Thread by @critconcepts

Allan Walkey on atrial fibrillation in ICU (#SCCM2023)

Arrigo 2015: cardioversion for secondary A-fib in ICU rarely works (43% remains sinus at 1hr, 23% at 24hr). Diltiazem was more effective at rate control than amio or digoxin.

Walkey 2016: Afib in sepsis? Association on outcome of BB vs CCB, BB vs digoxin, BB vs amiodarone compared. BB had slightly lower mortality than CCB.

Bosch 2021: BB achieved more HR reduction than amiodarone, CCB, and dioxin at 1hr (CCB was highest at 6 hours).

Davey 2005: magnesium was about as effective at rhythm control as amiodarone.

Remember untreated infection, volume status, presence of beta agonists.

Law 2022: phenylephrine vs norepinephrine for shock? HR difference was only about 2-6 only, even in the subgroup with RVR.

If you really think Afib is causing adverse effects, consider whether it's due to the rate (give BB or CCB), or due to loss of atrial kick (mag and/or amio).

44% have A-fib recur in 1 year after ICU afib, but anticoagulation acutely during sepsis is not associated with less stroke risk, only more bleeding (Walkey Chest 2014, Walkey JAMA Cardiology 2016). So just give followup.

(More specifically, Afib during sepsis entails a higher stroke risk long-term than sepsis without Afib, but not as high as A-fib occurring in other contexts.)

Bosch: new Afib in sepsis should be viewed as a (cardiac) organ failure, similar to AKI, respiratory failure, etc. It's a sign something is wrong and the cause needs to be addressed.

Conclusions: Rate control is usually first line; rhythm control as second line or in emergencies, but does entail stroke risk and may require anticoagulation. AC for rate control can usually be deferred.