Article Text
PDF
Abstract
Introduction Atrial septal defects (ASDs) are associated with atrial arrhythmias (AAs) however little is known about the arrhythmia substrate in these patients. Left atrial (LA) fibrosis, detected by cardiac MRI (CMR) is well described in patients with atrial fibrillation and structurally normal hearts but fibrosis has not been described in either atria in the ASD cohort.
Hypothesis We hypothesised that right atrial (RA) fibrosis is present in patients with ASDs and that the right atrium is important for arrhythmogenesis in this cohort.
Aims To evaluate the presence of bi-atrial fibrosis in patients with ASDs and to identify MRI parameters associated with the presence of atrial arrhythmias.
Methods Patients with uncorrected ASDs underwent CMR imaging with dedicated atrial sequences on a 1.5T MRI scanner. MR angiography was performed 90 s after infusion of 0.2 ml/kg Gadovist. 3D LGE imaging using a respiratory navigated, ECG triggered inversion recovery sequence was performed 20 min post contrast administration. The right and left atria were semi-automatically segmented using in house software. Fibrosis maps were generated from the LGE images using a maximum intensity projection technique. Fibrosis scores were generated by thresholding the maps using the image intensity ratio with a fibrosis cut off point of 0.97 times blood pool SI. Right and left atrial and ventricular dimensions were quantified from short and long axis cine imaging and aortic and pulmonary flows measured to calculate shunt fraction (Qp:Qs). Presence of atrial arrhythmia was documented for each patient.
Results 14 patients were included (52 3.7 years, 10 female). Four patients had a history of atrial arrhythmia (2 atrial flutter, 1 atrial fibrillation, 1 atrial flutter and fibrillation). Mean Qp:Qs was 2.3:1. RA area was significantly greater than LA area (33.1±7.5 vs 26.6±6.8 cm2, p=0.012). RA fibrosis burden was significantly greater than LA fibrosis burden (69.7%±13.1% vs 52.7±8.1%, p<0.001). A positive correlation was noted between the degree of right and left atrial fibrosis but this did not reach statistical significance (R=0.511). Presence of AAs was associated with RA size (p=0.011) but not fibrosis burden (figure 1).
A and B- Boxplots demonstrating differences in fibrosis burden and dimensions between RA and LA in ASD patients; C- Relationship between RA and presence of AAs
MRI analysis of right atrial fibrosis. A) 3D LGE images demonstrating areas of high SI (red arrows) in the RA wall; B) Fibrosis maps generated from LGE images above. Red star – Tricuspid annulus, green star – ASD. Red arrow indicating area of high SI in the anterolateral wall
Conclusions Right atrial fibrosis is present in ASD patients to a significantly greater degree than left atrial fibrosis. RA size is significantly greater in ASD patients with than without atrial arrhythmias. Further work is needed to fully define the arrhythmia substrate in terms of structural and electrical remodelling in these patients.