Stenting the Vertical Neonatal Ductus Arteriosus Via the Percutaneous Axillary Approach

Background/objective: Stenting the ductus arteriosus (DAS) has become an alternative to surgical systemic to pulmonary artery shunts in neonates with ductal-dependent pulmonary blood flow (PBF). Femoral approach for a vertical ductus can be difficult secondary to the acute angle and tortuous course, thus alternative access sites have been explored. Carotid access complications have been reported in 5%-10%. The extensive use of an axillary arterial approach in the United States has not been reported. The aim of this study is to describe our experience with DAS using the axillary approach.

Methods: We reviewed all patients with DAS with an axillary approach in neonates with ductal-dependent PBF (May 2017-May 2018) in our institution. Procedural reports, angiograms, and clinical records of all consecutive patients were reviewed. Procedural technique, procedural outcomes, adverse events, and post-hospital courses are reported.

Results: Seven consecutive patients who received DAS utilizing axillary approach. All patients had ductal-dependent PBF through a vertical, tortuous ductus. Five had pulmonary atresia or near atresia, one had compromised PBF due to dynamic subvalvar obstruction, and one had Tetralogy of Fallot with isolated left pulmonary artery. Axillary access with 3.3 or 4 French sheath was obtained using ultrasound guidance. Bare metal coronary stents were deployed successfully in all. Intra-procedure, one developed in stent thrombus requiring re-stenting. There were no procedural mortalities or major adverse events from axillary access. There is a steep learning curve. Hemostasis was achieved with manual compression. Two patients had reintervention at 6-8 weeks. All patients underwent successful planned surgeries.

Conclusions: This series suggests DAS in neonates utilizing an axillary approach is a feasible and effective alternative for establishing PBF. Axillary arterial approach may be preferred as there is no risk to neurological sequelae and very low risk of limb complications. Larger series are needed to validate this approach.