Atrial fibrillation after surgical aortic valve replacement: the dark side of safe procedure

Atrial fibrillation (AF) is the most common arrhythmia that occurs after cardiac surgery. Postoperative atrial fibrillation (POAF) has an incidence of 20–40%, depending on the type of procedure, with higher rates after valvular surgery (30–50%) [1–3]. Postoperative AF is more common in older patients, patients with concomitant valvular heart disease, hypertension, heart dysfunction and renal failure. The cause of postoperative AF is multifactorial and related to both the underlying atrial substrate and the acute inflammation, oxidative stress and sympathetic output associated with cardiac surgery [4]. Current European Society of Cardiology and European Association for Cardio-thoracic Surgery joint guidelines state that oral anticoagulation (OAC) therapy may be considered in patients with POAF (class IIb, lever of evidence B), after taking stroke risk and bleeding risk into consideration [5]. This recommendation is based on a few retrospective studies involving cardiac surgery patients, while the remaining evidence come from non-surgical AF patients. No specific recommendations for OAC in surgical aortic valve replacement (SAVR) patients with POAF have been published. In this issue of the Journal, we read a retrospective, population-based study including all isolated SAVR and combined SAVR and Coronary Artery Bypass Graft (CABG) patients with new onset of postoperative AF. Rezk et al. [6] explore associations between POAF, early-initiated OAC and long-term outcomes after SAVR and combined SAVR + CABG. They performed a retrospective, population-based study including all isolated SAVR and combined SAVR and CABG patients without a history of preoperative AF in Sweden 2007–2017. The study unrolled 10 892 patients, 7038 undergoing SAVR and 3854 combined SAVR and CABG. 44.5% of patients undergoing SAVR and 50.7% of patients undergoing SAVR and CABG manifest POAF. The patients who present POAF have some common characteristics like advanced age, renal failure, Hypertension, elevated CHA2DS2VASc and heart failure. During the follow-up, a total of 15.0% of the isolated SAVR and 12.9% SAVR + CABG patients died. POAF after SAVR was associated with a higher long-term risk of all-cause mortality, ischaemic stroke, thromboembolic event, heart failure hospitalization, recurrent AF and major bleeding. POAF after SAVR + CABG was associated with a higher risk of all-cause mortality, heart failure hospitalization and recurrent AF, but not with increased long-term risk of ischaemic stroke, thromboembolism or major bleeding. 67% of SAVR and 65.9% of SAVR + CABG received OAC treatment within 30 days after discharge. After 1 year, 30.5% of isolated SAVR and 24.9% of SAVR + CABG patients were dispensed OAC. In both groups, after multivariable adjustment, early-initiated OAC was not associated with reduced long-term risk for all-cause mortality, ischaemic stroke, thromboembolism or higher risk for major bleeding. The main findings of this large retrospective population-based cohort study were that POAF after isolated SAVR was associated with increased long-term risk for all-cause mortality, ischaemic stroke, thromboembolic events, heart failure hospitalization, recurrent AF and major bleeding, while POAF after SAVR + CABG was associated with death, heart failure and recurrent AF. Although these results agree with many studies, they are not univocal in the literature [7–9]. Most patients undergoing isolated SAVR were treated with anticoagulation after discharge. Most patients undergoing SAVR + CABG were treated with anticoagulation and antiplatelet therapy and this may explain how patients undergoing SAVR + CABG in antiplatelet therapy are protected from ischaemic stroke. The administration of oral anticoagulant therapy did not lead to major complications in both groups. This study, as already stated by the authors, has some important limitations. The study design exposes to risks of selection bias and residual confounding. The multicentre study does not allow standardizing the type of valve whether porcine or bovine, the model and the sizing. Another limitation of this multicentre study depends on the lack of standardized protocols in the management of anticoagulant therapy in patients undergoing SAVR with bioprostheses. The strengths of this study include the large nationwide study population, the real-life setting and the use of validated registries. We believe that the key to reducing the effects of POAF is monitoring. Indeed, we recommend monitoring patients undergoing cardiac surgery with multiple risk factors for developing AF such as advanced age, elevated CHA2DS2VASc, hypertension, valvular disease, chronic heart failure and renal dysfunction. In these patients, we perform a Holter monitor before the discharge between the fifth and seventh postoperative days. In a patient with POAF, we recommend performing a pharmacological or electrical cardioversion and in case of persistent AF we start an OAC. In case of POAF, we start an OAC and we perform a new Holter monitor control in 1–3 months. Strategies for rate control and rhythm control to treat postoperative AF were associated with equal numbers of days of hospitalization, similar complication rates and similarly low rates of persistent AF 60 days after onset. Neither treatment strategy showed a unique clinical advantage over the other [10]. In conclusion, we should not consider POAF as a discomfort but acknowledge it for what it really is: a severe pathological change that increased risk of death and stroke, increased hospital length of stay and increased healthcare costs.