Redundancy of foramen ovale flap may mimic fetal aortic coarctation

Abstract

To assess the relationship between presence of a redundant foramen ovale flap (RFOF), in the absence of a clearly restrictive foramen ovale, and ventricular disproportion, in three groups of fetuses: (1) those with a final diagnosis of aortic coarctation (CoA); (2) those referred for suspicion of ventricular disproportion and/or CoA which did not develop CoA postnatally; and (3) normal fetuses. This was a retrospective study including 73 fetuses: 12 with a final diagnosis of isolated CoA; 30 referred for suspicion of ventricular disproportion and/or CoA, which did not develop CoA postnatally; and 31 normal fetuses. Four-dimensional volume datasets and clips were assessed offline. Maximum diameters of the FOF (FOFD), left atrium (LAD), right atrium, left and right ventricles and, when available, aortic isthmus, were measured, as were areas of the FOF (FOFA), left atrium (LAA) and right atrium. The left/right ratios for all segments of the heart, as well as the FOFD/LAD ratio and FOFA/LAA ratio, were calculated. Regression analysis was performed to assess the relationship between RFOF and ventricular disproportion and means were compared by ANOVA. Repeatability was fair, with all variables having an intraclass correlation coefficient ≥ 83%. In the pooled group of fetuses with no CoA found at birth (normal fetuses plus those with ventricular disproportion (n = 61)), there was a significant linear correlation between redundancy of the FOF and degree of ventricular disproportion (P < 0.01 and P < 0.05 for diameter and area ratios, respectively). Categorizing the FOF redundancy, FOFD/LAD ratio ≥ 0.65 was significantly associated with ventricular disproportion (P = 0.006). Based on the degree of FOF prominence, we described four categories of redundancy, ranging from no redundancy/ventricular disproportion (Stage 0) to severe redundancy/ventricular disproportion with transient obstruction of the foramen ovale or mitral orifice (Stage III). Comparing cases without neonatal evidence of coarctation but FOFD/LAD ratio ≥ 0.65 vs those with neonatal evidence of coarctation, there was no statistically significant difference in the degree of ventricular disproportion or in the Z-score of the aortic isthmus maximum diameter. This study demonstrates that: (1) there is an association between RFOF and ventricular disproportion, independent of the association with a restrictive foramen ovale, and (2) the presence of a RFOF may mimic CoA. In fact, it causes both ventricular disproportion and a significant reduction in the diameter of the aortic isthmus, associated in some cases also with reversed isthmic flow. Future prospective studies are needed to evaluate whether focusing the sonologist's attention on the appearance of the foramen ovale may reduce the rate of false-positive diagnosis of CoA. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.