Abstract 2619: Exercise-Induced Appearance of Arteriovenous Fistula in Pulmonary Circulation: Novel Pathophysiology of Unexplained Dyspnea on Exertion?
Appearance of right to left interatrial shunting (IAS) through a patent foramen ovale during exercise has known to be accounted for a subset of patients presented with unexplained dyspnea on exertion (DOE). However, a role of exercise appearance of arteriovenous fistula in pulmonary circulation (AVF) which may cause ventilation circulation mismatch in these patients in DOE has not tested because invasive blood sampling or cardiopulmonary exercise testing cannot easily differentiate between IAS and AVF. To assess a clinical significance of AVF in patients with unexplained DOE, we differentiated AVF from IAS using agitated saline microbubble (MB) injections during supine bicycle exercise echocardiography (BE) and evaluated the characteristics of IAS and AVF. BE was performed for 38 consecutive outpatients referred for evaluation of unexplained DOE (age; 52±13, 34 females, data are mean±SD). MB were injected at rest, peak exercise, and recovery to assess IAS and AVF. Pulse oximetry and echocardiographic assessment of peak pulmonary artery pressures (PAP) based on peak tricuspid regurgitation velocities were performed at each stage. The delayed MB appearance in the left atrium at more than 5 beats after the right atrial opacification was defined as AVF and the earlier MB appearance was classified to IAS. A number of MB appeared in the left atrial <20 per heart beat were considered small. In 36 cases in which MB were successfully administered during BE, 3 small IAS (8%) were noted at baseline, but none of them exacerbated and no new IAS appeared at peak exercise. On the other hand, 3 small AVF (8%) were noted at baseline and additional 8 new small AVF appeared at peak exercise (22%), which were more frequently than IAS at peak exercise (P<0.01 with chi-square test). The AVF noted at baseline did not exacerbate during peak exercise and all newly developed AVF during peak exercise disappeared at recovery. Six out of 8 cases (75%) with the exercise-induced AVF were associated with exercise-induced pulmonary hypertension (PAP; 53±15 mmHg) and 3 (38%) were associated with exercise-induced hypoxemia. Our data suggest that exercise-induced AVF occurs more frequently than IAS in patients with unexplained DOE and this may contribute to DOE more frequently than previously thought.