The implantation of continuous left ventricular assist devices (LVADs) for circulatory support in heart failure patients has become routine. Approximately 30% of these patients require biventricular support,1 and in this study, only two long-term options are available: the total artificial heart (TAH), requiring that the native heart be excised, and the biventricular assist device (BVAD), using extracorporeal or implantable displacement pumps. We have reported a technique for the creation of an implantable BVAD.2 However, in patients with small chest dimensions, the right pump may become dislodged from its position or the flow may become obstructed after chest closure. In this study, we report a modification of the biventricular implantation of the HeartWare HVAD (HeartWare Inc.; Framingham, MA).
A 28-year-old man (body mass index [BMI], 18 kg/m2) with dilative cardiomyopathy presented with biventricular decompensation and beginning cardiogenic shock. Transesophageal echocardiography (TEE) showed tricuspid regurgitation of grades III and IV and dilated right ventricle. In addition, a small patent foramen ovale (PFO) with minimal shunt was diagnosed. According to our institutional criteria,3 biventricular support was the optimal solution in this case. We decided to implant two HeartWare HVAD (Framingham, MA) centrifugal pumps as described in detail elsewhere. 2,4 Briefly, one HeartWare pump was implanted into the apex of the LV and the outflow graft connected to the ascending aorta. Optimal position for the second HeartWare pump (as a right ventricular assist device [RVAD]) on the free wall of the right ventricle (at the point of maximal distance from the ventricular septum) was determined by TEE. The inflow cannula was placed in the right ventricular (RV) cavity and secured as described for the LVAD with two additional silicone rings (total 5 mm thickness) placed under the fixation ring to lower the part of the inflow cannula protruding into the RV. The outflow graft, narrowed before surgery from 10 to 5 mm of diameter, was sutured to the pulmonary artery. Both pumps were placed in the pericardium, and the chest was primarily closed. The RV improved after unloading and ejected up to 2 L/min. The pumps' speed was adjusted to achieve a flow of ∼5 L/min (left pump) and 3–4 L/min (right pump). After the patient's arrival on the intensive care unit, the right pump flow suddenly decreased to 2 L/min, and gas exchange was compromised requiring Fio2 of 1. In the following hours, the situation failed to stabilize. Transesophageal echocardiography showed significant right-left shunt through the PFO. The PFO was interventionally closed with an Amplatzer Septal Occluder (AGA-Medical, Golden Valley, MN); however, residual shunt of up to 5% still persisted, requiring Fio2 of 0.6. In addition, RV angiogram and repeated TEE showed compression of the right ventricle causing low flow of the right pump of maximally 2 L/min. On the next day, the patient was transferred to the operating room (OR). After bicaval cannulation and CPB initiation, the right atrium was opened, the Amplatzer device removed, and the PFO directly closed with sutures. The right pump was removed. As the right pump was placed into the right pleural cavity, a hole was created in the right-sided pericardium just opposite to the RA wall. The sewing ring of the HVAS was reinforced with two silicone rings and these three all-together sewn to the free right atrial wall. After punching a hole in the atrial wall, the pump was advanced through the pericardial hole and the rings into the right atrial cavity. In this way, the pericardium was fixed between the pump body and the rings (Figure 1). The anastomoses to the pulmonary artery and the outflow graft were left without modifications. The HVAD fixation ring on the free wall of the right ventricle was left in place, and the opening was closed with an individually designed titanium plug (Stephan Fittkau GmbH, Berlin, Germany), shown in Figure 2 and described by Potapov et al.5 The right pump flow increased immediately to 4–5 L/min, with subsequent increase of left pump flow also to up to 5 L/min. The postoperative course was uneventful, and the patient was discharged home. Targeted international normalized ration was 2.5–3.0. Within the next few months, no severe complications, suction events, or hospital admissions occurred. After 216 days on BVAD support, successful heart transplantation was performed.
This case report confirmed first the well-known finding that any atrial shunt detected preoperatively or intraoperatively should be directly closed. Compression of the right ventricle after chest closure may lead to an increase in right atrial pressure with increase of the shunt volume to the left atrium causing low pO2 and low flow of the right pump. Therefore, if interventional closure of a PFO fails, intraoperative revision is mandatory regardless of shunt volume.
Even more interesting should be the implantation site of the HeartWare pump cannula into the right atrium. An implantable centrifugal pump for RV support was investigated by Fukamachi et al.6 in animal experiments. Frazier et al. demonstrated feasibility of the TAH using intracorporeal continuous-flow pumps in animals, but both ventricles were excised. In one LVAD patient suffering from RV failure 2 weeks after surgery, they implanted the Jarvik 2000 as an RVAD into the right atrium.7 The implantation of two Heartware HVAD pumps as BVAD would offer the advantages of completely noiseless as well as more flexible and thinner drivelines compared with the pneumatically driven TAH or BVADs. Feasibility of implantation for the Heartware HVAD pumps as BVAD was reported.2,4 In these studies, the free wall of the right ventricle for the RVAD inflow cannula was used.2,4 Thanks to the small size of the HeartWare pump, it allows implantation in patients with low BMI, whereas in some patients with small chest dimensions, positioning on the free wall of the right ventricle will not be optimal and the surgeon will need a bail-out strategy.
This case showed that implantation of the HeartWare HVAD for RV support may be safely performed also in atrial position. Loforte et al.8 reported on a patient in whom the HVAD for RV support was placed on the diaphragmatic wall, which may have not solved the problem in this case.
In our opinion, a modification of the pump with a shorter inflow cannula would decrease the risk of inflow obstruction and decrease the height of the part of the pump placed outside the heart. This would make the use of the HVAD as RVAD or BVAD, in atrial and ventricular position, easier.
The authors thank Anne Gale, Editor in the Life Sciences, for editorial assistance.
1. Potapov EV, Loforte A, Weng Y, et al
: Experience with over 1000 implanted ventricular assist devices. J Card Surg
23: 185–194, 2008.
2. Hetzer R, Krabatsch T, Stepanenko A, et al
: Long-term biventricular support with the heartware implantable continuous flow pump. J Heart Lung Transplant
29: 822–824, 2010.
3. Potapov EV, Stepanenko A, Dandel M, et al
: Tricuspid incompetence and geometry of the right ventricle as predictors of right ventricular function after implantation of a left ventricular assist device. J Heart Lung Transplant
27: 1275–1281, 2008.
4. Strueber M, Meyer AL, Malehsa D, Haverich A: Successful use of the HeartWare HVAD rotary blood pump for biventricular support. J Thorac Cardiovasc Surg
140: 936–937, 2010.
5. Potapov EV, Stepanenko A, Hennig E, et al
: A titanium plug simplifies left ventricular assist device removal after myocardial recovery. J Heart Lung Transplant
29: 1316–1317, 2010.
6. Fukamachi K, Saeed D, Massiello AL, et al
: Development of DexAide right ventricular assist device: Update II. ASAIO J
54: 589–593, 2008.
7. Frazier OH, Myers TJ, Gregoric I: Biventricular assistance with the Jarvik FlowMaker: A case report. J Thorac Cardiovasc Surg
128: 625–626, 2004.
8. Loforte A, Montalto A, Della Monica PL, et al
: Biventricular support with the HeartWare implantable continuous flow pump: An additional contribution. J Heart Lung Transplant
29: 1443–1444, 2010.