A Retrospective Analysis of the Predictive Potential of Pre-Operative

Roughly thirty percent of people that undergo open heart surgery get an abnormal heart beat afterwards known as atrial fibrillation (AF). While not life threatening, this abnormal heart beat increases the likelihood of stroke and delays recovery. There are strategies to prevent post-operative AF, but they are costly and sometimes have undesirable side effects. Therefore, it would be best if we use these preventive treatments only in high risk patients.

We intend to develop a risk prediction model based on demographic and electrocardiogram (ECG) findings that will predicted who is likely to get AF. We will develop this model using data already available on patients who have undergone cardiac surgery. The development of this model will use the latest mathematical algorithms similar to those used to study genetic evolution. This type of model is capable of looking at many parameters in an unbiased way, so that only the strongest, independent predictors remain in the final model. Once, the model is developed, we will validate the model by comparing our predictions to actual outcomes previously recorded in the database.