Dabigatran in patients with atrial fibrillation after COVID-19 hospitalization: an update of the ANIBAL protocol.

COVID-19 increases the risk of atrial fibrillation (AF) and thrombotic complications, particularly in severe cases, leading to higher mortality rates. Anticoagulation is the cornerstone to reduce thromboembolic risk in patients with AF. Considering the risk of hepatotoxicity in patients with severe COVID-19 as well as the risk of drug-drug interactions, drug-induced hepatotoxicity and bleeding, the ANIBAL protocol was developed to facilitate the anticoagulation approach at discharge after COVID-19 hospitalization. However, since the publication of the original algorithm, relevant changes have occurred. First, treatment of COVID-19 pneumonia has been modified with the use of dexamethasone or remdesivir during the first week in patients that require oxygen therapy, and of dexamethasone and/or tocilizumab or baricitinib during the second week in patients that necessitate supplementary oxygen or with a high inflammation state, respectively. On the other hand, metabolic syndrome is common in patients with AF as well as metabolic-associated fatty liver disease, and this could negatively impact the prognosis of patients with COVID-19, including high transaminase levels in patients treated with immunomodulators. The EHRA guidelines update also introduce some interesting changes in drug-drug interaction patterns with the reduction of the level of the interaction with dexamethasone, which is of paramount importance in this clinical context. Considering the new information, the protocol, named ANIBAL II, has been updated. In this new protocol, the anticoagulant of choice in patients with AF after COVID-19 hospitalization is provided according to three scenarios: with/without dexamethasone treatment at discharge and normal hepatic function, transaminases ≤2 times the upper limit of normal, or transaminases >2 times the upper limit of normal.

Introduction

Atrial fibrillation (AF) is common in patients with COVID-19. Thus, a recent meta-analysis of 31 studies found that ~8% of patients with COVID-19 also have AF, and thus have a worse prognosis.1 The ANIBAL (Anticoagulación al alta tras una Neumonía Inducida por Covid-19 y Basada en las Alteraciones Laboratoriales de la función hepática y metabólica – Anticoagulation at COVID-19 pneumonia discharge based on hepatic and metabolic laboratory alterations) protocol recently published (Figure 1) considered dabigatran as the oral anticoagulant of choice in this clinical setting due to the lower risk of hepatotoxicity, drug–drug interactions and the availability of a specific reversal agent.2 However, as the therapeutic approach of COVID-19 pneumonia has substantially changed during this period, the ANIBAL protocol was updated. Whereas certain drugs used during the first months of the COVID-19 pandemic, such as hydroxychloroquine, azithromycin or lopinavir/ritonavir, have become obsolete, current hospital protocols include remdesivir as the antiviral therapeutic of choice during the first week of COVD-19 pneumonia.3,4 In addition, corticosteroids, particularly dexamethasone, have also been considered as the first-line therapy in patients hospitalized with COVID-19 pneumonia,5,6 with a modulation of the risk of drug–drug interactions according to the European Heart Rhythm Association (EHRA) update regarding the use of direct oral anticoagulants (DOACs) in patients with AF.7 The current protocol in our hospital (University Hospital Santa Lucía, Cartagena, Spain) for the treatment of COVID-19 pneumonia (Figure 2) is based on current recommendations according to available evidence.3–6 In addition, the anticoagulation protocol during hospitalization and at discharge according to different clinical circumstances is also presented (Figure 2). During the first week, remdesivir and dexamethasone are indicated in symptomatic patients that require oxygen therapy to improve prognosis. During the second week from the onset of symptoms of COVID-19 pneumonia, tocilizumab or baricitinib, according to C-reactive protein levels, are administered together with dexamethasone in case of disease progression, even at discharge. In this line, the EHRA update has reduced the impact of the interaction between dexamethasone and dabigatran or edoxaban,7 facilitating their use at the end of the hospitalization without the need for a heparin bridge.

Figure 1

Algorithm of management of patients with atrial fibrillation at discharge after COVID-19 pneumonia. The ANIBAL I protocol.

