Discussion
Risk Factors
Predisposing factors for device embolization may be: device too small, atrial septal aneurysm (ASA), thick septum secundum (> 10 mm), long tunnel5 (≥ 8 mm)6, deficient rims of surrounding tissue and device malpositioning3. In this case, we believe that the choice of a device too small, the presence of an ASA and a long tunnel could have favoured the device leaving from its site. The PFO tunnel was longer than measured at the TEE during the intervention (24 mm vs. 13-14 mm, respectively): the possible tunnel length underestimation could have led to the choice of an unsuitable device for the PFO characteristics and, therefore, to its displacement. The patient denied having performed manoeuvres potentially able to facilitate the device displacement, such as having coughed heavily or repeatedly7,8. A peculiarity of this case is the finding of this complication in presence of other congenital defects (BAV and previous AC). The presence of other congenital defects may have predisposed the device displacement (e.g. precluding an adequate device fixation). Furthermore, the surgical repair of the aortic coarctation may have favoured a more distal drop of the device. The device embolization in presence of multiple congenital defects represents a unique event in the literature, which gives thought to other possible mechanisms of the PFO occluder device displacement in patients with multiple congenital defects.
Symptoms
The mechanisms underlying the symptoms remain unclear. We assume that the onset of dyspnoea after the intervention was caused by the worsening of the right-left shunt (provoked by exercise) due to the PFO enlargement following the device displacement; this may be supported by the detection of a significant oxygen desaturation on the 6-Minutes Walking Test (98% to 92% after 2 minutes), an event already observed in other patients with PFO, defined as Provoked Exercise Desaturation (PED) 9. We excludedPlatypnea Orthodeoxia Syndrome (POS) , as the patient had a normal arterial oxygen saturation in standing position. We also presume that the chest pain radiated to the limbs, after having excluded statin-induced myopathy (normal creatine kinase), was the expression of an aortic syndrome with atypical features, due to the presence of the foreign body into the abdominal aorta. This may be confirmed by its disappearance after the device retrieval.
Follow-up
Lastly, we believe that if a TTE before discharge or a TTE with bubble study within 1 month from discharge had been performed, the problem would certainly have been identified earlier and with lower risk for the patient. Indeed, the current guidelines recommend TTE before discharge, TTE with bubble study after 1, 6, 12 months and, then, every 1-2 years3. However, a correct follow-up probably was hindered by the COVID-19 pandemic.
Conclusion
As shown, the embolization of a PFO occluder device is a very rare but possible complication (in this case reported for the first time in a patient with multiple congenital defects). For this reason, we strongly recommend paying attention to 1) risk factors before the intervention (PFO characteristics and multiple congenital defects) to correctly choose the device to be placed, and to 2) the clinical-echocardiographic follow-up to recognize and treat this severe complication as soon as possible.