Eur J Cardiothorac Surg 2003;24:834-836
© 2003 Elsevier Science NL
A new right ventricular assist device for right ventricular support
Stefan Christiansen*,
Stefan Brose,
Lütfü Demircan,
Rüdiger Autschbach
Department of Cardiothoracic Surgery, University of Aachen, Pauwelsstrasse 30, 52074 Aachen, Germany
Received 7 May 2003;
received in revised form 13 July 2003;
accepted 17 July 2003.
* Corresponding author. Tel.: +49-241-80-89221; fax: +49-241-80-82454
e-mail: schristiansen{at}ukaachen.de
 |
Abstract
|
|---|
Right ventricular support by mechanical devices for postcardiotomy right heart failure is still associated with a high mortality. We report on the first use of a new paracardiac microaxial blood pump for postcardiotomy right heart failure in two patients undergoing emergency coronary artery bypass grafting (the first patient for a myocardial infarction complicated by a left ventricular wall rupture, the second patient for a dissection of the right coronary artery after an interventional procedure).
Key Words: Right ventricular failure • Postcardiotomy failure • Right ventricular assist devices
|
1. Introduction
|
|---|
Isolated postcardiotomy right ventricular (RV) failure is rare but associated with a high mortality [1]. Only 15–35% of these patients survive to discharge from hospital [1,2]. We report on our first two patients undergoing RV support for postcardiotomy right heart failure with the new paracardiac RV microaxial blood pump (MBP, Recover 600®, Impella Cardiotechnik, Aachen, Germany).
|
2. Device description
|
|---|
The paracardiac RVMBP was approved by the MEDCERT Zertifizierungs- und Prüfungsgesellschaft für die Medizin GmbH, Hamburg, Germany (Process No.: PP-10752, Certificate No.: 10752GB41 1020523) for the use of a maximum duration of 7 days. It is a miniaturized rotary blood pump with a diameter of 6.4 mm and a weight of 11 g. The pump consists of a rotor driven by an incorporated brushless DC motor, the housing of the rotor, the inflow cage, the outflow cannula and the driveline. The inner volume of the pump is limited to 12 ml (avoiding the necessity to prime the pump), and the inner artificial and blood contacting surface to 65 cm2. At the maximal speed of 32,500 revolutions per minute, a flow of 6 l/min may be delivered under physiological pressure difference conditions. Flow can be adjusted to the patient's requirements by a variation of the rotational speed from 0 to 6 l/min. A pressure sensor is placed in front of the rotor measuring the pressure difference between the outflow and the inflow of the pump. Using the pressure difference and the rotational speed of the rotor, pump flow is calculated online. All these values are displayed on the driving console continuously. Purge fluid is used to prevent blood from entering the motor unit. Fig. 1 demonstrates a lab type of the RVMBP. The overall costs for this pump are 3500 Euro.
View larger version (92K):
[in this window]
[in a new window]
|
Fig. 1. The figure presents the implantation technique of the MBP for RV support. The inflow cage is implanted into the right atrium, and the outflow graft is anastomosed to the main pulmonary artery.
|
|
|
3. Clinical summary
|
|---|
3.1. Patient 1
A 57-year-old patient suffering from an acute myocardial infarction of the left ventricular (LV) anterior wall complicated by a LV wall rupture was referred for emergency coronary artery bypass grafting (CABG). CABG consisted of a left internal mammary artery bypass to the left anterior descending coronary artery (LAD) and closure of the perforation of the LV lateral wall. Weaning from extracorporeal circulation (ECC) was possible with catecholamines. Two days later, the patient presented a progredient hypotension despite continuously enhanced catecholamines. The echocardiographic study demonstrated a dilated, non-contractile right ventricle and a well-contracting, small left ventricle. The diagnosis of a predominant right heart failure was made and the RVMBP was implanted as depicted in Fig. 1. After reopening of the median sternotomy and complete anticoagulation (activated coagulation time >400 s), the right atrial appendage was side-clamped and a purse-string suture was performed around the appendage. After incision of the right atrium, the inflow cage was implanted into the right atrium and fixed with the purse-string suture. After that, the main pulmonary artery was side-clamped and the outflow graft anastomosed. The pump was started with a slow rotational speed for deairing via this anastomosis. Thereafter, the side-clamp was removed and the pump flow adjusted to 3.5–4 l/min. Heparin was partially antagonized by protamine. The activated coagulation time was adjusted to 160–200 s during the pump run. During the next days, the pump flow was reduced gradually controlling RV function by echocardiographic studies. After 3 days, RV function was completely recovered. Therefore, the RVMBP was explanted. We did not observe any thrombotic material at the pump or thromboembolic events during the pump run. The patient recovered well and was extubated on the 7th postoperative day. The overall drainage loss was 1850 ml. Altogether, 4 units of packed red blood cells and 6 units of fresh frozen plasma were administered. Unfortunately, the patient developed a renal failure requiring a prolonged postoperative hospital stay. After 35 days, the patient was discharged home in good overall condition.
