Background Magnetic resonance imaging (MRI) assists in delineating pulmonary venous (PV) anatomy prior to catheter ablation of atrial fibrillation (AF). Use of MRI parameters to predict early ablation success is not well studied.
Methods Thirty-four consecutive patients (age 56 ± 10.8, males 74%) with symptomatic, drug refractory AF (85% paraoxysmal, 15% persistent) underwent ECG-gated, breath-hold MRI (1.5 T Siemens Sonata) with 3D-reconstruction MRA. All had circumferential left atrial ablation. Cardiac chamber dilatation were graded normal, mild, moderate, or severe based on assessment of images and data from standard analysis software. Early ablation success was defined as freedom from symptoms and presence of sinus rhythm on ECG/Holter at follow-up.
Results Average left ventricular (LV) ejection fraction, LV end diastolic dimension, and left atrial (LA) dimension are 51% (± 10), 52 mm (± 7), and 42 mm (± 7), respectively. Eleven (32%) of the patients had a variant pulmonary venous anatomy (10 right and 1 left middle vein). The mean follow-up was 5.2 months. In this period, 50% remained symptom free and without evidence of recurrence. Twenty-two percent required repeat ablation procedures. Multivariate logistic regression identified age as an independent predictor of early outcome (OR 1.26, p = 0.019) whereas MRI derived parameters did not.
Conclusions Although MRI is helpful in delineating anomalous pulmonary vein anatomy, MRI derived anatomic parameters are not useful in predicting early AF ablation success.