Update on stroke prevention in atrial fibrillation
In a comprehensive talk, Hobbs considered a range of aspects relevant to the management of patients with atrial fibrillation (AF), the most common cardiac arrhythmia. As an arrhythmia, AF is a relative benign condition and some patients have no symptoms. The main reason to treat these patients is their nearly 5-fold increased risk of embolic stroke, as compared with those with normal sinus rhythm. Severity of AF-related stroke is generally worse than non-AF-related stroke, resulting in higher levels of disability. Hobbs postulated that at least half of strokes may be prevented, because they are due to AF and hypertension, and thus are preventable by interventions in primary care.
Stroke risk in AF can be mitigated with anticoagulant therapy. But before stroke risk can be treated, patients at risk need to be identified. Both the ESC and the UK NICE guidelines recommend opportunistic AF screening. This advice is based on the SAFE study, which demonstrated that opportunistic screening for AF was more cost-effective than systematic screening to detect new cases. The SAFER study subsequently determined that opportunistic screening is sufficient.
Once AF is detected, the next important step is to assess stroke risk, with the CHA₂DS₂-VASc score. The score advices on who should, and who should not receive anticoagulation therapy, to lower stroke risk. Vitamin K antagonists (VKA) such as warfarin show great efficacy in reducing strokes in AF, when compared with no treatment. The BAFTA study compared warfarin with low dose aspirin in the elderly, and showed that the risk of fatal or disabling stroke, intracranial haemorrhage or clinically significant arterial embolism was more than halved with warfarin, without increasing the risk of major haemorrhage. The benefit-risk profile actually improves with increasing age or with higher CHA₂DS₂-VASc scores. A caveat of VKA therapy is that it has a very narrow therapeutic window, and it has many interactions with foods and drugs. Thus, patients need to be monitored to ensure tight control of the level of anticoagulation. Time in therapeutic range should be above 60%, otherwise the risk is larger than when people are not treated at all.
The next thing to do is assess bleeding risk, with the HAS-BLED score. This score can be used to identify which risk factors can be modified. The ESC guidelines no longer speak of the score, because it was used to deny patients treatment, which is not what it should be used for. Anticoagulated AF patients have better outcomes, regardless of stroke and bleeding risk.
Hobbs also discussed whether there is a role for antiplatelet therapy in AF. As mentioned, BAFTA demonstrated that treatment with VKA and aspirin showed a similar risk of extracranial bleeding. A more recent meta-analysis of trials, however, has demonstrated that the efficacy of aspirin is lost in elderly patients of 82 years and older. ESC and NICE state that aspirin should not be offered for stroke prevention in AF. Dual antiplatelet therapy with aspirin and clopidogrel may be considered, only if anticoagulation is contraindicated or not tolerated, and if the person has CHA₂DS₂-VASc score ≥2.
The coagulation process can be targeted in many ways, and various therapeutic approaches have been developed to do so. While the NOACs rivaroxaban, apixaban and edoxaban inhibit factor Xa, dabigatran inhibits factor IIA, and VKA targets several factors in the coagulation cascade. The different NOACs have been evaluated in trials with different patient populations and different risk levels. Differences also exist in the pharmacodynamics of the members of this drug class. It is important to be aware of the differences to make the best choice for an individual patient. The ESC guidelines prefer NOACs over VKA, as the evidence base is strong, which now also includes a lot of real-world data.
Issues to consider when using NOACs include which member of the class to choose when initiating anticoagulation therapy. Some agents are more appropriate than others in patients with specific characteristics, such as previous CV disease, GI disorders or previous GI bleeding, or renal impairment. Some agents allow dose adjustments in specific cases. Drug interactions in general are sparse with NOACs. With regard to bleeding risk in the real world, data have been very consistent with the NOAC trial data. The drawback of real-world data is that preferential prescription bias cannot be ruled out, but the data suggest at least that if VKA is the reference, the NOACs are at least as safe with regard to major bleeds.
A fact of the real world as opposed to trials, is that patients have more comorbidities and polypharmacy. The effect of polypharmacy in the elderly on the efficacy and safety of apixaban was evaluated in a substudy of the ARISTOTLE trial. This showed that efficacy remained the same despite polypharmacy (comparing 0-5 drugs, with 6-8 and ≥9 drugs). The impact of apixaban on major bleedings declined with increasing polypharmacy, but the highest category still showed a benefit.
In peri- and post-operative management of patients undergoing elective surgery, it does not make a big difference for patients with AF whether they were treated with warfarin or NOAC. Similarly, in urgent situations, being on anticoagulation is not a major issue if surgery is needed. The level of coagulation may, however, be assessed. In the case of VKA, vitamin K supplementation can be given. Because of the short half-life of NOACs, it was thought that reversal of the anticoagulation effect was not needed, and trials have confirmed this. Still, reversal agents for NOACs are now available.
When treating patients with NOACs, there is no need to monitor their anticoagulation level. But it is important that patients adhere to treatment. Moreover, it is important keep in mind that CrCl should be measured (or otherwise calculated with the Cockcroft-Gault Calculator) rather than eGFR, because the trials were based on the former.
The GARFIELD-AF cohort study demonstrated that a lot of overtreatment is seen in AF patients with CHA₂DS₂-VASc of 0, while in the higher risk group still a quarter receives aspirin! Thus, Hobbs concluded that we need to focus on overcoming barriers to implement evidence-based practice.