Author + information
- Achille Gaspardone, MPhil, MD†∗ (, )
- Arianna Giardina, MD†,
- Maria Iamele, MD†,
- Gaetano Gioffrè, MD†,
- Mauro Polzoni, MD‡,
- Filippo Lamberti, MD†,
- Romolo Remoli, MD†,
- Gregory A. Sgueglia, MD†,
- Marco Papa, MD§ and
- Cesare Iani, MD‖
- †Divisione di Cardiologia, Ospedale S. Eugenio, Rome, Italy
- ‡Divisione di Anestesia e Rianimazione, Ospedale S. Eugenio, Rome, Italy
- §Dipartimento Cardio-Toracico, Unità di Cardiopatie Congenite e Strutturali, Istituto Scientifico San Raffaele, Milan, Italy
- ‖Divisione di Neurologia, Ospedale S. Eugenio, Rome, Italy
- ↵∗Divisione di Cardiologia, Dipartimento di Medicina, Ospedale S. Eugenio, ASL RMC, P.le dell’Umanesimo 10, 00144 Rome, Italy
To the Editor:
Atrial arrhythmias after percutaneous patent foramen ovale (PFO) closure, including atrial fibrillation (AF), has been consistently reported in different series suggesting a causal link between mechanical closure of PFO and the new onset of post-procedural arrhythmias (1–3). We have systematically assessed the effect of percutaneous PFO closure upon the development of post-procedural arrhythmias in 221 consecutive patients (144 women; mean age 48 ± 13 years) undergoing percutaneous PFO closure. At clinical evaluation, no patient had symptoms indicative or suggestive of arrhythmias. Indications for closure were the presence of significant (grade >1) basal or Valsalva-induced right-to-left atrial shunt (RLS) at contrast transthoracic echocardiography (TTE) associated with cryptogenic stroke (n = 75), repeated transient ischemic attacks (n = 108), severe migraine (≥4 attacks/monthly resistant to triple pharmacologic therapy including topiramate) with presence of multiple bilateral cerebral ischemic lesions at magnetic resonance imaging (n = 30), decompression illness (n = 4), platypnea-orthodeoxia syndrome (n = 2), and documented peripheral embolism (n = 2). All patients underwent 24-/48-h Holter monitoring or external loop recorder electrocardiography (ECG) recording 1 to 3 months before PFO closure and between 3 and 6 months after the procedure. All ECG recordings were manually reviewed. Supraventricular (SV) arrhythmias were categorized as sporadic ectopic beats (≥5% of total beats), AF (lasting ≥30 s), and SV tachycardia. Contrast TTE was performed at 1 and 6 months to evaluate residual RLS. To evaluate predictors of post-procedural arrhythmias, a binary logistic regression analysis was performed. The following dichotomized variables were considered as potential risk predictors in exploratory analysis: age, sex, body mass index, device type, pre-closure arrhythmias, atrial septal aneurysm and septal excursion, severity of RLS shunt, and immediate and 6-month residual shunt (Table 1). Variables included in the multivariable model were those showing a statistical association (p < 0.1) with post-procedural arrhythmias at univariate analysis. Data are reported as mean ± SD. A 2-sided p value <0.05 was required for statistical significance. The mean follow up was 14 ± 7.8 months.
Pre-closure SV arrhythmias were detected in 51 patients (23%) who were all asymptomatic. Transient episodes of AF were detected in 9 patients (4%); sporadic ectopic beats in 36 patients (16.3%), and ST in 6 patients (2.7%). Paroxysmal AF represented 18% of all arrhythmias. At 3- to 6-month Holter monitoring, SV arrhythmias were present in 48 patients (21.7%). Transient episodes of AF were detected in 11 patients (4.9%), sporadic ectopic beats in 33 patients (14.9%), and SV tachycardia in 4 patients (1.8%). AF represented 22.9% of all post-procedural arrhythmias. In synthesis, 29 patients presented SV arrhythmias before and after PFO closure, 21 patients had arrhythmias only before PFO closure and 18 patients had arrhythmias only after PFO closure. All the 9 patients presenting AF before PFO closure had the same arrhythmia after the procedure. At multivariate analysis, the only predictor of post-closure arrhythmias was the presence of pre-closure arrhythmias (p < 0.0001) (Table 1).
Our study has the unique peculiarity in being specifically designed to investigate the effect of percutaneous PFO closure upon post-procedural atrial arrhythmias. All patients included in the study underwent 1 to 3 months before closure to rhythm monitoring by using traditional 24-/48-h Holter-ECG recording or external loop recorder monitoring (one-third). ECG recording has been repeated 3 to 6 months after transcatheter PFO closure, a time interval considered optimal for the anatomical stabilization of the device. The main finding of our study is that percutaneous PFO closure is not per se an inductor for post-procedural atrial arrhythmias; indeed, the only predictor of post-procedural arrhythmias appear to be the presence of arrhythmias before PFO closure. Interestingly, the majority of patients with post-procedural arrhythmias had the same arrhythmias before PFO closure thus suggesting an intrinsic increased susceptibility to develop rhythm disturbances (4). Indeed, at multivariate analysis we could not identify any predictors of post-procedural arrhythmias except pre-closure arrhythmias. In particular, residual RLS, device type and size, atrial and PFO anatomical features were not predictive of post-procedural arrhythmias. This finding suggests that the device-related atrial mechanical stretch per se does not increase electrical vulnerability in nonsusceptible patients.
An interesting ancillary finding of this study, albeit non conclusive for the lack of a control group, is that the prevalence of atrial arrhythmias, including atrial fibrillation, in patients with PFO is unexpectedly higher than that estimated in the general population (5). This evidence suggests that, at least in a proportion of patients with PFO, asymptomatic unrecognized arrhythmias might play a pathogenetic role in systemic embolism. An obvious flaw of the present study is the technical limitation of the recording systems used that probably underestimated the true prevalence of arrhythmias. Furthermore, our study lacks a control group so no definite conclusions can be drawn regarding the prevalence of SV arrhythmias in patients with PFO.
In conclusion, PFO closure does not appear to be per se an inductor of post-procedural arrhythmias. The high prevalence of arrhythmias detected before and after PFO closure may be related to an increased atrial electrical vulnerability in patients with atrial septal abnormalities.
Please note: This study was supported by Cardiva ONLUS Italy. Dr. Gaspardone is a proctor for St. Jude Medical and Occlutech. Dr. Lamberti is a proctor for Biosense Webster and St. Jude Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2013 American College of Cardiology Foundation