants: a real-world nationwide cohort study’13 by Ching-Hui Loh and colleagues from the Hualien Tzu Chi Hospital in Hualien, Taiwan. They evaluate the fracture risk among adult patients newly diagnosed with atrial fibrillation treated with NOACs or warfarin. After matching, 19 414 patients were included, i.e. 9707 each in the NOAC and warfarin groups, with follow-up for 2.4 years. Compared with warfarin, overall NOACs were associated with a reduced fracture risk with a hazard ratio (HR) of 0.84. Subanalyses revealed for dabigatran a HR of 0.88, for rivaroxaban a HR of 0.81, and for apixaban a HR of 0.67. Thus, compared with warfarin, NOACs were associated with a reduced fracture risk among atrial fibrillation patients. Therefore, if oral anticoagulants are indicated, NOACs rather than warfarin should be considered to lower the risk of fractures. As outlined in an interesting Editorial by Raffaele De Caterina from the University of Pisa in Italy,14 further studies are needed to investigate the underlying mechanisms and confirm causality. Initially, atrial fibrillation is paroxysmal in nature and commonly triggered by alcohol,15,16 drugs, inflammation,17,18 or infection, as outlined in their article ‘One-year outcomes in atrial fibrillation presenting during infections: a nationwide registry-based study’ by Anna Gundlund and colleagues from the Gentofte Hospital in Hellerup, Denmark.19 The authors compared 1-year outcomes in 30 304 patients with infection-related atrial fibrillation with 90 912 with infection, but without atrial fibrillation. In infection-related atrial fibrillation, the 1-year risks of atrial fibrillation and thrombo-embolic events were 36.4% and 7.6%, respectively, but only 1.9% and 4.4%, respectively in infection without atrial fibrillation (Figure 2). Infection-related atrial fibrillation was associated with an increased long-term risk of atrial fibrillation and thrombo-embolic events compared with those with infection without atrial fibrillation, with an HR of 26 for atrial fibrillation and of 2.1 for thrombo-embolic events. Thus, during the first year after discharge, one-third of patients with infection-related atrial fibrillation had atrial fibrillation again. Infection-related atrial fibrillation was associated with increased risk of thrombo-embolic events compared with infection without atrial fibrillation. As such, atrial fibrillation related to infection may merit treatment and follow-up similar to that of atrial fibrillation not related to infection. The manuscript is accompanied by an Editorial by Christopher John Boos from the Poole Hospital in the United Kingdom.20 Cumulative incidence of atrial fibrillation, thromboembolic events, and death according to atrial fibrillation or not during infection (from Gundlund A, Olesen JB, Butt JH, Christensen MA, Gislason GH, Torp-Pedersen C, Køber L, Kümler T, Fosbøl EL. One-year outcomes in atrial fibrillation presenting during infections: a nationwide registry-based study. See pages 1112–1119). Cumulative incidence of atrial fibrillation, thromboembolic events, and death according to atrial fibrillation or not during infection (from Gundlund A, Olesen JB, Butt JH, Christensen MA, Gislason GH, Torp-Pedersen C, Køber L, Kümler T, Fosbøl EL. One-year outcomes in atrial fibrillation presenting during infections: a nationwide registry-based study. See pages 1112–1119). While the understanding of atrial fibrillation has evolved from the view of a sole rhythm disturbance, Christian Sohns and colleagues from the Herz- und Diabeteszentrum NRW, Ruhr-Universität Bochum in Bad Oeynhausen, Germany focus in their review entitled ‘Atrial fibrillation and cardiac fibrosis’ on atrial fibrosis that can be visualized using late gadolinium enhancement cardiac magnetic resonance imaging21 and show that it is a powerful predictor for the outcomes of the condition.22 However, while a strategy of individual and fibrosis-guided management of atrial fibrillation looks promising, results from prospective multicentre trails are pending. The authors provide an overview about the relationship between cardiac fibrosis and atrial fibrillation focusing on translational aspects, clinical observations, and fibrosis imaging to emphasize personalized management taking into account the individual amount and distribution of fibrosis.23 As health systems around the world increasingly look to measure and improve the value of care they provide to patients, being able to measure the outcomes that matter most to patients is vital. In their article ‘Development of an international standard set of outcome measures for patients with atrial fibrillation: a report of the International Consortium for Health Outcomes Measurement (ICHOM) atrial fibrillation working group’, William Henry Seligman and colleagues from the International Consortium for Health Outcomes Measurement or ICHOM in London, UK note that in order to support the shift towards value-based healthcare in atrial fibrillation, the ICHOM assembled an international working group of 30 volunteers, including health professionals and patient representatives, to develop a standardized minimum set of outcomes for benchmarking care delivery in clinical settings.24 Using an online-modified Delphi process, outcomes important to patients and health professionals were selected and categorized into (i) long-term consequences of disease outcomes; (ii) complications of treatment outcomes; and (iii) patient-reported outcomes. They identified demographic and clinical variables for use as case-mix risk adjusters, i.e. demographics, comorbidities, cognitive function, date of diagnosis, disease duration, prescribed medications, and atrial fibrillation procedures, as well as smoking, body mass index, alcohol intake, and physical activity. The standard set underwent an open review process in which over 80% of patients surveyed agreed with the outcomes captured by the standard set. Thus, implementation of these consensus recommendations could help institutions to monitor, compare, and improve the quality and delivery of chronic atrial fibrillation care. Their consistent definition and collection, using ICD codes where applicable, could also broaden the implementation of more patient-centric clinical outcomes research in atrial fibrillation. The editors hope that this issue of the European Heart Journal will be of interest to its readers. With thanks to Amelia Meier-Batschelet for help with compilation of this article.