AIM  To evaluate the ability of C₂HEST score to predict new-onset atrial fibrillation in patients aged over 75.

  METHODS  A retrospective cohort study was conducted on the clinical data of non atrial fibrillation patients aged>75 years who were hospitalized in the First and Second Medical Centers of the General Hospital of the People’s Liberation Army from January 1, 2008 to December 31, 2009. A total of 2560 patients aged (86 ± 4) years were included. The Kaplan-Meier survival curve was used to analyze the impact of different scores on disease-free survival time of patients, and the subject work characteristic curve was used to evaluate the predictive value of C₂HEST score in new onset atrial fibrillation.

  RESULTS  During the average (64 ± 43) month study period, 299 patients developed new episodes of atrial fibrillation. Compared with the non atrial fibrillation group, there were more men in the newly diagnosed atrial fibrillation group (P<0.01), a higher proportion of diabetes (P<0.05), and a higher proportion of coronary atherosclerotic heart disease, hypertension, chronic obstructive pulmonary disease, heart failure, hyperthyroidism, stroke or transient ischemic attack, and vascular disease (all P<0.01). At baseline, the C₂HEST score, HATCH score, CHADS₂, and CHA₂DS₂-VASc scores in the AF group were higher (all P<0.01). With the increase of C₂HEST score, the incidence rate of atrial fibrillation gradually increased, and the risk ratio gradually increased. The C₂HEST score (AUC=0.677, 95% CI: 0.643～0.711) was significantly superior to the HATCH score, CHADS₂ score, and CHA₂DS₂-VASc score in predicting the efficacy of new onset atrial fibrillation.

  CONCLUSION  C₂HEST score has good predictive value for new-onset atrial fibrillation in hospitalized patients over 75 years old. Given the high incidence of patients with high score, a more intensive screening to detect atrial fibrillation in this population may be considered.