PLoS ONE (Jan 2016)

Prediction of Thromboembolic Events in Heart Failure Patients in Sinus Rhythm: The Hong Kong Heart Failure Registry.

  • Jo-Jo Hai,
  • Pak-Hei Chan,
  • Yap-Hang Chan,
  • Carol-Ho-Yi Fong,
  • Duo Huang,
  • Wen-Hua Li,
  • Li-Xue Yin,
  • Chu-Pak Lau,
  • Hung-Fat Tse,
  • Chung-Wah Siu

DOI
https://doi.org/10.1371/journal.pone.0169095
Journal volume & issue
Vol. 11, no. 12
p. e0169095

Abstract

Read online

AIM:Heart failure (HF) increases the risk of thromboembolic events (TE). Study in a Caucasian population has shown that the CHA2DS2-VASc score predicts TE among HF patients without atrial fibrillation. We sought to assess the usefulness of the CHA2DS2-VASc score in predicting TE in an Asian population and refine the scoring system to improve its predictability of TE among HF patients in sinus rhythm. METHODS:A total of 1,202 consecutive patients who were admitted to our institution for new-onset HF from 2005 to 2012 and without atrial fibrillation or anticoagulation were retrospectively reviewed. RESULTS:The mean age was 77.6 ± 12.2 years and 51.7% were female. After 36.2 ± 30.1 months, 113 (9.4%) developed TE. The annual incidence was 0.54%, 1.54%, 2.98% and 5.04% per year in those who had a CHA2DS2-VASc score of 1, 2-3, 4-5 and ≥6, respectively. In multivariate analysis, age ≥75 years [Hazard ratio (HR) 2.59, 95% confidence interval (CI) 1.23-5.46, p = 0.012), chronic ischemic heart disease (HR 1.54, 95% CI 1.02-2.31, p = 0.040) and chronic kidney disease (HR 1.66, 95% CI 1.09-2.53, p = 0.018) independently predicted TE. Incorporation of chronic ischemic heart disease and chronic kidney disease into the CHA2DS2-VASc score significantly increased the area under the Receiver Operating Curve from 0.57 (95% CI 0.54-0.59) to 0.61 (95% CI 0.55-0.66; p = 0.022). CONCLUSION:The CHA2DS2-VASc score is useful for stratification of the risk of TE among HF patients in sinus rhythm. Incorporation of chronic ischemic heart disease and chronic kidney disease into the score modestly improves its predictive value.