3.2. Patient 2
A 64-year-old women suffered from a myocardial infarction due to an occlusion of the right coronary artery (RCA). Furthermore, the LAD also demonstrated a high-grade proximal stenosis. The interventional procedure for recanalization resulted in a dissection of the RCA. Therefore, the patient was referred for emergency CABG. Preoperatively the patient had to be resuscitated with drugs and chest compressions due to a ventricular fibrillation. Cardiac function was severely impaired, so that beating heart CABG was performed with ECC. The left internal mammary artery was anastomosed to the LAD and a reversed segment of the greater saphenous vein to the RCA. Despite extensive reperfusion (overall duration of ECC: 184 min) and implantation of an intraaortic balloon pump the patient could not be weaned from ECC due to a RV failure. Therefore, the RVMBP was implanted. After that, weaning from ECC was possible. Anticoagulation was done as described in patient 1. Furthermore, the patient suffered from an acute renal failure requiring a continuous veno-venous hemofiltration. Pump flow was reduced gradually, so that the RVMBP could be explanted on the 8th postoperative day after recovery of RV function without any signs of thromboembolic events or thrombus development within the pump. Total drainage loss was 1460 ml requiring transfusion of 4 units of packed red blood cells and 4 units of fresh frozen plasma. Unfortunately, the patient expired 2 days later because of a therapy-refractory ventricular fibrillation.
|
4. Discussion
|
|---|
RV failure has a multifactorial etiology with prolonged ECC, RV ischemia, pulmonary hypertension, transfusion of numerous blood products, and use of vasopressors being the most frequent causes [3]. Although rare, RV failure accounts for a mortality greater than 50% after circulatory support for postcardiotomy failure [2]. As one of the most important figures influencing mortality and nonweaning from temporary circulatory support [2], development of new RV support devices appears to be necessary considering the poor results with currently used devices. This becomes more and more true regarding the considerable number of complications with these RV support devices [1–3]. Other devices which are currently used for RV support are the Abiomed BVS 5000 and the Biomedicus pump. Samuels et al. [4] reported on their experiences with the Abiomed BVS 5000 in 45 patients (25 left heart support, ten RV support and ten biventricular assist devices), i.e. postcardiotomy shock in 36 patients (80%) and precardiotomy shock in nine patients (20%) for a mean of 8.3 days. Twenty-two patients (49%) were weaned and 14 (31%) discharged from hospital. The most common complications included bleeding (78%) and adverse neurologic events (22%). Transient neurologic events occurred with increasing frequency as the duration of support was extended beyond the first week. Similar results are reported by Noon et al. [5] for the Biomedicus pump. The pump was implanted in 141 patients for postcardiotomy failure for a mean of 3.8 days, i.e. in 110 patients for LV assistance, in eight patients for RV assistance, and in 23 patients for biventricular assistance. Altogether, 54% of all patients were weaned from cardiac support but only 22% were discharged. For isolated RV support these figures are 12.5% and 0%. Bleeding was described as a frequently occurring complication requiring reexploration for removal of excessive blood clots or relief of cardiac tamponade.
Apart from virtues of new blood pumps, they have to meet the safe and highly accepted standards of currently used devices. Important aspects of blood pumps are lifetime, costs, adaptability to diverse applications and patient requirements, rapid and easy deployment, thrombogenicity, flow characteristics and blood damage [4]. The major advantages of the paracardiac RVMBP are its small size (inner volume: 12 ml, inner surface: 65 cm2), the simple design, the low energy requirements, and the avoidance of a priming volume. These figures lead to a reliable pump function without technical failures and may result in a reduced number of transfusion requirements which is reported to be excessively high in patients with extracorporeal membrane oxygenation support [6]: 87.3% of patients undergoing extracorporeal membrane oxygenation support require blood transfusions.
