Article Text

Download PDFPDF

Ensuring adherence to therapy with anticoagulation in patients with atrial fibrillation
  1. Jeroen M Hendriks,
  2. Celine Gallagher,
  3. Prashanthan Sanders
  1. Centre for Heart Rhythm Disorders, University of Adelaide, South Australian Health and Medical Research Institute and Royal Adelaide Hospital, Adelaide, Australia
  1. Correspondence to Dr Jeroen M Hendriks, Centre for Heart Rhythm Disorders, University of Adelaide, Level 5 Mc Ewin Building, Royal Adelaide Hospital, North Terrace, Adelaide 5000, Australia; jeroen.hendriks{at}

Statistics from

Atrial fibrillation (AF) is a significant burden for healthcare systems due to increased morbidity and mortality rates. It is a major cause of ischaemic stroke, which is considered one of the most serious and disabling complications of AF. Stroke prevention with oral anticoagulation (OAC) by vitamin K antagonists (VKA) or non-vitamin K antagonists (NOAC) is therefore a significant component of AF management. Both VKAs and NOACs are effective for the prevention of stroke in AF; however, stringent adherence to the recommended treatment regimen is crucial, both from a prescriber’s and a patient’s perspective. Worldwide, stroke prevention in AF is suboptimal, with poor adherence to international guideline recommendations.1 Poor adherence is a major barrier to effective stroke prevention, reflected in inadequate time in therapeutic range (TTR) with VKA treatment. A proposed advantage of NOAC therapy is the lack of requirement for routine blood monitoring for therapeutic levels. However, this fact, along with a short plasma half-life of approximately 12 hours, means that adherence to therapy becomes even more crucial, and indeed non-persistence or discontinuation of OAC treatment can have devastating consequences.2

In this issue of the Journal, Jackevicius et al 3 published their findings on early non-persistence (within 6 months after initiation) with dabigatran and rivaroxaban in patients with AF. A retrospective cohort study was performed using linked administrative data from hospital admissions in Ontario, Canada, over a time period of 16 years (1998–2014). This cohort comprised patients aged ≥65 years with a diagnosis of AF prior to NOAC prescription. Given that both dabigatran and rivaroxaban were available on formulary from 2012 in Canada, all patients were considered new users of these drugs. Patients were classified into one cohort that initiated dabigatran (15 857) and a second cohort that initiated rivaroxaban (10 119) to compare non-persistence in each cohort. Non-persistence was defined as a gap of 14 days or greater between the specific agent prescription. This was considered a clinically relevant gap, given the correlation between short plasma half-life of these agents and the degree of anticoagulant effect, and consequently a significant point in time in which thromboembolic events may occur. The study demonstrates a high proportion of 6-month non-persistence with a significant difference between dabigatran (36.4%) and rivaroxaban (31.9%; p<0.001). After a mean follow-up of 533 days, the primary composite endpoint (stroke/transient ischemic attack (TIA) or death) occurred in 1740 (11.6%) in the dabigatran cohort vs 817 (8.6%) in the rivaroxaban cohort after a mean follow-up of 273 days, suggesting that non-persistence, independent of cohort, significantly increased the risk of the composite endpoint. Overall, discontinuation rates of either dabigatran or rivaroxaban were relatively high in this cohort, which is similar to previously reported ‘real-world’ data,4 although reasons for this are not discussed in the current study. Although there may be valid reasons for treatment cessation, it has also been suggested that this could be due to lack of knowledge by either the patient or the provider, or fear of potential complications of OAC therapy.5

This study provides important insights in the magnitude of non-persistence with NOAC therapy and the impact on adverse clinical events in a real-world setting compared with those presented by the large NOAC landmark trials (eg, RE-LY (Randomized Evaluation of Long Term Anticoagulant Therapy (RE-LY) With Dabigatran Etexilate) and ROCKET-AF (An Efficacy and Safety Study of Rivaroxaban With Warfarin for the Prevention of Stroke and Non-Central Nervous System Systemic Embolism in Patients With Non-Valvular Atrial Fibrillation)). However, it also raises a number of issues worthy of further exploration. First, higher levels of non-persistence were demonstrated in clinical practice compared with the landmark trials. A clinical trial setting includes strict follow-up and monitoring of endpoints. With the translation of evidence into practice, it is striking that follow-up and evaluation of patients on NOAC therapy may be less frequent and intensive in daily practice. Given that frequent monitoring is considered unnecessary in these agents, it is imperative to ensure safety and effectiveness prior to prescribing these agents to an individual patient. Prescribers should be aware of the risk of non-persistence in all patients, and especially in those without any continuous follow-up.

