Residual shunting after PFO closure: device size matters more than anatomy

Abstract Introduction Recent studies have shown that the prevalence of residual shunting following patente foramen ovale (PFO) closure ranges from 10% to 30%. Various factors have been identified as potential predictors of residual shunts, including the anatomical structure of the PFO tunnel, the presence of atrial septal aneurysms, and the size of the occluder device. Purpose This study aims to evaluate the incidence of residual shunting and identify its key predictors in patients who underwent PFO closure at a tertiary hospital centre. Methods A retrospective analysis was conducted on patients who underwent PFO closure at a tertiary hospital centre. The study assessed implantation success, device sizing, echocardiographic follow-up methods, and the detection of residual shunts. Pre-implantation transesophageal echocardiography (TEE) was performed to measure PFO tunnel size and identify the presence of septal aneurysms. Device size was recorded from hospital charts. Residual shunting was evaluated using transthoracic echocardiography (TTE) with a bubble test during the Valsalva maneuver at 6-month and 1-year follow-ups. Statistical analysis was conducted using the Chi-square test, Cramér’s V test, and linear multivariate regression to identify significant predictors of residual shunting. Results Between August 2016 and December 2024, 104 PFO closures were attempted, with successful device implantation in 100 patients (96.2%). Amplatzer occluder devices were used: 50 patients (51.0%) received an 18/25 mm device, 15 (15.3%) a 25/25 mm device, 9 (9.2%) a 25/30 mm device, 14 (14.3%) a 25/35 mm device, and 10 (10.2%) a 30/30 mm device; size data were unavailable for 2 cases. Pre-implantation TEE measurements were available in 54 patients (54.0%) for PFO tunnel width (3.9 ± 2.4 mm) and 51 patients (51.0%) for tunnel length (10.9 ± 3.6 mm); atrial septal aneurysm was reported in 33 patients (33.0%). During the first year, residual shunting was detected in 18 patients (18.0%) exclusively via TTE bubble test, with 78% occurring in those with larger devices. Beyond one year, persistent residual shunt was found in 11 patients (10.6%), with 73% in those with larger devices. Larger device size (p0.001) and left atrial disk (p0.001) were independent risk factors for residual shunting, with a significant association between device size and shunting (Chi-square, p0.001; Cramer’s V = 0.577). No significant association was found between TEE-derived PFO tunnel measurements and residual shunting. Conclusion PFO closure outcomes at our center align with published data, with an 18.0% incidence of residual shunting. All residual shunts were detected exclusively by the TTE bubble test, highlighting its superiority over Doppler-based methods for follow-up. Larger device size, particularly the left atrial disk, was the only significant predictor of residual shunting in our cohort, whereas anatomical PFO features were not predictive, contrasting with prior reports.