Flow and pressure characteristics, shear stress, blood exposure times to artificial surfaces and the size of the pump are important factors contributing to hemolysis [7]. Extracorporeal membrane oxygenation requires long and thin cannulas which in turn requires a higher pumping pressure resulting in an enhanced degree of hemolysis [7]. These theoretical assumptions are confirmed by Magovern et al. [6] who reported that platelet destruction and hemolysis are common in patients undergoing extracorporeal membrane oxygenation support. The reduced artificial surfaces, the short outflow cannula and the avoidance of a priming volume in this RVMBP may be reasons for the low degree of hemolysis of this pump.
The effects of pulsatile and nonpulsatile flow on pulmonary vasculature and function are an ongoing topic of debate. Whereas some authors described beneficial effects of pulsatile flow [8,9], other authors could not confirm these results [9,10]. A uniform finding of most of the studies is that pulsatile flow results in a decreased pulmonary vascular resistance compared to that seen with nonpulsatile flow [9]. This can be explained by an enhanced release of endothelium-derived relaxing factor by rhythmic stimulation of endothelial cells through oscillating changes in vessel wall shear stress [8]. But on the other hand, this effect of pulsatile flow does not appear to improve pulmonary gas exchange, as demonstrated by Brandes et al. [10]. They did not observe any advantages in the pulsatile flow group regarding peak inspiratory pressure, mean pulmonary artery pressure, oxygenation capacity, and development of pulmonary edema. Therefore, it seems to be justified to support the RV with nonpulsatile flow devices, especially when support duration is limited to a few days as in patients with postcardiotomy failure. Furthermore, it must be taken into consideration that pulmonary flow starts to be pulsatile with recovery of native RV function.
|
Acknowledgments
|
|---|
This study was supported by Bundesministerium für Bildung und Forschung, Germany.
|
References
|
|---|
- Moazami N., Pasque M.K., Moon M.R., Herren R.L., Bailey M.S., Lawton J.S., Damiano R.J. Mechanical support for isolated RV failure in post-cardiotomy patients. J Heart Lung Transplant 2003;22(Suppl 1):S206-S207.
- Kitamura M., Aomi S., Hachida M., Nishida H., Endo M., Koyanagi H. Current strategy of temporary circulatory support for severe cardiac failure after operation. Ann Thorac Surg 1999;68:662-665.[Abstract/Free Full Text]
- Chen J.M., Levin H.R., Rose E.A., Addonizio L.J., Landry D.W., Sistino J.J., Michler R.E., Oz M.C. Experience with right ventricular assist devices for perioperative right-sided circulatory failure. Ann Thorac Surg 1996;61:305-310.[Abstract/Free Full Text]
- Samuels E., Holmes E.C., Thomas M.P., Entwistle J.C., Morris R.J., Narula J., Wechsler A.S. Management of acute cardiac failure with mechanical assist: experience with the Abiomed BVS 500. Ann Thorac Surg 2001;71:S67-S72.[Abstract/Free Full Text]
- Noon G.P., Lafuente J.A., Irwin S. Acute and temporary ventricular support with Biomedicus centrifugal pump. Ann Thorac Surg 1999;68:650-654.[Abstract/Free Full Text]
- Magovern G.J., Simpson K.A. Extracorporeal membrane oxygenation for adult cardiac support: the Allegheny experience. Ann Thorac Surg 1999;68:655-661.[Abstract/Free Full Text]
- Kawahito K., Nose Y. Hemolysis in different centrifugal pumps. Artif Organs 1997;21:323-326.[Medline]
- Champsaur G., Vedrinne C., Martinot S., Tronc F., Robin J., Ninet J., Franck M. Flow-induced release of endothelium-derived relaxing factor during pulsatile bypass: experimental study in the fetal lamb. J Thorac Cardiovasc Surg 1997;114:738-745.[Abstract/Free Full Text]
- Hickey P.R., Buckley M.J., Philbin D.M. Pulsatile and nonpulsatile cardiopulmonary bypass: review of a counterproductive controversy. Ann Thorac Surg 1983;36:720-737.[Abstract]
- Brandes H., Albes J.M., Conzelmann A., Wehrmann M., Ziemer G. Comparison of pulsatile and nonpulsatile perfusion of the lung in an extracorporeal large animal model. Eur Surg Res 2002;34:321-329.[Medline]
This article has been cited by other articles:
|
|
|
|
 
A. Boening, C. Friedrich, D. Caliebe, and J. Cremer
Efficacy of intracardiac right ventricular microaxial pump support during beating heart surgery
Interactive CardioVascular and Thoracic Surgery,
September 1, 2004;
3(3):
495 - 498.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
1. Introduction
