Derivation of a simple clinical model to categorize the probability of acute myocardial infarction in patients with atrial fibrillation.

OBJECTIVE: Atrial fibrillation (AF) is independently associated with a higher risk of acute myocardial infarction (AMI). The occurrence of AMI in AF patients may lead to dismal prognosis. Risk assessment is a fundamental component of prevention for AMI. PATIENTS AND METHODS: 2419 consecutive patients with nonvalvular AF were enrolled in this retrospective study. A logistic regression analysis was performed on clinical variables to create a simple clinical prediction rule. The following nine variables and assigned scores (in brackets) were included in the prediction rule: age ≥65 years (1.0), heart failure (1.0), hypertension (1.0), diabetes mellitus (1.0), hyperlipidemia (0.5), history of stroke/TIA (0.5), vascular disease (1.0), current smoking (0.5), and resting heart rate >90 beats/min (1.0). Patients were considered to have a low probability if the score was ≤2.5, moderate if the score was 3.0 to 4.0, and high if the score was ≥4.5. The AMI unlikely was assigned to patients with scores <3.5 and AMI likely if the score was ≥3.5. To evaluate the score, we included an external validation cohort of 1810 nonvalvular AF patients from the Cardiology Center, Shandong Provincial Hospital affiliated to Shandong University, Jinan, China. RESULTS: The score showed a good ability in discriminating AF patients experiencing AMI both in the internal derivation cohort, with a c-index of 0.80 [95% Confidence Interval (CI) 0.77-0.83, p<0.001] and in the external validation cohort (c-index 0.73, 95% CI 0.69-0.77, p<0.001). Our scoring system offered significantly better predictive performance than the CHA2DS2-VASc score (c-index 0.80 vs 0.71, p<0.001). CONCLUSIONS: Our scoring system is a simple and accurate way of predicting the risk of AMI in AF patients. Therefore, more accurate targeting of preventive therapy will be allowed.