CrCl, creatinine clearance; LMWH, low-molecular-weight heparin; ULN, upper limit of normal. Adapted from ref.2

Figure 2

Therapeutic approach of patients hospitalized with COVID-19 pneumonia at the University Hospital Santa Lucía, Cartagena, Spain.

CrCl, creatinine clearance; CRP, C-reactive protein; PaFi, inspired oxygen fraction. Reproduced from ref.2 with permission.

Metabolic syndrome is a cluster of cardiometabolic risk factors, including abdominal obesity, hypertension, hyperglycaemia, hypertriglyceridaemia and low serum HDL cholesterol, in which insulin resistance, as well as prothrombotic and pro-inflammatory states, play a key role in the aetiopathogenesis.8,9 Metabolic syndrome is a very common condition. In fact, approximately one out of four adult individuals in the United States exhibit this condition, and this proportion is expected to increase in the following years due to the increasing elderly population, sedentarism and unbalanced dietary patterns.10 Metabolic syndrome is associated with a marked risk of developing cardiovascular disease and type 2 diabetes.11 Similarly, metabolic syndrome also increases the risk of AF up to three-fold, being more frequent as the components of metabolic syndrome increase, particularly in young and middle-age patients.12,13 In addition, metabolic syndrome worsens AF burden and outcomes, thus increasing the risk of AF recurrence after single-catheter ablation.14

Fatty liver associated with metabolic dysfunction, termed metabolic-associated fatty liver disease (MAFLD), is very common in clinical practice, affecting approximately one-quarter of the population worldwide. Remarkably, COVID-19 can negatively affect liver function, particularly in patients with pre-existing liver diseases, including MAFLD. Conversely, MAFLD patients have a higher risk of progression to severe COVID-19. Thus, amongst patients with COVID-19 pneumonia, patients with MAFLD could have a higher risk of disease progression, higher likelihood of abnormal liver function during hospitalization and a longer viral shedding time compared to patients without MAFLD. With regard to the pathogenesis, both conditions share inflammatory pathways and the presence of fibrosis in MAFLD patients independently increases the severity of COVID-19. The pattern of liver injury is mostly hepatocellular rather than cholestatic.15–21

Non-alcoholic fatty liver disease has been associated with arterial hypertension, arterial stiffness, aortic valvular sclerosis, atherosclerosis, coronary artery disease and AF. In fact, non-obese, non-alcoholic fatty liver disease induces early cardiac electrical and structural changes that could promote the development of AF.22,23 As a result, the presence of MAFLD in patients with AF may be particularly problematic in individuals with COVID-19. On the other hand, immunomodulators, such as tocilizumab and baricitinib, may increase transaminase levels that could aggravate the previous liver state of these patients.24,25

Considering the current therapeutic approach of COVID-19 pneumonia and the frequent concomitance of fatty liver disease in patients with metabolic syndrome and AF, we have updated the ANIBAL protocol (ANIBAL II protocol) with the aim to facilitate the anticoagulation approach at discharge amongst patients with AF after hospitalization for COVID-19 pneumonia.

Update of the criteria to be considered for anticoagulation in patients with AF at discharge after COVID-19 pneumonia

To perform the first ANIBAL protocol,2 the following statements were considered: the ‘biological’ safety of DOACs over vitamin K antagonists (VKA);26 the potential of COVID-19 to promote liver injury, including the antiviral and immunomodulatory therapy;27–31 renal function at discharge;7 the high thrombogenicity of the COVID-19;32,33 the risk of new-onset AF amongst patients with COVID-19;34–36 and the use of DOACs with a specific reversal agent.37 Although the majority of variables analysed for the ANIBAL I protocol2 continue to be topical, new elements have been updated to be included in the new protocol (Box 1).

‘Biological’ safety of DOAC versus vitamin K antagonist as no anticoagulation control is required.

Hepatotoxicity of COVID-19 and antiviral therapy in short-term and long-term evolution as well as the presence of possible hepatic steatosis associated with metabolic syndrome.

Treatment with dexamethasone at discharge may reduce plasmatic levels of DOAC when they are used concomitantly.

Renal function at discharge (eGFR according to Cockroft–Gault).

Thrombogenicity of COVID-19: choose the most potent DOAC, preferably a twice daily dose.

