Atrial fibrillation occurring initially during acute medical illness: the heterogeneous nature of disease, outcomes and management strategies

 

Atrial fibrillation (AF) occurring transiently in association with other acute medical conditions, such as surgery, infectious diseases and myocardial infarction, has provoked substantial debates. This common arrhythmia is often seen in the emergency department, intensive care unit (ICU) and others, and has been termed as ‘secondary’, ‘acute’, ‘transient’, ‘resolved’ or ‘new-onset’ AF, all of which suggests or implies that the arrhythmia is caused by a reversible aetiology and being self-limited.^1,2 However, ‘transient’ AF may not be conclusive as AF often re-occurs during follow-up or may be merely the first detection of paroxysmal or persistent AF.

Indeed, the episode of ‘transient’ AF in association with other acute medical conditions may indicate that the individual is predisposed to developing AF. Otherwise the patient may well have had silent episodes of AF but was unaware, and the acute medical presentation may be a result of decompensation related to the arrhythmia or represent as a resultant complication.

Perhaps more relevant is whether such ‘transient’ AF is prognostically important? A major debate has been focused on the implications on clinical outcomes, and the optimal management strategy, especially for stroke prevention, given that 78.7% of patients with transient AF can be categorised as being at moderate to high stroke risk (CHA₂DS₂-VASc score ⩾1).³

In this issue of the journal, a systemic review and meta-analysis study summarises the available literature on this topic, reporting the incidence and long-term recurrence of AF that has occurred among these patients with other underlying medical conditions, ranging from the ICU and beyond.⁴ McIntyre et al.⁴ show that the reported incidence of AF detected initially during acute medical illness ranges from 1% to 44%, with a higher incidence among patients from the ICU compared with those not from the ICU (3–44% vs. 1–22%). Approximately half of the studies subjects had recurrent AF at 5 years of follow-up.⁴ Nevertheless, the true rates of incidence and recurrence of AF could be higher, given that continuous ECG monitoring reveals more occult AF and this approach of arrhythmia monitoring was not systematically conducted across the summarised studies.⁴

As previously reported, the prevalence of transient AF accounts for 7.8–10.0% among all types of AF.^3,5 Approximately 50% of patients presenting with transient AF will revert spontaneously back to sinus rhythm during their hospital stay.⁶

McIntyre et al. have also performed analyses regarding the relationship between illness severity and AF incidence, which revealed no significance.⁴ Nevertheless, the latter interpretation should be viewed with caution. The method and intensity of cardiac rhythm monitoring, and the severity of illness may have a strong association in clinical practice. Also, treating patients in the ICU with advanced illness severity could be arbitrarily assessed. Indeed, the severity of illness is not easily quantified, and the diagnosis of AF would not only be subject to the type or intensity of monitoring, but may also relate to clinicians’ preference and emphasis, especially when considering the observational and retrospective nature of the summarised studies.

Apart from the incidence and recurrence of ‘transient’ AF, outcome events are related more to the decision-making strategy for holistic management of the patients, which was not covered in the study by McIntyre et al. per se.⁴ The incidence of outcome events, including ischaemic stroke, transient ischaemic attack (TIA) and mortality, in patients with ‘transient’ AF varies with different study populations and study designs.

For example, a recent retrospective study covering approximately 50,000 subjects from general practice contributing to the Health Improvement Network of the United Kingdom (UK) compared the outcome events among subjects with ‘transient’ AF, clinical AF and no AF.³ Patients with ‘transient’ AF seem to carry a higher risk of stroke or TIA compared with non-AF individuals, but a lower risk compared with clinical AF at approximately 3 years of follow-up. The reported incidences of stroke/TIA were 12.1, 16.7 and 7.4 per 1000 person-years for subjects with ‘transient’ AF, clinical AF and no AF, respectively (rate ratio (RR) 0.73, 95% confidence interval (CI) 0.65–0.81, when compared with ‘clinical’ AF; but RR 1.65, 95% CI 1.47–1.84, compared with non-AF patients).³ Another previous study (n=1409) has demonstrated that patients with ‘transient’ AF had similar risks of stroke (hazard ratio (HR) 1.13, 95% CI 0.82–1.57) and mortality (HR 1.00, 95% CI 0.87–1.15), comparable to patients with clinical AF.¹ This high events risk among subjects with ‘transient’ AF was also evident in the Loire Valley AF project.⁷

How do we approach the management of such patients with ‘transient’ AF?

