Prevalence and predictors of residual shunt in patient undergoing patent foramen ovale closure

Abstract Introduction Percutaneous closure of patent foramen ovale (PFO) aims at eliminating the chance of paradoxical embolism. However, a certain number of patients maintain a right-to-left shunt over time after the procedure. The proportion of patients undergoing PFO closure who remain with residual shunt is unknown, and the reasons for this finding are uncertain. Aim We aimed to determine the prevalence of residual shunts in PFO patients within 6 and 12 months after the procedure, and to investigate predictors of residual shunt. Methods We included patients undergoing PFO closure with any commercially available devices and with guidance by intracardiac echocardiography from October 2011 to March 2018, enrolled in single centre. Patients were followed up with transcranial doppler echocardiography at 6 months and at 12 months to assess the presence of residual shunt: grade 1 if 1-10 Micro-Embolic Signals (MES), grade 2 if 11-25 MES, grade 3 if 25 MES. Predictors of any residual shunt were explored with multivariate regression analysis. Candidate predictors were those variables that were significant at a p value level of 0.10 at univariate analysis. Results A total of 303 patients were included. Residual shunt at 6 months was detected in 176 patients (58%), in particular 87.5% grade 1, 6.25% grade 2 and 6.25% grade 3. Patients with residual shunt were more likely to present with coagulation disorders (36.6% vs 25.0%, p = 0.04) and long tunnels defined as tunnels ≥ 10 mm (80% vs 65.4%) compared with patients without residual shunts (Table 1). At multivariate analysis, including univariate predictors of residual shunt (i.e., body mass index, diabetes mellitus, coagulation disorders, left ventricular ejection fraction and long tunnels), the presence of a long tunnel emerged as the only independent predictor (odds ratio 1.90, 95% confidence interval 1.10 to 3.28; p = 0.022). Only 42 patients (24.4%) were still positive for shunt at 12 months, in particular 78.6% grade 1, 9.5% grade 2 and 11.9% grade 3. In this case, the multivariate analysis, which included univariate predictors of residual shunt (Table 2: body mass index, smoke, migraine, dyslipidemia, and septal aneurysm), showed that septal aneurism, defined as a protrusion of interatrial septum ≥ 15 mm, resulted as the only independent predictor (odds ratio 4.43, 95% confidence interval 1.97 to 9.98; p 0.001). Conclusions More than half of patients undergoing PFO closure presents a residual shunt at 6-month transcranial doppler echocardiography, even if most is grade 1. Patients with long tunnels are more likely to present with residual shunt. Moreover, only 24.4% of these patients still have a residual shunt at 12-month follow-up, and mostly of mild severity. In this case, the presence of a septal aneurysm appears to be a predictive factor for slow endothelialization. Extending dual antiplatelet therapy in these categories of patients might be considered.6-month predictors of residual shunt 12-month predictors of residual shunt