Page 81 - Read Online

Page 6 of 9 Rademacher et al. Vessel Plus 2022;6:39 https://dx.doi.org/10.20517/2574-1209.2021.138

[27]
Aasbo et al. , also pooling estimates from eight trials (some being the same as in the Gillespie meta-
analysis), reporting an OR 0.39 (0.21-0.76).

DISCUSSION
We present a systematic review of the literature on POAF and associated stroke risk, as well as on
anticoagulation for stroke prevention in patients with POAF after cardiac surgery. Eight meta-analyses of
the data on POAF and stroke were published in the last few years [Tables 1 and 3]. All of them report an
increased risk of stroke to be associated with POAF. The reported pooled odds ratios range from 1.36 to
4.09 for unadjusted estimates. This is consistent with the higher short- and long-term mortality that has also
been reported in the same meta-analyses and the underlying studies, respectively [15-18] . Importantly, this is
strong evidence that POAF and stroke are correlated. However, it is not proof that POAF is the cause of the
higher stroke incidence.

Atrial fibrillation and stroke share many risk factors, including age, smoking, hypertension, and prior
stroke, among others [11,30] . Furthermore, the most studied tool to estimate the risk of stroke in patients with
atrial fibrillation, the CHA DS -VASc score, has been found to predict the onset of POAF. A meta-analysis
2
2
summarizing the data of 12 studies with over 18,000 patients suggests CHA DS -VASc is an independent
2
2
predictor of POAF after cardiac surgery [OR 1.46 (1.25-1.72)] . The score consists of points for
[31]
hypertension, stroke, diabetes, and peripheral vascular disease, among others. In this context, it is evident
that even a strong association between POAF and stroke is not sufficient to establish causality. In fact, it
may be considered to support the notion that POAF is a risk marker, rather than a risk factor that warrants
clinical attention and possibly further investigation during follow-up . Butt et al. reported recently from
[32]
[33]
a large Danish cohort study that patients with CABG with and without POAF had similar long-term
thromboembolic risk. They compared long-term outcomes between patients with POAF and matched
nonsurgical patients with incident nonvalvular atrial fibrillation. Interestingly, patients with POAF after
CABG had markedly better long-term outcomes than nonvalvular atrial fibrillation patients in terms of
lower thromboembolic risk, lower mortality risk, and lower risk of recurrent hospitalization for AF. Other
investigators have also shown that the overall rate of stroke is not statistically different between post-cardiac
surgery patients with and without atrial fibrillation [34,35] .

Only two meta-analyses published to date reported pooled odds ratios from adjusted estimates of the
association of POAF and stroke (OR 1.25; 1.88) [12,14] . The level of adjustment is rather variable between the
studies, and the meta-analyses could only include a few studies providing adjusted estimates, as most studies
did not provide such information. Residual confounding remains a significant concern in the included
original studies.

Historically, POAF has been thought to be associated with an increased risk of thromboembolic events and
a negative impact on morbidity. To reduce the risk of stroke in the setting of POAF, some studies have
suggested using anticoagulants [25,36,37] . This is also reflected in guidelines proposed for the management of
POAF by the American College of Chest Physicians, the Canadian Cardiovascular Society, and the
European Association for Cardio-Thoracic Surgery [38-40] . Matos et al. found that post-CABG
[26]
anticoagulation for new atrial fibrillation in the STS Database was associated with increased bleeding events,
but no difference in stroke at 30 days postoperatively. In a recent review by Greenberg et al. , they
[41]
conclude that the benefits of anticoagulation in POAF may lose to the risk of bleeding in patients with
certain risk factors, including advanced age, uncontrolled hypertension, and history of bleeding. Similarly, a
report from the SWEDEHEART study showed harm and no benefit with anticoagulation in patients with
POAF . Two out of three meta-analysis pooling estimates from RCT on the use of amiodarone and beta-
[19]