The times they are a changin'1

This editorial refers to ‘Safety and efficacy of dual vs. triple antithrombotic therapy in patients with atrial fibrillation following percutaneous coronary intervention: a systematic review and meta-analysis of randomized clinical trials’†, by H.B. Golwala et al., on page 1726. Antithrombotic strategies in patients with atrial fibrillation (AF) undergoing percutaneous coronary intervention (PCI) are evolving. Oral anticoagulation (OAC) has been shown to be superior to dual antiplatelet therapy (DAPT) in preventing stroke in patients with AF, while DAPT has been shown to be superior to OAC in preventing stent thrombosis and ischaemic events in patients undergoing coronary stenting. In patients presenting both conditions, covering both risks by cumulating both treatments, the so-called triple therapy (TT), has been the rule for many years, supported by the clinical practice guidelines. However, studies, whether randomized or not, have consistently shown that TT dramatically increases the bleeding risk compared with a treatment combining OAC with only one antiplatelet agent, the so-called double therapy (DT).1,2 To better understand the risk of treatments, we need to understand whether the risks of stroke, stent thrombosis, and bleeding are at the same level at the same time. The risk of stent thrombosis is well documented and high during the first month after stenting. The risk decreases exponentially and becomes low 15 days after stent implantation. The risk of stroke in a chronic AF patient is steady and depends on the patient’s characteristics (graded by the CHA2DS2-VASc score) and there is no reason to believe that this risk may increase with coronary stenting. The risk of bleeding increases with the number of antithrombotic treatments and is high during the first 3 months, particularly the first month after treatment initiation. Considering this chronology of exposure to the three risks, it would make sense to cover the risk of stent thrombosis with adequate DAPT during the first month, if needed, with ticagrelor or prasugrel, withholding OAC during this period. At 1 month, OAC would be (re-)introduced to cover the risk of stroke when one antiplatelet agent is removed. This strategy using only two antithrombotics would limit the risk of bleeding (compared with TT) and prevent the ischaemic events according to the most likely time of occurrence. It must also be noted that DAPT, even if not optimal, is certainly not neutral in prevention of stroke in AF patients during the first month, just like OAC is not neutral in prevention of ischaemic events in a stabilized coronary patient. Unfortunately, this strategy of double treatment with a switch from one antiplatelet to OAC at 1 month has never been tested in a clinical trial and thus cannot be recommended. The question then becomes: how short can the duration of TT be in order to reduce the risk of bleeding with sufficient coverage of stent thrombosis risk? ISAR-TRIPLE suggested that 6 weeks was long enough (compared with 6 months) while another three studies (WOEST, PIONEER, and REDUAL) limited the duration of TT to the hospital phase in their study arm.3–6 Interestingly, the control arms of the three studies had different durations of TT using warfarin: a majority of patients received TT up to 1 year in WOEST, up to 6 months in PIONEER, and up to 3 months in REDUAL. We understand that the comparison of TT with DT was long in WOEST, intermediate in PIONEER, and the shortest in REDUAL. Nevertheless, all studies showed a reduction of bleeding complications with DT, including the study which had the shortest duration of TT in the control arm (REDUAL) and a low dose of non-vitamin K OAC (NOAC; dabigatran 110 mg b.i.d.) in the study arm. No single trial was adequately powered to rule out an increase of ischaemic events with DT vs. TT. The remaining question is about the price to pay, if any, in terms of ischaemic outcome when using DT with an OAC rather than TT, and if this price is worth the known excess bleeding with TT. In this issue of the journal, Golwala and colleagues performed a meta-analysis of the four pre-cited studies comparing DT (single antiplatelet agent plus OAC) and TTT (dual antiplatelet therapy plus OAC) in patients with AF who underwent PCI.7 Although we now have more meta-analyses published on this topic than we have individual studies, this work is important for several reasons. First, it was conducted by the knowledgeable investigators of the latest published trial who fully apprehend the pending issues of their own study. Secondly, with >5000 patients, there is a substantial gain of power to examine ischaemic outcomes, although we would need a 10-fold larger sample to ensure a reliable analysis on each ischaemic endpoint or each important subgroup. Thirdly, while waiting for other ongoing randomized studies, these results are useful to the practitioners who need to treat high bleeding risk patients and cannot run the risk of TT. The absence of a deleterious signal, statistically or numerically, on stent thrombosis, myocardial infarction, stroke, or mortality is an important finding. The 2014 ACC/AHA/HRS guidelines for AF note that OAC is primarily indicated for patients with a CHA2DS2-VASc ≥2 and there is no differentiation in the recommendation between warfarin