Abstract
Treatment options for advanced stages of congestive heart failure remain limited. Left ventricular assist devices (LVADs) have emerged as a means to support failing circulation. However, these devices are not without significant risk such as major open chest surgery. We utilized a novel approach for device placement at the aorto-left atria continuity as a site to create a conduit capable of accommodating a percutaneous LVAD system. We designed and developed an expandable nitinol based device for placement at this site to create a shunt between the LA and aorta. Our experiments support this anatomic location as an accessible and feasible site for accommodation of an entirely percutaneous LVAD. The novelty of this approach would bypass the left ventricle, and thereby minimize complications and morbidities associated with current LVAD placement.
1
Introduction
Congestive heart failure (CHF) is reaching epidemic proportions around the world with over 5 million people affected only in the U.S. . Left ventricular assist devices (LVADs) have dramatically increased in usage in the recent years for treating intractable CHF not only as a bridge to transplantation or recovery, but also as a destination therapy . However, present LVADs have significant limitations that prevent widespread adoption . An ideal LVAD device would be placed percutaneously, thereby eliminating the need for open heart surgery. It would have limited contact of arterial blood with artificial surfaces and would have no extracardiac circuits, thus minimizing the risk of thromboembolism. It would not require a ventricular incision and would assist circulation effectively, irrespective of valve function. An anatomic site at which oxygenated blood is in juxtaposition with major vessels would provide an ideal site for utilization and percutaneous delivery of the hemodynamic support to the failing circulation. We sought to establish such a site and further to create a stable shunt which will be reliably accessed and used as a platform for placement of a percutaneous LVAD system.
2
Methods and results
We utilized an innovative approach at a unique anatomic location at which the left atrium (LA) and aorta are in juxtaposition ( Fig. 1 ). This site is the continuity of the posterior aortic wall in the region of the non-coronary and left coronary cusps of the aortic valve with the atria. A tissue plane at this site separates oxygenated blood in the left atrium from the aorta.
We examined 40 formalin-fixed human hearts to confirm the presence of this aorto-left atrial continuity and to define its dimensions . A continuity plane of the posterior aorta (non-coronary cusp) with the left atrium (LA) existed and it was present in 40 out of 40 (100%) analyzed hearts. The aorto-left atria continuity had a mean length of 9.6 ± 2.0 mm and a mean height of 5.6 ± 1.6 mm.
Our innovation includes the design and development of a stable conduit for percutaneous placement in this aorto-left atrial continuity that would support pump placement. A custom, self- expandable nitinol conduit was developed (U.S. Patent Application No. 61-177075) for positioning in a fistula created between the aorta and LA in a feasibility study using a canine model. This conduit was an early prototype with a 12-Fr (0.156 inch) central lumen that, in future iterations, would accommodate a pump ( Fig. 2 ). It consists of two circumferential phalanges, each serving to mechanically anchor the conduit on the left atria and aorta side respectively, and a central pore which will be the firm platform of the created shunt.
In three canine experiments we successfully deployed a prototype of the invented device ( Fig. 3 ). A retrograde approach (retrograde puncture from the aorta to the LA) was used to access the aorto-LA continuity. The deployment process is illustrated in Fig. 4 . Fistula creation and deployment process were guided and monitored continuously by intracardiac echocardiogram (ICE) and fluoroscopy. To aid in identification of the noncoronary cusp an 8 F five millimeter, bipolar electrogram recording catheter was used. In all three canine experiments, we observed for a mean duration of 57 min and continuous ICE was used to monitor for any complications. No device dislodgment, pericardial effusion, or wall motion abnormalities suggestive of ischemia were observed in real time ICE assessment. Evaluation of canine hearts at necropsy did not reveal any aortic tear, aortic or atrial perforation, dissection and intraaortic or intracoronary thrombus.