Arrhythmogenicity of COVID-19: risk of new-onset atrial fibrillation that requires anticoagulation (choose DOAC to facilitate health education and adherence).

DOAC with specific reversal agent to reduce the impact of the underuse of the health system during the COVID-19 pandemic.

‘Biological’ safety of DOACs (versus VKA)

The introduction of DOACs in Spain has been restricted due to economic politics as reimbursement has been limited to some specific indications. This has led to a lower proportion of patients taking DOACs compared with other European countries.38 However, as patients taking VKA require periodic monitoring of anticoagulant activity (i.e. every 2–6 weeks, according to INR control) and this increases SARS-CoV-2 exposure, the Spanish Sanitary Authorities have reduced the restrictions of switching from VKA to DOACs, leading to a higher prescription of DOACs and, secondarily, to a reduction of outcomes. Of note, a recent study reported that, in Spain, the progressive higher prescription of DOACs has been associated with a decrease in the incidence of AF-related ischaemic stroke.39 It is not only important to reduce the risk of contagion for patients but also for healthcare professionals as they are an important risk group given their high exposure level. In fact, the main concern of healthcare professionals is contagion, particularly in the workplace.40 The use of DOACs instead of VKA thus reduces exposure to SARS-CoV-2 infection for both patients and professionals.

COVID-19-related hepatotoxicity, antiviral drugs and immunomodulatory therapy, previous hepatic disease such as MAFLD

Although uncommon, drug-induced liver injury associated with the use of tocilizumab, remdesivir or baricitinib has been described.24,25,28,29 In this context, the use of those drugs with a lower risk of drug–drug interactions may be preferable. Thus, no clinically significant interaction is expected between dabigatran and tocilizumab, baricitinib, or remdesivir (Table 1).41 With regard to oral anticoagulants in patients with AF and liver injury, a study showed that after 1-year follow-up, hepatic injury hospitalization rates were lower in initiators of DOACs versus in those taking warfarin (HR 0.57, 95% CI 0.46–0.71) and, of available DOACs, dabigatran had the lowest risk compared with warfarin (Figure 3).31 However, not all studies have shown differences regarding the risk of acute liver injury and DOACs compared with VKA in patients with AF.42

Table 1

Potential drug–drug interactions of anticoagulants with current COVID-19 therapies.

Treatment for COVID-19	Anticoagulant	Drug–drug interaction	Recommendation
Dexamethasone	VKA	The efficacy of VKA can be enhanced, promoting bleeding risk Potential clinically significant interaction	Close INR monitoring is recommended
	Apixaban	Apixaban is metabolized by CYP 3A4 and dexamethasone is a moderate CYP 3A4 inducer, decreasing apixaban levels Potential clinically significant interaction	Consider changing to another anticoagulant (i.e. LMWH) whilst concomitant treatment, up to 14 days after dexamethasone interruption
	Dabigatran	Dabigatran is not metabolized by CYP 450 The prodrug of dabigatran is a substrate of P-gp, and dexamethasone is an inducer of P-gp in pre-clinical models Potential clinically significant interaction
	Edoxaban	Edoxaban is minimally metabolized by CYP 3A4/5 Edoxaban is a substrate of P-gp, and dexamethasone is an inducer of P-gp in pre-clinical models Potential clinically significant interaction
	Rivaroxaban	Rivaroxaban is partly metabolized in the liver by CYP 3A4, CYP 1A2, and hydrolytic enzymes and partly eliminated unchanged in urine by P-gp and BCRP Dexamethasone is a moderate CYP 3A4 inducer and an inducer of P-gp Potential clinically significant interaction
	Enoxaparin	No clinically significant interaction is expected	No action is required
Tocilizumab	VKA	VKA are mainly metabolized by CYP 2C9 IL-6 (tocilizumab target) may suppress expression/activity of cytochromes Tocilizumab, per se, has no effect on cytochromes, but normalizes cytochrome activity (via IL-6 inhibition) Potential weak interaction	INR monitoring is recommended (dose increase may be required)
	Apixaban	Apixaban is metabolized by CYP 3A4 IL-6 (tocilizumab target) may suppress expression/activity of cytochromes Tocilizumab, per se, has no effect on cytochromes, but normalizes cytochrome activity (via IL-6 inhibition) Potential weak interaction	Dose increase may be required
	Rivaroxaban	Rivaroxaban is partly metabolized in the liver by CYP 3A4, CYP 1A2 and hydrolytic enzymes and partly eliminated unchanged in urine by P-gp and BCRP IL-6 (tocilizumab target) may suppress expression/activity of cytochromes Tocilizumab, per se, has no effect on cytochromes but normalizes cytochrome activity (via IL-6 inhibition) Potential weak interaction
	Dabigatran Edoxaban Enoxaparin	No clinically significant interaction is expected	No action is required
Baricitinib	VKA Apixaban Dabigatran Edoxaban Rivaroxaban Enoxaparin	No clinically significant interaction is expected	No action is required
Remdesivir	VKA Apixaban Dabigatran Edoxaban Rivaroxaban Enoxaparin	No clinically significant interaction is expected	No action is required

