Endocarditis risk of the USCI PDA umbrella for transcatheter closure of patent ductus arteriosus.
Academic Article
Overview
abstract
BACKGROUND: The USCI PDA Umbrella is a device to close patent ductus arteriosus (PDA) by a transcatheter technique. Human clinical trials have shown excellent efficacy in reducing or eliminating the PDA shunt, but concerns remain about the risk of infection with this device. The purpose of this study was to evaluate the risk of infection using an animal model. METHODS AND RESULTS: Susceptibility to developing endocarditis was tested by injecting a single intravenous dose of group L streptococcus. Ten piglets with a closed ductus served as controls. Two of these developed valvular vegetations. PDA was produced in 19 animals by balloon dilation of the ductus. Seven of 7 animals with PDA at the time of bacterial injection developed endarteritis of the ductus and valvular vegetations. A PDA Umbrella was placed in the remaining 12 animals, and bacteria were injected 2 weeks after device implantation. Infection was evident in the PDA Umbrella only in the single animal in which the Umbrella had embolized and been left in the left pulmonary artery. Three of the remaining 11 piglets had a significant residual leak, and all developed infection in the ductus and an additional valve. Similar to the control group, none of the animals with complete (n = 8) or nearly complete (n = 2) closure of the ductus by the PDA Umbrella had infection in or around the ductus, and only 1 had a valvular vegetation. CONCLUSIONS: In this animal model, presence of a significant PDA shunt (with or without a PDA Umbrella present) results in significantly increased susceptibility to endarteritis and endocarditis. The PDA Umbrella device does not appear to be susceptible to direct infection as early as 2 weeks after implantation if it is properly located in the ductus arteriosus. Animals with no shunt or a trivial shunt are no more susceptible to developing endocarditis 2 weeks after PDA Umbrella implantation than are controls.
