Predicting atrial fibrillation after cardiac surgery: a scoping review of associated factors and systematic review of existing prediction models

Clinical Predictors of Device-Detected Atrial Fibrillation During 2.5 Years After Cardiac Surgery: Prospective RACE V Cohort

BACKGROUND

Postoperative atrial fibrillation (POAF) is a frequent complication after cardiac surgery that is associated with late atrial fibrillation (AF) recurrences (late-POAF) and increased morbidity and long-term mortality.

OBJECTIVES

This study sought to determine device-detected POAF incidence and to identify clinical variables associated with POAF, both in patients with and without preoperative AF history.

METHODS

A total of 133 consecutive patients undergoing cardiac surgery were prospectively enrolled and continuously monitored with an implantable loop recorder for 2.5 years after surgery. Preoperative transthoracic echocardiography, 12-lead electrocardiogram, blood biomarkers, and clinical data were analyzed to develop prediction models for early- and late-POAF.

RESULTS

In patients without preoperative AF history, early-POAF within the first 90 postoperative days occurred in 41 (47.1%) of 87 patients. Late-POAF after the first 90 postoperative days occurred in 22 (25%) of 87 patients, and 20 of these patients also had early-POAF during the first 90 days (20 of 22 [91%]). Increased right atrial minimum volume indexed for body surface area (RAVImin) and early-POAF were independently associated with late-POAF. A prediction model for late-POAF, which included RAVImin >11 mL/m2, age >65 years, and early-POAF, achieved an area under the curve of 0.82 (95% CI: 0.72-0.92). For patients with preoperative AF-history, late-POAF recurrences were frequent (22 of 33 [67%]). Increased RAVImin was independently associated with a higher incidence of late-POAF.

CONCLUSIONS

In patients with and without AF history, late-POAF recurrences are frequent, including in patients undergoing surgical AF ablation. In patients with no history of AF, late-POAF might be predicted with excellent accuracy by using a combination of preoperative variables. In patients with a history of AF, signs of advanced AF substrate (eg, increased right atrial volumes) were associated with long-term AF recurrences. [Reappraisal of Atrial Fibrillation: Interaction Between Hypercoagulability, Electrical Remodeling, and Vascular Destabilisation in the Progression of AF; NCT03124576].

Research progress on predicting atrial fibrillation recurrence after radiofrequency ablation based on electrocardiogram-related parameters

Atrial fibrillation (AF) is the most common arrhythmia. It is associated with increased stroke risks, thromboembolism, and other complications, which are great life and economic burdens for patients. In recent years, with the maturity of percutaneous catheter radiofrequency ablation (RFA) technology, it has become a first-line therapy for AF. However, some patients still experience AF recurrence (AFR) after RFA, which can cause serious consequences. Therefore, it is critical to identify appropriate parameters that are predictive of prognosis and to be able to translate the parameters easily into the clinical setting. Here, we reviewed possible predicting indicators for AFR, focusing on all the electrocardiogram indicators, such as P wave duration, PR interval and so on. It may provide valuable information for guiding clinical works.

Multivariable prediction models for atrial fibrillation after cardiac surgery: a systematic review protocol

INTRODUCTION

Dozens of multivariable prediction models for atrial fibrillation after cardiac surgery (AFACS) have been published, but none have been incorporated into regular clinical practice. One of the reasons for this lack of adoption is poor model performance due to methodological weaknesses in model development. In addition, there has been little external validation of these existing models to evaluate their reproducibility and transportability. The aim of this systematic review is to critically appraise the methodology and risk of bias of papers presenting the development and/or validation of models for AFACS.

METHODS

We will identify studies that present the development and/or validation of a multivariable prediction model for AFACS through searches of PubMed, Embase and Web of Science from inception to 31 December 2021. Pairs of reviewers will independently extract model performance measures, assess methodological quality and assess risk of bias of included studies using extraction forms adapted from a combination of the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies checklist and the Prediction Model Risk of Bias Assessment Tool. Extracted information will be reported by narrative synthesis and descriptive statistics.

ETHICS AND DISSEMINATION

This systemic review will only include published aggregate data, so no protected health information will be used. Study findings will be disseminated through peer-reviewed publications and scientific conference presentations. Further, this review will identify weaknesses in past AFACS prediction model development and validation methodology so that subsequent studies can improve upon prior practices and produce a clinically useful risk estimation tool.

PROSPERO REGISTRATION NUMBER

CRD42019127329.