Second, although the superfluity of regular follow-up in NOAC treatment may be more appealing to individuals, the paradigm ‘prescribe and forget’ may be a potential risk and should be avoided. Raising awareness of such a scenario is important among healthcare professionals as well as among patients. The lack of requirement for routine blood monitoring in NOAC therapy may also imply loss of patient–physician contact time, which can be an invaluable source of information and education. A comprehensive integrated approach is needed to incorporate stroke prevention into daily practice, including structured follow-up in those patients on NOAC treatment to evaluate and prevent non-persistence and adverse effects of therapy.

The 2016 European Society of Cardiology guidelines on the management of AF recommends such integrated approach to AF care. This approach advocates for active patient involvement in treatment decisions and tailored patient education throughout the care process,6 which may improve the patient’s understanding of stroke prevention and encourage empowerment to self-management OAC treatment. Moreover, patients on OAC treatment may comprise an older and frail population, with high CHA2DS2-VASc scores (C - Congestive Heart Failure, H - Hypertension, A - Age 75 years or older, D - Diabetes Mellitus, S - Previous Stroke, TIA, or thromboembolism, V - Vascular disease, A - Age between 65-74, S - Sex category (female gender)). This requires a team approach and awareness of all healthcare professionals involved to focus on facilitating treatment continuation and monitoring for side effects. Ideally, patients on NOAC treatment should be seen on a regular basis to evaluate their treatment. Given the critical time to non-persistence is 4–8 months post-initiation,3 this would be paramount in the first year after initiation. Follow-up (eg, monitoring renal and liver function) and coordination of care may be undertaken by OAC specialists or experienced primary care practitioners, and nurse-coordinated clinics may be also considered in this regard (figure 1).7

Figure 1

Suggested initiation and structured follow-up pathway of patients on NOACs. Reproduced with permission from Heidbuchel et al 7 and Oxford University Press.

Third, the study demonstrates that the high rate of non-persistence was associated with a significantly increased risk of stroke, TIA or death by 80%. Conversely, in a large cohort of AF patients (n=64 661), it was demonstrated that improved adherence to stroke prevention was associated with a lower risk of stroke.8 These findings confirm the importance of evidence-based guideline adherence as well as persistence to OAC therapy, and advocate for regular evaluation of NOAC therapy. From a cost perspective one could argue that such frequent monitoring and follow-up is likely to increase the health burden. However, given the potentially irreversible and disabling character of stroke and consequent hospitalisation for clinical observation and/or rehabilitation, one can understand the valuable contribution of an integrated approach to OAC care. Specialised clinics aim to invest in the patient from the initial visit and have potential to be cost-effective in the long-term.9 10 Finally, the high mortality rate in those non-persistent to NOAC therapy is striking and worthy of further consideration. Although the reasons for this are not able to be elucidated from the presented data set, this calls for a comprehensive and integrated team approach to the management of AF patients to improve outcomes in this growing population.

In conclusion, non-persistence rates to NOAC treatment are high in this ‘real-world’ population. Although patients on NOAC treatment may not require routine blood monitoring to the same degree as those on VKA antagonists, it is clear that this should not be translated to a lack of requirement for monitoring of the individual. This study by Jackevicius et al emphasises that non-persistence of individuals on NOAC therapy can have potentially devastating consequences with increased risks for stroke/TIA or death. Comprehensive, integrated care in stroke prevention in AF with a multidisciplinary, patient-centred approach may contribute to improving persistence in OAC therapy and reduction of adverse events.



  • Contributors JMH, CG and PS were all involved in drafting and critically revising the editorial.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles