Simple scores to predict 1-year mortality in atrial fibrillation

Area of Science:

• Cardiology
• Clinical Epidemiology
• Biostatistics

Background:

• Traditional scores like CHADS2 and CHA2DS2-Vasc predict stroke and mortality in atrial fibrillation (AF) patients.
• More complex scores have been proposed for improved survival prediction in AF.
• This study evaluates the performance of various clinical scores for predicting 1-year mortality in AF patients.

Purpose of the Study:

• To compare the predictive accuracy of CHADS2, CHA2DS2-Vasc, R2-CHADS2, R2-CHA2DS2-Vasc, and CHA2DS2VASc-RAF scores for 1-year mortality in AF patients.
• To assess the impact of incorporating renal function (GFR or creatinine clearance) into risk scores.
• To determine if more complex scores offer clinically significant improvements over traditional scores for mortality prediction.

Main Methods:

• Retrospective analysis of AF patients from the BLITZ-AF register.
• Calculation of CHADS2, CHA2DS2-Vasc, R2-CHADS2, R2-CHA2DS2-Vasc, and CHA2DS2VASc-RAF scores.
• Comparison of score performance using receiver-operating characteristic (ROC) curves to predict 1-year all-cause and cardiovascular mortality.

Main Results:

• R2(ClCrea)-CHADS2 demonstrated the highest predictive accuracy for all-cause mortality (AUC 0.734).
• All tested scores showed association with all-cause mortality, but differences were not always significant.
• CHA2DS2VASc-RAF performed best for cardiovascular mortality (AUC 0.757), outperforming simpler scores.
• The widely used CHA2DS2-Vasc score showed high sensitivity for all-cause (AUC 0.695) and CV mortality (AUC 0.691).

Conclusions:

• Complex risk scores may offer marginal statistical benefits but lack clinically relevant advantages over simpler scores for 12-month mortality prediction in AF.
• The established CHA2DS2-Vasc score remains a valuable tool with high sensitivity for predicting both all-cause and cardiovascular mortality.
• Clinical decisions should consider the established utility and accessibility of the CHA2DS2-Vasc score alongside its predictive performance.