Right ventricular (RV) failure is a problematic complication of left ventricular assist device (LVAD) placement, occurring at a rate of ~20%, and portending a poor prognosis. Left ventricular assist device patients needing biventricular support have a dismal survival of no better than 60% at 1 year.1 Prediction of RV failure is difficult, and present preoperative risk scores provide no more than modest discrimination of acute RV failure risk.2 Thus, despite advancements in LVAD technology and patient selection, the burden of acute RV failure in patients supported with continuous-flow LVADs (CF-LVADs) remains significant. Management strategies for post-LVAD RV failure include pharmacologic agents (such as pulmonary vasodilators and inotropic agents), ultrafiltration, and temporary or permanent right ventricular assist devices (RVADs).2 Recently, a percutaneous RVAD called the TandemLife Protek Duo (TPD; TandemLife, Pittsburgh, PA) has been introduced. The Protek Duo (TPD) is a temporary RVAD placed via the right internal jugular vein, capable of providing up to 4.5 L of flow. We report a two-center experience using the TPD in 17 patients with RV failure, 12 of whom were post-LVAD implantation.
A detailed, retrospective database review of 17 patients receiving percutaneous RV support using the TPD was conducted. Patient clinical characteristics, demographics, and cardiopulmonary hemodynamics (N = 15 patients had hemodynamic measures) were obtained before the onset of TPD support.
Right ventricular failure was a clinical diagnosis made by the treating advanced heart failure surgeon or cardiologist. General indications for TPD support included ongoing elevation of right atrial (RA) pressure despite aggressive pulmonary vasodilator, diuretic, or inotrope support; inability to wean inotrope or vasopressor support while on CF-LVAD support despite acceptable left ventricular (LV) filling pressure and LVAD function; and clinical signs of RV dysfunction including hepatic congestion, lower extremity edema, or cardiorenal syndrome.
In brief, the implant technique for the TPD is as follows: the modified Seldinger technique is used to access the internal jugular vein (typically the right). After sheath placement (5F–9F), a balloon tip catheter is floated to wedge position under fluoroscopy. A 0.035 inch stiff wire is then inserted through the central lumen of the catheter (Lunderquist, Cook Medical, Bloomington, IN or equivalent). Although the balloon tip catheter is carefully removed under fluoroscopy, the wire tip is seated in the distal pulmonary artery and the neck access site is progressively dilated to 29F. Heparin is given and once a therapeutic-activated clotting time (300 seconds) is achieved, the TPD is then inserted over the stiff wire and then the two ports are clamped. “Wet-wet” connections are made in the usual way with a syringe used to provide fluid flow while the operator connects the tubing ends. Transesophageal echocardiography (TEE) or fluoroscopy is used to confirm whether the tip of the TPD catheter is in the pulmonary artery. The set speed is based on either of the following: TEE findings (to assess RV and LV size), LVAD flows, or provider discretion. Usually a “ramp protocol” is conducted, which consists of increasing pump speed until there are negligible increases in flow, development of tubing chatter, or the interventricular septum is midline. Then, the device is set on the lowest speed associated with near-maximal flow. The device is secured into place with multiple sutures, and a partial thromboplastin time of 50–70 seconds is maintained thereafter with unfractionated heparin. At St. Vincent Heart Center, cannula and device insertion were performed by the heart failure cardiologist in a hybrid operating room (OR), with cooperation from cardiothoracic surgery. The Newark Beth Israel heart failure cardiology team performed these procedures with local anesthesia and moderate sedation with fluoroscopic guidance in a cardiac catheterization suite.
Echocardiography, TPD and LVAD flows, and laboratories were used to assess RV recovery or the ability to wean TPD support. Speeds were sequentially turned down, and patients deemed suitable for explant had their devices removed at the bedside after placement of a purse-string stitch with subsequent brief manual pressure.
Categorical data are presented as n (%) and continuous data as mean ± standard deviation. All patients provided consent or waiver of informed consent was granted according to Internal Review Board policy before data acquisition.
From October 2014 to March 2016, 17 patients underwent TPD implant for management of clinical RV failure. Of these 17 patients, 12 had a durable LVAD in place; two patients had post-heart transplant (one early postoperative and one several years post with rejection); one was supported on both a Tandem LVAD and TPD, while two had predominant RV failure supported on TPD alone. Table 1 below describes the baseline and demographic characteristics of these patients, on a mean length of support of 10.5 ± 6.5 days (range: 0–24 days).
