Impact of Occluder Device Type on Success of Percutaneous Closure of Atrial Septal Defects-A Medium-Term Follow-up Study

Methods: Three hundred nine consecutive patients with presumed paradoxical embolism and scheduled for percutaneous closure of patent foramen ovale or atrial septal defect between 1997 and 2006 were considered for this study. Device implantation failed in seven patients (nonstable device position), so 302 patients (46 +/- 12 years, 190 males) formed the study group using Amplatzer (TM) (n = 208), Starflex (TM) (n = 61), and Cardiastar (TM) (n = 33) occluders. Follow-up transesophageal echocardiography was performed 1, 3, and 6 months after implantation. Results: Periinterventional complications occurred in two patients (cardiac arrhythmias). There were more residual shunts in the Starflex (TM) and Cardiastar (TM) group than in the Amplatzer (TM) group at 6-month follow-up (8/61 vs. 7/33 vs. 8/208, P = 0.0005). Performing logistic regression, the type of occluder device was a significant risk factor for the presence of residual shunt 6 months after implantation (P = 0.0033; Cardiastar (TM) vs. Amplatzer (TM) OR 6.346, 95% CI 1.998 - 20.156; Starflex (TM) vs. Amplatzer (TM) OR 4.369, 95% CI 1.444 - 13.222). During mean follow-up of 31 +/- 8 months 16 recurrent thromboembolic events occurred; the annual recurrence was 2.1% for combined end-point stroke, TIA, and peripheral embolism (Amplatzer (TM): 1.3%, Starflex (TM): 3.9%, and Cardiastar (TM): 3.6%, P = 0.0467). The presence of an atrial septal aneurysm was the only significant risk factor (P = 0.0168, OR 3.664, 95% CI 1.263 - 10.630) for the occurrence of thromboembolic events. Conclusions: Percutaneous closure of patent foramen ovale or atrial septal defect is a safe procedure with little incidence of peri- and postprocedural complications. There is a significant difference between the Amplatzer (TM), Cardiastar (TM), and Starflex (TM) occluders in regard to complete closure of the defects and annual recurrence of thromboembolic events. (J Interven Cardiol 2009;22:503-510).