# C-Reactive Protein-to-Albumin Ratio (CAR) and Left Atrial Diameter Predicts New-Onset Atrial Fibrillation in Chronic Coronary Syndrome: A Retrospective Cohort Study

## TL;DR

A study finds that combining a blood marker of inflammation and heart size can predict new-onset atrial fibrillation in heart disease patients.

## Contribution

The study introduces a novel dual-marker model combining CAR and LAD for predicting NOAF in CCS patients.

## Key findings

- Patients with high CAR and LAD had 2.67 times higher risk of NOAF compared to those with low CAR and LAD.
- The combined CAR–LAD model showed superior predictive accuracy (AUC = 0.731) for NOAF.
- The model provided consistent results across most subgroups and greater net clinical benefit.

## Abstract

Background/Objectives: New-onset atrial fibrillation (NOAF) frequently develops in patients with chronic coronary syndrome (CCS) and is associated with adverse cardiovascular outcomes. The C-reactive protein–to–albumin ratio (CAR) reflects systemic inflammation, whereas left atrial diameter (LAD) indicates structural cardiac remodeling. Their combined predictive role for NOAF in CCS remains uncertain. This study evaluated the predictive value of combined CAR and LAD for NOAF in CCS patients. Methods: We retrospectively analyzed 2431 CCS patients treated at the Third Affiliated Hospital of Sun Yat-sen University between 2012 and 2019. The primary endpoint was NOAF occurrence during follow-up. Receiver operating characteristic (ROC) analysis determined exploratory cutoff values for CAR (0.0429) and LAD (33.96 mm). Patients were categorized into four groups: Group 1 (low CAR–low LAD), Group 2 (high CAR–low LAD), Group 3 (low CAR–high LAD), and Group 4 (high CAR–high LAD). Cox proportional hazards, Kaplan-Meier, and subgroup analyses were conducted to evaluate associations with NOAF risk. Results: During a median follow-up of 4.96 years, 93 NOAF events were identified. Compared with the Group 1, patients with higher CAR and LAD showed significantly elevated NOAF risk (HR = 2.67, 95%CI 1.99–3.57, p < 0.001). The combined CAR–LAD model demonstrated superior predictive accuracy (AUC = 0.731, 95% CI = 0.654–0.765; p < 0.001) and consistent effects across most subgroups. Decision curve analysis confirmed greater net clinical benefit for the combined model. Conclusions: The integration of CAR and LAD serves as a simple, non-invasive, and effective tool for predicting NOAF in CCS patients. This dual-marker model facilitates early identification of high-risk individuals and support personalized preventive strategies in clinical practice.