Introduction: Tetralogy of Fallot (TOF), the most common
cyanotic congenital heart disease in adults, has excellent long-term
survival. However, many patients (30-45%) develop late arrhythmias.
Previous studies have identified predictors of arrhythmia (atrial or
ventricular) using clinical markers that predate arrhythmia onset for
many years. Our objective was to develop predictive models for incident
atrial arrhythmias or any arrhythmia (atrial and ventricular) within two
years of clinical evaluation and diagnostic testing. Methods: A
single-center retrospective cohort study of 188 adults with repaired
TOF. We included only patients with results from ECG and echocardiogram
data in the required time interval (3-24 months before first arrhythmia
for cases; 24 months of follow-up for controls). Predictive
multivariable models for risk of atrial arrhythmia using logistic
regression with a least absolute shrinkage and selection operator
(LASSO) penalty were developed. Results: Of 41 demographic,
surgical, and diagnostic variables, six were selected as having
predictive value for atrial arrhythmia based on cross validation. The
factors with the greatest predictive value in decreasing order were
moderate / severe tricuspid regurgitation (adjusted odds ratio (OR)
149.42), QRS fragmentation (OR 28.08), severe pulmonary regurgitation
(OR 8.22), RV systolic dysfunction (OR 2.95), 1
degree AV block, and age at time of surgical repair.
Conclusion: Predictors for atrial or any arrhythmia in our
study were similar, suggesting a common pathophysiologic substrate,
i.e., abnormal right ventricle anatomical function and
electrophysiologic properties (conduction and repolarization).