Comparative assessment of published atrial fibrillation stroke risk stratification schemes for predicting stroke, in a non-atrial fibrillation population: The Chin-Shan Community Cohort Study Gregory Y.H. Lip a, ⁎, Hung-Ju Lin b , Kuo-Liong Chien a, b, c, ⁎⁎, Hsiu-Ching Hsu c , Ta-Chen Su c , Ming-Fong Chen c , Yuan-Teh Lee c, d a University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Birmingham, B18 7QH, United Kingdom b Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan c Institute of Epidemiology & Preventive Medicine, College of Public Health, National Taiwan University, Taipei, Taiwan d Institute of Clinical Medical Science, Chinese Medical University, Taichung, Taiwan abstract article info Article history: Received 2 August 2012 Received in revised form 15 September 2012 Accepted 22 September 2012 Available online 13 October 2012 Keywords: Stroke risk stratification Atrial fibrillation CHADS 2 Background: In patients at high risk of stroke, such as atrial fibrillation (AF), there has been great interest in developing stroke risk prediction schemes for identifying those at high risk of stroke. Stroke risk prediction schemes have also been developed in non-AF populations, but are limited by lack of simplicity, which is more evident in schemes used in AF populations. We hypothesized that contemporary stroke risk stratifica- tion schemes used in assessing AF patients could predict stroke and thromboembolism in a non-AF commu- nity population, comparably to that seen in AF populations. Methods: We tested the CHADS 2 and CHA 2 DS 2 -VASc schemes, as well as the AF stroke risk stratification schemes from the Framingham study, Rietbrock et al., 2006 ACC/AHA/ESC guidelines, the 8th American Col- lege of Cardiology (ACCP) guidelines and NICE, for predicting stroke in a large community cohort of non-AF subjects, the Chin-Shan Community Cohort Study. Results: The tested schemes had variable classification into low, moderate and high risk strata, with the pro- portion classified as low risk ranging from 5.4% (Rietbrock et al. to 59.0% (CHADS 2 classical). Rates of stroke also varied in those classified as ‘low risk’ ranging from 1.1% (Rietbrock et al. to 3.5% (Framingham). All com- mon risk schemes had broadly similar c-statistics, ranging from 0.658 (Framingham) to 0.728 (CHADS 2 clas- sical) when assessed as a continuous risk variable for predicting stroke in this population, with clear overlap between the 95% CIs. In an exploratory analysis amongst AF subjects in our population, the c-statistics were broadly similar to those seen in non-AF subjects. Conclusion: Contemporary stroke risk stratification schema used for AF can also be applied to non-AF populations with a similar (modest) predictive value. Given their simplicity (e.g. CHADS 2 score), these scores could potentially be used for a ‘quick’ evaluation of stroke risk in non-AF populations, in a similar manner to AF populations. © 2012 Elsevier Ireland Ltd. All rights reserved. 1. Introduction In patients at high risk of stroke, such as atrial fibrillation (AF), there has been great interest in developing stroke risk prediction schemes for identifying those at high risk of stroke. Stroke risk predic- tion schemes have also been developed in non-AF populations, but are limited by lack of simplicity [1], which is more evident in schemes used in AF populations, such as the CHADS 2 and CHA 2 DS 2 -VASc schemes [2,3]. The CHADS 2 scheme is an amalgamation of stroke risk factors identified from 2 trial-based stroke risk stratification schemes, the AF Investigators and the SPAF-1 schemes [2]. However, the CHADS 2 scheme has many limitations, and does not include many stroke risk factors [4,5]. To complement the CHADS 2 scheme, the CHA 2 DS 2 -VASc has been developed [3], by being more inclusive (rather than exclu- sive) of stroke risk factors. The CHA 2 DS 2 -VASc scheme has been shown to be as good as (and possibly better) than the CHADS 2 scheme in predicting high risk patients with AF who develop stroke and thromboembolism (TE) [6,7], but performs particularly well in identifying those patients with AF who are ‘truly low risk’ of throm- boembolism, who do not need any antithrombotic therapy [6–10]. International Journal of Cardiology 168 (2013) 414–419 ⁎ Corresponding author. Tel.: +44 121 5075080; fax: +44 121 5544083. ⁎⁎ Corresponding author at: Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan. Tel.: +886 2 23123456x62830; fax: +886 2 23511955. E-mail addresses: g.y.h.lip@bham.ac.uk (G.Y.H. Lip), klchien@ntu.edu.tw (K.-L. Chien). 0167-5273/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2012.09.148 Contents lists available at ScienceDirect International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard