Refining age stratum of the CHEST score for predicting incident atrial fibrillation in a hospital-based Chinese population.

BACKGROUND

The CHEST score (C: coronary artery disease [CAD] / chronic obstructive pulmonary disease [COPD] (1 point each); H: Hypertension; E: Elderly (Age≥75, doubled); S: Systolic heart failure (doubled); T: Thyroid disease (hyperthyroidism)) has been validated to predict incident atrial fibrillation (AF). Its performance in the hospital-based Chinese population has never been evaluated.

METHODS

Risk factors for incident AF were investigated in a hospital-based population. Comparison of the CHEST score and other clinical scores with the capacity of predicting incident AF was conducted using area under the curves (AUC), net reclassification index (NRI), integrated discriminative improvement (IDI), and decision curve analysis (DCA). An age-stratified criterion was used to refine the CHEST score to form a modified CHEST score (mCHEST). The performance of the mCHEST score was also evaluated.

RESULTS

A total of 23,523 patients entered the study with 520 developed AF during 2.84 ± 3.56 years of follow-up. Risk factors for incident AF included age, male sex, hypertension, CAD, COPD, previous ischemic stroke, hyperthyroidism, and heart failure. Age ≥65 years has significantly increased the risk of AF, which was considered as the age cutoff for a modified CHEST score (mCHEST). The risk of AF increased by 89% per one-point increase of the mCHEST score. The mCHEST score showed better predictive performance (AUC of 0.809) compared with the original CHEST (AUC of 0.752), CHADS-VASc (0.756), HATCH (0.722), and HAVOC (0.758) scores, also as estimated by IDI, NRI and DCA. Among those enrolled after 2012, the mCHEST score had numerically higher AUC (0.849) compared with the CHEST score (0.826) and the other scores.

CONCLUSION

In a hospital-based Chinese population, by refining the age strata of the original CHEST score, the mCHEST score had significantly increased predictive accuracy and discriminative capability for incident AF. The clinical benefits of the application of novel mCHEST score needs further validation in multiple settings.