INR, International Normalized Ratio; LMWH, low-molecular-weight heparin; P-gp, P-glycoprotein; VKA, vitamin K antagonists.

Adapted from ref.41

Figure 3

Risk of liver injury hospitalization by type of direct oral anticoagulant (versus warfarin).

CI, confidence interval; HR, hazard ratio.

Adapted from ref.31

On the other hand, we consider that the systematic search of MAFLD in patients with metabolic syndrome with imaging techniques, particularly echography, and if positive, to rule out the presence of hepatic fibrosis with indirect methods (i.e. FibroScan)43 may not be useful in routine practice. A better approach seems to be to consider that patients with AF and metabolic syndrome may have or at least are at high risk of MAFLD. In this context, it would be preferable to prescribe a first-line oral anticoagulant, such as dabigatran, that is not metabolized by CYP 450 and is less harmful to the liver.31,41

Use of DOACs with a low risk of dexamethasone interaction

Dexamethasone is a moderate CYP 3A4 inducer and an inducer of P-glycoprotein and may potentially have a clinically significant interaction with DOACs, leading to a reduction of serum concentrations of DOACs and, consequently, to a reduction of their efficacy.41 This is very relevant considering the high thrombogenicity of COVID-19.32,33 The EHRA update states the low risk of drug–drug interactions with those DOACs that are not preferentially metabolized by the CYP 450 pathway. In this context, the widespread use of dexamethasone in patients with COVID-19, either initially or later in case of progression, including at discharge, promotes the use of dabigatran or edoxaban due to the low risk of drug–drug interactions.41,44,45

Remarkably, this change allows the use of dabigatran or edoxaban immediately at discharge, without requiring low-molecular-weight heparin bridge therapy. This is the most important novelty compared to the first ANIBAL protocol.2

Renal function at discharge (according to Cockcroft–Gault formula)

First, it should be emphasized that the estimation of glomerular filtration rate should be performed by means of the Cockcroft–Gault formula as this was the method used in the pivotal clinical trials with DOACs (versus warfarin) to determine renal function, and approval by the regulatory agencies was performed considering this method.46–49

In addition, it has been reported that, in elderly patients, the use of other methods, such as the CKD-EPI equation instead of the Cockcroft–Gault formula, to calculate renal function could overestimate the glomerular filtration rate, leading to mistakes in the dosage of DOACs and a higher risk of side effects.50

Finally, it has been shown that the use of some DOACs, such as dabigatran and rivaroxaban but not of apixaban, could be associated with a lower risk of renal adverse events compared to warfarin.51 However, it should be also considered that, amongst patients with chronic kidney disease, the risk of stroke or bleeding could differ according to the DOAC used.52

Thrombogenicity and COVID-19

Patients with COVID-19 have a higher risk of pulmonary embolism and deep vein thrombosis.53,54 Although treatment with low-molecular-weight heparin reduced the risk of pulmonary embolism or venous thrombosis during COVID-19 hospitalization, the dose to be used remains a subject of debate, particularly in critically ill patients, as well as the identification of risk factors that may recognize those patients that would benefit more from high-dose low-molecular-weight heparin.55 Similarly, it remains unclear which is the best ambulatory thromboprophylaxis approach after discharge.56,57

The European Society of Cardiology recommends that, in patients with AF during COVID-19 pandemic, unless contraindicated, full therapeutic anticoagulation for the prevention of stroke should be administered to men and women with a CHA2DS2-VASc of ≥2/3, and anticoagulation should be considered in case of CHA2DS2-VASc 1/2.33 As a result, it is not clear whether anticoagulation should also be recommended during COVID-19 due to the high thrombogenicity of this period and whether this should be maintained over time after COVID-19 recovery.

