Comparison of CHA2DS2-VASc, C2HEST, HAT2CH2, SYNTAX, GRACE, and SYNTAX II Scores for Predicting New-Onset Atrial Fibrillation Complicating Acute Myocardial Infarction-Uncorrected Proof

OBJECTIVE
This study evaluated the most effective scoring system for predicting new-onset atrial fibrillation (NOAF) during acute myocardial infarction (AMI). Identifying the best predictive tool may help clinicians select the most appropriate personalized treatment based on individual risk scores to prevent NOAF complicating AMI.

METHOD
A total of 2,206 patients diagnosed with AMI between June 2021 and January 2023 were included in this study. After excluding cases with missing data, univariable and multivariable analyses were conducted on 1,672 patients to assess the association between baseline characteristics and the development of atrial fibrillation. The CHA2DS2-VASC (Congestive heart failure, Hypertension, Age ≥ 75 years, Diabetes mellitus, Stroke/TIA/thromboembolism, Vascular disease, Age 65-74 years, Sex category), C2HEST (Coronary artery disease, Chronic obstructive pulmonary disease, Hypertension, Elderly [age ≥ 75], Systolic heart failure, Thyroid disease), HAT2CH2 (Hypertension, Age > 75, Stroke/TIA, Chronic obstructive pulmonary disease, Heart failure), SYNTAX (Synergy Between PCI with Taxus and Cardiac Surgery), GRACE 2.0 (Global Registry of Acute Coronary Events), and SYNTAX II scores were calculated for each patient.

RESULTS
Receiver operating characteristic (ROC) analysis showed that the SYNTAX score (SxS) had the highest predictive value for NOAF during AMI, with an area under the curve (AUC) of 0.785 (95% confidence interval [CI]: 0.767-0.802, P < 0.001), followed by the SYNTAX II score (SxSII) with an AUC of 0.747 (95% CI: 0.728-0.765, P < 0.001), and the GRACE 2.0 risk score (RS) with an AUC of 0.740 (95% CI: 0.721-0.758, P < 0.001). It was shown that the modified scores (created by incorporating hemoglobin A1c [HbA1c] levels), the primary independent predictive parameter in this study, into the existing risk models demonstrated higher predictive value for NOAF (C-statistic: 0.784-0.794).

CONCLUSION
Combining HbA1c levels with SxS yielded the highest diagnostic performance for predicting NOAF during AMI. In this study, while SxS outperformed other risk models, the GRACE 2.0 and SxSII scores also demonstrated relatively strong predictive value and were superior to the CHA2DS2-VASC, C2HEST, and HAT2CH2 scores for predicting NOAF in the setting of AMI.