11th September 2014: Perioperative Atrial Fibrillation and the Long-term Risk of Ischemic Stroke

Presented By Dr Justus Groen

Welcome to the first journal club of the 2014 – 2015 academic year.

The paper this week is perhaps an unusual topic for a neurosciences and trauma critical care unit – the epidemiology of stroke in patients who develop atrial fibrillation following elective surgery. However, it is part of an dilemma which confronts us daily: how do we manage the autonomic consequences of a physiologically stressed state. We have visited this broad, complex topic in a previous journal club (rate control in sepsis) and will return to it, not just in journal clubs, but also during teaching on stress cardiomyopathy and the extracranial consequences of acute encephalopathy.

We know that the development of arrhythmia during critical illness is associated with worse outcomes. For example, a large study using similar methodology to this week’s paper examined the association between AF during sepsis (editorial) and a range of long term sequelae. It is a complicated field and confounded by the usual suspects which plague ICM research, but critically ill patients who develop AF have a worse outcome.

In some respects restating the fact that AF leads to embolic stroke is hardly necessary, the epidemiology of AF and stroke has been known for a long time, it was one of the results from the Framingham Study, amongst others. Why, therefore, should it be surprising that AF around the time of cardiac and non-cardiac surgery has a similar association. In fact the smaller studies which preceded this week’s study of nearly 25,000 patints had led to an equivocal position.

The simple conclusion from this paper is that we need to pay close attention to the prevention and treatment of arrhythmias occurring in our peri-operative patients and treat AF aggressively when it occurs. But we do that anyway, don’t we?

Even so, the risk of embolic stroke is probably not something which can be completely addressed in the peri-operative period. The development of post-op AF demonstrates a patient with a myocardium more prone to AF and who therefore needs closer follow-up, it is probably the information we provide to GPs which needs to improve as well as our rapid use of antiarrhythmics.

The less comfortable conclusion is that this study puts comedy glasses, nose and moustache on the elephant already on the unit – we disregard the long term cardiovascular consequences of our management decisions to our patient’s peril.

How can we mitigate the risk. It seems likely, although we do not know for sure, that rapid cardioversion of perioperative AF should improve long term outcome. There are ample resources to help with the management of peri-operative AF, see below, and we will talk about this during journal club.

A further interesting question is whether we should provide perioperative prophylaxis against AF. This should be viewed against the results of the POISE studies which examined β-blockers and clonidine as a means of preventing peri-op MIs, complicated studies which do not give a clear signal yet. Choosing patients at risk of AF will be the first step and the paper this week has some thoughts about this.


A video produced by JAMA to accompany the article provides a useful summary of the article (embedded from this page).



NICE Guidance on the management of AF.
Paul Marik, a journal club regular, has a good review on the management of AF on ICU.
Critical Care Research and Practice review from 2014.