Risk of arrhythmias and COVID-19

COVID-19 affects both the respiratory system and the cardiovascular system. Patients with a prior cardiovascular condition have a higher risk of mortality and severe COVID-19. In addition, cardiovascular complications of COVID-19 include arrhythmias, particularly AF, cardiac injury, myocarditis, heart failure, arterial/venous thrombosis and acute coronary syndrome.58

A recent study that analysed 160 consecutive patients hospitalized due to COVID-19 showed that patients with new-onset AF had a higher incidence of thromboembolic events, bleeding, a combined endpoint of thrombosis and death, and longer hospital stays, indicating that an active search of AF should be performed in patients hospitalized with COVID-19 and the need of anticoagulation in case of development of AF.59

DOACs and specific reversal agents

During the COVID-19 pandemic, particularly during the lockdown period, there has been a reduction of hospital admissions for cardiovascular disease, including acute coronary syndrome, heart failure or AF, leading to an increase in adverse outcomes.60,61 This could be related, at least in part, to the lack of seeking for medical attention, despite symptoms, due to concerns about COVID-19 contagion. Thus, a study that analysed data from nationwide Danish registries showed a lower incidence of new-onset AF during the 3 weeks of lockdown compared with the previous year, with a 47% drop in total numbers, but with a higher risk of ischaemic stroke and death amongst patients with new-onset AF.61 As a result, during the COVID-19 pandemic, there has been an underdiagnosis of AF, increasing the risk of severe complications.

In order to reduce the risk of complications during the COVID-19 pandemic, and in some way to counteract the harmful effects of the underuse of health systems, the use of DOACs, such as dabigatran, which have a specific agent that rapidly reverses the effects of the drug in case of life-threatening bleeding, should be considered as first choice.37

Update of the protocol

With all this new information, we have proposed an updated protocol, the ANIBAL II algorithm for the management of patients with AF at discharge after COVID-19 pneumonia (Figure 4). This algorithm has three scenarios: Scenario 1: with or without dexamethasone and normal hepatic function; Scenario 1: with or without dexamethasone and transaminases ≤2 × upper limit of normal (ULN); Scenario 1: with or without dexamethasone and transaminases >2 × ULN. In those patients with transaminases >2 × ULN, therapeutic doses of low-molecular-weight heparin are recommended as is switching to oral anticoagulation in case of transaminases ≤2 × ULN. In all other cases, dabigatran is recommended if the estimated glomerular filtration rate is ≥30 mL/min or edoxaban 30 mg is recommended if the estimated glomerular filtration rate >15 to <30 mL/min.

Figure 4

Algorithm of management of patients with atrial fibrillation at discharge after COVID-19 pneumonia. The ANIBAL II protocol.

CrCl, creatinine clearance; LMWH, low-molecular-weight heparin; SPC, summary of product characteristics; ULN, upper limit of normality

Conclusions

The application of the ANIBAL I protocol has had a positive impact in routine practice as it has allowed a holistic approach in the management of anticoagulation in patients with AF after COVID-19 hospitalization, considering different important variables with relevant impact on the clinical course of patients such as hepatic function and the risk of drug–drug interactions, drug-induced hepatotoxicity, or bleeding, as well as considering the availability of a specific reversal agent. The impact of the ANIBAL I protocol is currently being analysed in a Spanish multicentre study. However, as the management of COVID-19 pneumonia has been revised in the last months (i.e. using remdesivir, tocilizumab, baricitinib and dexamethasone) and as greater knowledge about the impact that COVID-19 may have on patients with AF and MAFLD has been obtained, the protocol has been updated as delineated herein.