and NOACs, but they recommend that a bare metal stent should be used to minimize the duration of DAPT therapy, although it is a weak recommendation.1 In a 2016 update to the PCI Guidelines, the recommendation for treating patients with AF who have undergone PCI and are at high bleeding risk is that TT should remain in place for 1 month.8 After 1 month, one of the antiplatelet agents should be discontinued. The 2016 ESC Guidelines for the management of AF state that patients with AF undergoing PCI for stable coronary artery disease with low risk of bleeding should be treated with TT for 1 month, DT (OAC plus aspirin or clopidogrel) for 11 months, and thereafter only OAC. If these patients are at a high risk of bleeding, the algorithm changes to DT for 5 months.2 For AF patients with acute coronary syndrome (ACS) who are at a low risk of bleeding, TT should be used for the first 6 months and then DT for the next 6 months. For ACS patients at high risk of bleeding, TT should be given for 1 month and then DT for 11 months. After 12 months, both high and low bleeding risk patients are recommended to be treated with OAC as monotherapy. In the 2017 ESC Guidelines on DAPT, DT from hospital discharge is presented as an alternative to TT and only in the high bleeding risk patient (see Figure 1).9 Top: 2017 ESC recommendations on DAPT.9 Bottom: proposal for a simplified version of these recommendations in 2018. Top: 2017 ESC recommendations on DAPT.9 Bottom: proposal for a simplified version of these recommendations in 2018. These recommendations are complex and pose new questions. First, they lag behind the evidence, which now comes from four positive independent studies and several consistent meta-analyses including the one published in this issue, showing the superiority of DT over TT from the point of view of the net clinical benefit, which is the patient’s point of view. This strategy of DT has not been endorsed yet by the guidelines as the preferred strategy. The guidelines are not rapidly updated and 4 years on average may separate two sets of guidelines. In essence, the guidelines are conservative, adding to the gap between guidelines and evidence. Because nowadays evidence is rapidly presented, promoted, discussed, and evaluated by our community, the opinions are often made before the next revision of guidelines occur. AF-stented patients are a good example where changes in practice precede changes in guidelines. Quitting aspirin at hospital discharge in these patients is not the first option in the guidelines but is becoming the preferred option for many practitioners. The reason is that the decision is taken directly by the physicians concerned, in hospital, physicians who are well aware of the most recent literature, the interventional cardiologists in particular, who recommend the antithrombotic treatment as soon as the patient leaves the catheterization laboratory with a freshly implanted stent. Times are changing, and changes in practice arise from our educated community, possibly tomorrow also under the pressure of the patients themselves. The old model of printed guidelines, slowly adjusted every 4 years by a group of knowledgeable experts, may now have outlived their usefulness. The model is less adapted to the rapid evolution of science in many areas, the social network communication of these changes, and the adoption by our well-informed colleagues, whose judgement is probably as valuable as expert opinions provided behind closed doors every 4 years. More studies will follow these four studies. Since the four studies were underpowered to examine the impact of DT on ischaemic events, data from the ongoing AUGUSTUS and ENTRUST AF-PCI are expected to contribute to the totality of the evidence to evaluate the efficacy of such strategies. Nevertheless, evidence has now reached the point to induce a change in practice with solutions that limit serious bleeding events without excess ischaemic risk. The unanswered questions now seem more important than this question of DT vs. TT. Which antiplatelet agent shall we stop early: aspirin or a P2Y12 antagonist? What is the place of ticagrelor or prasugrel in high thrombotic risk patients undergoing stenting when they have AF? Can we safely remove the last antiplatelet agent 1 year after stenting in this AF population on OAC? Should we preferably use NOACs with an available antidote? Which NOAC shall we use preferably and at what dose? Can we apply the same rules to valvular or TAVI patients? We need more studies, but now on issues other than TT. Conflict of interest: N.H. reports research grants to the Institution or consulting/lecture fees from Philips, Bayer, Laboratoires Servier, Novartis Pharma, Astra Zeneca, Bristol Myers Squibb, MSD, Fédération Française de Cardiologie, and ICAN. G.M. reports research grants to the Institution or consulting/lecture fees from Abbott, Amgen, Actelion, AstraZeneca, Bayer, Boehringer Ingelheim, Boston-Scientific, Bristol-Myers Squibb, Beth Israel Deaconess Medical, Brigham Women’s Hospital, Cardiovascular Research Foundation, Daiichi-Sankyo, Idorsia, Lilly, Europa, Elsevier, Fédération Française de Cardiologie, ICAN, Medtronic, Journal of the American College of Cardiology, Lead-Up, Menarini, MSD, Novo-Nordisk, Pfizer, Sanofi, Servier, The Mount Sinai School, TIMI Study Group, and WebMD.