Evaluation of sympathetic function with PET 11C-hydroxyephedrine (HED) and ammonia (13N-NH3) in a canine pacing model of atrial fibrillation

229 Aim: Introduction 11C-hydroxyephedrine (HED) is a norepinephrine analog used for positron emission tomography (PET) imaging of the sympathetic nervous system. HED uptake mechanism is through uptake-1 and to a lesser extent uptake-2, transport into cardiac sympathetic nerve terminals. Pacing the left ventricle of a canine heart can produce atrial remodeling and the development of fibrotic tissue caused by atrial fibrillation (Afib). The use of HED and 13N-ammonia (NH3) allows pre-clinical characterization of the paced left atrium tissue substrate producing atrial fibrillation. Purpose: To evaluate sympathetic function in the left atrium of pre- and post-paced canine hearts. Methods: Ten dogs underwent HED and NH3 PET baseline scans followed by the surgical placement of pacing wires to induce left atrial fibrosis. After five weeks of rapid ventricular pacing (220 to 240 bpm) all dogs underwent HED and NH3 final scans and then were euthanized. PET perfusion imaging at baseline and after pacing involved the administration of 108 ± 28.8 MBq NH3 at the start of a 10-minute dynamic acquisition for baseline and final scans. HED scans were started 50 minutes after the NH3 injection to allow for a decay of 5 half-lives. PET sympathetic neuronal imaging at baseline and pacing involved the administration of 334 ± 22 MBq HED at the start of a 50-minute dynamic acquisition for the baseline scan and 30-minute dynamic acquisition after pacing. All scans were performed on a Biograph mMR hybrid PET-MR system. Regions of interest (ROI) were placed over the left atrium (LA) wall, left ventricle (LV) wall and LV blood cavity. HED and NH3 peak standardized uptake values (SUVpeak) of the LA and LV were measured at baseline and pacing scans. SUVs were measured at 25-30 min after HED, and at 5-10 min after NH3 injection, and normalized to SUVmean values in the LV blood cavity. Corridor-4DM software was used to calculate ungated left ventricle volumes. Paired t-tests were used to compare baseline values to those after pacing. P-values of less than 0.05 were considered significant. Results: Ventricular remodeling was demonstrated with LV cavity volumes increased by 1.59±0.27 times after pacing (p=0.006). Left atrium and left ventricle HED SUVpeak were consistently increased by 1.50±0.27 (p=0.0002) and 1.13±0.20 (p=0.062) times (respectively) suggesting altered SNS activity. Corresponding NH3 SUVpeak values were not significantly changed in the LA (×1.14±0.21) nor the LV (×1.03±0.20), confirming that the observed changes in HED uptake were not due to blood flow effects. Conclusion Rapid ventricular pacing resulted in LV dilatation and altered focal SNS activity in this canine model of atrial fibrillation. Future studies are needed to correlate these findings with measures of atrial fibrosis. Research support Ontario Research Fund (RE07-021) and Canadian Arrhythmia Network (NCE-15-P06-001).