The substantial variability in AF recurrence rates and the incidence of outcome events may reflect the heterogeneity of this condition, which could be attributed to different study populations (including illness severity, etc.), the frequency and approach of cardiac rhythm assessments and the duration of study follow-up. Patients in whom AF is considered as ‘transient’ or ‘resolved’ might in fact have paroxysmal or persistent subtypes of AF which tends to recur. Reversing of the acute condition may not dependably protect against long-term AF recurrence. Thus any re-occurrence of AF would again confirm a higher risk of stroke or TIA, especially in the presence of risk factors.

The management of AF occurring transiently involves temporal and long-term considerations. In the acute phase, the primary strategy includes assessment for new haemodynamic collapse and drivers of arrhythmia, which could be corrected by cardioversion, ventricular rate control, reversal of AF triggers and electrolyte disturbances. Stroke prevention is also important in this acute phase, given such patients may be dehydrated or combined with other stroke risk factors.⁸

For long-term management, the decision-making for stroke prevention is more debatable given the dearth of solid evidence with regard to ‘transient’ AF in which this arrhythmia is considered as ‘resolved’. Currently, no guidelines have provided an explicit recommendation on anticoagulation treatment for this condition.^9,10 Only the most recent American Heart Association/American College of Cardiology/Heart Rhythm Society guidelines recommend anticoagulation with warfarin for patients with new-onset, transient AF combined with acute coronary syndrome with a CHA₂DS₂-VASc score of 2 or greater.¹¹ Two observational studies including transient AF demonstrated that anticoagulation was associated with a decreased incidence of stroke.^12,13 In another retrospective study (n=2304), long-term anticoagulation in ‘transient’ AF was not beneficial for stroke risk reduction but was associated with a 70% higher risk of bleeding.¹⁴ Nevertheless, there is still a deficiency of supportive evidence for long-term thromboembolism prophylaxis strategies in patients with ‘transient’ AF.^13–15 The inherent limitation of these studies, e.g. small sample size, retrospective or observational design, should be recognised.

The meta-analysis by McIntyre et al. did not report prescription rates of anticoagulants in the studies on ‘transient’ AF patients, although 24 out of 1409 patients from the Framingham study were taking anti-thrombotic agents at baseline.⁴ Evidence from the UK general practice study in 2016 demonstrated that among patients with ‘resolved’ AF and a CHA₂DS₂-VASc score of 1 or greater, the proportion receiving current anticoagulant treatment was 14.3%; while among patients with definitive AF, this proportion was 71.8%.³

The lack of evidence and uncertainty on management strategies for this condition call for more studies into this ‘not so rare’ condition. A holistic approach to AF management is proposed, referred to as the ABC pathway: ‘A’ avoid stroke with anticoagulation; ‘B’ better symptom management; ‘C’ cardiovascular and comorbidity risk management, including lifestyle changes.¹⁶ Application of the ABC pathway has been associated with a significant improvement in death and hospitalisation for patients with AF,¹⁷ thus supporting the concept of integrated care or holistic management. The same should be applied to ‘transient AF’ patients.

Funding

This work was supported by China Scholarship Council (201708110232).

Conflict of interest: GYHL is a consultant for Bayer/Janssen, BMS/Pfizer, Medtronic, Boehringer Ingelheim, Novartis, Verseon and Daiichi-Sankyo. He is a speaker for Bayer, BMS/Pfizer, Medtronic, Boehringer Ingelheim and Daiichi-Sankyo. No fees have been directly received personally. The other authors report no conflicts of interest directly related to this paper.

 

The opinions expressed in this article are not necessarily those of the Editors of the European Journal of Preventive Cardiology or of the European Society of Cardiology.