The TPD was successfully weaned in 23% (n = 4) of patients without need for home inotropes or urgent transplant due to RV failure. In 35% of patients (n = 6), the device could not be weaned and patients required either a surgical RVAD (CentriMag, Abbott Laboratories, Illinois) or durable RVAD (HeartWare device, Heartware, Medtronic, Minnesota). The remaining 41% of patients (n = 7) did not survive on RV-TH support, confirming the poor prognosis of RV failure.
Complications occurred in six patients (35%): one patient had epistaxis and hematemesis related to systemic anticoagulation; one patient had injury to left internal jugular due to inability to advance the catheter past the RV due to tortuous anatomy; two patients had intracranial bleeds on systemic anticoagulation; and two patients had bleeding at the catheter insertion site after placement toward the beginning of the implant experience.
Aside from paracorporeal or intracorporeal devices requiring surgical placement, there are presently only two available percutaneous options for RV support—the Impella RP (Abiomed, Danvers, MA) and the TPD. Results from the Impella RP trial in patients with RV failure (after LVAD or postcardiotomy) demonstrated that this device provided an average flow of 3 L, with improvements in RA pressures, cardiac index, and need for inotropic support.3 However, the Impella RP is placed via the femoral vein, preventing ambulation. In contrast, the TPD has the advantage of rapid internal jugular placement, allowing ambulation in carefully monitored subjects, and the use of an external ultrasonic flow probe that directly measures flow in the blood tubing during support.
Unfortunately, even with adequate reported TPD pump flow, more than 40% of patients in the present case series died. The mortality outcomes in this experience are similar to the unplanned RVAD experience reported by Columbia University.4 However, the potential benefits of this less invasive of RV support include avoidance of repeat sternotomy, and therefore a possibly less technically challenging future operation for transplant candidates, as well as the possibilities of preventing prolonged ventilation and costly selective pulmonary vasodilator therapy. The ease of percutaneous insertion may allow consideration of earlier support versus other surgical alternatives, improving outcomes in the future. In our opinion, the poor outcomes of RV failure are related to late treatment rather than early intervention. Severe right heart failure leads to hepatic congestion and dysfunction, which leads to coagulopathy. In these settings, it is not surprising that there were several hemorrhagic complications. With increased comfort with this intervention, it is likely that earlier placement will be considered, therefore resulting in improved outcomes. In many ways, this is analogous to the push to place left-sided mechanical circulatory support in candidates before “crash and burn” interagency for mechanically assisted circulatory support (INTERMACS) 1 status.
Clinical trials of the TPD alone or in comparison with the Impella RP have not been undertaken. The current series underscores the need for a clinical trial and highlights important considerations during device placement and patient management. As with any new device, there is a significant operator learning curve. The complication profile is driven by the requirement for anticoagulation, as well as the 29F catheter size, potential difficulties securing the device, and ensuring access at an appropriate point in the jugular vein to avoid complications due to anatomic and cannulation considerations.
Limitations of the current study include its two-center, retrospective nature, small numbers of patients, heterogeneous population, and the criteria for RVAD utilization and its weaning protocols were not prespecified. In addition, we do not have complete invasive hemodynamic and echocardiographic data before and after LVAD/RVAD. In spite of these limitations, the current study provides additional information regarding the feasibility of the TPD percutaneous RVAD in patients with severe RV failure such as post-LVAD or with rejection after heart transplant.
Critically important to the successful utilization of these devices will be to identify best practices for not only predicting but also managing RV failure, including prospectively determining which patients may benefit from planned biventricular support. One successfully described strategy is to place a temporary RVAD when unable to wean from cardiopulmonary bypass.5 Further data are required for a prospective evaluation of RV perioperative device management. This experience may provide equipoise in conducting a trial to determine the optimal strategy for the application of the TPD in RV failure.
1. Kirklin JK, Naftel DC, Pagani FD, et al.Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 14951504,
2. Lampert BC, Teuteberg JJRight ventricular failure after left ventricular assist devices. J Heart Lung Transplant 2015.34: 11231130,
3. Anderson M, Morris L, Tang D, et al.Impella RP Post Approval Study: First multi-center, prospective market approval results for the Impella RP in patients with right ventricular failure. J Heart Lung Transplant 2017.36: S64S65,
4. Takeda K, Naka Y, Yang JA, et al.Outcome of unplanned right ventricular assist device support for severe right heart failure after implantable left ventricular assist device insertion. J Heart Lung Transplant 2014.33: 141148,
5. Deschka H, Holthaus AJ, Sindermann JR, et al.Can perioperative right ventricular support prevent postoperative right heart failure in patients with biventricular dysfunction undergoing left ventricular assist device implantation? J Cardiothorac Vasc Anesth 2016.30: 619626,