Intended for healthcare professionals

CCBYNC Open access
Research

Risk of stroke and transient ischaemic attack in patients with a diagnosis of resolved atrial fibrillation: retrospective cohort studies

BMJ 2018; 361 doi: https://doi.org/10.1136/bmj.k1717 (Published 09 May 2018) Cite this as: BMJ 2018;361:k1717

Opinion: The importance of asking the right research question

Re: Risk of stroke and transient ischaemic attack in patients with a diagnosis of resolved atrial fibrillation: retrospective cohort studies

I would like to comment on the article's conclusion that all patients with resolved atrial fibrillation (AF) should continue oral anticoagulation.

A clear definition of resolved AF is not established. In this study's sample, 22.8% of patients with resolved AF had recurrence (without any guarantee that all recurances were detected). Arguably, those patients never had resolved AF. Assuming a possibility of identifying "true" resolved AF the possibility that those had a lower or "near normal" risk of cardiogenic cerebral ischemia is still open. The usage of resolved AF in the study's population seems to reflect to a relevant degree British physician's creativity in meeting the quality standards of the the Quality and Outcomes Framework.

The reported incidence rate of stroke or TIA in those with resolved AF is 1.2 per 100 person-years (pyrs). This estimate includes haemorrhagic stroke (why?), a guess at the ischemic stroke & TIA rate would thus be 1 per 100 pyrs. Benefit of oral anticoagulation has been modelled to begin at a rate of 0.9 for DOCAs and 1.7 per 100 pyrs for VKAs [1]. But those estimates have a relevant degree of uncertainty and individual risk estimation (e.g. with the CHADS-VASc score) at these low-risk ranges is related to a high degree of uncertainty as well.

The study demonstrates that, whatever exacltly "resolved AF" might be (pathophysiologically), it is associated with a lower, but stil relevant risk of ischemic stroke. For advancement, we have to figure out more closely what it is, as it likely could contribute to better ischemic stroke risk stratification. Introduction of an undefined concept into guidelines is premature.

[1] European Heart Journal - Cardiovascular Pharmacotherapy (2017) 3, 37–41

Competing interests: No competing interests

19 May 2018
Felix S Wicke
Physician
Institute of General Practice Johann Wolfgang Goethe University Frankfurt/Main, Germany