Postoperative atrial fibrillation is associated with late mortality independently of patient comorbidities: a systematic review and reconstructed individual patient data meta-analysis

Abstract

Abstract Background Early postoperative atrial fibrillation (POAF) is the most common complication within the first days after cardiac surgery. It is hypothesized to result from perioperative stressors such as inflammation, oxidative stress, and increased adrenergic tone, along with an advanced arrhythmogenic substrate resulting from advanced age and cardiovascular comorbidities including hypertension, heart failure, or chronic obstructive pulmonary disease (COPD).[1, 2] Although prior studies have demonstrated an association between in-hospital POAF and late mortality or stroke, it is questionable whether these long-term adverse events result from the occurrence of early-POAF or, rather, from the advanced cardiovascular risk profile that partly underlies the development of early-POAF.[3] Purpose To investigate the association of early-POAF with late mortality (primary outcome) and stroke (secondary outcome) in patients undergoing cardiac surgery after adjustment for age and the cardiovascular risk profile. Methods A systematic search, covering PubMed, EMBASE, and the Cochrane Library, was conducted to identify studies reporting on late mortality after cardiac surgery in patients with and without new-onset POAF during hospitalization. Articles meeting the criteria, which included Kaplan-Meier (KM) curves along with risk tables, were included for a pooled analysis of late mortality and stroke. Individual time-to-event data were reconstructed from the KM curves and incorporated into a multivariable frailty Cox model. The model included adjustments for age, sex, preoperative cardiovascular comorbidity profile (hypertension, diabetes mellitus [DM], peripheral artery disease [PAD], history of stroke, heart failure, and COPD), type of surgery, and a frailty term (study variable). Results In total, 28 studies were included in the analysis for late mortality (N=112931), and 9 studies in the analysis for late stroke (N=35004). Overall, 31039 patients had early-POAF with a pooled incidence of 29.7% (95% CI: 25.7-34.1%). Patients developing POAF were older (68.5 vs. 64.1 years old, p<0.001), had a more frequent history of PAD (9.5% vs. 7.8%, p=0.026), a history of stroke (9.3% vs. 7.6%, p<0.001), heart failure (17.2% vs. 15.1%, p<0.001), and COPD (12.5% vs. 9.7%, p<0.001). Unadjusted analysis showed that POAF was strongly associated with increased late mortality (Hazard Ratio [HR]=1.64, 95% CI: 1.59 to 1.69, p<0.001) and late stroke (HR=1.46, 95% CI: 1.36 to 1.58, p<0.001) (Figure 1). POAF was an independent predictor for late mortality (HR=1.46, 95% CI: 1.36 to 1.58, p<0.001) but not for late stroke (HR=1.10, 95% CI: 0.85 to 1.43, p=0.48) after adjustment for age, cardiovascular comorbidity profile, type of surgery, and frailty (study term). Conclusion Early-POAF after cardiac surgery is an independent predictor of mortality in the years following discharge after cardiac surgery, regardless of age, cardiovascular comorbidity profile, and type of surgery.Figure 1